Core Submission Dossier PTJA06
Polatuzumab Vedotin (POLIVY®) in Combination with Bendamustine and Rituximab is Indicated for the Treatment of Adult Patients with Relapsed/Refractory Diffuse Large B-Cell Lymphoma (DLBCL) who are Not Candidates for Hematopoietic Stem Cell Transplant
Submitted by: Roche Registration GmbH
Contact details for administrative purposes Name of contact person: Dr Nasrin Vassel
Address of contact: F. Hoffmann-La Roche Ltd Pharmaceuticals Division Grenzacherstrasse 124 CH-4070 Basel, Switzerland
Email address: [email protected]
For agency completion Date of receipt: 14th of November 2019 Version: 2; amended dossier reflection CHMP discussions Project Identifier: PTJA06
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.
PTJA06 – Submission Dossier by Roche Registration GmbH
Table of Contents
LIST OF TABLES ...... 8
LIST OF FIGURES ...... 13
1. SUMMARY ...... 19 1.1 Health Problem and Unmet Medical Need ...... 19 1.2 Basis for the Assessment of Polatuzumab Vedotin (POLIVY) in combination with Bendamustine and Rituximab ...... 20 1.3 Clinical Efficacy and Safety ...... 23 1.4 Conclusions ...... 25
2. DESCRIPTION AND TECHNICAL CHARACTERISTICS OF THE TECHNOLOGY ...... 26 2.1 Characteristics of the Technology ...... 27 2.2 Investments and Tools Required ...... 31 2.2.1 Requirements to Use the Technology ...... 32 2.3 Regulatory Status of the Technology ...... 34 2.3.1 European Union ...... 34 2.3.2 United States...... 35 2.4 Current Use of the Technology...... 37
3. HEALTH PROBLEM AND CURRENT CLINICAL PRACTICE ...... 37 3.1 Overview of the Disease or Health Condition ...... 38 3.1.1 Disease Overview ...... 38 3.1.2 Symptoms and Burden of the Disease ...... 38 3.1.3 Classification of the Disease and Prognosis ...... 39 3.1.4 Epidemiology...... 42 3.2 Clinical Management of the Disease or Health Condition ...... 45 3.2.1 First-line DLBCL Treatment ...... 45 3.2.2 Treatment of R/R DLBCL ...... 46 3.2.2.1 Treatment of R/R DLBCL Patients Eligible for SCT ...... 46 3.2.2.2 Treatment of R/R DLBCL Patients Ineligible for SCT ...... 46 3.3 Unmet Medical Need ...... 54
2 PTJA06 – Submission Dossier by Roche Registration GmbH
3.4 Target Population and the Proposed Positioning of the Technology in the Patient Pathway of CARE ...... 