PTJA11-Cefiderocol-Core-Submission-Dossier-V1.0.Pdf

Home , Aminopenicillin, Cefiderocol, Morganellaceae

Core Submission Dossier PTJA11

Cefiderocol For the treatment of infections due to aerobic Gram-negative organisms in adults with limited treatment options

Submitted by: Shionogi

Contact details for administrative purposes Shionogi BV

33 Kingsway London WC2B 6UF

Email address: [email protected]

For agency completion Date of receipt: 14-04-2020

Version 3: Ameded dossier reflecting additional PK/PD analysis. Identifier: PTJA11

Disclaimer: The sole responsibility for the content of this document lies with the submitting manufacturer and neither the European Commission nor EUnetTHA are responsible for any use that may be made of the information contained therein.

Abbreviations

A Aerobic AAT Appropriate antibacterial therapy ABC transporter ATP-binding cassette transporter ABSSSI Acute bacterial skin and skin structure infection Ac-BSI Acinetobacter spp. Bacteraemia AE Adverse event AET Appropriate empirical therapy ALAT Asociación Latinoamericana del Tórax AMK Amikacin AMR Antimicrobial resistance AN Anaerobic AR Antimicrobial-resistance AS Antimicrobial susceptibility AST Antimicrobial susceptibility tests AT Antibacterial therapy ATS American Thoracic Society AUC Area under the curve BAT Best available therapy BD Becton Dickinson BIA Budget impact analysis BIM Budget impact model BAL Bacterial β-lactamase BLI β-lactamase inhibitor BSI Bloodstream infection BSIMRS Bloodstream infection mortality risk score CAI Community-acquired infection CarbNS Carbapenem non-susceptible CASR Carbapenem- and ampicillin-sulbactam-resistant CAZ Ceftazidime/avibactam CDC Centres for Disease Control and Prevention CDI Clostridium difficile infection CFU Colony forming unit CHMP Committee for Medicinal Products for Human Use CI Confidence interval

cIAI Complicated intra-abdominal infection CLSI US Clinical & Laboratory Standards Institute CNSE Carbapenem-non-susceptible Enterobacteriaceae CPE Carbapenemase Producing Enterobacteriaceae CPFX Ciprofloxacin CR Carbapenem-resistant CRAB Carbapenem-resistant A. baumannii CRAc Carbapenem-resistant Acinetobacter spp. CRE Carbapenem-resistant Enterobacteriaceae CRGNB Carbapenem-resistant Gram-negative CRGNIs Carbapenem-resistant Gram-negative infections CRPA Carbapenem-resistant P. aeruginosa CSE Carbapenem-susceptible Enterobacteriaceae CSPA Carbapenem-susceptible P. aeruginosa CTX Cefotaxime cUTI Complicated urinary tract infection DALYs Disability-adjusted life-years DBO Diazabicyclooctane DGI German Society for Infectious Diseases Association DRG Disease-related groups EA Early assessment EAU European Association of Urology ECDC European Centre for Disease Prevention and Control EEA European Economic Area eHRB Emerging highly antibacterial resistant bacteria EMA European Medicines Agency EMEA Europe, Middle East, and Africa EOT End of treatment ERS European Respiratory Society ESBLs Extended-spectrum β-lactamases ESCMID European Society of Clinical Microbiology and Infectious Diseases ESICM European Society of Intensive Care Medicine EU European Union EUCAST European Committee on Antimicrobial Susceptibility Testing

FA Facultative anaerobic FDA U.S. Food and Drug Administration FUP Follow-up G3CREC Third-generation cephalosporin-resistant E. coli G3CSEC third-generation cephalosporin-susceptible E. coli GDP Gross domestic product Spanish Society of Infectious Diseases and Clinical GEIH-SEIMC Microbiology GNO Gram-negative organisms GVD Global Value Dossier HAI Hospital-acquired infection HAP Hospital acquired pneumonia HAS Haute Authorite de la Sante HCAI Healthcare-associated infection HCAP Healthcare-associated pneumonia Hr Hour HTA Health Technology Assessment HTAB Health Technology Assessment Body (c)IAI (complicated) Intra-abdominal infection IAT Inappropriate antibacterial therapy ICD International Classification of Disease ICU Intensive care unit ID-CAMHB Iron-depleted cation-adjusted Mueller Hinton broth IDSA Infectious Disease Society IET Inappropriate empiric therapy IMP IMP-type carbapenemases IPM/CS Imipenem/Cilastatin IQR Inter-quartile range IRAB Imipenem-resistant Acinetobacter baumannii ITT Intention-to-treat IV Intravenous KAPE Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter KPC Klebsiella pneumoniae carbapenemase LOS Length of stay

MA Marketing Authorization MAA Marketing Authorization Application MBLs Metallo-β-lactamases MCO Managed care organization MCR-1 Plasmid-mediated colistin-resistance MDR Multidrug resistant MDRA Multidrug resistant Acinetobacter MDRAB Multidrug resistant A. baumannii MDRP Multidrug resistant P. aeruginosa MDS Multidrug-sensitive ME Microbiologically evaluable (HD) MEPM (high dose) Meropenem MIC Minimum inhibitory concentration mITT Microbiological intention to treat MoA Mode of action NDA New drug application NDM New Delhi metallo-β-lactamase NHS National Health Service NI Nosocomial infections NICE National Institute of Health and Care Excellence NR Non-resistant NS Non- survivors OM Osteomyelitis OMT Outer membrane transporters OR Odds ratio OXA Oxacillinase PBC Positive blood culture PBPs Penicillin-binding proteins PCR Polymerase Chain Reaction PD Pharmacodynamic PDCO Paediatric Committee PDR Pan-drug-resistant PEG Percutaneous endoscopic gastroscopy PER Pseudomonas extended resistant β-lactamases PK Pharmacokinetic

PP Per Protocol PTA Probability of target attainment q8h Every 8 hours R Resistant RCT Randomized controlled trial RESP Respiratory tract RR Relative risk r-GNR Resistant gram-negative rod RTI Respiratory tract infection S Survivors SAE Serious adverse event SC Subcutaneous SCCM Society of Critical Care Medicine SD Standard deviation SEFH Spanish Society of Hospital Pharmacies Spanish Society of Preventive Medicine, Public Health and SEMPSPH Hygiene SICU Surgical intensive care unit SIS Surgical Infection Society SMC Siderophore monobactam conjugate sNDA Supplemental new drug application SOC Standard of care spp Species SSI Surgical site infection TOC Test of Cure tRNA Transfer ribonucleic acid UTI Urinary tract infection VAP Ventilator-acquired pneumonia VABP Ventilator-associated bacterial pneumonia VIM Verona integrin-encoded metallo-β-lactamase w/wo With or without WHO World Health Organization WSES World Society of Emergency Surgery XDR Extensively drug-resistance

Contents

EXECUTIVE SUMMARY ...... 14 1 Description and technical characteristics of the technology ...... 23 1.1 Characteristics of the technology ...... 25 1.1.1 Cefiderocol Structure ...... 26 1.1.2 Mechanism of action and cell entry ...... 27 1.1.3 Stability against β-lactamases ...... 28 1.2 Regulatory status of the technology ...... 29 2 Health problem and current clinical practice ...... 31 2.1 Overview of the disease or health condition ...... 32 2.1.1 Overview of Gram-negative bacteria ...... 33 2.1.2 Antimicrobial resistance ...... 35 2.1.3 Overview of infection sites ...... 39 2.1.4 Risk and prognostic factors for MDR and CR infections ...... 41 2.1.5 Epidemiology ...... 42 2.1.6 Mortality ...... 47 2.1.7 Quality of Life ...... 48 2.1.8 Disability Adjusted Life Years (DALYs) ...... 48 2.1.9 Delayed effective therapy ...... 49 2.2 Target population ...... 52 2.3 Clinical management of the disease or health condition ...... 57 2.3.1 Key information on currently available treatments in Europe ...... 59 2.3.2 Site-specific vs. pathogen-specific guidelines...... 63 2.3.3 Specific recommendations...... 63 2.3.4 Specific considerations of CR infections ...... 63 2.4 Comparators in the assessment ...... 87 2.4.1 General considerations ...... 87 2.4.2 Selection of relevant comparators for the assessment ...... 89 3 Current use of the technology...... 94 3.1 Current use of the technology...... 95 3.2 Reimbursement and assessment status of the technology ...... 96 4 Investments and tools required...... 97 4.1 Requirements to use the technology ...... 98 4.1.1 Conditions for use ...... 99 4.1.2 Good stewardship and societal considerations...... 99 5 Clinical effectiveness and safety ...... 102 5.1 Identification and selection of relevant studies ...... 105 5.1.1 PRISMA Chart ...... 111 5.1.2 Study categorisation ...... 111 5.2 Relevant studies ...... 112 5.3 Main characteristics of studies...... 134 5.3.1 APEKS-cUTI STUDY ...... 144 5.3.2 APEKS-NP STUDY ...... 151 5.3.3 CREDIBLE-CR STUDY ...... 155 5.3.4 Summary of compassionate use cases and published evidence ...... 160 5.4 Individual study results (clinical outcomes) ...... 164 5.4.1 Individual study results (in vitro surveillance outcomes) ...... 164 5.4.2 Individual study results (PK/PD data, study report S-649266-CPK-004- B) ...... 189

5.4.3 Retrospective analysis of cefiderocol and comparators by population PK/PD simulation ...... 192 5.4.4 Clinical study results (clinical outcomes) ...... 194 5.4.5 Resistance against Cefiderocol ...... 233 5.5 Individual study results (safety outcomes) ...... 259 5.5.1 Overall safety results: pooled analysis and individual studies: APEKS- cUTI, APEKS-NP, and CREDIBLE CR ...... 259 5.5.2 Safety analyses by clinical trial ...... 265 5.6 Conclusions ...... 287 5.6.1 Evidence to support use of cefiderocol in patients with infections by suspected MDR/CR pathogens: ...... 289 5.6.2 Evidence to support use of cefiderocol in patients with infections by confirmed CR pathogens: ...... 291 5.6.3 Quality of Life ...... 293 5.6.4 Comparators ...... 293 5.7 Strengths and limitations ...... 297 5.7.1 Risk of bias assessment ...... 297 5.7.2 Discussion ...... 301 REFERENCES ...... 306

List of Figures

Figure 1: Cefiderocol structure ...... 27 Figure 2: Cefiderocol mechanism of cell entry ...... 28 Figure 3: Antibacterial activity against β-lactamase-producing pathogens ...... 28 Figure 4: Classification of Gram-negative bacteria ...... 34 Figure 5: Global burden of AMR ...... 35 Figure 6: Mechanisms of beta lactam bacterial resistance ...... 37 Figure 7: Hospital-acquired infections in acute care hospitals (EU/EEA 2011-2012) ...... 40 Figure 8: Worldwide carbapenem resistance ...... 42 Figure 9: Prevalence of CR Gram-negative infections in the EU-5 ...... 43 Figure 10: Epidemiology of carbapenemases in EU 5 ...... 44 Figure 11: Confirmed carbapenemase-producing Enterobacteriaceae isolates (Public Health England: 2008–17) ...... 45 Figure 12: Distribution of carbapenem resistance mechanisms in Enterobacteriaceae species in the Europe...... 45 Figure 13: Summary of effect of appropriate versus inappropriate initial antibacterial therapy on mortality ...... 51 Figure 14: Summary of effect of delay versus no delay in receiving initially appropriate antibacterials on mortality ...... 51 Figure 15: Summary of effect of appropriate versus inappropriate therapy on treatment failure ...... 52 Figure 16 - Treatment of patients with highly suspected infection by CR or other MDR GN pathogens ...... 55 Figure 17: Treatment of patients with confirmed infection by carbapenem-resistant or other MDR Gram-negative pathogen ...... 55 Figure 18: Current treatment approach for bacterial infections ...... 57 Figure 19: Current clinical reasoning for the treatment of serious MDR Gram-negative infections ...... 59 Figure 20 - Search strategy for OVD MEDLINE ALL ...... 106 Figure 21 - PRISMA flow diagram of record selection process ...... 111 Figure 22: APEKS-cUTI study design ...... 144 Figure 23: Subject disposition (all randomized subjects) ...... 145 Figure 24: Distribution of uropathogens (mITT population) ...... 150 Figure 25: APEKS-NP study design and patient flow ...... 152 Figure 26: Patient demographics and baseline characteristics ...... 153 Figure 27: CREDIBLE-CR study design and patient flow ...... 156 Figure 28: Subjects disposition (all randomized subjects) ...... 157 Figure 29: APEKS-cUTI study design and endpoints ...... 196 Figure 30: Primary efficacy results: Composite outcome at TOC in the MITT population .. 197 Figure 31: Primary efficacy results: Composite outcome at TOC by predefined subgroups198 Figure 32: Maximum Network Chart for Network Meta-analysis ...... 209 Figure 33: Network Diagram for Microbiological Eradication Secondary Outcome ...... 209 Figure 34: Microbiological Eradication Rates at TOC - Frequentist Analysis ...... 210 Figure 35: Microbiological Eradication Rates at TOC - Bayesian Analysis ...... 210 Figure 36: Network Diagram for Clinical Cure Outcome ...... 210 Figure 37: Clinical cure rates at TOC - Frequentist Analysis ...... 211 Figure 38: Clinical Cure rate at TOC - Bayesian Analysis ...... 211 Figure 39: Clinical cure rates at FU - Frequentist Analysis ...... 211 Figure 40: APEKS-NP study design ...... 213 Figure 41: All-cause Mortality (mITT) ...... 214 Figure 42: Primary efficacy results: Day 14 All-cause Mortality by Subgroups ...... 215 Figure 43: Day 14 and Day 28 all-cause mortality according to MIC for meropenem ...... 218 Figure 44: Microbiological eradication by MIC at EOT ...... 219 Figure 45: CREDIBLE CR study design ...... 223

Figure 46: Clinical cure by Clinical Diagnosis and time point ...... 224 Figure 47: Microbiological eradication by Clinical Diagnosis and time point ...... 224 Figure 48: Clinical and Microbiological Outcomes at TOC in Enterobacteriaceae by Carbapenemase or Porin Channel Mutation (CR Micro-ITT Population) ...... 226 Figure 49: Clinical and Microbiological Outcomes in Metallo Β-lactamase Producing Gram- negative Pathogens (CR Micro-ITT Population) ...... 226 Figure 50: All-cause Mortality Rates by Type of Infection ...... 227 Figure 51: Mortality rates comparison across studies ...... 230 Figure 52: Network Diagram for Safety Analysis ...... 271 Figure 53: Safety Analysis for All Adverse Events - Frequentist Analysis ...... 271 Figure 54: Network for safety analysis for Treatment related AEs ...... 271 Figure 55: safety analysis for Treatment related AEs – Frequentist analysis ...... 271

List of Tables

Table 1: Features of the technology ...... 25 Table 2: Administration and dosing of the technology ...... 25 Table 3: Regulatory status of the technology ...... 29 Table 4: List of the highest priority bacteria (WHO) ...... 35 Table 5: In vitro activity profile of antibacterials for GN Infections with limited treatment options ...... 39 Table 6: Most common CR causal pathogens across available EU-5 data sources ...... 43 Table 7: Proportion of CR infection sites in the EU-5...... 44 Table 8: Overview of disease burden according to the infection site ...... 46 Table 9: In Vitro Gram-negative activity profiles ...... 54 Table 10a: Relevant guidelines for diagnosis and management – MDR/GN Bacteria ...... 65 Table 11b: Relevant guidelines for diagnosis and management – HAP/VAP(HCAP) ...... 70 Table 11c: Relevant guidelines for diagnosis and management – cUTI ...... 75 Table 11d: Relevant guidelines for diagnosis and management – BSI/Sepsis ...... 78 Table 11e: Relevant guidelines for diagnosis and management- cIAI ...... 83 Table 12: Cefiderocol assessment ...... 90 Table 13: Overview of the reimbursement status of the technology in European countries . 96 Table 14: Databases and information sources searched ...... 107 Table 15: Inclusion and exclusion criteria ...... 109 Table 16: List of all relevant studies ...... 113 Table 17: Study characteristics ...... 135 Table 18: Patient demographics and baseline characteristics (mITT population) ...... 147 Table 19: Patient demographics and baseline characteristics (mITT population) ...... 154 Table 20: Top 5 baseline Gram-negative pathogens, n (%) ...... 154 Table 21: Patient demographics and baseline characteristics (ITT population) ...... 158 Table 22: Summary of study regimen for Gram-negative pathogen at day 1 and day 2 (CR- mITT population) ...... 159 Table 23: Baseline Gram-negative pathogens, n (%) ...... 160 Table 24: Patient demographics and baseline characteristics ...... 162 Table 25: SIDERO Surveillance studies ...... 165 Table 26: In vitro activity data for all tested clinical strains (SIDERO-WT-2014/2015/2016 and Proteeae) of cefiderocol (at MIC of 4mg/L) versus ceftazidime-avibactam, ceftolozane- tazobactam, and colistin ...... 167 Table 27: In vitro activity of cefiderocol and comparators against Gram-negative bacilli isolated by 55 clinical laboratories in Europe in 2015 (n=5352) ...... 170 Table 28: In vitro activity of cefiderocol and comparators against non-fermenters ...... 172 Table 29: Breakpoints for non-susceptibility used in definition of DTR (μg/mL) ...... 173 Table 30: Susceptibility of cefiderocol and comparators to pathogens ...... 173 Table 31: In vitro activity data for CR Gram-negative pathogens (SIDERO-WT-2016-2017) of cefiderocol versus ceftazidime-avibactam, ceftolozane-tazobactam and colistin ...... 174 Table 32: Number of MEM-NS isolates by year and species ...... 175 Table 33: Number of MEM-NS isolates by country and species ...... 175 Table 34: Susceptibility breakpoints according to the CLSI (cefiderocol) and/or EUCAST (all comparators) ...... 176 Table 35: Percentage of susceptibility of MEM-NS A. baumannii complex by country ...... 177 Table 36: Percentage of susceptibility of MEM-NS P. aeruginosa complex by country ...... 177 Table 37: Percentage of susceptibility of MEM-NS K. pneumoniae by country ...... 177 Table 38: Percentage of susceptibility of other MEM-NS Enterobacteriaceae by country .. 177 Table 39: In vitro activity data for all tested clinical strains (SIDERO-CR 2014-2016) of cefiderocol versus ceftazidime-avibactam, ceftolozane-tazobactam, and colistin ...... 178 Table 40: MIC of cefiderocol and comparators in Germany ...... 179 Table 41: MIC of cefiderocol and comparators in Greece ...... 180 Table 42: MIC of cefiderocol and comparators in Spain ...... 181

Table 43: MIC of cefiderocol and comparators against in United Kingdom and Ireland ..... 182 Table 44: Activity of antimicrobial agents tested against carbapenem-resistant P. aeruginosa and S. maltophilia ...... 183 Table 45: MIC of cefiderocol and comparators for MDR-GN isolated ...... 184 Table 46: Number of cefiderocol non-susceptible isolated in global surveillance studies (MIC ≥8 μg/mL)...... 185 Table 47: EUCAST breakpoints for cefiderocol ...... 186 Table 48: Susceptibility to Cefiderocol and comparators in all sites of infections for MDR3 pathogens ...... 187 Table 49- Theoretical success of antibacterial therapy in Gram‐negative 3MDR pathogens in gastrointestinal site of infections (A) Pneumonia; (B) cUTI; (C) BSI; (D) Gastrointestinal .. 187 Table 50: Summary table Theoretical percentage of success for Gram‐negative antibacterial therapy on aerobic Gram‐negative pathogens in different infection type ...... 188 Table 51: PTA per infectious disease renal function, and dose ...... 190 Table 52. Estimated CFR for MIC distributions corresponding to Enterobacterales and Pseudomonas spp. More simulation results for corresponding PTA, MIC and T>MIC target values are shown in Appendix C. The applied MIC distributions can be seen in Appendix D of the study report...... 193 Table 53: Endpoint Analysis as per EUnetHTA Request ...... 194 Table 54: Summary for Composite of Clinical and Microbiological Outcome by Time Point (Microbiological Intent-to-Treat Population) ...... 197 Table 55: Composite of Clinical Response and Microbiological Outcome per Pathogen at TOC (microbiological ITT population) ...... 199 Table 56: Summary of Clinical Outcomes per Subject by Time Point (Microbiological Intent- to-Treat Population) ...... 200 Table 57: Summary of Clinical Outcome per Uropathogen (E. coli, K. pneumoniae, P. aeruginosa, and P. mirabilis) by Time Point (Microbiological ITT Population) ...... 201 Table 58: Summary of Microbiological Outcome per Subject by Time Point (Microbiological ITT Population) ...... 203 Table 59: Summary of Microbiological Outcome per Uropathogen (E. coli, K. pneumoniae, P. aeruginosa, P. mirabilis) by Time Point (Microbiological ITT Population)...... 205 Table 60: Day 14 All-cause Mortality (mITT and ME-PP Populations) ...... 214 Table 61: Secondary Endpoints (mITT Population) ...... 216 Table 62: Secondary Endpoints (mITT Population) ...... 216 Table 63: Clinical and microbiological outcome per baseline pathogen ...... 217 Table 64: Microbiological and Clinical Outcome for the Meropenem-non-susceptible Subgroup (mITT Population) ...... 219 Table 65: Susceptibility and effectiveness model predicting outcomes for Cefiderocol versus comparators in UTI ...... 221 Table 66: Susceptibility and effectiveness model predicting outcomes for Cefiderocol versus comparators in Pneumonia ...... 221 Table 67: Clinical cure and microbiological eradication by baseline CR-pathogen ...... 225 Table 68: Summary for All-cause Mortality in the Study (Intent to treat Population) ...... 227 Table 69: Summary for all-cause mortality overall by pathogens subgroup (Enterobactereacea and non-fermenters) ...... 229 Table 70: CREDIBLE-CR study: Mortality subgroup Analysis for Subjects with A. baumannii (safety population) ...... 229 Table 71: Mortality and serious adverse events ...... 231 Table 72: Summary of MIC shift ...... 234 Table 73a: Methods of data collection and analysis of Mortality ...... 235 Table 80b: Methods of data collection and analysis of Clinical outcomes ...... 237 Table 80c: Methods of data collection and analysis of Composite microbiological eradication and cure ...... 245 Table 80d: Methods of data collection and analysis of Microbiological outcomes ...... 249 Table 80e: Methods of data collection and analysis of Susceptibility rates ...... 258

Table 81: Dose and Duration of Exposure to cefiderocol* (Number of Patients by Indication) ...... 259 Table 82: Subjects with Treatment Related Adverse Events by System Organ Class and Preferred Term (All Phase II/III Studies) Safety Population ...... 261 Table 83: Summary of duration of exposure (safety population) ...... 266 Table 84: Summary of treatment-emergent adverse events (safety population) ...... 266 Table 85: Number (%) of subjects with adverse events by maximum severity (safety population) ...... 268 Table 86: Number (percent) of subjects with serious adverse events (SAEs) by organ class and preferred term (safety population) ...... 269 Table 87: Number (%) of subjects with treatment-related serious adverse events (SAEs) 270 Table 88: Overview of Treatment-emergent Adverse Events (Safety Population) ...... 272 Table 89 – Number (percent) of subjects with serious adverse events (SAEs) by organ class and preferred term (safety population) ...... 275 Table 90: Overview of Treatment-emergent Adverse Events (Safety Population) ...... 279 Table 91: Subjects with Treatment-related Adverse Events by Preferred Term (Safety Population) ...... 280 Table 92: Subjects with Serious Adverse Events by System Organ Class and Preferred Term (Safety Population) ...... 281 Table 93: Limitations to detect adverse events in clinical trial programmes ...... 283 Table 94: Methods of data collection and analysis of AE, TEAE and SAE ...... 284 Table 95 - Comparator overview ...... 294 Table 96: Risk of bias on study level – Randomized trials with cefiderocol ...... 300

EXECUTIVE SUMMARY

AMR is a major, growing threat to public health Anti-microbial resistance (AMR) is a major and growing threat to global public health [1]. Infection by multi-drug-resistant (MDR) and particularly carbapenem-resistant (CR) pathogens is associated with a high mortality rate and increased morbidity and economic burden [2, 3]. In 2015 in the EU, 33,110 deaths were attributable to infections due to antibacterial-resistant bacteria [4] and it has been estimated that, if significant action is not taken by the year 2050, 10 million lives will be lost each year due to AMR [1].

Nosocomial MDR infections (including CR), caused by Gram-negative (GN), aerobic bacteria including CR Escherichia coli, CR Klebsiella pneumoniae, CR Pseudomonas aeruginosa, CR Acinetobacter baumannii (WHO priority pathogens; [5-7]) and intrinsically CR Stenotrophomonas maltophilia [8, 9] are particularly relevant, as there are limited treatment options for these and particularly for carbapenem resistant pathogens [5, 6]. They primarily occur in vulnerable hospitalised patients resulting in hospital acquired pneumonia and ventilator acquired pneumonia (HAP/VAP), bloodstream infections (BSI), complicated urinary tract infection (cUTI) and complicated intra-abdominal infections (cIAI), amongst other infections [10-12]. Currently, an antibacterial susceptibility test (AST) is needed for a definitive prescription, which can take more than 3 days [13, 14], so patients with infections involving resistant pathogens are more difficult to treat and therefore, patients are more likely to receive multiple courses of inappropriate therapies before an effective treatment is initiated. This delay can lead to increased mortality and clinical burden, poorer outcomes, increasing the likelihood of developing new resistances [15-21]. Furthermore, where CR infection is suspected in critically ill patients, an antibacterial regimen is started immediately, despite incomplete information on pathogen susceptibility, with the antibacterial (or combination of antibacterials) that has the highest likelihood of success. The selection of antibacterial(s) should be guided by knowledge of local epidemiology (local resistance profile and local pathogen distribution), as well as by site of infection and patient specific factors, such as severity of illness and previous antibacterial exposure or comorbidities. Treatment may be de-escalated to a more targeted treatment once the AST results have been obtained [13]. This further emphasizes the critical importance of susceptibility testing and the need for antibacterials with a wider spectrum of activity targeting MDR/PDR strains, especially because studies found that inappropriate initial treatment and the subsequent delay in effective treatment results in worse outcomes including increased mortality, length of stay and treatment costs [15-21].

Given the lack of treatment options, there are few defined standard of care (SoC) or guidelines defining the most appropriate treatment strategy for MDR and CR Gram-negative bacteria (GNB) [22]. Specific treatments for MDR-GNB infections are predominantly multiple drug combinations that include one or more of the following: aminoglycosides (amikacin, gentamicin); tetracyclines (tigecycline; eravacycline; minocycline) carbapenems (e.g. ertapenem; imipenem-cilastatin, meropenem; meropenem/vaborbactam; imipenem/relebactam/cilastatin); β-lactamase inhibitor combinations (ceftazidime/avibactam; ceftolozane/tazobactam); fosfomycin; or polymyxins (colistin and polymyxin B) [23-27].

Carbapenems, due to their potency, broad-spectrum activity, and less frequent resistance, have until recently for reasons of antimicrobial stewardship, been reserved for use in treatment of patients with resistant bacterial infections that could not be treated with other beta-lactams. Current treatment options, for treatment of CR pathogens [28], have either suboptimal efficacy (e.g. carbapenems), limited pathogen and/or mechanism of resistance coverage (e.g. ceftolozane/tazobactam; ceftazidime/avibactam; meropenem/vaborbactam) [29, 30] and/or significant safety and tolerability concerns [e.g. colistin, tigecycline]) [31-34].

Even recently approved combinations of cephalosporins with established β-lactam/β- lactamase inhibitors have activity against MDR Gram-negative infections, including P. aeruginosa, but their limitations include a lack of activity against metallo-β-lactamase- producing organisms and these new antibacterials remain vulnerable to resistance mechanisms due to porin channel mutations or overexpression of efflux pumps [28, 35-43].

Despite having high rates of renal toxicity, the broad Gram-negative spectrum of colistin and polymyxin B mean that they are still used in the absence of alternative effective treatment options for increasingly emerging CR in Gram-negative bacteria [31].

New treatments that can overcome the known resistance mechanisms, are therefore needed, contributing to more effective eradication of MDR pathogens and increase antibacterial diversity, thus, supporting good stewardship and the overall effectiveness of the existing arsenal of antibacterials.

CEFIDEROCOL overcomes the 3 main mechanisms of antibacterial resistance present in Gram-negative pathogens and is active on WHO critical priority pathogens

Cefiderocol is the first siderophore cephalosporin [44] to be approved. The cephalosporin core of cefiderocol exerts it’s activity through inhibition of Gram-negative bacterial cell wall biosynthesis leading to cell lysis. Its unique molecular structure catecholate siderophore moiety, exploits the bacteria’s own active iron uptake mechanism via siderophores to enter the periplasmic space of GNB where it exerts its bactericidal activity. This is a novel mechanism of bacterial cell entry which means that cefiderocol, unlike other antibacterials, bypasses pathways traditionally used by other antibacterials such as efflux pumps or porin channels, which bacteria can regulate to reduce their exposure to antibacterials. Cefiderocol also has a higher stability to both serine- and metallo-type β -lactamases, key enzymes rendering resistance to β–lactam antibacterials, including carbapenems. All these factors contribute to cefiderocol’s unique breadth of activity and efficacy, covering a wide range of aerobic, GN bacteria, demonstrated by its potent activity (both in vitro and in-vivo) against all three WHO priority CR pathogens (Enterobacteriaceae, A. baumannii and P. aeruginosa) [29, 30, 45-49]. In addition, cefiderocol has in vitro activity against intrinsically CR Stenotrophomas maltophilia and Burkholderia cepacia [30].

The dosing regimen of cefiderocol is 2g administered every 8 hours by IV infusion over 3 hour period, with treatment duration dependent on the site of infection, e.g. 5-10 days for cUTI and cIAI and 7-14 days for hospital-acquired pneumonia, but treatment up to 21 days may be required [50].

The indication for cefiderocol is expected to be:

 Fetcroja is indicated for the treatment of infections due to aerobic Gram-negative organisms in adults with limited treatment options.

This indication will therefore be pathogen focused, not restricted to any specific site of infection and supports the use of cefiderocol in two types of patients:

 Hospitalised patients with suspected (but prior laboratory confirmation) MDR/CR infection who are critically ill and require immediate antibacterial treatment that provides full cover against CR pathogens and potential resistant mechanisms, to avoid the risk of rapid clinical deterioration (with the option to de-escalate to a more targeted treatment when the pathogen and susceptibility profile is subsequently confirmed)

 Hospitalised patients where CR infection has been confirmed and cefiderocol is best option based on pathogen susceptibility information and/or where other treatment choices are inappropriate (efficacy, contra-indication or tolerability).

OVERVIEW OF PRE-CLINICAL AND CLINICAL EVIDENCE Unlike other therapeutic areas, the evaluation of the effectiveness of an antibacterial relies on the combined consideration of in vitro, Pharmacokinetic (PK)/Pharmacodynamic (PD) and clinical data. Cefiderocol’s favourable in vitro minimum inhibitory concentrations (MICs) correlate well with in vivo efficacy in PK/PD in vivo efficacy in validated animal models of infection, including MDR pathogens. Randomized clinical trials in patients with complicated urinary tract infections (cUTI) [51], nosocomial pneumonia (HAP/VAP/HCAP), and BSI have provided confirmation of the good efficacy and safety of cefiderocol in key target patient populations.

In vitro evidence shows cefiderocol has activity in >95% of CR Gram-negative isolates

In vitro activity of cefiderocol has been studied in two large surveillance studies (SIDERO- WT/Proteeae and SIDERO-CR 2014/2016) [29, 30, 45, 46] and many country specific smaller similar studies. The SIDERO-WT study tested the in vitro antibacterial activity of cefiderocol against Gram-negative bacteria [29]. A total of 30,459 clinical isolates of Gram-negative bacilli were systematically collected from USA, Canada, and 11 European countries between 2014 and 2017. Cefiderocol demonstrated activity against 99.5% of Gram-negative isolates at a MIC of 4 mg/L. Isolates were less susceptible to the comparators including colistin (95.5%), ceftazidime-avibacatam (90.2%) and ceftolozane-tazobactam (84.3%).

In the SIDERO-CR-2014-2016 study [30], which was a global study of 52 countries, focusing only on CR isolates, cefiderocol demonstrated potent in vitro activity at a MIC of 4 mg/L against 96.4% of isolates of carbapenem-nonsusceptible pathogens including all of the WHO priority pathogens and Stenotrophomas maltophilia. Cefiderocol was found to provide a wider Gram- negative coverage, and more potent in vitro antimicrobial activity than comparators including ceftazidime/avibactam (39.8%), ceftolozane/tazobactam (37%), and colistin (91.5%).

PK/PD studies predict (probability >90%) that the dosing regimen achieves a concentration of free drug in plasma > MIC for 75% dosing period As for other cephalosporins, %fT>MIC is the best predictor of efficacy for cefiderocol. A dosing regimen delivering 75% T>MIC succeeded achieving at least 1 log10 kill reducing the number of viable bacterial cells in both murine thigh infection and murine lung infection by at least 90% regardless of the isolate used to induce the infection (E. coli, K. pneumoniae, P. aeruginosa, A. baumannii or S. maltophilia).

A 3-compartment model was used to describe the plasma concentrations of cefiderocol. A 3- compartment pharmacokinetic population model was developed based on pharmacokinetic data from healthy volunteers, patients with renal impairment and patients from the clinical trials. Probability of Target Attainment (PTA) for 75% fT>MIC was above 97% for a MIC of 4 mg/L regardless of the site of infection or the renal function. In the epithelial lining fluid (ELF), PTA for 75% fT>MIC was above 88% for a MIC of 4 mg/L confirming the adequacy of the dosing regimen in the different patient populations. The dosing regimen therefore ensures sufficient drug exposure to maximise the efficacy of cefiderocol.

Evidence from a streamlined clinical trial programme supports the in vitro data

An improved in vitro potency in addition to a well-characterized favorable PK/PD profile are crucial to achieve both adequate exposure to the antibacterial over the MIC for the pathogen, and clinical cure in patients infected with drug-resistant pathogens [52]. Therefore, clinical studies in antimicrobials, provide only supportive safety and efficacy evidence to the pivotal in vitro and PK/PD data. Furthermore, in the context of antibacterial resistance, the standard clinical trial approach aiming at demonstrating superiority over existing treatments is not feasible. Treatment options for MDR infections do not allow a superiority trial and it would be unethical to wihthold effective treatment to pateints in such trials [52]. Hence, clinical trials have an important role to confirm clinical efficacy, but a limited role in providing comparative evidence outside the trial, as only pathogens that fall within the in vitro spectrum of the tested treatments and comparators are included in the study. This is particularly relevant for antimicrobial treatment selection in the absence of antibiogram.

The clinical efficacy and safety of cefiderocol was demonstrated in 2 randomised double- blinded clinical trials and 1 open label, descriptive study.

The APEKS-NP study compared treatment with cefiderocol against the combination of high- dose (HD), prolonged infusion meropenem in patients with nosocomial pneumonia caused by MDR Gram-negative pathogens. Three hundred (300) patients were randomized 1:1 to cefiderocol or HD meropenem, a regimen only used in more difficult-to-treat pathogens which optimizes exposure and efficacy for meropenem. Cefiderocol met the primary endpoint of non- inferiority in all-cause mortality (ACM) at day 14 versus HD meropenem (12.4% for cefiderocol and 11.6% for HD meropenem; [95 % CI: -6.6, 8.2]) and similar results maintained for ACM at Day 28 and end of study (EOS). Rates of clinical cure and microbiological eradication at TOC were also similar between the treatment groups. Although patients with CR-pathogens known prior to randomization were excluded from the study, in a meropenem-nonsusceptible subgroup (MIC>8mg/L) later identified, the rates of ACM at Day 14 were 17.1% in the

cefiderocol group and 20.0% in the HD meropenem group. Adverse events were similar between cefiderocol and HD meropenem and cefiderocol safety profile was consistent with other cephalosporins.

APEKS-cUTI was an international, multicenter, randomised, double-blind, active-controlled, parallel-group, non-inferiority study to investigate the efficacy and safety of cefiderocol vs imipenem/cilastatin (IPM/CS) in cUTI caused by Gram-negative MDR pathogens in hospitalisd adults [51, 53]. 448 patients were randomized, of whom 300 received cefiderocol and 148 received IPM/CS. The primary efficacy endpoint was the composite of clinical response and microbiological response rate at TOC assessment, in the MITT (microbiological Intent-to-treat) population. The results demonstrated that 73% of patients in the cefiderocol group achieved the primary endpoint, vs only 55 % of patients in the IPM/CS group, with an adjusted treatment difference of 18.6% (95 % CI: 8.2, 28.9). This difference showed superiority in favour of cefiderocol in a post-hoc analysis. Adverse events were similar in type and rate between treatment groups and cefiderocol safety profile was consistent with other cephalosporins.

A Network Meta-Analysis (NMA) was feasible for cUTI, given the similarity of patients and pathogens included across trials. All results were consistent with APEKS-cUTI trial and showed no statistically significant differences compared to ceftazidime/avibactam and ceftolozane/tazobactam in a similar patient population with similar pathogen distribution.

The CREDIBLE CR study was a small, exploratory, open label, randomised, descriptive study to evaluate efficacy of cefiderocol and best available therapy (BAT) in critically ill patients with confirmed CR infections, but was not designed or powered for statistical comparison between arms. The study included 150 severely ill patients, (48 allocated to BAT) consistent with compassionate use cases, with a range of infection sites including nosocomial pneumonia, cUTI, BSI/sepsis. Many patients had end stage comorbidities and had failed multiple lines of therapy. Clinical and microbiological outcomes were similar between the 2 arms, but there were marked imbalances in some baseline clinical relevant characteristics and pathogen distribution of the cefiderocol and BAT arms.

Cefiderocol has proven efficacy in complex compassionate use cases to date More than 200 patients to date have been treated with cefiderocol within the compassionate use programme around the world, highlighting the unmet medical need for alternative antibacterials active against CR Gram-negative pathogens. Confirmed information on 74 patients who have completed treatment in this program showed that over 60% of the severely ill patients infected with CR Gram-negative pathogens survived when no other treatment option was available to them.

In the absence of AST results, cefiderocol is estimated to provide better predicted susceptibility rates and projected clinical success rates considering the European Gram-negative pathogen epidemiology When critically ill patients require immediate treatment in the absence of AST results, the likelihood of treatment success with cefiderocol and comparators can be predicted through an effectiveness model based on estimates of pathogen prevalence for the specific site of infection, combined with pathogen susceptibility results for each infection site (taken from the SIDERO surveillance studies); relying on the drug’s ability to achieve effective concentrations at the site of infection. Such methodologies are used when ethical considerations limit the prospective clinical evaluation of treatments by randomized control trials, i.e. where the risk of exposing patients to potentially ineffective drugs in a clinical trial setting is too great.

Results from this effectiveness model showed that cefiderocol is expected to have higher predicted susceptibility rates than comparators across different infection sites in the European prevalent Gram-negative bacteria, and higher projected treatment success rates in cUTI and pneumonia. These were consistent with trials results from APEKS cUTI and APEKS NP for cefiderocol, but not for comparators as it includes pathogens for which they are not susceptible. This modelling approach highlights the limitations of the existing clinical trials, and the potential difference for the effectiveness rates, when antimicrobials are used in the absence of AST.

Cefiderocol presents a safety profile consistent with other cephalosporins

The clinical safety for cefiderocol was established in the three randomised clinical trials, including 549 treated patients, and showed a similar profile compared to other cephalosporins. Pooled adverse event analyses showed that there were overall less treatment emergent adverse events with cefiderocol (344/549 [67.1%]) vs comparators (252/347 [72.6%]), as well as less treatment related AEs, (56/549 [10.2%]) with cefiderocol vs compartors (45/347 [13.0%]).

In the nosociomial pneumonia study treatment-emergent adverse events (TEAEs) and treatment-related TAEs were balanced between arms. SAEs occurred in 36% of patients treated with cefiderocol and 30% of patients treated with meropenem. The most frequently observed AE was urinary tract infection (15.5% in cefiderocol and 10.7% in meropenem group), hypokalemia (10.8% vs 15.3%) and anemia (8.1% vs 8%). In the cUTI study the proportion of patients who experienced at least one adverse event (AE) was lower in the cefiderocol group than in the IPM/CS group (41 % vs 51%). The most

frequently observed AEs were gastrointestinal, such as diarrhoea [experienced by 4.3% and 6.1% of cefiderocol- and IPM/CS-treated subjects, respectively], and there was an numerical increased incidence of C. difficile colitis in the IPM/CS arm compared with cefiderocol. Serious adverse events (SAE) occurred less in cefiderocol-treated patients than in IPM/CS-treated patients (5% vs 8%).

CR study (CREDIBLE-CR): The cefiderocol group had a lower incidence of AEs and treatment-related AEs, but a higher incidence of death, SAEs and discontinuation due to AEs, than was observed for patients receiving BAT. The incidence of treatment-related AEs leading to discontinuation was similar between treatment groups. A blinded adjudication committee concluded that none of the deaths in the cefiderocol arm was due to a drug-related AE, although one death due to acute kidney injury in the BAT arm was attributed to colistin-based therapy. Furthermore, whereas the mortality rate in the cefiderocol group was consistent with previous studies in similar populations the evidence suggests that the mortality rate in the BAT group was unexpectedly low for the population randomised.

CONCLUSION

Cefiderocol is an innovative, effective and well tolerated treatment for aerobic GN infections in patients with limited treatment options. Cefiderocol overcomes the common resistance mechanisms of GN pathogens and covers a broad range of aerobic, GN bacteria including all three WHO priority CR pathogens (Enterobacteriaceae, A. baumannii and P. aeruginosa) and the CR Stenotrophomas maltophilia and Burkholderia cepacia. It provides an important alternative for physician managing patients with MDR/CR infections.

Cefiderocol’s favourable in vitro MICs across all relevant pathogens correlates well with in vivo efficacy in PK/PD analyses. Randomized clinical trials in patients with cUTI, nosocomial pneumonia (HAP/VAP/ HCAP), and BSI and sepsis have provided confirmation of the good efficacy and safety of cefiderocol in key target patient populations, alongside compassionate use case reports. The combination of in vitro, PK/PD, and clinical data predicts that cefiderocol has a greater likelihood of obtaining clinical success rates, in patients with suspected MDR/CR infections than relevant comparators across different infection sites. Cefiderocol provides an important new option for treating critically ill, hospitalised patients where MDR/CR infection is suspected and time to effective treatment must be minimised, and for patients where an MDR/CR infection has been confirmed and it is the most appropriate option, due to pathogen susceptibility or where other treatment choices are inappropriate.

1 Description and technical characteristics of the technology

Summary of the characteristics of the technology  Cefiderocol is the first siderophore cephalosporin [44] to be approved. It’s unique molecular structure and novel mechanism of cell entry allow it to overcome the three major resistance mechanisms found in Gram-negative pathogens (i.e., degradation by β-lactamase enzymes, porin channel mutations and overexpression of efflux pumps):

o Cefiderocol has improved stability to hydrolysis by β-lactamases, including all 4 types of carbapenemases, key enzymes rendering resistance to β–lactam antibacterials, including carbapenems.

o Cefiderocol exploits the bacteria’s need for iron and mimics the action of bacterial own siderophores. A chelate complex with free iron is formed, which is then actively transported into the bacterial cell via iron transporters, circumventing pathways traditionally used by other antibacterials such as efflux pumps or porin channels, which bacteria can regulate to reduce their exposure to antibacterials.

 Cefiderocol is active against a wider range of aerobic, GN bacteria than its comparators (including all WHO priority pathogens: CR Enterobacteriaceae, CR P. aeruginosa and CR A. baumannii). In addition, cefiderocol is also active against intrinsically CR Stenotrophomas maltophilia and Burkholderia cepacia.

 In Europe, Shionogi seeks a pathogen-focused indication for cefiderocol, and it is expected to be approved for the treatment of infections due to aerobic GN organisms in adults with limited treatment options. Within this indication, it is proposed that cefiderocol offers most value in two clinical scenarios, and evidence for cefiderocol and its relevant comparators is provided for each:

o Hospitalised patients with suspected (but prior laboratory confirmation) MDR/CR infection who are critically ill and require immediate antibacterial treatment that provides full cover against CR pathogens and potential resistant mechanisms, to avoid the risk of rapid clinical deterioration (with the option to de-escalate to a more targeted treatment when the pathogen and susceptibility profile is subsequently confirmed).

o Hospitalised patients where CR infection has been confirmed and cefiderocol is best option based on pathogen susceptibility information and/or where other treatment choices are inappropriate (efficacy, contra-indication or tolerability).

 Cefiderocol was approved by the U.S. Food and Drug Administration (FDA) on November 14, 2019, for treatment of cUTI in adult patients with limited or no alternative treatment options. Based on the results of the recently presented APEKS NP study Shionogi is preparing a sNDA submission to FDA for approval of cefiderocol for pneumonia in 2020.

 Evaluation of the effectiveness of an antibacterial requires the integrated analysis of in vitro, PKPD and clinical data.

o Two large susceptibility studies, SIDERO-WT/Protea and SIDERO-CR 2014/2016), showed cefiderocol to have activity against 99.5% of GN isolates and 96.2% of CR GN isolates respectively. This was higher than other tested antibacterials, according to CLSI breakpoints. This was replicated in several small country specific studies, with consistency results.

o Cefiderocol’s favourable in vitro MICs correlate well with in vivo efficacy in PK/PD analyses conducted.

o Three clinical trials (APEK cUTI, APEKS NP and CREDIBLE CR) have provided confirmation of the efficacy and safety of cefiderocol in key infection types: cUTI, nosocomial pneumonia (HAP/VAP/HCAP), and BSI.

o In the absence of AST results, and in an integrated effectiveness model analysis of European pathogen epidemiology, in vitro/in vivo data, and clinical data, cefiderocol provides the best predicted susceptibility rates and projected clinical success rates considering for the EU setting.

 Cefiderocol provides an important new option for treating critically ill, hospitalised patients where MDR/CR infection is suspected and time to effective treatment must be minimised, and also for patients where an MDR/CR infection has been confirmed and it is the most appropriate option, due to pathogen susceptibility or where other treatment choices are inappropriate

1.1 Characteristics of the technology

1. In Table 1 provide an overview of the technology.

Table 1: Features of the technology

Non-proprietary name Cefiderocol

Proprietary name FETCROJA

Marketing Shionogi B.V., Amsterdam, Netherlands authorisation holder

Class Antibacterials for systemic use

Active substance(s) Siderophore cephalosporin

Pharmaceutical Powder for concentrate for solution for infusion (powder for concentrate). formulation(s) White to off-white powder.

ATC code J01DI04 cefiderocol

Mechanism of action Cefiderocol is a siderophore cephalosporin. In addition to passive diffusion through outer membrane porin channels, cefiderocol can bind to extracellular free iron via its siderophore side chain, allowing active transport into the periplasmic space of Gram-negative bacteria through siderophore uptake systems. Cefiderocol subsequently binds to penicillin binding proteins (PBPs), inhibiting bacterial peptidoglycan cell wall synthesis which leads to cell lysis and death.

2. In Table 2, summarise the information about administration and dosing of the technology.

Table 2: Administration and dosing of the technology

Method of administration Intravenous use; administered by intravenous infusion over 3 hours.

Doses 1 g/vial; the recommended dose for individuals with normal renal function is 2g over 3h infusion

Dosing frequency Every 8 hours (three times daily)

Average length of a course of 3-hour infusion of 2g; Overall duration of treatment is in treatment accordance with the site of infection.

Anticipated average interval Each treatment cycle lasts 8 hours; 3h of infusion and then 5h between courses of treatments until the next cycle begins.

Anticipated number of repeat For complicated urinary tract infections including courses of treatments pyelonephritis and complicated intra-abdominal infections the

recommended treatment duration is 5 to 10 days. For hospital- acquired pneumonia including ventilator-associated pneumonia the recommended treatment duration is 7 to 14 days. Treatment up to 21 days may be required.

Dose adjustments Dose adjustments are necessary for patients with renal impairment (reduced dose) or augmented renal function (increased dose)

3. State the context and level of care for the technology (for example, primary healthcare, secondary healthcare, tertiary healthcare, outside health institutions or as part of public health or other).

Cefiderocol is administered by intravenous infusion over 3h every 8h. It is intended for hospitalised, critically ill patients and therefore, is intended for hospital-use only.

4. State the claimed benefits of the technology, including whether the technology should be considered innovative.

Cefiderocol is the first siderophore cephalosporin [44] to be approved. Its unique molecular structure catecholate siderophore moeity, exploits the bacteria’s own active iron uptake mechanism via siderophores to enter the periplasmic space of GNB where it exerts its bactericidal activity by inhibiting bacteria cell wall synthesis. This is a novel mechanism of bacterial cell entry which means that, unlike other antibacterials, cefiderocol bypasses pathways traditionally used by other antibacterials such as efflux pumps or porin channels, which bacteria can regulate to reduce their exposure to antibacterials. Cefiderocol also has a higher stability to both serine- and metallo-type β -lactamases, key enzymes rendering resistance to β–lactam antibacterials, including carbapenems. All these factors contribute to cefiderocol’s unique breadth of activity and efficacy, covering a wide range of aerobic, GN bacteria, demonstrated by its potent activity (both in vitro and in vivo) against all three WHO priority CR pathogens (Enterobacteriaceae, A. baumannii and P. aeruginosa) [29, 30, 45-49]. In addition, cefiderocol has in vitro activity against intrinsically CR Stenotrophomas maltophilia and Burkholderia cepacia [30].

1.1.1 Cefiderocol Structure

Cefiderocol has a pyrrolidinium group on the C-3 side chain, which improves antibacterial activity and stability against β-lactamases (Figure 1) [48]. The major difference in the chemical

structure of cefiderocol and other cephalosporins is the addition of a chlorocatechol group on the end of the C-3 side chain, which confers the siderophore activity (Figure 1) [48].

Figure 1: Cefiderocol structure

Source: Sato, 2019 [54]

Cefiderocol has been granted with an ATC code J01DI04: cefiderocol.1 The J01D group (other β-lactam antibacterials) comprises beta-lactam antibacterial, other than penicillins.

1.1.2 Mechanism of action and cell entry

The low levels of free iron in the human body during an infection induce pathogens to upregulate iron acquisition factors, such as secretion of iron-binding small molecules called siderophores into their environment and production of membrane-bound active iron transporters [44, 55-58]. Bacterial siderophores tightly bind to host iron, forming a chelated iron complex, which then penetrates through the outer membrane via active iron transporters located in the GN outer membrane [44, 54].

Cefiderocol exploits the bacteria’s need for iron for cell growth and uses the bacteria’s own active iron uptake mechanism to enter the periplasmic space of GN bacteria where it binds to penicillin binding proteins (PBPs), inhibiting the bacterial cell wall synthesis causing killing the bacteria [44, 48, 59].

1 https://www.whocc.no/ddd/lists_of_new_atc_ddds_and_altera/new_atc/?order_by=1

Cefiderocol has been designed to chelate iron to form an iron complex similar to a bacterial catecholate siderophore (Figure 1) [44, 54]. When bound to iron, the cefiderocol iron complex mimics a bacterial siderophore-iron complex and therefore is actively transported through the outer membrane using the bacteria’s active iron transporters, bypassing pathways traditionally used by other antibacterials such as efflux pumps or porin channels, which bacteria can regulate to reduce their exposure to antibacterials [44, 54, 60]. Even in the absence of forming a complex with iron, cefiderocol can still function as other antibacterials, entering the bacterial periplasm via passive diffusion through porin channels (Figure 2) [59].

Cefiderocol activity against bacterial strains with porin channel mutations and overexpression of efflux pumps has been demonstrated in two in vitro studies [61, 62].

Figure 2: Cefiderocol mechanism of cell entry

Source: Image adapted from Zhanel 2019 [44]

1.1.3 Stability against β-lactamases

Figure 3: Antibacterial activity against β-lactamase-producing pathogens

*Color-coding based on the pathogen susceptibility: Green – Activity reported, yellow – undetermined activity reported, and red – no clinically relevant activity reported. Source: Thalhammer F, 2018 [63], Theuretzbacher, 2019 [64]

The structure of cefiderocol also presents higher stability to hydrolysis across a wide range of bacterially produced β-lactamase enzymes (including carbapenemases of the serine and metallo-β-lactamase classes) and thus overcomes the primary mechanism of bacterial resistance to beta–lactam antibacterials, without adding a β-lactamase inhibitor.

The image above (Figure 3) shows that whilst there are effective alternatives for Extended Spectrum Β-lactamase (ESBL), there are limited treatment options for serine cabapenemases, and metallo-bectalactamases, as well as other mechanisms of resistance. This is particulary relevant for non-fermenters Pseudomonas aeruginosa, Acinetobacter baumanii, and Stenotrophomonas maltophilia.

All these factors contribute to cefiderocol’s unique breadth of activity and demonstrated efficacy, covering a wide range of aerobic, GN bacteria and cefiderocol’s demonstrated potent activity (both in vitro and in vivo) against all three WHO priority carbapenem resistant pathogens (Enterobacterales, A. baumannii and P. aeruginosa). In addition, cefiderocol has in vitro activity against carbapenem resistant Stenotrophomas maltophilia and Burkholderia cepacia.

1.2 Regulatory status of the technology

1. Complete Table 3 with the marketing authorisation status of the technology.

Table 3: Regulatory status of the technology

Launched Organisation (Expected) (yes/no). Verbatim wording of the (expected) issuing Date of If no include indication(s) approval approval proposed date of launch FETROJA is a cephalosporin Not launched. November 14, FDA antibacterial indicated in patients 18 Expected to be 2019 years of age or older who have limited launch in Q1 2020

or no alternative treatment options, for the treatment of complicated urinary tract infections (cUTI), including pyelonephritis caused by susceptible Gram-negative microorganisms Fetcroja is indicated for the treatment of Not launched infections due to aerobic Gram-negative EMA May, 2020 Expected to launch organisms in adults with limited in 2H, 2020 treatment options EMA: European Medicines Agency, FDA: U.S. Food and Drug Administration

2. State any other indications not included in the assessment for which the technology has marketing authorisation.

Not applicable. This corresponds to a new marketing authorization for a new chemical entity.

3. State any contraindications or groups for whom the technology is not recommended.

It is recommended that Fetcroja should be used to treat patients that have limited treatment options only after consultation with a physician with appropriate experience in the management of infectious diseases.

Cefiderocol should not be given to the following patients:

 Patients with hypersensitivity to the active substance or to any of the excipients listed

 Patients with hypersensitivity to any cephalosporin antibacterial medicinal product.

 Patients with severe hypersensitivity (e.g. anaphylactic reaction, severe skin reaction) to any other type of beta-lactam antibacterial agent (e.g. penicillins, monobactams or carbapenems).

The safety and efficacy of cefiderocol in children below 18 years of age has not yet been established. No data are available. Patients with Central Nervous System infections were also not included in the cefiderocol clinical trials.

4. List the other countries in which the technology has marketing authorisation.

Currently cefiderocol only has marketing authorization in the United States of America, under the brand name of Fetroja, with the indication mentioned in Table 3.

2 Health problem and current clinical practice

Summary of issues relating to the health problem and current clinical practice  AMR is a major and growing, threat to global public health. Infections by MDR (and particularly CR) pathogens are associated with a high mortality rate and increased morbidity and economic burden. In 2015 in the EU, 33,110 deaths were attributable to infections due to antibacterial-resistant bacteria and it has been estimated that, if significant action is not taken by the year 2050, 10 million lives will be lost each year due to AMR. [4]

 MDR and particularly CR infections are predominantly caused by Gram-negative, aerobic bacteria. Given the increasing prevalence of resistant pathogens, WHO has declared that the availability of new treatments for CR Gram-negative pathogens to be a critical priority for CR strains of Enterobacteriaceae, Pseudomonas aeruginosa and Acinetobacter baumannii including them on the WHO “Priority One CRITICAL List” [65].

 Current treatment options, including antibacterial combinations, have either suboptimal efficacy (carbapenems), limited pathogen and mechanism of resistance coverage (ceftolozane/tazobactam; ceftazidime/avibactam; meropenem-vaborbactam), and/or significant safety and tolerability concerns (e.g. colistin, tigecycline).

o Treatment of confirmed CR infections must be tailored to each patient based on results of AST results, knowledge of hospital and regional pathogen epidemiology and patient specific factors, such as severity of concomitant illness and infection. Therefore, few guidelines exist in Europe identifying a specific treatment approach in CR infection.

o resistance to many antimicrobial classes almost invariably reduces the probability of adequate empirical coverage, with possible unfavourable consequences

o In the absence of an AST results (which commonly takes 3 days) and suspected MDR/CR infection, treatment should start immediately in critically ill patients to avoid risk of rapid deterioration. However, resistance to many antimicrobial classes almost invariably reduces the probability of adequate initial treatment, with possible unfavourable consequences vs those where the pathogen is susceptible, and therefore easier to treat with available treatments

(46.5% vs. 11.8%, p < 0.001). These patients experience delays in receiving effective treatment which may lead to worse outcomes, increased mortality, and increased length of hospital stay and treatment costs.

 The integrated cefiderocol data-set (susceptibility, PK/PD and clinical trials) shows it has a greater likelihood of treatment success in patients with MDR/CR infections than relevant comparators. Cefiderocol therefore provides an important new option for treating critically ill, hospitalised patients where MDR/CR infection is suspected and time to effective treatment must be minimised, and also for patients where an MDR/CR infection has been confirmed and it is the most appropriate option, due to pathogen susceptibility or where other treatment choices are inappropriate.

 The availability of cefiderocol as an additional effective agent against Gram-negative bacteria can contribute to good antibacterial stewardship by allowing physicians to increase the diversity of prescribing, reducing selective pressure on any one agent and minimize development of resistance.

2.1 Overview of the disease or health condition

1. Define the disease or health condition in the scope of this assessment.

The human body harbours over 1000 types of bacteria constituting the normal microbial flora. In healthy individuals, these bacteria do not usually cause infection and exist on the host for long periods without causing harm [66]. However, invasion of the host by pathogenic microorganisms that proliferate results in tissue injury [67]. Pathogenic bacteria can infect any part of the human body. Infections can be acquired in the community setting (community- acquired infection [CAI]), acquired in hospital setting (Hospital-acquired infection – [HAI] or nosocomial infection) or acquired in long-term care facilities (Healthcare-associated infection [HCAI]) such as intensive care wards, ambulatory settings, nursing homes or rehabilitation facilities. Infections caused by multidrug-resistant bacteria (MDR) are more likely to be a HAI.

Nosocomial infections primarily occur in vulnerable hospitalised patients. These patients are often ≥ 50 years of age, likely to be severely ill, e.g. transplanted patients, possibly in intensive care units (ICU), or undergoing chemotherapy, or patients who have compromised immunogenicity, and generally wuth multiple comorbidities (e.g. heart disease, diabetes or kidney disease) [68, 69]. Infections caused by MDR pathogens can occur at many sites including the urinary tract (complicated urinary tract infections [cUTI]), lungs (Hospital-

acquired and ventilator-associated pneumonia [HAP/ VAP]), blood (bloodstream infections [BSI]), and intra-abdominal sites (complicated intra-abdominal infections [cIAI]).

The International Classification of Disease (ICD, version 10) contains separate codes for pathogens as well as for infections sites. ICD-codes for nosocomial Gram-negative infections are included as an appendix [70].

2.1.1 Overview of Gram-negative bacteria

Gram-negative bacteria can be classified into fermenters and non-fermenters based on their ability to ferment glucose [71]. While non-fermenting Gram-negative bacteria (non-fermenters) are usually found in nature, they are harmful when colonizing and infecting immunocompromised people or when the infections are a consequence of trauma or invasive procedures (e.g. surgery, intravenous catheters, respiratory care equipment or endotracheal tubes) [71]. Bacteria are also differentiated based on cellular morphology (most commonly bacilli and cocci) and oxygen requirements (aerobes and anaerobes) (Figure 4) [72].

Figure 4: Classification of Gram-negative bacteria

Pathogens highlighted in blue are of interest when exploring the topic of carbapenem resistance Source: The Ohio State University, 2017 [73]; Adeolu, 2016 [74]

2.1.2 Antimicrobial resistance

Anti-microbial resistance (AMR) is a major, and growing, threat to global public health. Treatment of pathogens has become more and more challenging due to the emergence of resistance, especially to carbapenems which are usually reserved for use where other alternative options have failed [5, 6, 75-79]. AMR is estimated to contribute to 700,000 deaths every year globally, with 33,110 lives lost per year in Europe [1, 4, 80]. The burden of infections with bacteria resistant to antibacterials in the European population is comparable to that of influenza, tuberculosis and HIV/AIDS combined [81]. It has been estimated that, if significant action is not taken, by the year 2050 10 million lives will be lost each year due to AMR (Figure 5) [1]. While the global consumption in antibacterials is predicted to rise three-fold by 2030, the current treatment options may address only a subset of resistance mechanisms [1].

Figure 5: Global burden of AMR

Source: O’Neill [1]

An overview of medically important GN bacteria classified based on WHO’s priority criteria for Drug development is provided in Table 4 [1].

Table 4: List of the highest priority bacteria (WHO)

Priority Pathogen

Critical Carbapenem-resistant Acinetobacter baumannii, Carbapenem-resistant Pseudomonas aeruginosa,

Carbapenem-resistant/3rd generation cephalosporin-resistant Enterobacterales, (including Klebsiella pneumonia, Escherichia coli, Enterobacter spp., Serratia spp., Proteus spp., and Providencia spp, Morganella spp.) High Clarithromycin-resistant Helicobacter pylori, Fluoroquinolone-resistant Campylobacter Fluoroquinolone-resistant Salmonella spp 3rd generation cephalosporin-resistant/fluoroquinolone-resistant Neisseria gonorrhoeae, Medium Ampicillin-resistant Haemophilus influenzae Fluoroquinolone-resistant Shigella spp

Source: WHO 2017 [5, 6]

GN pathogens are challenging to treat due to their potential intrinsic resistance to antibacterials and the emergence of acquired resistance [75]. This development has been recognized to be a major public health threat. Outbreaks of infections with resistant strains have been reported in several European countries [82, 83]).

Facultative anaerobes E. coli, A. baumannii, K. pneumoniae and P. aeruginosa, are a cause of great concern with regard to antimicrobial resistance (AMR) [8, 79]. Non- fermenters such as P. aeruginosa, A. baumannii, and S. maltophilia, are often resistant to a large number of antibacterial treatments and also differ in their pathogenic potential and transmissibility [84].

The widespread use of antibacterials has led to new mechanisms of resistance to develop [85]. Bacteria have adjusted by producing new types of β-lactamases, which can cleave the otherwise resistant carbapenems [86]. Additional resistance-causing mutations can modify and/or downregulate the cell wall proteins, porin channels and other molecules that the antibacterials use to enter and kill the bacteria. Some bacteria have acquired the ability to up-regulate efflux pumps to eliminate the antibacterial faster. Figure 6 illustrates the main mechanisms of beta lactam bacterial resistance.

Additionally, bacteria have developed the ability to transfer resistant genes not only vertically, but also horizontally to other members of their own or even different species [87] [88, 89] [90], thus became a global problem of interspecies transmission.

Carbapenems, due to their potent efficacy, broad-spectrum activity, and relative resistance to hydrolysis by the majority of β-lactamases, are usually reserved for use when other options have failed [28].

Figure 6: Mechanisms of beta lactam bacterial resistance

Β-lactamases and carbapenemases are the most common cause of resistance to beta- lactam antibacterials (e.g. penicillin) in GN bacteria [75, 91]. Β-lactamases hydrolyse the beta-lactam ring of beta-lactam antibacterials and render them inactive [92, 93]. They can be classified into four molecular classes (A-D). Carbapenemases, a subset of β- lactamases that hydrolyse carbapenems as well as almost all beta-lactam antibacterials, include enzymes from classes A, B and D [86, 94]. Class A carbapenemases, the most common carbapenemases primarily identified in Enterobacterales can hydrolyse carbapenems as well as cephalosporins, penicillins, and aztreonam [86, 91, 95]. Class B carbapenemases, found in K. pneumoniae and A. baumannii, usually exhibit resistance to penicillins, cephalosporins, carbapenems, and the available β-lactamase inhibitors [91, 96]. Β-lactamases from class D, also known as OXA β-lactamases, can confer resistance to penicillins, cephalosporins, extended-spectrum cephalosporins, and carbapenems (OXA-type carbapenemases) and are poorly inhibited by currently available β-lactamase inhibitors. These enzymes are expressed in A. baumannii and P. aeruginosa [91, 95, 97]. While resistance based on carbapenemases is mostly acquired, it can be intrinsic in some species, such as S. maltophilia [98].

Changes to porin channels, reducing the permeability of the outer membrane, is a common mechanism of intrinsic antibacterial resistance in GN bacteria [99]. These changes include both reducing the number of porin channels and altering their conformation resulting in a reduced ability of antimicrobial agents to cross the outer membrane and reach the intracellular antibacterial target [75]. In Enterobacterales, P. aeruginosa, and A. baumannii, reductions in porin expression significantly contribute to resistance to carbapenems and cephalosporins [99].

Another common mechanism of resistance is overexpression of efflux pumps. Efflux pumps actively pump antibacterials from the bacterial cytoplasm, inhibiting their function within the cell [99]. When overexpressed, efflux pumps can confer high levels of resistance to previously clinically useful antibacterials such as carbapenems [100]. Efflux pumps are a common mechanism of resistance in non-fermenting bacteria such as P. aeruginosa, and A. baumannii [101].

All these mechanisms can co-exist in the same organism.

Furthermore, while use of colistin was largely abandoned due to the high rates of renal toxicity, in recent years, the increasing emergence of CR GN bacteria has led to its clinical renaissance [102], given its wider spectrum of activity and lack of effective alternatives. However, resistance to colistin is rapidly increasing. In Europe for instance, 28% of CR K. pneumoniae have been identified as resistant to colistin [103]. Also, reports of resistance to the recently approved treatments such as ceftolozane/tazobactam and ceftazidime/avibactam have raised concerns [42, 43].

Table 5 shows the limited in vitro efficacy of several antibacterials against different GN bacteria (Enterobactereacea, Pseudomonas aeruginosa, Acinetobacter baumannii, Stenotrophomonas maltophilia), particularly for CR non-fermenters and pathogens, including serine carbapenemases. ([104]). Stenotrophomonas maltophilia is intrinsically resistant against carbapenems [98, 105].

Table 5: In vitro activity profile of antibacterials for GN Infections with limited treatment options

CP: carbapenemase; CR: carbapenem resistant. OXA-48, KPC, MBL, AmpC: types of carbapenemases. Green – Activity reported, yellow – undetermined activity reported, and red – no clinically relevant activity reported. Source: Adapted from Thalhammer F. (2018). [63]

2.1.3 Overview of infection sites

Aerobic GN pathogens are the most common causes of nosocomial infections and are most commonly seen pneumonia, BSI and UTI, which together represent over 50% of the HAI in Europe.

Figure 7: Hospital-acquired infections in acute care hospitals (EU/EEA 2011-2012)

HAI, hospital-acquired infection Source: ECDC, 2012 [10]

2.1.3.1 Nosocomial Pneumonia

According to the European Centre for Disease Prevention and Control (ECDC) prevalence survey database, pneumonia is the most common infection site accounting for 23% of HAIs [10], of which 67.6% are caused by GN pathogens, and can be defined as HAP, VAP or healthcare-associated pneumonia (HCAP) (Figure 7) [106]. The main causal pathogens of HAP/VAP include Pseudomonas aeruginosa, Acinetobacter spp. and Enterobacterales [11]. Patients with HAP/VAP are at risk of experiencing acute respiratory failure and may require mechanical ventilation [107]. In the 2012 ECDC report, intensive care unit (ICU)-acquired pneumonia was reported to be associated with 5,495 deaths annually resulting in an attributable mortality rate of approximately 3.5% [108].

2.1.3.2 Complicated urinary tract infection (cUTI)

According to the ECDC prevalence survey database, UTI accounts for 19% of HAIs, of which 76.7% are caused by GN pathogens [10, 106]. The main causal pathogens of cUTI include Escherichia coli, Enterococcus spp., Klebsiella spp., Pseudomonas aeruginosa and Proteus spp. [11]. Patients with cUTI can develop bacteraemia and sepsis in 10% to 30% of cases, with risk of death reaching up to 40% [109, 110].

2.1.3.3 Bloodstream infection and sepsis

Bloodstream infections account for 11% of HAIs, of which 43.8% are caused by Gram- negative pathogens [10, 106]. It is defined as the presence of bacteria in the blood and can be also referred to as bacteraemia. In some cases, it can result in sepsis developing, which is a life-threatening condition mediated by the inflammatory response to infection [27]. The main

causal pathogens of BSI include Staphylococcus spp., Streptococcus pneumonia, Enterobacterales and Enterococcus spp. [11]. In the 2012 ECDC report, ICU-acquired BSI was reported to be associated with 4,505 deaths annually resulting in an attributable mortality rate of approximately 5% ([108].

2.1.3.4 Complicated intra-abdominal infection

Complicated intra-abdominal infections account for 1 to2.8% of CR infections [111] [4, 112]. The percentage of IAI caused by Gram-negative pathogen is 15.9% [106]. It is associated with either abscess formation or peritonitis [113]. cIAI generally extends beyond local viscera into peritoneal or retroperitoneal spaces and are associated with systemic signs and symptoms of illness [114]. The main causal pathogens of cIAI are Enterobacterales, Streptococci and Anaerobes (particularly Bacteroides fragilis) [23].

2.1.4 Risk and prognostic factors for MDR and CR infections

2.1.4.1 Risk factors

Risk factors for CR Gram-negative infections consist of a combination of patient clinical setting/healthcare exposure and patient-level characteristics [115-117] and include risk factors that are common to all nosocomial infections (e.g. long term hospitalisation, invasive procedures, long-term ventilation, or depressed host immune system), and some are more specific to CR infections (e.g. previous colonization or infection with CR pathogen, prior exposure to carbapenems, and recent hospitalisation in a endemic CR infections country, or where there was a recent outbreak). Risk factors can vary by infection site (e.g. ventilation is more frequently reported in pneumonia). [85, 118-122].

A summary of the most commonly reported risk factors according to different pathogens is included as an appendix [123] (see Table 6.1:Most commonly reported risk factors per pathogen).

2.1.4.2 Prognostic factors

Time to effective therapy impacts patient’s overall outcomes. Delays in the determination of the pathogen identity and AST results frequently lead to inadequate initial treatment, which causes increased morbidity and mortality. The impact of treatment delay of appropriate treatment was analysed in an SLR [124, 125] reviewing 145 studies and considering three types of outcome comparisons: delay vs. no delay in receiving appropriate therapy, duration of delay of appropriate therapy, and appropriate vs. inappropriate initial therapy. A delay in patients receiving appropriate effective treatment was shown to lead to worse patient

outcomes, including higher mortality rates. Early treatment with appropriate initial therapy represents an important prognostic factor in the treatment of patients with GN infections with limited treatment options. This is further detailed in this section

2. Present an estimate of prevalence and/or incidence for the disease or health condition including recent trends.

2.1.5 Epidemiology

The rate of infections caused by multidrug-resistant (MDR) bacteria continues to increase and limit the utility of existing antibacterial agents. In its surveillance report (2018), European Centre for Disease Prevention and Control (ECDC) reported an increase in resistance to currently available treatments across some Gram-negative pathogens between 2015 and 2018 [126]. ECDC estimate that nearly 700,000 infections and 33,000 deaths in the EU and European Economic Area (EEA) in 2015 are a consequence of MDR bacterial infection [4]. Carbapenem-resistance (CR) in Pseudomonas aeruginosa, Klebsiella pneumoniae and Acinetobacter spp. contributed significantly to the number of estimated deaths (in total approximately 9,000 deaths).

Reports on CR isolates are highly heterogeneous across the globe (Figure 8), but the prevalence of carbapenem resistance has been found to be particularly high in Mediterranean countries, South America and Asia-Pacific countries, with the exception of Japan [127, 128].

Figure 8: Worldwide carbapenem resistance

Source: CDC 2013[80]; ECDC 2017[79]; Mendes et al.[129]; Kiratisin et al.[130]

In the EU-5, the number of CR Gram-negative infections has been reported to be 65,592 in 2015, 123,069 in 2018 and 124,630 in 2019 with P. aeruginosa and A. baumannii as the most

frequently diagnosed CR pathogens [4, 79, 111]. Across EU-5 countries, prevalence of CR Gram-negative infections is reported to range between 0.14 per 100,000 in the UK to 3.05 per 100,000 in Italy (Figure 9) [4].

Figure 9: Prevalence of CR Gram-negative infections in the EU-5

3,05 Cases/100,000

1,20 1,07 0,64 0,31 0,14

France Germany Italy Spain UK EU-5 average

Source: Cassini, 2018[4]

Prevalence estimates are available from multiple sources, generated thorugh different methodologies. Furthermore, pathogen resistance is a constantly evolving, and therefore, results may vary significantly with time, and region/country. Also relevant to account is the fact that the epidemiology varies across the different pathogens, and infections sites:

 Non-fermenters P. aeruginosa and Acinetobacter spp. are the most common pathogens. P. aeruginosa was found in 17% to 61% of CR infections and Acinetobacter spp. in 19% to 50%. The second most common CR pathogen is K. pneumoniae (6% to 20% of infections) followed by E.coli (0.1% to 2.8%) [4, 79, 111].

 The most prevalent CR Gram-negative infection site is the respiratory tract with reported ranges from approximately 41% [4] to 57%[111], followed by UTI and BSI/Sepsis (Table 7).

Table 6: Most common CR causal pathogens across available EU-5 data sources

Pathogen % of causal pathogen for CR Gram-negative infections

P. aeruginosa 17%-60.7% A. baumannii 19%-50% K. pneumoniae 6%-20.0% S. maltophilia 1%a E. coli 0.1%-2.8%

A proportion of S. maltophilia that caused HAIs Suetens 2018[106] Sources: ECDC 2018[79]; Cassini et al, 2018[4] and DRG 2017[111]

Table 7: Proportion of CR infection sites in the EU-5

Infection site % of infection sites for CR Gram-negative infection

Respiratory tract 41.3%-57% Urinary tract 17.0%-19.1% Bloodstream 11.2%-21% Abdomen 2.0% Skin/wound 10.7%-12.8% Other 7.8% Sources: Cassini et al, 2018[4] ; DRG 2017[111]

While there appears to be geographical variation in different types of carbapenemases, recent surveillance study reports an overall increase in these enzymes.

While carbapenem resistance affects both non-fermenters and fermenters in all regions, mechanisms of resistance appear to vary geographically [48, 128].

Analyses from SIDERO-CR surveillance studies [131] confirmed the diversity in carbapenemases across Europe, reporting prevalences of carbapenemas producing

Enterobacteriaceae (CRE), P. aeruginosa (CRP), and A. baumannii (CRA) (Figure 10).

Overall there is an increase in the prevalence of isolates with carbapenemases with significant divrsity (Figure 11) [103] and non-carbapenemase mechanisms of resistance are present in a significant proportion of isolates, particularly in E. coli. (Figure 12) [103, 132]

Figure 10: Epidemiology of carbapenemases in EU 5

Source: Shionogi data on file (Data adapted from SIDERO-CR study)[131]

Figure 11: Confirmed carbapenemase-producing Enterobacteriaceae isolates (Public Health England: 2008–17)

Source: ESPAUR, 2019[132]

Figure 12: Distribution of carbapenem resistance mechanisms in Enterobacteriaceae species in the Europe

Source: Nordmann, 2019 [128] 3. Describe the symptoms and burden of the disease or health condition for patients.

Multi Drug Resistant Gram-negative infections primarily occur in vulnerable hospitalized patients. These pateints are often ≥ 50 years of age, severely transplanted patients, possibly in intensive care units (ICU), or undergoing chemotherapy, or patients who have compromised immunogenicity, and generally wuth multiple comorbidities (e.g. heart disease, diabetes or kidney disease) [68, 69].

The clinical burden of bacterial infection has an impact on key outcomes such as longer treatment, extended hospital admission, additional healthcare professional time, healthcare resource use, adverse events, greater disability (morbidity) and increased risk of death

(mortality) [76]. The need to treat patients empirically before pathogen susceptibility has been confirmed means that this initial treatment choice in MDR is often inappropriate and this can have a significant impact on the individual patient due to the negative clinical consequences of a delay on effective treatment [133, 134].

An overview of signs and symptoms of common CR infections by infection site is provided in Table 8.

Table 8: Overview of disease burden according to the infection site

Site of infection Signs and Symptoms2 Mortality

pneumonia Dyspnoea a 5,495 annual number of deaths in Europe due to ICU-acquired Productive cough a pneumonia (2008–12)[108] Fever a Chest pain a Loss of appetite a Attributable mortality rate: ~3.5% cUTI Fever b Can develop bacteraemia and sepsis in 10% to 30% of cases, Increased urinary frequency b with risk of death reaching up to 40%[109, 110] and urgency b Haematuria b Dysuria b Suprapubic/flank pain b BSI Fever c 4,505 Annual number of deaths in Europe due to ICU-acquired Chills c bloodstream infections (2008–12) [108] Tachycardia c Tachypnoea c Potential complications: Attributable mortality rate: ~5% Infective endocarditis d Osteomyelitis d Infectious arthritis d Septic shock/sepsis d sepsis Dyspnoea e A rate of hospital mortality for sepsis: 17%-26% in severe cases Confusion e [135] Tachycardia e Fever/shivering/feeling very cold e Extreme pain e Extrapolation to global estimates: ~ 5.3 million deaths annually Clammy/sweaty skin e from sepsis

cIAI Fever f Severe infections: mortality rate of 30-50% Tachycardia f Tachypnoea f In case of sepsis: mortality rate > 70%g Hypotension f Abdominal pain f Nausea and vomiting f Diarrhea f Abdominal fullness e Obstipation e Sources: a. https://www.blf.org.uk/support-for-you/pneumonia/symptoms ; b. Sabih et al, 2019[136]; c. MedlinePlus - Medical Dictionary[137] d. Hassoun et al, 2017 [138];

Symptoms of MDR (including CR) Gram-negative infections vary according to the infection site, but for the same infection site, are no different than that caused by other serious infections.

2 Symptoms of MDR (including CR) Gram-negative infections do not differ from those of other serious infections.

2.1.6 Mortality

Multidrug resistant infections, including CR, are associated with 1.6 to 5.0 times higher mortality risk compared non-MDR/CR infections [21, 139, 140]. Mortality rates can reach up to 70% in the most severe cases such as bacteraemia [141]. In Europe, the mortality associated with MDR and CR Gram-negative infections is estimated to be 35% [142-148] .

The extent of the clinical burden of infections with Gram-negative pathogen depends on the severity of infection but generally the burden increases when coinciding with resistant pathogens. The risk of mortality is more than doubled when the cause of an infection is MDR Gram-negative bacilli, in comparison to susceptible organisms [134] For carbapenem- resistant Gram-negative infections, mortality has been estimated to range between 26-44% in one meta-analysis [149], and between 30-75% in another review of studies [150].

Clinical outcomes and burden from Gram-negative bacterial infection can vary depending on the site of infection:

HAP/VAP: Mortality rate estimates in patients with pneumonia ranged from 48.6% to 64.7% [115]. The crude mortality rate associated with VAP has been observed to range from 25% to 76% [151] but mortality directly attributed to VAP could be less than 10% because patients with VAP are already being treated for life-threatening illnesses and may die from the comorbid disease [152-154].

BSI: Hospital-acquired BSI has been associated with substantial morbidity and mortality [155, 156]. According to ECDC, patients with BSIs due to carbapenem-resistant Enterobacteriaceae have mortality rates reaching 50% [157].

In Europe, sepsis caused by the most frequent resistant bacteria is responsible for approximately 25,000 deaths per year, and that two-thirds of these are due to Gram-negative pathogens [158].

UTI: Patients with cUTI can develop, in 10% to 30% of the cases, bacteraemia being associated with a mortality rate ranging between 30% and 40% [110].

The clinical burden of Gram-negative bacilli infections varies depending on the causal pathogen. Infection by Gram-negative pathogens, and specifically MDR Gram-negative pathogens such as E. coli, K. pneumoniae, P. aeruginosa, and Acinetobacter spp. can result in significant clinical burden due to the increase in the length of hospital stay, lack of clinical efficacy, treatment-related adverse events, morbidity and mortality. The reported hospital mortality rates were highest for A. baumannii (23.4 to 50%) and P. aeruginosa (50 to 59.5%),

followed by K. pneumonia (14.4 to 24%) and E. coli (2.5%) [115] [159] [160] [142, 161]. However, quantifiable research at the pathogen level is limited and influenced by global variation in epidemiology, small study sizes and varying definitions of resistance to antimicrobials, leading to difficulties with cross-pathogen comparisons of the pathogen- specific impact of an AMR Gram-negative infection.

Patient factors such as health status and functional status can further contribute to the clinical burden of Gram-negative infection. Mortality associated with CR infections can reach up to 100% in severe cases such as mechanically ventilated patients with bacteraemia [115]. In addition, admission to a hospital with a high prevalence of MDR Gram-negative pathogens and inpatient stay due to invasive procedures (e.g. surgery, ventilators, catheters) increases the risk of infection and thus the risk of poor clinical outcome if the procedure [134].

2.1.7 Quality of Life

There is limited and confounded information available on the impact of infections over the quality of life of these patients, as these are severely ill patients who are frequently treated in ICU units and may be intubated and unconscious, and unable to complete these questionnaires. The quality of life of these patients is also impacted by their underlying disease, and most importantly by the severity of the infection and the infection site (i.e. patients with BSI and sepsis are expected to have lower quality of life compared to a patient with cUTI). The fact that these patients are hospitalised already has detrimental impact on their quality of life. The ward in the hospital also impacts the patient’s quality of life (i.e. patients on ICU or isolation, are expected to have lower quality of life compared to general ward), although this may be correlated with the severity of the infection and underlying condition. All these factors make investigating quality of life in antimicrobial clinical trials difficult and infrequent. However, any therapy that resolves the infection and/or reduces length of hospitalization is expected to improve patient’s quality of life.

2.1.8 Disability Adjusted Life Years (DALYs)

The estimated burden of infections with antibacterial-resistant bacteria in Europe is substantial compared with that of other infectious diseases [4]. A study based on EARS-Net data from 2015 estimated that infections due to antibacterial-resistant bacteria3 accounted for 33,110 attributable deaths and 874,541 DALYs [4]. Infections with colistin-resistant or CR pathogens

The included antibacterial resistance-bacterium combinations were colistin-resistant, carbapenem-resistant, or multidrug-resistant Acinetobacter spp; vancomycin-resistant Enterococcus faecalis and Enterococcus faecium; colistin-resistant, carbapenem-resistant, or third- generation cephalosporin-resistant Escherichia coli; colistin-resistant, carbapenem-resistant, or third-generation cephalosporin-resistant Klebsiella pneumoniae; colistin-resistant, carbapenem-resistant, or multidrug-resistant Pseudomonas aeruginosa; meticillin-resistant Staphylococcus aureus (MRSA); and penicillin-resistant and macrolide-resistant Streptococcus pneumoniae

accounted for 38.7% of the total DALYs. The highest burden in terms of lost DALYs and deaths was noted in Italy and Greece.

The burden due to DALYs associated with antibacterial-resistant bacteria including CR and colistin-resistant infections is reported to have increased between 2007 and 2015. The proportion of the DALYs due to all CR infections increased from 18% in 2007 to 28% in 2015. With regards to specific pathogens, the proportion of the DALYs due to CR K. pneumoniae and CR E. coli doubled from 4.3% in 2007 to 8.79% in 2015.

In terms of infection sites, the highest DALYs burden was associated with BSI reaching up to 71,201 DALYs, and with respiratory infections, reaching up to 19,132 DALYs. The main CR pathogen contributing to DALY was P. aeruginosa except in Italy, where the most burdensome pathogen was K. pneumoniae. The annual number of DALYs attributable to P. aeruginosa ranged from 1,576 to 34,717. In Italy, CR K. pneumoniae was associated with 37,394 DALYs.

2.1.9 Delayed effective therapy

Given that conventional pathogen identification and AST results can take up to 3 days to provide a diagnostic result, the current treatment approach for patients with bacterial infections suspected to be caused by an MDR pathogen, involves initial administration of empiric therapy with wider-spectrum of activity antimicrobial followed by de-escalation to targeted therapy when AST results are available [13, 14]. However, in many instances, the antibiogram is not retrieved. The Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals 2011–2012 indicated that between 40.2% and 80.5% of HAIs are documented with microbiological results [11]. The percentage of pathogens with known AST results is reported to vary between 47.4% and 100% [11].

Increasing antibacterial resistance has made the empiric antibacterial selection more difficult particularly as fewer appropriate treatments for resistant pathogens are available [162]. As a result, many patients with severe bacterial infections receive inappropriate therapy and consequently experience delays in receiving appropriate effective therapy. As the severity of infection increases, patients are more likely to be cycled through a number of inappropriate therapies in the attempt to successfully treat the infection. According to two recent systematic literature reviews ((1) 2015, n=27 and (2) 2019, n=122), patients receiving inappropriate empiric treatment were reported to have a higher mortality risk [163, 164].

 A systematic literature review including studies on the incidence and outcome of inappropriate in-hospital empiric antibacterials for severe infections published

between 2004 and 2014 reported that the percentage of inappropriate empiric antibacterial treatment ranges between 14.1% and 78.9% [163].

 A retrospective cohort study including 40,137 patients with Enterobactereacea in UTI, pneumonia or sepsis reported that patients with CR Enterobactereacea (CRE) were three times more likely to receive inappropriate empiric treatment (IET) than non-CRE (46.5% vs. 11.8%, p < 0.001) [165].

 A systematic literature review (2007-2018, n=37) assessing the impact of delay in appropriate antibacterial therapy for patients with severe bacterial infections treated in hospital settings concluded that approximately 27% of patients experience delays [166].

A delay in effective treatment of an infection may lead to sepsis, a life-threatening condition, irrespective of the initial infection site. A range of studies have confirmed that inappropriately treated patients had 5-times higher mortality risk, twice longer hospital stays and increased risk of readmission, compared to patients receiving appropriate initial therapy. Moreover, patients who fail initial therapies and reach last resort antibacterials are exposed to additional burden associated with severe adverse events and toxicity [167].

In a more recent (2019) systematic literature review, Bassetti et al reported significantly lower mortality rates in patients with appropriate therapy compared to those with inappropriate therapy (OR 0.44 [95% CI, 0.39–0.50]) and these findings were consistent across all time points (Figure 13) [164]. In a pooled subgroup analysis, mortality rates were significantly lower in patients with bacteraemia, sepsis and septic shock in patients with pneumonia who had received appropriate therapy compared to those having inappropriate treatment [164]. This burden increases with resistant pathogens, whereby patients with CR P. aeruginosa infections who receive initial inappropriate treatment have mortality risk that is twice as high as that seen in susceptible patients (27.3% vs 13.8% respectively) [15].

In another recent systematic literature review of 37 studies by Zasowki et al. (2019), patients receiving initial appropriate therapy had significantly lower mortality rates (OR 0.57, 95% CI: 0.45–0.72]) in comparison to those receiving initial inappropriate treatment and a consequent delay in effective treatment (Figure 14) [166]. These findings were consistent across all time points. Published literature reports that patient prognosis worsens with each day and hour of delay in appropriate treatment. A retrospective cohort study including 480 patients with BSIs due to carbapenemase-producing Enterobacteriaceae (CPE) reported an increase in mortality risk with each day of delay (HR=1.02; (95% CI: 1.01, 1.04; p < .0001) [168].

Figure 13: Summary of effect of appropriate versus inappropriate initial antibacterial therapy on mortality

Source: Bassetti, 2019[164]

Figure 14: Summary of effect of delay versus no delay in receiving initially appropriate antibacterials on mortality

Source: Zasowski, 2019[166]

Inappropriate antibacterial therapy is associated with higher rates of treatment failure. Bassetti et al assessed the impact of appropriate versus inappropriate initial antibacterial therapy on the treatment failure. The findings suggest that patients receiving appropriate had a significantly lower incidence of treatment failure compared to patients with inappropriate therapy (OR 0.33; 95% CI: 0.16, 0.66) (Figure 15) [164]. These findings were consistent

across all time points and subgroup of patients with UTIs or acute pyelonephritis and with bacteremia or sepsis (Figure 15) [164].

Figure 15: Summary of effect of appropriate versus inappropriate therapy on treatment failure

Source: Bassetti, 2019[164]

2.2 Target population

1. Describe the target population and the proposed position of the target population in the patient pathway of care.

The indication for cefiderocol is expected to be:  Fetcroja is indicated for the treatment of infections due to aerobic Gram-negative organisms in adults with limited treatment options. Limited treatment options can be pragmatically translated into infections by MDR (including CR) pathogens.

This indication will therefore be pathogen focused, not restricted to any specific site of infection and predicts 2 different populations:

 Hospitalised critically ill patients with suspected (but prior AST results availability) MDR/CR infection where effective treatment should be administered as soon as possible (followed by de-escalation to a more targeted treatment when the pathogen and susceptibility profile is subsequently confirmed), resistance to many antimicrobial classes almost invariably reduces the probability of adequate empirical coverage, with possible unfavourable consequences. In this light, readily

available patient’s medical history and updated information about the local microbiological epidemiology remain critical for defining the baseline risk of MDR- GNB infections and firmly guiding empirical treatment choices, with the aim of avoiding both undertreatment and overtreatment ([13, 14] and Clinical guidelines overview). Given its wide Gram-negative spectrum of activity, and safety profile, cefiderocol would be an appropriate treatment choice for these patients.

 Hospitalised patients where CR infection has been confirmed the selection of treatment is predominantly based on AST results regarding pathogen, its mechanism of resistance, and susceptibility results for the different antibacterials tested. Based on its in vitro data, cefiderocol would be an appropriate option, in aerobic Gram-negative pathogens, particularly non-fermenters such as P. aeruginosa, A. baumannii and S. maltophilia, and presence of metallo-β- lactamases, where there is limited in vitro activity from newer regimens, and other treatment choices may be inappropriate due to safety and tolerability concerns.

The treatment of these patients will require an expert and complex clinical reasoning, taking into account the peculiar characteristics of the target population, but also the need for adequate empirical coverage and the more and more specific enzyme-level activity of novel antimicrobials with respect to the different resistance mechanisms of MDR-GNB, resulting to variations in the use of specific treatments even within regions of countries [169]. Thus, treatment decisions differ for patients with suspected or confirmed infection by MDR/CR pathogens.

2. Provide a justification for the proposed positioning of the technology and the definition of the target population.

The intended indication for cefiderocol is for the treatment of aerobic, Gram-negative infections in adults with limited treatment options (i.e. confirmed or suspected MDR infections, including CR infections).

Cefiderocol can be used when the antibacterial susceptibility results have been obtained and show that no other treatment is likely to have an effect against the disease pathogen, particularly in non-fermenters and Acinetobacter baumanii, Stenotrophomonas maltophilia and Pseudomonas aeruginosa, as well as in the metallo-carbapenemases in Enterobacteriaceae, as per Table 9 below. It can also be used earlier, as a pre-emptive treatment to ensure appropriate antibacterial coverage as early as possible given its wider spectrum of activity as per image below in Gram-negative pathogens.

Table 9: In Vitro Gram-negative activity profiles

Predicted clinical activity based on CSLI breakpoints; *Color-coding based on the pathogen susceptibility Source: Thalhammer F, 2018 [63], Theuretzbacher, 2019 [64]

The target population thus comprises two groups:

Critically ill, adult patients with highly suspected infection by a carbapenem-resistant or other MDR Gram-negative pathogen.

 These patients often require immediate treatment. Initiation of treatment for these patients cannot be deferred until antibiogram is available.

 Knowledge of local pathogen epidemiology and patient-specific factors can support initial antibacterial treatment decisions.

 Aligned with stewardship recommendations, when results from susceptibility tests are available, de-escalation to other treatments should occur whenever possible to avoid undertreatment and overtreatment.

 For these patients, cefiderocol has demonstrated broad efficacy according to current evidentiary standards for antimicrobials (in vitro, PK/PD and clinical data, see sections 5.3-5.5.). They include patients with common infections such as cUTI, pneumonia, blood infection/sepsis, and IAI.

 In line with good antimicrobial stewardship, cefiderocol should be regarded as a first treatment choice in this context, replacing current treatment attempts with carbapenems in high dose and/or in combinations and where there is high suspicion of susceptibility to cefiderocol.

 Placing cefiderocol in the management of hospitalised patients with suspected difficult to treat GNI as first treatment choice, replacing current treatment attempts

with carbapenems in high dose and/or in combinations, or recently approved medicines, followed by de-escalation to reduce the risk of development of resistance [170]. This strategy reduces time to effective treatment and is associated with lower mortality as well as LOS in patients with severe sepsis and septic shock Fout! Verwijzingsbron niet gevonden.[171, 172]. In line with good antimicrobial stewardship, cefiderocol should be regarded as an early, targeted treatment based on advanced risk determination methods (Figure 16) [124].

Figure 16 - Treatment of patients with highly suspected infection by CR or other MDR GN pathogens

Adult patients with confirmed infection by a carbapenem-resistant Gram-negative pathogen or multidrug resistant Gram-negative pathogen including carbapenem-resistant Enterobacteriaceae (CRE) such as K. pneumoniae and E. coli, and non-fermenters such as A. baumannii, P. aeruginosa, and S. maltophilia.

 For these patients, cefiderocol has demonstrated broad efficacy according to current evidentiary standards for antimicrobials (in vitro, PK/PD and clinical data, see section 5.4). They include patients with common infections such as cUTI, pneumonia, blood infection/sepsis, and IAI. (Figure 17)

 In line with good antimicrobial stewardship, cefiderocol should be regarded as a first treatment choice in this context, replacing current treatment attempts with colistin in combination with several other antimicrobials from different classes, which carry a very substantial side effect burden (see chapter 2.1).

Figure 17: Treatment of patients with confirmed infection by carbapenem-resistant or other MDR Gram-negative pathogen

3. Estimate the size of the target population. Include a description of how the size of the target population was obtained and whether it is likely to increase or reduce over time.

As outlined in section 2.1 MDR/CR, resistances are increasing with estimates of deaths related to serious infections being updated frequently. The European Centre for Disease Prevention and Control (ECDC) estimate that nearly 700,000 infections and 33,000 deaths in the EU and European Economic Area (EEA) in 2015 are a consequence of MDR bacterial infection [4] WHO the predicted annual deaths by AMR is expected to rise from 700,000 cases in 2014 to 10,000,000 in 2050 (WHO, 2014, Review on AMR [78]. The prevalence of resistance to last- resort antibacterials, particularly with regards to carbapenems and colistin, has been increasing globally. An alarming spread of CR Gram-negative infections through healthcare facilities has been reported and is expected to transfer to the community [8, 76]. Even for the more recently approved antibacterials such as ceftolozane/tazobactam and ceftazidime/avibactam, there have been reported cases of resistance [40-43]. This is also expected to expand to the community, similarly to what has been previously observed with ESBL-producing pathogens, via environment and traveling [76, 173]. For example, in 2018, Sweden and Norway reported a cluster of returning travellers who carried or were infected with carbapenemase (OXA-48)-producing K. pneumoniae that were associated with hospital admissions in Gran Canaria [173].

The size target population for cefiderocol in Europe is difficult to estimate, as incidence strongly depends on:

 epidemiology data (local resistance profile and local pathogen distribution, which is constantly evolving),  potentially regional outbreaks may occur changing substantially the patient numbers,  patient population characteristics and risk factors (e.g. travels to CR endemic countries),  introduction of new medicines with overlapping activity profile that will change the unmet need,  how data is reported: there is a wealth of epidemiology information available, but using different methodologies, increasing the uncertainty of the actual numbers of MDR or CR infections,  also, the 2 different target populations are interlinked and self-exclusive.

Cassini, et al, [4] estimates that there are 107,801 Carbapenem resistant infections in Europe, not account for those caused by Stenotrophomonas maltophilia, and as mentioned in the previous section, this is likely to increase in the future [4]. Only a Dynamic infection disease modeling considering the mentioned factors, upcoming treatments and future trends could provide plausible predictions [123], but still with significant degree of uncertainty.

2.3 Clinical management of the disease or health condition

1. Describe the clinical pathway of care for different stages and /or subtypes of the disease being considered in the assessment.

With the emergence of antibacterial resistance, antimicrobial stewardship programs have been put in place in various healthcare settings to attempt achieve as rapid as possible, identification of pathogens causing bacterial infections and the most appropriate treatment. Treatment is determined by clinical state and local epidemiology to minimise the chance of ineffective therapy. The current treatment approach for patients with bacterial infections when there is suspicion of MDR pathogen, involves initial administration of empiric therapy with wide-spectrum antimicrobial (or combination of antibcterials) followed by de-escalation to targeted antimicrobials antibacterials when the antibiogram is available (i.e., identification of the underlying pathogen and susceptibility testing (Figure 18)) [13].

Figure 18: Current treatment approach for bacterial infections

Treatment selection is based on information on pathogen identification as well as susceptibility and mechanism of resistance. According to proportions of main pathogens in the infection and in vitro susceptibility of potential treatments the treatment with the highest predicted treatment success is being selected [13].

Resistance to many antimicrobial classes in MDR pathogens almost invariably reduces the probability of adequate empirical coverage, with possible unfavorable consequences. Timely administration of antibacterials is vital to improve patient’s outcomes [14] and in line with antimicrobial stewardship, includes treatment with the most appropriate drug regimen [174, 175]. While microbiological testing is carried out (this can take up to 3 days), early clinical decisions are based on environmental and patient factors including clinical state and local epidemiology to minimize the chance of ineffective therapy. Empirical treatment with wide- spectrum antibacterials is usually administered to severely ill patients when a quick treatment decision is required [13]. However, this creates pressure for the selection or development of resistant organisms over time [13, 176].

If the patient is confirmed to have multidrug resistant (MDR) including carbapenem resistant (CR) infection, treatment is selected based on the susceptibility results. [170]. Good antibacterial stewardship mandates more restrictive use of antibacterials and regardless which treatment is being used, guidelines always urge to de-escalate the treatment whenever possible [170] (see below).

Clinical reasoning for the treatment of suspected MDR-GNB infections in critically ill patients aims to reduce time to effective therapy [124]. Current standard practice for this population is not well defined, and highly variable across different geographies and infection sites. Traditionally however, carbapenems in higher dose regimens tha toptimizes exposure, and/or combination with other antibacterials, have been used but with limited success in resistant pathogens [177]. Recently approved antibacterials such as ceftazidime/avibactam and ceftolozane/tazobactam, have also been used in this setting, as already proposed by Bassetti et al. [177]

However, there is still no defined standard of care for the treatment of severe MDR (including CR) Gram-negative infections. A recent literature review of current and upcoming therapeutic approaches for severe MDR Gram-negative infections in critically-ill patients reported that the availability of newly approved antibacterials such as ceftolozane/tazobactam, ceftazidime/avibactam, meropenem/vaborbactam, plazomicin and eravacycline, have addressed some challenges due to antimicrobial resistance [177]. However, these treatment options are reported to have suboptimal activity against some pathogens especially against CR A. baumannii and against carbapenem-resistant Enterobacteriaceae (CRE) of novel beta- lactam/β-lactamase inhibitors is dependent of the type of carbapenemase conferring resistance to carbapenems [177]. The existing therapies for MDR including CR infections include newly approved beta-lactam/β-lactamase inhibitor combinations such as ceftolozane/tazobactam, ceftazidime/avibactam and meropenem/vaborbactam, novel aminoglycoside plazomicin and a novel fluorocycline eravacycline. Other treatments include polymyxins (polymyxin E [colistin] and polymyxin B), glycylcyclines (e.g., tigecycline), and aminoglycosides [177]. Figure 19 gives an overview of current clinical reasoning for the treatment of serious MDR-GNB infections in critically-ill patients.

Figure 19: Current clinical reasoning for the treatment of serious MDR Gram-negative infections

• Ceftazidime/avibactam (as preferred empiricalchoice when both KPC and OXA carbapenemases are reported locally) ormeropenem/vaborbactam • Although in the lack of high-level evidence, forboth empirical and targeted treatmenta combination with old (collistin, polymyxin B, tigecycline, old aminoglycosides, fosformycin) ornovel agents (plazomicin, eravacycline, double BL- CRE BLI combinations) could be considered in the attempt of delayig emergence of restistance, after having carefully balanced potentional additional toxicity on a case- by-case basis (expert opinion) • In case of resistance to novel BL-BLI, considerpolymyxins-based oraminoglycosides- based combination with carbepenems and/or(tigecyclineoreravacycline) and/or fosformycin • Considerconcomitant adminitration of inhaled polymyxins/aminoglycosides when they are used intravenously forVAP

• Ceftolozane/tazobactam (as preferred empirical choice in absence of concomitant risk of CRE) or ceftazidime/avibactam • For empirical therapy, administera second anti-pseudomonal agent (an aminoglycosideora polymyxin orfosformycin) • Although in the lack of high-level evidence, fortargeted therapy combination with old (collistin, polymyxin B, old aminoglycosides, fosformycin) ornovel agents CRPA (plazomicin, double BL-BLI combinations) could be considered in the attempt of delayingemergence of restistance, after having carefully balanced potential additional toxicity on a case-by-case basis (expert opinion) • In case of restistance to novel BL-BLI, considerpolymyxins-based oraminoglycosides- based combinations with carbapernems and/or fosformycin and/orrifampin • Considerconcomitant administration of inhaled polymyxins/aminoglycosides when they are used intravenously forVAP

• Administera polymyxin as the backbone agent • Considercombination with old (carbapenems, old aminoglycosides, tigecycline, CRAB fosformycin, rifampin) ornovel agents (plazomicin, eravacyclin) • Considerconcomitant administration of inhaled polymyxins/aminoglycosides when they are used intravenously forVAP

DR-GNB, Multi-drug resistant Gram-negative bacteria; CRE, carbapenem-resistant Enterobacterales; CRPA, carbapenem- resistant Pseudomonas aeruginosa; CRAB, carbapenem-resistant Acinetobacter baumannii; BL-BLI, b-lactam/b-lactamase inhibitors; VAP, ventilator-associated pneumonia. Source: Bassetti, 2019[177]

2.3.1 Key information on currently available treatments in Europe

As outlined before, treatment options for infections with MDR/CR aerobic Gram-negative pathogens are very limited. Susceptibility tests have shown that to date broad coverage, including pathogens affecting patients with limited treatment options (such as CR A. baumannii, CR P. aeruginosa, S. maltophilia, and CR Enterobacteriaceae) is only achieved by cefiderocol [29, 30]. In a recent analysis of the global clinical antibacterial pipeline by WHO, cefiderocol was reported to be the only antibacterial providing coverage against all three critical priority pathogens: CR A baumannii, CR P aeruginosa, and CR Enterobacteriaceae (Figure 7) [64].

Ceftazidime/avibactam is a recently approved combination of a well-known beta-lactam with a novel β-lactamase inhibitor for cIAI, cUTI, HAP/VAP and aerobic Gram-negative infections in adults with limited treatment options (EU)[178]. It is active against class A (e.g., KPC) and class D (e.g., OXA) carbapenemase-producing CRE and has demonstrated activity against some CR P. aeruginosa isolates [177]. Recent results of in vitro study, SIDERO-WT, reported

a poor activity of ceftazidime/avibactam against CR A. baumannii with minimum inhibitory concentration (MIC50) for meropenem-non-susceptible A. baumannii of 32 mg/L3 [46] and stenotrophomonas maltophilia. Currently widely available on the EU countries, with reimbursement.

Ceftolozane/tazobactam is a novel combination of a beta-lactam antimicrobial with a well known β-lactamase inhibitor, with EMA approval for cIAI and cUTI [179]. It has demonstrated a potent in vitro activity against CR P. aeruginosa isolates; however, without activity against CRE [177]. Tazobactam (β-lactamase inhibitor) protects ceftolozane from degradation by Class A β-lactamase enzymes [179], but has not demonstrated activity against KPC Class A carbapenemases, and Class B (metallo-), or Class D β-lactamases [179]. Currently widely available on the EU countries, with reimbursement.

Meropenem/vaborbactam is a novel combination of a well know carbapenem in a higher dose, and a novel β-lactamase inhibitor approved for cIAI, cUTI, HAP/VAP, and infections due to aerobic Gram-negative organisms in adults with limited treatment options [180]. It has activity against class A (e.g., KPC) carbapenemase-producing CRE. Vaborbactam has limited activity against Class D β-lactamases and no activity against Class B (metallo-) β-lactamases and does not improve the activity of meropenem against CR A. baumannii, P. aeruginosa or S. maltophilia [181]. However, is it not yet reimbursed in most of the European markets. Currently approved by EMA, but not yet reimbursed in many countries and therefore, not widely available on the European countries.

Eravacycline is a novel synthetic fluorocycline that was approved by EMA for the treatment of cIAI [182]. It has demonstrated activity against Gram-negative pathogens including CRE and CR A. baumannii with exception of P. aeruginosa and Burkholderia cepacia [177, 183, 184]. Currently approved by EMA, but not yet reimbursed in many countries and therefore, not widely available on the European countries.

While colistin once was abandoned due to the high rates of renal toxicity in recent years, the increasing emergence of MDR Gram-negative bacteria appears to have led to its reintroduction in clinical practice [102]. Colistin has antibacterial activity against a wide variety of Gram-negative pathogens including E. coli, Klebsiella spp., Enterobacter spp., P. aeruginosa, and Acinetobacter spp. [185]. Some Gram-negative pathogens such as Proteus spp., Providencia spp. and most isolates of Serratia spp. are intrinsically resistant to colistin [185]. While it covers a broad spectrum of Gram-negative pathogens, colistin is associated with severe adverse events [102, 134]. Among the more severe adverse events are neurotoxicity, nephrotoxicity, and ototoxicity [102, 134]. Renal failure is reported to reach up

to 60% in patients treated with colistin[186]. A recent systematic literature review (n=224) including data on 33,573 patients reported that the overall rate of nephrotoxicity in patients treated with polymyxins was 0.277 (95% CI: 0.252, 0.303). Nephrotoxicity rates were found to differ between patients treated with CMS, colistin and PMB (0.260 [95% CI: 0.216, 0.30]), 0.274 [95% CI: 0.239, 0.312] and 0.348 [95% CI: 0.301, 0.397], respectively; p=0.016) [187].

Aminoglycosides have been frequently used for the treatment of CR infections, particularly in case of polymyxin resistance [177]. However, their efficacy is hindered by their impaired safety profile (i.e., nephrotoxicity and ototoxicity) and increasing rates of resistance [177] [188]. While nephrotoxicity often can be reversed, the hearing loss is irreversible [188]. Aminoglycosides have been also associated with neuromuscular blockade [189].

Tigecycline, a glycylcycline antibacterial, is active against CRE and CR A. baumannii [177]. P. aeruginosa is inherently resistant to tigecycline with > 90% of pathogens reported to be resistant to it [177, 190]. Of note, tigecycline is reported to have been associated with increased mortality in comparison with other regimens in patients with VAP [177]. Currently approved by EMA for cIAI and cABSSI.

Relebactam/imipenem/cilastatin has recently been granted positive CHMP opinion for approval in Europe for Gram-negative infections in patients with limited treatment options. It shows activity against resistant strains of P. aeruginosa and K. pneumoniae carbapenemase in CR bacterial infections, but not those of A. baumannii or S. maltophilia (Merck and CID) or metallo-betalactamases. Currently with positive CHMP opinion and not yet widely available on the European countries.

While clinical guidelines on the management of multidrug resistant (MDR) including carbapenem resistant (CR) Gram-negative infections are usually included in infection-site specific treatment guidelines, they have not yet caught up to date with the new EMA regulatory guidance that is focused on pathogens. Therefore, there is a lack of integrated recommendations for the management of these resistant infections looking at pathogens, regardless of infection site, and therefore a lack of well-defined standard of care. A survey from 2017 including >100 hospitals in the US and Europe, reported that almost half of the respondents (54/111, 48.6%) had no guidelines for the treatment of infections caused by CR Gram-negative pathogens [22]. As outlined before the treatment of Gram-negative infections is based on multiple factors, including the underlying condition, infection sites, and local epidemiology, but most importantly taking into account local resistance. This leads to variations in the use of specific treatments even within regions of countries.

The management of carbapenem-resistant Gram-negative infections is particularly challenging due to the paucity of antimicrobials active against these bacteria; various treatment options are used, very often in double or triple combinations with no consensus on the most appropriate treatment strategy [22]. Current treatment options have limited activity and/or have safety concerns. Recently marketed treatment options only partially cover carbapenem resistance, and for example, none can address all 3 critical pathogens identified by the WHO [38, 191]; ceftazidime/avibactam covers K. pneumoniae carbapenemase (KPC)- producing carbapenem-resistant Enterobacteriaceae (CRE), ceftolozane/tazobactam covers P. aeruginosa carbapenem-resistant strains and tigecycline, rather used as combination therapy, covers carbapenem-resistant Enterobacteriaceae and A. baumannii [192, 193]. Polymyxins provide additional treatment options; recently re-introduced as a last alternative due to the increasing CR resistance, they provide broader coverage for carbapenem-resistant Gram-negative infections [33], but polymyxin resistance is already prevalent both in North America and Europe [102] and these agents have serious side effects, such as nephrotoxicity and neurotoxicity [194, 195]. Trimethoprim/sulfamethoxazole is considered as the treatment of choice for S. maltophilia infections, though limited by toxicities [196].

In summary, available treatments consist of last resource drugs and multiple antibacterial combinations, often with limited pathogen and/or mechanism of resistance coverage, and/or with significant safety/tolerability concerns (e.g. colistin, tigecycline) [23-27].

2. Clinical guidelines overview

Search approach

A targeted search for relevant guidelines and review articles on the treatment of Gram- negative bacteria was conducted in December 2019 and complemented by Internet searches for national guidelines from Denmark, England (UK), France, Germany, Italy, Norway, Spain and Sweden. Focus of selection and analysis were recommendations for treatment of infections with MDR Gram-negative bacteria as well as for CR infections in hospital settings [197]. Key results are provided in the tables below.

Summary of findings

Recommendations refer either to most common infection sites (e.g. pneumonia, sepsis, IAI, cUTI) or, more recently, to pathogen types, e.g., infections with MDR/CR Gram-negative bacteria (UK: [122] Spain: [198, 199] Italy Klebs:[200]) or more specific, ESBL-producing bacteria or CR bacteria [201].

Fewer guidelines refer to specific treatments [202, 203]. As clinical data are rare, recommendations are often based on evidence derived from in vitro susceptibility testing results, case studies, observational studies, and expert opinion, which is standard and the most appropriate approach in antimicrobial treatment decisions. In general, all guidelines recommend that treatment should be started early for suspected infections (within hours), support the antibacterial de-escalation strategy and recommend that empirical antibacterial therapy should be implemented in accordance with local microbiological data and previous treatment.

2.3.2 Site-specific vs. pathogen-specific guidelines

While infection-site-specific guidelines have been issued regularly for many years, recommendations for treatment of MDR/CR Gram-negative infections dependent on the type of resistance and pathogen with reference to the infections site have been developed at an increasing rate in recent years by International/European/National Societies [23-25, 122, 201, 204].

2.3.3 Specific recommendations

Mono- and combination therapies of carbapenems (e.g. meropenem, imipenem, ertapenem), or in combination with polymixins (colistin), colistin in combination with tigecycline, or newer treatments such as ceftolozane/tazobactam prevail. Accordingly for CR Gram-negative infections, guideline recommendations include combination treatment of colistin with meropenem or with tigecycline, ceftazidime/avibactam, high dose tigecycline, fosfomycin and colistin [23-25, 27, 122, 201].

An overview of recent guideline recommendations for MDR/CR Gram-negative infections as well as for respective infection sites is provided in Table 10a-Table 11e.

The guidelines identified for MDR/CR Gram-negative infections are aligned in terms of recommendations on the first line empiric therapies, use of mono- or combination therapies and recommended antibacterial class(es) across indications, but may differ in use of specific antibacterials within the same class. There was no evidence of incompatible recommendation (e.g., one country recommending a treatment that another country specifically excludes).

2.3.4 Specific considerations of CR infections

International consensus guidelines recommend that patients with CR Gram-negative infections including Enterobacteriaceae (CRE), A. baumannii (CRAB) and P. aeruginosa (CRPA) are managed with polymyxin B or colistin in combination with one or more additional

agents to which the pathogen is susceptible. If additional susceptible agents are unavailable, polymyxin B or colistin should be used in combination with a non-susceptible agent (e.g., a carbapenem) in patients with CRE and CRPA, and in monotherapy in patients with CRAB [24]. Recent national guideline recommend targeted combination therapies according to type of carbapenemases [122, 200, 205] and include newer agents ceftazidime/avibactamand ceftolozane-tazobactam. Therapeutic approaches with infusion of high-dose antimicrobials is considered for meropenem [201, 205] and ceftolozane/tazobactam [204] in high risk patients.

Table 10a: Relevant guidelines for diagnosis and management – MDR/GN Bacteria

Name of society/organisation issuing guidelines Date of Country (s) Summary of Summary of issue or to which recommendations recommendations last update guideline -MDR infections- -CR infections- applies AWMF (Registry Nr. 092/001 (S3 guideline: 2018 Germany Reports on measures to NR Strategies to ensure rational use of antibacterials in prevent or reduce resistant hospitals) germs in hospitals. [206] 1 Paul-Ehrlich-Gesellschaft für Chemotherapie e.V. 2019 Germany Pneumonia/Sepsis: CR Enterobacteriaceae (PEG) • meropenem plus colistin Pneumonia/sepsis: AWMF 082-006 • colistin plus tigecycline • meropenem plus colistin Chapter 16. Infections with MDR gram neg sticks (potentially plus fosfomycin) • colistin plus tigecycline (Stäbchen) – ESBL-producing bacteria, (potentially plus fosfomycin); Carbapenemase-producing Enterobact083eriaceae, Carbapenem-resistant Acinetobacter baumannii CR Acinetobacter-baumannii [201]2 Pneumonia/sepsis/wound/cUTI: colistin, sulbactam, high dose tigecycline; (cotrimoxazole only for cUTI)

Infectious Diseases Working (2015) Stenotrophomonas maltophilia Pseudomonas aeruginosa Party (AGIHO) of the German Society of (Publication (Europe) pneumonia: pneumonia: Haematology and year) • High dose trimethoprim– • Combination therapy Medical Oncology (DGHO) sulfamethoxazole; Piperacillin (±tazobactam),

[207] • tigecycline-based therapy ceftazidime, imipenem/ cilastatin, meropenem and cefepime Gruppo Italiano Trapianto Midollo Osseo (GITMO), 2015 Italy NR CRKp-targeted antibacterial Associazione Microbiologi Clinici Italiani (AMCLI), therapy: combination therapy Società Italiana Malattie Infettive e Tropicali (SIMIT), including at least two among and the Centro Nazionale Trapiantt (CNT) colistin/polymyxin B, [200] tigecycline and gentamicin; addition of meropenem, and eventually fosfomycin preferred Helsedirektoratet. Antibiotikabruk i sykehus, 2018 Norway Resistant to 3rd generation Resistant to 3rd generation Nasjonal faglige retningslinje. cephalosporins: cephalosporins [208] 3 • Meropenem • Colistin

• Imipenem / cilastatin • Tigecycline

• Doripenem or Ertapenem [Not for Pseudomonas or Acinetobacter]

Infectious Diseases and Clinical Microbiology 2015 Spain Fosfomycin (as part of a NR (SEIMC) 2015 combination regimen including [199] at least one more active agent)

Spanish Society of Chemotherapy (2018) Spain • Colistin, NR [198] • Amikacin

“Antibacterial selection in the treatment of acute (Publication • Ceftolozane/ tazobactam invasive infections by Pseudomonas aeruginosa: year) • High doses of β-lactam Guidelines by the Spanish Society of Chemotherapy” antibacterials (Review article) Spanish Society of Transplantation (SET)/ Group for 2018 Spain • Amoxicillin-clavulanic acid NR Study of Infection in Transplantation of the Spanish • Piperacillin-tazobactam Society of Infectious Diseases and Clinical • Meropenem Microbiology (GESITRA-EIMC)/ Spanish Network for • Aztreonam Research in Infectious Diseases (REIPI) • Tigecycline [204] • Fosfomycin • Ceftazidime-avibactam • Ceftolozane-tazobactam • Colistin Spanish Society of Infectious Diseases and Clinical 2019 Spain (ESBL)-producing KPC-producing Microbiology (SEIMC) and the Spanish Association Enterobacteriaceae: Enterobacteriaceae: of Haematology and Hemotherapy (SEHH) • at least two active drugs from • beta-lactam/β-lactamase [205] 4 the options included in the inhibitor (BLBLI) antibiogram (meropenem, • Piperacillin-tazobactam colistin, tigecycline, and meropenem (extended fosfomycin and infusion) aminoglycosides)

AmpC-producing strains with meropenem MICs < Enterobacteriaceae 16 mg/L: • Cefepime and fluoroquinolones

• Piperacillin-tazobactam • combination regimen should include high-dose meropenem (extended infusion)

KPC-producing or OXA-48- producing Enterobacteriaceae: • Ceftazidime-avibactam

Extensively drug-resistant (XDR) and pandrug-resistant (PDR): • single-agent treatment, prioritizing the use of (in following order) beta-lactams, sulbactam (in infections due to A. baumannii) and colistin.

XDR or PDR P. Aeruginosa infections: • Ceftolozane-tazobactam or ceftazidime-avibactam

ESBL-producing intestinal bacteria Knowledge base 2014 Sweden • Cephalosporins and Cephalosporins should not be with proposals for management to limit the spread of quinolones reduced by increasing the Enterobacteriaceae with ESBL consumption of carbapenems,

[209] 5 • Alternative: Piperacillin- as this increases the risk of tazobactam for sepsis, selection of carbapenem- pneumonia, abdominal resistant strains. infections; also, to reduce cephalosporins consumption British Society for Antimicrobial 2018 UK ESBL and AmpC-producing IKPC-carbapenemase: Colistin Chemotherapy/Healthcare Infection Society/British Enterobacteria & meropenem (if unknown/S in Infection Association. Joint Working Party • Meropenem, imipenem or past) [122] ertapenem (Consider tigecycline to colistin or ceftazidime/avibactam to Susceptibility of past/current meropenem) infection not known (inpatient setting): OXA-48: Aztreonam or • Meropenem and imipenem ceftazidime Ceftazidime/ or Meropenem-sparing: avibactam if R or unknown. temocillin (if urinary), ceftolozane/ tazobactam Metallo-carbapenemase: Susceptibility of past/current Fosfomycin and colistin, infection known, along with consider tigecycline, Use co- urinary infection: trimoxazole if • Co-amoxiclav or Stenotrophomonas piperacillin/tazobactam or gentamicin or amikacin 1https://www.awmf.org/uploads/tx_szleitlinien/092-001l_S3_Strategien-zur-Sicherung-rationaler-Antibiotika-Anwendung-im-Krankenhaus_2019-04.pdf 2https://www.awmf.org/uploads/tx_szleitlinien/082-006l_S2k_Parenterale_Antibiotika_2019-08.pdf 3https://www.helsedirektoratet.no/Retningslinjer/Antibiotika-i-sykehus

4https://seimc.org/contenidos/documentoscientificos/seimc-dc-2019-Febrile_Neutropenia.pdf 5https://www.folkhalsomyndigheten.se/contentassets/f4df42e7e643414ba3499a9ee1801915/esbl-producerande-tarmbakterier.pdf

Table 11b: Relevant guidelines for diagnosis and management – HAP/VAP(HCAP)

Name of society/organisation Date of Country (s) Summary of recommendations Summary of issuing guidelines issue or to which -MDR infections- recommendations last guideline -CR infections- update applies ERS/ESICM/ESCMID/ALAT 2017 Europe • No septic shock: single agent (carbapenem, Combination therapy; similar guidelines EU cephalosporin, piperacillin/ tazobactam or approach as in MDR patients [210] fluoroquinolone) (carbapenem-resistant • Severely ill or in septic shock: combination therapy Enterobacteriaceae) (antipseudomonal β-lactam plus a second agent such as an aminoglycoside or an antipseudomonal fluoroquinolone or, in some cases polymyxins) pro. medicin Information til 2020 Denmark No specific recommendation for MDR infections NR sundhedsfaglige - Antibiotikavejledning Pneumonia: Phenoxymethylpenicillin or [211] 1 Clarithromycin HAP: Piperacillin/Tazobactam or Cefuroxime Mycoplasma and Chlamydophila pneumonia: Clarithromycin or Roxithromycin Legionella pneumonia: Clarithromycin or Roxithromycin or Ciprofloxacin Chlamydophila psittaci: Doxycycline

The French Society of 2017 France Late pneumonia ≥5 days or nonfermenting GNB: in NR Anaesthesia and Intensive Care (2018 case of ESBL: Imipenem-cilastatin or meropenem + Medicine and the French publication amikacin or ciprofloxacin Society of Intensive year) HAP (when no other antibacterials can be used): Care nebulised colimycine (sodium colistiméthate) and/or [212] aminoglycosides

Leone et al: Summary of French guidelines for the prevention, diagnosis and treatment of hospital‐acquired pneumonia in ICU (Review article) AWMF (Registry Nr: 020-013) Update Germany At risk of MDR: Colistin in combination with (S3 guideline Epidemiology, 2017 • Piperacillin/Tazobactam Aminoglycosides diagnosis and therapy of adult or or patients with nosocomial • Cefepime Fosfomycin, pneumonia) • Ceftazidime or [213] 2 or Carbapenem • Imipenem/Cilastatin or • Meropenem Ceftazidime/Avibactam (depending on in vitro-Tests and Plus/minus adverse reactions) • Fluorchinolon (Ciprofloxacin, Levofloxacin) or • Aminoglycosides (Gentamicin, Tobramycin, Amikacin)

In suspected MRSA: combination with Glycopeptid or Oxazolidinone, Vancomycin or Linezolid AWMF (Registry Nr: 082-006) 2019 Germany • Nosocomial pneumonia: ESBL strains with additional (Calculated initial parenteral Group 3a cephalosporins, aminopenicillin / β- resistance to carbapenems, therapy for bacterial diseases in lactamase inhibitor combinations, • Colistin in combination adults - update 2018) • or pneumococcal fluoroquinolones therapy [201] 3 However, it should be noted that the use of group 3 • Ceftazidime / Avibactam. cephalosporins increases the selection of vancomycin- resistant enterococci (VRE), ESBL-producing Enterobacteriaceae and beta-lactam antibacterial- resistant Acinetobacter spp. Fluoro-quinolones should also be used with caution due to the frequent resistance selection. “Guidelines for the management 2014 Spain • Patients without frailty: Pseudomonas aeruginosa: of community-acquired Outpatient setting - Amoxicillin/clavulanate or Piperacillin/tazobactam or pneumonia in the elderly cefditoren + clarithromycin or moxifloxacin or imipenem or meropenem or patient” levofloxacin cefepime + levofloxacin or [214] •Treatment at admission - Amoxicillin/ clavulanate or ciprofloxacin or amikacin or ceftriaxone + azithromycin or moxifloxacin or tobramycin levofloxacin • Patients with frailty: Mild cases - Amoxicillin/clavulanate or ceftriaxone + azithromycin or moxifloxacin or levofloxacin Moderate-severe cases – Ertapenem or amoxicillin/clavulanate

• Enterobacteriaceae/anaerobes: Ertapenem or amoxicillin/clavulanate Infectious Diseases and Clinical 2015 Spain KPC-producing Klebsiella pneumonia: NR Microbiology (SEIMC) 2015 • Combination therapy: carbapenem (see preferred [199] drug and recommended dose below) plus one or two fully active drugs (including colistin, tigecycline, and aminoglycoside or fosfomycin, the latter preferably as a third drug) is recommended if the carbapenem MIC is ≤8 mg/L; this applies mainly to patients with severe infections caused by Spanish Society of 2018 Spain VAP or Enterobacteriaceae with MIC ≥1 mg/L: P. aeruginosa: High-dose Transplantation (SET)/ Group • Tigecycline ceftolozane-tazobactam could for Study of Infection in be prescribed to solid organ Transplantation of the Spanish transplantation (SOT) recipients Society of Infectious Diseases and Clinical Microbiology (GESITRA-EIMC)/ Spanish Network for Research in Infectious Diseases (REIPI) [204] Spanish Society of Infectious 2019 Spain • Cefepime NR Diseases and Clinical • Piperacillin-tazobactam Microbiology (SEIMC) and the • Imipenem or meropenem Spanish Association of +/- Haematology and Hemotherapy • Fluoroquinolones, aminoglycosides, colistin (SEHH)

[205] 4 In critically ill patients, or patients previously colonized/infected with multidrug-resistant gram- negative bacilli, it is advisable to use a dual therapy strategy, according to local epidemiology. NICE guideline [NG139]: 2019 UK • Piperacillin with tazobactam NR Pneumonia (hospital-acquired): • Ceftazidime antimicrobial prescribing” • Ceftriaxone [215] 5 • Cefuroxime • Meropenem • Ceftazidime/ avibactam

Antibacterials to be added if suspected or confirmed MRSA infection (dual therapy with an IV antibacterial for empirical therapy) • Vancomycin, Teicoplanin, Linezolid 1https://pro.medicin.dk/Specielleemner/Emner/318019 2https://www.awmf.org/uploads/tx_szleitlinien/020-013l_S3_Nosokomiale_Pneumonie_Erwachsener_2017-11.pdf 3https://www.awmf.org/uploads/tx_szleitlinien/082-006l_S2k_Parenterale_Antibiotika_2019-08.pdf 4https://seimc.org/contenidos/documentoscientificos/seimc-dc-2019-Febrile_Neutropenia.pdf 5https://www.nice.org.uk/guidance/ng139/resources/pneumonia-hospitalacquired-antimicrobial-prescribing-pdf-66141727749061

Table 11c: Relevant guidelines for diagnosis and management – cUTI

Name of Date of Country (s) to Summary of recommendations Summary of recommendations society/organisation issue or which -MDR infections- -CR infections- issuing guidelines last update guideline applies European Association of 2018 Europe No specific recommendation for MDR No specific recommendation for CR Urology EAU [25] infections. infections.

Recommendations: • 3rd generation cephalosporin as empirical treatment of cUTI with systemic symptoms

Comparison of Guidelines Europe (15 Dosage according to resistance pattern NR antibacterial treatment from 2012- countries) • Ampicillin iv guidelines for urinary tract 2017 • Gentamicin infections in 15 European (one from • Amoxicillin / clavulanic acid countries: Results of an 2004, • Trimethoprim / sulfamethoxazole online survey ([216] Serbia) • Cefuroxime iv • Ciprofloxacin

pro. medicin Information til 2019 Denmark No specific recommendation for MDR NR sundhedsfaglige - infections Antibiotikavejledning [211] 1

Complicated cystitis/ Acute pyelonephritis: Pivmecillinam or Ciprofloxacin in penicillin allergy With urosepsis: Mecillinam +/- Gentamicin or Ampicillin +/- Gentamicin or Piperacillin/Tazobactam UTI in catheter carriers: Pivmecillinam or Ciprofloxacin Update on a proper use of 2015 France Severe pyelonephritis: ESBL-producing NR systemic fluoroquinolones Enterobacteriaceae 1st-line treatment, in adult patients – systematically with an aminoglycoside Recommendations [202] French Infectious Diseases Updated France Combination treatment: β-lactam, NR Society (SPILF) 2015, aminoglycoside [217] changes ESBL-E: Amikacin (risk of cross-resistance is decided in substantially lower with amikacin than with 2017 gentamicin or tobramycin). included AWMF (Registry Nr: 082- 2018 Germany nosocomial acquired or catheter-associated • Ceftolozane / tazobactam 006) (Calculated initial (2019 UTIs •Group 3b cephalosporins, including • Ceftazidime / avibactam parenteral therapy for publication the cephalosporin / BLI combinations bacterial diseases in adults year) ceftolozane / tazobactam and ceftazidime / - update 2018) avibactam, or 4 (cefepime), [201]2 • Group 2 or group 3 fluoroquinolones • Group 1 carbapenems

Spanish Society of Clinical 2016 Spain • Patients with healthcare acquired acute NR Microbiology and Infectious polynephritis (non-severe and severe): Diseases (SEIMC) Antipseudomonal carbapenem plus [218]3 ceftolozane-tazobactam or piperacillin- tazobactam • Severe sepsis: Amikacin • CA APN (complicated and uncomplicated): APN with specific risk factors for ESBL producing Enterobacteriaceae: First choice: ertapenem is an acceptable option, Second choice: other carbapenems or piperacillin- tazobactam • CA-APN with penicillin allergy: Amikacin or sodium fosfomycin Spanish Society of 2019 Spain • Cefepime NR Infectious Diseases and • Piperacillin-tazobactam Clinical Microbiology • Imipenem or carbapenem (SEIMC) and the Spanish Association of • Consider the addition of an aminoglycoside Haematology and or glycopeptide in critically ill patients, those Hemotherapy (SEHH) with indwelling urinary catheters, and/or a [205]4 history of colonization/infection with multidrug- resistant microorganisms

British Society for 2018 UK Pyelonephritis and cUTI caused by MDR Antimicrobial GNB: Chemotherapy/ • Meropenem or, ceftolozane/tazobactam or Healthcare Infection temocillin Society/British Infection • Piperacillin/tazobactam Association • Amikacin Joint Working Party • Ceftazidime/avibactam or non-b-lactam [122] agents in combination with meropenem

1https://pro.medicin.dk/Specielleemner/Emner/318019 2https://www.awmf.org/uploads/tx_szleitlinien/082-006l_S2k_Parenterale_Antibiotika_2019-08.pdf 3https://pdfs.semanticscholar.org/95c8/f85c6122ced97ad0d4076427b4fcba7e0214.pdf 4https://seimc.org/contenidos/documentoscientificos/seimc-dc-2019-Febrile_Neutropenia.pdf

Table 11d: Relevant guidelines for diagnosis and management – BSI/Sepsis

Name of society/organisation issuing guidelines Date of Country (s) Summary of Summary of issue or to which recommendations recommendations last update guideline -MDR infections- -CR infections- applies Surviving Sepsis Campaign Update International Combination therapy: two drugs Combination therapy: [27] 2018 (Europe and from different classes of broad coverage North antibacterials- usually a β- antibacterial + pathogen- America) lactam with a specific agent:

fluoroquinolone, Broad-spectrum aminoglycoside or macrolide carbapenem (e.g., meropenem, imipenem/cilastatin or doripenem) or extended- range penicillin/β- lactamase inhibitor combination (e.g., piperacillin/tazobactam or ticarcillin/clavulanate. third- or higher generation cephalosporins can also be used, especially as part of a multidrug regimen.

pro. medicin Information til sundhedsfaglige - 2019 Denmark No specific recommendation for NR Antibiotikavejledning (bacterial MDR infections https://www.pro.medicin.dk. section Revised Septic shock: Ampicillin + [211]1 13.01.2020) Gentamicin or Piperacillin / Tazobactam + Gentamicin

Suspected/detected Enterococci: add Vancomycin

Suspected/detected Pseudomonas aeruginosa: Ceftazidime + Gentamicin AWMF (Registry Nr: 079/001) (Prevention, diagnosis, 2010 Germany Treatment of Sepsis: NR therapy and aftercare of sepsis) (currently It is recommended to use a [219] 2 under Pseudomonas-effective revision) antibacterial ureidopenicillins (piperacillin) or third-party or fourth generation cephalosporins (ceftazidime or cefepime) or carbapenems (imipenem or meropenem) under consideration use local resistance patterns Infectious Diseases Working Party of the German 2012 Germany CVC-related bloodstream NR Society of Haematology and Medical Oncology infections (CRBSI) caused by [220] Stenotrophomonas maltophilia: • Co-trimoxazole

Infectious Diseases Working Party of the German Society 2014 Germany Neutropenic patients with NR of Haematology and Medical sepsis: Oncology • Imipenem/cilastatin or [221] piperacillin/ tazobactam • Combination treatment with aminoglycoside in case of severe sepsis

Helsedirektoratet, Antibiotikabruk i sykehus, Nasjonal Update Norway • Broad-spectrum beta-lactam No specific faglig retningslinje 2018 antibacterial if suspected recommendation for CR https://www.helsedirektoratet.no/Retningslinjer/Antibiotika- resistant microorganisms infections i-sykehus • Aminoglycosides (gentamicin or tobramycin) [208]3 for severe sepsis and septic shock • Septic shock and suspected gram-negative aetiology: gentamicin or ciprofloxacin

Infectious Diseases and Clinical Microbiology (SEIMC) 2015 Spain • Enterobacteriaceae: NR 2015 carbapenem [199] • Nosocomial sepsis and previous receipt of cephalosporins: fluoroquinolones or carbapenems Spanish Society of Transplantation (SET)/ Group for Study 2018 Spain • Enterobacteriaceae: • SOT recipients of Infection in Transplantation of the Spanish Society of Tigecycline diagnosed with BSI and/or Infectious Diseases and Clinical Microbiology (GESITRA- • mild infections: Carbapenem pneumonia caused by P. EIMC)/ Spanish Network for Research in Infectious monotherapy (extended- aeruginosa resistant to Diseases (REIPI) infusion) carbapenems and other β- [204] lactams, if the strain shows in vitro susceptibility:

High-dose ceftolozane- tazobactam

Spanish Society of Clinical Microbiology and Infectious 2018 Spain Based on local epidemiology: NR Diseases (SEIMC), Spanish Society of Intensive Care piperacillin-tazobactam, Medicine and Coronary Units (SEMICYUC) carbapenems, a fourth- [222] generation cephalosporin, aztreonam, quinolones or aminoglycosides Spanish Society of Infectious Diseases and Clinical 2019 Spain Use of carbapenems is Microbiology (SEIMC) and the Spanish Association of recommended for patients with Haematology and Hemotherapy (SEHH) sepsis or septic shock criteria [205]4 The UK joint specialist societies guideline on the diagnosis 2016 UK Suspected cases: NR and management of acute Ceftriaxone/cefotaxime meningitis and meningococcal sepsis in immunocompetent >60 years old & adults” immunocompromised patients: [223] Ampicillin/amoxicillin + cephalosporin GN diplococci: continued Ceftriaxone/cefotaxime Suspected ESBL infection: Meropenem Confirmed Neisseria meningitidis: Ceftriaxone/cefotaxime,

benzylpenicillin, ciprofloxacin (if not given ceftriaxone)

1https://pro.medicin.dk/Specielleemner/Emner/318019 2https://www.awmf.org/uploads/tx_szleitlinien/079-001l_S2k_Sepsis_2010-abgelaufen.pdf 3https://www.helsedirektoratet.no/Retningslinjer/Antibiotika-i-sykehus 4https://seimc.org/contenidos/documentoscientificos/seimc-dc-2019-Febrile_Neutropenia.pdf

Table 11e: Relevant guidelines for diagnosis and management- cIAI

Name of Date of Country (s) to Summary of recommendations Summary of recommendations society/organisation issue or which -MDR infections- -CR infections- issuing guidelines last guideline update applies World Society of 2017 Global Ceftolozone/tazobactam and Tigecycline (against carbapenemase- Emergency Surgery ceftazidime/avibactam (both in combination producing Enterobacteriaceae and (WSES) and World Society with metronidazole) Stenotrophomonas maltophilia) of Abdominal Compartment -Aminoglycosides with β-lactams -Ceftazidime/avibactam (against Syndrome (WSACS) -Polymyxins and fosfomycin (in critically ill carbapenemase producing K. patients) pneumoniae) [224]

Check update?

The Surgical Infection 2017 USA • High risk patients with Pseudomonas NR Society ([24] aeruginosa: ceftolozane-tazobactam + metronidazole

• High-risk patients with Klebsiella pneumoniae carbapenemase (KPC)-producing Enterobacteriaceae: ceftolozane-avibactam + metronidazole

pro. medicin Information til 2019 Denmark No specific recommendation for MDR NR sundhedsfaglige - infections Antibiotikavejledning Piperacillin / tazobactam + metronidazole + https://www.pro.medicin.dk. fluconazole [211]1 or

Cefuroxime + metronidazole + fluconazole in penicillin allergy

Update on a proper use of 2015 France Shigella sonnei diarrhea: ciprofloxacin or NR systemic fluoroquinolones ofloxacin in adult patients – First-line use of fluoroquinolones not Recommendations recommended in IAI; instead ciprofloxacin or [202] ofloxacin

Société française 2015 France • Amoxicillin / clavulanic acid + gentamicin NR d'anesthésie et de • Cefotaxime or ceftriaxone + imidazoles. réanimation (Sfar)

[225]2

Severe IAI: piperacillin / tazobactam gentamicin.

Haute Autorité de Santé 2019 France In IA EBLSE infections: piperacillin- NR (HAS), la Société de tazobactam pathologie infectieuse de • With septic shock: carbapenem (imipenem langue française (SPILF) et or meropenem). la Société de réanimation IA Enterobacteriaceae resistant to C3G by de langue française (SRLF) hyperproduction of cephalosporinase and [226]3 without production of ESBL: cefepime combined with metronidazole or ornidazole Serious IAI: piperacillin-tazobactam combined with amikacin AWMF (Registry Nr: 082- 2018 Germany The following antibacterials are recommended The following antibacterials are 006) (Calculated initial if suspected pathogens are suspected: recommended if suspected pathogens parenteral therapy for are suspected to be CR MRSA Tigecycline bacterial diseases in adults Tigecycline Linezolid+ - update 2018) Colistin Vancomycin+ [201]4 Ceftazidim/Avibactam VRE Tigecycline Meropenem (High doses)

Linezolid+ ESBL Tigecycline (E. coli, Ceftolozane/Tazobactam Klebsiella Ceftazidim/Avibactam spp.) Imipenem Meropenem Ertapenem

Fosfomycin (no Monotherapy) Acinetobacter Colistin spp. Tigecycline Sulbactam Pseudomonas Imipenem, Meropenem spp. Piperacillin/Tazobactam Cefepime Gentamicin, Amikacin Ciprofloxacin2, Levofloxacin2 Ceftolozane/Tazobactam Ceftazidim/Avibactam

1https://pro.medicin.dk/Specielleemner/Emner/318019 2https://www.infectiologie.com/UserFiles/File/medias/Recos/2014-inf-intra-abdo-SFAR.pdf 3https://www.infectiologie.com/UserFiles/File/spilf/recos/2019-synthese-infections-enterobacteries.pdf 4https://www.awmf.org/uploads/tx_szleitlinien/082-006l_S2k_Parenterale_Antibiotika_2019-08.pdf

2.4 Comparators in the assessment

1. Based on the alternatives presented, identify the technologies to be used as comparator(s) for the assessment.

2.4.1 General considerations

In general, in antibacterial research and development, in vitro, PK/PD models and clinical trials provide integrated sources of information for comparative analysis of effectiveness.

Clinical trials can provide reliable information regarding comparative efficacy when the pathogens have confirmed or expected susceptibility to both drugs. This is consistent with prescription based on AST results, which occurs in patients with confirmed CR infections. In this setting, Network meta-analysis (NMA) if feasible provide additional reliable information of comparative effectiveness, in the absence of direct comparative data, but rendered to show no significant differences between treatments included in the NMA, as all clinical trials are designed as non-inferiority and conducted in a population where all pathogens are expected to be susceptible to both treatments.

However, in patients with infections suspected to be caused by MDR/CR pathogens, clinical trials only provide limited comparative evidence regarding the efficacy of new antibacterials. This is because trials must include only pathogens for which the tested agents and comparators are effective, as it would be unethical to knowingly allow patients to have ineffective treatment. In this setting, standard NMAs also provide little information, as they never account for pathogens not susceptible to the treatment regimens included in the network. A comparison of efficacy against all relevant comparators can only be obtained from in vitro surveillance studies. Hence approaches integrating all available evidence from in vitro, PK/PD and clinical data (such as effectiveness models), are the necessary to predict susceptibility rates and clinical effectiveness rates.

Specific findings supporting the chosen comparators

The guideline review revealed that the comparators used in the clinical trials of cefiderocol were commonly used across different countries. For suspected MDR infections, carbapenems are (alone or in combination) recommended throughout, while for confirmed MDR/CR resistance, colistin-based regimens are considered (A. Baumannii, S. maltophilia, metalloβ- lactamases). Newer treatments are recommended in case of P. Aeruginosa (ceftolozane/tazobactam) and Enterobactereacea (ceftazidime/avibactam) for suspected and confirmed MDR pathogens.

This review results confirm that the comparators in clinical trials APEKS-NP, APEKS-cUTI and CREDIBLE-CR reflect the guideline recommendations for the different target populations. The trials have been designed in line with recommendations from regulatory authorities and allowed clinicians to select specific treatments (“best available care”) in the CR population (CREDIBLE-CR trial), for which different treatment options may be combined to combat very difficult-to-treat infections.

Comparator in APEKS-NP (Meropenem - high-dose and prolonged infusion) For APEKS-NP high-dose and prolonged infusion (HD) of meropenem was the selected comparator, as per FDA recommendations, given the severity of the population expected in the trial and likelihood of including patients with infections resistant to carbapenems (confirmed after trials inclusion). This regimen optimises exposure and time over MIC for Carbapenems, can be active in pathogens with MIC up to 16mg/L and has shown to improve prognosis in severe infections compared with short-term infusions (Microbiology (SEIMC) and the Spanish Association of Haematology and Hemotherapy (SEHH), Gudiol 2019)). In alignment with this strategy, some guidelines recommend high doses of β-lactam antibacterials for treatment of acute invasive infections by Pseudomonas aeruginosa (Mensa, Antibacterial selection in the treatment of acute invasive infections by Pseudomonas aeruginosa: Guidelines by the Spanish Society of Chemotherapy” (2018).

The fact that this HD meropenem was a regimen not used in other clinical trials as comparator, as well as including pathogens which other antibacterials were not active against (e.g. A. baumannii) did not allow a network to be built and therefore, an NMA could not be performed in patients with nosocomial infections. For more information on this topic please refer to the systematic literature review and feasibility assessment for a network meta-analysis of treatments of Gram-negative bacterial infections [227].

Comparator in APEKS-cUTI (imipenem/cilastatin) Given the probability of resistance to 3rd-class cephalosporins, imipenem / cilastatin was among the recommended treatments (e.g. Helsedirektoratet. Antibiotikabruk i sykehus, Kortversjon av Nasjonal faglige retningslinje for antibiotikabruk i sykehus 2014 [208]) and chosen as the comparator alongside imipenem, a recommended and commonly used carbapenem for the treatment of cUTI.

In addition to the trial-based comparison, an NMA was possible in cUTI, given the overlapping pathogen profile and similar patient baseline characteristics, across all relevant published studies. Treatments for suspected MDR Gram-negative infection were identified through a

systematic review of the literature. Trials focusing on efficacy and safety of current treatments for Gram-negative urinary tract infections were identified (see APEKS-cUTI section of chapter 5.4.3), including all trials that compared any parenteral antibacterials for the treatment of Gram-negative infection to placebo or another parenteral antibacterial. Based on selection according to bacteria, the potential network was designed in line with the intended label for cefiderocol. The NMA results were consistent with APEKS-cUTI trial results and expectedly found no statistically significant difference between cefiderocol and other comparators, given the considerations in section 2.4.1. For more information on this topic please refer to the relevant appendix [227].

This NMA provides supportive comparative information for patients with infections caused by confirmed CR resistant pathogens.

Comparator in CREDIBLE-CR (BAT based on combinations with colistin) As requested by regulatory authorities, the comparator in CREDIBLE-CR was BAT in order to enable the variable treatment approaches required where there are very limited options and given variable local epidemiology and resistance patterns. The heterogeneity of BAT drug combinations reflects the current treatment reality and selection of treatment choice according to most likely effective treatment in a given place and setting, consisting predominantly of colistin based regimens.

2.4.2 Selection of relevant comparators for the assessment

Following the EMA’s guidance and expected label approval, the comparators are defined predominantly based on pathogens (as opposed to infection sites). To address the high priority pathogens and based on recommendations and susceptibility tests, the following comparators are most relevant for each target treatment population

- Suspected MDR infection - carbapenems (including meropenem and imipenem, in monotherapy with high dose & prolonged infusion or combinations), ceftolozane/tazobactam, and ceftazidime/avibactam.

- Confirmed MDR infection- colistin-based (combination) regimens for (A. Baumannii, S. maltophilia, and other Gram-negative pathogens containing metalloβ-lactamases);

ceftolozane/tazobactam (P. Aeruginosa), ceftazidime/avibactam (Enterobactereacea)

Given the important role of in vitro surveillance studies for antibacterials, an assessment of cefiderocol needs to be based on a combination of comparisons of surveillance data and clinical evidence, as outlined below (Table 12).

Table 12: Cefiderocol assessment

Population Comparator Data source Result (cefiderocol vs. comparator) Suspected High dose Meropenem SIDERO WT Broader coverage of Gram-negative, MDR/CR surveillance aerobic pathogens. Lower MIC value and preserved efficacy in the presence of carbapenemases. APEKS-NP Non-inferior with regard to mortality RCT (primary outcome) and all clinical and microbiological secondary outcomes. High dose Meropenem Integrated Cefiderocol presents higher weighed Ceftalozane-tazobactam, epidemiology susceptibility rates in cUTI, Ceftazidime-avibactam and in vitro pneumonia, BSI, and gastrointestinal data analysis samples vs comparators High dose Meropenem Effectiveness Cefiderocol presents higher Ceftalozane-tazobactam, model likelihood of clinical and Ceftazidime-avibactam integrating microbiological effectiveness in epidemiology, pneumonia and cUTI vs in vitro data comparators. and clinical data Imipenem/Cilastatin APEKS-cUTI Non-inferior to comparator, but RCT proven superiority in a post-hoc analysis, on the primary endpoint of combined microbiological eradication / clinical cure at TOC, and secondary endpoint microbiological eradication at TOC. Ceftalozane-tazobactam, network meta- In similar patient populations with ceftazidime-avibactam, analysis for similar pathogen distribution across doripenem, cUTI different trials, and consistent with imipenem/cilastatin APEKS-cUTI there was statistical significant difference in microbiological eradication at TOC vs Imipenem/cilastatin, but in all other endpoints there was no statistically significant difference, including clinical cure rates and adverse events Ceftolozane/tazobactam SIDERO WT Lower MIC90 (0.25 vs. 8 for surveillance Pseudomonas, 0.25 vs. 32 for Acinetobacter, 1 vs. 64 for Enterobacteriaceae)4 Higher % isolates susceptible to cefiderocol Ceftazidime/avibactam SIDERO WT Same MIC90 for Enterobacteriaceae surveillance (1 vs. 1), otherwise superiority of cefiderocol Higher % isolates susceptible to cefiderocol Confirmed Colistin-based SIDERO CR Higher % isolates susceptible to CR (combination) regimens surveillance cefiderocol; Similar in vitro efficacy. (most relevant for for A. Colistin is known to have severe side Baumanii, S. maltophilia, effects, especially kidney toxicity. Resistances against colistin have

4 Longshaw et al., 2019 ID Week

pathogens with metalloβ- been reported to increase in lactamases) epidemiological studies. Ceftolozane/tazobactam SIDERO CR Higher percent susceptibility for (most relevant for P. surveillance cefiderocol against Acinetobacter aeruginosa, except and Pseudomonas across all pathogens with metalloβ- included countries (MEM-NS lactamases) pathogens)5 Ceftazidime/avibactam SIDERO CR Higher percent susceptibility for (most relevant for surveillance cefiderocol against Acinetobacter Enterobacterales, except and Pseudomonas across all pathogens with metalloβ- included countries (MEM-NS lactamases) pathogens) Best available therapy CREDIBLE-CR Descriptive results only. Evidence of (BAT), predominantly eradication of resistant pathogens. (combination) regimens Numerical, non-significant (most relevant for A. disadvantage with regard to mortality Baumanii, S. maltophilia, for cefiderocol compared to BAT. pathogens with metalloβ- lactamases)

The important question, raised during the scoping process, was: Given the large amount of heterogeneity in the treatment recommendation and the limited number of comparators in the surveillance data and the clinical studies, how can clinicians determine when to use cefiderocol over another potential candidate?

The answer combines the intended label with the target populations, as follows:

The indication for cefiderocol is expected to be:  Fetcroja is indicated for the treatment of infections due to aerobic Gram-negative organisms in adults with limited treatment options.

This indication will therefore be pathogen focused, not restricted to any specific site of infection and supports the use of cefiderocol in two types of patients:

5 Sato et al. 2019 ID Week

Thus, clinicians will encounter three types of patients in their clinical practice that would be treated according to the identified guidelines:

1) Patients with infections for which there are sufficient treatment options listed in the guidelines. These fall out of the scope of the cefiderocol label and are thus not relevant for the current assessment.

2) At the other end of the spectrum, patients with confirmed MDR/CR infections, for which the antibacterial susceptibility test shows that there are no other options but cefiderocol. These patients would gain an important new, last-resort option with cefiderocol.

3) Patients with suspected MDR/CR options, for which local surveillance data indicate that many of the currently available comparators will not provide cover against certain possible carbapenem-resistant pathogens, and who are critically ill and at risk of clinical deterioration. These patients would gain a new treatment option to reduce the risk of insufficient pathogen coverage leading to a delay in appropriate treatment and consequent clinical deterioration

The clinician could optimize the chances of success by considering different treatment options based on their indications, the MICs and breakpoints published by EUCAST, and the outcomes in trials of susceptible patient populations. Based on the local epidemiology, the clinician would then select an agent (or combination of agents) that would maximize the likelihood to cover the suspected pathogen.

All this data was integrated into an effectiveness model, where European epidemiological data for MDR pathogen prevalence rates for specific infection sites was used alongside with results from the SIDERO surveillance studies, and clinical cure rates from clinical trials, to estimate the likelihood of success of cefiderocol compared with the most relevant comparators.

The results indicate that cefiderocol would have the highest likelihood of success of clinical cure and microbiological eradication at these infection sites. For more information please see section 5.4.1 and 5.4.3

These calculations would have to be adjusted for individual cases by taking into account local variability of pathogen frequencies, but the approach illustrates a practical solution for the complex challenge of optimizing treatments in patients with suspected MDR/CR infections.

Once the antibacterial susceptibility test becomes available, doctors should again follow the guideline recommendations and de-escalate the treatment to the choice with the narrowest and specific spectrum for the identified pathogen.

In summary, the combined consideration of international guidelines, the growing unmet need of antimicrobial resistance, the fact that delays in appropriate treatment cause worse outcomes, indicate that cefiderocol constitutes a valuable addition to the current treatment landscape of Gram-negative pathogens for patients with limited treatment options.

3 Current use of the technology

Summary of issues relating to current use of the technology  Cefiderocol is not yet approved in Europe with the only current use being in compassionate use programmes. Over 200 patients globally were treated to date under the compassionate use programme of cefiderocol, underlining the clear unmet need in patients with highly resistant infections with no treatment options.

o The criteria for fulfilling these requests are highly restrictive. All other available treatments must be ruled out through susceptibility testing and/or where there is evidence of treatment failure (efficacy or safety).

o In addition, patients must be unable to enrol in clinical studies of cefiderocol.

 Case reports for three patients in the compassionate use programme have been published.

o A patient was treated successfully for endocarditis due to extensively drug resistant (XDR) Pseudomonas aeruginosa.

o A patient with multiple comorbidities and a complicated intra-abdominal infection (IAI) due to MDR Pseudomonas aeruginosa was released from hospital care within six weeks of completion of cefiderocol treatment.

o A patient with VAP and BSI caused by XDR Acinetobacter baumannii and carbapenemase-producing Klebsiella pneumoniae had potentially serious organ failure from older anti-infectives. Six weeks after cefiderocol administration, chest X-rays showed complete resolution of infection.

 An abstract submitted (not accepted) for ECCMID 2020 summarizes results from a case series of seven severely ill patients with CR Acinetobacter infections treated with cefiderocol. The two patients who died had received cefiderocol for only two days prior to death.

 While the compassionate use program is restricted to the use of cefiderocol for the treatment of XDR infections with no other options, the EMA-approved indication will be broader and encompass early targeted treatment of suspected MDR/CR/difficult-to-treat infections in addition to treatment of confirmed resistant pathogens.

3.1 Current use of the technology

1. Describe the experience of using the technology, for example the health conditions and populations, and the purposes for which the technology is currently used. Include whether the current use of the technology differs from that described in the (expected) authorisation.

Since cefiderocol is currently not approved in Europe, the only use has been within the compassionate use program. This program is registered in https://www.clinicaltrials.gov/ registry under NCT03780140: Expanded Access to Cefiderocol for the Intravenous Treatment of Severe Gram-Negative Bacterial Infections. Expanded access may be provided for cefiderocol for qualified patients who have limited treatment options and are not eligible for a clinical trial.

Case reports for three patients in the compassionate use programme have been published. o A patient was treated successfully for endocarditis due to extensively drug resistant (XDR) Pseudomonas aeruginosa. o A patient with multiple comorbidities and a complicated intra-abdominal infection (IAI) due to MDR Pseudomonas aeruginosa was released from hospital care within six weeks of completion of cefiderocol treatment. o A patient with VAP and BSI caused by XDR Acinetobacter baumannii and carbapenemase-producing Klebsiella pneumoniae had potentially serious organ failure from older anti-infectives. Six weeks after cefiderocol administration, chest X-rays showed complete resolution of infection.

To date over 200 patients have been treated with cefiderocol through this programme. Detailed information on 74 patients which have completed treatment with cafiderocol are presented in section 5.4 and are part of the data pacage that substantiates the efficacy of cefiderocol in patients with confirmed CR infections alongside CREDIBLE CR.

The criteria for compassionate use of cefiderocol are highly restrictive. All other available treatments must be ruled out through susceptibility testing, and/or there must be evidence of treatment failure (efficacy or safety). Enrolled patients will have confirmed CR infection and are likely to be consistent with the target population where patients have confirmed CR infections. However, EMA-approved indication will be broader and encompass both the treatment of confirmed resistant infection and patients with infections by suspected MDR pathogens.

2. Indicate the scale of current use of the technology, for example the number of people currently being treated with the technology, or the number of settings in which the technology is used.

200 patients have been treated to date with cefiderocol in this programme. Results from 74 patients are available (see section 5.4).

Compassionate use of cefiderocol is restricted to seriously ill, hospitalised patients. Cefiderocol is and will be used in critically ill hospitalised patients, many of whom will be treated in ICU units. These patients will often be unconscious, and on many occasions require ventilation (intubation). This is consistent with existing intravenous use of antibacterials in critically ill hospitalised patients.

3.2 Reimbursement and assessment status of the technology

1. Complete Table 13 with the reimbursement status of the technology in Europe.

Table 13: Overview of the reimbursement status of the technology in European countries

Country and Status of recommendation If positive, level of reimbursement* issuing (positive/negative/ongoing/not organisation assessed) NA NA NA Include a reference to any publicly available guidance documents *For example, full reimbursement or only partial reimbursement. If partial reimbursement gives a percentage of reimbursement.

4 Investments and tools required

Summary of issues relating to the investments and tools required to introduce the technology

 The use of cefiderocol is for hospital use only, and is not expected to require any specialized equipment, or to demand additional resources beyond the standard ability to store, prepare and administer intravenous infusion treatments, alongside susceptibility testing to cefiderocol and standard monitoring microbiological evaluation tests, as it is current practice with all hospital use antibacterials in nosocomial infections.

 Cefiderocol is formulated as a freeze-dried (lyophilized) powder (1g/vial) for powder for concentrate for solution for infusion. Following reconstitution, it is administered as a 3- hour infusion of 2g every eight hours.

o Consideration should be given to official guidance on the appropriate use of antibacterial agents. Treatment should commence for pathogens highly suspected to be susceptible to cefiderocol, and susceptibility should be confirmed through appropriate diagnostic testing as soon as possible.

o It is recommended that cefiderocol should be used to treat patients that have limited treatment options only after consultation with a physician with appropriate experience in the management of infectious diseases.

o Cefiderocol may be used in combination with other antibacterial agents active against anaerobic pathogens and/or Gram-positive pathogens when these are known or suspected to be contributing to the infectious process.

o Dose adjustments are necessary for patients with renal impairment, but not for hepatic impairment. No adjustment is required in elderly populations. The safety and efficacy of cefiderocol in children below 18 years of age has not yet been established.

 Treatment of severe MDR-GNB infections in critically ill patients will require an expert and complex clinical reasoning, taking into account the peculiar characteristics of the target population, but also the need for adequate empirical coverage and the more and more specific enzyme-level activity of novel antimicrobials with respect to the different resistance mechanisms of MDR-GNB.

 The unmet need of developing additional effective antibacterials is accompanied by needs to further establish improved antimicrobial stewardship programs, to provide an

alternative treatment option when there are limited effective alternatives, and reducing the likelihood of initial inappropriate pre-emptive therapy, while the identification of pathogens and their susceptibility patterns is not available.

o For hospitalised patients with infection by suspected (but unconfirmed by an antibiogram) GN MDR/CR pathogens who are critically ill and cannot wait for results of an AST, cefiderocol will fill an important unmet need providing a more likely initial appropriate treatment. With these factors, therefore it is likely that patient outcomes improve, and length of stay associated with reduced time to effective therapy are minimised. Early appropriate treatment with cefiderocol is more likely to avoid delays in effective treatment. This may reduce healthcare resource utilization and costs associated with prolonged stay in, or admission to, intensive care units and/or delays in hospital discharge.

o Cefiderocol can be added to a combination regimen in case of multiple pathogens with diverse Gram status or administered alone for the treatment of Gram-negative MDR bacteria.

 Additional reductions in healthcare resource utilization are possible, when considering the broad context of AMR and how cefiderocol as a new treatment option contributes to it; i.e., possible fewer hospital quarantine due to spread of CR bacteria (within hospitals and across countries), potential reduction of complications when treating immunosuppressed patients (e.g., cancer patients with febrile neutropenia, who require effective antimicrobial treatment options). These additional values of new antibacterials have recently been highlighted by several European initiatives (see EEPRU report in the UK) focusing on innovative reimbursement models that aim to capture the full value that new antibacterials convey.



4.1 Requirements to use the technology 1. If any special conditions are attached to the regulatory authorisation more information should be provided, including reference to the appropriate sections of associated documents (for example, the EPAR and SPC). Include:  conditions relating to settings for use, for example inpatient or outpatient, presence of resuscitation facilities  restrictions on professionals who can use or may prescribe the technology  conditions relating to clinical management, for example patient monitoring, diagnosis, management and concomitant treatments.

4.1.1 Conditions for use

Cefiderocol is used as IV infusion and intended for hospital use only in patients that are hospitalised. It is recommended that cefiderocol should be used to treat patients that have limited treatment options only after consultation with a physician with appropriate experience in the management of infectious diseases.

Consideration should be given to official guidance on the appropriate use of antibacterial agents. Treatment should commence for pathogens highly suspected to be susceptible to cefiderocol, and susceptibility should be confirmed through appropriate diagnostic testing as soon as possible

4.1.2 Good stewardship and societal considerations

As applicable to all antibacterials, cefiderocol should be used according to good stewardship practices. Such use holds the promise to decrease the total use of resources in the hospital and area, due to the ability of new, infective antimicrobials to decrease resistances to existing last-resort agents and to avoid hospital shutdowns.

Antimicrobials thus provide additional value to society that is not measured in clinical trials. These elements of value have been summarized in a recent report based on a multi- stakeholder conference in 2017 [228-231]. They include transmission, insurance, diversity, novel action, enablement, and spectrum value. “Transmission value” refers to the fact that novel antibacterials contribute to slowing the spread of resistant genes. Insights from the evaluations of vaccines can be useful to support a quantitative assessment of this value. “Insurance value” means that the general population is protected from catastrophic outbreaks by having a last-resort antibacterial available. This value is independent of the actual quantities used in the clinics and applies even if a new, last-resort treatment is being kept for emergency use only. “Diversity value” means that additional treatments increase the value of older treatments over time, because those become effective. Few other therapeutic areas deal with medicines that provide such value, and standard HTA procedures are optimized to compare alternative options in head-to-head comparisons but do not yet account for such additional, synergistic value. “Novel-action value” refers to the fact that antibacterials with new mechanisms of action are especially valuable due to the lower risk of cross-resistance. “Enablement value” arises when the availability of antibacterials ensures safe surgeries and chemotherapies. Due to the availability of effective treatments, this can easily be taken for granted and may be challenged by resistant bacteria in the foreseeable future. “Spectrum value” means that antibacterials with a narrower (i.e.,“more specific”)

spectrum are more valuable for the treatment of susceptible pathogens than those with a broader spectrum due to the lower risk of inducing additional resistances.

These additional elements of value due to improved resource utilization are important considerations for the evaluation of novel antibacterials, which have the potential to lead to substantial savings of resources if used in accordance with good stewardship practices.

Finally, antimicrobial stewardship includes the rapid identification of bacterial infections and treatment with the most appropriate drug regimen [174, 175]. Improved early-detection and characterization methods for bacterial infections used in hospital environments in which cefiderocol and/or other modern antibacterials are available can thus lead to further resource optimization, by allowing antibacterials use to be optimized for patients with infections susceptible to certain treatments. While these resources are not a requirement for the use of cefiderocol, they may help to optimize its use and effectiveness.

No special conditions are attached to the regulatory authorisation for cefiderocol with respect to settings for use of cefiderocol. It will be used in critically ill hospitalised patients, many of whom will be treated in ICU units. This is consistent with normal intravenous use of antibacterials in these patients.

Clinical Particulars of the SmPC [232] highlight the following information:

 Consideration should be given to official guidance on the appropriate use of antibacterial agents. Treatment should commence for pathogens highly suspected to be susceptible to cefiderocol, and susceptibility should be confirmed through appropriate diagnostic testing as soon as possible.

 It is recommended that Fetcroja should be used to treat patients that have limited treatment options only after consultation with a physician with appropriate experience in the management of infectious diseases

 Cefiderocol may be used in combination with antibacterial agents active against anaerobic pathogens and/or Gram-positive pathogens when these are known or suspected to be contributing to the infectious process.

 Dose adjustments are necessary for patients with renal impairment, but not for hepatic impairment. No adjustment is required in elderly populations. The safety and efficacy of cefiderocol in children below 18 years of age has not yet been established.

Pathogen susceptibility should be determined as quickly as possible to identify the most appropriate antibacterial to achieve microbiological eradication. AST is an integral part of antibacterial treatment in the hospital. Use of cefiderocol will not require any capital investment for AST beyond standard diagnostic microbiology lab equipment and consumables.

Shionogi is working with several diagnostics manufacturers and the EUCAST Development Laboratory (EDL) to ensure a standard Kirby–Bauer test assay (disc-diffusion sensitivity assay) is available for cefiderocol. This widely used assay will be accredited and cross- referenced to both microdilution (BMD) Minimum Inhibitory Concentration (MIC) results, with corresponding EDL breakpoints and inhibition zone diameter correlates. If microbiology laboratories have the appropriate KB discs, AST determination for cefiderocol can be conducted alongside other drugs with no need for specialised equipment or growth media. There will also be a available methodology for BMD MIC determination which in the manual read version requires no specialist devices other than standard incubation and plate reading equipment.

Shionogi also has collaborations with several diagnostics manufacturers to develop other AST technologies, including epsilometer strips (eTest), inclusion in various automated AST panels and ready-made BMD strips for MIC determination. These should be commercially available shortly after launch.

2. Describe the equipment required to use the technology.

Cefiderocol is provided as a 1g powder for concentrate for solution for infusion. For each complete infusion, 2 vials are needed. It needs to be stored in a refrigerator (2 to 8°C) and should be stored in the original carton in order to protect it from light (see SmPC). Each vial is for single use only.

The use of cefiderocol is not expected to require any other specialized equipment, or to demand additional resources beyond those already required to administer an intra-venous antibacterial to hospitalised patients and to determine pathogen susceptibility.

3. Describe the supplies required to use the technology.

The powder should be reconstituted with 10 mL of either sodium chloride 9 mg/ml (0.9%) solution for injection or 5% dextrose injection taken from the 100 mL bags that will be used to prepare the final infusion solution. Standard infusion equipment and training is required.

5 Clinical effectiveness and safety

Summary of the effectiveness (combined evaluation of in vitro, PK/PD, and clinical data)  The anti-microbial efficacy of cefiderocol has been investigated in several major in vitro susceptibility studies (SIDERO-WT/Proteeae 2014/2015/2016 surveillance studies), including both European and US clinical isolates. In addition multiple smaller, country specific replicate studies were also conducted with similar results. These studies support the use of cefiderocol in both target populations: suspected and confirmed infections in several infection sites with MDR/CR/difficult-to-treat pathogens.

o Patients with infection caused by suspected MDR pathogen: The SIDERO- WT-2014-2016 study (which tested the in vitro antibacterial activity of cefiderocol against Gram-negative bacteria in 30,459 isolates across the world that included MDR and difficult-to-treat strains), cefiderocol demonstrated potent inhibition activity against 99.5% of Gram-negative isolates at a MIC of 4 mg/L (as defined by CLSI) including European clinical isolates, of K. pneumoniae, P. aeruginosa, A. baumannii, S. maltophilia and B. cepacia complex. Isolates were less susceptible to the comparators including ceftazidime-avibactam (90.2%) and ceftolozane-tazobactam (84.28%).

o Patients with infection caused by a confirmed CR pathogen: The SIDERO- CR-2014-2016 study (which tested the in vitro antibacterial activity of cefiderocol against CRE and MDR non-fermenters), cefiderocol demonstrated potent in vitro activity at a MIC of 4 mg/L (as defined by CSLI) against 96.2% of isolates of carbapenem non-susceptible pathogens including all of the WHO priority pathogens and Stenotrophomonas maltophilia. Cefiderocol was found to have a wider Gram-negative coverage, and more potent in vitro activity than comparators (cefepime, ceftazidime/avibactam, ceftolozane/tazobactam, ciprofloxacin, colistin, and meropenem) against a range of CR-GN isolates, including those non-susceptible to colistin.

 The antimicrobial efficacy results from the in vitro studies are further supported by in-vivo studies in animal models showing that cefiderocol penetrates into the target tissues at therapeutic doses.

 The clinical efficacy and safety of cefiderocol was demonstrated in two randomised double-blinded clinical trials and one open label, randomised descriptive study. The two randomised double-blinded clinical trials (APEKS NP and APEKS cUTI)

provide confirmatory clinical evidence of cefiderocol’s efficacy and safety in patients with suspected MDR/difficult-to-treat infections at risk of carbapenem resistance. Reports of compassionate use cases also contribute to the overall efficacy characterization of cefiderocol.

o APEKS-NP trial, compared treatment with cefiderocol against the high-dose, prolonged infusion (HD) meropenem in patients with nosocomial pneumonia caused by MDR Gram-negative pathogens. Cefiderocol met the primary endpoint of non-inferiority in ACM at day 14 versus HD meropenem (12.4% for cefiderocol and 11.6% for meropenem; (95 % CI: -6.6, 8.2)) and similar results were obtained between arms for ACM at Day 28 and EOS. Rates of clinical cure and microbiological eradication at TOC and other time points were also similar between the treatment groups.

o APEKS-cUTI, cefiderocol demonstrated an adjusted treatment difference vs imipenem/cilastatin of 18.6% (95 % CI: 8.2, 28.9), which proven superiority in a post-hoc analysis, in cUTI caused by Gram-negative MDR pathogens in hospitalized adults, in the primary composite endpoint (microbiological eradication and clinical cure).

 A Network Meta-Analysis (NMA) was feasible for cUTI, given the similarity of patients and pathogens included across trials. All results showed no statistically significant difference compared with ceftazidime/avibactam and ceftolozane/tazobactam in a similar patient population with similar pathogen distribution.

 Furthermore, in an effectiveness model that incorporated European pathogen epidemiology and susceptibility rates (based on EUCAST breakpoints), and clinical trials data, cefiderocol provides the best predicted susceptibility rates and estimated clinical and microbiological success rates, in the absence of an antibiogram for the critically ill patients with infections caused by suspected MDR pathogen infection requiring immediate treatment.

 The third, small, descriptive, exploratory, open-label study in patients with confirmed carbapenem-resistant pathogen infections, supports cefiderocol’s use in the confirmed-resistant population alongside with the compassionate use cases:

o CREDIBLE CR study was a small, exploratory, randomised, open label, descriptive study, conducted to evaluate efficacy in patients with confirmed CR infections for cefiderocol and BAT, but not designed or powered for statistical comparison between arms. The study included 150 severely ill patients,

consistent with compassionate use cases (many patients had end stage comorbidities and had failed multiple lines of therapy), with a range of infection sites including nosocomial pneumonia, cUTI, BSI/sepsis. Clinical and microbiological outcomes were similar between the 2 arms, but there were marked clinical differences in some baseline characteristics and pathogen distribution of the cefiderocol and BAT arms.

Summary of safety  Overall, cefiderocol was generally well tolerated, and the safety profile of cefiderocol was found to be consistent with that of other cephalosporin antibacterials. The clinical safety for cefiderocol as observed in the three randomised clinical trials, including 549 treated patients.

o Pooled adverse event analyses there overall less treatment emergent adverse events with cefiderocol (344/549 [67.1%]) vs comparators (252/347 [72.6%]). The most common adverse reactions for cefiderocol were diarrhoea (8.2%), constipation (4.6%), pyrexia (4.0%) and UTI (4.7%).

o In the total sample, 56/549 (10.2%) patients treated with cefiderocol experienced treatment related AEs and 45/347 (13.0%) patients treated with comparators.

 The clinical safety for cefiderocol has been investigated in three randomised clinical trials, two specific to different infection sites and one specific to carbapenem-resistant pathogens.

o HAP/VAP/ HCAP study (APEKS-NP): Overall, TEAEs and treatment- related TEAEs were balanced between treatment arms. Serious adverse events occurred in 36% of patients treated with cefiderocol and 30% of patients treated with meropenem. The most frequently observed AE was urinary tract infection (15.5% in cefiderocol group and 10.7% in meropenem group), hypokalaemia (10.8% in cefiderocol group and 15.3% in meropenem group) and anaemia (8.1% in cefiderocol group and 8% in meropenem group).

o cUTI study (APEKS-cUTI): The proportion of patients who experienced at least one adverse event (AE) was lower in the cefiderocol group than in the IPM/CS group (41 % vs 51%). Gastrointestinal disorders, such as diarrhoea and constipation, were the most common adverse events and

there was an increased incidence of C. difficile colitis in the imipenem/cilastatin arm compared with cefiderocol. Serious adverse events (SAE) occurred in a numerically lower proportion of cefiderocol- treated patients than of IPM/CS-treated patients (5% vs 8%). The most frequently observed AEs were gastrointestinal, such as diarrhoea [experienced by 4.3% (13/300) and 6.1% (9/148) of cefiderocol- and IPM/CS-treated subjects, respectively.

o CR study (CREDIBLE-CR): The cefiderocol group had lower incidence of AEs and treatment-related AEs, but higher incidence of death, SAEs and discontinuation due to AEs, compared with BAT. The incidence of treatment-related AEs leading to discontinuation was similar between treatment groups. An imbalance in mortality was observed in the cefiderocol arm compared to BAT (18/49 vs 5/25). No deaths were found to be causally associated with cefiderocol through assessment by the investigator and two independent committees. Furthermore, whereas the mortality rate in the cefiderocol group was consistent with previous studies in similar populations the evidence suggests that the mortality rate in the BAT group was unexpectedly low for the population randomised. No single factor that would explain the imbalance was identified. Small patient numbers and multiple confounders preclude definitive conclusions.

 Like in any other beta-lactam antibacterial, patients who have a history of hypersensitivity to carbapenems, penicillins or other beta-lactam antibacterial medicinal products may also be hypersensitive to cefiderocol. Before initiating therapy with cefiderocol, careful inquiry should be made concerning previous hypersensitivity reactions to beta-lactam antibacterials.

5.1 Identification and selection of relevant studies

The research question for this assessment is: “Do patients with aerobic Gram-negative infections and limited treatment option benefit from cefiderocol as an additional treatment option?”

Cefiderocol is expected to be approved for adult patients with aerobic, Gram-negative infections with limited treatment options. This indication comprises:

 critically ill patients with suspected infection by a carbapenem-resistant Gram- negative pathogen or other Gram-negative pathogen difficult to treat with limited treatment options

and

 patients with confirmed infection by a carbapenem-resistant Gram-negative pathogen or other Gram-negative pathogen difficult to treat with limited treatment options.

As outlined in chapter 2.4, clinical trials can only provide limited evidence regarding the efficacy of new antibacterials because trials must focus on pathogens for which the tested agents and comparators are effective. Comparison of efficacy against all relevant comparators in the antibacterial setting can only be obtained from in vitro surveillance studies. Unlike in other therapeutic areas, the evaluation of the effectiveness of an antibacterial relies on the combined consideration of in vitro, PK/PD and clinical data.

To identify all relevant studies for cefiderocol and comparators for the two patient populations a comprehensive systematic literature review was conducted comprising in vitro and in-vivo studies, as well as any comparative or non-comparative studies and RCTs (including cross- over RCTs).

The search strategy was broad to ensure that all relevant studies were captured. The only restriction was that cefiderocol or any of its synonyms had to be an intervention in the study. The search strategy is shown below in Figure 20.

Figure 20 - Search strategy for OVD MEDLINE ALL

Through this methodology, all studies of cefiderocol in various populations were captured and a complete study pool defined as the primary evidence base:

 Surveillance studies provide evidence for the expected efficacy of cefiderocol compared to other treatment options for patients with suspected MDR/CR infection.

o Site-specific clinical studies in patients with suspected MDR/CR infections amend the evidence pool.

o In addition, to amend the evidence for cefiderocol versus comparators for the population with suspected infection, a systematic literature review was conducted, to retrieve data for a potential network meta-analyses of cefiderocol versus approved comparators in the indication cUTI and nosocomial pneumonia. Details and results are described in chapter 5.4.3.

 Surveillance studies from confirmed infections provide evidence for expected efficacy of cefiderocol versus comparators in patients with suspected MDR/CR infection.

o The evidence is amended by subpopulations from RCTs, descriptive clinical study and by compassionate use cases.

The combined evidence is appropriate for the assessment of cefiderocol considering the scope outlined in EUnetHTA project plan.

In addition, a systematic literature was conducted to evaluate the feasibility of conducting 2 NMAs: 1 in cUTI and another in nosocomial pneumonia, identifying all relevant comparators and their clinical data for each NMA. All information on this SLR is available in the appendix [227].

1. State the databases and trial registries searched and, when relevant, the platforms used to do this.

A literature search in the following databases and information resources (Table 14) identified a total of 428 records.

Table 14: Databases and information sources searched

Database / information source Interface / URL Coverage MEDLINE ALL Ovid SP Biomedical journal literature

Database / information source Interface / URL Coverage https://www.ncbi.nlm.nih.gov/p PubMed Biomedical journal literature ubmed/ Embase OvidSP Biomedical journal literature Cochrane Central Register of Randomized and quasi- Wiley Cochrane Library Controlled Trials (CENTRAL) randomized controlled trials Cochrane Database of Wiley Cochrane Library Systematic reviews Systematic Reviews Database of Abstracts of Reviews CRD website Systematic reviews of Effects (DARE) Health Technology Assessment CRD website Health technology assessment (HTA) http://apps.webofknowledge.co Science, social science, arts, Web of Science m/ humanities http://apps.webofknowledge.co Life sciences and biomedical BIOSIS Citation Index m/ research Records for registered clinical ClinicalTrials.gov https://www.clinicaltrials.gov/ct studies WHO International Clinical Trials http://www.who.int/ictrp/en/ Trial registration data sets Registry Platform (WHO ICTRP)

The PubMed search was restricted to records not yet fully indexed for MEDLINE. The trials register sources listed above (ClinicalTrials.gov and ICTRP) were searched to identify information on studies in progress.

Recent research published as conference abstracts were identified by searching Embase (which indexes a significant number of conference publications). In addition, where the following conferences (identified as highly relevant by the research team) were not indexed in Embase from 2016 to 2019, we conducted hand-searches for abstracts via conference webpages:

 European Congress of Clinical Microbiology & Infectious Diseases (ECCMID)

 IDWeek

 European Respiratory Society (ERS) International Congress.

Reference lists of any relevant reviews or systematic reviews for eligible records were also checked.

2. State the date the searches were done and any limits (for example date, language) placed on the searches.

The searches were conducted between 7th October 2019 and 11th October 2019. A supplementary set of abstracts relevant to cefiderocol from the IDWeek 2019 conference (October 2-9, Washington DC, USA) was obtained in a grey literature search on December 19, 2019.

3. Include as an appendix the search terms and strategies used to interrogate each database or registry.

The report “Systematic Searches and Study Selection to Identify Clinical and Non- Clinical Evidence Available for Cefiderocol” contains the full strategies (including search dates) for all sources searched [233].

4. In Table 15, state the inclusion and exclusion criteria used to select studies and justify these.

Table 15: Inclusion and exclusion criteria

Criterion Inclusion criteria Exclusion criteria

 Cell based (bacteria, human or animal) Population  Animal (healthy and infected) NA  Human (healthy and infected) Any other Intervention  Cefiderocol intervention  Any intervention, placebo or best standard of care Comparators NA  Studies with no comparator  Clinical cure  Microbiological eradication  All-cause mortality  Adverse events Outcomes NA  Microbiology  Pharmacodynamic  Pharmacokinetic  Toxicology

Criterion Inclusion criteria Exclusion criteria

 RCTs of any duration  Cross-over RCTs if data are presented at time of  Systematic cross-over reviews, reviews, Study designs  Any comparative or non-comparative studies both opinion pieces (for prospective and retrospective listing in the final  In vitro studies report only)  In vivo studies Limits No date or language limits

Note that the search strategy shown above included all cefiderocol references, regardless of the measured outcomes. A second screening step then assigned the identified publications to different topics of interest (see below).

5. Provide a flow chart showing the number of studies identified and excluded. The PRISMA statement can be used; the PRISMA flow chart is included below, as an example.

Figure 21 shows the PRISMA flow diagram of the numbers of records included and excluded at each stage of the selection process.

Following deduplication, 254 records remained for assessment. Twelve records were excluded after an assessment of the information in the title and abstract. 242 full text documents were assessed and 129 records were included.

Additional eligible records identified through reference checking of reviews or systematic reviews which had not been identified through the database searches or hand searched conferences, have been included in the PRISMA flow diagram as identified from other sources.

5.1.1 PRISMA Chart

Figure 21 - PRISMA flow diagram of record selection process

5.1.2 Study categorisation

Studies were first categorised into the following five primary categories:

 In vitro  In vivo  Clinical  Multiple categories  Modelling simulation  Systematic review protocol

Then, for each study, the type of outcome data was reported:

 Efficacy  Frequency of spontaneous resistance  Method reproducibility  Mode of action  Pharmacokinetic/pharmacodynamic  Pharmacokinetic  Pharmacodynamic  Safety

Additionally, based on the identified scope, we screened the identified publications for possible results regarding hospital utilization and quality of life. No such studies were identified, as expected in the context of antibiotic treatment trials.

Of the 129 identified records:

 39 were conducted in vitro, two of which were letters reporting data on cefiderocol  37 records reported in vitro investigations of specifically identified clinical isolates  16 records reported in vivo investigations  25 records reported details of clinical studies, eight of these records reported protocol information only  Five records reported details of mixed study investigations e.g. both in vitro and in vivo  Six records reported details of modelling simulations  One record reported details of a Cochrane systematic review protocol which cefiderocol is an eligible intervention The supplementary search of the IDWeek 2019 conference (October 2-6, 2019), the presentations of which became available after the cut-off date for the SLR (October 7- 11, 2019), yielded 12 additional relevant presentations involving company staff from the manufacturer of cefiderocol. These presentations were added to the table of identified studies and screened for relevant content. References were added to Table 16 – List of all relevant studies - below.

5.2 Relevant studies

1. In Table 16 provide a list of the relevant studies identified.

Table 16: List of all relevant studies

Study reference/ID Available documentation* Status (ongoing**/ complete) In vitro Surveillance SIDERO WT Tsuji M, Hackel M, Yamano Y, Echols R, Longshaw C, Manissero D, et al. Cefiderocol in vitro activity against Continuous gram-negative clinical isolates collected in Europe: result from three SIDERO-WT surveillance studies between collection of world- 2014-2017. In: ECCMID, 2019. wide isolates (currently ca. 38k Hackel MA, Tsuji M, Yamano Y, Echols R, Karlowsky JA, Sahm DF. In vitro activity of the siderophore isolates) cephalosporin, cefiderocol, against a recent collection of clinically relevant gram-negative bacilli from North America and Europe, including carbapenem-nonsusceptible isolates (sidero-wt-2014 study). Antimicrob Agents Chemother. 2017;61(9):1-9.

Ito A, Kuroiwa M, Rokushima M, Hackel M, Sahm D, Tsuji M, et al. Characterization of isolates showing high MICs to cefiderocol from global surveillance study SIDERO-WT-2014. In: ASM Microbe 2019, San Francisco; 2019.

Kazmierczak KM, Tsuji M, Wise MG, Hackel M, Yamano Y, Echols R, et al. In vitro activity of cefiderocol, a siderophore cephalosporin, against a recent collection of clinically relevant carbapenem-non-susceptible Gram- negative bacilli, including serine carbapenemase- and metallo-β-lactamase-producing isolates (SIDERO-WT- 2014 Study). Int J Antimicrob Agents. 2019;53(2):177-84.

Tsuji M, Hackel M, Yamano Y, Echols R, D S. Surveillance of cefiderocol in vitro activity against gram-negative clinical isolates collected in Europe: SIDERO-WT-2014. In: ECCMID, 2017.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. Global surveillance of cefiderocol (S-649266) against Gram- negative clinical strains collected in North America: SIDERO-WT-2014. In: ASM Microbe 2017, New Orleans; 2017.

Tsuji M, Kazmierczak K, Hackel M, Echols R, Yamano Y, D S. Cefiderocol susceptibility against globally isolated meropenem-non-susceptible gram-negative bacteria containing serine-and metallo-carbapenemase genes: SIDERO-WT-2014 and -2015. In: ASM Microbe, San Francisco; 2019.

Yamano Y, Tsuji M, Echols R, Hackel M, Sahm D. In vitro activity of cefiderocol against gram-negative clinical isolates from respiratory specimens: Sidero-WT-2014. Am J Respir Crit Care Med. 2018;197

Tsuji M, Hackel M, Echols R, Yamano Y, D S. In vitro activity of cefiderocol against gram-negative clinical isolates collected from urinary track source: SIDERO-WT-2014/SIDERO-WT-2015 In: IDWeek, 2017.

Nguyen S, Hackel M, Hayes J, Sahm D, Echols R, Tsuji M, et al. In vitro antibacterial activity of cefiderocol against an international collection of carbapenem-non-susceptible gram-negative bacteria isolated from respiratory, blood, skin/soft tissue and urinary sources of infection: SIDERO-WT2014-2016. In: ECCMID, 2019.

Mackenzie T, Nguyen S, Haynes J, Hackel M, Echols R, Sahm D, et al. Cefiderocol activity against North American clinical isolates SIDERO-WT-2014–2017. In: ASM Microbe 2019, San Francisco; 2019.

Nguyen S, Hackel M, Hayes J, Sahm D, Echols R, Tillotson G, et al. In vitro antibacterial activity of cefiderocol against carbapenem-non-susceptible gram-negative bacteria from hospitalized patients in the United States: SIDERO-WT-2014–2017. In: ASM Microbe 2019, San Francisco; 2019.

Karlowsky JA, Hackel MA, Tsuji M, Yamano Y, Echols R, Sahm DF. In vitro activity of cefiderocol, a siderophore cephalosporin, against gram-negative bacilli isolated by clinical laboratories in north america and europe in 2015-2016: SIDERO-WT-2015. Int J Antimicrob Agents. 2019;53(4):456-66.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. In vitro antibacterial activity of cefiderocol (S-649266) against gram-negative clinical strains collected in North America and Europe: SIDERO-WT-2015. In: ASM Microbe 2017, New Orleans; 2017.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. Global surveillance of cefiderocol against gram-negative clinical strains collected in North America: SIDERO-WT-2015. In: IDWeek, 2018.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. In vitro antibacterial activity of cefiderocol against gram-negative clinical strains collected in North America and Europe: SIDERO-WT-2016. In: ASM Microbe 2019, San Francisco; 2019. SIDERO CR Hackel MA, Tsuji M, Yamano Y, Echols R, Karlowsky JA, Sahm DF. In vitro activity of the siderophore Completed cephalosporin, cefiderocol, against carbapenem-nonsusceptible and multidrug-resistant isolates of gram- negative bacilli collected worldwide in 2014 to 2016. Antimicrob Agents Chemother. 2018;62(2):1-13.

Tsuji M, Kazmierczak K, Hackel M, Wise M, Echols R, Sahm D, et al. Cefiderocol susceptibility and geographical analysis against globally isolated meropenem non-susceptible gram-negative bacteria containing serineand metallo-carbapenemase gene In: ECCMID, 2019.

Ito A, Hackel M, Sahm D, Tsuji M, Y Y. Characterization of isolates showing high MICs to cefiderocol from global surveillance study SIDERO-CR-2014/2016. In: ECCMID, 2019.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. In vitro activity of cefiderocol against globally collected carbapenem-resistant gramnegative bacteria isolated from urinary track source: SIDERO-CR-2014/2016. In: IDWeek, San Diego; 2017. 1199

Tsuji M, Hackel M, Yamano Y, Echols R, Longshaw C, Manissero D, et al. Cefiderocol in vitro activity against Gram-negative clinical isolates collected in Europe: result from SIDERO-CR-2014/2016. In: ECCMID, 2019.

Yamano Y, Tsuji M, Hackel M, Echols R, Sahm D. In vitro activity of cefiderocol against gram-negative clinical isolates collected from Asia and South Pacific in 2014-2016 (SIDERO-CR Study). Int J Antimicrob Agents. 2017;50(Supplement 2):S48.

Yamano Y, Tsuji M, Hackel M, Echols R, D S. In vitro activity of cefiderocole against globally collected carbapenem-resistant Gramnegative bacteria including isolates resistant to ceftazidime/avibactam, ceftolozane/tazobactam and colistin: SIDERO-CR-2014/2016 study. In: ECCMID, 2017. Independent Kresken M, Berwian E, Wernicke S, Frank A, G A. In vitro activity of cefiderocol against gram-negative Completed Studies pathogens circulating in Germany. In: ECCMID, 2019.

Falagas ME, Skalidis T, Vardakas KZ, Legakis NJ, Hellenic Cefiderocol Study G. Activity of cefiderocol (S- 649266) against carbapenem-resistant gram-negative bacteria collected from inpatients in Greek hospitals. J Antimicrob Chemother. 2017;72(6):1704-08.

Falagas ME, Skalidis T, Vardakas K, N L. Activity of cefiderocol (S-649266) against carbapenem-resistant gram- negative bacteria collected from inpatients in Greek hospitals. In: ECCMID, 2017.

Tsuji M, Yamaguchi T, Nakamura R, Kanazawa S, Ito-Horiyama T, Sato T, et al. S-649266, a novel siderophore cephalosporin: in vitro activity against gram-negative bacteria isolated in Japan including carbapenem resistant strains. In: IDWeek, San Diego; 2015.

Hackel M, Tsuji M, Echols R, D S. In vitro antibacterial activity of s-649266 against gram-negative clinical strains collected in North America and Europe. In: IDWeek, 2016.

Hsueh SC, Huang YT, Liao CH, Lee YJ, Hsueh PR. In vitro activities of cefiderocol (S-649266), ceftazidimeavibactam, ceftolozane-tazobactam, and other comparative drugs against pseudomonas aeruginosa, acinetobacter baumannii and stenotrophomonas maltophilia associated with bloodstream infections. Int J Antimicrob Agents. 2017;50(Supplement 2):S48-S49.

Hsueh S-C, Lee Y-J, Huang Y-T, Liao C-H, Tsuji M, Hsueh P-R. In vitro activities of cefiderocol, ceftolozane/tazobactam, ceftazidime/avibactam and other comparative drugs against imipenem-resistant pseudomonas aeruginosa and acinetobacter baumannii, and stenotrophomonas maltophilia, all associated with bloodstream infections in Taiwan. J Antimicrob Chemother. 2019;74(2):380-86.

Mushtaq S, Sadouki Z, Vickers A, Livermore D, N W. In vitro activity of cefiderocol against extensively drug- resistant pseudomonas aeruginosa and acinetobacter baumannii from the UK. In: ECCMID, 2019.

Mushtaq S, Vickers A, Hussain A, Livermore D, N W. In vitro activity of cefiderocol (S-649266) against multidrug- resistant enterobacteriaceae from the UK. In: ECCMID, 2017. Other Mode of Action Coppi M, Antonelli A, Baglio G, Giani T, GM R. Activity of cefiderocol on a reference collection of Completed carbapenemase-,ESBL-, and acquired class C β-lactamases-producing gram-negative pathogens. In: ECCMID, 2019.

Dobias J, Denervaud V, Poirel L, P N. Activity of the novel siderophore cephalosporin cefiderocol (S-649266) against Gramnegative pathogens. In: ECCMID, 2017.

Hackel MA, Tsuji M, Yamano Y, Echols R, Karlowsky JA, Sahm DF. Reproducibility of broth microdilution MICs for the novel siderophore cephalosporin, cefiderocol, determined using iron-depleted cation-adjusted Mueller- Hinton broth. Diagn Microbiol Infect Dis. 2019;94(4):321-25.

Huband MD, Ito A, Tsuji M, Sader HS, Fedler KA, Flamm RK. Cefiderocol mic quality control ranges in iron- depleted cation-adjusted mueller-hinton broth using a clsi m23-a4 multi-laboratory study design. Diagn Microbiol Infect Dis. 2017;88(2):198-200.

Ito A, Kohira N, Bouchillon SK, West J, Rittenhouse S, Sader HS, et al. In vitro antimicrobial activity of S-649266, a catechol-substituted siderophore cephalosporin, when tested against non-fermenting gram-negative bacteria. J Antimicrob Chemother. 2016;71(3):670-7.

Ito A, Nishikawa T, Matsumoto S, Fukuhara N, Nakamura R, Tsuji M, et al. S-649266, a novel siderophore cephalosporin: II. Impact of active transport via iron regulated outer membrane proteins on resistance selection. Abstract F-1563. In: ICAAC, 2014.

Ito A, Nishikawa T, Matsumoto S, Yoshizawa H, Sato T, Nakamura R, et al. Siderophore cephalosporin cefiderocol utilizes ferric iron transporter systems for antibacterial activity against pseudomonas aeruginosa. Antimicrob Agents Chemother. 2016;60(12):7396-401.

Ito A, Sato T, Ota M, Takemura M, Nishikawa T, Toba S, et al. In vitro antibacterial properties of cefiderocol, a novel siderophore cephalosporin, against gram-negative bacteria. Antimicrob Agents Chemother. 2018;62(1):1- 11.

Ito A, Ishibashi N, Ishii R, Tsuji M, Maki H, Sato T, et al. Changes of responsible iron-transporters for the activity of cefiderocol against pseudomonas aeruginosa depending on the culture conditions. In: ASM Microbe 2019, San Francisco; 2019.

Ito A, Nishikawa T, Ishii R, Kuroiwa M, Ishioka Y, Kurihara N, et al. Mechanism of Cefiderocol high MIC mutants obtained in non-clinical FoR studies In: IDWeek, 2018.

Luscher A, Moynie L, Auguste PS, Bumann D, Mazza L, Pletzer D, et al. TonB-dependent receptor repertoire of pseudomonas aeruginosa for uptake of siderophore-drug conjugates. Antimicrob Agents Chemother. 2018;62(6):1-11.

Nordmann P, Vazquez-Rojo L, L P. Stability of cefiderocol against clinically-significant broad-spectrum oxacillinases. In: ECCMID, 2018.

Poirel L, Kieffer N, Nordmann P. Stability of cefiderocol against clinically significant broad-spectrum oxacillinases. Int J Antimicrob Agents. 2018;52(6):866-67.

Robertson G, Henderson A, Paterson DL, P H. In vitro activity of cefiderocol (S-649266) against clinical isolates of burkholderia pseudomallei. In: ECCMID, 2019.

Takemura M, Ito A, Nishikawa T, Oota M, Horiyama T, Satou T, et al. Stability and low induction potential of cefiderocol against chromosomal AmpC β-lactamases of pseudomonas aeruginosa and enterobacter cloacae. In: ECCMID, 2018.

Tsuji M, Hackel M, Yamano Y, Echols R, D S. Correlations between cefidercol broth microdilution MICs and disk diffusion inhibitory zone diameters among target gram-negative organisms. In: ECCMID, 2018.

Tsuji M, Ito A, Toba S, Nishikawa T, Oota M, Kanazawa S, et al. S-649266, a novel siderophore cephalosporin: binding affinity to PBP and in vitro bactericidal activity. In: ECCMID, Copenhagen; 2015.

Tsuji M, Jakielaszek C, LCL M. S-649266, a novel siderophore cephalosporin: in vitro activity against biothreat pathogen. In: IDWeek, 2016.

Tsuji M, Nakamura R, Kohira N, Ito A, Sato T, Y Y. S-649266, a novel siderophore cephalosporin: in vitro combination effect of S-649266 and other antibacterials against gram-negative bacteria. In: ECCMID, 2016.

Rolston KVI, Gerges B, Raad I, Aitken SL, Reitzel R, R P. In vitro activity of cefiderocol and comparator agents against gramnegative isolates from cancer patients. In: IDWeek, 2018.

Ito A, Ota M, Nakamura R, Tsuji M, Sato T, Y Y. In vitro and in vivo activity of cefiderocol against stenotrophomonas maltophilia clinical isolates. In: IDWeek, 2018.

Tsuji M, Ito-Horiyama T, Nakamura R, Sato T, Y Y. S-649266, a Novel Siderophore Cephalosporin: Pharmacodynamic assessment by using MIC in Iron-depleted Cation-adjusted Mueller Hinton Broth (ID- CAMHB) In: IDWeek, 2016.

Tsuji M, Nakamura R, Sato T, Hackel M, Sahm D, Echols R, et al. Use of iron depleted mueller hinton broth (IDMHB) for microdilution testing of S649266, a novel siderophore cephalosporin. In: ECCMID, 2016.

Ito A, Miyagawa S, Ishibashi N, Nishikawa T, Kohira N, Sato T, et al. Anti-Acinetobacter activity of cefiderocol (S-649266), a novel siderophore cephalosporin: bactericidal activity due to penicillin-binding proteins inhibition and antibacterial activity against globally collected clinical isolates, abstr SATURDAY-115. In: ASM Microbe 2017, New Orleans; 2017. Resistance Aoki T, Yoshizawa H, Yamawaki K, Yokoo K, Sato J, Hisakawa S, et al. Cefiderocol (S-649266), A new Completed siderophore cephalosporin exhibiting potent activities against pseudomonas aeruginosa and other gram-

negative pathogens including multi-drug resistant bacteria: structure activity relationship. Eur J Med Chem. 2018;155:847-68.

Aoki T, Yoshizawa H, Yamawaki K, Yokoo K, Sato J, Hisakawa S, et al. Cefiderocol (S-649266): A new siderophore cephalosporin exhibiting potent activities against pseudomonas aeruginosa and other gram negative-pathogens including multi-drug resistant bacteria: structure activity relationship. Abstr. Pap. Am. Chem. Soc. 2017;254

Boyd S, Anderson K, Albrecth V, Campbell D, Karlsson MS, JK R. In vitro activity of cefiderocol against multidrug resistant carbapenemase-producing gramnegative pathogens. In: IDWeek, 2017.

Delgado-Valverde M, Conejo MC, Serrano-Rocha L, Fernandez-Cuenca F, AP H. Activity of cefiderocol (S- 649266), a new siderophore cephalosporin, against multidrug-resistant Enterobacteriaceae, Acinetobacter baumannii, Pseudomonas aeruginosa and Stenotrophomonas maltophilia. In: ECCMID, 2019.

Dobias J, Denervaud-Tendon V, Poirel L, Nordmann P. Activity of the novel siderophore cephalosporin cefiderocol against multidrug-resistant Gram-negative pathogens. Eur J Clin Microbiol Infect Dis. 2017;36(12):2319-27.

Ishii Y, Horiyama T, Nakamura R, Fukuhara N, Tsuji M, Yamano Y, et al. S-649266, a novel siderophore cephalosporin: III. Stability against clinically relevant β-lactamases. In: ICAAC, 2014. 1557

Ito A, Nishikawa T, Ota M, Ito-Horiyama T, Ishibashi N, Sato T, et al. Stability and low induction propensity of cefiderocol against chromosomal AmpC β-lactamases of pseudomonas aeruginosa and enterobacter cloacae. J Antimicrob Chemother. 2018;73(11):3049-52.

Ito-Horiyama T, Ishii Y, Ito A, Sato T, Nakamura R, Fukuhara N, et al. Stability of novel siderophore cephalosporin S-649266 against clinically relevant carbapenemases. Antimicrob Agents Chemother. 2016;60(7):4384-6. Jacobs M, Bbajaksouzian S, Good C, Hujer K, Hujer A, R B. In vitro activity of cefiderocol (S-649266), a siderophore cephalosporin, against carbapenem-susceptible and resistant non-fermenting gram-negative bacteria. In: ECCMID, 2018.

Jacobs MR, Abdelhamed AM, Good CE, Rhoads DD, Hujer KM, Hujer AM, et al. ARGONAUT-I: activity of cefiderocol (S-649266), a siderophore cephalosporin, against gram-negative bacteria, including carbapenem- resistant nonfermenters and enterobacteriaceae with defined extended-spectrum β-lactamases and carbapenemases. Antimicrob Agents Chemother. 2019;63(1):1-9.

Jacobs MR, Abdelhamed AM, Good CE, Rhoads DD, Hujer KM, Hujer AM, et al. In vitro activity of cefiderocol (S-649266), a siderophore cephalosporin, against enterobacteriaceae with defined extended-spectrum β- lactamases and carbapenemases. In: IDWeek, 2018.

Kanazawa S, Sato T, Kohira N, Ito-Horiyama T, Tsuji M, Yamano Y. Susceptibility of imipenem-susceptible but meropenem-resistant blaIMP-6-carrying enterobacteriaceae to various antibacterials, including the siderophore cephalosporin cefiderocol. Antimicrob Agents Chemother. 2017;61(7):1-3.

Kohira N, West J, Ito A, Ito-Horiyama T, Nakamura R, Sato T, et al. In vitro antimicrobial activity of a siderophore cephalosporin, S-649266, against enterobacteriaceae clinical isolates, including carbapenem-resistant strains. Antimicrob Agents Chemother. 2016 ;60(2) :729-34.

Kohira N, Nakamura R, Ito A, Nishikawa T, Ota M, Sato T, et al. Resistance acquisition studies of cefderocol by serial passage and in vitro pharmacodynamic model under human simulated exposure. In: ASM Microbe, Atlanta; 2018.

Kohira N, Oota M, Nishikawa T, Kuroiwa M, Ishioka Y, Naoko K, et al. Frequency of resistance acquisition and resistance mechanisms to cefiderocol. In: ASM Microbe, Atlanta; 2018.

Tsuji M, Ito A, Nakamura R, Yamano Y, J S. S-649266, a novel siderophore cephalosporin: in vitro activity against gram-negative bacteria including multidrug-resistant strains. In: IDWeek, 2014. 252

Tsuji M, Kazmierczak K, Hackel M, Echols R, Yamano Y, D S. Cefiderocol (S-649266) susceptibility against globally isolated meropenem non-susceptible gram-negative bacteria containing serine- and metallo- carbapenemase genes. In: ASM Microbe, New Orleans; 2017.

Tsuji M, Matsumoto S, Kanazawa S, Nakamura R, Sato T, Y Y. S-649266, a novel siderophore cephalosporin: in vitro activities against multidrug-resistant and carbapenem resistant gram-negative pathogens in an in vitro pharmacodynamic model. In: IDWeek, San Diego; 2015. 767 Yamano Y, Tsuji M, Hackel M, Sahm D, R E. Mode of action of cefiderocol, a novel siderophore cephalosporin, active against highly resistant gram-negative bacteria including carbapenem-resistant strains of Enterobacteriaceae and non-fermenting bacteria. In: ECCMID, 2017.

Tsuji M, Hackel M, Echols R, Tillotson G, Fam D, Yamano Y, et al. In vitro and in vivo activity of cefiderocol against burkholderia cepacia complex clinical isolates. In: ECCMID, 2019.

Tsuji M, Hackel M, Echols R, Yamano Y, D S. The in vitro activity of cefiderocol, a novel siderophore cephalosporin, against a global collection of stenotrophomonas maltophilia. In: ECCMID, 2017.

Tsuji M, Hackel M, Yamano Y, Echols R, D S. Cefiderocol, a novel siderophore cephalosporin: in vitro activity against Stenotrophomonos maltophillia isolated globally. In: ECCMID, 2018.

A. Ito, H. Yoshizawa, R. Nakamura, M. Tsuji, Y. Yamano, Shimada J. S-649266, a novel siderophore cephalosporin: I. In vitro activity against gram-negative bacteria including multidrug-resistant strains. ICAAC Abstracts. 2014 PK/PD data Animal models Thigh Monogue ML, Tsuji M, Yamano Y, Echols R, Nicolau DP. Efficacy of humanized exposures of cefiderocol (S- Completed 649266) against a diverse population of gram-negative bacteria in a murine thigh infection model. Antimicrob Agents Chemother. 2017;61(11):1-10.

Nakamura R, Toba S, Ito A, Tsuji M, Yamano Y, Shimada J. S-649266, a novel siderophore cephalosporin. V. Pharmacodynamic assessment in murine thigh infection models, abstr F1-1559. In: ICAAC, 2014.

Ghazi IM, Monogue ML, Tsuji M, Nicolau DP. Pharmacodynamics of cefiderocol, a novel siderophore cephalosporin, in a pseudomonas aeruginosa neutropenic murine thigh model. Int J Antimicrob Agents. 2018;51(2):206-12.

Horiyama T, Toba S, Nakamura R, Tsuji M, Yamano Y, Shimada J. S-649266, a novel siderophore cephalosporin: vii. Magnitude of pk/pd parameter required for efficacy in murine thigh infection model. In: ICAAC, 2014. 1561

Monogue M, Tsuji M, Yamano Y, Echols R, DP N. In vivo efficacy of humanized exposures of cefiderocol compared with cefepime (FEP) and meropenem (MEM) against Gram-negative bacteria in a murine thigh model. In: IDWeek, 2017.

Stainton S, Monogue M, Tsuji M, Yamano Y, Echols R, DP N. Efficacy of humanized cefiderocol exposures over 72 hours against a diverse group of gram-negative isolates in the neutropenic murine thigh infection model. In: IDWeek, 2018.

Stainton SM, Monogue ML, Tsuji M, Yamano Y, Echols R, Nicolau DP. Efficacy of humanized cefiderocol exposures over 72 hours against a diverse group of gram-negative isolates in the neutropenic murine thigh infection model. Antimicrob Agents Chemother. 2019;63(2):1-7. Lung Horiyama T, Singley CM, Nakamura R, Tsuji M, Roger E, Rittenhouse S, et al. S-649266, a novel siderophore Completed cephalosporin: viii. Efficacy against pseudomonas aeruginosa and acinetobacter baumannii in rat lung infection model with humanized exposure profile of 2 gram dose with 1 hour and 3 hours infusion. In: ICAAC, 2014. 1556

Matsumoto S, Singley CM, Hoover J, Nakamura R, Echols R, Rittenhouse S, et al. Efficacy of cefiderocol against carbapenem-resistant gram-negative bacilli in immunocompetent-rat respiratory tract infection models recreating human plasma pharmacokinetics. Antimicrob Agents Chemother. 2017;61(9):1-8.

Takemura M, Matsumoto S, Miyagawa S, Satou T, Tsuji M, Y Y. Efficacy of humanized cefiderocol exposure against Stenotrophomonas maltophilia in a rat respiratory tract infection mode. In: ECCMID, 2018.

Takemura M, Nakamura R, Satou T, Tsuji M, Y Y. In vivo pharmacokinetic/pharmacodynamic (PK/PD) assessment of cefiderocol against Stenotrophomonas maltophilia in a neutropenic murine lung infection model. In: ECCMID, 2018. Lung & Thigh Horiyama T, Toba S, Nakamura R, Tsuji M, Yamano Y, Shimada J. S-649266, a novel siderophore Completed cephalosporin: vi. Magnitude of pk/pd parameter required for efficacy in murine lung infection model. In: ICAAC, 2014. 1560

Nakamura R, Ito-Horiyama T, Takemura M, Toba S, Matsumoto S, Ikehara T, et al. In vivo pharmacodynamic study of cefiderocol, a novel parenteral siderophore cephalosporin, in murine thigh and lung infection models. Antimicrob Agents Chemother. 2019;63(9)

Yamano Y, Nakamura R, Sato T, Tsuji M, R E. Good correlation of cefiderocol between in vivo efficacy murine thigh/lung infection models and mic determined in iron-depleted conditions. In: IDWeek, 2017. Kidney Matsumoto S, Kanazawa S, Nakamura R, Tsuji M, Sato T, Y Y. In vivo efficacy of cefiderocol against Completed carbapenem-resistant gram-negative bacilli in murine urinary tract infection models. In: IDWeek, 2017. Multiple Nakamura R, Toba S, Tsuji M, Yamano Y, Shimada J. S-649266, a novel siderophore cephalosporin: IV. In vivo Completed efficacy in various murine infection models. In: ICAAC, 2014. 1558

Ito A, Ota M, Nakamura R, Tsuji M, Sato T, Y Y. In vitro and in vivo activity of cefiderocol against stenotrophomonas maltophilia clinical isolates. In: IDWeek, 2018.

Tsuji M, Hackel M, Echols R, Tillotson G, Fam D, Yamano Y, et al. In vitro and in vivo activity of cefiderocol against burkholderia cepacia complex clinical isolates. In: ECCMID, 2019. Human Thigh Ghazi IM, Monogue ML, Tsuji M, Nicolau DP. Humanized exposures of cefiderocol, a siderophore Completed cephalosporin, display sustained in vivo activity against siderophore-resistant pseudomonas aeruginosa. Pharmacology. 2018;101(5-6):278-84. Lung Shionogi. Cefiderocol Concentrations in the Lungs of Hospitalized Patients With Bacterial Pneumonia. Identifier: Terminated due to NCT03862040. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2019. Available from slow enrolment on https://ClinicalTrials.gov/show/NCT03862040. November 26, 2019 Kidney Contreras DA, Fitzwater SP, Nanayakkara DD, Schaenman J, Aldrovandi GM, Garner OB, et al. Co-infections of two strains of NDM-1 and OXA-232 co-producing klebsiella pneumoniae in a kidney transplant patient. Antimicrob Agents Chemother. 2019;16:1-12.

Echols R, Katsube T, F A, C J, Krenz H. S-649266, a siderophore cephalosporin for Gram-negative bacterial infection: pharmacokinetics and safety in subjects with renal impairment. In: ESCMID, Denmark; 2015.

Katsube T, Echols R, Arjona Ferreira JC, Krenz HK, Berg JK, Galloway C. Cefiderocol, a siderophore cephalosporin for gram-negative bacterial infections: pharmacokinetics and safety in subjects with renal impairment. J Clin Pharmacol. 2017;57(5):584-91.

Katsube T, Kawaguchi N, Echols R, Wajima T. Population pharmacokinetic and pharmacokinetic/pharmacodynamic analyses of cefiderocol in subjects without infection and patients with complicated urinary tract infection and acute uncomplicated pyelonephritis. In: IDWeek, 2017.

Katsube T, Miyazaki S, Narukawa Y, Hernandez-Illas M, Wajima T. Drug-drug interaction of cefiderocol, a siderophore cephalosporin, via human drug transporters. Eur J Clin Pharmacol. 2018;74(7):931-38. Clinical Studies Early Katsube T, Saisho Y, Shimada J, Furuie H. Intrapulmonary pharmacokinetics of cefiderocol, a novel siderophore Completed cephalosporin, in healthy adult subjects. J Antimicrob Chemother. 2019;74(7):1971-74.

Miyazaki S, Katsube T, Shen H, Tomek C, Narukawa Y. Metabolism, excretion, and pharmacokinetics of [14 C]- cefiderocol (S-649266), a siderophore cephalosporin, in healthy subjects following intravenous administration. J Clin Pharmacol. 2019;59(7):958-67.

Saisho Y, Katsube T, White S, Fukase H, Shimada J. Pharmacokinetics, safety, and tolerability of cefiderocol, a novel siderophore cephalosporin for gram-negative bacteria, in healthy subjects. Antimicrob Agents Chemother. 2018;62(3):1-12.

Sanabria C, Migoya E, Mason JW, Stanworth SH, Katsube T, Machida M, et al. Effect of cefiderocol, a siderophore cephalosporin, on QT/QTc interval in healthy adult subjects. Clin Ther. 2019;41(9):1724-36.e4.

Shimada J, Saisho Y, Katsube T, White S, Fukase H. S-649266, a novel cephalosporin for gram negative bacterial infection: pharmacokinetics (PK), safety and tolerability in healthy subjects. In: ICAAC, 2014 2014. F- 1564

Shiro M, Katsube T, Narukawa Y, E M. Metabolism and excretion of [14C]-cefiderocol, a siderophore cephalosporin, and drug-drug interaction potential via transporters of cefiderocol in healthy subjects. In: ECCMID, 2018.

Kawaguchi N, Katsube T, Echols R, Wajima T. Population pharmacokinetic analysis of cefiderocol, a parenteral siderophore cephalosporin, in healthy subjects, subjects with various degrees of renal function, and patients with complicated urinary tract infection or acute uncomplicated pyelonephritis. Antimicrob Agents Chemother. 2018;62(2):1-11.

RCT Results Blind Bass A, Echols R, Portsmouth S, A H. Heterogeneity of recent phase 3 cUTI clinical trials with new antibacterials. Completed In: IDWeek 2018.

Portsmouth S, van Veenhuyzen D, Echols R, Machida M, Ferreira JCA, Ariyasu M, et al. Cefiderocol versus imipenem-cilastatin for the treatment of complicated urinary tract infections caused by gram-negative uropathogens: a phase 2, randomised, double-blind, non-inferiority trial. Lancet Infect Dis. 2018;18(12):1319- 28.

Portsmouth S, van Veenhuyzen D, Echols R, Mitsuaki M, Camilo Arjona Ferreira J, Ariyasu M, et al. Cefiderocol compared with imipenem/cirastatin in the treatment of adults with complicated urinary tract infections with or without pyelonephritis or acute uncomplicated pyelonephritis: results from a multicenter, double-blind, randomized study. In: ECCMID, 2017.

Portsmouth S, Van Veenhuyzen V, Echols R, Machida M, Arjona Ferreira JC, Ariyasu M, et al. Clinical response of cefiderocol compared with imipenem/cilastatin in the treatment of adults with complicated urinary tract

infections with or without pyelonephritis or acute uncomplicated pyelonephritis: results from a multicenter, double blind, randomized study (APEKS-cUTI). In: IDWeek, 2017.

Shionogi. A randomized study in hospitalised patients with complicated urinary tract infections caused by gram- negative bacteria to compare the efficacy and safety of S-649266 to imipenem/cilastin, both administered by intravenous infusion. Identifier: EUCTR2014-000914-76-HU. In: London [internet]. European Medicines Agency: 2014. Available from http://www.who.int/trialsearch/Trial2.aspx?TrialID=EUCTR2014-000914-76-HU.

Shionogi. A Study of Efficacy/Safety of Intravenous S-649266 Versus Imipenem/Cilastatin in Complicated Urinary Tract Infections. Identifier: NCT02321800. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2014. Available from https://ClinicalTrials.gov/show/NCT02321800.

Matsunaga Y, Echols R, Katsube T, Yamano Y, Ariyasu M, Nagata T. Cefiderocol (S-649266) for nosocomial pneumonia caused by gram-negative pathogens: study design of apeks-NP, a phase 3 double-blind parallel- group randomized clinical trial. Am J Respir Crit Care Med. 2018;197

Shionogi. Clinical Study of S-649266 for the Treatment of Nosocomial Pneumonia Caused by Gram-negative Pathogens. Identifier: NCT03032380. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2017. Available from https://ClinicalTrials.gov/show/NCT03032380. Open-label Shionogi. Study of S-649266 or Best Available Therapy for the Treatment of Severe Infections Caused by Completed Carbapenem-resistant Gram-negative Pathogens. Identifier: NCT02714595. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2016. Available from https://ClinicalTrials.gov/show/NCT02714595. Recruiting Shionogi. RCT Cefiderocol vs BAT for Treatment of Gram Negative BSI. Identifier: NCT03869437. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2019. Available from https://ClinicalTrials.gov/show/NCT03869437.

Compassionate Edgeworth JD, Merante D, Patel S, Young C, Jones P, Vithlani S, et al. Compassionate use of cefiderocol as Ongoing program. Use adjunctive treatment of native aortic valve endocarditis due to extremely drug-resistant pseudomonas 200 patients aeruginosa. Clin Infect Dis. 2019;68(11):1932-34. treated to date. Three case reports Stevens RW, Clancy M. Compassionate use of cefiderocol in the treatment of an intraabdominal infection due published. to multidrug resistant pseudomonas aeruginosa: a case report. Pharmacotherapy. 2019;24:1-14.

Trecarichi EM, Quirino A, Scaglione V, Longhini F, Garofalo E, Bruni A, et al. Successful treatment with cefiderocol for compassionate use in a critically ill patient with XDR acinetobacter baumannii and KPC-producing klebsiella pneumoniae: a case report. J Antimicrob Chemother. 2019;01:3399–401.

Expanded Access Shionogi. Expanded Access to Cefiderocol for the Intravenous Treatment of Severe Gram Negative Bacterial Ongoing Infections. Identifier: NCT03780140. In: ClinicalTrials.gov [internet]. Bethesda: US National Library of Medicine: 2018. Available from https://ClinicalTrials.gov/show/NCT03780140. Modelling/SLR protocol Katsube T, Tenero D, Wajima T, T I. S-649266 modeling and simulation for prediction of efficacy and dose Completed optimization. In: IDWeek, Philadelphia; 2014.

Katsube T, Wajima T, Ishibashi T, Arjona Ferreira JC, Echols R. Pharmacokinetic/pharmacodynamic modeling and simulation of cefiderocol, a parenteral siderophore cephalosporin, for dose adjustment based on renal function. Antimicrob Agents Chemother. 2017;61(1):1-12.

Katsube T, Wajima T, Ishibashi T, Arjona Ferreira JC, R E. S-649266 dose adjustment for patients with impaired renal function. In: ECCMID, 2015.

Katsube T, Wajima T, Ishibashi T, Arjona Ferreira JC, R. E. Dose adjustment of S-649266, a siderophore cephalosporin, for patients requiring haemodialysis. . In: ECCMID, 2016.

McCullough A, Scott AM, Macindoe C, Clark J, Hansen MP, Beller EM, et al. Adverse events in patients taking cephalosporins versus placebo for any indication. Cochrane Database Syst Rev. 2016(11) IDweek 2019 Jason M. Pogue, Hemanth Kanakamedala, Yun Zhou, Bin Cai. Burden of Illness in Carbapenem-resistant Acinetobacter baumannii Infections in US Hospitals (2014 to 2018)

Ryan K. Shields, Hemanth Kanakamedala, Yun Zhou, Bin Cai. Burden of Illness in Patients with Urinary Tract Infections with or without Bacteremia Caused by Carbapenem-resistant Gram-negative Pathogens in US Hospitals (2014 to 2018)

Thomas Lodise, Hemanth Kanakamedala, Wen-Chun Hsu, Bin Cai. Association between Days to Initiate Appropriate Therapy and Hospital Length of Stay among Adult Hospitalized Patients with Gram-negative Bloodstream Infections (GN-BSI)

Thomas Lodise, Hemanth Kanakamedala, Wei-Chun Hsu, Bin Cai. Analysis of Adult, Hospitalized Patients with Carbapenem-resistant (CR) Gram-negative Bloodstream Infections (GN-BSIs) due to Lactose Fermenters (LFs) and Non-Lactose Fermenters (NLFs): Is there a Difference in Outcomes?

Simon Portsmouth, Roger Echols, Mitsuaki Machida, Juan Camilo Arjona Ferreira, Mari Ariyasu, Tsutae Den Nagata. Efficacy and Safety of Cefiderocol According to Renal Impairment in Patients with Complicated Urinary Tract Infection (cUTI) in a Phase 2 Study

Simon Portsmouth, Kiichiro Toyoizumi, Tsutae Den Nagata, Glenn S. Tillotson, Roger Echols. Structured Patient Interview in Complicated Urinary Tract Infections to Assess Clinical Outcomes versus Investigator’s Evaluation in the APEKS-cUTI Study.

Takafumi Sato, Masakatsu Tsuji, Krystyna M. Kazmierczak, Meredith Hackel, Roger Echols, Yoshinori Yamano, Daniel F Sahm. Cefiderocol Susceptibility Against Molecularly Characterized Carbapenemase-Producing Gram- negative Bacteria in North America and Europe between 2014 and 2017: SIDERO-WT-2014 to -2016 Studies.

Sonia Rao, Sean Nguyen, Melinda Soriano, Jennifer Hayes, Meredith Hackel, Daniel Sahm, Glenn Tillotson, Roger Echols, Masakatsu Tsuji, Yoshinori Yamano. In Vitro Antibacterial Activity of Cefiderocol against a Multi- national Collection of Carbapenem-non-susceptible Gram-negative Bacteria from Respiratory Infections: SIDERO-WT-2014–2017.

Chris Longshaw, Masakatsu Tsuji, Meredith Hackel, Daniel F. Sahm, Yoshinori Yamano. In Vitro Activity of Cefiderocol (CFDC), a Novel Siderophore Cephalosporin, Against Difficult-to-Treat Resistant (DTR) Gram- negative Bacterial Pathogens from the Multi-national Sentinel Surveillance Study, SIDERO-WT (2014–2017).

Yoshinori Yamano, Masakatsu Tsuji, Roger Echols. Synergistic Effect of Cefiderocol Combined with Other Antibiotics Against Cefiderocol High MIC Isolates from the Multi-national SIDERO-WT Studies.

Takayuki Katsube, Roger Echols, Toshihiro Wajima. Prediction of Cefiderocol Pharmacokinetics and Probability of Target Attainment in Pediatric Subjects for Proposing Dose Regimens.

Richard G. Wunderink (Presenting Author), Yuko Matsunaga, Mari Ariyasu, Roger Echols, Anju Menon, Tsutae Den Nagata. Efficacy and Safety of Cefiderocol versus High-Dose Meropenem in Patients with Nosocomial Pneumonia – Results of a Phase 3 Randomized, Multicenter, Double-Blind, Non-Inferiority Study. (Late- breaking abstract).

*Include references to all linked documents and indicate the expected date of publication for any unpublished clinical studies **Include expected date of completion

5.3 Main characteristics of studies

For a rigorous comprehension of the evidence package, it is fundamental to understand the challenges associated to the design of clinical trials to assess the efficacy of antibacterials. The development of clinical trials can be challenging for a number of reasons [234, 235]:

. Clinical trials for antimicrobials must be designed as non-inferior studies. The clinical studies must focus on specific pathogens for which the tested agents and comparators are effective; otherwise, they would be un-ethical. . For serious bacterial diseases, there is a need to urgently initiate early targeted antibacterial drug therapy, which may obscure the effect of the antibacterial drug under study because patients receive effective antibacterial therapy before enrolling in the trial. . Patients with serious acute bacterial diseases can be acutely ill (e.g., delirium in the setting of acute infection) and obtaining informed consent and performing other trial enrolment procedures in a timely fashion may be difficult. . There may be diagnostic uncertainty with respect to the aetiology of the patients’ underlying disease, including identifying a bacterial aetiology. . There may be a need for concomitant antibacterial drug therapy with a spectrum of activity that may overlap with the antibacterial drug being studied. . The recruitment of patients with infections due to specific pathogens and with limited treatment options that would be required for inferential testing is challenging. MDR/CR pathogens are still rare, and this rigorous selection strategy must be applied to include patients with resistant pathogens. Otherwise large patient numbers and subgroup analyses are required. . A comparison of efficacy against all relevant comparators can only be obtained from in vitro surveillance studies. The evaluation of the effectiveness of an antibacterial is derived from the combined consideration of in vitro, PK/PD and clinical data.

1. In Table 17, describe the main characteristics of the studies.

2. For each study provide a flow diagram of the numbers of patients moving through the trial.

3. For each study provide a comparison of patients (including demographic, clinical and social information [if applicable]) in treatment arms at baseline.

Cefiderocol studies are summarized with patient flow diagram and comparison of patients after Table 17.

Table 17: Study characteristics

Study Objective Study design Eligibility criteria Intervention and Primary outcome Secondary reference/ID Comparator measure and outcome (N enrolled) follow-up time measures and point follow-up time points SIDERO-WT To calculate the In vitro surveillance Study tested the in Cefiderocol, To determine the  Annual analysis; analysis* indices related to vitro antibacterial ceftazidime- minimum inhibitory  Cumulative the antibacterial activity of cefiderocol avibactam (CZA), concentration recurrent annual activity of against Gram-negative ceftolozane- (MIC) of analysis cefiderocol and the bacteria clinically tazobactam (C/T), cefiderocol against  Analysis in the ratio of susceptible isolated from medical colistin (CST), Gram-negative difficult to treat strains of institutions in the EU cefepime (FEP), bacteria clinically pathogens cefiderocol and and USA meropenem isolated from  Analysis of MDR other reference (MEM), and medical institutions 3 and MDR 4 compounds based ciprofloxacin (CIP) in the EU and USA pathogens on the breakpoint (30,459 Gram- criteria of Clinical negative isolates) + and Laboratory Standards Institute (CLSI) standards. SIDERO-CR-2014- To test the in vitro In vitro surveillance Study tested the in Cefiderocol, To determine the  Annual analysis; 2016 study* activity of vitro antibacterial ceftazidime- minimum inhibitory  Cumulative cefiderocol and activity of cefiderocol avibactam (CZA), concentration recurrent annual comparators against CRE and MDR ceftolozane- (MIC) of analysis

against a collection non-fermenters tazobactam (C/T), cefiderocol against  Analysis in the of 1,873 clinical (defined as resistant to colistin (CST), CRE and MDR difficult to treat isolates of Gram- carbapenems, cefepime (FEP), non-fermenters pathogens negative bacilli fluoroquinolones, and meropenem (defined as provided by a aminoglycosides) (MEM), and resistant to worldwide network collected globally ciprofloxacin (CIP) carbapenems, of laboratories (52 (1,873 MDR and fluoroquinolones, countries) in 2014- CarbNS isolates and 2016, using current Gram-negative aminoglycosides) CLSI broth Bacilli) + collected globally microdilution methodology

Several To determine In vitro surveillance to investigate the in Cefiderocol, To determine independent cefiderocol activity vitro antimicrobial ceftazidime- MIC50 and MIC90 national validations against difficult-to- activity of cefiderocol avibactam (CZA), of the studies* treat CR pathogens and that of ceftolozane- antibacterials for gathered from commercially available tazobactam (C/T), the tested bacterial various countries comparator colistin (CST), isolates and their including Germany, antibacterials against cefepime (FEP), respective Greece, Italy, a collection of meropenem resistance Spain, UK/Ireland, contemporary, clinical, (MEM), and percentages and the US carbapenem-resistant ciprofloxacin (CIP), Gram-negative tigecycline bacteria from

inpatients from various hospitals

Independent world- To determine In vitro To evaluate Cefiderocol, To determine wide collection* cefiderocol activity antimicrobial activity of aztreonam, MIC50 and MIC90 against difficult-to- cefiderocol and other amikacin, of the treat CR pathogens Gram-negative cefepime, antibacterials for gathered from a antibiotics (aztreonam, ceftazidime, the tested bacterial worldwide amikacin, cefepime, ceftazidime– isolates and their collection ceftazidime, avibactam, respective ceftazidime– ceftolozane– resistance avibactam, tazobactam, percentages ceftolozane– ciprofloxacin, tazobactam, meropenem, ciprofloxacin, colistin, and meropenem, colistin, tigecycline and tigecycline) against a panel of multidrug-resistant bacterial isolates from human clinical sources with characterized antibacterial resistance mechanisms.

Identify To investigate In vitro resistance To reveal cefiderocol Cefiderocol, To determine mechanisms of features of and mechanism of features relating to ceftazidime- MIC50 and MIC90 resistance studies* cefiderocol, namely action studies antibacterial activity avibactam (CZA), of the antibacterial against AmpC ceftolozane- antibacterials for activity against overproducers, tazobactam (C/T), the tested bacterial AmpC stability against AmpC colistin (CST), isolates and their overproducers, b-lactamases, and cefepime (FEP), respective stability against propensity for AmpC meropenem resistance AmpC b- induction for E. (MEM), and percentages lactamases and cloacae and P. ciprofloxacin (CIP), propensity for aeruginosa. tigecycline AmpC induction using Pseudomonas aeruginosa and Enterobacter cloacae.

Single ascending To evaluate the Randomized, Healthy adult subjects 70 subjects split Evaluate the dose safety, tolerability double-blind, into two study safety, tolerability, (SAD)/multiple and PK of placebo controlled, parts and PK dose (MAD) study cefiderocol in 70 ascending single

(1203R2111) healthy Japanese and multiple dose and Caucasian study adult subjects

The renal Evaluate the A multi-centre, Healthy adult 38 subjects Evaluate the impairment study influence of renal open-label, non- subjects and enrolled in 5 influence of renal (1222R2113) impairment and randomized study subjects with various cohorts impairment and hemo-dialysis on degrees of renal hemo-dialysis PK impairment onPK

APEKS-cUTI To compare the International, multi Adults (≥18 years) who Cefiderocol The primary Secondary composite outcome centre, had a symptomatic compared to efficacy endpoint endpoints were efficacy and safety randomised, cUTI, defined as a imipenem/cilastatin was the composite safety, clinical and of cefiderocol with double-blind, clinical syndrome (N=452 of clinical response microbiological IPM/CS in a subject Phase II, active- characterized by randomised 2:1 for and response at early population cUTI by controlled, parallel- pyuria and a cefiderocol) microbiological assessment, end of MDR Gram- group, non- documented or response at the treatment, and negative inferiority suspected microbial test of cure (TOC) follow-up, pathogens, with or pathogen on culture of assessment in microbiological and without urine or blood, MITT population, clinical response pyelonephritis or accompanied by local defined as 7 days per-pathogen and acute and systemic signs (±2 days) after the per-patient at early uncomplicated and symptoms, end of antibacterial assessment, end of pyelonephritis at including fever (i.e., treatment. treatment, test of the Test of Cure temperature ≥ 38ºC), cure, and follow-up. (TOC, defined as 7 chills, malaise, flank days following the pain, back pain, and / End of Treatment or costovertebral angle (EOT). pain or tenderness that (in the case of cUTI

with or without pyelonephritis) occurred in the presence of a functional or anatomical abnormality of the urinary tract or in the presence of catheterization and who required hospitalization for the IV treatment of cUTIs were enrolled in the study. Number of patients with acute uncomplicated pyelonephritis was restricted APEKS-NP To compare all- Phase 3, Adults (≥18 years) who Cefiderocol plus The primary Secondary cause mortality at multicentre have a documented linezolid, endpoint was all- endpoints included Day 14 of (multinational), nosocomial compared to high cause mortality at safety, clinical and cefiderocol with double-blind, pneumonia dose, prolonged Day 14 microbiological high-dose, parallel-group, (HABP/VABP/HCABP) infusion outcomes at the prolonged infusion randomized, caused by an aerobic meropenem (plus test of cure (TOC),

meropenem, in active-controlled Gram-negative linezolid for at least clinical and adults with study pathogen only, or in 5 days, and up to microbiological hospital-acquired combination with an 21 days) outcomes at early bacterial aerobic Gram-positive (N=300, assessment, end of pneumonia (HAP), or anaerobic randomised 1:1) treatment, and ventilator- pathogen, and who follow-up, all-cause associated require hospitalization mortality at day 28, bacterial for the parenteral during treatment, pneumonia (VAP), (intravenous) and at follow-up, or healthcare- treatment of the and resource associated infection may be utilization. bacterial enrolled in the study pneumonia (HCAP) caused by Gram- negative pathogens

CREDIBLE-CR The primary Phase 3, Adult patients (≥18 Cefiderocol The primary Secondary objective of descriptive, multi years) with gram- compared with endpoints were: endpoints included: CREDIBLE CR centre, open label, negative pathogen best available Clinical outcome Clinical cure per study was to parallel group, infection, with therapy (BAT) per patient at TOC in assess at TOC, the randomized study evidence of (N=152, patient/pathogen at patients with clinical outcome of carbapenem randomised 2:1 to EOT, and TOC HAP/VAP/HCAP treatment with resistance prior to cefiderocol); BAT (cUTI)Microbiologic or BSI/sepsis cefiderocol and randomisation was chosen by the outcome (for Gram- Microbiologic BAT in adult investigator before negative pathogen)

patient’s hospital randomization and eradication per per acquired could include up to patient at TOC in patient/pathogen at pneumonia 3 different patients with cUTI EOT, TOC, and (HAP)/ventilator medicines; FUP (HAP / VAP / associated cefiderocol could HCAP or pneumonia be added 1 other BSI/sepsis) (VAP)/healthcare- molecule Safety associated pneumonia (HCAP) or bloodstream infections/sepsis (BSI/sepsis) caused by carbapenem- resistant Gram- negative pathogens. Only descriptive statistics were performed. Compassionate Expanded Access Cefiderocol has The criteria for fulfilling Over 200 patients Clinical cure per use studies to Cefiderocol for been provided these requests are have been treated patient the Intravenous upon request from highly restrictive with cefiderocol Treatment of attending including that all other through this Severe Gram- physicians to available treatments programme patients with must be ruled out

negative Bacterial serious CR Gram- through susceptibility Infections negative infections testing and/or who have no other evidence of treatment treatment options failure in efficacy or safety, and patients must be unable to enrol in clinical studies of cefiderocol *Detailed comparisons of patients and patient flow diagrams are not available for in vitro studies +Ongoing studies, to date (January 2020), 38288 samples have been tested in SIDERO-WT program.

5.3.1 APEKS-cUTI STUDY

A Multicenter, Double-blind, Randomized, Clinical Study to Assess the Efficacy and Safety of Intravenous S-649266 (Cefiderocol) in Complicated Urinary Tract Infections With or Without Pyelonephritis or Acute Uncomplicated Pyelonephritis Caused by Gram-Negative Pathogens in Hospitalized Adults in Comparison With Intravenous Imipenem/Cilastatin.

APEKS-cUTI was an international, multicenter, randomised, double-blind, Phase II, active- controlled, parallel-group, non-inferiority study to investigate the efficacy and safety of intravenous cefiderocol vs imipenem/cilastatin (IPM/CS) in cUTI with or without pyelonephritis or acute uncomplicated pyelonephritis (restricted to ≤ 30%) caused by Gram-negative pathogens in hospitalized adults with MDR infections (Figure 22). 448 patients were randomized, of whom 300 received cefiderocol and 148 received IPM/CS (Figure 23). The primary efficacy endpoint was the composite of clinical response and microbiological response at TOC assessment in MITT population according to FDA guidance document. Secondary endpoints were safety, clinical and microbiological response at EA, EOT and FU, microbiological and clinical response per-pathogen and per-patient at EA, EOT, TOC and FU. Safety was assessed daily while the subject was hospitalized and specifically at EOT, TOC, and FU.

Figure 22: APEKS-cUTI study design

Figure 23: Subject disposition (all randomized subjects)

Overall, 495 patients were screened for inclusion. One subject withdrew during screening and 8.5% (42/495) of subjects were screen failures (mostly for lack of symptoms and signs confirming eligibility). The flow diagram of the numbers of patients moving through the trial is provided in Figure 23.

Of the 452 subjects randomized, 448 subjects were treated and 421 subjects completed the study: 93.4% (283/303) of subjects in the cefiderocol group and 92.6% (138/149) of subjects in the IPM/CS group. The most frequent reasons in the total population for discontinuing from the study were “lost to follow up” (3.1% [14/452]) and “withdrawal by subject” (1.3% [6/452]). By treatment group, 3.3% (10/303) of subjects in the cefiderocol group and 2.7% (4/149) of subjects in the IPM/CS group were “lost to follow up,” and 1.0% (3/303) of subjects in the

cefiderocol group and 2.0% (3/149) of subjects in the IPM/CS group did not complete the study due to “withdrawal by subject.”

Completion of treatment was defined as achieving 5 or more days of study treatment. The most frequent reasons for subjects not completing treatment (2.4% [11/452] in the total population) were withdrawal by subject (0.7% [2/303] of subjects in the cefiderocol group and 1.3% [2/149] of subjects in the IPM/CS group) and “other” (1.0% [3/303] of subjects in the cefiderocol group and 0.7% [1/149] of subjects in the IPM/CS group).

5.3.1.1 Demographics and baseline characteristics

The demographics and baseline characteristics are provided in the overview Table 18 below.

The mean age of the Micro-ITT Population was 62.0 years (range 18 to 93 years), and 55.0% (204/371) of subjects were ≥ 65 years. For clinical diagnosis at baseline, 25.3% (94/371) of subjects had cUTI with pyelonephritis, 47.7% (177/371) of subjects had cUTI without pyelonephritis, and 27.0% (100/371) of subjects had acute uncomplicated pyelonephritis. Complicated urinary tract infection was complicated most commonly by obstructive uropathy (33.2% [123/371] of subjects). For the severity of disease, 18.9% (70/371) of subjects had severe disease as judged by the investigator, and 71.2% (264/371) of subjects had moderate disease. The remaining subjects had mild disease. A greater proportion of subjects in the cefiderocol group (19.8% [50/252]) had severe disease compared with subjects in the IPM/CS group (16.8% [20/119]). This may be due to a greater proportion of subjects in the cefiderocol group with a diagnosis of cUTI compared with the IPM/CS group that had more acute uncomplicated pyelonephritis.

Prior infection history was reported for 40.4% (150/371) of subjects, and the most frequently reported prior infection was cUTI (29.9% [111/371] of subjects). Prior antimicrobial medication treatments were reported for 9.4% (35/371) of subjects, and most received treatment for UTI (7.5% [28/371] of subjects).

Baseline subject characteristics for the Micro-ITT Population were broadly similar to the ME Population, ITT Population, and Safety Population.

Table 18: Patient demographics and baseline characteristics (mITT population)

Patient baseline characteristics were generally well balanced between the 2 treatment arms (Table 18) and were consistent with more complicated infections whereby 7% of patients had BSI [51, 236]. The main pathogen reported at the baseline in the microbiologically evaluable population was E. coli (Figure 24) [51, 236]. The cefiderocol-treated group had 53% cefepime- and levofloxacin-resistant K. pneumoniae strains and 17% cefepime-resistant and 38% levofloxacin-resistant E. coli strains; which were similar in the IPM/CS group [51, 236].

Figure 24: Distribution of uropathogens (mITT population)

MITT, modified intent-to-treat. Source: Portsmouth, 2018 [51]; Data on file [236]

The most frequently reported Gram-negative uropathogens isolated at baseline for both cUTI with or without pyelonephritis or acute uncomplicated pyelonephritis in the Micro-ITT Population were E. coli (60.3% [152/252] of subjects in the cefiderocol group and 66.4% [79/119] of subjects in the IPM/CS group) and K. pneumoniae (19.0% [48/252] of subjects in the cefiderocol group and 21.0% [25/119] of subjects in the IPM/CS group) (Figure 24). In subjects diagnosed with cUTI with or without pyelonephritis, E. coli was isolated in 51.3% (96/187) of subjects in the cefiderocol group and 60.7% (51/84) of subjects in the IPM/CS group, and K. pneumoniae was isolated in 22.5% (42/187) of subjects in the cefiderocol group and 23.8% (20/84) of subjects in the IPM/CS group. In subjects diagnosed with acute uncomplicated pyelonephritis, E. coli was isolated in 86.2% (56/65) of subjects in the cefiderocol group and 80.0% (28/35) of subjects in the IPM/CS group, and K. pneumoniae was isolated in 9.2% (6/65) of subjects in the cefiderocol group and 14.3% (5/35) of subjects in the IPM/CS group.

Gram-negative uropathogens isolated at baseline for the ME Population were similar to the Micro-ITT Population in subjects diagnosed with cUTI with or without pyelonephritis or acute uncomplicated pyelonephritis for both E. coli and K. pneumoniae and in both treatment groups.

A similar distribution was noted in the ME Population. E. coli was the most frequently identified Gram-negative pathogen isolated from the blood culture for the Micro-ITT Population: 14/252 (5.6%) subjects in the cefiderocol group and 7/119 subjects (5.9%) in the IPM/CS group.

For E. coli, the most frequent uropathogen, MIC distribution for cefiderocol in both treatment groups was similar and all strains were susceptible to IPM. For K. pneumoniae at baseline, a lower percentage (86.7% [39/45]) of isolates were susceptible to IPM in the cefiderocol group

compared with the IPM/CS group (95.7% [22/23] of isolates). A greater percentage of K. pneumoniae isolates were susceptible to cefepime (44.4% [20/45] in the cefiderocol group compared with 34.8% [8/23] in the IPM/CS group). E. coli and K. pneumoniae rates of resistance to levofloxacin were similar between treatment groups. For the MIC of other Gram- negative baseline uropathogens, the numbers of subjects with isolates were too small to make a meaningful comparison between treatment groups.

The summary of MIC of baseline Gram-negative pathogens isolated from the blood culture that were the same as the uropathogen at baseline in ME Population is consistent with the findings in the Micro-ITT Population [237].

5.3.2 APEKS-NP STUDY

A Multicenter, Randomized, Double-blind, Parallel-group, Clinical Study of S-649266 (Cefiderocol) Compared With Meropenem for the Treatment of Hospital-acquired Bacterial Pneumonia, Ventilator-associated Bacterial Pneumonia, or Healthcare-associated Bacterial Pneumonia Caused by Gram-negative Pathogens.

The APEKS-NP study was a Phase 3, multicenter, randomized, double-blind, parallel-group, active-controlled study to assess the efficacy and safety of cefiderocol vs high dose prolonged infusion (HD) meropenem in subjects with nosocomial pneumonia caused by Gram-negative bacteria. Subjects meeting eligibility criteria and assessed by the investigator as requiring 7 to 14 days of intravenous (IV) treatment in the hospital were randomized (1:1) to either cefiderocol, 2 g, administered IV over 3 hours every 8 hours (q8h) or meropenem, 2 g, administered IV over 3 hours, q8h. The dose of meropenem was increased from the labelled dose of 1 g to 2 g and extended to a 3-hour infusion to optimize the exposure to meropenem thereby the antibacterial activity of meropenem in this MDR pathogens, at risk of carbapenem resistance. Dose adjustment based on renal function was required for cefiderocol and comparator. Linezolid was administered for at least 5 days to subjects in both arms to provide coverage for methicillin-resistant Staphylococcus aureus (MRSA), maintain the study blind and, in the cefiderocol arm, provide coverage for Gram-positive bacteria. The recommended duration of treatment with IV study drugs was 7 to 14 days in the hospital, but treatment could have been extended up to 21 days based on the investigator’s clinical assessment of the subject.

Figure 25: APEKS-NP study design and patient flow

EA, early assessment; EOS, end of study; EOT, end of treatment; PE, Primary Endpoint; FUP, follow up; HD, high dose; Q8h, every 8 hours; TOC, test of cure

A total of 300 subjects (150 in the cefiderocol group and 150 in the HD meropenem group) were randomized 1 to 1 to cefiderocol or HD meropenem. All randomized subjects who received at least 1 dose of study treatment were included in the Intent-to-treat (ITT)/Safety population (298 subjects: 148 in the cefiderocol group and 150 in the HD meropenem group). The primary efficacy population was the mITT population (145 in the cefiderocol group and 147 in the HD meropenem group), which included all ITT subjects with evidence of a Gram- negative infection of the lower respiratory tract and those who had evidence of a lower respiratory tract infection but whose culture or other diagnostic tests did provide a microbiological diagnosis; subjects with Gram-positive only infections were excluded. Figure 26 below provides an overview of the patient flow.

Figure 26: Patient demographics and baseline characteristics

5.3.2.1 DEMOGRAPHICS AND BASELINE CHARACTERISTICS

Demographic characteristics of gender, age, race, and region in the ITT population were generally similar between the treatment groups (Table 19).[238] Most subjects were male (68.2% in the cefiderocol group and 69.3% in the HD meropenem group) and white (68.9% in the cefiderocol group and 66.7% in the HD meropenem group). The mean age was 64.7 years in the cefiderocol group and 65.6 years in the HD meropenem group; 27.0% of subjects in the cefiderocol group and 31.3% in the HD meropenem group were ≥ 75 years of age. Most subjects were enrolled from Europe (66.9% in the cefiderocol group and 66.7% in the HD meropenem group); 29.1% of subjects in the cefiderocol group and 29.3% in the HD meropenem group were enrolled from the Asia-Pacific region.

Baseline characteristics of clinical diagnosis, ventilation status, baseline pathogens, and blood culture status in the ITT population were also similar between the treatment groups. The percentage of subjects with VABP was 40.5% in the cefiderocol group and 43.3% in the HD meropenem group; the percentage with HABP was 40.5% in the cefiderocol group and 40.7% in the HD meropenem group, and the percentage with HCABP was 18.9% in the cefiderocol group and 16.0% in the HD meropenem group. Subjects on ventilation at baseline represented 61.5% of the cefiderocol group and 58.0% of the HD meropenem group.

Table 19: Patient demographics and baseline characteristics (mITT population)

ITT Population Cefiderocol HD meropenem (N=148) (N=150) Gender (Male), n (%) 101 (68.2) 104 (69.3) Age, mean 64.7 65.6 Race (White), n (%) 102 (68.9) 100 (66.7) VABP 60 (40.5) 65 (43.3) Clinical diagnosis at HABP 60 (40.5) 61 (40.7) baseline, n (%) HCABP 28 (18.9) 24 (16) Ventilation status at Ventilated 91 (61.5) 87 (58) randomisation, n (%) Non-ventilated 57 (38.5) 63 (42) ≤ 15 75 (50.7) 78 (52) APACHE II, n (%) 16-19 32 (21.6) 26 (17.3) ≥ 20 41 (27.7) 46 (30.7) Mild renal impairment 44 (29.7) 37 (24.7) (50-80 ml/min) Moderate renal impairment 29 (19.6) 32 (21.3) Renal function, n (%) (30-50 ml/min) Severe renal impairment 20 (13.5) 20 (13.3) (<30 ml/min) APACHE, Acute Physiology, Age, Chronic Health Evaluation; HD, high dose; Source: Data on file [239]

Most subjects in both treatment groups had only Gram-negative pathogens at baseline (76.4% in the cefiderocol group and 70.0% in the HD meropenem group). The most frequently occurring Gram-negative pathogen in both treatment groups at baseline was Klebsiella pneumoniae (32.4% in the cefiderocol group and 29.3% in the HD meropenem group), followed by Pseudomonas aeruginosa (16.2% and 16.0% in the cefiderocol and HD meropenem groups, respectively) and Acinetobacter baumannii (15.5% and 16.0% in the cefiderocol and HD meropenem group, respectively), as shown in Table 20. Blood cultures positive for Gram-negative pathogens were observed in 5.4% of subjects in the cefiderocol group and 6.7% of the HD meropenem group. The mean APACHE II score was 16.1 in the cefiderocol group and 16.3 in the HD meropenem group.

Table 20: Top 5 baseline Gram-negative pathogens, n (%)

ITT Population Cefiderocol HD meropenem (N=148) (N=150)

Klebsiella pneumoniae 48 (32.4) 44 (29.3) Pseudomonas aeruginosa 24 (16.2) 24 (16) Acinetobacter baumannii 23 (15.5) 24 (16) Escherichia coli 19 (12.8) 22 (14.7) Enterobacter cloacae 7 (4.7) 8 (5.3) Source: Data on file [239]

5.3.3 CREDIBLE-CR STUDY

A Descriptive, Open-label, Multicenter, Randomized, Clinical Study of cefiderocol or Best Available Therapy for the Treatment of Severe Infections Caused by Carbapenem-resistant Gram-negative Pathogens

The small, descriptive CREDIBLE-CR study is a pathogen-focused randomised clinical trial that investigated the efficacy and safety of cefiderocol versus an individualized best available therapy (BAT) in 150 seriously ill patients with confirmed carbapenem-resistant (CR) Gram- negative infections, independent of the host’s infection site (HCAP/HAP/VAP, cUTI, BSI/sepsis are included). The objective of the study was to provide descriptive evidence of the efficacy and safety of cefiderocol for the target population of patients with CR infections, including the non-fermenters. The study was conducted in patients with evidence of CR Gram- negative infections at 100 sites in 17 countries covering 4 regions: Asia-Pacific, Europe, North America, and South America.

This study was not designed or powered to conduct hypothesis but to start gaining experience in patients with CR infections, with life-threatening, or end-of-life conditions with a high risk of mortality, often failing multiple lines of therapy (i.e. salvage therapy). No stratification for pathogen or terminal disease was done and differences in baseline characteristics between the two arms were observed. Study design and patient flow are presented in Figure 27: CREDIBLE-CR study design and patient flow

The study key inclusion criteria were [240, 241]:

. Patients who were diagnosed with HAP/VAP/HCAP, BSI or sepsis, or cUTI and . Documented or suspected CR Gram-negative infections [240, 241].

Key exclusion criteria were [240, 241]:

. Effective antibacterial regimen for the current CR infection within 72 hrs prior to randomization for a continuous duration of ≥24 hrs in cUTI, or ≥36 hrs in HAP/VAP/HCAP or BSI/sepsis, . Moderate or severe hypersensitivity or allergic reaction to any beta-lactam antibacterial . Requirement of >3 systemic antibacterials for the treatment of the current infection if randomised to the BAT arm . Infections: endocarditis, osteomyelitis, meningitis and co-infections with invasive mold . Conditions: cystic fibrosis/bronchiectasis, refractory septic shock or severe neutropenia (<100 cells/μL blood)

Patients with Acute Physiology and Chronic Health Evaluation II (APACHE II) score > 30Patients (n = 152) were randomised 2:1 to either cefiderocol, 2 g, administered IV over 3 hours every 8 hours (q8h) or BAT (

Figure 28) [240, 241]. Patients with cUTI received cefiderocol as monotherapy, whereas for patients with HAP/VAP/HCAP or BSI/sepsis, physicians could choose to add one additional antibacterial [240, 241]. Patients were stratified by primary clinical diagnosis (HAP/VAP/HCAP, BSI/sepsis, cUTI), APACHE II score (≤15 or ≥16–≤30 at screening), and region (North America, South America, Europe, Asia-Pacific) [240]. Of note, the stratification did not account for pathogens at baseline or other severity indicators such as mechanical ventilation status, shock, and location in the intensive care unit (ICU).

Best Available Therapy was chosen by the investigator before randomization, and could include up to three antibacterials with Gram-negative coverage used in combination [240, 241]. Best Available Therapy was chosen by the investigator before randomization, and could include up to three antibacterials with Gram-negative coverage used in combination [240, 241]. Due to the enrolment of patients with a broad range of CR Gram-negative bacteria and infection types, BAT was considered to be the appropriate comparator reflecting the variation in the combination of treatments within the clinical practice [240, 241]. This was also in accordance with the regulatory guidance by EMA [240].

Figure 27: CREDIBLE-CR study design and patient flow

Figure 28: Subjects disposition (all randomized subjects)

5.3.3.1 BASELINE CHARACTERISTICS

The population in the CREDIBLE-CR study was expected to be very heterogeneous as it was a pathogen-focussed study which included subjects with many underlying conditions, different infection sites and infections due to a variety of Gram-negative CR pathogens also including non-fermenters (Acinetobacter spp. and Stenotrophomonas spp) [240, 242]. The study included a substantial number of patients with life-threatening, or end-of-life conditions with a high risk of mortality reflecting a compassionate use scenario. Baseline demographics were generally balanced between the 2 treatment arms (Table 21) with some clinical exceptions that can influence the results [242]. There was a higher proportion of patients of ≥ 65 years old (63.4% vs 44.9%) and patients with moderate (22.8% vs 16.3%) and severe renal impairment (19.8% vs 14.3%) in cefiderocol group than in BAT arm (Table 21) [242]. Due to the heterogeneity of the population the treatment groups do not appear to be balanced for baseline characteristics such as shock (which has a major impact on mortality) in the subgroup of subjects with A. baumannii infections. Twenty-six out of 150 (17%) patients in CREDIBLE- CR trial had polymicrobial infections at baseline, and all patients with 3-4 co-pathogens were randomised to the cefiderocol arm (Table 21).

Table 21: Patient demographics and baseline characteristics (ITT population)

Cefiderocol BAT Parameter n=101 n=49 Sex, (%) Men (%) 65.3 71.4 Median 69.0 (19, 92) 62.0 (19, 92) Age, y ≥65 (%) 63.4 44.9 69.4 (4.6, CrCl, mL/min Median (min, max) 59.2 (9.4, 539.6) 270.8) CrCl renal grading group in mL/min, n <50 (Moderate and Severe) (%) 42.6 30.6 (%) HAP/VAP/HCAP (%) 44.6 44.9 Clinical diagnosis at BSI/sepsis (%) 29.7 34.7 baseline, n (%) cUTI (%) 25.7 20.4 APACHE II score Median (min, max) 15 (2, 29) 14 (2, 28) SOFA Score Median (min, max) 4.0 (0, 17) 4.0 (0, 16) Clinical Pulmonary Median (min, max) 5.0 (2, 9) 5.0 (0, 7) Infection Score BAT, best available therapy; BSI, bloodstream infection; cIAI, complicated intra-abdominal infection; cUTI, complicated urinary tract infection; HAP, hospital-acquired pneumonia; HCAP, healthcare-associated pneumonia; ITT, intent-to-treat; VAP, ventilator-associated pneumonia. Source: Data on file [242]

5.3.3.2 Baseline Study Drug Regimen

In the cefiderocol group, 82.5% (66/80) of the subjects received monotherapy, while 28.9% (11/38) of the subjects in the BAT group received monotherapy (Table 22). A colistin-based regimen was given to 65.8% (25/38) of the subjects in the BAT group. Other than colistin monotherapy (received by 6 subjects in the BAT group), 5 subjects in the BAT group received other monotherapy (amikacin, ceftazidime/avibactam, doripenem, fosfomycin, and gentamicin).

Colistin was a prohibited medication in the cefiderocol group; however, one patient received cefiderocol and colistin. In the cefiderocol group only 1 additional drug for Gram-negative pathogens could be added; however, one patient received cefiderocol, gentamicin, and tigecycline. Overall there was a significant diversity of regimens in combination with colistin.

Table 22: Summary of study regimen for Gram-negative pathogen at day 1 and day 2 (CR- mITT population)

5.3.3.3 Baseline Pathogens

The study was not considering pathogens as a stratification factor, therefore, there was an imbalance on the baseline pathogens, where cefiderocol arm contained more patients with multiple pathogens and more non-fermenters, particularly Stenotrophomonas spp. All infections caused by S. maltophilia were randomised to the cefiderocol arm (Table 23) [242].

Table 23: Baseline Gram-negative pathogens, n (%)

Cefiderocol BAT Diagnosis (N = 86) (N = 44) Pathogen [a] n (%) n (%) All Infection Sites Combined N' = 86 N' = 44 Acinetobacter baumannii 39 (45.3) 17 (38.6) Klebsiella pneumoniae 34 (39.5) 16 (36.4) Pseudomonas aeruginosa 17 (19.8) 12 (27.3) Escherichia coli 6 (7.0) 3 (6.8) Stenotrophomonas maltophilia 5 (5.8) 0 Acinetobacter nosocomialis 2 (2.3) 0 Enterobacter cloacae 2 (2.3) 0 Acinetobacter radioresistens 1 (1.2) 0 Chryseobacterium indologenes 1 (1.2) 0 Klebsiella oxytoca 1 (1.2) 0 Klebsiella variicola 1 (1.2) 1 (2.3) Serratia marcescens 1 (1.2) 0 Enterobacter asburiae 0 1 (2.3) Morganella morganii 0 1 (2.3) BAT, best available therapy; Micro-ITT, microbiological intent-to-treat Source: Data on file Of note, of the subjects with HAP/VAP/HCAP, 64.3% in the cefiderocol group and 47.6% in the BAT group had A. baumannii at baseline (Attachment: R2131_CREDIBLE-CR Final Study Summary) [243].

A comparison of the characteristics of APEKS-NP and CREDIBLE-CR and CREDIBLE-CR with HAP/VAP/HCAP (ITT population) is on file [243]. An analysis of APEKS NP subgroup with CR infection is presented in chapter 5.4.3.

5.3.4 Summary of compassionate use cases and published evidence

Cefiderocol has been provided upon request from attending physicians to patients with serious CR Gram-negative infections who have no other treatment options [244]. The criteria for fulfilling these requests are highly restrictive including that all other available treatments must be ruled out through susceptibility testing and/or evidence of treatment failure in efficacy or safety, and patients must be unable to enroll in clinical studies of cefiderocol [244]. Data for

74 patients completed cefiderocol therapy in compassionate use [244], and are presented here.

5.3.4.1 Patient characteristics in the compassionate use program

The mean age of patients in the compassionate use program was 46.8 years which is lower compared to other studies of cefiderocol, mainly due to the inclusion of children and infants in the compassionate use program [245]. There were 9/74 patients who were < 18 years with the youngest patients of 5 months old [245]. In addition to the infection sites included in the clinical trials of cefiderocol, compassionate use program also included patients with bone infections (18.9%) (Table 24) [245]. With regards to causal pathogens, non-fermenter species accounted for almost all isolates, with the most common being P. aeruginosa (n = 30); A. baumannii (n = 24), Achromobacter xylosoxidans (n = 10), Burkholderia cepacia complex (n = 9), Enterobacterales (n = 9), and S. maltophilia (n = 3) (Table 24) [244]. Eight patients had mixed infections with various MDR organisms [244]. All isolates were MDR with some being pan-resistant to currently available classes of antimicrobial agents [244]. The median duration of cefiderocol treatment was 21 days (range: 1-94 days) and patients received up to seven concomitant therapies alongside cefiderocol including polymixins (43.2%), cephalosporins (33.8%), carbapenems (23%), β-lactams (21.6%), sulphonamides (18.9%) and aminoglycosides (14.9%) [245]. (details of compassionate use program are provided in chapter 3.).

Patients in compassionate us program are only eligible if all other available treatments have been ruled out through susceptibility testing and/or evidence of treatment failure in efficacy or safety. TheTheThe severity of infections included in compassionate and based on the pathogen distribution reminds of the infections from the patient populationThe severity of infection included in CREDIBLE CR and is more severe than observed in APEKS trials.

Table 24: Patient demographics and baseline characteristics

Parameter Cefiderocol N=74 Gender (Male), n (%) 44 (59.5) Age, mean 46.8 ≥65 years, n (%) 19 (25.7) Infection type at BSI 21 (28.4) baseline, n (%) Pneumonia and respiratory infection 25 (33.8) Bone infection 14 (18.9) Bacteraemia 4 (5.4) Sepsis 3 (4.1) cUTI 1 (1.4) Other 6 (8.1) Most common Pseudomonas aeruginosa 31 (41.9) pathogen at diagnosis, n (%) Acinetobacter baumannii 22 (29.7) Burkholderia cenocepacia 10 (13.5) Klebsiella pneumoniae 6 (8.1) Escherichia coli 1 (1.4) BSI, bloodstream infection; cUTI, complicated urinary tract infection Source: NDA briefing document[244]; Data on file [245]

5.3.4.2 Published case reports Case reports for three patients from the expanded access program have been published so far.

 A patient was treated successfully for endocarditis due to extensively drug resistant (XDR) Pseudomonas aeruginosa.(Edgeworth et al., 2019)[246]

 A patient with multiple comorbidities and a complicated intra-abdominal infection (IAI) due to MDR Pseudomonas aeruginosa was released from hospital care within six weeks of completion of cefiderocol treatment. (Stevens et al., 2019)[247]

 A patient with VAP and BSI caused by XDR Acinetobacter baumannii and carbapenemase- producing Klebsiella pneumoniae had potentially serious organ failure from older anti- infectives. Six weeks after cefiderocol administration, chest X-rays showed complete resolution of infection (Trecarichi et al., 2019)[248]

Single and Multiple Dose Study A single-center, randomised, double-blind, placebo-controlled, ascending single and multiple dose study to evaluate the safety, tolerability and PK of cefiderocol in 70 healthy Japanese and Caucasian adult subjects. In the single-dose cohort, single doses of 100 mg, 250 mg, 500 mg, 1 g, and 2 g over a 1-hour infusion were tested. A single dose of 4 g was planned but was not initiated according to the study protocol dose escalation guidelines; a cohort would not proceed if the predicted maximum plasma concentration (Cmax) exceeds a 10-fold lower exposure than the rat no-observed-adverse-effect-level (C0 = 1660 μg/mL). In the multiple- dose cohort, once daily doses of 1 and 2 g over a 1-hour infusion on Day 1 followed by q8h doses of 1 and 2 g over a 1-hour infusion for 8 days on Days 2 to 9 and a once daily dose of 1 and 2 g over a 1-hour infusion on Day 10 were tested. The active drug and the placebo were administered to 6 subjects and 2 subjects, respectively, in each single-dose group and 8 subjects and 2 subjects, respectively, in each multiple-dose group [249].

Renal Impairment Study A multicenter, open-label, nonrandomised study to evaluate the PK, safety and tolerability of cefiderocol in subjects with varying degrees of renal impairment and in subjects with normal renal function. The PK of a single-dose 1 g of cefiderocol in subjects with mild, moderate, or severe renal impairment, or end-stage renal disease (ESRD) requiring haemodialysis (HD) was compared with that of healthy subjects with normal renal function who were demographically matched with moderate renal impairment. A total of 38 subjects were enrolled in 5 cohorts.

The clearance of cefiderocol with HD was determined based on plasma concentration data both before and after HD. Renal function was classified at screening visit based on creatinine clearance estimated by Cockcroft-Gault equation (CrCl) for subjects with normal renal function (≥ 90 mL/min) and estimated glomerular filtration rate (eGFR) using the modification of diet in renal disease (MDRD) equation for subjects with renal impairment (mild, 60 to < 90; moderate, 30 to < 60; severe, 15 to < 30 mL/min/1.73 m2). A single-dose of 1 g cefiderocol over a 1-hour infusion was administered to subjects with normal renal function or mild, moderate or severe renal impairment. Subjects with ESRD requiring HD were dosed approximately 1 to 2 hours after completion of a HD session on Day 1, and 2 hours prior to start of HD following at least a 72-hour washout period [249].

5.4 Individual study results (clinical outcomes)

1. Describe the relevant endpoints, including the definition of the endpoint, and method of analysis (Table 73a - Table 80e).

2. Provide a summary of the study results for each relevant comparison and outcome.

Due to the need to consider in-vitro data in combination with PK/PD and supportive trial data for an assessment of a novel antibacterial, such as cefiderocol, the data of the individual studies cannot be summarized across trials. Each clinical study used its own comparator and was conducted in different patient populations. For this reason, no overall results summary table is shown here, and section 5.4 has been modified to account for the specific circumstances.

The individual study results, including all requested stratifications by pathogen and time-point during the scoping process, are discussed in sections 5.4.1 through 5.4.3 below; the standard dossier section 5.4 has thus been split to accommodate all relevant results (in-vitro, PK/PD, and clinical).

In addition, for the APEKs trials, a feasibility analysis for an NMA was conducted. This feasibility analysis [227], showed that an NMA was feasible for APEKS-cUTI. Appendix F of that document contains the summary tables of all relevant outcomes across the trials included in the NMA.

5.4.1 Individual study results (in vitro surveillance outcomes)

In vitro activity of cefiderocol has been studied in large-scale multinational surveillance and small independent national studies [250]. Large multinational surveillance studies include SIDERO-WT studies initiated in North America and Europe and SIDERO-CR program collecting CR isolates from Europe, North America, South America, and the Asia-Pacific region (Table 25) [250]. In the section below, the results are reported for all global Gram-negative isolates. Of note, the susceptibility to cefiderocol was assessed based on the CLSI breakpoints. The EUCAST breakpoints are expected to be determined in February after the Committee for Medicinal Products for Human Use (CHMP) opinion. Data for the US clinical strains is reported in the appendix and data for European strains will be included after the availability of EUCAST breakpoints [251].

In addition, several independent validation studies carried out to determine cefiderocol activity have included collections of difficult-to-treat CR pathogens gathered from various countries

including Italy, Germany, Greece, Spain, UK/Ireland, Switzerland, and the US [250]. The list of these studies is reported in section “Study categorisation.”

Table 25: SIDERO Surveillance studies

SIDERO-WT in vitro studies Scope Systematic surveillance studies of cefiderocol in vitro activity compared to key antibacterials against a total of 30,459 Gram-negative isolates collecting isolates from three consecutive 12-month periods from 2014 to 2015 (SIDERO-WT-2014), from 2015 to 2016 (SIDERO-WT-2015), and from 2016 to 2017 (SIDERO-WT-2016) as well as cumulative. Geographic North America and Europe location Comparator Ceftolozane/tazobactam, ceftazidime/avibactam, cefepime, ciprofloxacin, treatments polymyxin E (colistin), and meropenem Included Carbapenem susceptible and carbapenem non-susceptible pathogens pathogens (CarbNS): Enterobacteriaceae (including but not limited to Escherichia coli, K. pneumoniae, Enterobacter spp., Citrobacter spp., Serratia spp.), non- fermenters (including but not limited P. aeruginosa, A. baumannii, S. maltophilia, B. cepacia), and Proteeae (M. morgannii, P. vulgaris, P. mirabilis). SIDERO-CR Scope In vitro study from 2014 – 2016 evaluating the activity of cefiderocol against a total of 1,873 MDR and CarbNS isolates Gram-negative Bacilli. Geographic World-wide (Europe, North America, Latin America, Asia, South Pacific, location Africa, and the Middle East) Comparator Ceftolozane/tazobactam, ceftazidime/avibactam, cefepime, ciprofloxacin, treatments polymyxin E (colistin), and meropenem

Included CarbNS Enterobacteriaceae, MDR A. baumannii, MDR P. aeruginosa, S. pathogens maltophilia and B. cepacia. The test isolates of MDR non-fermenters were defined to be resistant to meropenem, amikacin and ciprofloxacin. Source: Longshaw 2019 [47]; Hackel 2017 [29]; Hackel 2018 [30]

In vitro efficacy has been demonstrated in several independent world-wide pathogen collections:

1. A total of 30,459 clinical isolates of Gram-negative bacilli were systematically collected from USA, Canada, and 11 European countries between 2014 and 2017. All isolates were

sent to a central laboratory, IHMA (Schaumburg, Illinois), where the isolates were further evaluated and stored.

a. The SIDERO-WT analysis (study report S-649266-EB-344-N) was an extensive effort to determine susceptibility of cefiderocol and relevant comparators against cabapenem-susceptible and carbapenem-resistant pathogens. The purpose of this study was to calculate the indices related to the antibacterial activity of cefiderocol and the ratio of susceptible strains of cefiderocol and other reference compounds based on the breakpoint criteria of Clinical and Laboratory Standards Institute (CLSI) standards. MICs were determined by broth microdilution for a panel of 7 antibacterials, including cefiderocol, ceftazidime-avibactam (CZA), ceftolozane-tazobactam (C/T), colistin (CST), cefepime (FEP), meropenem (MEM), and ciprofloxacin (CIP) according to the Clinical & Laboratory Standards Institute (CLSI). Included herein are results from the total sample and from the European subsample (overall and non-fermenters) [29, 45, 46, 49, 250].

b. A subsequent analysis focused on a Difficult-to-treat resistant (DTR) subset of pathogens, which were non-susceptible to fluoroquinolones (CIP), extended-spectrum cephalosporins (FEP), and carbapenems (MEM) according to CLSI M100-E28:2018 breakpoints. (Longshaw et al., Poster presentation, IDWeek 2019 [47]).

c. A molecular analysis based on the same collection investigated acquired carbapenem-hydrolyzing enzymes (carbapenemases) identified in meropenem- non-susceptible (MEM-NS) strains and antibacterial susceptibility by year and country for the included strains. (Sato et al, Poster presentation, IDWeek 2019).

2. The SIDERO-CR-2014-2016 study (protocol S-649266-EF-115-N) included CR Enterobacteriaceae and MDR non-fermenters (defined as resistant to carbapenems, fluoroquinolones, and aminoglycosides) and demonstrated the potent in vitro activity of cefiderocol against these pathogens [30].

3. Several independent national validation studies were carried out to determine cefiderocol activity against difficult-to-treat CR pathogens gathered from various countries including Germany, Greece, Italy, Spain, UK/Ireland, and the US [252-257]. In these studies, cefiderocol demonstrated consistent activity against Gram-negative pathogens regardless of the geographic origin.

4. A validation study by a Swiss team of scientists confirmed cefiderocol activity in an independent world-wide collection of Gram-negative pathogens [258].

5. Several studies followed up on the surveillance results and aimed to characterize rare resistant pathogens and identify their mechanisms of resistance. [61, 62] (Ito et al., Poster presentation, ASM Microbe 2019).

Results from these studies and relevant published sub-analyses involving European samples are summarized in the following sections, followed by a summary analysis of the expected efficacy of cefiderocol based on in vitro susceptibility results, compared to other treatment options.

It is important to highlight that there these studies are continuously being updated and new isolates analysed and incorporated, with correspondent publications following, showing the results by year, cumulative, or for specific groups of pathogens of interest.

5.4.1.1 1a) SIDERO-WT results for all Gram-negative isolates[45, 49, 250]

In the SIDERO-WT in vitro studies, cefiderocol demonstrated activity against the majority of

Gram-negative isolates at MIC of <4 µg/mL (only MIC90 for B. multivorans was 32 µg/mL) with higher coverage rates than other comparators included in these studies [250]. The SIDERO- WT program included 4 multinational surveillance analyses testing a total of 9205 Gram- negative bacterial clinical isolates in 2014–2015, 8954 in 2015–2016, and 10 470 in 2016– 2017 [250] and continues to include more isolates every year. To date (January, 2020), 30,459 459samples have been tested. Of note, >99% of isolates had low cefiderocol MIC values in each testing period [250]. The latest surveillance SIDERO-WT study (2016-2017) showed that cefiderocol demonstrated activity against 99.45% of GN pathogens at MIC of 4 mg/L compared to 90.2% for ceftazidime-avibactam, 84.28% for ceftolozane-tazobactam, and 95.49% for colistin (Table 26) [49].

With regards to in vitro activity across different pathogens, cefiderocol demonstrated potent in vitro activity against Enterobacteriaceae (99.9%) and non-fermenters including A. baumannii, P. aeruginosa, S. maltophilia, and B. cepacia (98.53%) which was higher than for other available treatments (Table 26) [49].

Table 26: In vitro activity data for all tested clinical strains (SIDERO-WT-2014/2015/2016 and Proteeae) of cefiderocol (at MIC of 4mg/L) versus ceftazidime-avibactam, ceftolozane-tazobactam, and colistin

Ceftolozane / tazobactam Polymyxin E Ceftazidime / Cefiderocol %S MIC ≤2 µg/mL for Organism (colistin) % S avibactam % S % Enterobacteriaceae, ≤4 MIC ≤2 µg/mL MIC ≤8 µg/mL µg/mL for non-fermenters All Gram-negative 95.49b 99.45 84.28 90.20 (N=30,459) (n=25372) Enterobacteriaceae 96.54c 99.86 91.43 99.23 (N=20,949) (n=16026) Non-fermentersa 93.67d 98.53 68.52 70.33 (N=9,510) (n=9346) CR Enterobacteriaceae 75.55c 98.16 8.40 77.67 (N=654) (n=581) (MEPM MIC ≥ 2 µg/mL) CR 86.85d (N=4,331) 97.57 34.61 40.96 (n=4208) (MEPM MIC ≥ 4 µg/mL) CR P. aeruginosa (N=1,154) 99.91 98.35 76.08 75.38 (MEPM MIC ≥ 4 µg/mL) CR A. baumannii (N=1,891) 94.87 85.14 7.77 16.23 (MEPM MIC ≥ 4 µg/mL) S. maltophilia 99.82 78.17 34.27 42.88 (N=1,173) Source: [49]. CarbNS - carbapenem non-susceptible; MEPM - meropenem; MIC - minimum inhibitory concentration. Green: More than 80% susceptible; yellow: between 60-80% susceptible, red: less than 60% susceptible. a Non-fermenters include P. aeruginosa, S. maltophilia, Burkholderia spp, and Acinetobacter spp. b Burkholderia spp, Proteeae and Serratia spp. were excluded because they are intrinsically resistant to Polymyxin E (Colistin) c Serratia spp. and Proteeae was excluded. d Burkholderia spp was excluded.

5.4.1.1.1 European sub-sample

The in vitro activities of cefiderocol and six comparators are summarized in Table 27 for the 5352 isolates from European clinical laboratories from the 2015 collection. (Karlowsky et al.,

2018). The concentration of antimicrobial agent inhibiting 50% (MIC50) and 90% (MIC90) of Enterobacteriaceae isolates tested against cefiderocol were 0.25 and 1 mg/L for European isolates (MIC range ≤0.002-8 mg/L). Cefiderocol inhibited 99.9% (6005/6013) of all isolates of Enterobacteriaceae tested, from European clinical laboratories, at a concentration (MIC) of ≤4 mg/L. Of the eight isolates of Enterobacteriaceae with cefiderocol MICs of ≥8 mg/L, three were European isolates (two isolates of E. coli and one isolate of Citrobacter freundii), all with cefiderocol MICs of 8 mg/L. Each isolate of Enterobacteriaceae with a cefiderocol MIC ≥8 mg/L was from a unique clinical laboratory location. Seven of the eight isolates with cefiderocol MICs of 8 mg/L were susceptible to both meropenem and ceftazidime-avibactam compared with six isolates susceptible to colistin, four isolates susceptible to ciprofloxacin, and only one isolate susceptible to cefepime and ceftolozane-tazobactam. Against meropenem-non- susceptible (MIC ≥2 mg/L) isolates of Enterobacteriaceae from Europe (n = 196), cefiderocol

MIC50 and MIC90 values were 2 and 4 mg/L, respectively; 99.6% (245/246) of all meropenem- non-susceptible Enterobacteriaceae had MICs to cefiderocol of ≤4 mg/L.

Table 27: In vitro activity of cefiderocol and comparators against Gram-negative bacilli isolated by 55 clinical laboratories in Europe in 2015 (n=5352)

An updated analysis of the European sample (Shionogi, data on file) investigated the in vitro activity of cefiderocol and comparators specifically against non-fermenters (SIDERO-WT- 2014-2016; European isolates).

Table 28: In vitro activity of cefiderocol and comparators against non-fermenters

Source: Shionogi, data on file.

 Cefiderocol demonstrated greater potency than all comparators against the pathogens P.

aeruginosa and A. baumannii, based on MIC50 and MIC90 values:

o Against P. aeruginosa, and based on MIC90 values, cefiderocol (MIC90 0.5 mg/L) was 4 times more potent than colistin and ≥8 times more potent than all other comparators.

o The activity of cefiderocol against A. baumannii (MIC90 2 mg/L) was ≥32 times greater than cefepime, ceftazidime/avibactam, ceftolozane/tazobactam, and meropenem, and was 4 times greater than colistin.

o Cefiderocol (MIC90 0.25 mg/L) also demonstrated activity against S. maltophilia that was ≥256 times more potent than cefepime, ceftazidime/avibactam, ceftolozane/tazobactam, and meropenem, and 32 times more potent than ciprofloxacin and colistin.

o All comparators showed lower activity than cefiderocol (MIC90 0.5 mg/L) against B. cepacia complex, with cefiderocol being ≥16 times more potent.

The cefiderocol MIC90 against CarbNS-P. aeruginosa was 1 mg/L and, with the exception of

colistin (MIC90 2 mg/L) and ciprofloxacin (MIC90 >8 mg/L), comparator MIC90s were ≥64 mg/L.

Cefiderocol maintained activity against CarbNS-A. baumannii (MIC90 2 mg/L), and demonstrated >4 times greater potency than all comparators.

5.4.1.2 1b) SIDERO-WT-based analysis of difficult-to-treat resistant (DTR) pathogens [47]

All antibacterials were tested in cation-adjusted Mueller-Hinton Broth (CAMHB) except cefiderocol, for which iron-depleted CAMHB was used. Susceptibility was determined

according to CLSI interpretive breakpoints (CLSI M100-E28: 2018) except CST, where EUCAST breakpoints were used (Table 29).

Pathogens were defined as ‘Difficult-to-Treat Resistant’ (DTR) if they were non-susceptible to fluoroquinolones (CIP), extended-spectrum cephalosporins (FEP), and carbapenems (MEM) according to CLSI M100-E28:2018 breakpoints (Table 29). Pathogens were defined as carbapenem non-susceptible if they had MICs to meropenem of >1 μg/mL (Enterobacterales); >2 μg/mL (Pseudomonas spp./ Acinetobacter sppspp.); >4 μg/mL (Burkholderia cepacia complex). Stenotrophomonas maltophilia was considered inherently non-susceptible to carbapenems, however, an arbitrary breakpoint of >4 μg/mL was used. Among 30,459 Gram- negative isolates collected between 2014 and 2017, 9.3% were non-susceptible to FEP, MEM, and CIP and could be defined as DTR.

Table 29: Breakpoints for non-susceptibility used in definition of DTR (μg/mL)

Cefiderocol demonstrated activity in 94.5% of DTR A. baumannii, 99.8% of P. aeruginosa and 98.3% of Enterobacterales [47]. Susceptibility of these pathogens were lower for other available treatments (Table 30) [47].

Table 30: Susceptibility of cefiderocol and comparators to pathogens

Ceftazidime / Ceftolozane / Pathogen Cefiderocola % Colistina % avibactama % tazobactama

DTR Enterobacterales 98.3 78.2 2.05 68.2 (n=573)

DTR P. aeruginosa 99.8 49.5 48.8 98.3 (n=470)

DTR A. baumannii 94.5 14.2 5.8 85 (N=3,451)

Table 31 below shows that 98.7% of CarbNS Enterobacteriaceae, and 96.4% of CarbNS non- fermenters were calculated to be sensitive to cefiderocol at a MIC of ≤4 μg/mL.

Table 31: In vitro activity data for CR Gram-negative pathogens (SIDERO-WT-2016-2017) of cefiderocol versus ceftazidime-avibactam, ceftolozane-tazobactam and colistin

Ceftazidime / Ceftolozane / Pathogen Cefiderocola % Colistina % avibactama % tazobactama CarbNSb Enterobacteriaceae 98.7 81.3 11.1 71d (225) CarbNSb non- fermentersc 96.4 39.8 37.0 91.5e (1427) CarbNSb P. aeruginosa 100 75.6 77.3 97.3 (406)

CarbNSb A. baumannii 91 11.0 9.0 90.8 (565)

S. maltophilia 100 38.8 31.4 86 (405) CarbNS, carbapenem-non-susceptible a Ratios (%) susceptible strains were calculated by using the following MIC criteria: Cefiderocol MIC ≤4 μg/mL, ceftazidime/avibactam MIC ≤8 μg/mL, ceftolozane/tazobactam MIC ≤2 μg/mL for Enterobacteriaceae, ≤4 μg/mL for non- fermenters, colistin MIC ≤2 μg/mL. b CR strain was defined as meropenem MIC ≥2 μg/mL for Enterobacteriaceae, ≥4 μg/mL for non-fermenters c Non-fermenters include P. aeruginosa, S. maltophilia, Burkholderia spp, and Acinetobacter spp. d Serratia spp. and Proteeae were excluded. e Burkholderia spp. was excluded. Source: Tsuji 2019[49]

The DTR phenotype was most frequently observed in Acinetobacter spp. (55.5%), followed by Burkholderia spp. (19%), Pseudomonas aeruginosa (9.5%) and Enterobacterales (2.7%). From 1173 S. maltophilia isolates tested, 60.7% were non-susceptible to meropenem, cefepime and ciprofloxacin, however, trimethoprim-sulfamethoxazole was not tested and could be considered a treatment option for these infections, thus we were not able to state how many isolates might be considered DTR.

In summary, the results from this analysis showed that cefiderocol demonstrated potent activity against ‘Difficult-to-Treat Resistant’ Gram-negative pathogens which leave physicians with limited options for high efficacy, low toxicity first-line treatment.

5.4.1.3 1c) Study of acquired carbapenem-hydrolyzing enzymes (carbapenemases) identified in meropenem-non-susceptible (MEM-NS) strains [48] In a subgroup analysis of SIDERO-WT-2014-2016 studies including meropenem-non- susceptible strains, cefiderocol demonstrated potent in vitro activity irrespective of the presence of specific carbapenemases [48].The isolates were stratified per resistance determinants detected through the conventional polymerase chain reaction (PCR) method and

included VIM-, NDM-1-, KPC-, and OXA-producing Enterobacteriaceae, VIM-, IMP-, and GES- producing P. aeruginosa, and OXA-, GES- and NDM-producing A. baumannii [48].

In all, 3691 Gram-negative isolates of MEM-NS A. baumannii complex, P. aeruginosa, K. pneumoniae, other Klebsiella spp., Serratia marcescens, Enterobacter spp., Citrobacter spp., and Escherichia coli, from SIDERO-WT-2014 (Year 1: 2014–2015), SIDERO-WT-2015 (Year 2: 2015–2016), and SIDERO-WT-2016 (Year 3: 2016–2017) were molecularly characterized. Information on the number of isolates by year and by country of collection is shown in Table 32 and Table 33, respectively.

Table 32: Number of MEM-NS isolates by year and species

Table 33: Number of MEM-NS isolates by country and species

5.4.1.3.1.1 Detection of β-lactamase genes Screening for the carriage of genes encoding carbapenemases and sequencing are described by Kazmierczak et al. Briefly, multiplex polymerase chain reaction assays were used to screen oxacillin carbapenemase (OXA)-23-like, OXA-24/40-like, OXA-48-like, OXA-58-like, K. pneumoniae carbapenemase (KPC), imipenemase metallo-β-lactamase (IMP), Verona integron-encoded metallo-β-lactamase (VIM), New Delhi metallo-β-lactamase (NDM), Sao

Paulo metallo-β-lactamase (SPM), Guiana-extended-spectrum β-lactamase (GES) and German imipenemase (GIM) in Acinetobacter spp.; KPC, GES, OXA-24/40-like, IMP, VIM, NDM, SPM, and GIM in P. aeruginosa; and KPC, GES, OXA-48-like, IMP, VIM, and NDM in Enterobacteriaceae.

Genes encoding KPC, GES, IMP, VIM, and NDM carbapenemases were sequenced. Among GES subtypes, GES-2, -4, -5, -6, -11, -12, -14, -15, -16, -18, 20, and -24 were considered carbapenemases. All experiments were conducted at a central laboratory (International Health Management Associates, Inc. in Schaumburg, IL, USA), where the isolates were stored.

5.4.1.3.1.2 Minimum inhibitory concentration (MIC) data Antimicrobial susceptibility data reported in the SIDERO studies were used. MICs were determined by the broth microdilution method according to the CLSI guidelines. For MIC determination, iron-depleted cation-adjusted Mueller–Hinton broth (ID-CAMHB) medium was used for testing of cefiderocol and CAMHB was used for testing of ceftazidime-avibactam, ceftolozane-tazobactam, meropenem, cefepime, and colistin.

5.4.1.3.1.3 Susceptibility criteria Susceptibility to each antibacterial agent was determined according to the CLSI M100-S29. For the purpose of comparison, breakpoint values of ceftazidime-avibactam and ceftolozane- tazobactam for P. aeruginosa were also applied to isolates of A. baumannii complex for which breakpoint values have not been defined by the CLSI (Table 34).

Table 34: Susceptibility breakpoints according to the CLSI (cefiderocol) and/or EUCAST (all comparators)

The following tables (Table 35 - Table 38) summarize the county-by country variability in the frequency of resistant strains of A. baumannii, P. aeruginosa, K. pneumoniae, and Enterobacteriaceae.

Table 35: Percentage of susceptibility of MEM-NS A. baumannii complex by country

Table 36: Percentage of susceptibility of MEM-NS P. aeruginosa complex by country

Table 37: Percentage of susceptibility of MEM-NS K. pneumoniae by country

Table 38: Percentage of susceptibility of other MEM-NS Enterobacteriaceae by country

This analysis found the molecular diversity to be high in carbapenemases. Carbapenemase detection rates, especially in P. aeruginosa and K. pneumoniae, greatly varied across countries. The presence of metallo-carbapenemases, both NDM and VIM, is noteworthy in some countries.

The results show that cefiderocol demonstrated potent in vitro activity against MEM-NS strains, including isolates with reduced susceptibility to colistin, irrespective of the presence of either serine-type or metallo-type carbapenemases. Of the antibacterials tested, only

cefiderocol was broadly active against all species of MEN-NS clinical isolates regardless of the geographic origin.

5.4.1.4 2) SIDERO-CR-2014-2016 study (protocol S-649266-EF-115-N)

The SIDERO-CR-2014-2016 study including CR Enterobacteriaceae and MDR non- fermenters (defined as resistant to carbapenems, fluoroquinolones, and aminoglycosides)

demonstrated the potent in vitro activity of cefiderocol with a MIC90 ranging between 0.25 and 8 µg/mL [30, 250]. For MIC of ≤ 4 µg/mL, cefiderocol showed activity against 96.2% of these pathogens and demonstrated higher in vitro activity than other available treatments (Table 39) [30, 250]. Cefiderocol inhibited the growth of 97.0% of CR Enterobacteriaceae, 99.2% MDR P. aeruginosa, 90.9% MDR A. baumanii and 100% S. maltophila at a concentration of 4 mg/L (Table 39) [30, 250].

Table 39: In vitro activity data for all tested clinical strains (SIDERO-CR 2014-2016) of cefiderocol versus ceftazidime-avibactam, ceftolozane-tazobactam, and colistin

Ceftazidime / Ceftolozane / Pathogen Cefiderocola % Colistina % avibactama % tazobactama CarbNSb Enterobacteriaceae 97.0 77.0 1.7 77.8c (1022) MDR P. aeruginosa 99.2 36.3 24.1 99.6 (262)

MDR A. baumannii 90.9 NA NA 94.6 (368)

S. maltophilia 100 NA NA NA (217) CarbNS, carbapenem-non-susceptible; MDR, multi drug resistant; NA, susceptibility breakpoints not available a Ratios (%) susceptible strains were calculated by using the following MIC criteria: Cefiderocol MIC ≤4 μg/mL, ceftazidime/avibactam MIC ≤8 μg/mL, ceftolozane/tazobactam MIC ≤2 μg/mL for Enterobacteriaceae, ≤4 μg/mL for non- fermenters, colistin MIC ≤2 μg/mL. b CR strain was defined as meropenem MIC ≥2 μg/mL for Enterobacteriaceae, ≥4 μg/mL for non-fermenters c Includes 39 Serratia species that are intrinsically resistant to colistin Source: Hackel, 2018 [30]; Yamano [250]

In addition to demonstrating high activity of cefiderocol against different drug-resistant species, the SIDERO-CR study showed antibacterial activity against isolates stratified per resistance determinants detected through the PCR method [250]. This includes VIM-, NDM-, KPC-, and OXA-producing Enterobacteriaceae, VIM-producing P. aeruginosa, MEPM-non- susceptible but acquired β-lactamase negative P. aeruginosa, OXA-23 and OXA-24/40 carbapenemase-producing A. baumannii.

5.4.1.5 3) Independent international validation studies

5.4.1.5.1 Germany [259]

Collection I comprised 213 first isolates from patients collected during a multicenter surveillance study conducted by the Paul-Ehrlich-Society in 2013, namely 146 Enterobacterales (including 17 ESBL-producing strains), 13 Acinetobacter baumannii group isolates, and 54 Pseudomonas aeruginosa. Collection II included 59 carbapenemase producing Enterobacterales from our stock collection. Minimum inhibitory concentrations (MICs) of cefiderocol and comparative antibacterial agents were determined using the microdilution method according to the standard ISO 20776-1. The provisional CLSI breakpoint of cefiderocol for susceptibility is ≤4 mg/L.

Cefiderocol inhibited 99% of the collection I at ≤4 mg/L (Table 40). MIC50/90 values of cefiderocol for Enterobacterales isolates were 0.12/1 mg/L. However, cefiderocol was more active against ESBL-negative isolates than against ESBL-producing Enterobacterales (isolates with MIC >1 mg/L: 4/129 [3.1%] ESBL-negative isolates vs 7/17 [41%] ESBL- producing isolates). In contrast, cefiderocol inhibited all Acinetobacter isolates at 0.12 mg/L and all P. aeruginosa isolates at 1 mg/L (Table 40). The highest cefiderocol MICs observed for collection II strains were 16 mg/L. Cefiderocol inhibited all seven carbapenemase-

producing A. baumannii at 0.25 mg/L. MIC50/90 values for Enterobacterales (n=30) and P. aeruginosa (n=22) were 1/4 mg/L and 0.5/2mg/L, respectively.

Table 40: MIC of cefiderocol and comparators in Germany

5.4.1.5.2 Greece [252]

A total of 471 (445 meropenem resistant and 26 meropenem intermediate) isolates, collected from ICUs and wards of 18 Greek hospitals, were included [282 Enterobacteriaceae (244 K. pneumoniae, 1 Klebsiella oxytoca, 14 Enterobacter cloacae, 11 Providencia stuartii, 7 E. coli, 4 Proteus mirabilis and 1 Serratia marcescens) and 189 non-fermentative Gram-negative bacteria (107 A. baumannii and 82 P. aeruginosa). Table 41 shows the summary data of the

MIC range, MIC50 and MIC90 of the antibacterials for the tested bacterial isolates and their respective resistance percentages.

Resistance to colistin was observed in 154 isolates [including: 91 K. pneumoniae isolates (37.2% of all K. pneumoniae); 45 A. baumannii isolates (42.1% of all A. baumannii); 1 P.

aeruginosa isolate and 2 E. coli isolates]. The MIC range, MIC50 and MIC90 of cefiderocol did not differ between colistin-resistant and colistin-susceptible A. baumannii isolates.

Table 41: MIC of cefiderocol and comparators in Greece

5.4.1.5.3 Italy [260]

42 MDR strains, previously characterized for their β-lactamases content, including 13 Klebsiella pneumoniae, 9 Escherichia coli, 5 Proteus mirabilis, 6 Pseudomonas aeruginosa, 6 Acinetobacter baumannii, 2 Enterobacter cloacae complex and 1 Aeromonas spp., were tested for susceptibility to cefiderocol and comparators.

The cefiderocol MIC50 and MIC90 values (0.5 mg/L and 4 mg/L, respectively) were significantly lower than comparators. In particular, cefiderocol showed a good in vitro activity (MIC≤4 mg/L)

against: i) 20 carbapenemase-producing Enterobacterales (8 KPC, 3 VIM, 1 NDM, 4 OXA-48, 2 OXA-232, 2 NMC-A/IMI); ii) 6 ESBL-producing Enterobacterales (TEM-52,TEM-92, PER1,VEB-6 + TEM-52, CTX-M-15, CTX-M-65); iii) one CMY-producing P. mirabilis; iv) 6 carbapenemase-producing P. aeruginosa (3 VIM, 1 FIM, 1 GES, 1 IMP);v) 5 A. baumannii (2 OXA-58, 1 OXA-23, 1 OXA-24, 1 ISAba1-OXA-51).

Cefiderocol was less active against a FOX-7-producing K. pneumoniae (MIC, 8 mg/L), an NDM5-producing E. coli (MIC, >64 mg/L), an OXA-23-producing A. baumannii (MIC, >64 mg/L) and a PER-producing Aeromonas spp. (MIC, >64 mg/L).

High cefiderocol MIC values were not associated with any specific β-lactamase class.

5.4.1.5.4 Spain [257]

231 clinical isolates of Enterobacteriaceae (121 ESBL-and/or carbapenemase-producing K. pneumoniae, and 4 carbapenemase-producing E. cloacae), 80 A. baumannii, six P. aeruginosa, and 20 S. maltophilia were tested. Cefiderocol showed a potent in vitro activity

against the isolates analyzed, with MIC50 and MIC90 values between 0.125-8 mg/L and 0.5-8 mg/L, respectively, and 98% of isolates were inhibited at ≤4 mg/L. Only five isolates showed a MIC of cefiderocol >4 mg/L, three ST2/OXA-24/40-producing A. baumannii, oneST114/ VIM- 1-producing E. cloacae, and one ST114 E. cloacae producing VIM-1 plus OXA-48. All KPC- 3-producing K. pneumoniae were susceptible to cefiderocol, even those resistant to ceftazidime/avibactam (Table 42).

Table 42: MIC of cefiderocol and comparators in Spain

5.4.1.5.5 United Kingdom and Ireland [253]

The test panel were 305 clinical Enterobacteriaceae submitted between 2008 -2016 and all but 2 were from UK hospitals. The 2 non-UK isolates were from Ireland. The panel was selected to represent diverse carbapenemase producers and those with carbapenem resistance via combinations of porin loss with AmpC or ESBL activity. Carbapenemase genes were identified by PCR or by whole genome sequencing.

Carbapenem resistance due to porin loss combined with ESBL or AmpC activity was inferred from their previous susceptibility results and the absence of carbapenemases.

Comparator antibacterials comprised meropenem, ceftazidime, ceftazidime-avibactam, cefepime, ceftolozane-tazobactam, aztreonam, colistin, amikacin, ciprofloxacin and tigecycline. MICs were interpreted using CLSI guidelines where available, or EUCAST breakpoints for ceftazidime-avibactam, tigecycline, and colistin.

Table 43: MIC of cefiderocol and comparators against in United Kingdom and Ireland

Table 44: Activity of antimicrobial agents tested against carbapenem-resistant P. aeruginosa and S. maltophilia

5.4.1.6 4) Independent validation study by Swiss scientists based on world-wide pathogens [258]

A total of 753 clinical multidrug-resistant isolates were evaluated in this study. They were representative of the most widespread and broad-spectrum mechanisms of resistance currently observed worldwide in Gram-negative bacteria. The strains were collected from hospitals worldwide (42 countries) from 2000 to 2016, with a majority dating from the 2012– 2016 period. They were of various origins (not always recorded) but mostly from urines, broncho-alveolar specimens, blood, pus, and stools (Table 45).

Table 45: MIC of cefiderocol and comparators for MDR-GN isolated

5.4.1.7 5) In vitro studies investigating resistance of pathogens against cefiderocol [61, 62] [261]

Cefiderocol activity against strains with porin channel mutations and overexpression of efflux pumps has been demonstrated in two in vitro studies [61, 62].

A study assessing contribution of active iron transporters and binding ability to PBPs of cefiderocol to its antibacterial/bactericidal activity against K. pneumoniae and E. coli compared to meropenem and ceftazidime, reported that neither porin mutations nor single iron transport mutations result in clinically relevant increases of cefiderocol MICs, probably due to the active siderophore transport system and β-lactamase stability [62].

Another in vitro study assessed contribution of chelating ability with iron (III) and the utilization of iron transporters through the outer membrane to the in vitro activity of cefiderocol against P. aeruginosa compared to other siderophore compounds (hydroxypyridone-substituted siderophore monobactams BAL30072, MB-1 and SMC-3176) [61]. The results suggest that cefiderocol is active against the mutants with multiple transporters with MIC of 2 µg/mL while other siderophore beta-lactams demonstrated lower activity. Of note, cefiderocol antibacterial activity was not affected by major efflux pump MexAB-OprM of P. aeruginosa, which is known to confer multidrug resistance (MDR) [61].

In the latest update of resistance investigations of the SIDERO-WT collection, Ito et al. reported overall low rates of resistance against cefiderocol (Ito et al., Poster presentation,

ASM Microbe 2019). The authors followed up with additional characterizations of the mutant strains identified in the SIDERO-WT-2014 subsample (red box in Table 46 below).

Table 46: Number of cefiderocol non-susceptible isolated in global surveillance studies (MIC ≥8 μg/mL)

Among the 38 isolates tested, 25 strains were Acinetobacter baumannii possessing PER β- lactamase isolated in Russia (18 isolates), Turkey (6 isolates) and Sweden (1 isolate), and 5 isolates were Klebsiella pneumoniae possessing NDM carbapenemase from Turkey. The addition of avibactam resulted in a ≥4-fold decrease of cefiderocol MIC to ≤0.5 μg/mL in 33 isolates that did not harbor NDM. Against the 5 isolates containing NDM, the addition of either DPA or avibactam did not decrease the MIC of cefiderocol, but the addition of both avibactam and DPA showed ≥8-fold decrease of the MIC to cefiderocol to ≤0.5 μg/mL. In contrast, meropenem MIC showed ≥64-fold decrease with the addition of DPA alone against these NDM-producing isolates. Additional testing of a further collection of 46 PER-producing isolates from IHMA confirmed that 43 had cefiderocol MIC of ≤4 μg/mL.

The authors concluded that the findings suggest that the high MICs observed were unlikely due to the presence of PER or NDM alone.

5.4.1.8 Summary Analysis: Expected comparative susceptibility

To determine susceptibility of Gram-negative bacteria to cefiderocol, multinational surveillance studies (SIDERO) were conducted over four consecutive years (2014 to 2018) using systematically collected clinical isolates from approximately 100 clinical laboratories in North American and European countries. A separate multinational surveillance study of Proteeae clinical isolates was also conducted. The antibacterial activity of cefiderocol was determined in iron-depleted cation-adjusted Mueller-Hinton broth medium (ID-CAMHB) medium, a method approved by the Clinical and Laboratory Standards Institute (CLSI). Comparators were tested in parallel using standard cation- adjusted Mueller-Hinton medium according to CLSI recommendations.

In February 2020, EUCAST defined a new clinical breakpoint for cefiderocol of 2 μg/mL for P. Aeruginosa and Enterobacterales. For A. baumanii and S. maltophilia, was proposed insufficient evidence (IE) refering to PK-PD breakpoints of 2 μg/mL (table below), which were used for this analysis. For the comparators, EUCAST breakpoint (version 9.0) were used in the analysis. In the absence of species specific breakpoint, PK/PD breakpoint were applied. For colistin, PK/PD breakpoint were not available so the analysis considered the Pseudomonas breakpoint of 2 μg/mL as an arbitrary breakpoint for Stenotrophomonas sp. and Burkholderia sp.

Table 47: EUCAST breakpoints for cefiderocol Species Sensitive (≤) Resistant (>) PK-PD breakpoints 2 μg/mL 2 μg/mL Enterobacterales 2 μg/mL 2 μg/mL Pseudomonas Aeruginosa 2 μg/mL 2 μg/mL Acientobacter baumanii 2 μg/mL 2 μg/mL Stenotrophomonas maltophilia 2 μg/mL 2 μg/mL

In total and for all infection sites, 20911 isolates were collected between 2013 and 2018 from 11 European countries. Out of the 20911 isolates, Enterobacterales represented 66.5 % of the pathogens, Acinetobacter spp. 12.7%, Burkholderia sp 0.7 %, Pseudomonas aeruginosa 16.1% and Stenotrophomonas maltophilia 3.9%.

Susceptibility for cefiderocol and the comparators was estimated in different subgroups of pathogens, suspected MDR/CR infections were defined as pathogens resistant to both ciprofloxacin and cefepime simultaneously.

Theoretical success in suspected MDR/CR infections was estimated for each antibacterial agent tested by combining the ECDC Epidemiological data of GN pathogen distribution in each individual infection site with the susceptibility to each antibacterial agent.

Table 48 and Table 49 (A-D) below summarize the results of such analyses for four different infection sites and the respective relevant comparators:

Table 48: Susceptibility to Cefiderocol and comparators in all sites of infections for MDR3 pathogens

Table 49- Theoretical success of antibacterial therapy in Gram‐negative 3MDR pathogens in gastrointestinal site of infections (A) Pneumonia; (B) cUTI; (C) BSI; (D) Gastrointestinal (A)

(B)

(C)

(D)

Regardless of the infection sites, cefiderocol demonstrated the highest theoretical success compared with meropenem, ceftolozane/tazobactam, ceftazidime/avibactam or colistin,for pre-emptive therapy in suspected MDR/CR infections.

Table 50: Summary table Theoretical percentage of success for Gram‐negative antibacterial therapy on aerobic Gram‐negative pathogens in different infection type

5.4.2 Individual study results (PK/PD data, study report S-649266-CPK-004-B)

This section summarizes the methodology, underlying assumptions, and main findings from extensive population PK/PD modelling efforts for ceficerocol. The full study report of the PK/PD population model (S-649266-CPK-004-B) is included in the submission.

5.4.2.1 Model description

A population pharmacokinetic (PK) analysis was performed to develop a model using a total of 3427 plasma concentration data of cefiderocol from the single ascending dose (SAD)/multiple dose (MAD) study (1203R2111), the renal impairment study (1222R2113), the phase 2 APEKS-cUTI study (1409R2121), the phase 3 CREDIBLE-CR study (1424R2131), and the phase 3 APEKS-NP study (1615R2132).

A 3-compartment model was used to describe the plasma concentrations of cefiderocol. The covariates explored included creatinine clearance calculated by Cockcroft-Gault equation (CrCL), body weight, age, albumin concentration, aspartate aminotransferase , alanine aminotransferase, total bilirubin, sex, race, infection (no infection, complicated urinary tract infection [cUTI] or acute uncomplicated pyelonephritis [AUP] in the phase 2 study, cUTI in the CREDIBLE-CR study, bloodstream infections/sepsis [BSI/sepsis], either hospital-acquired pneumonia [HAP]/ventilator-associated pneumonia [VAP]/healthcare-associated pneumonia [HCAP] in the CREDIBLE-CR study, and HAP/VAP/HCAP in the APEKS-NP study), and ventilation (with or without mechanical ventilation during PK sampling). CrCL was the most significant covariate on cefiderocol total clearance (CL), as expected. Observed plasma cefiderocol concentrations were adequately described by the developed final model.

5.4.2.2 Results

Individual maximum concentration (Cmax), daily area under the concentration-time curve (AUC) at steady state, percentage of time for which free drug concentration in plasma exceeds minimum inhibitory concentration (MIC) over dosing interval (%fT>MIC), and the %fT>MIC with MIC of 4 μg/mL (%fT>4) were predicted using Monte Carlo simulation of 1000 virtual patients for each infection sites. The ELF concentrations of cefiderocol in patients with pneumonia were predicted using a developed ELF model based on the ELF concentrations in 20 healthy subjects and 7 ventilated patients with pneumonia. The ELF compartment was linked with plasma compartment in the population PK model. The probabilityPTAPTAprobabilityPTA of target attainment for cefiderocol in ELF was also estimated usingwithwithusingwith Monte-Carlo simulations. A thousand virtual patients with pneumonia were generated for the simulations [251].

Probability of target attainment (PTA) were simulated to achieve 75%T>MIC for the different patient population in plasma and in ELF, depending on the renal function and dose adjustment required.

Table 51: PTA per infectious disease renal function, and dose

Target PK variable Infection disease MIC (µg/mL) a fT>MIC Renal Function Regimen 0.25 0.5 1 2 4 8 16

75% Plasma HAP/VAP/HCAP Augmented 2 g q6h 100 100 100 100 99.7 94.5 60.4

Normal 2 g q8h 100 100 100 99.9 98.9 87.1 43.4

Mild 2 g q8h 100 100 100 100 99.8 97.0 69.7

Moderate 1.5 g q8h 100 100 100 100 99.9 98.7 83.3

Severe 1 g q8h 100 100 100 100 100 99.9 90.7

ESRD 750 mg q12h 100 100 100 100 100 99.6 86.3

BSI/sepsis Augmented 2 g q6h 100 100 100 100 99.4 91.3 49.6

Normal 2 g q8h 100 100 100 99.9 97.3 80.6 32.6

Mild 2 g q8h 100 100 100 99.9 99.6 94.4 57.7

Moderate 1.5 g q8h 100 100 100 100 99.9 98.0 74.8

Severe 1 g q8h 100 100 100 100 100 99.8 84.8

ESRD 750 mg q12h 100 100 100 100 100 99.2 79.2

cUTI/AUP Augmented 2 g q6h 100 100 100 100 99.9 96.9 73.3

Normal 2 g q8h 100 100 100 100 99.6 93.6 56.3

Mild 2 g q8h 100 100 100 100 99.8 98.4 81.2

Moderate 1.5 g q8h 100 100 100 100 100 99.6 90.4

Severe 1 g q8h 100 100 100 100 100 100 95.9

ESRD 750 mg q12h 100 100 100 100 100 100 91.6

ELF HAP/VAP/HCAP Augmented 2 g q6h 100 100 100 99.8 92.1 55.9 11.0

Normal 2 g q8h 100 100 100 99.6 88.5 45.3 6.8

Mild 2 g q8h 100 100 100 99.8 94.1 62.1 16.2

Moderate 1.5 g q8h 100 100 100 100 96.5 67.6 19.3

Severe 1 g q8h 100 100 100 99.9 98.0 75.4 27.1

ESRD 750 mg q12h 100 100 100 99.9 94.7 64.7 21.9 Augmented: CrCL ≥ 120 mL/min (120 to < 150 = 50%; ≥ 150 = 50%). Normal: CrCL 90 to < 120 mL/min. Mild: CrCL 60 to < 90 mL/min. Moderate: CrCL 30 to < 60 mL/min. Severe: CrCL 15 to < 30 mL/min. ESRD: CrCL 5 to < 15 mL/min.

PTA for 75% fT>MIC was above 97% for a MIC of 4 mg/L regardless of the site of infection or the renal function. In the ELF, PTA for 75% fT>MIC was above 88% for a MIC of 4 mg/L confirming the adequacy of the dosing regimen in the different patient populations.

5.4.3 Retrospective analysis of cefiderocol and comparators by population PK/PD simulation A retrospective analysis was performed comparing the probability of target attainment (PTA) for cefiderocol, ceftolozane/tazobactam and meropenem against Enterobacterales and Pseudomonas aeruginosa in a representative patient population at risk of MDR or carbapenem resistant infections. Published pharmacokinetic (PK) models for meropenem and ceftolozane/tazobactam, and an existing model for cefiderocol, were used with standard dosage regimens for simulating individual PK data. The intial list of comparators included ceftazidime/avibactam and colistin, but this proved not possible to include:  the model implemented for ceftazidime-avibactam could not be appropriately validated.  the model for colistin, required information about the correlation matrix, and the nature of the parameter values reported in the original colistin model article, which was not made available by the original model’s authors.

PTA for clinically relevant pharmacokinetic/pharmacodynamic (PK/PD) targets was calculated from steady state PK profiles for a range of minimum inhibitory concentrations (MICs). The calculated PTAs in plasma for the 3 antimicrobials were above 95% at their respective MIC corresponding to their EUCAST breakpoints confirming published results. Cumulative fractions of response (CFRs) were also calculated to estimate using European MIC distributions from the SIDERO surveillance selected for being resistant to two antibiotic classes (quinolone and cephalosporins) and thereby representative of a patient population at risk of MDR infections. CFR analysis was performed on selected European isolates already resistant to cefepime and ciprofloxacin. In this patient population infected with suspected MDR/CR pathogens, CFRs were 97.4% and 99.8% for cefiderocol for Enterobacterales and Pseudomonas spp. respectively. The cumulative fractional responses for cefiderocol against Enterobacterales and Pseudomonas spp. are considerably higher than seen for meropenem and ceftolozane-tazobactam. Despite the simulation of high dose, extended infusion of meropenem CFRs were 91.2% for Enterobacterales but only

68.4% for Pseudomonas spp. The dose simulated for ceftolozane/tazobactam is also a high dose applied for the treatment of nosocomial pneumonia however due to the selected suspected MDR/CR isolates CFRs were only 67.2% for Enterobacterales and 55.2% for Pseudomonas spp.

Table 52. Estimated CFR for MIC distributions corresponding to Enterobacterales and Pseudomonas spp. More simulation results for corresponding PTA, MIC and T>MIC target values are shown in Appendix C. The applied MIC distributions can be seen in Appendix D of the study report.

Cefiderocol Meropenem Ceftolozane-tazobactam*

MIC distribution: Cumulative fraction of response, CFR (%) xxx

Enterobacterales 97.4 91.2 67.2 Pseudomonas spp. 99.8 68.4 55.2

Meaningful comparisons could be made between the performances of the models for cefiderocol, meropenem and ceftolozane-tazobactam. The simulations showed a superior performance of cefiderocol against Enterobacterales and Pseudomonas spp in terms of cumulative fraction of response when compared with meropenem and ceftolozane/tazobactam (Table 52). It should be noted, though, that the meropenem model exhibited a very long terminal half-life for the drug in plasma, probably reflecting that the experimental data originate from a patient population where the majority of the patients had bloodstream infections.

5.4.4 Clinical study results (clinical outcomes) Each section begins with a summary of the results for the respective primary endpoint, followed by summaries of relevant secondary endpoints in the order shown below. Detailed results (e.g., stratifications by pathogen) are provided in 5.4.3 [262, 263]. The section on APEKS-cUTI contains additional results from a network-meta-analysis.

The Table 53 below summarizes the endpoint analyses requested by EUnetHTA in the scoping process. In the following sections, the results for each of these endpoints are presented, if applicable.

EA: early assessment; EOT: end of treatment; TOC: test of cure; FUP: follow up; EOS: end of study; NR: not relevant; NA: not applicable

Table 53: Endpoint Analysis as per EUnetHTA Request

Study Endpoint/Analysis Available Stratifi- Stratification Primary time points cation by pathogen endpoint? by infection site APEKS- Clinical outcome EA, EOT, NA No. Pathogen No cUTI TOC, FUP specific data Composite EA, EOT, detailed Yes (at TOC) microbiological TOC, and FUP results on file eradication and [262] cure Microbiological EA, EOT, No eradication TOC, FUP All-cause mortality NR No at day 14 and 28

Changes of MIC or At EOS appearance of resistant bacteria

Network meta- Clinical Cure Not relevant analysis (NMA) and microbiological eradication at TOC and FU

APEKS-NP Clinical outcome TOC NA Main No pathogens Microbiological EA, EOT, Main No eradication TOC, FUP pathogens All-cause Day 14, day Yes Yes, day 14 mortality 28 ACM CREDIBLE- Clinical outcome EOT, TOC, FU Yes Main Yes, for CR (only pathogens HAP/VAP/HCAP descriptive and non- and BSI/sepsis results) Microbiological EOT, TOC, FU Yes fermenters Yes, for cUTI eradication All-cause mortality Day 14, day Yes No (part of safety 28, EOS assessment in study protocol) Blue font: primary endpoint

5.4.4.1 APEKs-cUTI

APEKS-cUTI was a Phase 2, multicentre (multinational), double-blind, randomised, active- controlled, parallel-group study including 452 patients diagnosed with complicated urinary tract infection [51] Clinicaltrials.gov Record NCT02321800).

The Figure 29 below summarizes the trial design and displays relevant endpoints.

Figure 29: APEKS-cUTI study design and endpoints

5.4.4.1.1 APEKS-cUTI primary efficacy endpoint: composite clinical and microbiological response at TOC in the mITT population

Primary efficacy endpoint analysis

In accordance to the FDA guidelines, the primary efficacy endpoint is the composite of clinical outcome and microbiological outcome at TOC. The response rate for the primary efficacy endpoint was 72.6% (183/252) of subjects in the cefiderocol group and 54.6% (65/119) of subjects in the IPM/CS group.

In a post-hoc analysis, the adjusted treatment difference was 18.58% (95% CI; 8.23%, 28.92%) in favor of cefiderocol (Figure 30) demonstrated superiority; it and met the criterion for noninferiority at the prespecified 20% and 15% margins (the lower limit of the 95% CI was 8.23% and exceeded both -15% and -20%). In addition, as it exceeded zero, which is consistent with the superiority of cefiderocol compared with IPM/CS this was further confirmed to be statistically significant (p=0.0004). Similar results were observed in the sensitivity analysis (composite clinical and microbiological response in ME population) [51, 236].

Table 54: Summary for Composite of Clinical and Microbiological Outcome by Time Point (Microbiological Intent-to-Treat Population)

Figure 30: Primary efficacy results: Composite outcome at TOC in the MITT population (Clinical and microbiological response) [a]Treatment difference (cefiderocol minus imipenem/cilastatin) is the adjusted estimate of the difference in the responder rate between the 2 treatment arms, calculated using a stratified analysis with Cochran- Mantel-Haenszel weights based on the stratified factor at baseline (cUTI with or without pyelonephritis vs acute uncomplicated pyelonephritis). CI, confidence interval; MITT, modified intent-to- treat; TOC, test of cure. Source: Portsmouth, 2018 [51]; Data on file [236]

Treatment differences by clinical diagnosis were consistent with the treatment difference in the mITT population, with cefiderocol demonstrating higher efficacy rates than IPM/CS in patients with cUTI with or without pyelonephritis, and with acute uncomplicated pyelonephritis (Figure 31) [51, 236]. The treatment differences by gender and age were also consistent with the treatment difference for the primary analysis (Figure 31) [51, 236].

Figure 31: Primary efficacy results: Composite outcome at TOC by predefined subgroups

aTreatment difference (cefiderocol minus imipenem/cilastatin) is the adjusted estimate of the difference in the responder rate between the 2 treatment arms, calculated using a stratified analysis with Cochran-Mantel-Haenszel weights based on the stratified factor at baseline (cUTI with or without pyelonephritis vs acute uncomplicated pyelonephritis); bMITT population included all patients who received at least one dose of study drug and had a qualifying baseline Gram-negative uropathogen (≥1×10⁵ CFU/mL). CFU, colony forming units; CI, confidence interval; cUTI, complicated urinary tract infection; MITT, modified intent-to-treat; NI, non- inferiority; TOC, test of cure. Source: Portsmouth S, et al. Lancet Infect Dis 2018;18:1319–28.

Composite clinical and microbiological response at TOC across different pathogens

For E. coli and Klebsiella spp. pneumoniae. the most frequently observed pathogens, the treatment difference between both arms was preserved. The composite of microbiological eradication and clinical response was higher in the cefiderocol group and the treatment difference was statistically significant with a treatment difference of 15.53% for E. coli and 25.91% for Klebsiella spp. (Table 55) [51, 236]. Other uropathogens occurred at a low frequency (in less than 10 patients in at least 1 of the groups) and therefore a statistical

comparison could not be made for either clinical response or microbiological eradication [51, 236].

Table 55: Composite of Clinical Response and Microbiological Outcome per Pathogen at TOC (microbiological ITT population) Percent Response % (n/N) Treatme Imipene nt Pathogen Cefidero m/ Differen col Cilastati ce n (95%CI) 74.0 58.4 15.53 E. coli (108/146) (45/77) (2.42-28.64) * K. 73.9 48.0 25.91 pneumoni (34/46) (12/25) (2.58-49.25) * ae P. 46.7 50.0 -3.33 aeruginos (7/15) (2/4) (NA) a P. 69.2 69.23 0.0 (0/1) mirabilis (9/13) (NA) CI - confidence Interval; NA, not available; * significant difference

Additional detailed stratified results for the primary endpoint by uropathogen are on file[262, 263].

5.4.4.1.2 APEKS-cUTI secondary endpoint: Clinical Outcome

Clinical Outcome per Subject at EA, EOT, TOC, and FUP

At TOC, clinical response was 89.7% (226/252) of subjects in the cefiderocol group and 87.4% (104/119) of subjects in the IPM/CS group. At FUP, sustained clinical response was higher in the cefiderocol group (81.3% [205/252] of subjects) than in the IPM/CS group (72.3% [86/119] of subjects), with an adjusted treatment difference of 9.02% (95% CI; -0.37%, 18.41%) (Table 56).

Clinical response rates for the ME Population (Table 14.2.5.1.2 of the CSR) were similar to the Micro-ITT Population, with sustained clinical response at FUP in the cefiderocol group (85.1% [194/228]) higher than in the IPM/CS group (78.3% [83/106]). Results were otherwise similar for both treatment groups and assessment time points for this population.

Table 56: Summary of Clinical Outcomes per Subject by Time Point (Microbiological Intent-to-Treat Population)

Stratified analyses: Clinical Outcome by Baseline Uropathogens at EA, EOT, TOC, and FUP

The summary of clinical outcome per uropathogen by time point (4 major uropathogens: E. coli, K. pneumoniae, P. aeruginosa, and P. mirabilis) for the Micro-ITT Population is shown in Table 57. For each of the major pathogen cefiderocol demonstrated no significant treatment difference in clinical outcome at any of the assessment time point. Additional detailed stratified analyses per pathogen and time point are on file [262, 263].

Table 57: Summary of Clinical Outcome per Uropathogen (E. coli, K. pneumoniae, P. aeruginosa, and P. mirabilis) by Time Point (Microbiological ITT Population)

5.4.4.1.3 APEKS-cUTI secondary endpoint: Microbiological outcome

The microbiological eradication rate in the mITT Population (Table 58) was statistically significantly higher at TOC in the cefiderocol group (73.0% [184/252] of subjects) compared with the IPM/CS group (56.3% [67/119] of subjects). The adjusted treatment difference of 17.25% (95% CI; 6.92%, 27.58%) in favor of the cefiderocol group was statistically significant and clinically meaningful. Results for both treatment groups were similar at EA and at EOT. The sustained microbiological eradication rate at FUP was also higher in the cefiderocol group (57.1% [144/252] of subjects) compared with the IPM/CS group (43.7% [52/119] of subjects). The adjusted treatment difference of 13.92% (95% CI; 3.21%, 24.63%) in favor of the cefiderocol group was statistically significant and clinically meaningful.

Table 58: Summary of Microbiological Outcome per Subject by Time Point (Microbiological ITT Population)

Results for both treatment groups showed no significant difference at EA and at EOT. The sustained microbiological eradication rate at FUP was also statistically significantly higher in the cefiderocol group (57.1% [144/252] of subjects) compared with the IPM/CS group (43.7% [52/119] of subjects).

The adjusted treatment difference of 17.83% (95% CI; 7.42%, 28.24%) for the microbiological eradication rate at TOC between the treatment groups in the ME Population (Table 14.2.2.1.2 of the CSR) was similar to the results in the Micro-ITT Population.

Microbiological Outcome per Uropathogen at EA, EOT, TOC, and FUP

The summary of microbiological outcome per uropathogen by time point (for the 4 most frequent uropathogens E. coli, K. pneumoniae, P. aeruginosa, and Proteus mirabilis) for the Micro-ITT Population is shown in Table 59.

Summaries of microbiological outcomes per uropathogen by time point for all uropathogens and per uropathogen group by time point are on file[262, 263]. For the most frequently isolatedisolated uropathogens, E. coli and K. pneumoniae, eradication at EA and EOT was not different between the treatment groups. For E. coli at TOC and FUP an adjusted treatment difference of 16.77% and 18.10%, respectively, was demonstrated, and this difference is consistent with the microbiological responses in the overall population. For K. pneumoniae, an adjusted treatment difference of 23.00% at TOC was observed, followed by a treatment difference of 6.33% at FUP.

These results demonstrate the microbiological efficacy of cefiderocol, which is consistently better than IPM/CS for these uropathogens.

Similar responses were seen in the ME Population for both pathogens; however, for K. pneumoniae, the difference between the 2 treatment groups for sustained eradication at FUP was minimal (3.13%) [237].

Table 59: Summary of Microbiological Outcome per Uropathogen (E. coli, K. pneumoniae, P. aeruginosa, P. mirabilis) by Time Point (Microbiological ITT Population)

5.4.4.1.4 APEKS-cUTI secondary endpoint: New Infection and Superinfection during the Study

No new infections were noted. Superinfection, defined as an uropathogen emerging during study drug therapy, was limited to a single occurrence of E. coli in 1 subject (Subject 143-002) in the cefiderocol group. The subject had E. coli isolated from the urine at the EA visit (Table 14.2.4.1.1 of the CSR). Of note, this subject had P. aeruginosa alone isolated at baseline and was treated for 10 days with cefiderocol. The E. coli superinfection was sensitive to levofloxacin, cefepime, and IPM, and the MIC for cefiderocol was 0.12 μg/mL. Both E. coli and P. aeruginosa were eradicated at TOC, and P. aeruginosa alone was isolated at FUP.

There were 8.3% (21/252) of subjects in the cefiderocol group and 15.1% (18/119) of subjects in the IPM/CS group who had new uropathogens that emerged after the EOS drug therapy in the mITT Population [262, 263]. Numbers of isolates for each pathogen identified and tested for susceptibility were small (the largest number was for E. coli, with 8 isolates in the cefiderocol group and 3 isolates in the IPM/CS group); hence no meaningful comparisons or conclusions could be made.

In conclusion, the cUTI study demonstrated that in a hospitalized population of 448 patients, with multiple comorbidities and difficult-to-treat infections caused by MDR pathogens sensitive to imipenem, cefiderocol was non-inferior to a standard-of-care antibacterial comparator, IPM/CS. Although the study was only designed to demonstrate non-inferiority, the findings of a post-hoc analysis were consistent with superiority for cefiderocol. The adjusted treatment difference favored cefiderocol and the lower limit of the 95 % confidence interval exceeded 0. The absolute difference of 18.58% in the composite primary endpoint was supported across all analyzed populations and clinical diagnostic groups with cUTI. The magnitude of the observed treatment differences is considered clinically important. Sensitivity analyses also showed that cefiderocol had better microbiological efficacy than IPM/CS in predefined subgroups and in patients infected with E. coli and K. pneumoniae, the most prevalent uropathogens.

5.4.4.2 APEKS-cUTI Network Meta-Analysis (NMA)

Feasibility for an NMA was conducted based on APEKS cUTI study (full detailes of the SLR and feasibility assessment can be found in [227]). The patient populations and most importantly, the pathogens included in the different trials were similar, enabling a small NMA to be conducted based on data from a YHEC SLR and feasibility assessment. The following analyses were conducted using a fixed effects model (frequentist analysis) as well as a Bayesian analysis:

 Microbiological eradication at TOC and at FU visit

 Clinical cure at TOC visit and FU visits

 Any adverse event

 Any ‘drug related’ adverse event

It should be noted that several of the outcomes and timepoints were infeasible due to insufficient reporting and 100% events (the equivalent of 0 events). Full report of the NMA analysis are provided as an attachment [264].

Also, the Figure 32 below reflects the maximum network diagram, of which TANGO II study was then removed because of significant differences the baseline patient population (patients were included after CR confirmation), therefore preventing the inclusion of TANGO I and ZEUS trials in the network:

Figure 32: Maximum Network Chart for Network Meta-analysis

The most extensive network could be constructed for the microbiological eradication secondary outcome is displayed in Figure 33 below:

Figure 33: Network Diagram for Microbiological Eradication Secondary Outcome

Legends: BAT: best available therapy C_T: ceftazalone-tazobactam CZA: ceftaz idime-avibactam DOR: doripenem FDC: cefiderocol IPM_CIL: imipenem/cilastatin LVX: levofloxacin

Consistent with the clinical trials results, the results in Figure 34 show a trend favouring cefiderocol, and there were significant differences in microbiological eradication rates at TOC between cefiderocol and imipenem/cilastatin and BAT in the frequentist analysis, respectively. This is due to superior results in the APEKS-cUTI for cefiderocol vs IPM-CIL (73% vs 56%) compared to Vasquez for CZA vs IPM-CIL (67% vs 63%). The Bayesian analysis (on the right) is consistent with frequentist analysis, showing the same trends but without reaching statistical significant difference:

Figure 34: Microbiological Eradication Rates at TOC - Frequentist Analysis

Legends: BAT: best available therapy C_T: ceftazalone-tazobactam CZA: ceftaz idime-avibactam DOR: doripenem FDC: cefiderocol IPM_CIL: imipenem/cilastatin LVX: levofloxacin

Figure 35: Microbiological Eradication Rates at TOC - Bayesian Analysis

Similar results and trends were obtained for the microbiological eradication at follow-up in a smaller network.

Clinical cure endpoint was evaluated in a smaller network as shown in Figure 36 below.

Figure 36: Network Diagram for Clinical Cure Outcome

The safety analysis based on “any adverse event” resulted in the largest network (Figure 37):

Figure 37: Clinical cure rates at TOC - Frequentist Analysis

Legends: BAT: best available therapy C_T: ceftazalone-tazobactam CZA: ceftaz idime-avibactam DOR: doripenem FDC: cefiderocol IPM_CIL: imipenem/cilastatin LVX: levofloxacin

Figure 38: Clinical Cure rate at TOC - Bayesian Analysis

Figure 39: Clinical cure rates at FU - Frequentist Analysis

The results did not show any statistically significant difference in clinical cure rates at TOC between cefiderocol and comparators both frequentist (Figure 37) and Bayesian analysis. However, in the analysis for clinical cure in the FU visit, show

a trend favouring cefiderocol, and there were statistically significant differences in clinical cure rate at FU between cefiderocol and imipenem/cilastatin in the frequentist analysis (Figure 39). Again.this is due to the superior results in APEKS cUTI for cefiderocol vs IPM-CIL (81% vs 72%) compared to Vasquez for CZA vs IPM-CIL (74% vs 67%). The Bayesian analysis (Figure 38) is consistent with frequentist analysis, showing the same trends but without reaching statistical significant difference.

5.4.4.3 APEKs-NP clinical outcomes

The APEKS-NP study compared treatment with cefiderocol against high-dose, prolonged infusion (HD) meropenem in patients with nosocomial pneumonia caused by suspected MDR Gram-negative pathogens. 300 patients were randomized 1:1 to cefiderocol or HD meropenem, a regimen only used in more difficult-to-treat pathogens which optimizes exposure and efficacy for meropenem (Figure 40).

Figure 40: APEKS-NP study design

The dose of meropenem was increased from the labeled dose of 1 g to 2 g and extended to a 3-hour infusion to optimize the antibacterial activity of meropenem, at the request of regulators. [265].

5.4.4.3.1 APEKS-NP primary efficacy endpoint: Day-14 ACM

Cefiderocol demonstrated noninferiority to high-dose extended infusion meropenem with regard to all-cause mortality at Day 14. The all-cause mortality rate was 12.4% (18/145 subjects) for the cefiderocol group and 11.6% (17/146 subjects) for the HDhigh-doseHD meropenem group, demonstrating the noninferiority of cefiderocol, as the upper limit of the 95% CI was < 12.5% (95% CI: −6.6, 8.2) (Figure 41).

Figure 41: All-cause Mortality (mITT)

[a] Treatment difference (cefiderocol minus meropenem) is the adjusted estimate of the difference in the all-cause mortality rate at Day 14 and Day 28 between the 2 treatment arms based on Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor.; [b] The 95% CI (2-sided) is based on a stratified analysis using Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor. The CI was calculated using a normal approximation to the difference between 2 binomial proportions (Wald method). Source: Data on file [239]

Table 60: Day 14 All-cause Mortality (mITT and ME-PP Populations)

HD Treatment Differencea Cefiderocol Meropenem Total Differenc Population n/N’ (%) n/N’ (%) n/N’ (%) e (%) 95% CIb p-value mITT N = 145 N = 147 N = 292 0.8 (−6.6, 8.2) 0.0020c 18/145 17/146 (11.6) 35/291 0.8321d (12.4) (12.0) ME-PP N = 105 N = 101 N = 206 −0.3 (−9.4, 8.7) nc 13/105 13/100 (13.0) 26/205 (12.4) (12.7) mITT excl N = 145 N = 147 N = 292 −1.3 (−10.1, nc meropenem 9/91 (9.9) 10/90 (11.1) 19/181 7.5) resistante (10.5) APACHE II = Acute Physiology and Chronic Health Evaluation II; CI = confidence interval; CLSI = Clinical and Laboratory Standards Institute; Day 14 ACM = all-cause mortality at Day 14 since first infusion of study drug; excl = excluding; ME-PP = microbiologically-evaluable per-protocol; mITT = modified intent-to-treat; n = number of subjects who died; nc = not calculated; N = number of subjects in the analysis set; N’= number of subjects with known survival status [a] Treatment difference (cefiderocol minus meropenem) is the adjusted estimate of the difference in the all-cause mortality rate at Day 14 and Day 28 between the 2 treatment arms based on Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor. [b] The 95% CI (2-sided) is based on a stratified analysis using Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor. The CI was calculated using a normal approximation to the difference between 2 binomial proportions (Wald method). [c] p-value for non-inferiority hypothesis. [d] p-value for the superiority hypothesis. [e] Meropenem-resistant subjects were those subjects whose baseline Gram-negative pathogens were resistant to meropenem based on CLSI susceptibility results. Subjects who did not have any susceptibility results available at baseline based on CLSI were not included for this analysis. Source: Tables 14.2.1.1.1, 14.2.1.1.3, and 14.2.1.1.4

The sensitivity analysis of Day 14 all-cause mortality using the ME-PP population is in support of the noninferiority finding in the primary efficacy population (Table 60). In a supplementary analysis of the primary endpoint, in which subjects who were resistant to HD meropenem were

excluded from the mITT population, Day 14 all-cause mortality was 9.9% in the cefiderocol group and 11.1% in the HD meropenem group (Table 60).

Subgroup analyses revealed no statistically significant differences between the included groups (Figure 42).

Figure 42: Primary efficacy results: Day 14 All-cause Mortality by Subgroups

Source: Data on file [239]

5.4.4.3.2 Secondary efficacy endpoints

Rates of microbiological eradication and clinical cure at TOC confirmed the non-inferiority between the treatments (Table 61). The microbiological eradication at TOC was 47.6% (59/124) in the cefiderocol group and 48.0% (61/127) in the HD meropenem group, and the clinical cure at TOC was 64.8% (94/145) in the cefiderocol group and 66.7% (98/147) in the HD meropenem group.

Table 61: Secondary Endpoints (mITT Population)

HD Treatment Comparison Cefiderocol Meropenem Total (N = 145) (N = 147) (N = 292) Differenc Endpoint n/N’ (%) n/N’ (%) n/N’ (%) e (%) 95% CI Microbiological 61/127 eradication at 59/124 (47.6) 120/251 (47.8) -1.4 a (-13.5, 10.7) a (48.0) TOC Clinical cure at 98/147 94/145 (64.8) 192/292 (65.8) -2.0 a (-12.5, 8.5) a TOC (66.7) Day 28 all- 30/146 30/143 (21.0) 60/289 (20.8) 0.5b (-8.7, 9.8)b cause mortality (20.5) EOS all-cause 34/146 38/142 (26.8) 72/288 (25.0) 3.6b (-6.3, 13.4)b mortality (23.3) APACHE II = Acute Physiology and Chronic Health Evaluation II; CI = confidence interval; EOS = end of study; mITT = modified intent=to-treat; TOC = test of cure [a] Treatment difference (cefiderocol minus meropenem) is the adjusted estimate of the difference in the eradication rate or cure rate between the 2 treatment arms. The adjusted difference estimates and the 95% CIs (2-sided) were calculated using a stratified analysis with Cochran-Mantel-Haenszel weights based on the stratified factors at baseline, infection type (HABP/VABP/HCABP), and APACHE II score (≤ 15 and ≥ 16). [b] Treatment difference (cefiderocol minus meropenem) is the adjusted estimate of the difference in the all-cause mortality rate at Day 28 or at the EOS visit between the 2 treatment arms based on Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor. The 95% CI (2-sided) is based on a stratified analysis using Cochran-Mantel Haenszel weights using APACHE II score (≤ 15 and ≥ 16) as the stratification factor. The CI is calculated using a normal approximation to the difference between 2 binomial proportions (Wald method). Source: Tables 14.2.2.1.1, 14.2.3.1.1, 14.2.1.1.1, and 14.2.4.1.1

Cefiderocol has demonstrated similar all-cause mortality at Day 28, clinical and microbiological outcomes to HD meropenem (Table 62) [239]. The microbiological eradication at TOC was 47.6% (59/124) in the cefiderocol group and 48.0% (61/127) in the HD meropenem group and the clinical cure at TOC was 64.8% (94/145) in the cefiderocol group and 66.7% (98/147) in the HD meropenem group [239]. All-cause mortality at Day 28 and at EOS was also similar between the treatment groups [239].

Table 62: Secondary Endpoints (mITT Population)

HD meropenem

Cefiderocol (N = 147) Endpoint (N = 145) n/N’ (%) n/N’ (%)

Microbiological eradication at TOC 59/124 (47.6) 61/127 (48.0)

Clinical cure at TOC 94/145 (64.8) 98/147 (66.7)

Day 28 all-cause mortality 30/143 (21.0) 30/146 (20.5)

EOS all-cause mortality 38/142 (26.8) 34/146 (23.3)

EOS, end of study; TOC, test of cure Source: Data on file [239]

Efficacy data across different pathogens A similar response between cefiderocol and HD meropenem was observed across different pathogens (Table 63) [239].

Table 63: Clinical and microbiological outcome per baseline pathogen

HD Treatment comparison Cefiderocol Meropenem Difference (n=145) 95% CI (n=147) (%) Clinical cure K. pneumoniae 31/48 (64.6) 29/44 (65.9) −1.3 (−20.8, 18.1) at TOC (mITT) P. aeruginosa 16/24 (66.7) 17/24 (70.8) −4.2 (−30.4, 22.0) A. baumannii 12/23 (52.2) 14/24 (58.3) −6.2 (−34.5, 22.2) E. coli 12/19 (63.2) 13/22 (59.1) 4.1 (−25.8, 33.9) Microbiological K. pneumoniae 22/48 (45.8) 24/44 (54.5) −8.7 (−29.1, 11.7) eradication P. aeruginosa 9/24 (37.5) 11/24 (45.8) −8.3 (−36.1, 19.5) at TOC (mITT) A. baumannii 9/23 (39.1) 8/24 (33.3) 5.8 (−21.7, 33.2) E. coli 10/19 (52.6) 11/22 (50.0) 2.6 (−28.0, 33.3) HD, high-dose; TOC, test of cure; Source: Data on file [239]

5.4.4.3.3 Efficacy data based on susceptibility to meropenem

In a subgroup analysis including a small sample of patients with meropenem-non-susceptible Gram-negative pathogens (as per CLSI break point of 8mg/L), post hoc analyses of subjects with values of > 16 μg/mL, > 32 μg/mL, and > 64 μg/mL, showed a trend of lower mortality in the cefiderocol group than in the meropenem group at Day 14 and Day 28; however, the sample sizes are too small to draw definitive conclusions (Figure 43) [239].

Figure 43: Day 14 and Day 28 all-cause mortality according to MIC for meropenem

HD, high dose; MIC, minimum inhibitory concentration; Source: Data on file [239]

Microbiological and clinical outcomes at TOC in the subgroup of meropenem-nonsusceptible subjects are shown in Table 64. The meropenem–nonsusceptible subgroup includes intermediate and resistant categories of susceptibility. At TOC, the microbiological eradication rate was 40.0% (14/35) in the cefiderocol group and 33.3% (10/30) in the meropenem group, and the clinical cure rate was 57.1% (20/35) in the cefiderocol group and 56.7% (17/30) in the meropenem group.

Table 64: Microbiological and Clinical Outcome for the Meropenem-non- susceptible Subgroup (mITT Population)

HD Treatment Cefiderocol Meropenem Total Comparisonb Meropenem- (N = 145) (N = 147) (N = 292) nonsusceptibl n/N’ (%) n/N’ (%) n/N’ (%) Differenc e Status = Yesa (N’ = 35) (N’ = 30) (N’ = 65) e (%) 95% CI Microbiological (-16.7, eradication at 14 (40.0) 10 (33.3) 24 (36.9) 6.7 30.1) TOC Clinical cure at (-23.7, 20 (57.1) 17 (56.7) 37 (56.9) 0.5 TOC 24.6) CI = confidence interval; CLSI = Clinical and Laboratory Standards Institute; EOS = end of study; mITT = modified intent=to-treat; N’ = number of meropenem-nonsusceptible subjects; TOC = test of cure [a] The meropenem-nonsusceptible status for subjects was Yes if for any baseline Gram-negative pathogens (including Stenotrophomonas maltophilia) the CLSI results were nonsusceptible to meropenem. Subjects who did not have any susceptibility results available at baseline based on CLSI were not included for this analysis. [b] Treatment difference is cefiderocol minus meropenem. The 95% CIs (2-sided) of treatment difference were calculated using a normal approximation to the difference between the 2 binomial proportions (Wald method). The CIs for cure rates within a visit with less than 10 subjects in any treatment arm are not presented. Source: Tables 14.2.2.1.4 and 14.2.3.1.4

When analyzing microbiological eradication rates based on different CLSI MIC for meropenem in the non-susceptible group, data suggests that cefiderocol retains microbiological eradication as MIC for meropenem increases, whereas for it HD meropenem decreased (Figure 44).

Figure 44: Microbiological eradication by MIC at EOT

EOT, end of treatment; HD, high dose; MIC, minimum inhibitory concentration; Source: Data on file [239]

Feasibility for NMA in nosocomial pneumonia: A feasibility assessment was carried out for an NMA for APEKS-NP trial. It proved not to be possible to conduct an NMA, given that the comparator used in APEKS NP trial (HD meropenem) was not used in other trials alone, and there was no bridging study. Even though the molecule is the same, this higher dose and prolonged infusion optimizes efficacy of meropenem. In addition, APEKS NP included difficult to treat pathogens such as Acinetobacter baumannii, which are not included in other clinical trials because they are not susceptible to the newer drugs. For full information on the feasibility assessment please refer to [227].

5.4.4.4 Comparative analysis of estimated success rates considering the European pathogen epidemiology in the population with suspected MDR/CR infections

In the absence of antibiogram, cefiderocol provides the best predicted susceptibility rates and estimated success rates considering the European pathogen epidemiology

When critically ill patients require immediate treatment in the absence of AST, the likelihood of treatment success with cefiderocol and comparators can be predicted through a simple effectiveness model, that projects the clinical trials outcomes in terms of microbiological eradication and clinical cure for each of antimicrobials, for a scenario where an antimicrobial prescription is required in the absence of an antibiogram for a suspected MDR pathogen. This analysis is therefore based on epidemiology (pathogen prevalence estimates for the specific site of infection, taken from eCDC) and pathogen susceptibility results (taken from the SIDERO studies, when selecting pathogens already resistant to ciprofloxacin and cefepime), relying on drug’s ability to achieve effective concentrations to the infections site. This weighed susceptibility is then overlaid with the individual relevant antimicrobial outcomes in the clinical trials for each infection site.

Results for cUTI and pneumonia are shown below [266-269].

Table 65: Susceptibility and effectiveness model predicting outcomes for Cefiderocol versus comparators in UTI

Microbiological Projected Microbiological Clinical Cure at Projected Clinical Susceptibility eradication at TOC (m- eradication at TOC in the TOC from Cure at TOC in UTI *EPI ITT) from clinical trials Suspected population clinical trials the Suspected antimicrobial cefiderocol 94.28% 73.00%1 68.82% 89.70%1 84.57% ceftolozane /tazobactam 63.87% 80.40%2 51.35% 92.00%2 58.76% ceftazidime /avibactam 84.79% 77.40%3 65.63% 70.20%3 59.53%

Source: 1-APEKS cUTI trial; 2- EPAR for Zerbaxa [270] ; 3 RECAPTURE [271], Results from this effectiveness model analysis showed that cefiderocol has a higher predicted susceptibility rates in the European prevalent Gram-negative bacteria than comparators in cUTI and higher projected treatment success rates both microbiological eradication and clinical cure (Table 65). Given the higher susceptibility rates for cefiderocol, these results are generally consistent with actual results from the APEKS cUTI trial, but not for comparators as the analysis included pathogens for which they are not susceptible, situation that can occur in the need to immediate treatment in the absence of an antibiogram.

Table 66: Susceptibility and effectiveness model predicting outcomes for Cefiderocol versus comparators in Pneumonia

Microbiological Projected Projected Microbiological Clinical Cure at eradication at Clinical Cure Pneumonia eradication at TOC in the TOC from Susceptibility TOC (m-ITT) from at TOC in the Suspected population clinical trials *EPI clinical trials Suspected antimicrobial cefiderocol 92.70% 47.60% 44.13% 64.8% 60.07% meropenem 58.29% 50.00%1 29.15% 65.1% 37.95% (MIC>8mg/mL) ceftolozane/tazobactam 47.55% 73.10% 34.76% 54.4% 25.87% ceftazidime/avibactam 65.27% 54.00% 35.25% 68.8% 44.91%

Sources for cefiderocol and comparator data: APEKS-NP (1- for meropenem only results of the subgroup MIC>8mg/mL were considered); EPAR for Zerbaxa [270], Torres 2018 [272], Torres 2019 [273]

Furthermore, for pneumonia, results from this effectiveness model analysis showed that cefiderocol has a higher predicted susceptibility and higher predicted treatment success rates from both a clinical and microbiological perspective. These results are generally consistent with actual results from the APEKS NP clinical trials, but not for comparators as the analysis included pathogens for which they are not susceptible (Table 66). Even though a similar breakpoint of 8mg/L was considered for both APEKS NP and susceptibility analysis in SIDERO studies, the high dose, prolonged infusion meropenem regimen used in APEKS NP trial, showed to be effective in pathogens with MICs up to 16mg/ml, reason why the results observed in this effectiveness model and the clinical trial for the meropenem susceptible group are different.

Such methodologies are required, when ethical considerations limit clinical trials design to intendedly risk exposing patients to potentially ineffective drugs. Also, since NMAs are based on the non-inferiority clinical trials results (which excludes non-susceptible pathogens) the results obtained between the 2 methodologies are therefore understandably different, but consistent:

 the effectiveness model highlights the potential difference in effectiveness between different drugs, obtained when antimicrobial prescription is needed in the absence of antibiogram

 the NMA reinforces the notion that similar results are obtained between drugs when comparing effectiveness in similar patient population with similar pathogen distribution (i.e. pathogen is sensitive to both drugs, and both drugs reach the infection site in effective concentrations, which can occur when antibiogram is available and prescription is targeted).

Also to note that the results of this effectiveness model will vary according with the local epidemiology, and changes in susceptibility patterns for each of the drugs.

5.4.4.5 CREDIBLE-CR

The CREDIBLE CR study was a small, descriptive, randomised, open label, descriptive study conducted to evaluate efficacy in patients with confirmed CR infections for cefiderocol and BAT, not designed or powered for statistical comparison between arms (Figure 45). The study included 150 severely ill patients randomised 2:1 between the treatment groups, consistent with compassionate use cases, with a range of infection sites including nosocomial pneumonia, cUTI, BSI/sepsis. Many patients had end stage comorbidities and had failed multiple lines of therapy.

This study, alongside with the compassionate use cases inform the efficacy of cefiderocol in the population with confirmed CR infection.

Figure 45: CREDIBLE CR study design

5.4.4.5.1 Primary endpoint analysis

Primary endpoint for HAP/VAP/HCAP and BSI/sepsis: clinical cure rate

Primary endpoint for cUTI: microbiological outcome

Results of clinical cure and microbiological eradication were similar between arms in each point in time, with the highest differences being observed in patients with cUTI and follow-up visit. One should remeber that this is a descriptive study without any formal comparison, and furthermore, the number of patients in each group is too small to derive any conclusions other than that cefiderocol demonstrated activity, from both a clinical and microbiological outcomes, in all 3 infection sites (Figure 46 and Figure 47) [242].

Figure 46: Clinical cure by Clinical Diagnosis and time point

BAT, best available therapy; BSI, bloodstream infection; cUTI, complicated urinary tract infection; EOT, end of treatment; FU, follow-up; HAP, hospital-acquired pneumonia; HCAP, healthcare-associated pneumonia; Micro- ITT, microbiological intent-to-treat; TOC, time of cure; VAP, ventilator-associated pneumonia; Source: Data on file [242]

Figure 47: Microbiological eradication by Clinical Diagnosis and time point

5.4.4.5.2 Efficacy data across different pathogens

Outcomes by pathogen were broadly similar between the treatment groups. Cefiderocol has demonstrated efficacy across all main pathogens [242].

Table 67: Clinical cure and microbiological eradication by baseline CR-pathogen

CR Micro-ITT population Cefiderocol BAT (n=80) (n=38) 42/80 Clinical cure 19/38 (50.0) (52.5) 16/37 CR A. baumannii 9/17 (52.9) (43.2) CR P. aeruginosa 7/12 (58.3) 5/10 (50.0) 18/27 CR K. pneumoniae 6/12 (50.0) (66.7) 25/80 Microbiological eradication 9/38 (23.7) (31.3) 10/37 CR A. baumannii 5/17 (29.4) (27.0) CR P. aeruginosa 1/12 (8.3) 2/10 (20.0) 13/27 CR K. pneumoniae 3/12 (25.0) (48.1) Source: Data on file [242]

Against the most difficult-to-treat pathogens with New Delhi metallo-β-lactamase (NDM), metallo-betalactamases or porin channels mutations, cefiderocol showed to be an effective treatment presenting similar or better clinical and microbiological outcomes than BAT. There were eight NDM producing Enterobacteriaceae in the cefiderocol arm and four in the BAT arm. Six out of eight patients in cefiderocol arms had a clinical cure and microbiological response. Of the four in the BAT arm, none responded (Figure 48). There were 14 KPC producers in the cefiderocol group and seven in the BAT group. The clinical and microbiological responses were similar between treatment groups (Figure 48). Porin channel mutations were present in 15 pathogens in the cefiderocol group and 9 in the BAT group with similar clinical responses. Microbiological eradication was demonstrated in seven out of 15 pathogens in cefiderocol arm and one out of nine in BAT arm (Figure 48). In patients with infections caused by metallo- betalactamase producing Gram-negative pathogens, cefiderocol demonstrated benefits for both clinical cure and microbiological responses (Figure 49), but again, numbers are too small to derive any conclusion.

Figure 48: Clinical and Microbiological Outcomes at TOC in Enterobacteriaceae by Carbapenemase or Porin Channel Mutation (CR Micro-ITT Population)

*OMPK35/36-deficient. Only patients with molecular data are included.

Figure 49: Clinical and Microbiological Outcomes in Metallo Β-lactamase Producing Gram-negative Pathogens (CR Micro-ITT Population)

5.4.4.5.3 CREDIBLE-CR all-cause Mortality Data

Mortality was evaluated as part of safety assessment in the study, however, as per EUnetHTA request, it is presented within the efficacy outcomes.

An imbalance in mortality favouring the BAT arm was observed at all time points in the study. Table 68 and Figure 50 includes a summary of death in all subjects and by infection type at each time point. Twenty-eight-day mortality represents a fixed time point for all patients and is

a conventional endpoint to assess both safety and efficacy in antibacterial studies. The Day 28 mortality for all subjects (i.e., all infection sites combined) was 24.8% in the cefiderocol group and 18.4% in the BAT group. This difference was observed in pneumonia and BSI patients, but not in subjects with cUTI.

Through EOS (Day 49), mortality for all subjects (ie, all infection sites combined) was 33.7% in the cefiderocol group and 18.4% in the BAT group.

Table 68: Summary for All-cause Mortality in the Study (Intent to treat Population)

Cefiderocol BAT Infection Site (N = 101) (N = 49) All-cause Mortality Rate n/N (%) 95% CI n/N (%) 95% CI All Infection Sites Combined N' = 101 N' = 49 Day 14 19/101 (18.8) (11.7, 27.8) 6/49 (12.2) (4.6, 24.8) Day 28 25/101 (24.8) (16.7, 34.3) 9/49 (18.4) (8.8, 32.0) Through EOS 34/101 (33.7) (24.6, 43.8) 9/49 (18.4) (8.8, 32.0)

Figure 50: All-cause Mortality Rates by Type of Infection

BAT, best available therapy; BSI, bloodstream infection; cUTI, complicated urinary tract infection; HAP, hospital- acquired pneumonia; HCAP, healthcare-associated pneumonia; VAP, ventilator-associated pneumonia; Source: Data on file [242]

The time stratification analysis of the mortality data show that the imbalance occurs outside the treatment effective period: 4 deaths occurred in cefiderocol arm only at very early stages of treatment (up to day 3 when there was an early assessment), and 9 occurred after after TOC, which are more likely to be associated with the underlying condition of the patient

Considering the 2:1 randomization, there was no significant difference in mortality rates between day 4 and 28 (Table 68).

For completion of mortality information, there were 7 subjects in the CREDIBLE-CR study who died after study completion are provided on file [274]. There were 2 subjects treated with cefiderocol (Subjects 2AA001 and 3HK002) and 5 subjects treated with best available therapy (BAT; Subjects 3HN001, 3HJ001, 3HJ004, 3HM003, and 3FG010) who died after study completion. Neither of the 2 cefiderocol-treated subjects, but 3 of the 5 subjects treated with BAT (3HN001, 3HJ004, and 3FG010) had Acinetobacter baumannii as a causative pathogen at baseline.

The population in the CREDIBLE-CR study was designed to be very heterogeneous as it was a pathogen-focussed study which included subjects with many underlying conditions, different infection sites and infections due to a variety of Gram-negative pathogens. The study was relatively small (101 subjects treated with cefiderocol and 49 subjects treated with BAT) and due to the heterogeneity of the population the treatment groups do not appear to be balanced for baseline characteristics such as shock (which has a major impact on mortality) in the subgroup of subjects with A. baumannii infections. It is likely that the mortality imbalance observed is due to a variety of factors related to baseline imbalances. [275]

When considering baseline pathogens, mortality was lower in cefiderocol-treated subjects than BAT-treated subjects for the Enterobacteriaceae and the higher mortality was seen for the non- fermenters. Many subjects had co-infection with multiple non-fermenters (Table 69). The difference seen in non-fermenters was mostly due to the difference seen with A. baumannii. The mortality rate for subjects with P. aeruginosa alone without Acinetobacter spp. as a co- pathogen was the same in each treatment group being 18.2% (2/11 subjects) for cefiderocol and 18.2% (2/11 subjects) for BAT [50].

Table 69: Summary for all-cause mortality overall by pathogens subgroup (Enterobactereacea and non-fermenters)

Source: Response to D90 [276]

In subjects with A. baumannii infection and a history of shock (both shock at baseline and a history of shock within 31 days of baseline), mortality rates were much higher than in subjects without a history of shock in both treatment groups [277]. The proportion of subjects with a history of shock was higher for the cefiderocol group than for the BAT group and so given the high mortality rates reported for subjects with shock, the increased incidence of a history of shock in cefiderocol-treated subjects with A. baumannii may provide an explanation for some of the difference in mortality rates between the treatment groups in the CREDIBLE-CR study (Table 70).

Table 70: CREDIBLE-CR study: Mortality subgroup Analysis for Subjects with A. baumannii (safety population)

Cefiderocol (N=39) BAT (N=17) All-cause All-cause mortality mortality Subgroup N’/N (%) n/N' (%) N’/N (%) n/N' (%) Overall 39 19/39 (48.7) 17 3/17 (17.6) Shock within 31 days of baseline Yes 9/39 (23.1) 7/9 (77.8) 1/17 (5.9) 1/1 (100) No 30/39 (76.9) 12/30 (40.0) 16/17 (94.1) 2/16 (12.5) Shock ongoing at baseline Yes 7/39 (17.9) 6/7 (85.7) 1/17 (5.9) 1/1 (100) No 32/39 (82.1) 13/32 (40.6) 16/17 (94.1) 2/16 (12.5) ICU at baseline Yes 32/39 (82.1) 15/32 (46.9) 8/17 (47.1) 1/8 (12.5) No 7/39 (17.9) 4/7 (57.1) 9/17 (52.9) 2/9 (22.2) BAT = best available therapy; ICU = intensive care unit; N = number of subjects with A baumannii.; N’ = number of subjects in subgroup

5.4.4.5.4 Comparison of CREDIBLE-CR mortality with other studies

Whereas the mortality rate in the cefiderocol group was consistent with previous studies in similar populations with high levels of A. baumannii infections ([142, 278, 279]), the mortality rate in the BAT group was substantially lower than expected from previous studies (Figure 51) [142, 244, 275, 278-283]. The reason for the lower than expected mortality in the BAT group is not clear but is likely also due to a variety of factors related to baseline imbalances and other anomalies (such as the low mortality associated with high APACHE II and SOFA scores). The evidence suggests that the mortality rate in the BAT group was unexpectedly low for the population randomised and that the mortality in the cefiderocol group was consistent with what has been reported in previous studies.

Figure 51: Mortality rates comparison across studies

No other factors that indicated disease severity were identified that clearly contributed to the mortality imbalance seen between treatments [275].

In conclusion, the difference in mortality between treatments in the CREDIBLE-CR study still cannot be fully explained. However, there are plausible factors contributing to the mortality difference in this pathogen-focussed study. The 2:1 randomisation, the small study size, and the heterogeneity of the patient population, particularly the inclusion of multiple infection sites and diverse comorbidities, means that it was difficult to ensure that treatment groups were balanced for all baseline factors. Other than a history of shock, and possibly low WBC count, no individual baseline characteristic has been identified which could clearly be linked to this

imbalance. A mortality imbalance was not observed in the APEKS-NP study overall, including in the subset of subjects infected with A. baumannii [276]. The evidence suggests that the mortality rate in the BAT group was unexpectedly low for the population randomised and that the mortality in the cefiderocol group was consistent with what has been reported in previous studies.

5.4.4.6 Compassionate Use

To date, over 200 patients were treated with cefiderocol, worldwide, within compassionate use programme. Data for 74 patients who have completed therapy is available, of which only 3 positive outcomes are published to date [246-248]. This programme included patients with a diversity of infections beyond those presented in the clinical trials, with a baseline patient characteristics consistent with that of CREDIBLE CR as per previously detailed in section 5.3

5.4.4.6.1 Clinical efficacy and safety Over 60% of the patients receiving cefiderocol survived when no other possible treatment option were available to them [244]. Of these, 17 died due to their underlying infection, 6 died for reasons other than the original bacterial infection, and other causes of death remained unknown [244]. However, none of the observed deaths were considered to be related to cefiderocol [245]. Cefiderocol has demonstrated a manageable safety profile with the longest use being more than 90 days in a renal transplant patient where no apparent safety issues were observed [244].

Table 71: Mortality and serious adverse events

Mortality, n (%) Cefiderocol (n=74) Overall mortality 27 (36.5) Overall mortality by pathogen Acinetobacter baumannii 12/22 (54.5) Klebsiella pneumoniae 2/7 (28.6) Burkholderia cenocepacia 4/10 (40) Pseudomonas aeruginosa 9/31 (29) Abnormal LFTs 4 (5.4) Multiple organ failure 6 (8.1) Acute renal failure 3 (4.1) Cardiac arrest 2 (2.7) Sepsis or septic shock 5 (6.8) LFT, liver function test; SAE, serious adverse event; Source: NDA briefing document[244]; Data on file [245]

5.4.4.7 Published case reports

Case reports for three patients from the expanded access program have been published so far.

 A patient was treated successfully for endocarditis due to extensively drug resistant (XDR) Pseudomonas aeruginosa.(Edgeworth et al., 2019)[246]

 A patient with VAP and BSI caused by XDR Acinetobacter baumannii and carbapenemase-producing Klebsiella pneumoniae had potentially serious organ failure from older anti-infectives. Six weeks after cefiderocol administration, chest X-rays showed complete resolution of infection (Trecarichi et al., 2019)[248]

5.4.5 Resistance against Cefiderocol 5.4.5.1 In vitro resistance development

Resistance development to Cefiderocol has been investigated using the standard in vitro experiments to determine the frequency of spontaneous resistance and to observe the adaptation of pathogens during serial passaging. Frequency of spontaneous resistance: The frequency of spontaneous resistance of E. coli, E. cloacae, K. pneumoniae, and P. aeruginosa (8 strains in total) was determined in the presence of 10 × MIC of cefiderocol. If resistant mutants were isolated, the in vitro activity of cefiderocol against the mutant strains was determined and compared to the susceptibility of the parent strains. The magnitude of the order of frequency of the resistance for cefiderocol was similar to ceftazidime with a frequency of 10−7 to 10−8 except for P. aeruginosa for which the frequency ranged from 10−6 to 10−8. Cefiderocol MIC increase was shown to be associated with the mutation in the upstream region of pvdS (pyoverdine synthesis gene) and fadD3 (fatty acyl-CoA synthetase) in P. aeruginosa, and baeS, envZ, ompR (all are 2-component signal transduction gene), and exbD (biopolymer transport gene) in K. pneumoniae.

Resistance acquisition assay by serial passage: Resistance acquisition was evaluated for K. pneumoniae, and P. aeruginosa (5 strains in total) by a 10 times serial passage in two different media. The MIC of cefiderocol increased in general 1 to 4-fold but for one strain up to 8-fold.

Resistance acquisition by using an in vitro pharmacodynamic model: To estimate the risk of emergence of cefiderocol-resistant mutants during the treatment of patients, in vitro PD models simulating the free concentration-time curves in human plasma was used. The simulated concentration-time curves were determined for a 2-g cefiderocol q8h administration with 3-hour infusion, 2-g/0.5 g CAZ/AVI q8h administration with 2-hour infusion, and 1-g MEPM q8h administration with 1-hour infusion. Against all 3 strains, cefiderocol showed rapid bacterial reduction within 4 hours. Regrowth was observed for one strain, but no growth was observed with a MIC of ≥ 10 mg/L and no resistant colonies to cefiderocol were detected at the 24- and 72-hour time points.

5.4.5.2 MIC shifts during therapy

During the clinical studies, 4-fold MIC increases have been observed with both cefiderocol and the comparators. Table 72 is summarising the MIC shift observed:

Table 72: Summary of MIC shift

APEKS-cUTI APEKS-NP CREDIBLE CR

imipene Cefidero Cefidero HD Cefiderocol BAT m/ col col meropenem cilastatin (N=252) (N=145) (N=147) (N=101) (N=49) (N=119)

Nb patients with 4- 7 3 9* 9 15** 5 fold MIC increase

% patients with 4- 2.8% 2.5% 6.2% 6.1% 14.8% 10.2% fold MIC increase

*1 subject with postbaseline MIC>4 mg/L**only 3 with postbaseline MIC>4mg/L, in the APEKS-cUTI none of the strains had an MIC>4 mg/L postbaseline.

In the 3 clinical studies, a similar percentage of subjects with 4-fold MIC increase was observed in both the cefiderocol and the comparator treatment group. Only in 4 subjects the MIC observed postbaseline was above the unbound concentration of cefiderocol in plasma of 4 mg/L. Molecular characterisation of the strains with increased MICs is not completed yet.

In conclusion, as for other antibacterial, in vitro resistance development was observed for cefiderocol similar to ceftazidime. In clinical trials, MIC’s increase was also observed with the same magnitude in both treatment gourps (ie cefiderocol and comparators). The HD meropenem was not sufficient to fully repress increase in MICs during treatment of patient with nosocomial pneumonia.

Table 73a: Methods of data collection and analysis of Mortality

Study Endpoint definition Method of analysis reference/ID APEKS-NP All-cause mortality at Day 14 since first ACM by treatment group will be calculated as the proportion of patients who experienced mortality infusion of study drug (ACM): All-cause regardless of the cause at or before Day 14. The adjusted estimates of the difference in ACM at Day mortality rate at Day 14 since first infusion of 14 between cefiderocol and meropenem will be presented along with 95% confidence intervals (CIs) study drug will be calculated as the proportion based on a stratified analysis using Cochran-Mantel-Haenszel (CMH) weights. The CI will be 2-sided. of patients who experienced mortality Cochran-Mantel-Haenszel weights will be calculated with APACHE II score (≤ 15 and ≥ 16) as the regardless of the cause at or before Day 14 stratified factor. since first infusion. Sensitivity analysis for missing Day 14 ACM status will be implemented as follows: subjects with ACM at Day 28: All-cause mortality rate at Day unknown mortality status at Day 14 in the cefiderocol group will be imputed as “Death” while any 28 since first infusion of study drug will be subject with unknown mortality status at Day 14 in the Meropenem arm will be imputed as “Alive “. calculated as the proportion of patients who experienced mortality regardless of the cause The estimates of the difference in the ACM at Day 14 between cefiderocol and meropenem will be at or before Day 28 since first infusion. presented along with 95% confidence intervals (CIs) (Wald method) if data warrant. If the number of . subjects within a subgroup is less than 10 in any treatment arm, only the difference in the ACM between the two treatment arms (no CI) will be presented. The CI will be 2-sided. Similar analysis will also be carried out for Day 28 all-cause mortality.

Analysis for Day 14 ACM will also be performed by excluding subjects who are meropenem resistant in the mITT analysis population as a supplementary analysis. Subjects who are meropenem resistant will be determined from central laboratory culture results

Analyses of ACM at Day 14 will be presented for the following subgroups: Clinical diagnosis, Gender, Race, Age, Region and Baseline clinical characteristics.

ACM rate during treatment and follow-up ACM rate during treatment and follow-up period (until EOS) will be calculated as the proportion of period (until EOS) patients who experienced mortality regardless of the cause at or before EOS since the first infusion.

If a subject discontinues from the study before this period and survival information was not available, then the survival status for this endpoint for the subject will be unknown. All-cause until EOS visit will be analysed in a similar way to the Primary Efficacy endpoint described CREDIBLE-CR All-cause mortality at Day 14 and Day28 for All-cause mortality rate with 95% CI at Day 14, Day 28 and overall by treatment group will be HAP/VAP/HCAP and BSI/sepsis. calculated as the proportion of subjects who experienced mortality regardless of the cause at or before Day 14 and Day 28, respectively. In this analysis, deaths occurring after EOS will not be used for analysis and any subject who does not have vital status information at Day 14 and 28 will not be included in the analysis. This analysis will be performed for both CR MITT and ITT Population. Survival time (HAP/VAP/HCAP, BSI/sepsis) In addition, for CR MITT Population, subgroup analysis regarding all-cause mortality at Day 28 will be

performed. For the survival time up to End of Study (EOS), the survival curve using Kaplan-Meier method by treatment group will be presented. For the subjects whose vital status is survival at EOS, the subjects will be treated as right-censored at EOS. For the subjects whose vital status is not collected or unknown, the subjects will be treated as right-censored at last visit day.

All-cause mortality rate at Day 14, Day 28 and overall including death after EOS will be calculated by treatment group for ITT population.

EA: Early Assessment, EOT: End of Treatment, TOC: Test of Cure, FUP: Follow-up, MAX

Table 80b: Methods of data collection and analysis of Clinical outcomes

Study Endpoint definition Method of analysis reference/ID APEKS-cUTI Clinical response at EA Subject reported symptoms identified at baseline will be assessed at EA, EOT, TOC, and FUP utilizing Clinical Cure: Resolution or improvement of a Structured Subject Interview (see Appendix 3 of the study protocol [237]) that will evaluate whether baseline signs and symptoms of cUTI at EA or the symptom is still present (and if so the degree of that symptom, i.e., mild, moderate, or severe) or return to pre-infection baseline if known. returned to baseline. Clinical Failure: No apparent response to therapy, persistence of signs and/or symptoms Clinical response will be determined by the investigator based upon resolution or improvement of of cUTI infection beyond pre-infection baseline, clinical signs and symptoms of cUTI prior to receiving any potentially effective antibacterial therapy or reappearance of signs and/or symptoms, at for cUTI and subject reported symptoms noted in the Structured Subject Interview. Baseline or before the EA. symptoms associated with anatomic abnormalities that predisposes to cUTI do not need to be Indeterminate: Lost to follow-up such that a resolved for a consideration of a successful responder. determination of clinical response (cure, improvement, or failure) cannot be made. For the primary analysis, the clinical response will be a dichotomy (cure or failure) based on the Clinical response at EOT and TOC clinical outcome as assessed by the investigator taking into consideration objective data Clinical Cure: Resolution or improvement of (temperature, WBC, urinalysis) and patient reported symptoms noted in the Structured Patient baseline signs and symptoms of cUTI, or return Interview. to pre-infection baseline if known, at EOT and TOC. The clinical outcome of interest at EA, EOT, and TOC will be the proportion of subjects who have a Clinical Failure: No apparent response to clinical outcome of cure. The outcome of interest at FUP will be the proportion of subjects with therapy, persistence of signs and/or symptoms sustained clinical cure. of cUTI infection beyond pre-infection baseline, or reappearance of signs and/or symptoms, at or before the EOT and/or TOC visit. Indeterminate: Lost to follow-up such that a determination of clinical response (success or failure) cannot be made.

Clinical response at FUP Sustained Clinical Cure: All pre-therapy signs and symptoms of cUTI show no evidence of recurrence after administration of the last dose of study drug. Failure: Patients carried forward from TOC. Relapse: Signs and/or symptoms of cUTI that were absent at TOC reappear at the FUP. Indeterminate: Lost to follow-up such that a determination of clinical response (success or failure) cannot be made APEKS-NP Clinical response The clinical outcomes will be assessed by the investigator according to the described criteria at EA, Clinical Cure: Resolution or substantial EOT and TOC. improvement of baseline signs and symptoms of pneumonia, including a reduction in The clinical response rate at Early Assessment, End of Treatment and Test of Cure will be calculated Sequential Organ Failure Assessment (SOFA) as the proportion of subjects who have a clinical outcome of cure. The adjusted estimate of the and Clinical Pulmonary Infection (CPIS) scores, difference in the cure rate between the 2 treatment groups will be presented along with the adjusted and improvement or lack of progression of 95% CIs based on the CMH weights: diagnosis and APACHE II score. In addition, the number and chest radiographic abnormalities such that no proportion of subjects having clinical outcome as failure and indeterminate will be summarized by additional antibacterial therapy is required for treatment group. the treatment of current infection at the EA and EOT visits, and no antibacterial therapy is Results will be presented per Infection site, Pathogen and Non-fermenters required for the treatment of the current infection at the TOC. Clinical Failure: No apparent response to therapy; persistence or worsening of baseline signs and/or symptoms of pneumonia; reappearance of signs and/or symptoms of

pneumonia; development of new signs and/or symptoms of pneumonia requiring antibacterial therapy other than, or in addition to, study treatment therapy; progression of chest radiographic abnormalities; or death due to pneumonia. Indeterminate: Lost to follow-up such that a determination of clinical cure/failure cannot be made. APEKS-NP Sustained Clinical Cure: Continued resolution The clinical outcome at FU will be assessed by the investigator according to the or substantial improvement of baseline signs described criteria. and symptoms of pneumonia, such that no antibacterial therapy has been required for the The cure rate at FU will be calculated as the proportion of subjects with clinical outcome of sustained treatment of pneumonia in a subject assessed clinical cure. In addition, the number and proportion of subjects having clinical outcome as relapse, as cured at TOC. clinical failure and indeterminate will be summarized by treatment group. Relapse: Recurrence of signs and/or The same analysis method as described above for clinical outcome per subject at EA, EOT and TOC symptoms of pneumonia, appearance of new will be performed for the clinical outcome per subject FU. signs and/or symptoms of pneumonia, or new The outcome will be tabulated for each treatment group. The adjusted estimate of the difference in chest radiographic evidence of pneumonia in a the response rate between the 2 treatments arms along with the adjusted 95% CIs based on the CMH subject assessed as cured at TOC. weights will be presented. Clinical Failure: Clinical failure at TOC will be carried forward regardless of lost to follow-up. Indeterminate: Lost to follow-up, such that a determination of clinical sustained cure/relapse cannot be made, or subject received additional antibacterial therapy for the treatment of the current infection. CREDIBLE-CR HAP/VAP/HCAP: Efficacy Criteria for Infection Site Specific Clinical Outcomes assessed at EA, EOT, and TOC

● Clinical Cure: Resolution or The clinical outcomes will be assessed by the investigator according to the described criteria substantial improvement of baseline established for each infection site at EOT and TOC. In case treatment duration is extended beyond signs and symptoms of pneumonia 14 days, an additional clinical outcome will be assessed on Day 14. including a reduction in SOFA and CPIS scores, and improvement or lack Sequential Organ Failure Assessment score (SOFA) and its change from baseline will be summarized of progression of chest radiographic by treatment group per infection site at Baseline, EOT, TOC, and FU. Change from baseline will also abnormalities such that no additional be summarized. In addition, SOFA score regardless of primary infection diagnosis will be analysed in antibacterial therapy is required for the a similar manner. treatment of the current infection. For the HABP/VABP/HCABP subjects, CPIS at EOT, TOC, and FU will be summarized by treatment

● Clinical Failure: No apparent group. response to therapy; persistence or worsening of baseline signs and/or symptoms of pneumonia; reappearance of signs and/or symptoms of pneumonia; development of new signs and/or symptoms of pneumonia requiring antibacterial therapy other than, or in addition to, study treatment therapy; progression of chest radiographic abnormalities; or death due to pneumonia.

● Indeterminate: Lost to follow-up such that a determination of clinical cure/failure cannot be made. cUTI ● Clinical Cure: Resolution or substantial improvement of baseline

signs and symptoms of cUTI, or return to pre-infection baseline if known, such that no additional antibacterial therapy is required for the treatment of the current infection.

● Clinical Failure: No apparent response to therapy; persistence or worsening of baseline signs and/or symptoms of cUTI; or reappearance of signs and/or symptoms of cUTI; development of new signs and/or symptoms of cUTI requiring antibacterial therapy other than, or in addition to, study treatment therapy; or death due to cUTI.

● Indeterminate: Lost to follow-up such that a determination of clinical cure/failure cannot be made. BSI/Sepsis ● Clinical Cure: Resolution or substantial improvement of baseline signs and symptoms including a reduction in SOFA score, such that no additional antibacterial therapy is required for the treatment of BSI/sepsis. Patients with bacteraemia must have eradication of bacteraemia caused by the Gram-negative

pathogen.

● Clinical Failure: No apparent response to therapy; persistence or worsening of baseline signs and/or symptoms, reappearance of signs and/or symptoms; development of new signs and/or symptoms requiring antibacterial therapy other than, or in addition to, study treatment therapy; or death due to BSI/sepsis.

● Indeterminate: Lost to follow-up such that a determination of clinical cure/failure cannot be made.

CREDIBLE-CR HAP/VAP/HCAP: Efficacy Criteria for Infection Site Specific Clinical Outcomes assessed at FUP will be determined

● Sustained Clinical Cure: Continued according to the described criteria above. resolution or substantial improvement of baseline signs and symptoms of pneumonia, such that no additional antibacterial therapy is required for the treatment of pneumonia in a patient assessed as cured at TOC.

● Relapse: Recurrence of signs and/or symptoms of pneumonia, appearance of new signs and/or symptoms of pneumonia, or new chest radiographic evidence of pneumonia in a patient

assessed as cured at TOC.

● Indeterminate: Lost to follow-up such that a determination of clinical sustained cure/relapse cannot be made, or patient received additional antibacterial therapy for the treatment of the current infection. ● Clinical Failure: Clinical failure at TOC will be carried forward regardless of lost to follow-up cUTI ● Sustained Clinical Cure: Continued resolution or improvement of baseline signs and symptoms of cUTI, or return to pre-infection baseline if known, in a patient assessed as cured at TOC.

● Relapse: Recurrence of signs and/or symptoms of cUTI, or appearance of new signs and/or symptoms of cUTI in a patient assessed as cured at TOC.

of lost to follow-up BSI/Sepsis ● Sustained Clinical Cure: Continued resolution or substantial improvement of baseline signs and symptoms associated with reduction in SOFA score, such that no additional antibacterial therapy is required for the treatment of the patient’s original BSI/sepsis in a patient assessed as cured at TOC.

● Relapse: Recurrence of signs and/or symptoms of BSI/sepsis, or appearance of new signs and/or symptoms of the patient’s original BSI/sepsis in a patient assessed as cured at TOC.

● Indeterminate: Lost to follow-up such that a determination of clinical sustained cure/relapse cannot be made, or patient received additional antibacterial therapy for the treatment of the current infection. ● Clinical Failure: Clinical failure at TOC will be carried forward regardless of lost to follow-up EA: Early Assessment, EOT: End of Treatment, TOC: Test of Cure, FUP: Follow-up, MAX

Table 80c: Methods of data collection and analysis of Composite microbiological eradication and cure

Study Endpoint definition Method of analysis reference/ID APEKS-cUTI Clinical and microbiologic response: The primary composite efficacy endpoint is based on the outcome (response or failure) for both the Resolution or improvement of the symptoms of clinical and microbiologic response at TOC. cUTI present at trial entry (and no new symptoms) and the demonstration that The composite outcome is a “response” if both the clinical and microbiologic outcome are responses. bacterial pathogen found at trial entry is reduced to fewer than 104 CFU/mL on urine Clinical resolution assessed by the investigator will be defined based in part on the graded response culture at the TOC (microbiological response). to the structured subject interview about the current status of the subject’s symptoms that had been recorded at the time of randomization, and the absence of any new symptoms related to the cUTI. Clinical or microbiologic failure: Symptoms of cUTI present at trial entry have not Definition of clinical and microbiological outcome based on possible combinations of microbiological completely resolved or new symptoms have outcome and clinical outcome is presented below. developed, the subject has died, or the urine culture taken at the TOC grows greater than or At EA, EOT and TOC: equal to 104 CFU/mL of the original pathogen

identified at trial entry. Per Subject Microbiological Composite of Clinical and Clinical Outcome Outcome Microbiological Outcome

Microbiological eradication Clinical cure Response

Microbiological eradication Clinical failure Failure

Microbiological eradication Indeterminate Indeterminate

Microbiological failure Clinical cure Failure

Microbiological failure Clinical failure Failure

Microbiological failure Indeterminate Failure

Indeterminate Clinical cure Indeterminate

Indeterminate Clinical failure Failure

Indeterminate Indeterminate Indeterminate

At FUP:

Per Subject Microbiological Composite of Clinical and Clinical Outcome Outcome Microbiological Outcome

Sustained eradication Sustained clinical cure Response

Sustained eradication Clinical failure Failure

Sustained eradication Clinical relapse Failure

Sustained eradication Indeterminate Indeterminate

Microbiological failure Sustained clinical cure Failure

Microbiological failure Clinical failure Failure

Microbiological failure Clinical relapse Failure

Microbiological failure Indeterminate Failure

Indeterminate Sustained clinical cure Indeterminate

Indeterminate Clinical failure Failure

Indeterminate Clinical relapse Failure

Indeterminate Indeterminate Indeterminate

CREDIBLE-CR The definition for composite of clinical and For the composite clinical and microbiological outcome, the outcomes will be summarized and the microbiological outcome based on possible response rate with 95% CI at EOT, TOC, and FU will be calculated per infection site by treatment combinations per subject microbiological group as the proportion of subjects who have both clinical cure and microbiological eradication. outcome and clinical outcome is shown in the

tables for EOT and TOC, and in separate tables Clinical and Microbiological Outcome: EOT and TOC for FUP Per Subject Microbiological Composite of Clinical and Clinical Outcome Outcome Microbiological Outcome Eradication Clinical cure Response

Eradication Clinical failure Failure

Eradication Indeterminate Indeterminate

Persistence Clinical cure Failure

Persistence Clinical failure Failure

Persistence Indeterminate Failure

Indeterminate Clinical cure Indeterminate

Indeterminate Clinical failure Failure

Indeterminate Indeterminate Indeterminate

Composite Outcome: Follow-up

Per Subject Microbiological Composite of Clinical and Clinical Outcome Outcome Microbiological Outcome

Sustained eradication Sustained clinical cure Response

Sustained eradication Clinical failure Failure

Sustained eradication Clinical relapse Failure

Sustained eradication Indeterminate Indeterminate

Persistence Sustained clinical cure Failure

Persistence Clinical failure Failure

Persistence Clinical relapse Failure

Persistence Indeterminate Failure

Recurrence Sustained clinical cure Failure

Recurrence Clinical failure Failure

Recurrence Clinical relapse Failure

Recurrence Indeterminate Failure

Indeterminate Sustained clinical cure Indeterminate

Indeterminate Clinical failure Failure

Indeterminate Clinical relapse Failure

Indeterminate Indeterminate Indeterminate

EA: Early Assessment, EOT: End of Treatment, TOC: Test of Cure, FUP: Follow-up, MAX

Table 80d: Methods of data collection and analysis of Microbiological outcomes

Study Endpoint definition Method of analysis reference/ID APEKS-cUTI Eradication: A urine culture shows the An overall per subject microbiological outcome will be determined at EA, EOT, TOC, bacterial uropathogen(s) identified at baseline and FUP. In addition, per pathogen microbiological outcomes will be determined for baseline at ≥ 105 CFU/mL are reduced to < 104 uropathogens. New pathogens that emerge after study therapy is started will also be assessed. Per CFU/mL. subject and per pathogen microbiological outcomes will be assessed only for Gram-negative Persistence: A urine culture shows that the uropathogens which are identified with quantitative measurements by the local laboratory and original bacterial uropathogen(s) identified at confirmed by the central microbiology laboratory. If the pathogen is not sent to central microbiology baseline at ≥ 105 CFU/mL grows ≥ 104 laboratory, the outcome will be only assessed by the local laboratory. CFU/mL. Indeterminate: No urine culture or a urine For the subjects who have Gram-positive uropathogens at baseline, the Gram-positive uropathogens culture that cannot be interpreted for any will be shown in the listing of local microbiological test and not be considered in either per pathogen reason. microbiological outcome or per subject microbiological outcome. Subjects who used non-study antibacterial drug therapy with Gram-negative coverage and thus may have a potential effect on outcome evaluation in patients with cUTI were treated as microbiological failure at all the following analysis visits after the use of the non-study antibacterial drug therapy regardless of outcome above.

As shown below, subjects who experience eradication of all baseline Gram-negative uropathogen(s) at EA, EOT, and TOC will be considered” microbiological eradication” and subjects who experience persistence of any baseline Gram-negative uropathogen will be considered” microbiological failures”. Subjects whose experiences are other than above will be considered” indeterminate”.

The microbiological response rate at EA, EOT and TOC will be calculated as the proportion of subjects who experience eradication at EA, EOT and TOC respectively by treatment group. The adjusted estimate of the difference in the response rate between the 2 treatment groups will be presented along with the 95% CIs based on a stratified analysis using the CMH weights: infection diagnosis (HABP/VABP/HCABP) and APACHE II score (≤ 15 and ≥ 16). In addition, the number and proportion of subjects having microbiological outcome as persistence and indeterminate will be summarized by treatment group.

APEKS-cUTI Sustained Eradication: A urine culture Assessment of baseline Gram-negative pathogens at FUP includes sustained eradication obtained after documented eradication at the TOC, up to and including the FUP, shows At FU, the per subject microbiological outcome for subjects who experience sustained eradication of that the bacterial uropathogen(s) all baseline Gram-negative pathogens will be considered as “sustained eradication” and subjects who identified at baseline at ≥105 CFU/mL remain experience recurrence of any baseline Gram-negative pathogens will have a per subject <104 CFU/mL. microbiological outcome of “recurrence”. Subjects who show persistence of any baseline Gram-

• Persistence: A urine culture obtained any time negative pathogens will have a per subject microbiological outcome of “persistence”. Subjects whose after TOC, up to and including the FUP, grows experiences are other than above at FU will be considered “indeterminate”. ≥104 CFU/mL of the original uropathogen. If there are no available culture results, the outcomes for pathogens that persisted at TOC are carried forward to the FUP. • Recurrence: A urine culture obtained any time after documented eradication at the TOC, up to and including the FUP, grows ≥104 CFU/mL of the original uropathogen.

APEKS-NP Eradication: Absence of the baseline Gram- The microbiological outcomes by baseline pathogens will be determined according to the described negative pathogen from an criteria at EA, EOT and TOC. appropriate clinical specimen. Presence of colonizers or contaminants associated Subjects who experience eradication of all baseline Gram-negative with a baseline pathogen will be associated pathogen(s) at EA, EOT and TOC their per subject microbiological outcome with microbiological outcome of will be considered “eradication” and subjects who experience persistence of any baseline eradication. If it is not possible to obtain an Gram-negative pathogens, per subject microbiological outcome will be considered appropriate clinical culture, and the “persistent”.” Subjects whose experiences are other than the above will be considered subject has a successful clinical outcome; the “indeterminate.” response will be presumed as eradication.

Persistence: Continued presence of the baseline Gram-negative pathogen from an appropriate clinical specimen. Persistence at End of Treatment or Test of Cure will be carried forward. Indeterminate: No culture obtained from an appropriate clinical specimen or if the microbiological outcome is eradication after additional antibacterial therapy for the treatment of the current infection.

APEKS-NP Sustained Eradication: Absence of the The microbiological outcomes by baseline pathogens will be determined according to the described baseline Gram-negative pathogen from an criteria at FU. appropriate clinical specimen after TOC. Presence of colonizers or contaminants At FU, the per subject microbiological outcome for subjects who experience sustained eradication of associated with a baseline pathogen will be all baseline Gram-negative pathogens will be considered as “sustained eradication” and subjects who associated with microbiological outcome of experience recurrence of any baseline Gram-negative pathogens will have a per subject sustained eradication. If it is not possible to microbiological outcome of “recurrence”. Subjects who show persistence of any baseline Gram- obtain an appropriate clinical culture, and the

subject has a successful clinical response after negative pathogens will have a per subject microbiological outcome of “persistence”. Subjects whose TOC, the response will be presumed experiences are other than above at FU will be considered “indeterminate”. eradication. Recurrence: Recurrence of the baseline Gram-negative pathogen from an appropriate clinical specimen taken after TOC, and the TOC culture was negative. Persistence: Persistence of any baseline Gram-negative pathogen from an appropriate specimen. Indeterminate: No culture obtained from an appropriate clinical specimen or if the microbiological outcome is eradication after the subject received additional antibacterial therapy for the treatment of the current The microbiologic response rate at FU will be calculated as the proportion of subjects who experience infection. sustained eradication of all baseline Gram-negative pathogens after documented eradication at the TOC. The same analysis method as described above for microbiological outcome per subject at EA, EOT and TOC will be performed for the microbiologic outcome per subject at FU. The outcome will be tabulated for each treatment group. The adjusted estimate of the difference in the response rate between the 2 treatments arms along with the adjusted 95% CIs based on the CMH weights will be presented. CREDIBLE-CR HAP/VAP/HCAP The microbiological outcomes by baseline pathogen will be determined by the sponsor according to

● Eradication: Absence of the baseline the described criteria established for each infection site at EA, EOT, and TOC. In case treatment Gram-negative pathogen from an duration is extended beyond 14 days, an additional microbiological outcome will be assessed on Day appropriate clinical specimen. If it is 14. An overall per-subject microbiological outcome will also be determined based on the individual not possible to obtain an appropriate microbiological outcomes for each baseline pathogen. Emergent (i.e., non-baseline) pathogens are clinical culture and the patient has a considered separately, and do not affect the per-subject microbiological outcome.

successful clinical outcome, the Subjects who experience eradication of all baseline Gram-negative pathogens at EOT and TOC will response will be presumed as be considered “Eradication” and subjects who experience persistence of any baseline Gram-negative eradication. pathogen will be considered “persistence.” Subjects whose experiences are other than above will be

● Persistence: Continued presence of considered “indeterminate.” At FU, subjects who experience sustained eradication of all baseline the baseline Gram-negative pathogen Gram-negative pathogens after documented eradication at the TOC will be considered “sustained from an appropriate clinical specimen. eradication” and subjects who experience eradication at TOC, but recurrence of any baseline Gram- negative pathogen will be considered as” recurrence”, and subjects who are considered as ● Indeterminate: No culture obtained “persistence” at TOC will be “persistence.” Subjects whose experiences are other than above will be from an appropriate clinical specimen considered “indeterminate.” (see Table below) or additional antibacterial therapy for the treatment of the current infection.

cUTI Per Subject Microbiological ● Eradication: A urine culture shows Visit Definition Outcome the baseline Gram-negative EOT, TOC Eradication Eradication of all baseline Gram-negative uropathogen found at entry at ≥ 105 pathogens CFU/mL are reduced to < 104 CFU/mL. Persistence Persistence of any baseline Gram-negative pathogens ● Persistence: A urine culture shows that the baseline Gram-negative Indeterminate Other than those above uropathogen found at entry at ≥ 105 FU Sustained eradication Sustained eradication of all baseline Gram- CFU/mL grows ≥ 104 CFU/mL. negative pathogens after documented ● Indeterminate: No urine culture eradication at the TOC

obtained or additional antibacterial Persistence Persistence of any baseline Gram- at the TOC therapy for the treatment of the current Recurrence Recurrence of any baseline Gram-negative infection. pathogens for subject’s eradication at the TOC BSI/Sepsis Indeterminate Other than those above ● Eradication: Absence of the baseline EOT = End of Treatment; FU = Follow-up; TOC = Test of Cure Gram-negative pathogen from a blood

culture and/or other primary source.

● Persistence: Continued presence of the baseline Gram-negative pathogen from a blood culture or other primary source.

● Indeterminate: No culture obtained or additional antibacterial therapy for the treatment of the current infection.

HAP/VAP/HCAP The microbiological outcomes by baseline pathogen will be determined according to the described

● Sustained Eradication: Absence of criteria established for each infection site at FUP. the baseline Gram-negative pathogen from an appropriate clinical specimen after TOC. If it is not possible to obtain an appropriate clinical culture and the patient has a successful clinical response after TOC, the response will be presumed eradication.

● Recurrence: Recurrence of the baseline Gram-negative pathogen from an appropriate clinical specimen taken after TOC and the TOC culture is negative.

● Indeterminate: No culture obtained from an appropriate clinical specimen or patient received additional antibacterial therapy for the treatment of the current infection.

● Persistence: Persistence at TOC will be carried forward. cUTI

● Sustained Eradication: A culture taken any time after documented eradication at TOC, and a urine culture obtained at FUP shows that the baseline uropathogen found at entry at ≥105 CFU/mL remains < 104 CFU/mL.

● Recurrence: A culture taken any time after documented eradication at TOC, up to and including FUP that grows the baseline uropathogen ≥ 104 CFU/mL

● Indeterminate: No urine culture or patient received additional antibacterial therapy for the treatment of the current infection. ● Persistence: Persistence at TOC will be carried forward. BSI/Sepsis ● Sustained Eradication: Absence of the baseline Gram-negative pathogen from a blood culture or other primary source after TOC.

● Recurrence: Recurrence of the baseline Gram-negative pathogen

from a blood culture or other primary source after TOC and the TOC culture is negative.

● Indeterminate: No culture or patient received additional antibacterial therapy for the treatment of the current infection. ● Persistence: Persistence at TOC will be carried forward. CREDIBLE-CR New Pathogens New pathogens that emerge on or after Day 3 will be categorized as either superinfection or new

● Superinfection: The identification infection as follows: Superinfection and new infection will be listed by Gram-negative pathogen and from an appropriate clinical specimen the others. of a new pathogen from the original infection site. This new pathogen must be associated with new or persisting signs and symptoms of infection.

● New Infection: The identification from an appropriate clinical specimen of a new pathogen from an infection site different from the original infection site. This new pathogen must be associated with new or persisting signs and symptoms of infection.

EA: Early Assessment, EOT: End of Treatment, TOC: Test of Cure, FUP: Follow-up, MAX

Table 80e: Methods of data collection and analysis of Susceptibility rates

Study Endpoint definition Method of analysis reference/ID SIDERO-WT Minimum inhibitory concentrations (MICs) of The range and concentration of each antimicrobial agent tested is listed below: SIDERO-CR cefiderocol, cefepime, ceftazidime-avibactam, ceftolozane-tazobactam, ciprofloxacin, colistin, and meropenem, were determined by broth microdilution.

SIDERO-WT Percent susceptibility (%) calculation Percent susceptibility (%) was calculated according to CLSI interpretive criteria where available, and SIDERO-CR the FDA interpretive criteria for ceftazidime-avibactam. In the absence of any CLSI or FDA breakpoints for colistin tested against Enterobacteriaceae, the EUCAST susceptible breakpoint of ≤2 μg/mL was applied.

5.5 Individual study results (safety outcomes)

1. Describe the relevant endpoints, including the definition of the endpoint and methods of analysis (Table 94).

Endpoints are described in Dossier Table 94.

2. For the technology, and the comparator, tabulate the total number of adverse events, frequency of occurrence (as a %), absolute and relative risk and 95% CI reported in each of the clinical studies. Categorise the adverse events by frequency, severity and system organ class.

This section summarizes the safety outcomes in the overall sample, based on regulatory documents, followed by the reporting of results of safety assessments in each clinical study.

5.5.1 Overall safety results: pooled analysis and individual studies: APEKS-cUTI, APEKS-NP, and CREDIBLE CR

The cefiderocol clinical development program to date includes information from 6 completed clinical pharmacology studies, a completed Phase 2 study, two Phase 3 studies and cases of compassionate Table 81 summarises the dose and exposure of patients to cefiderocol within the clinical trials, where nearly half the patients are from APEKS cUTI that per protocol design had a maximum treatment duration of 14 days. As so, the vast majority of patients were treated with cefiderocol between 7 to 14 days for cUTI. APEKS NP and CREDIBLE presented longer treatment durations; 52 patients receivd treatment with cefiderocol between 14 and 22 days (mostly coming from CREDIBLE CR study).

Table 81: Dose and Duration of Exposure to cefiderocol* (Number of Patients by Indication)

APEKS NP Study

cUTI study (Dose CREDIBLE-CR study (Dose 2g cefiderocol 3 Total (Dose 2g Duration 2g cefiderocol 3 (Dose 2g cefiderocol 3 times daily (every 8 cefiderocol 3 times of times daily (every times daily (every 8 hours) hours) for 7-14 days (may daily) exposure 8 hours) for 7-14 days (may be be extended up to 21 (days) for 7-14 days) extended up to 21 days)) days)) <5 8 8 14 30 5 to <7 10 7 4 21 7 to ≤14 435 277 61 97 * >14 to 31 ≤21 5 16 52

>21 0 9 2 11

Source: EU Risk Managing Plan for Fetcroja [284]; cUTI dose was given over 1 hour; CREDIBLE-CR and APEKS NP dose was given over 3 hours

Table 82 summarizes treatment-related adverse events for each trial and for the total patient population studied. An additional detailed summary of all treatment-emergent adverse events is on file[262, 263].

Pooled adverse event analyses there overall less treatment emergent adverse events with cefiderocol (344/549 [67.1%]) vs comparators (252/347 [72.6%]). The most common adverse reactions for cefiderocol were diarrhoea (8.2%), constipation (4.6%), pyrexia (4.0%) and UTI (4.7%).

In the total sample, 56/549 (10.2%) patients treated with cefiderocol experienced treatment related AEs and 45/347 (13.0%) patients treated with comparators.

Table 82: Subjects with Treatment Related Adverse Events by System Organ Class and Preferred Term (All Phase II/III Studies) Safety Population

cUTI Study CREDIBLE-CR Study APEKS NP Study All Studies Cefiderocol Imipenem/Cilastatin Cefiderocol BAT Cefiderocol Meropenem Cefiderocol Comparator System Organ Class N=300 N=148 N=101 N=49 N=148 N=150 N=549 N=347 - Preferred Term n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) Subjects with any Treatment Related AEs 27 (9.0) 17 (11.5) 15 (14.9) 11 (22.4) 14 (9.5) 17 (11.3) 56 (10.2) 45 (13.0) Blood and lymphatic system disorders 0 0 0 0 0 2 (1.3) 0 2 (0.6) - Disseminated intravascular coagulation 0 0 0 0 0 1 (0.7) 0 1 (0.3) - Thrombocytopenia 0 0 0 0 0 1 (0.7) 0 1 (0.3) Cardiac disorders 0 1 (0.7) 0 0 0 0 0 1 (0.3) - Tachycardia 0 1 (0.7) 0 0 0 0 0 1 (0.3) Ear and labyrinth disorders 0 0 0 0 1 (0.7) 0 1 (0.2) 0 - Ear discomfort 0 0 0 0 1 (0.7) 0 1 (0.2) 0 Gastrointestinal disorders 9 (3.0) 5 (3.4) 4 (4.0) 1 (2.0) 3 (2.0) 5 (3.3) 16 (2.9) 11 (3.2) - Diarrhoea 4 (1.3) 3 (2.0) 2 (2.0) 0 3 (2.0) 5 (3.3) 9 (1.6) 8 (2.3) - Nausea 3 (1.0) 1 (0.7) 0 0 0 0 3 (0.5) 1 (0.3) - Vomiting 1 (0.3) 1 (0.7) 0 1 (2.0) 0 0 1 (0.2) 2 (0.6) - Abdominal pain upper 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Ascites 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Constipation 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Dry mouth 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Stomatitis 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Upper gastrointestinal haemorrhage 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Lip oedema 0 1 (0.7) 0 0 0 0 0 1 (0.3) General disorders and administration site 5 (1.7) 0 2 (2.0) 0 0 2 (1.3) 7 (1.3) 2 (0.6) conditions - Oedema peripheral 2 (0.7) 0 0 0 0 0 2 (0.4) 0

cUTI Study CREDIBLE-CR Study APEKS NP Study All Studies Cefiderocol Imipenem/Cilastatin Cefiderocol BAT Cefiderocol Meropenem Cefiderocol Comparator System Organ Class N=300 N=148 N=101 N=49 N=148 N=150 N=549 N=347 - Preferred Term n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) - Infusion site pain 2 (0.7) 0 0 0 0 0 2 (0.4) 0 - Feeling hot 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Oedema 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Pyrexia 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Infusion site erythema 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Hyperthermia 0 0 0 0 0 1 (0.7) 0 1 (0.3) - Multiple organ dysfunction syndrome 0 0 0 0 0 1 (0.7) 0 1 (0.3) Hepatobiliary disorders 0 1 (0.7) 0 0 1 (0.7) 1 (0.7) 1 (0.2) 2 (0.6) - Hepatic failure 0 0 0 0 1 (0.7) 0 1 (0.2) 0 - Hepatic function abnormal 0 1 (0.7) 0 0 0 0 0 1 (0.3) - Hepatocellular injury 0 0 0 0 0 1 (0.7) 0 1 (0.3) Immune system disorders 1 (0.3) 0 0 1 (2.0) 0 0 1 (0.2) 1 (0.3) - Drug hypersensitivity 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Anaphylactic reaction 0 0 0 1 (2.0) 0 0 0 1 (0.3) Infections and infestations 4 (1.3) 6 (4.1) 2 (2.0) 2 (4.1) 3 (2.0) 6 (4.0) 9 (1.6) 14 (4.0) - Clostridium difficile colitis 1 (0.3) 4 (2.7) 1 (1.0) 0 0 0 2 (0.4) 4 (1.2) - Oral candidiasis 1 (0.3) 0 0 0 1 (0.7) 0 2 (0.4) 0 - Candiduria 2 (0.7) 0 0 0 0 0 2 (0.4) 0 - Clostridium difficile infection 0 0 0 0 1 (0.7) 2 (1.3) 1 (0.2) 2 (0.6) - Pseudomembranous colitis 0 0 1 (1.0) 1 (2.0) 0 0 1 (0.2) 1 (0.3) - Sepsis 0 0 0 1 (2.0) 1 (0.7) 0 1 (0.2) 1 (0.3) - Fungal infection 0 1 (0.7) 0 0 0 0 0 1 (0.3) - Septic shock 0 0 0 1 (2.0) 0 0 0 1 (0.3) - Systemic candida 0 0 0 0 0 1 (0.7) 0 1 (0.3) - Vaginal infection 0 1 (0.7) 0 0 0 0 0 1 (0.3) - Urinary tract infection fungal 0 0 0 0 0 1 (0.7) 0 1 (0.3) - Pseudomonas infection 0 0 0 0 0 1 (0.7) 0 1 (0.3)

cUTI Study CREDIBLE-CR Study APEKS NP Study All Studies Cefiderocol Imipenem/Cilastatin Cefiderocol BAT Cefiderocol Meropenem Cefiderocol Comparator System Organ Class N=300 N=148 N=101 N=49 N=148 N=150 N=549 N=347 - Preferred Term n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) - Candida infection 0 0 0 0 0 1 (0.7) 0 1 (0.3) Investigations 5 (1.7) 2 (1.4) 8 (7.9) 2 (4.1) 4 (2.7) 4 (2.7) 17 (3.1) 8 (2.3) - Alanine aminotransferase increased 1 (0.3) 0 3 (3.0) 0 2 (1.4) 1 (0.7) 6 (1.1) 1 (0.3) - Gamma-glutamyltransferase increased 4 (1.3) 1 (0.7) 0 0 2 (1.4) 0 6 (1.1) 1 (0.3) - Aspartate aminotransferase increased 0 0 3 (3.0) 0 2 (1.4) 1 (0.7) 5 (0.9) 1 (0.3) - Transaminases increased 0 0 1 (1.0) 0 1 (0.7) 0 2 (0.4) 0 - Liver function test increased 0 0 2 (2.0) 0 0 0 2 (0.4) 0 - Hepatic enzyme increased 1 (0.3) 0 0 1 (2.0) 0 2 (1.3) 1 (0.2) 3 (0.9) - Blood creatinine increased 0 1 (0.7) 1 (1.0) 0 0 0 1 (0.2) 1 (0.3) - Blood pressure increased 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Blood creatine increased 0 0 0 1 (2.0) 0 0 0 1 (0.3) - Blood alkaline phosphatase increased 0 1 (0.7) 0 0 0 0 0 1 (0.3) Metabolism and nutrition disorders 0 0 1 (1.0) 1 (2.0) 0 0 1 (0.2) 1 (0.3) - Hypokalaemia 0 0 1 (1.0) 0 0 0 1 (0.2) 0 - Metabolic acidosis 0 0 0 1 (2.0) 0 0 0 1 (0.3) Musculoskeletal and connective tissue disorders 1 (0.3) 0 0 0 0 0 1 (0.2) 0 - Myalgia 1 (0.3) 0 0 0 0 0 1 (0.2) 0 Nervous system disorders 1 (0.3) 4 (2.7) 1 (1.0) 1 (2.0) 3 (2.0) 0 5 (0.9) 5 (1.4) - Dysgeusia 1 (0.3) 1 (0.7) 1 (1.0) 0 0 0 2 (0.4) 1 (0.3) - Headache 0 3 (2.0) 0 0 1 (0.7) 0 1 (0.2) 3 (0.9) - Dizziness 0 0 0 0 1 (0.7) 0 1 (0.2) 0 - Paraesthesia 0 0 0 0 1 (0.7) 0 1 (0.2) 0 - Status epilepticus 0 0 0 1 (2.0) 0 0 0 1 (0.3) Psychiatric disorders 0 0 0 0 1 (0.7) 0 1 (0.2) 0 - Confusional state 0 0 0 0 1 (0.7) 0 1 (0.2) 0 Renal and urinary disorders 0 0 0 5 (10.2) 0 0 0 5 (1.4) - Acute kidney injury 0 0 0 4 (8.2) 0 0 0 4 (1.2)

ALT = alanine aminotransferase; AST = aspartate aminotransferase; BAT = best available therapy; INC = increase from baseline; PT-INR = prothrombin time-international normalized ratio; ULN = upper limit of normal; Percentage is calculated using N’ as the denominator, where N’ is the number of subjects with valid postbaseline measurements.

5.5.2 Safety analyses by clinical trial

5.5.2.1 APEKS cUTI

Cefiderocol was generally safe and well-tolerated in the cUTI study, with a safety profile consistent with other cephalosporin antibacterials. Adverse events (AEs) and serious adverse events (SAEs) were comparable between the cefiderocol and imipenem groups. The safety profile of cefiderocol supports its use in cUTI.

5.5.2.1.1 Extent of Exposure

Safety Analysis Population

Of 452 subjects randomized, 448 received at least 1 dose of the study drugs and were included in the Safety Population (99.0% [300/303] of subjects in the cefiderocol group and 99.3% [148/149] of subjects in the IPM/CS group) (Table 82). Of the subjects in the Safety Population, 93.4% (283/303) of randomized subjects in the cefiderocol group and 92.6% (138/149) of randomized subjects in the IPM/CS group completed the study.

Subjects were excluded from the Safety Population for no study drug infusion (1.0% [3/303] of subjects in the cefiderocol group and 0.7% [1/149] of subjects in the IPM/CS group). Study blind was broken for 4 subjects. All four were unblinded before the database was locked to evaluate potential suspected unexpected serious adverse reactions.

Duration of Study Treatment

The duration of treatment exposure in the Safety Population is shown in Table 83. Treatment duration was similar between the treatment groups and consistent with the ITT and Micro-ITT populations. A similar percentage of subjects received less than 5 days of treatment (2.7% in both treatment groups). A median of 9.0 days of treatment for both groups suggests the majority of subjects received an adequate duration of therapy, and no differences between the treatment groups were observed.

Table 83: Summary of duration of exposure (safety population)

5.5.2.1.2 Brief Summary of Adverse Events

Incidence rates for AEs, treatment-related AEs, and SAEs were numerically lower in the cefiderocol group compared with the IPM/CS group in the Safety Population (Table 84). Adverse events and SAEs related to study drug are referred to as “treatment-related” in the tables.

Table 84: Summary of treatment-emergent adverse events (safety population)

Cefiderocol (N=300) Imipenem/cilastatin (N=148) Safety Event n (%) n (%) Any AE 122 (41.0%) 76 (51.0%) Any drug-related AEa 27 (8.7%) 17 (11.5%) Discontinuation due to AEb 5 (1.7%) 3 (2.0%) Any SAEs 14 (4.7%) 12 (8.1%) Deathsc 1 (0.3%) 0 (0%)

[a] Considered treatment-related by the investigator; [b] SAEs for cefiderocol: C. difficile, hypersensitivity (itching), increased hepatic enzymes, diarrhea; [c] Death due to cardiac arrest considered unrelated to study drug by investigator.

AE - adverse event; SAE - serious adverse event; Source: Portsmouth, 2018[51]

Discontinuations due to AEs were reported for 1.7% (5/300) of subjects in the cefiderocol group compared with 2.0% (3/148) of subjects in the IPM/CS group.

One death due to cardiorespiratory arrest was observed in the cefiderocol treatment group; however, this SAE was considered not related to the study drug by both the investigator and the sponsor [51, 236].

5.5.2.1.3 Incidence of Adverse Events

Adverse events were most frequently reported in the gastrointestinal disorders SOC as shown in Table 82. Of the AEs reported in at least 2% of subjects, diarrhea, hypertension, constipation, infusion site pain, headache, nausea, hypokalemia, insomnia, renal cyst, infusion site erythema, abdominal pain upper, cardiac failure, C. difficile colitis, and vaginal infection were seen less frequently in the cefiderocol group than in the IPM/CS group (Table 82).

Cough and vomiting were reported more frequently in the cefiderocol group than in the IPM/CS group. Cough was reported in 2.3% (7/300) of subjects in the cefiderocol group compared with 0.7% (1/148) of subjects in the IPM/CS group. Of note, cough was mild in severity in 5 of 7 subjects and moderate in 2 of 7 subjects in the cefiderocol group, and the single incidence of cough in the IPM/CS group was mild. There were no reports of severe cough. Vomiting was reported in 2.0% (6/300) of subjects in the cefiderocol group compared with 1.4% (2/148) of subjects in the IPM/CS group (all mild in severity). There were no other notable differences between the treatment groups.

The incidence rate of treatment-related AEs (considered treatment-related by the investigator) was 9.0% (27/300) of subjects in the cefiderocol group and 11.5% (17/148) of subjects in the IPM/CS group (Table 82).

5.5.2.1.4 Severity of Adverse Events

The percentage of subjects with mild AEs was approximately the same for each treatment: 25.7% (77/300) of subjects in the cefiderocol group and 24.3% (36/148) of subjects in the IPM/CS group. However, a lower percentage of subjects in the cefiderocol group had moderate AEs (13.0% [39/300] of subjects) compared with the IPM/CS group (23.6% [35/148] of subjects) and severe AEs (2.0% [6/300] of subjects in the cefiderocol group compared with 3.4% [5/148] of subjects in the IPM/CS group) (Table 85).

Table 85: Number (%) of subjects with adverse events by maximum severity (safety population)

5.5.2.1.5 Relationship

9.0% (27/300) of subjects had AEs reported as related to treatment in the cefiderocol group and 11.5% (17/148) of subjects in the IPM/CS group (Table 82).

5.5.2.1.6 Other Serious Adverse Events

Serious adverse events were reported in 4.7% (14/300) of subjects in the cefiderocol group and 8.1% (12/148) of subjects in the IPM/CS group (Table 86). The most frequently reported SAE was C. difficile colitis (0.7% [3/448] of subjects in the total population), with 0.3% (1/300) of subjects in the cefiderocol group and 1.4% (2/148) of subjects in the IPM/CS group. The SAEs of C. difficile colitis in 1 subject in the cefiderocol group (0.3% [1/300]) and in 1 of the 2 subjects in the IPM/CS group (0.7% [1/148]) were considered by the investigator to be treatment related (Table 87).

Table 86: Number (percent) of subjects with serious adverse events (SAEs) by organ class and preferred term (safety population)

Table 87: Number (%) of subjects with treatment-related serious adverse events (SAEs)

5.5.2.2 APEKS cUTI NMA safety analysis

The safety NMA analysis was only possible to be performed for All AEs and Treatment related AEs. Results for both endpoints are presented in this section. Full information on the feasibility assessment and NMA analysis can be found in [227] and [285].

Figure 52: Network Diagram for Safety Analysis

Results for both safety analysis were non-significant, except for the results observed in the APEKS cUTI vs Imipenem/cilastatin in the frequentist analysis for all AEs (Figure 53 to Figure 55)

Figure 53: Safety Analysis for All Adverse Events - Frequentist Analysis

Legends: BAT: best available therapy C_T: ceftazalone-tazobactam CZA: ceftaz idime-avibactam DOR: doripenem FDC: cefiderocol IPM_CIL: imipenem/cilastatin LVX: levofloxacin

Figure 54: Network for safety analysis for Treatment related AEs

Figure 55: safety analysis for Treatment related AEs – Frequentist analysis

5.5.2.3 APEKs-NP: SAFETY

5.5.2.3.1 Extent of Exposure

Of the 300 randomised subjects, 298 received at least one dose of study drug and were included in the safety population: 148 subjects in the cefiderocol group (2 g q8h, 3-hour infusion, or equivalent renally adjusted dose) and 150 in the HD meropenem group (2 g q8h, 3-hour infusion, or equivalent renally adjusted dose). The median (range) duration of treatment was 10.0 (2-22) days in the cefiderocol group and 8.5 (1-22) days in the HD meropenem group [238]. Most subjects in both treatment groups had 7-14 days of exposure (65.5% [97/148] and 73.3% [110/150] in cefiderocol and HD meropenem groups, respectively).

Adverse events occurred in 87.8% (130/148) of subjects in the cefiderocol group and 86.0% (129/150) of the HD meropenem group (Table 88). SAEs occurred in 36.5% (54/148) of subjects in the cefiderocol group and 30.0% (45/150) in the meropenem group.

Adverse events leading to death occurred in 26.4% (39/148) of subjects in the cefiderocol group and 23.3% (35/150) in the meropenem group. Treatment-related AEs, treatment-related SAEs, discontinuations due to AEs, and discontinuations due to treatment-related AEs differed between treatment groups by < 2%.

Table 88: Overview of Treatment-emergent Adverse Events (Safety Population)

Cefiderocol HD Meropenem

(N = 148) (N = 150) Difference of Subjects # of Subjects # of Proportion Adverse Event Category n (%) events n (%) events (95% CI)

TEAEs 130 (87.8) 582 129 (86.0) 537 1.8 (-5.8, 9.5)

Treatment-related TEAEs 14 (9.5) 24 17 (11.3) 22 -1.9 (-8.8, 5.1)

TEAEs leading to death 39 (26.4) 49 35 (23.3) 50 3.0 (-6.8, 12.8)

Treatment-emergent SAEs 54 (36.5) 102 45 (30.0) 96 6.5 (-4.2, 17.2)

Treatment-related SAEs 3 (2.0) 6 5 (3.3) 6 -1.3 (-5.0, 2.4)

Discontinuation due to TEAEs 12 (8.1) 18 14 (9.3) 19 -1.2 (-7.6, 5.2)

Discontinuation due to 2 (1.4) 4 2 (1.3) 3 0.0 (-2.6, 2.6) treatment-related TEAEs CI = confidence interval; TEAEs = treatment emergent adverse events; SAEs = serious adverse events; Percentage was calculated using the number of subjects in the column heading as the denominator. Adverse events that started on or after the first dose date of the study drug and up to ‘End of Study’ were defined as treatment-emergent. Confidence intervals were calculated using the Wilson score method. Source: APEKS-NP Study Synopsis[238]

Adverse events with the highest frequency in the cefiderocol group (urinary tract infection [15.5%], hypokalemia [10.8%], diarrhea [8.8%], and anemia [8.1%]) were also the most frequent AEs in the high-dose meropenem group (hypokalemia [15.3%], urinary tract infection [10.7%], diarrhea [8.7%], and anemia [8.0%]) [262, 263].

5.5.2.3.2 Overview of TEAEs

Most subjects in the cefiderocol group and meropenem group experienced at least 1 TEAE (87.8% [130/148] and 86.0% [129/150], respectively) (Table 88). SAEs were reported in 36.5% (54/148) in the cefiderocol group and 30.0% (45/150) in the meropenem group. Overall, treatment-related TEAEs and SAEs, TEAEs leading to death and discontinuation were reported with similar frequency in the two treatment groups.

5.5.2.3.3 Common TEAEs

The most commonly reported TEAEs (i.e. TEAEs reported in ≥5% of subjects in either treatment group) are summarised by PT in Table 119-9. All TEAEs are reported by SOC and PT in a safety data on file[262, 263]. The most commonly reported TEAEs were from the following SOCs:

 Infections and Infestations: in 40.5% (60/148) and 35.3% (53/150) of subjects in the cefiderocol and meropenem groups, respectively

 Metabolism and nutrition disorders: in 29.1% (43/148) and 31.3% (47/150) of subjects in the cefiderocol and meropenem groups, respectively.

Specifically, the most common TEAEs were urinary tract infection in the cefiderocol group (in 15.5% [23/148] of subjects compared with 10.7% [16/150] in the meropenem group) and hypokalaemia in the meropenem group (in 15.3% [23/150] of subjects compared with 10.8% [16/148] in the cefiderocol group). Most TEAEs were reported with similar frequency in the two treatment groups. TEAEs reported more frequently (>4% difference between treatment groups) in the cefiderocol group than in the meropenem group were: urinary tract infection (in 15.5% [23/148] vs. 10.7% [16/150] of subjects) and hypomagnesaemia (in 5.4% [8/148] vs. 0.7% [1/150] of subjects). TEAEs reported less frequently in the cefiderocol group than in the meropenem group (>4% difference between treatment groups) were: hypokalaemia (in 10.8% [16/148] vs. 15.3% [23/150] of subjects), hepatic enzyme increased, hyponatraemia and decubitus ulcer (each in 2.7% [4/148] vs. 6.7% [10/150] of subjects), and hypotension (in 1.4% [2/148] vs. 6.7% [10/150] of subjects).

5.5.2.3.4 TEAEs by severity

Overall, the proportions of subjects experiencing mild, moderate or severe TEAEs was 23.5% (70/298), 29.5% (88/298) and 33.9% (101/298), respectively. The incidence of severe TEAEs was 37.8% (56/148) in the cefiderocol group compared with 30.0% (45/150) in the high-dose meropenem group.

5.5.2.3.5 Severe TEAEs

The most common severe TEAEs were: cardiac arrest was reported in 4.7% (7/148) in the cefiderocol group and 3.3% (5/150) in the high-dose meropenem group, and pneumonia, was reported in 4.7% (7/148) and 2.0% (3/150), respectively; brain oedema was reported in 0.7% (1/148) subjects in the cefiderocol group compared with 3.3% (5/150) in the meropenem group.Treatment-related TEAEs

Treatment-related TEAEs are presented by SOC and PT in Table 86. Overall, the incidence of treatment-related TEAEs was 9.5% (14/148) in the cefiderocol group and 11.3% (17/150) in the meropenem group. The most common treatment-related TEAE was diarrhoea, reported for 2.0% (3/148) subjects in the cefiderocol group compared with 3.3% (5/150) subjects in the meropenem group.

All treatment-related TEAEs associated with increases in liver enzyme in the cefiderocol group were transient and resolved or were resolving during the study. Overall, the majority of treatment-related TEAEs were either mild (n=15) or moderate (n=20), while 11 were severe.

5.5.2.3.6 Deaths

The primary objective of this study was to compare all-cause mortality between the 2 groups at Day 14 after start of study drug therapy in the mITT population. All-cause mortality rates for the mITT population are reported in the efficacy section.

5.5.2.3.7 Other SAEs

All SAEs reported during the study are presented by SOC and PT on file [238]. Overall, the frequency of SAEs was 36.5% (54/148) in the cefiderocol group compared with 30.0% (45/150) in the meropenem group. Overall, the most common SAE was cardiac arrest, reported in 4.7% (7/148) in the cefiderocol group compared with 3.3% (5/150) in the meropenem group.

Overall, treatment-related SAEs were reported in 2.0% (3/148) in the cefiderocol group compared with 3.3% (5/150) in the meropenem group [238].

Table 89 – Number (percent) of subjects with serious adverse events (SAEs) by organ class and preferred term (safety population)

5.5.2.3.8 TEAEs leading to study treatment discontinuation

All TEAEs leading to study treatment discontinuation are presented by SOC and PT in Table 119-13. TEAEs leading to study treatment discontinuation were reported for 8.1% (12/148) in the cefiderocol group and 9.3% (14/150) in the meropenem group. Alanine aminotransferase increased was the most frequently reported TEAE leading to discontinuation, in 2/148 (1.4%) subjects in the cefiderocol group. Hepatic enzymes increased was reported in no subjects in the cefiderocol group and 5/150 (3.3%) in the HD meropenem group. All other TEAEs leading to discontinuation were reported at most in 1 subject in either treatment group.

5.5.2.3.9 Conclusions for APEKS-NP Study

Overall, the types and frequency of TEAEs for cefiderocol were generally similar to high-dose meropenem and consistent with safety profile of cephalosporin class of antibacterials.

5.5.2.4 CREDIBLE-CR

ADVERSE EVENTS AND SERIOUS ADVERSE EVENTS

5.5.2.4.1 Treatment-Emergent Adverse Events

Over 90% of the subjects in each treatment group had at least 1 adverse event (Table 90). The incidence of treatment-related adverse events was 14.9% in the cefiderocol group and 22.4% in the BAT group. The incidence of adverse events with an outcome of death by the end of the study was 33.7% in the cefiderocol group and 18.4% in the BAT group. Of note, none of the deaths in the cefiderocol group were considered related to study treatment by either the investigator or Shionogi. The percentage of reported serious adverse events was 49.5% in the cefiderocol group and 46.9% in the BAT group. Overall, 6 subjects experienced treatment-related serious adverse events (1 in the cefiderocol group and 5 in the BAT group). The percentage of discontinuations due to adverse events was 9.9% in the cefiderocol group and 6.1% in the BAT group.

Table 90: Overview of Treatment-emergent Adverse Events (Safety Population)

Cefiderocol BAT (N = 101) (N = 49) Subjects Events Subjects Events Adverse Event Category n (%) n' n (%) n' AEs 92 (91.1) 634 47 (95.9) 311 Treatment-related AEs 15 (14.9) 27 11 (22.4) 16 Death 34 (33.7) 45 9 (18.4) 14 SAEs 50 (49.5) 92 23 (46.9) 36 Treatment-related SAEs 1 (1.0) 1 5 (10.2) 7 Discontinuation due to AEs 10 (9.9) 12 3 (6.1) 3 Discontinuation due to treatment-related 3 (3.0) 3 2 (4.1) 2 AEs AEs = adverse events; BAT = best available therapy; SAEs = serious adverse events

Percentage is calculated using the number of subjects in the column heading as the denominator. Adverse events that started after the first dose of the study drug and up to End of Study visit are defined as treatment-emergent. One subject received cefiderocol after completion of BAT; this subject is included under BAT in this table. Source: CREDIBLE-CR Final Study Summary [243] Treatment-emergent Adverse Events Reported in Either Treatment Group

The most frequently (> 10% of subjects in either arm) reported adverse events were diarrhea, pyrexia, septic shock, and vomiting. Adverse events reported more frequently (> 5% difference between the treatment groups) in the cefiderocol group than in the BAT group were diarrhea, alanine aminotransferase increased, aspartate aminotransferase increased, pleural effusion, and chest pain. Adverse events reported less frequently (> 5% difference between the treatment groups) in the cefiderocol group than in the BAT group were hypokalemia, hyperkalemia, rash, and depression.

Transient elevations in liver enzymes due to cefiderocol cannot be excluded; however, like other cephalosporins, the elevations are reversible after discontinuing the study drug, and none have resulted in serious hepatotoxicity.

All 6 chest pain cases were in the cefiderocol group and were reviewed. The majority of the events of chest pain were considered to be noncardiovascular in nature and not related to cefiderocol. Most of the remaining adverse events occurred at a low frequency, suggesting manifestations of the subjects’ underlying disease. Overall, no safety signals related to cefiderocol use were observed.

5.5.2.4.2 Treatment-related Adverse Events

Overall there were 15 (14.9%) of patients with treatment related AEs in the cefiderocol arm and 11 (22.4%) in the BAT arm.

Diarrhea (2.0%), liver function test abnormal (2.0%), alanine aminotransferase increased (3.0%), and aspartate aminotransferase increased (3.0%) were the most frequently reported treatment-related treatment-emergent adverse events in the cefiderocol group, while acute kidney injury (8.2%) was the most frequently reported treatment-related treatment-emergent adverse event in the BAT group.

Only 1 (1%; increase in transaminases) in cefiderocol arm and 5 (10.2%) in the BAT arm were considered serious (1 anaphilactic reaction; 1 septic shock; 1 metabolic acidosis; 1 status epilepticus; 2 acute renal failure; and 1 respiratory arrest). The higher incidence of treatment- related SAE in the BAT group was due to use of antibacterials with known renal toxicity in BAT group. In the BAT group, 5 SAEs of renal impairment (acute kidney injury and renal disorder) considered related to colistin and tobramycin were reported.

Table 91: Subjects with Treatment-related Adverse Events by Preferred Term (Safety Population)

Cefiderocol BAT (N = 101) (N = 49) Preferred Term n (%) n (%) Subjects with treatment-related AEs 15 (14.9) 11 (22.4) Alanine aminotransferase increased 3 (3.0) 0 Aspartate aminotransferase increased 3 (3.0) 0 Diarrhoea 2 (2.0) 0 Liver function test abnormal 2 (2.0) 0 Ascites 1 (1.0) 0 Blood creatinine increased 1 (1.0) 0 Blood pressure increased 1 (1.0) 0 Clostridium difficile colitis 1 (1.0) 0 Drug eruption 1 (1.0) 0 Dysgeusia 1 (1.0) 0 Hypertension 1 (1.0) 0 Hypokalaemia 1 (1.0) 0 Oedema 1 (1.0) 0 Pleural effusion 1 (1.0) 0

Cefiderocol BAT (N = 101) (N = 49) Preferred Term n (%) n (%) Pseudomembranous colitis 1 (1.0) 1 (2.0) Pyrexia 1 (1.0) 0 Rash 1 (1.0) 0 Transaminases increased 1 (1.0) 0 Upper gastrointestinal haemorrhage 1 (1.0) 0 Acute kidney injury 0 4 (8.2) Anaphylactic reaction 0 1 (2.0) Blood creatine increased 0 1 (2.0) Hepatic enzyme increased 0 1 (2.0) Metabolic acidosis 0 1 (2.0) Renal disorder 0 1 (2.0) Respiratory arrest 0 1 (2.0) Sepsis 0 1 (2.0) Septic shock 0 1 (2.0) Status epilepticus 0 1 (2.0) Vomiting 0 1 (2.0) AEs = adverse events; BAT = best available therapy

Percentage was calculated using the number of subjects in the column heading as the denominator. Adverse events that started after the first dose of the study drug and up to End of Study visit were defined as treatment-emergent. Although a subject may have had 2 or more adverse events, the subject was counted only once within a System Organ Class category. The same subject may have contributed to 2 or more Preferred Terms in the same System Organ Class category. One subject received cefiderocol after completion of BAT; this subject is included under BAT in this table. The most frequently reported treatment-related treatment- emergent adverse events are shown in bold. Source: CREDIBLE-CR Final Study Summary [243]

5.5.2.4.3 Serious Adverse Events

Septic shock was the most frequently reported serious adverse event in both the cefiderocol (11.9%; 12/101 subjects) and BAT (12.2%; 6/49 subjects) groups (Table 92).

Table 92: Subjects with Serious Adverse Events by System Organ Class and Preferred Term (Safety Population)

Cefiderocol BAT System Organ Class (N = 101) (N = 49) Preferred Term n (%) n (%) Subjects with SAEs 50 (49.5) 23 (46.9) Blood and lymphatic system disorders 1 (1.0) 1 (2.0) Anaemia 0 1 (2.0) Febrile neutropenia 1 (1.0) 0 Cardiac disorders 6 (5.9) 4 (8.2) Bradycardia 1 (1.0) 1 (2.0) Cardiac arrest 4 (4.0) 2 (4.1) Cardiac failure congestive 1 (1.0) 0 Myocardial infarction 1 (1.0) 0 Pulseless electrical activity 0 1 (2.0) Gastrointestinal disorders 5 (5.0) 0 Abdominal pain 1 (1.0) 0 Abdominal pain upper 1 (1.0) 0 Gastrointestinal haemorrhage 1 (1.0) 0 Intestinal ischaemia 1 (1.0) 0 Lower gastrointestinal haemorrhage 1 (1.0) 0 Pancreatitis 1 (1.0) 0 Small intestinal obstruction 1 (1.0) 0 General disorders and administration site conditions 7 (6.9) 3 (6.1) Chills 1 (1.0) 0 General physical health deterioration 0 1 (2.0) Multi-organ failure 2 (2.0) 2 (4.1) Pyrexia 3 (3.0) 0

Cefiderocol BAT System Organ Class (N = 101) (N = 49) Preferred Term n (%) n (%) Sudden death 1 (1.0) 0 Hepatobiliary disorders 3 (3.0) 0 Chronic hepatic failure 1 (1.0) 0 Hepatic failure 1 (1.0) 0 Hepatitis 1 (1.0) 0 Immune system disorders 0 1 (2.0) Anaphylactic reaction 0 1 (2.0) Infections and infestations 29 (28.7) 11 (22.4) Bacteraemia 3 (3.0) 0 Bacterial infection 1 (1.0) 0 Device related infection 0 1 (2.0) Empyema 1 (1.0) 1 (2.0) Endocarditis 0 1 (2.0) Enterococcal bacteraemia 1 (1.0) 0 Enterococcal infection 2 (2.0) 0 Meningitis 0 1 (2.0) Necrotising fasciitis 0 1 (2.0) Osteomyelitis 1 (1.0) 0 Osteomyelitis acute 0 1 (2.0) Pneumonia 5 (5.0) 1 (2.0) Pneumonia bacterial 1 (1.0) 0 Renal abscess 1 (1.0) 0 Sepsis 3 (3.0) 0 Septic shock 12 (11.9) 6 (12.2) Systemic candida 1 (1.0) 0 Urinary tract infection 1 (1.0) 0 Urosepsis 1 (1.0) 0 Investigations 5 (5.0) 3 (6.1) Liver function test abnormal 4 (4.0) 3 (6.1) Transaminases increased 1 (1.0) 0 Metabolism and nutrition disorders 3 (3.0) 1 (2.0) Hyponatraemia 1 (1.0) 0 Metabolic acidosis 2 (2.0) 1 (2.0) Neoplasms benign, malignant and unspecified (incl cysts 1 (1.0) 0 and polyps) Lung neoplasm malignant 1 (1.0) 0 Nervous system disorders 3 (3.0) 2 (4.1) Dizziness 1 (1.0) 0 Hypoaesthesia 1 (1.0) 0 Neurological decompensation 1 (1.0) 0 Paraesthesia 1 (1.0) 0 Quadriplegia 0 1 (2.0) Status epilepticus 0 1 (2.0) Renal and urinary disorders 6 (5.9) 2 (4.1) Acute kidney injury 3 (3.0) 2 (4.1) Anuria 1 (1.0) 0 Nephrolithiasis 1 (1.0) 0 Oliguria 2 (2.0) 0 Respiratory, thoracic and mediastinal disorders 7 (6.9) 2 (4.1) Acute respiratory failure 1 (1.0) 1 (2.0) Chronic obstructive pulmonary disease 1 (1.0) 0 Obstructive airways disorder 1 (1.0) 0 Pneumonia aspiration 2 (2.0) 0 Respiratory arrest 0 1 (2.0) Respiratory failure 2 (2.0) 0 Vascular disorders 2 (2.0) 2 (4.1) Hypotension 2 (2.0) 1 (2.0) Shock 1 (1.0) 1 (2.0) BAT = best available therapy; SAEs = serious adverse events

have had 2 or more adverse events, the subject was counted only once within a System Organ Class category. The same subject may have contributed to 2 or more Preferred Terms in the same System Organ Class category.

One subject received cefiderocol after completion of BAT; this subject is included under BAT in this table.

Source: CREDIBLE-CR Final Study Summary [243]

5.5.2.4.4 Adverse Events Leading to Death

Adverse events leading to death are summarized in the efficacy section (5.4.5 and the summary study report [243]) There were 20.6% (21/102) of patients in the cefiderocol group and 6.3% (3/48) of patients in the BAT group with deaths classified in the SOC of Infections and Infestations. After reviewing the details for each individual patient, this imbalance in mortality between the treatment groups was not considered a safety issue, considering the complicated comorbidities and difficult-to-treat infections in this patient population. As per request of EUnetHTA, mortality data is included in the efficacy section.

5.5.2.4.5 Discussion

Limitations to Detect Adverse Reactions in Clinical Trial Development Programmes

The limitations on adverse drug reactions (ADR) detection are based on the information in Table 93 for patients with Gram-negative infections (including complicated urinary tract infection) and considering patients with at least 7 days of treatment with cefiderocol. In addition to the differences in adverse event reporting which occur in open label vs double-blind studies, the study populations in the three studies are very different, with the cUTI study subjects more clinically stable than those in the APEKS NP and CREDIBLE-CR studies.

Table 93: Limitations to detect adverse events in clinical trial programmes

Source: EU Risk Management Plan [284]

Table 94: Methods of data collection and analysis of AE, TEAE and SAE

Study Endpoint definition Method of analysis reference/ID APEKs-cUTI Adverse Events (AE) or The severity of an event was graded according to the APEKs-NP Treatment-emergent Adverse following definitions: CREDIBLE-CR Events (TEAE) • Mild: A finding, or symptom was minor and did not An AE was defined as any interfere with usual daily activities untoward medical occurrence in • Moderate: The event was discomfort and caused a subject administered a interference with usual daily activity or affected clinical pharmaceutical product status (including investigational drug) • Severe: The event caused interruption of the subject's during a clinical investigation. An usual daily activities or had a clinically significant effect AE could therefore be any unfavourable and unintended The relationship of an event to the study drug was sign (including an abnormal determined according to the following criteria: laboratory finding), symptom, – Related: An AE that can be reasonably unplanned procedure, or disease explained that the study drug caused the AE. temporally associated with the For example, the occurrence of the AE cannot use of an investigational product, be explained by other causative factors, but whether considered related to can be explained by pharmacological effect of the investigational product. the study drug, such as a similar event had been reported previously, or increasing/decreasing the dose affects the occurrence or seriousness of the AE, etc.

 Not Related: An AE that cannot be reasonably explained that the study drug caused the AE Unless otherwise noted, the summary of AEs will be performed for events of treatment emergent. An expected treatment-related AE was any AE that was consistent with the current Investigator's Brochure for cefiderocol. Expectedness A treatment-related AE is considered expected if it is listed in Expected Adverse Reactions in Section “Undesirable Effects” of “SUMMARY OF DATA AND GUIDANCE FOR INVESTIGATORS” in the current investigator's brochure for cefiderocol. The expected adverse reactions of comparators will be those found in the EMA SmPC. The expected adverse reactions of linezolid will be those found in the local SmPC. Expectedness will be assessed by the sponsor.

Clinical Laboratory Adverse Events

For any abnormal laboratory test results (haematology, blood chemistry, or urinalysis) or other safety assessments (e.g., physical examination, vital signs) that are worsening from baseline or occur thereafter in the course of the study, the investigator or sub- investigator will consider whether these results are clinically significant. Abnormal laboratory test results are defined as values outside the reference range. Any test results which are clinically significant at the discretion of the investigator or sub investigator are to be recorded as AEs. If an abnormal laboratory finding is associated with disease or organ toxicity, the investigator should report only the disease or organ toxicity as AEs. These AEs should also be assessed as to whether they meet the definition of seriousness and reported accordingly. The investigator or sub-investigator will consider test results to be clinically significant in the following circumstances:

 Test result leads to any of the outcomes included in the definition of an SAE.  Test result leads to discontinuation from the study.  Test result leads to a concomitant drug treatment or other therapy.  Test result requiring additional diagnostic testing or other medical intervention.  Test result meeting the management criteria for liver function abnormalities identified in the Appendix 6 of the statistical analysis plan (SAP). Liver Abnormalities Management and Discontinuation Criteria for Abnormal Liver Function tests have been designed to ensure subject safety and evaluate liver event aetiology. The investigator or sub-investigator must review study subject laboratory results as they become available to identify if any values meet the criteria in Appendix 6. When any test result meets the management criteria for liver function abnormalities, the results of further assessments and required FUP. Serious Adverse Events (SAE) The severity of a SAE was graded to the listed criteria. An SAE is defined by regulation as any AE occurring at any dose

that results in any of the following outcomes:

 Death  Life-threatening condition  Hospitalization or prolongation of existing hospitalization for treatment  Persistent or significant disability/incapacity  Congenital anomaly/birth defect  Other medically important condition

5.6 Conclusions

1. Provide a general interpretation of the evidence base considering the benefits associated with the technology relative to those of the comparators.

Cefiderocol is an innovative siderophore cephalosporin antibacterial with a unique molecular structure designed to provide high stability to carbapenemases and use bacteria’s own mechanism of iron uptake. Both these attributes enable cefiderocol to overcome three main mechanisms of beta-lactam bacterial resistance (degradation by β-lactamase enzymes, porin channel mutations, and overexpression of efflux pumps), which is translated in a wide activity spectrum against aerobic Gram-negative pathogens, including the MDR and WHO critically important carbapenem resistant strains of Enterobactereacea, A. baumanii and P. aeruginosa, as well as intrinsically CR S. maltophilia. MDR pathogens are difficult to treat, have limited treatment options, and no existing treatment provides both comprehensive coverage and good safety profile. Cefiderocol therefore constitutes an effective and safe treatment option for patients with serious infections in the presence of world-wide growing resistances.

MDR infections, including those resistant to carbapenems, primarily occur in vulnerable hospitalised patients. The treatment of MDR-GNB infections in critically ill patients presents many challenges are associated with poorer outcomes including increased mortality, increased length of stay and healthcare resource utilization, compared to non-resistant pathogens. Since an effective treatment should be administered as soon as possible, resistance to many antimicrobial classes almost invariably reduces the probability of adequate empirical coverage, with possible unfavorable consequences in terms of increased mortality, length of stay and healthcare reseource utilization.

In this light, readily available patient’s medical history and updated information about the local microbiological epidemiology remain critical for defining the baseline risk of MDR-GNB infections and firmly guiding empirical treatment choices, with the aim of avoiding both undertreatment and overtreatment. Treatment of severe MDR-GNB infections in critically ill patients requires a expert and complex clinical reasoning, taking into account the peculiar characteristics of the target population, but also the need for adequate empirical coverage and the more andmore specific enzyme-level activity of novel antimicrobials with respect to the different resistance mechanisms of MDR-GNB.

Due to the urgent need to develop new treatments based on the underlying pathogens rather than the infection site, the EMA label is expected to authorize cefiderocol to be used for treatment of infections due to aerobic Gram-negative organisms in adults with limited treatment options

Assessment of the effectiveness of cefiderocol is based on the integration of in vitro, PK/PD and clinical data. Large susceptibility studies have confirmed cefiderocol wider Gram-negative spectrum, and being a more potent antimicrobial agent than comparators. It’s very favourable minimum inhibitory concentrations (MICs) have been shown to correlate well with in vivo efficacy and randomized clinical trials in patients with cUTI, nosocomial pneumonia, and BSI have provided confirmation of the efficacy and safety of cefiderocol in key target patient populations. These reflect the label and are pathogen focused, not restricted to any specific site of infection and supports the use of cefiderocol in two types of patients:

Cefiderocol is a time-dependent cephalosporin. Preclinical studies showed that cefiderocol has linear pharmacokinetics, primarily urinary excretion, an elimination half-life of 2–3 hours, and a protein binding of 58% in human plasma. The probability of a target attainment at ≥75% of the dosing interval during which the free drug concentration exceeds the minimum inhibitory concentration (ƒT/MIC) for bacterial strains with an MIC of ≤4 μg/mL was greater than 90% at the therapeutic dose of 2 g over 3-hour infusion every 8 hours in most patients. Only renal function markers were found to be influential covariates for the pharmacokinetics of cefiderocol for patients with altered renal function. Dose adjustment is recommended for patients with impaired and augmented renal function.

The potent activity of cefiderocol was confirmed in an extensive series of in vitro studies, against clinical isolates from surveillance studies, and in animal infection models. The SIDERO-WT study showed ccefiderocol to have activity against 99.5% of Gram-negative isolates at a MIC of 4 mg/L, while the SIDERO-CR study, which only includes CR isolates, showed cefiderocol to have potent in vitro activity at a MIC of 4 mg/L against 96.2% of isolates of carbapenem-non-susceptible pathogens including all of the WHO priority pathogens and Stenotrophomonas maltophilia. In both studies, cefiderocol was found to give wider Gram-

negative coverage, and to be a more potent in vitro antibacterial agent than comparators. The results confirm cefiderocol overcomes multiple mechanisms or resistance and to be stable against the 4 known classes of β-lactamases, including serine carbapenemases, with potency which is equal to or greater than comparators.

An improved in vitro potency in addition to a well-characterized favorable PK/PD profile are crucial to achieve both adequate exposure to the antibacterial over the MIC for the pathogen, and clinical cure in patients infected with drug-resistant pathogens [52]. Therefore, clinical studies in antimicrobials, provide only supportive safety and efficacy evidence to the pivotal in-vitro and PK/PD data. Furthermore, in the context of antibacterial resistance, the standard clinical trial approach aiming at demonstrating superiority over existing treatments is not feasible. Treatment options for MDR infections do not allow a superiority trial and it would be unethical to wihthold effective treatment to pateints in such trials [52]. Hence, clinical trials have an important role to confirm clinical efficacy, but a limited role in providing comparative evidence outside the trial, as only pathogens that fall within the in-vitro spectrum of the tested treatments and comparators are included in the study. This is particularly relevant for antimicrobial treatment selection in the absence of antibiogram.

The clinical evidence to support the use of cefiderocol is based on 2 randomised, double blinded clinical trials, and 1 descriptive open-label study. Data from an NMA, an effectiveness model and compassionate use cases complement the body of confirmatory clinical data.

Low likelihood of in treatment development of resistance against cefiderocol was demonstrated by the fact that only very few and moderate increases in the cefiderocol MIC were seen over the treatment course, usually requiring more than 1 simultaneous mutation to increase the MIC. Cefiderocol also presents low likelihood of generating cross resistance, given that the main resistant pathway identified in in vitro studies was related with the siderophore ion uptake.

5.6.1 Evidence to support use of cefiderocol in patients with infections by suspected MDR/CR pathogens:

SIDERO WT provides evidence to support the use of cefiderocol in the group of critically ill patients with infections suspected to be caused by a MDR pathogen, who require immediate treatment. These patients would benefit from the availability of an additional effective antibiotic treatment providing full cover against carbapenem-resistant pathogens while pathogen susceptibility is being confirmed.

The SIDERO-WT study tested the in-vitro antibacterial activity of cefiderocol against Gram- negative bacteria [29]. A total of 30,459 clinical isolates of Gram-negative bacilli were systematically collected from USA, Canada, and 11 European countries between 2014 and 2017. Cefiderocol demonstrated activity against 99.5% of Gram-negative isolates at a MIC of 4 mg/L. Isolates were less susceptible to the comparators including colistin (95.5%), ceftazidime-avibacatam (90.2%) and ceftolozane-tazobactam (84.3%).

In a retrospective analysis comparing the probability of target attainment (PTA) for cefiderocol, ceftolozane/tazobactam and meropenem against Enterobacterales and Pseudomonas aeruginosa in a representative patient population at risk of MDR or carbapenem resistant infections, the cumulative fractions of response (CFRs) calculated using European MIC distributions from the SIDERO surveillance for cefiderocol against Enterobacterales and Pseudomonas spp. are considerably higher than seen for meropenem and ceftolozane- tazobactam.

In patients with infections suspected to be caused by MDR/CR pathogens, clinical trials only provide limited comparative evidence regarding the efficacy of new antibacterials. This is because trials must include only pathogens for which the tested agents and comparators are effective, as it would be unethical to knowingly allow patients to have ineffective treatment. In this setting, standard NMAs also provide little information, as they never account for pathogens not susceptible to the treatment regimens included in the network. A comparison of efficacy against all relevant comparators can only be obtained from in-vitro surveillance studies. Hence approaches integrating all available evidence from in vitro, PK/PD and clinical data (such as effectiveness models), are the necessary to predict susceptibility rates and clinical effectiveness rates.

APEKS-cUTI compared cefiderocol with imipenem/cilastatin (IPM/CS) in cUTI caused by Gram-negative MDR pathogens in hospitalized adults. The study was designed to demonstrate non-inferiority, with the primary efficacy endpoint being the composite of clinical response and microbiological response at TOC. 73% of patients in the cefiderocol group achieved the primary endpoint, vs only 55 % of patients in the IPM/CS group, with an adjusted treatment difference of 18.6% (95 % CI: 8.2, 28.9). A further post-hoc analysis confirmed superiority in favour of cefiderocol.

Given the similarority of patients and pathogens included in across trials, a NMA was conducted to compare the result of APEKS cUTI with relevant comparator studies. Results showed no statistically significant difference between the APEKS cUTI result and results from

studies of ceftazidime/avibactam and ceftolozane/tazobactam conducted in a similar population with a similar pathogen distribution.

The APEKS-NP study compared treatment with cefiderocol against high-dose and prolonged infusion (HD) meropenem in patients with nosocomial pneumonia caused by MDR Gram- negative pathogens. Cefiderocol met the primary endpoint of non-inferiority in ACM at day 14 versus HD meropenem (12.4% vs 11.6%; (95 % CI: -6.6, 8.2)). APEKS-NP used an improved meropenem regimen (both high dose and prolonged infusion time) to optimize its exposure and efficacy. This meant that a NMA was not possible because previously published meropenem studies had used a lower dose of meropenem.

The results of the two randomized, double-blind APEKS trials combined provide highly reliable and clinically relevant evidence to support the use of cefiderocol in patients with suspected MDR pathogens with limited treatment options.

Furthermore, in an analysis incorporating European pathogen epidemiology and susceptibility data, cefiderocol provides the best predicted susceptibility rates and estimated clinical and microbiological success rates regardless of the infection site, in the absence of an antibiogram for the critically ill patients with suspected MDR pathogen infection requiring immediate treatment.

Combining these results and clinical data in an effectiveness model, show that cefiderocol has a greater likelihood of achieving microbiological eradication and clinical cure, in the patients with suspected MDR/CR infections than relevant comparators across for cUTI and pneumonia. In the absence of antibiogram, cefiderocol provides an effective option for treating critically ill, hospitalised patients where CR infection is suspected and time to effective treatment must be as short as possible, increasing the likelihood of providing an initial appropriate therapy and potentially avoiding worse morbidity and mortality outcomes associated with delayed effective therapy.

5.6.2 Evidence to support use of cefiderocol in patients with infections by confirmed CR pathogens:

Data from the SIDERO-CR study indicate that cefiderocol maintains high activity in the presence of beta-lactamases, carbapenemases, and strains with porin channel mutations or efflux-pump overexpression. Patients with confirmed MDR/CR infections, for whom the antibiogram indicates susceptibility for cefiderocol thus gain an additional treatment option with equal or higher susceptibility than comparators.

Clinical trials can provide more reliable information regarding comparative efficacy when the pathogens have confirmed or expected susceptibility to both drugs. This is consistent with prescription based on AST results, which occurs in patients with confirmed CR infections. In this setting, Network meta-analysis (NMA) if feasible provide additional reliable information of comparative effectiveness.

Evidence of clinical efficacy of cefiderocol in patients with a confirmed CR infection comes from three sources; the APEKS NP study, the CREDIBLE CR study and the cefiderocol compassionate use programme:

In a small subgroup of patients participating in the APEKS-NP that was non-susceptible to meropenem considering a breakpoint of 8mg/L (MIC), similar results in terms of mortality, clinical and microbiological outcomes were achieved between arms. However, when looking into the stratification for pathogens with MIC >16 mg/mL, patients on cefiderocol had reduced mortality and higher clinical cure rates. The HD prolonged infusion meropenem regimen in this trial, increased exposure in terms of time and concentration to the infection site, increasing the likelihood of effectiveness, even on pathogens with MIC up to 16mg/mL.

The CREDIBLE CR study was a small, randomised, open label, descriptive, exploratory, study conducted to evaluate efficacy in patients with confirmed CR infections given cefiderocol or BAT. The study was not designed or powered for statistical comparison between arms. The study included 150 severely ill patients, consistent with compassionate use cases, with a range of infection sites including nosocomial pneumonia, cUTI, BSI/sepsis. Many patients had end stage comorbidities and had failed multiple lines of therapy. Cefiderocol and BAT were shown to be effective in terms of clinical and microbiological outcomes in these patients, particularly for cefiderocol also showing clinical and microbiological efficacy regardless of carbapenemases present in the pathogen causing the infection. However, there were marked clinically relevant differences in some baseline characteristics and pathogen distribution of the cefiderocol and BAT arms. An imbalance in mortality was observed in the cefiderocol arm

compared to BAT (18/49 vs 5/25), which was not considered to be related with safety signals. No deaths were found to be causally associated with cefiderocol through assessment by the investigator and two independent committees. No single factor that would explain the imbalance was identified. Small patient numbers and multiple confounders preclude definitive conclusions. Additional analyses revealed that mortality in the treatment arm was similar to other studies in the context, while the BAT arm performed better than all reported studies, particularly for non-fermenters. The reasons for this are not understood.

Compassionate use program: More than 200 patients were treated with cefiderocol within the compassionate use programme around the world, demonstrating the unmet medical need. Confirmed information on 74 patients who have completed treatment in the compassionate use program showed that over 60% of the severely ill patients infected with CR Gram-negative pathogens survived when no other treatment option was available to them.

The overall mortality across the compassionate use programs and CREDIBLE CR was similar, 36.5% and 33.7% respectively, supporting the notion that the population recruited into the CREDIBLE-CR trial, included severely ill patients with a very poor prognosis, similar to those applying for compassionate use and other similar studies reported on literature.

5.6.3 Quality of Life

Patients with these severe nosocomial infections are frequently treated in ICU units, often unconscious, and on many occasions require ventilation (intubation), preventing investigation of patient-reported outcomes. Because the most severely ill patients cannot complete questionnaires, this can lead to systematic under-reporting QoL data of the most severe courses of illness. The fact that these patients are hospitalised already has decrimental impact on their quality of life. The ward in the hospital also impacts the patient’s quality of life (i.e patients on ICU or isolation, are expected to have lower quality of life compared to general ward), although this may be correlated with the severity of the infection and underlying condition. All these factors make investigating quality of life in antimicrobial clinical trials difficult and infrequent. The microbiological outcomes and mortality have thus been deemed to be most relevant, also by regulators. No PROs are, therefore, reported in the dossier. However, any therapy that resolves the infection and/or reduces length of hospitalization is expected to improve patient’s quality of life.

5.6.4 Comparators

The in vitro data and combination of the in vitro, PK/PD, and clinical data show that cefiderocol outperforms all relevant comparators with regard to the likelihood of obtaining microbiological

eradication in the population with suspected MDR/CR infections. While clinical evidence is restricted to a limited number of comparators that were deemed to be relevant in the specific context by regulators, an NMA in the cUTI indication and the additional in vitro data reveal favourable outcomes of cefiderocol compared with all relevant available treatments (Table 95).

Table 95 - Comparator overview Population Comparator Data source Result (cefiderocol vs. comparator) Suspected High dose Meropenem SIDERO WT Broader coverage of Gram-negative, MDR/CR surveillance aerobic pathogens. Lower MIC value and preserved efficacy in the presence of carbapenemases. APEKS-NP Non-inferior with regard to mortality RCT (primary outcome) and all clinical and microbiological secondary outcomes. High dose Meropenem Integrated Cefiderocol presents higher weighed Ceftalozane-tazobactam, epidemiology susceptibility rates in cUTI, Ceftazidime-avibactam and in-vitro pneumonia, BSI, and gastrointestinal data analysis samples vs comparators High dose Meropenem Effectiveness Cefiderocol presents higher Ceftalozane-tazobactam, model likelihood of clinical and Ceftazidime-avibactam integrating microbiological effectiveness in epidemiology, pneumonia and cUTI vs in-vitro data comparators. and clinical data Imipenem/Cilastatin APEKS-cUTI Non-inferior to comparator, but RCT proven superiority in a post-hoc analysis, on the primary endpoint of combined microbiological eradication / clinical cure at TOC, and secondary endpoint microbiological eradication at TOC. Ceftalozane-tazobactam, network meta- In similar patient populations with ceftazidime-avibactam, analysis for similar pathogen distribution across doripenem, cUTI different trials, and consistent with imipenem/cilastatin APEKS-cUTI there was statistical significant difference in microbiological eradication at TOC vs Imipenem/cilastatin, but in all other endpoints there was no statistically significant difference, including clinical cure rates and adverse events Ceftolozane/tazobactam SIDERO WT Lower MIC90 (0.25 vs. 8 for surveillance Pseudomonas, 0.25 vs. 32 for

Acinetobacter, 1 vs. 64 for Enterobacteriaceae)6 Higher % isolates susceptible to cefiderocol Ceftazidime/avibactam SIDERO WT Same MIC90 for Enterobacteriaceae surveillance (1 vs. 1), otherwise superiority of cefiderocol Higher % isolates susceptible to cefiderocol Confirmed Colistin-based SIDERO CR Higher % isolates susceptible to CR (combination) regimens surveillance cefiderocol; Similar in-vitro efficacy. (most relevant for for A. Colistin is known to have severe side Baumanii, S. maltophilia, effects, especially kidney toxicity. pathogens with Resistances against colistin have metallobeta-lactamases) been reported to increase in epidemiological studies. Ceftolozane/tazobactam SIDERO CR Higher percent susceptibility for (most relevant for P. surveillance cefiderocol against Acinetobacter aeruginosa, except and Pseudomonas across all pathogens with included countries (MEM-NS metallobeta-lactamases) pathogens)7 Ceftazidime/avibactam SIDERO CR Higher percent susceptibility for (most relevant for surveillance cefiderocol against Acinetobacter Enterobacterales, except and Pseudomonas across all pathogens with included countries (MEM-NS metallobeta-lactamases) pathogens) Best available therapy CREDIBLE-CR Descriptive results only. Evidence of (BAT), predominantly eradication of resistant pathogens. (combination) regimens Numerical, non-significant (most relevant for A. disadvantage with regard to mortality Baumanii, S. maltophilia, for cefiderocol compared to BAT. pathogens with metallobeta-lactamases)

2. Provide a general interpretation of the evidence base considering the harms associated with the technology relative to those of the comparators.

The presented data demonstrate that cefiderocol has a similar safety profile compared to other cephalosporins.

Pre-clinical studies showed that single and multiple doses of cefiderocol tested were well tolerated in both healthy subjects and those with renal impairment. Furthermore, neither QT interval prolongation nor drug–drug interaction via organic anion transporters was demonstrated in healthy subjects.

The clinical safety for cefiderocol was established in the three randomised clinical trials, including 549 treated patients, and showed a similar profile compared to other cephalosporins.

6 Longshaw et al., 2019 ID Week 7 Sato et al. 2019 ID Week

Pooled adverse event analyses showed that there were overall less treatment emergent adverse events with cefiderocol (344/549 [67.1%]) vs comparators (252/347 [72.6%]), as well as less treatment related AEs, (56/549 [10.2%]) with cefiderocol vs compartors (45/347 [13.0%]). The large APEKS trials with active comparators showed that TEAEs and treatment-related TAEs were overall balanced between arms. In APEKS-NP, adverse events were experienced by most subjects in both treatment groups with SAE rates being slightly higher in the cefiderocol group (36.5%) than in in the meropenem group (30%). In the APEKS-cUTI trial, serious adverse events (SAE) occurred less in cefiderocol-treated patients than in IPM/CS- treated patients (5% vs 8%).

In the confirmed carbapenem-resistant CREDIBLE-CR trial, the cefiderocol group had lower incidence of AEs and treatment-related AEs, but imbalance in mortality, SAEs and discontinuation due to AEs, compared with BAT was observed. The incidence of treatment- related AEs leading to discontinuation was similar between treatment groups. A blinded adjudication committee concluded that none of the deaths was due to a drug-related AE.

The SPC details all potential risks associated with drug interactions or potential harms with drug use in special categories of patients.

5.7 Strengths and limitations

1. Summarise the internal validity of the evidence base, considering the study quality, the validity of the endpoints used as well as the overall level of evidence. Include a statement about the consistency of the results in the evidence base.

5.7.1 Risk of bias assessment

Unlike therapeutic areas, in vitro studies are key sources of data to substantiate clinical use of the antibacterials. Traditionally this falls outside the scope of bias assessment, as theoretically the risk of bias is considered minimal. In this case, same isolates were tested for all comparators, the methodology used was based on standard defined methods, and data was reported. The manufacturer provided the study protocol, and several publications for this assessment; thus, the possibility of selective outcome reporting is regarded as low. In summary, robustness of the study is ensured through large number of isolate samples, testing same sample for all comparators. The study shows high internal validity with low risk of bias.

In addition to the in vitro and PK/PD data, the evidence base in the population with suspected MDR/difficult-to-treat infections is amended by two RCTs, the APEKS cUTI and APEKS NP trials. Both studies were multicentre, multinational, double-blind, randomized, active-controlled studies.

APEKS-cUTI was a Multicentre, Double-blind, Randomized, Clinical Study to Assess the Efficacy and Safety of Intravenous S-649266 (Cefiderocol) in Complicated Urinary Tract Infections with or without Pyelonephritis or Acute Uncomplicated Pyelonephritis Caused by Gram-Negative Pathogens in Hospitalized Adults in Comparison with Intravenous Imipenem/Cilastatin. Randomization was stratified according to the patient’s clinical diagnosis, (cUTI with or without pyelonephritis and AUP) and region (North America, European Union, Russia, and Japan plus the rest of world). Randomization used a computer-generated randomization list (IXRS® provider), an interactive web or voice response system (IWRS/IVRS) was used to assign a total of 450 patients to identification numbers for which treatment has already been randomly assigned. Patients and investigator, site personnel, the sponsor, and the sponsor’s designees involved in blinded monitoring, data management, or other aspects of the study were blinded to treatment assignment. The site pharmacist or qualified designee who prepared the intravenous infusion solution was the only study site personnel with the identification of the study drug assignments for that site. Generation of randomization sequence and allocation concealment are considered adequate for this

study. Performance and detection bias were minimized through the described blinding and alignment of infusion duration. mITT population proportions were comparable in both arms with 252/300 for cefiderocol and 119/148 for IPM/CS thus reducing the likelihood of attrition bias. The main study publication (Portsmouth et al 2018) reported primary outcome (composite outcome at TOC) by predefined subgroups and microbiological and clinical secondary outcomes at the predefined time points EA, EOT, TOC, FU as well as any AE, treatment-related AEs, SAEs, AEs leading to discontinuation, deaths and AEs in at least 2% of patients in either treatment group. The manufacturer provided the study protocol, SAP, CSR and study synopsis for this assessment; thus, the possibility of selective outcome reporting is regarded as low. In summary, robustness of the study is ensured through randomization and stratification, blinding and large number of patients. The study shows high internal validity with low risk of bias at the study level.

APEKS-NP was a Phase 3, Multicentre, Randomized, Double-blind, Parallel-group, Clinical Study of Cefiderocol Compared with Meropenem for the Treatment of Hospital-acquired Bacterial Pneumonia, Ventilator-associated Bacterial Pneumonia, or Healthcare-associated Bacterial Pneumonia Caused by Gram-negative Pathogens. Treatments were randomized to subject identification numbers by the interactive response technology (IRT) provider in a 1:1 fashion to cefiderocol or meropenem. An IRT was used to assign a total of 300 subjects to identification numbers for which treatment has already been randomly assigned. Randomization was performed by the stratified randomization method using their infection type (HABP, VABP, and HCABP) and APACHE II score (≤ 15 and ≥ 16) as allocation factors. Linezolid infusion did not require blinding and was labelled with the study number, subject’s identification number, and infusion rate and drug name. Patients and investigator, site personnel, the sponsor, and the sponsor’s designees involved in blinded monitoring, data management, or other aspects of the study were blinded to treatment assignment. The site pharmacist or qualified designee who prepared the intravenous infusion solution was the only study site personnel with the identification of the study drug assignments for that site. Generation of randomization sequence and allocation concealment are considered adequate for this study. Performance and detection bias were minimized through the described blinding and alignment of infusion duration. mITT population proportions were comparable in both arms with 145/148 for cefiderocol and 147/150 for high dose meropenem, equally the microbiologically-evaluable Per-protocol (ME-PP) population was balanced (105 for cefiderocol and 101 for high dose meropenem), thus reducing the likelihood of attrition bias. Results of the study have been presented in an international clinical conference, but have not yet been published as a manuscript and no results have been posted at

clicnitrials.gov yet, however, the manufacturer provided the study protocol, SAP, and in the absence of available CSR at the date of EUnetHTA submission, the manufacturer also provided study synopsis and all the documentation submitted to EMA for this assessment thus possibility of selective outcome reporting is regarded as low. Thus, robustness of the study is ensured through randomization and stratification, blinding and large number of patients. The study shows high internal validity with low risk of bias at study level.

CREDIBLE-CR was a Multicentre, Randomized, Open-label Clinical Study of S-649266 or Best Available Therapy for the Treatment of Severe Infections Caused by Carbapenem- resistant Gram-negative Pathogens. The study is a small descriptive study, with no inferential analysis planned. The treatments were randomized to subject identification numbers by the IXRS® provider in a 2:1 fashion, i.e. to cefiderocol and BAT, respectively. An interactive web or voice response system (IWRS/IVRS) was used to assign patients to identification numbers for which treatment has already been randomly assigned. Randomization was performed by the stochastic minimization method using the infection site (HAP/VAP/HCAP, cUTI, and BSI/sepsis), APACHE II score (≤15 and ≥16), and region (N. America, S. America, Europe, and Asia- Pacific) as allocation factors, but did not account for pathogen stratification or other clinically relevant factors. To avoid deterministic allocation based on the ongoing allocation results, probabilistic allocation was incorporated [Pocock SJ, Simon R. Sequential Treatment Assignment with Balancing for Prognostic Factors in the Controlled Clinical Trial. Biometrics 1975; 31: 103-15.]. Planned proportions were approximately 50% with HAP/VAP/HCAP; cUTI no more than 30% and the remainder of patients were enrolled under the BSI/sepsis diagnosis. The randomization ratio of patients between treatment groups based on clinical diagnosis was maintained through the allocation factor of clinical diagnosis at the time of randomization. BAT was the standard of care for CR infections at each enrolling study site and could include up to three antibiotics with Gram-negative coverage used in combination. The comparator BAT could not be defined in the protocol and BAT was determined by the site investigator based on the assessment of the patient’s clinical condition and had to be determined by the investigator prior to randomization. The dosage of BAT was adjusted according to the local country-specific label. De-escalation of BAT was allowed. Concomitant antibiotics were allowed if the patients had a confirmed/suspected Gram-positive or anaerobic co-infection (e.g., vancomycin, daptomycin, linezolid, clindamycin, or metronidazole). Performance and detection bias cannot be ruled out due to the open-label design. mITT population proportions were comparable in both arms and populations were balanced reducing the likelihood of attrition bias. Results of the study have not been published yet and

no results have been posted at clinitrials.gov yet but have been presented to FDA for an Advisory Committee Meeting and made publicly available in the briefing book. Furthermore, the manufacturer provided the study protocol, SAP, results summary and in the absence of available CSR at the date of EUnetHTA submission, the manufacturer also provided study synopsis and all the documentation submitted to EMA, for this assessment thus possibility of selective outcome reporting is regarded as low.

In summary, the CREDIBLE-CR study is a small open-label, randomized, multinational, parallel-group, Phase 3 clinical trial designed as descriptive study. Through its open-label design, small number of patients and non-inferential design, the study shows unclear internal validity and high risk of bias at study level.

Table 96: Risk of bias on study level – Randomized trials with cefiderocol

Blinding

Staff

Study

Adequate generation of of generation Adequate sequence randomization allocation Adequate concealment Patient Treating individual of Reporting of independent outcomes results of bias aspects No other on study of bias Risk level unclear> unclear> unclear> unclear> unclear>

RCTs

APEKS NP yes yes yes yes yes yes low

APEKS cUTI yes yes yes yes yes yes low

Descriptive Trial

+ CREDIBLE CR yes no no no yes* no high *Results of the study have not been published yet and no results have been posted at clicnitrials.gov yet. The manufacturer provided the study protocol, SAP, CSR and study synopsis for this assessment thus possibility of selective outcome reporting is

regarded as low. +Several unbalances detected after study conclusion

5.7.2 Discussion

Unlike other therapeutic areas, the evaluation of an antimicrobial relies on the combined consideration of in vitro, PK/PD and clinical data. This is because of the primary importance of confirming pathogen susceptibility, and theoretically, if the pathogen is susceptible to the antimicrobial and it has adequate exposure in the infection site, the antibacterial therapy should be effective. The main evidence supporting efficacy of cefiderocol against a wide range of Gram-negative, aerobic pathogens thus comes from several large in-vitro surveillance studies, which were further confirmed by independent national studies in five European countries (Germany, Italy, Greece, Switzerland, UK). PK/PD studies showed that cefiderocol could reach target tissues in adequate concentrations at the recommended dosing regimen. Clinical trials served to confirm efficacy predicted based on the in vitro and PK/PD results.

In vitro testing was performed in iron-depleted broth, a standardized methodology that has been independently validated and approved. In vitro testing results are critical for clinical decision making, and the low MIC values reported from the studies together with the favourable PK/PD data indicate that cefiderocol is likely to will demonstrate clinical activity against the target Gram-negative, aerobic pathogens regardless of the infection site.

A clinical study in healthy volunteers [8] showed that the penetration ratio of cefiderocol into ELF was comparable with that of ceftazidime in critically ill patients (0.229 based on free plasma using a protein unbound fraction of 0.9).

The fraction of time during the dosing interval where free concentration exceeded the MIC (fT>MIC) for a PD target was reported to be 75%. PK/PD modelling confirmed that with probabilities of 97% in plasma and 88% in ELF free cefiderocol concentration of 4 mg/L could be achieved using the recommended dosing regimen. Outcomes of the APEKS-NP trial, which focused on pneumonia, lent further support to cefiderocol’s adequate penetration into lung tissues.

In general, clinical trials can only provide very limited evidence regarding the efficacy of new antibiotics in a real-world population of patients with suspected MDR/CR-resistant pathogens, because trials must focus on pathogens for which the tested agents are effective; otherwise, they would be un-ethical. Because trials thus focus on pathogens that fall within the in vitro spectrum of the tested treatments and comparators, it is difficult to conduct network-meta- analyses based on these trials. Since each trial excludes patients for which a poor outcome for the candidate treatment is expected, meta-analysis will only provide answers about the

8 https://academic.oup.com/jac/article/74/7/1971/5435733

efficacy of the treatment in patients with susceptible pathogens, but it cannot tell which treatment would have the best chances of success in an overall, un-tested population. Such results can only be obtained from in vitro surveillance studies. This implies that the clinical trial programs for antibiotics have an important role to confirm clinical efficacy, but a limited role in providing comparative evidence.

Formal risk-of-bias assessments performed for this dossier showed that the internal validity of the clinical trials differed in the two target populations:

 In the population with infections that were suspected to be MDR/CR the randomized and double-blinded APEKS trials provided strong evidence for non- inferiority of cefiderocol compared to respective treatment options when the pathogens are susceptible to both cefiderocol and comparator, and also confirmed the potential benefit of a wider pathogen Gram-negative coverage vs comparators.

o The non-inferiority designs were necessary due to the fact that it would be unethical to withhold effective treatments from the comparator group.

o The results are not only relevant for today’s use of antibiotics in the clinic, but also for a future in which resistances are projected to increase further and there will be many patients who would need new, effective treatment options, such as cefiderocol. In such a future scenario, cefiderocol treatment would expected to be superior compared to treatment with ineffective agents due to pan- resistant pathogens.

 The clinical results in the confirmed carbapenem-resistant populations show lower levels of internal validity, while the compassionate use program illustrates the relevance in the current treatment landscape.

o Patients with carbapenem non-susceptible pathogens in the APEKS-NP study were treated as effectively with cefiderocol as with higher-dosage meropenem. Due to the increase in dosage, meropenem maintained its efficacy in this group, leading to similar treatment results in both groups. Numerical evidence showed that cefiderocol maintained activity in high MIC (>16) to meropenem-non- susceptible infections.

o The CREDIBLE-CR trial was a small, randomised, non-blinded, non-inferential, exploratory descriptive study to start to gain experience in patients with confirmed CR infections, severe often end-stage comorbidities, often after failing multiple lines of therapy (i.e., salvage therapy context). No stratification was made for pathogen or presence of terminal disease (e.g., disseminated

cancers, end-stage organ failure). Given the small number of patients included in the trial, there are differences in the baseline characteristics of the treatment arms (older, more severe renal impairment in cefiderocol arm).

o The 74 compassionate use studies included open-label treatment of patients who had received other lines of treatment. The requests for cefiderocol treatment indicate the urgent clinical need for additional treatment options. Initial results from these cases confirm that cefiderocol shows clinical activity in severely ill patients with very limited options.

2. Provide a brief statement of the relevance of the evidence base to the scope of the assessment.

Overall, the level of evidence supporting cefiderocol treatment is higher in the suspected resistant population than in the confirmed CR populations, due to the robustness of the APEKS studies ensured through blinding, randomization, large number of patients, and adequate control group. The systematic evaluations show high internal validity with low risk of bias at study level and thus stronger confirmatory clinical results from the APEKS trials.

Population and Comparators: The presented results present the most relevant evidence base for the assessment of cefiderocol. The most relevant comparators in line with current regulatory recommendations were considered in in vitro studies and appropriate comparators reflected in the comparator arms of the clinical trials. As with all antibiotic therapies for patients with Gram-negative aerobic infections suspected to be MDR/CR/difficult-to-treat or confirmed CR-resistant, the patient populations included in the trials had a high unmet medical need.

The trial populations in APEKS (cUTI and NP) trials and CREDIBLE-CR study represented patients eligible for treatment also observed in clinical practice. Complicated urinary tract infections, nosocomial pneumonia and sepsis are the most common infections observed with Gram-negative aerobic MDR/CR/difficult-to-treat pathogens.

In line with ethical standards, most effective comparators (IMP/CS and HD meropenem) were chosen for the APEKS studies. Following good stewardship practice and at request of EMA BAT was decided to be appropriate to be administered in CREDIBLE-CR study.

Outcomes and timing: Standard outcomes for antibiotic treatment such as clinical and microbiological outcomes as well as composite clinical and microbiological endpoints and microbiological and clinical response per-pathogen and per-patient at different time points (early assessment, end of treatment (EOT), test of cure (TOC), follow-up (FUP)) were

collected in the respective trials. Clinical endpoints were in line with site of infections and severity of disease:  For cUTI the primary efficacy endpoint was the composite of clinical response and microbiological response at the test of cure (TOC).

 For nosocomial pneumonia (HAP/VAP/CAP) the primary outcome was all-cause mortality at day 14.

 For HAP/VAP/CAP and bloodstream infections/sepsis in CREDIBLE CR study primary endpoint was clinical cure at TOC, for cUTI it was microbiological outcomes at TOC.

All of these endpoints are main outcomes routinely assessed for antimicrobial studies according to current regulatory standards. All endpoints considered in the trials adequately measure relevant outcomes and follow established practice. Quality of life could not be assessed for the stated reasons.

A full clinical assessment of cefiderocol’s value needs to consider several important pieces of contextual information:

 Delays in appropriate antibiotic therapy lead to worse clinical outcomes. This means that an additional treatment that can target pathogens with a high unmet need can lead to more effective early treatment and improved clinical outcomes.

 Resistance rates are increasing. A dramatic slump in the development of new antibiotic treatments in the past two decades lead to lack of treatment options for current and future resistances.

o New treatments that show non-inferiority with all available treatments can turn out to become life-saving last-resort options in the future, when more and more pathogens have become resistant to the existing options.

o In addition to the static assessment of the current treatment landscape, a dynamic assessment that includes future trends is necessary to fully understand the current and future benefit of new antibiotic treatments.

Overall, the evidence provided in this dossier supports the clinical benefit of cefiderocol as an additional treatment option for patients with Gram-negative, aerobic infections with limited treatment options. As is true for all antibiotics, clinical use of cefiderocol will be based on the

integration of in vitro susceptibility data, hospital-wide antibiograms, monitoring resistance trends, and individual patient needs.

Within the expected pathogen based indication, it is proposed that cefiderocol offers most value in two clinical scenarios:

 Hospitalised patients with suspected MDR/CR infection who are at risk of rapid deterioration and require antibiotic treatment that provides full cover against carbapenem-resistant pathogens in the period while pathogen susceptibility is being confirmed.

 Hospitalised patients with a confirmed MDR/CR infection where existing treatment options are inappropriate because of pathogen susceptibility, contraindications or poor tolerability

Given the growing threat from MDR/CR infection and the limitations of currently available treatment options both populations have a high unmet medical need. Advances in fast diagnostics will allow clinicians to make decisions about effective treatment options earlier and earlier. The recent advent of several new treatment options, together with such early diagnostics holds promise to improve outcomes for critically ill patients and slow down the further spread of resistant pathogens. Economic evaluations of antibiotics based on these clinical data will need to take the full spectrum of benefits into account (e.g., enablement of chemotherapy, high-risk surgeries).

REFERENCES

1. O’Neill J. TACKLING DRUG-RESISTANT INFECTIONS GLOBALLY: FINAL REPORT AND RECOMMENDATIONS. 2016. 2. World Economic Forum. The Global Risks Report 2018 13th Edition. 2018. 3. Review on Antimicrobial Resistance. Tackling Drug-Resistant Infections Globally: Final Report and Recommendations. 2016. 4. Cassini A, Hogberg LD, Plachouras D, Quattrocchi A, Hoxha A, Simonsen GS, et al. Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysis. The Lancet Infectious diseases. 2018. 5. World Health Organization. Global priority list of antibiotic-resistant bacteria to guide research, discovery, and development of new antibiotics. 2017. 6. World Health Organization. WHO publishes list of bacteria for which new antibiotics are urgently needed 2017 [Available from: http://www.who.int/mediacentre/news/releases/2017/bacteria-antibiotics-needed/en/ 7. World Health Organization. Antibacterial agents in clinical development - An analysis of the antibacterial clinical development pipeline, including tuberculosis. 2017. 8. European Centre for Disease Prevention and Control. Antimicrobial resistance surveillance 2015.; 2015. 9. European Centre for Disease Prevention and Control. Surveillance atlas of infectious diseases 2015 [Available from: http://atlas.ecdc.europa.eu/public/index.aspx?Instance=GeneralAtlas. 10. (ECDC) ECfDPaC. Point prevalence survey database (HAI-Net) 2012 [Available from: https://www.ecdc.europa.eu/en/healthcare-associated-infections-acute-care-hospitals/surveillance- disease-data/database. 11. (ECDC) ECfDPaC. Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals 2011-2012. 2013. 12. Weinstein RA, Gaynes R, Edwards JR, System NNIS. Overview of nosocomial infections caused by gram-negative bacilli. Clinical infectious diseases. 2005;41(6):848-54. 13. Kaye KS. Antimicrobial de‐escalation strategies in hospitalized patients with pneumonia, intra‐abdominal infections, and bacteremia. Journal of hospital medicine. 2012;7(S1):S13-S21. 14. Bauer K, Perez K, Forrest G, Goff D. Review of Rapid Diagnostic Tests Used by Antimicrobial Stewardship Programs. CID. 2014;59 (suppl 3):S134-S45. 15. Dantas RC, Ferreira ML, Gontijo-Filho PP, Ribas RM. Pseudomonas aeruginosa bacteraemia: independent risk factors for mortality and impact of resistance on outcome. J Med Microbiol. 2014;63(Pt 12):1679-87. 16. Kang CI, Kim SH, Kim HB, Park SW, Choe YJ, Oh MD, et al. Pseudomonas aeruginosa bacteremia: risk factors for mortality and influence of delayed receipt of effective antimicrobial therapy on clinical outcome. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2003;37(6):745-51. 17. Lodise TP, Jr., Patel N, Kwa A, Graves J, Furuno JP, Graffunder E, et al. Predictors of 30-day mortality among patients with Pseudomonas aeruginosa bloodstream infections: impact of delayed appropriate antibiotic selection. Antimicrobial agents and chemotherapy. 2007;51(10):3510-5. 18. Micek ST, Lloyd AE, Ritchie DJ, Reichley RM, Fraser VJ, Kollef MH. Pseudomonas aeruginosa bloodstream infection: importance of appropriate initial antimicrobial treatment. Antimicrobial agents and chemotherapy. 2005;49(4):1306-11. 19. Osih RB, McGregor JC, Rich SE, Moore AC, Furuno JP, Perencevich EN, et al. Impact of empiric antibiotic therapy on outcomes in patients with Pseudomonas aeruginosa bacteremia. Antimicrobial agents and chemotherapy. 2007;51(3):839-44.

20. Shorr AF, Zilberberg MD, Micek ST, Kollef MH. Predictors of hospital mortality among septic ICU patients with Acinetobacter spp. bacteremia: a cohort study. BMC infectious diseases. 2014;14:572. 21. Zavascki AP, Barth AL, Goncalves AL, Moro AL, Fernandes JF, Martins AF, et al. The influence of metallo-beta-lactamase production on mortality in nosocomial Pseudomonas aeruginosa infections. The Journal of antimicrobial chemotherapy. 2006;58(2):387-92. 22. Papst L, Beovic B, Pulcini C, Durante-Mangoni E, Rodriguez-Bano J, Kaye KS, et al. Antibiotic treatment of infections caused by carbapenem-resistant Gram-negative bacilli: an international ESCMID cross-sectional survey among infectious diseases specialists practicing in large hospitals. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases. 2018;24(10):1070-6. 23. Sartelli M, Catena F, Abu-Zidan FM, Ansaloni L, Biffl WL, Boermeester MA, et al. Management of intra-abdominal infections: recommendations by the WSES 2016 consensus conference. World Journal of Emergency Surgery. 2017;12(1):22. 24. Mazuski JE, Tessier JM, May AK, Sawyer RG, Nadler EP, Rosengart MR, et al. The Surgical Infection Society Revised Guidelines on the Management of Intra-Abdominal Infection. Surgical infections. 2017;18(1):1-76. 25. Torres A, Niederman MS, Chastre J, Ewig S, Fernandez-Vandellos P, Hanberger H, et al. International ERS/ESICM/ESCMID/ALAT guidelines for the management of hospital-acquired pneumonia and ventilator-associated pneumonia: Guidelines for the management of hospital- acquired pneumonia (HAP)/ventilator-associated pneumonia (VAP) of the European Respiratory Society (ERS), European Society of Intensive Care Medicine (ESICM), European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and Asociacion Latinoamericana del Torax (ALAT). The European respiratory journal. 2017;50(3). 26. Kalil AC, Metersky ML, Klompas M, Muscedere J, Sweeney DA, Palmer LB, et al. Management of Adults With Hospital-acquired and Ventilator-associated Pneumonia: 2016 Clinical Practice Guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2016;63(5):e61-e111. 27. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive care medicine. 2017;43(3):304-77. 28. Papp-Wallace KM, Endimiani A, Taracila MA, Bonomo RA. Carbapenems: past, present, and future. Antimicrobial agents and chemotherapy. 2011;55(11):4943-60. 29. Hackel MA, Tsuji M, Yamano Y, Echols R, Karlowsky JA, Sahm DF. In vitro activity of the siderophore cephalosporin, cefiderocol, against a recent collection of clinically relevant Gram- negative bacilli from North America and Europe, including carbapenem-nonsusceptible isolates (SIDERO-WT-2014 Study). Antimicrobial agents and chemotherapy. 2017;61(9):e00093-17. 30. Hackel MA, Tsuji M, Yamano Y, Echols R, Karlowsky JA, Sahm DF. In vitro activity of the siderophore cephalosporin, cefiderocol, against carbapenem-nonsusceptible and multidrug-resistant isolates of Gram-negative bacilli collected worldwide in 2014 to 2016. Antimicrobial agents and chemotherapy. 2018;62(2):e01968-17. 31. Communications AH. A systematic review of rates of nephrotoxicity and neurotoxicity in patients treated with polymyxins [Data on file]. 2019. 32. Falagas ME, Kasiakou SK, Kofteridis DP, Roditakis G, Samonis G. Effectiveness and nephrotoxicity of intravenous colistin for treatment of patients with infections due to polymyxin- only-susceptible (POS) gram-negative bacteria. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2006;25. 33. Gales AC, Jones RN, Sader HS. Contemporary activity of colistin and polymyxin B against a worldwide collection of Gram-negative pathogens: results from the SENTRY Antimicrobial Surveillance Program (2006-09). The Journal of antimicrobial chemotherapy. 2011;66(9):2070-4.

34. Zusman O, Altunin S, Koppel F, Dishon Benattar Y, Gedik H, Paul M. Polymyxin monotherapy or in combination against carbapenem-resistant bacteria: systematic review and meta-analysis. Journal of Antimicrobial Chemotherapy. 2016;72(1):29-39. 35. Vazquez JA, Gonzalez Patzan LD, Stricklin D, Duttaroy DD, Kreidly Z, Lipka J, et al. Efficacy and safety of ceftazidime-avibactam versus imipenem-cilastatin in the treatment of complicated urinary tract infections, including acute pyelonephritis, in hospitalized adults: results of a prospective, investigator-blinded, randomized study. Curr Med Res Opin. 2012;28(12):1921-31. 36. Wagenlehner FM, Sobel JD, Newell P, Armstrong J, Huang X, Stone GG, et al. Ceftazidime- avibactam Versus Doripenem for the Treatment of Complicated Urinary Tract Infections, Including Acute Pyelonephritis: RECAPTURE, a Phase 3 Randomized Trial Program. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2016;63(6):754-62. 37. Eckmann C, Solomkin J. Ceftolozane/tazobactam for the treatment of complicated intra- abdominal infections. Expert Opin Pharmacother. 2015;16. 38. Giancola SE, Mahoney MV, Bias TE, Hirsch EB. Critical evaluation of ceftolozane-tazobactam for complicated urinary tract and intra-abdominal infections. Ther Clin Risk Manag. 2016;12:787-97. 39. Plant AJ, Dunn A, Porter RJ. Ceftolozane-tazobactam resistance induced in vivo during the treatment of MDR Pseudomonas aeruginosa pneumonia. Expert Review of Anti-infective Therapy. 2018;16(5):367-8. 40. Barnes MD, Winkler ML, Taracila MA, Page MG, Desarbre E, Kreiswirth BN, et al. Klebsiella pneumoniae carbapenemase-2 (KPC-2), substitutions at Ambler position Asp179, and resistance to ceftazidime-avibactam: unique antibiotic-resistant phenotypes emerge from β-lactamase protein engineering. MBio. 2017;8(5):e00528-17. 41. Karaiskos I, Lagou S, Pontikis K, Rapti V, Poulakou G. The "Old" and the "New" Antibiotics for MDR Gram-Negative Pathogens: For Whom, When, and How. Front Public Health. 2019;7:151-. 42. Haidar G, Philips NJ, Shields RK, Snyder D, Cheng S, Potoski BA, et al. Ceftolozane- Tazobactam for the Treatment of Multidrug-Resistant Pseudomonas aeruginosa Infections: Clinical Effectiveness and Evolution of Resistance. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2017;65(1):110-20. 43. Katchanov J, Asar L, Klupp E-M, Both A, Rothe C, König C, et al. Carbapenem-resistant Gram- negative pathogens in a German university medical center: Prevalence, clinical implications and the role of novel β-lactam/β-lactamase inhibitor combinations. PloS one. 2018;13(4):e0195757-e. 44. Zhanel GG, Golden AR, Zelenitsky S, Wiebe K, Lawrence CK, Adam HJ, et al. Cefiderocol: A Siderophore Cephalosporin with Activity Against Carbapenem-Resistant and Multidrug-Resistant Gram-Negative Bacilli. Drugs. 2019;79(3):271-89. 45. Karlowsky JA, Hackel MA, Tsuji M, Yamano Y, Echols R, Sahm DF. In Vitro Activity of Cefiderocol, a Siderophore Cephalosporin, Against Gram-Negative Bacilli Isolated by Clinical Laboratories in North America and Europe in 2015-2016: SIDERO-WT-2015. International journal of antimicrobial agents. 2019;53(4):456-66. 46. Kazmierczak KM, Tsuji M, Wise MG, Hackel M, Yamano Y, Echols R, et al. In vitro activity of cefiderocol, a siderophore cephalosporin, against a recent collection of clinically relevant carbapenem-non-susceptible Gram-negative bacilli, including serine carbapenemase-and metallo-β- lactamase-producing isolates (SIDERO-WT-2014 Study). International journal of antimicrobial agents. 2019;53(2):177-84. 47. Longshaw C, Tsuji M, Hackel M, Sahm D, Yamano Y, editors. In VitroActivity of Cefiderocol (CFDC), a Novel Siderophore Cephalosporin, Against Difficult-to-Treat Resistant (DTR) Gram-negative Bacterial Pathogens from the Multinational Sentinel Surveillance Study, SIDERO-WT (2014–2017). IDWeek2019; 2019; Washington, DC, USA. 48. Sato T, Tsuji M, Kazmierczak K, Hackel M, R E, Yamano Y, et al., editors. Cefiderocol Susceptibility Against Molecularly Characterized Carbapenemase-producing Gram-negative Bacteria in North America and Europe Between 2014 and 2017: SIDERO-WT-2014–2016 Studies. IDWeek2019; 2019; Washington, DC, USA.

49. Tsuji M, Hackel M, Echols R, Yamano Y, Sahm D, editors. In Vitro Antibacterial Activity of Cefiderocol Against Gram-negative Clinical Strains Collected in North America and Europe: SIDERO-WT-2016. ASM Microbe 2019; 2019; San Francisco, CA, USA. 50. Shionogi. Day 180 Response to Questions-Labelling - EMEA/H/C/004829. 2020. 51. Portsmouth S, van Veenhuyzen D, Echols R, Machida M, Ferreira JC, Ariyasu M, et al. Cefiderocol versus imipenem-cilastatin for the treatment of complicated urinary tract infections caused by Gram-negative uropathogens: a phase 2, randomised, double-blind, non-inferiority trial. The Lancet Infectious diseases. 2018. 52. Rex JH, Talbot GH, Goldberger MJ, Eisenstein BI, Echols RM, Tomayko JF, et al. Progress in the Fight Against Multidrug-Resistant Bacteria 2005-2016: Modern Noninferiority Trial Designs Enable Antibiotic Development in Advance of Epidemic Bacterial Resistance. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2017;65(1):141-6. 53. Shionogi. Clinical study protocol: 1409R2121 (APEK-cUTI study) [Data on file]. 2015. 54. Sato T, Yamawaki K. Cefiderocol: discovery, chemistry, and in vivo profiles of a novel siderophore cephalosporin. Clinical Infectious Diseases. 2019;69(Supplement_7):S538-S43. 55. Curtis NA, Eisenstadt RL, East SJ, Cornford RJ, Walker LA, White AJ. Iron-regulated outer membrane proteins of Escherichia coli K-12 and mechanism of action of catechol-substituted cephalosporins. Antimicrobial agents and chemotherapy. 1988;32(12):1879-86. 56. Bullen JJ. The significance of iron in infection. Rev Infect Dis. 1981;3(6):1127-38. 57. Krewulak KD, Vogel HJ. Structural biology of bacterial iron uptake. Biochim Biophys Acta. 2008;1778(9):1781-804. 58. Page MG. The Role of Iron and Siderophores in Infection, and the Development of Siderophore Antibiotics. Clinical Infectious Diseases. 2019;69(Supplement_7):S529-S37. 59. Tillotson GS. Trojan Horse Antibiotics-A Novel Way to Circumvent Gram-Negative Bacterial Resistance? Infectious diseases. 2016;9:45-52. 60. Buchanan SK. Bacterial metal detectors. Mol Microbiol. 2005;58(5):1205-9. 61. Ito A, Ishibashi N, Kitanishi K, Osaki H, Sato T, Tsuji M, et al. Contribution of chelating ability with iron (III) and the utilization of iron transporters through the outer membrane to the in vitro activity of cefiderocol against P. aeruginosa. Microbe 20172017. 62. Ito A, Toba S, Nishikawa T, Kohira N, Sato T, Tsuji M, et al. Contribution of Active Iron Transporters and Binding Ability to Penicillin Binding Proteins od Cefiderocol (S-649266) to Its Antibacterial/Bactericidal Acitvity against Klebsiellla pneumoniae and Escherichia coli. Microbe 20172017. 63. Thalhammer F. Infektiologie &gastroenterologie-Hepatologie. 2018. 64. Theuretzbacher U, Gottwalt S, Beyer P, Butler M, Czaplewski L, Lienhardt C, et al. Analysis of the clinical antibacterial and antituberculosis pipeline. The Lancet Infectious Diseases. 2019;19(2):e40-e50. 65. WHO. WHO publishes list of bacteria for which new antibiotics are urgently needed 2017 [Available from: https://www.who.int/news-room/detail/27-02-2017-who-publishes-list-of-bacteria- for-which-new-antibiotics-are-urgently-needed. 66. Davis CP. Normal flora. Medical Microbiology 4th edition: University of Texas Medical Branch at Galveston; 1996. 67. Peterson JW. Bacterial pathogenesis. Medical Microbiology 4th edition: University of Texas Medical Branch at Galveston; 1996. 68. BioTrends Research Group. Treatment Trends: Gram-negative Infections 2015 (US). 2015. 69. Arlington Medical Resources. Hospital antibiotic market guide. 2014. 70. Shionogi. ICD-codes for nosocomial Gram-negative infections. Appendix A1 of the Dossier. 71. Mahon L, Manuselis. Textbook of Diagnostic Microbiology - E-Book. Saunders2014.

72. Exner M, Bhattacharya S, Christiansen B, Gebel J, Goroncy-Bermes P, Hartemann P, et al. Antibiotic resistance: What is so special about multidrug-resistant Gram-negative bacteria? GMS Hygiene and Infection Control. 2017;12:1-24. 73. The Ohio State University. Gram negative organisms flowchart 2017 [Available from: HTTP://RX.OSUMC.EDU/ASP2/MICROBIOLOGY/GRAMNEGORGANISMS.PDF. 74. Adeolu M, Alnajar S, Naushad S, Gupta RS. Genome-based phylogeny and taxonomy of the ‘Enterobacteriales’: proposal for Enterobacterales ord. nov. divided into the families Enterobacteriaceae, Erwiniaceae fam. nov., Pectobacteriaceae fam. nov., Yersiniaceae fam. nov., Hafniaceae fam. nov., Morganellaceae fam. nov., and Budviciaceae fam. nov. International journal of systematic and evolutionary microbiology. 2016;66(12):5575-99. 75. Munita JM, Arias CA. Mechanisms of Antibiotic Resistance. Microbiol Spectr. 2016;4(2). 76. Centers for Disease Control and Prevention. Antibiotic resistance threats in the United States. 2013. 77. Shionogi. Infectious disease education (internal training) [Data on file]. 2017. 78. World Health Organization. Antimicrobial resistance - global report on surveillance. 2014. 79. European Centre for Disease Prevention and Control. Surveillance of antimicrobial resistance in Europe 2017.; 2018 November 2018. 80. Prevention C-CfDCa. ANTIBIOTIC RESISTANCE THREATS in the United States, 2013. 2013. 81. ECDC. Infographic: Antibiotic resistance – an increasing threat to human health 2018 [28.01.2020]. Available from: https://www.ecdc.europa.eu/en/publications-data/infographic- antibiotic-resistance-increasing-threat-human-health. 82. Weterings V, Zhou K, Rossen JW, van Stenis D, Thewessen E, Kluytmans J, et al. An outbreak of colistin-resistant Klebsiella pneumoniae carbapenemase-producing Klebsiella pneumoniae in the Netherlands (July to December 2013), with inter-institutional spread. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2015;34(8):1647-55. 83. Poirel L, Dobias J, Dénervaud V, Nordmann P. Activity of the novel siderophore cephalosporin cefiderocol against multidrug-resistant Gram-negative pathogens. ECCMID; Vienne2017. 84. Schreckenberger PC, Daneshvar, M.I., Weyant, R.S., Hollis, D.G.,. Acinetobacter, Achromobacter, Chryseobacterium, Moraxella, and other nonfermentative Gram-negative rods. Manual of Clinical Microbiology. 9th Ed. ed2007. p. 770-802. 85. Holmes AH, Moore LS, Sundsfjord A, Steinbakk M, Regmi S, Karkey A, et al. Understanding the mechanisms and drivers of antimicrobial resistance. Lancet (London, England). 2016;387(10014):176-87. 86. Queenan AM, Bush K. Carbapenemases: the versatile β-lactamases. Clinical microbiology reviews. 2007;20(3):440-58. 87. Watanabe M, Iyobe S, Inoue M, Mitsuhashi S. Transferable imipenem resistance in Pseudomonas aeruginosa. Antimicrobial agents and chemotherapy. 1991;35(1):147-51. 88. Paton R, Miles RS, Hood J, Amyes SG, Miles RS, Amyes SG. ARI 1: beta-lactamase-mediated imipenem resistance in Acinetobacter baumannii. Int J Antimicrob Agents. 1993;2(2):81-7. 89. Scaife W, Young HK, Paton RH, Amyes SG. Transferable imipenem-resistance in Acinetobacter species from a clinical source. The Journal of antimicrobial chemotherapy. 1995;36(3):585-6. 90. Yano H, Kuga A, Okamoto R, Kitasato H, Kobayashi T, Inoue M. Plasmid-encoded metallobeta-lactamase (IMP-6) conferring resistance to carbapenems, especially meropenem. Antimicrobial agents and chemotherapy. 2001;45(5):1343-8. 91. Eiamphungporn W, Schaduangrat N, Malik AA, Nantasenamat C. Tackling the antibiotic resistance caused by class A β-Lactamases through the use of β-Lactamase inhibitory protein. International journal of molecular sciences. 2018;19(8):2222.

92. Ruppé É, Woerther P-L, Barbier F. Mechanisms of antimicrobial resistance in Gram-negative bacilli. Annals of intensive care. 2015;5(1):21. 93. Madigan MT, Clark DP, Stahl D, Martinko JM. Brock biology of microorganisms 13th edition: Benjamin Cummings; 2010. 94. Bush K, Jacoby GA. Updated Functional Classification of β-Lactamases. Antimicrobial agents and chemotherapy. 2010;54(3):969-76. 95. Drawz SM, Bonomo RA. Three decades of β-lactamase inhibitors. Clinical microbiology reviews. 2010;23(1):160-201. 96. Walsh TR, Toleman MA, Poirel L, Nordmann P. Metallo-β-lactamases: the quiet before the storm? Clinical microbiology reviews. 2005;18(2):306-25. 97. Walther-Rasmussen J, Høiby N. OXA-type carbapenemases. Journal of Antimicrobial Chemotherapy. 2006;57(3):373-83. 98. Sanchez MB. Antibiotic resistance in the opportunistic pathogen Stenotrophomonas maltophilia. Front Microbiol. 2015;6:658. 99. Fernández L, Hancock RE. Adaptive and mutational resistance: role of porins and efflux pumps in drug resistance. Clinical microbiology reviews. 2012;25(4):661-81. 100. Blair JM, Webber MA, Baylay AJ, Ogbolu DO, Piddock LJ. Molecular mechanisms of antibiotic resistance. Nature reviews microbiology. 2015;13(1):42-51. 101. Gutiérrez O, Juan C, Cercenado E, Navarro F, Bouza E, Coll P, et al. Molecular epidemiology and mechanisms of carbapenem resistance in Pseudomonas aeruginosa isolates from Spanish hospitals. Antimicrobial agents and chemotherapy. 2007;51(12):4329-35. 102. Poirel L, Jayol A, Nordmann P. Polymyxins: Antibacterial Activity, Susceptibility Testing, and Resistance Mechanisms Encoded by Plasmids or Chromosomes. Clin Microbiol Rev. 2017;30(2):557- 96. 103. Grundmann H, Glasner C, Albiger B, Aanensen DM, Tomlinson CT, Andrasevic AT, et al. Occurrence of carbapenemase-producing Klebsiella pneumoniae and Escherichia coli in the European survey of carbapenemase-producing Enterobacteriaceae (EuSCAPE): a prospective, multinational study. The Lancet Infectious diseases. 2017;17(2):153-63. 104. Choi JJ, McCarthy MW. Cefiderocol: a novel siderophore cephalosporin. Expert Opin Investig Drugs. 2018;27(2):193-7. 105. Meletis G. Carbapenem resistance: overview of the problem and future perspectives. Ther Adv Infect Dis. 2016;3(1):15-21. 106. Suetens C, Latour K, Kärki T, Ricchizzi E, Kinross P, Moro ML, et al. Prevalence of healthcare- associated infections, estimated incidence and composite antimicrobial resistance index in acute care hospitals and long-term care facilities: results from two European point prevalence surveys, 2016 to 2017. Eurosurveillance. 2018;23(46). 107. U.S. Food and Drug Administration. Guidance for industry hospital-acquired bacterial pneumonia and ventilator-associtaed bacterial pneumonia: developing drugs for treatment. 2014. 108. (ECDC) ECfDPaC. Incidence and attributable mortality of healthcare-associated infections in intensive care units in Europe, 2008-2012. 2018. 109. Nicolle LE. Complicated urinary tract infection in adults. Can J Infect Dis Med Microbiol. 2005;16(6):349-60. 110. Wagenlehner FM, Lichtenstern C, Rolfes C, Mayer K, Uhle F, Weidner W, et al. Diagnosis and management for urosepsis. Int J Urol. 2013;20(10):963-70. 111. Decision Resources Group. Market access landscape report prepared for Shionogi. 2017. 112. Cai B, Echols, R., Morgan, G., Van Veenhuyzen, D., Ariyasu, M., Sawada, T., Murray, J., Den Nagata, T., Gupta, V.,. National Prevalence and Regional Variation of Carbapenem-resistant Gram- negative Bacteria in the Ambulatory and Acute Care Settings in the United States in 2016. ID Week 2017; Abstract #650892017. 113. Solomkin JS, Mazuski JE, Bradley JS, Rodvold KA, Goldstein EJ, Baron EJ, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: guidelines by the

Surgical Infection Society and the Infectious Diseases Society of America. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2010;50(2):133-64. 114. U.S. Food and Drug Administration. Complicated intra-abdominal infections: developing drugs for treatment. 2015. 115. Crusio R, Rao S, Changawala N, Paul V, Tiu C, van Ginkel J, et al. Epidemiology and outcome of infections with carbapenem-resistant Gram-negative bacteria treated with polymyxin B-based combination therapy. Scandinavian journal of infectious diseases. 2014;46(1):1-8. 116. Pouch SM, Kubin CJ, Satlin MJ, Tsapepas DS, Lee JR, Dube G, et al. Epidemiology and outcomes of carbapenem-resistant Klebsiella pneumoniae bacteriuria in kidney transplant recipients. Transpl Infect Dis. 2015;17(6):800-9. 117. Satlin MJ, Jenkins SG, Walsh TJ. The global challenge of carbapenem-resistant Enterobacteriaceae in transplant recipients and patients with hematologic malignancies. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2014;58(9):1274-83. 118. Rogers BA, Aminzadeh Z, Hayashi Y, Paterson DL. Country-to-country transfer of patients and the risk of multi-resistant bacterial infection. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2011;53(1):49-56. 119. Epelboin L, Robert J, Tsyrina-Kouyoumdjian E, Laouira S, Meyssonnier V, Caumes E. High Rate of Multidrug-Resistant Gram-Negative Bacilli Carriage and Infection in Hospitalized Returning Travelers: A Cross-Sectional Cohort Study. J Travel Med. 2015;22(5):292-9. 120. Cornaglia G, Giamarellou H, Rossolini GM. Metallo-beta-lactamases: a last frontier for beta- lactams? The Lancet Infectious diseases. 2011;11(5):381-93. 121. Gupta N, Limbago BM, Patel JB, Kallen AJ. Carbapenem-resistant Enterobacteriaceae: epidemiology and prevention. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2011;53(1):60-7. 122. Hawkey PM, Warren RE, Livermore DM, McNulty CAM, Enoch DA, Otter JA, et al. Treatment of infections caused by multidrug-resistant Gram-negative bacteria: report of the British Society for Antimicrobial Chemotherapy/Healthcare Infection Society/British Infection Association Joint Working Party. Journal of Antimicrobial Chemotherapy. 2018;73(suppl_3):iii2-iii78. 123. Shionogi. Systematic Literature Review of Patient Risk Factors for Carbapenem Resistant Infection. Appendix B1 of the dossier. 124. Shionogi. Delay of treatment systematic literature review. Report. Appendix B2 of the dossier. 125. Shionogi. Delay of treatment systematic literature review. Appendices. Appendix B3 of the dossier. . 126. (ECDC) ECfDPaC. Surveillance of antimicrobial resistance in Europe 2018. 2019. 127. Nordmann P, Naas T, Poirel L. Global Spread of Carbapenemase-producing Enterobacteriaceae. Emerging infectious diseases. 2011;17:1791-8. 128. Nordmann P, Poirel L. Epidemiology and Diagnostics of Carbapenem Resistance in Gram- negative Bacteria. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2019;69(Supplement_7):S521-S8. 129. Mendes RE, Mendoza M, Banga Singh KK, Castanheira M, Bell JM, Turnidge JD, et al. Regional Resistance Surveillance Program Results for 12 Asia-Pacific Nations (2011). Antimicrobial agents and chemotherapy. 2013;57(11):5721. 130. Kiratisin P, Chongthaleong A, Tan TY, Lagamayo E, Roberts S, Garcia J, et al. Comparative in vitro activity of carbapenems against major Gram-negative pathogens: results of Asia-Pacific surveillance from the COMPACT II study. International Journal of Antimicrobial Agents. 2012;39(4):311-6. 131. Shionogi. Additional evidence to support access of cefiderocol - Figure adapted from SIDERO-CR study [Data on file]. 2019.

132. England PH. English Surveillance Programme for Antimicrobial Utilisation and Resistance (ESPAUR) (2018-2019). 2019. 133. Raman G, Avendano E, Berger S, Menon V. Appropriate initial antibiotic therapy in hospitalized patients with gram-negative infections: systematic review and meta-analysis. BMC infectious diseases. 2015;15(1):395. 134. ReAct Group. Burden of resistance to multi-resistant Gram-negative bacilli (MRGN). 2008. 135. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. Am J Respir Crit Care Med. 2016;193(3):259-72. 136. Sabih A, Leslie SW. Complicated Urinary Tract Infections. StatPearls [Internet]: StatPearls Publishing; 2019. 137. U.S. National Library of Medicine. MedlinePlus - Medical Dictionary 2017 [Available from: https://medlineplus.gov/mplusdictionary.html. 138. Hassoun A, Linden PK, Friedman B. Incidence, prevalence, and management of MRSA bacteremia across patient populations—a review of recent developments in MRSA management and treatment. Critical care. 2017;21(1):211. 139. Aloush V, Navon-Venezia S, Seigman-Igra Y, Cabili S, Carmeli Y. Multidrug-resistant Pseudomonas aeruginosa: risk factors and clinical impact. Antimicrobial agents and chemotherapy. 2006;50(1):43-8. 140. Hirakata Y, Yamaguchi T, Nakano M, Izumikawa K, Mine M, Aoki S, et al. Clinical and bacteriological characteristics of IMP-type metallo-beta-lactamase-producing Pseudomonas aeruginosa. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2003;37(1):26-32. 141. Perez F, Van Duin D. Carbapenem-resistant Enterobacteriaceae: a menace to our most vulnerable patients. Cleve Clin J Med. 2013;80(4):225-33. 142. Paul M, Daikos GL, Durante-Mangoni E, Yahav D, Carmeli Y, Benattar YD, et al. Colistin alone versus colistin plus meropenem for treatment of severe infections caused by carbapenem-resistant Gram-negative bacteria: an open-label, randomised controlled trial. The Lancet Infectious diseases. 2018;18(4):391-400. 143. Tumbarello M, De Pascale G, Trecarichi EM, De Martino S, Bello G, Maviglia R, et al. Effect of Aerosolized Colistin as Adjunctive Treatment on the Outcomes of Microbiologically Documented Ventilator-Associated Pneumonia Caused by Colistin-Only Susceptible Gram-Negative Bacteria. Chest. 2013;144(6):1768-75. 144. Petrosillo N, Giannella M, Antonelli M, Antonini M, Barsic B, Belancic L, et al. Clinical experience of colistin-glycopeptide combination in critically ill patients infected with Gram-negative bacteria. Antimicrobial agents and chemotherapy. 2014;58(2):851-8. 145. French Institute for Public Health Surveillance. Morbidity and mortality of multiresistant bacterial infections to antibiotics in France in 2012. Study Burden BMR, report - June 2015 [In French]. 2015. 146. Retamar P, Portillo MM, Lopez-Prieto MD, Rodriguez-Lopez F, de Cueto M, Garcia MV, et al. Impact of inadequate empirical therapy on the mortality of patients with bloodstream infections: a propensity score-based analysis. Antimicrobial agents and chemotherapy. 2012;56(1):472-8. 147. Muggeo A, Guillard T, Barbe C, Thierry A, Bajolet O, Vernet-Garnier V, et al. Factors associated with carriage of carbapenem-non-susceptible Enterobacteriaceae in North-Eastern France and outcomes of infected patients. The Journal of antimicrobial chemotherapy. 2017;72:1496-501. 148. Pena C, Suarez C, Gozalo M, Murillas J, Almirante B, Pomar V, et al. Prospective multicenter study of the impact of carbapenem resistance on mortality in Pseudomonas aeruginosa bloodstream infections. Antimicrobial agents and chemotherapy. 2012;56(3):1265-72. 149. Falagas ME, Tansarli GS, Karageorgopoulos DE, Vardakas KZ. Deaths attributable to carbapenem-resistant Enterobacteriaceae infections. Emerg Infect Dis. 2014;20(7):1170-5.

150. Tischendorf J, de Avila RA, Safdar N. Risk of infection following colonization with carbapenem-resistant Enterobactericeae: A systematic review. Am J Infect Control. 2016;44(5):539- 43. 151. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med. 2002;165(7):867-903. 152. Barbier F, Andremont A, Wolff M, Bouadma L. Hospital-acquired pneumonia and ventilator- associated pneumonia: recent advances in epidemiology and management. Curr Opin Pulm Med. 2013;19(3):216-28. 153. Bekaert M, Timsit JF, Vansteelandt S, Depuydt P, Vesin A, Garrouste-Orgeas M, et al. Attributable mortality of ventilator-associated pneumonia: a reappraisal using causal analysis. Am J Respir Crit Care Med. 2011;184(10):1133-9. 154. Timsit JF, Zahar JR, Chevret S. Attributable mortality of ventilator-associated pneumonia. Curr Opin Crit Care. 2011;17(5):464-71. 155. Laupland KB, Church DL. Population-based epidemiology and microbiology of community- onset bloodstream infections. Clin Microbiol Rev. 2014;27(4):647-64. 156. Laupland KB, Kirkpatrick AW, Delaney A. Polyclonal intravenous immunoglobulin for the treatment of severe sepsis and septic shock in critically ill adults: a systematic review and meta- analysis. Critical care medicine. 2007;35(12):2686-92. 157. European Centre for Disease Prevention and Control. Carbapenem-resistant Enterobacteriaceae 2016 [Available from: https://ecdc.europa.eu/sites/portal/files/media/en/publications/Publications/carbapenem- resistant-enterobacteriaceae-risk-assessment-april-2016.pdf 158. European Centre for Disease Prevention and Control E. The bacterial challenge: time to react. 2009. 159. Shionogi. Premier Healthcare Database Study. 2017. 160. Perez F, Endimiani A, Ray AJ, Decker BK, Wallace CJ, Hujer KM, et al. Carbapenem-resistant Acinetobacter baumannii and Klebsiella pneumoniae across a hospital system: impact of post-acute care facilities on dissemination. The Journal of antimicrobial chemotherapy. 2010;65(8):1807-18. 161. Kwon KT, Oh, W.S., Song, J-H., CHang, H-H., Jung, S-I., Kim, S-W., Ryu, S-Y., Heo, S-T., Jung, S.S., Rhee, J-Y., et al., . Impact of imipenem resistance on mortality in patients with Acinetobacter bacteraemia. Journal of Antimicrobial Chemotherapy. 2007;59(3):525-30. 162. Dellit TH, Owens RC, McGowan JE, Gerding DN, Weinstein RA, Burke JP, et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship. Clinical Infectious Diseases. 2007;44(2):159-77. 163. Marquet K, Liesenborgs A, Bergs J, Vleugels A, Claes N. Incidence and outcome of inappropriate in-hospital empiric antibiotics for severe infection: a systematic review and meta- analysis. Crit Care. 2015;19(1):63-. 164. Bassetti M. A systematic review on the impact of appropriate versus inappropriate initial antibiotic therapy on the outcomes of patients with severe bacterial infections [Draft, Data on file]. 2019. 165. Zilberberg MD, Nathanson BH, Sulham K, Fan W, Shorr AF. Carbapenem resistance, inappropriate empiric treatment and outcomes among patients hospitalized with Enterobacteriaceae urinary tract infection, pneumonia and sepsis. BMC infectious diseases. 2017;17(1):279-. 166. Zasowski EJ. A systematic review of the effect of delayed appropriate antibiotic treatment on the outcomes of patients with severe bacterial infections [Draft, data on file]. 2019. 167. Caliendo AM, Gilbert DN, Ginocchio CC, Hanson KE, May L, Quinn TC, et al. Better tests, better care: improved diagnostics for infectious diseases. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2013;57 Suppl 3:S139-70.

168. Gutiérrez-Gutiérrez B, Salamanca E, de Cueto M, Hsueh P-R, Viale P, Paño-Pardo JR, et al. Effect of appropriate combination therapy on mortality of patients with bloodstream infections due to carbapenemase-producing Enterobacteriaceae (INCREMENT): a retrospective cohort study. The Lancet Infectious Diseases. 2017;17(7):726-34. 169. ENGLAND P. Framework of actions to contain. Appendix A2 of the dossier. 170. Masterton RG. Antibiotic de-escalation. Crit Care Clin. 2011;27(1):149-62. 171. Garnacho-Montero J, Gutiérrez-Pizarraya A, Escoresca-Ortega A, Corcia-Palomo Y, Fernández-Delgado E, Herrera-Melero I. De-escalation of empirical therapy is associated with lower mortality in patients with severe sepsis and septic shock. Intensive care medicine. 2014;40. 172. Viasus D, Simonetti AF, Garcia-Vidal C, Niubó J, Dorca J, Carratalà J. Impact of antibiotic de- escalation on clinical outcomes in community-acquired pneumococcal pneumonia. Journal of Antimicrobial Chemotherapy. 2016;72(2):547-53. 173. RISK R. Carbapenemase-producing (OXA-48) Klebsiella pneumoniae ST392 in travellers previously hospitalised in Gran Canaria, Spain. 2018. 174. Luyt CE, Brechot N, Trouillet JL, Chastre J. Antibiotic stewardship in the intensive care unit. Crit Care. 2014;18(5):480. 175. Infectious Disease Society of America. Promoting antimicrobial stewardship in human medicine 2017 [Available from: www.idsociety.org/Stewardship_Policy/. 176. Paul M, Dickstein Y, Raz-Pasteur A. Antibiotic de-escalation for bloodstream infections and pneumonia: systematic review and meta-analysis. Clinical Microbiology and Infection. 2016;22(12):960-7. 177. Bassetti M, Peghin M, Vena A, Giacobbe DR. Treatment of infections due to MDR Gram- negative bacteria. Frontiers in Medicine. 2019;6. 178. EMA. Zavicefta Summary of Product Characteristics [Available from: https://www.ema.europa.eu/documents/product-information/zavicefta-epar-product- information_en.pdf. 179. EMA. Zerbaxa Summary of Product Characteristics [Available from: https://www.ema.europa.eu/documents/product-information/zerbaxa-epar-product- information_en.pdf. 180. EMA. Vabomere Summary of Product Characteristics [Available from: https://www.ema.europa.eu/documents/product-information/vabomere-epar-product- information_en.pdf. 181. Zhanel GG, Lawrence CK, Adam H, Schweizer F, Zelenitsky S, Zhanel M, et al. Imipenem– relebactam and meropenem–vaborbactam: two novel carbapenem-β-lactamase inhibitor combinations. Drugs. 2018;78(1):65-98. 182. EMA. Xerava Summary of Product Characteristics [Available from: https://www.ema.europa.eu/documents/product-information/xerava-epar-product- information_en.pdf. 183. Abdallah M, Olafisoye O, Cortes C, Urban C, Landman D, Quale J. Activity of eravacycline against Enterobacteriaceae and Acinetobacter baumannii, including multidrug-resistant isolates, from New York City. Antimicrobial agents and chemotherapy. 2015;59(3):1802-5. 184. Sutcliffe J, O'Brien W, Fyfe C, Grossman T. Antibacterial activity of eravacycline (TP-434), a novel fluorocycline, against hospital and community pathogens. Antimicrobial agents and chemotherapy. 2013;57(11):5548-58. 185. Landman D, Georgescu C, Martin DA, Quale J. Polymyxins revisited. Clinical microbiology reviews. 2008;21(3):449-65. 186. Wilson K, Bartlett C, Eales J, Edwards M. Literature review of burden of illness of serious, multi-drug resistant gram-negative infection. Final report. 2016. 187. Communications AH. A systematic review of rates of nephrotoxicity and neurotoxicity in patients treated with polymyxins 2019.

188. Lautermann J, Schacht J, Jahnke K. [Aminoglycoside toxicity. Pathomechanisms, clinical management and means of prevention]. HNO. 2003;51(4):344-51; quiz 52. 189. U.S. Food and Drug Administration. Zemdri - label. 2018. 190. Stein GE, Craig WA. Tigecycline: a critical analysis. Clinical infectious diseases. 2006;43(4):518-24. 191. Goodlet KJ, Nicolau DP, Nailor MD. Ceftolozane/tazobactam and ceftazidime/avibactam for the treatment of complicated intra-abdominal infections. Ther Clin Risk Manag. 2016;12:1811-26. 192. U.S. Department of Veterans Affairs Pharmacy Benefits Management Services. Ceftolozane/tazobactam (Zerbaxa (R)) - National Drug Monograph. 2015. 193. Zavascki AP, Bulitta JB, Landersdorfer CB. Combination therapy for carbapenem-resistant Gram-negative bacteria. Expert Rev Anti Infect Ther. 2013;11(12):1333-53. 194. Endo Pharmis Inc. Coly-Mycin (R). 2009. 195. West-ward Pharms Int. Polymyxin B Sulfate (R) label. 2012. 196. Hand E, Davis H, Kim T, Duhon B. Monotherapy with minocycline or trimethoprim/sulfamethoxazole for treatment of Stenotrophomonas maltophilia infections. The Journal of antimicrobial chemotherapy. 2016;71(4):1071-5. 197. Shionogi. Targeted Literature Search for Treatment Guidelines (Gram-negative infections, including MDR/CR). 2020. 198. Mensa J, Barberan J, Soriano A, Llinares P, Marco F, Canton R, et al. Antibiotic selection in the treatment of acute invasive infections by Pseudomonas aeruginosa: Guidelines by the Spanish Society of Chemotherapy. Rev Esp Quimioter. 2018;31(1):78-100. 199. Rodriguez-Bano J, Cisneros JM, Cobos-Trigueros N, Fresco G, Navarro-San Francisco C, Gudiol C, et al. Diagnosis and antimicrobial treatment of invasive infections due to multidrug-resistant Enterobacteriaceae. Guidelines of the Spanish Society of Infectious Diseases and Clinical Microbiology. Enferm Infecc Microbiol Clin. 2015;33(5):337.e1-.e21. 200. Girmenia C, Viscoli C, Piciocchi A, Cudillo L, Botti S, Errico A, et al. Management of carbapenem resistant Klebsiella pneumoniae infections in stem cell transplant recipients: an Italian multidisciplinary consensus statement. Haematologica. 2015;100(9):e373-6. 201. Bodmann KF, Grabein B, Kresken M, Derendorf H, Stahlmann R, Ott SR, et al. S2k Leitlinie - Kalkulierte parenterale Initialtherapie bakterieller Erkrankungen bei Erwachsenen – Update 2018. In: (PEG) P-E-GfCeV, editor.: AWMF online; 2019. 202. Chidiac C, group Sw. Update on a proper use of systemic fluoroquinolones in adult patients (ciprofloxacin, levofloxacin, moxifloxacin, norfloxacin, ofloxacin, pefloxacin. SPILF.). Med Mal Infect. 2015;45(9):348-73. 203. Tsuji BT, Pogue JM, Zavascki AP, Paul M, Daikos GL, Forrest A, et al. International Consensus Guidelines for the Optimal Use of the Polymyxins: Endorsed by the American College of Clinical Pharmacy (ACCP), European Society of Clinical Microbiology and Infectious Diseases (ESCMID), Infectious Diseases Society of America (IDSA), International Society for Anti-infective Pharmacology (ISAP), Society of Critical Care Medicine (SCCM), and Society of Infectious Diseases Pharmacists (SIDP). Pharmacotherapy. 2019;39(1):10-39. 204. Aguado JM, Silva JT, Fernandez-Ruiz M, Cordero E, Fortun J, Gudiol C, et al. Management of multidrug resistant Gram-negative bacilli infections in solid organ transplant recipients: SET/GESITRA-SEIMC/REIPI recommendations. Transplant Rev (Philadelphia). 2018;32(1):36-57. 205. Gudiol C, Aguilar-Guisado M, Azanza JR, Candel FJ, R C, Carratala J, et al. Consensus document of the Spanish Society of Infectious Diseases and Clinical Microbiology (SEIMC) and the Spanish Association of Hematology and Hemotherapy (SEHH) on the management of febrile neutropenia in patients with hematologic malignancies. 2019. 206. de With K, Wilke K, von Kern W, Strauß R, Kramme E, Friedrichs A, et al. S3 Leitlinie - Strategien zur Sicherung rationaler Antibiotika-Anwendung im Krankenhaus R-Nr. 092/001. In: (DGI) DGfIeV, editor.: AWMF online; 2019.

207. Maschmeyer G, Carratala J, Buchheidt D, Hamprecht A, Heussel CP, Kahl C, et al. Diagnosis and antimicrobial therapy of lung infiltrates in febrile neutropenic patients (allogeneic SCT excluded): updated guidelines of the Infectious Diseases Working Party (AGIHO) of the German Society of Hematology and Medical Oncology (DGHO). Ann Oncol. 2015;26(1):21-33. 208. Helsedirektoratet. Antibiotika i sykehus - Nasjonal faglig retningslinje 2018 [Available from: https://www.helsedirektoratet.no/Retningslinjer/Antibiotika-i-sykehus. 209. Folkhälsomyndigheten. ESBL-producerande tarmbakterier - Kunskapsunderlag med förslag till handläggning för att begränsa spridningen av Enterobacteriaceae med ESBL. 2014. 210. Torres A, Niederman MS, Chastre J, Ewig S, Fernandez-Vandellos P, Hanberger H, et al. International ERS/ESICM/ESCMID/ALAT guidelines for the management of hospital-acquired pneumonia and ventilator-associated pneumonia: Guidelines for the management of hospital- acquired pneumonia (HAP)/ventilator-associated pneumonia (VAP) of the European Respiratory Society (ERS), European Society of Intensive Care Medicine (ESICM), European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and Asociacion Latinoamericana del Torax (ALAT). European Respiratory Journal. 2017;50(3):09. 211. Johansen IS, Gerstoft J, Helweg-Larsen J, Engberg JH, Frimodt-Moller N, Bocher S, et al. Vejledning i brug af antibiotika 2019 [Available from: https://pro.medicin.dk/Specielleemner/Emner/318019. 212. Leone M, Bouadma L, Bouhemad B, Brissaud O, Dauger S, Gibot S, et al. Hospital-acquired pneumonia in ICU. Anaesth Crit Care Pain Med. 2018;37(1):83-98. 213. Dalhoff K, Abele-Horn M, Andreas S, Deja M, Ewig S, Gastmeier P, et al. S3 Leitlinie - Epidemiologie, Diagnostik und Therapie erwachsener Patienten mit nosokomialer Pneumonie – Update 2017 R-Nr. 020-013. In: e.V. DGfAuI, editor.: AWMF online; 2017. 214. Gonzalez-Castillo J, Martin-Sanchez FJ, Llinares P, Menendez R, Mujal A, Navas E, et al. Guidelines for the management of community-acquired pneumonia in the elderly patient. Revista Espanola de Quimioterapia. 2014;27(1):69-86. 215. Anonymous. NG139 Pneumonia (hospitalacquired): antimicrobial prescribing. In: NICE, editor. 2019. 216. Malmros K, Huttner BD, McNulty C, Rodriguez-Bano J, Pulcini C, Tangden T, et al. Comparison of antibiotic treatment guidelines for urinary tract infections in 15 European countries: Results of an online survey. Int J Antimicrob Agents. 2019;54(4):478-86. 217. Caron F, Galperine T, Flateau C, Azria R, Bonacorsi S, Bruyere F, et al. Practice guidelines for the management of adult community-acquired urinary tract infections. Med Mal Infect. 2018;48(5):327-58. 218. de Cueto M, Aliaga L, Alos J-I, Canut A, Los-Arcos I, Martinez JA, et al. Diagnosis and treatment of urinary tract infection: Clinical guidelines of the Spanish Society of Clinical Microbiology and Infectious Diseases (SEIMC) In: (SEIMC) SSoCMaID, editor. 2016. 219. Anonymous. S2 Leitlinie - Prävention, Diagnose, Therapie und Nachsorge der Sepsis R-Nr. 079-001. In: e.V. DIVfI-uNDaDS-G, editor.: AWMF online; 2010 (expired, new Guideline on the way). 220. Hentrich M, Schalk E, Schmidt-Hieber M, Chaberny I, Mousset S, Buchheidt D, et al. Central venous catheter-related infections in hematology and oncology: 2012 updated guidelines on diagnosis, management and prevention by the Infectious Diseases Working Party of the German Society of Hematology and Medical Oncology. Ann Oncol. 2014;25(5):936-47. 221. Penack O, Becker C, Buchheidt D, Christopeit M, Kiehl M, von Lilienfeld-Toal M, et al. Management of sepsis in neutropenic patients: 2014 updated guidelines from the Infectious Diseases Working Party of the German Society of Hematology and Medical Oncology (AGIHO). Ann Hematol. 2014;93(7):1083-95. 222. Chaves F, Garnacho-Montero J, Del Pozo JL, Bouza E, Capdevila JA, de Cueto M, et al. Diagnosis and treatment of catheter-related bloodstream infection: Clinical guidelines of the Spanish Society of Infectious Diseases and Clinical Microbiology and (SEIMC) and the Spanish Society of

Spanish Society of Intensive and Critical Care Medicine and Coronary Units (SEMICYUC). Med. 2018;42(1):5-36. 223. McGill F, Heyderman RS, Michael BD, Defres S, Beeching NJ, Borrow R, et al. The UK joint specialist societies guideline on the diagnosis and management of acute meningitis and meningococcal sepsis in immunocompetent adults. Journal of Infection. 2016;72(4):405-38. 224. Sartelli M, Chichom-Mefire A, Labricciosa FM, Hardcastle T, Abu-Zidan FM, Adesunkanmi AK, et al. The management of intra-abdominal infections from a global perspective: 2017 WSES guidelines for management of intra-abdominal infections. World Journal Of Emergency Surgery. 2017;12:29. 225. Montravers P, Dupont H, Leone M, J-M. C, Sfar., SRLF., et al. Prise en charge des infections intraabdominales. In: (Sfar) Sfdaedr, editor. 2015. 226. Anonymous. Antibiothérapie des infections à entérobactéries et à Pseudomonas aeruginosa chez l’adulte: place des carbapénèmes et de leurs alternatives. In: (HAS) HAdS, editor. 2019. 227. Shionogi. SLR and feasibility assessment for NMA report and apendix. Appendix B5 of the dossier. 228. Ferraro JS, Towse, A., Mestre-Ferrandiz, J.,. Incentives for new drugs to tackle anti-microbial resistance2017. 229. Karlsberg Schaffer S, West P, Towse A, Henshall C, Mestre-Ferrandiz J, Masterson R, et al. Additional Elements of Value for Health Technology Assessment Decisions. Office of Health Economics; 2017. 230. Schaffer SK, West, P., Towse, A., Henshall, C., Mestre-Ferrandiz, J., Masterton, R., Fischer, A, . Assessing the vaalue of new antibiotics: additional elements of value for health technology assessment decisions. Office of Health Economics 2017. 231. Interventions EPRUiEEiHSC. Framework for value assessment of new antimicrobials. 2018. 232. Shionogi. Summary of product characteristics. 233. Shionogi. SLR of Cefiderocol clinical and non-clinical evidence. Appendix B4 of the dossier. 234. U.S. Food and Drug Administration. Guidance for Industry Antibacterial Therapies for Patients With Unmet Medical Need for the Treatment of serious Bacterial Diseases 2013. 235. (EMA) EMA. Guideline on the evaluation of medicinal products 4 indicated for treatment of bacterial infections, Rev. 3. 2019 January 14. 236. Shionogi. Clinical study report: 1409R2121 (APEK-cUTI study) [Data on file]. 2017. 237. Shionogi. APEKs-cUTI CSR 1409R2121. See attachments folder A. [Data on File]. In press 2016. 238. Shionogi. APEKs-NP Study Synopsis. [Data on File]. In press. 239. Shionogi. Study synopsis: 1615R2132 (APEK-NP study) [Data on file]. 240. Bassetti M, Ariyasu M, Binkowitz B, Den Nagata T, Echols RM, Matsunaga Y, et al. Designing A Pathogen-Focused Study To Address The High Unmet Medical Need Represented By Carbapenem- Resistant Gram-Negative Pathogens–The International, Multicenter, Randomized, Open-Label, Phase 3 CREDIBLE-CR Study. Infection and Drug Resistance. 2019;12:3607. 241. Shionogi. Clinical study protocol: 1424R2131 (CREDIBLE-CR study) [Data on file]. 2016. 242. Shionogi. Final study summary: 1424R2131 (CREDIBLE-CR) [Data on file]. 243. Shionogi. CREDIBLE-CR Final Study Summary. Included in attachments folder C of the dossier. 2019. 244. FDA. ANTIMICROBIAL DRUGS ADVISORY COMMITTEE CEFIDEROCOL BRIEFING DOCUMENT NDA # 209445. 2019 October 16. 245. Shionogi. Cefiderocol compassionate use programme [Data on file]. 2019. 246. Edgeworth JD, Merante D, Patel S, Young C, Jones P, Vithlani S, et al. Compassionate Use of Cefiderocol as Adjunctive Treatment of Native Aortic Valve Endocarditis Due to Extremely Drug- resistant Pseudomonas aeruginosa. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2019;68(11):1932-4.

247. Stevens RW, Clancy M. Compassionate Use of Cefiderocol in the Treatment of an Intraabdominal Infection Due to Multidrug-Resistant Pseudomonas aeruginosa: A Case Report. Pharmacotherapy. 2019;39(11):1113-8. 248. Trecarichi EM, Quirino A, Scaglione V, Longhini F, Garofalo E, Bruni A, et al. Successful treatment with cefiderocol for compassionate use in a critically ill patient with XDR Acinetobacter baumannii and KPC-producing Klebsiella pneumoniae: a case report. The Journal of antimicrobial chemotherapy. 2019;74(11):3399-401. 249. Shionogi. Cefiderocol - Summary of Clinical Pharmacology (2.7.2). 250. Yamano Y. In vitro activity of cefiderocol against a broad range of clinically important gram- negative bacteria. Clinical Infectious Diseases. 2019;69(Supplement_7):S544-S51. 251. EUCAST. Rationale for EUCAST clinical breakpoints. 2020. 252. Falagas ME, Skalidis T, Vardakas KZ, Legakis NJ. Activity of cefiderocol (S-649266) against carbapenem-resistant Gram-negative bacteria collected from inpatients in Greek hospitals. The Journal of antimicrobial chemotherapy. 2017;72(6):1704-8. 253. Mushtaq S, Vickers A, Hussain A, Livermore D, Woodford N. In-vitro activity of cefiderocol(S- 649266) against multidrug-resistant Enterobacteriaceae from the UK. Public Health England2017. 254. Jacobs MR, Abdelhamed AM, Good CE, Rhoads DD, Hujer KM, Hujer AM, et al. ARGONAUT-I: activity of cefiderocol (S-649266), a siderophore cephalosporin, against Gram-negative bacteria, including carbapenem-resistant nonfermenters and Enterobacteriaceae with defined extended- spectrum β-lactamases and carbapenemases. Antimicrobial agents and chemotherapy. 2019;63(1):e01801-18. 255. Coppi M AA, Baglio G, Giani T, Rossolini GM, editor P1856 Activity of cefiderocol on a reference collection of carbapenemase-, ESBL-, and acquired class C beta-lactamases-producing Gram-negative pathogens. 29th ECCMID 2019; Amsterdam. 256. Kresken M BE, Wernicke S, Frank A, Anlauf G. , editor P1859 In vitro activity of cefiderocol against Gram-negative pathogens circulating in Germany. 29th ECCMID; 2019; Amsterdam, Netherlands. 257. Delgado-Valverde M MC, Serrano-Rocha L, Fernández-Cuenca F, Hernandez A., editor P1861 Activity of cefiderocol (S-649266), a new siderophore cephalosporin, against multidrug-resistant Enterobacteriaceae, Acinetobacter baumannii, Pseudomonas aeruginosa and Stenotrophomonas maltophilia. 29th ECCMID; 2019; Amsterdam, Netherlands. 258. Dobias J, Denervaud-Tendon V, Poirel L, Nordmann P. Activity of the novel siderophore cephalosporin cefiderocol against multidrug-resistant Gram-negative pathogens. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2017;36(12):2319-27. 259. Kresken M BE, Wernicke S, Frank A, G A. In vitro activity of cefiderocol against gram-negative pathogens circulating in Germany. ECCMID; April 13-16, 2019; Amsterdam2019. 260. Coppi M AA, Baglio G, Giani T, GM R. Activity of cefiderocol on a reference collection of carbapenemase-,ESBL-, and acquired class C beta-lactamases-producing gram-negative pathogens. ECCMID; April 13-16, 2019; Amsterdam2019. 261. Ito A KM, Rokushima M, Hackel M, Sahm D, Tsuji M, et al. Characterization of isolates showing high MICs to cefiderocol from global surveillance study SIDERO-WT-2014. ASM Microbe; June 20-24, 2019; San Francisco2019. 262. Shionogi. APEKS-cUTI Safety stratified results. 263. Shionogi. Safety - Pooled analyses Appendix 5.5 provided in attachments folder K_Fetcroja studies. 264. Shionogi. cUTI NMA Report. Provided in attachment folder A of the dossier. 265. Shionogi. Clinical study protocol: 1615R2132 (HAP/VAP study) [Data on file]. 2016. 266. Shionogi. Sidero Studies. [Data on File]. In press. 267. Shionogi. Cefiderocol EU core deck. 2020. 268. Shionogi. Competitor Table. [Data on File]. In press.

269. Shionogi. Analysis of susceptibility of Cefiderocol and comparators in European isolates and modelling chances of therapeutic success in different infection types. 2020. 270. EMA. Assessment report Zerbaxa. 2019. 271. Shionogi. CEA model - susceptibility data EUnetHTA. [Data on File]. In press 2020. 272. Torres A, Zhong N, Pachl J, Timsit JF, Kollef M, Chen Z, et al. Ceftazidime-avibactam versus meropenem in nosocomial pneumonia, including ventilator-associated pneumonia (REPROVE): a randomised, double-blind, phase 3 non-inferiority trial. The Lancet Infectious diseases. 2018;18(3):285-95. 273. Torres A, Rank D, Melnick D, Rekeda L, Chen X, Riccobene T, et al. Randomized Trial of Ceftazidime-Avibactam vs Meropenem for Treatment of Hospital-Acquired and Ventilator- Associated Bacterial Pneumonia (REPROVE): Analyses per US FDA-Specified End Points. Open Forum Infect Dis. 2019;6(4):ofz149. 274. Shionogi. Response to CHMP Day 180 List of Outstanding Issues (Clinical Efficacy and Safety) - EMEA/H/C/4829. 2020. 275. Shionogi. Response to CHMP Day 180 List of Outstanding Issues (Clinical benefit-risk) - EMEA/H/C/4829. 2020. 276. Shionogi. Response to Day 90 List of Questions. 2020. 277. Kadri SS, Lai YLE, Ricotta EE, Strich JR, Babiker A, Rhee C, et al. External Validation of Difficult-to-Treat Resistance Prevalence and Mortality Risk in Gram-Negative Bloodstream Infection Using Electronic Health Record Data From 140 US Hospitals. Open Forum Infect Dis. 2019;6(4):ofz110. 278. Durante-Mangoni E, Signoriello G, Andini R, Mattei A, De Cristoforo M, Murino P, et al. Colistin and rifampicin compared with colistin alone for the treatment of serious infections due to extensively drug-resistant Acinetobacter baumannii: a multicenter, randomized clinical trial. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2013;57(3):349-58. 279. Sirijatuphat R, Thamlikitkul V. Preliminary study of colistin versus colistin plus fosfomycin for treatment of carbapenem-resistant Acinetobacter baumannii infections. Antimicrobial agents and chemotherapy. 2014;58(9):5598-601. 280. Motsch J, Murta de Oliveira C, Stus V, Koksal I, Lyulko O, Boucher HW, et al. RESTORE-IMI 1: A Multicenter, Randomized, Double-blind Trial Comparing Efficacy and Safety of Imipenem/Relebactam vs Colistin Plus Imipenem in Patients With Imipenem-nonsusceptible Bacterial Infections. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2019. 281. Wunderink RG, Giamarellos-Bourboulis EJ, Rahav G, Mathers AJ, Bassetti M, Vazquez J, et al. Effect and Safety of Meropenem-Vaborbactam versus Best-Available Therapy in Patients with Carbapenem-Resistant Enterobacteriaceae Infections: The TANGO II Randomized Clinical Trial. Infect Dis Ther. 2018;7(4):439-55. 282. McKinnell JA, Dwyer JP, Talbot GH, Connolly LE, Friedland I, Smith A, et al. Plazomicin for Infections Caused by Carbapenem-Resistant Enterobacteriaceae. N Engl J Med. 2019;380(8):791-3. 283. van Duin D, Lok JJ, Earley M, Cober E, Richter SS, Perez F, et al. Colistin Versus Ceftazidime- Avibactam in the Treatment of Infections Due to Carbapenem-Resistant Enterobacteriaceae. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2018;66(2):163-71. 284. Shionogi. EU Risk Management Plan for Fetcroja. 2019. 285. Shionogi. Indirect treatment comparison to assess the efficacy and safety of treatments for gram-negative infections of the urinary tract. Attachments folder A: cUTI NMA report. [Data on File]. In press 2020.

APPENDICES AND ATTACHMENTS

Please see separate file attached with submission dossier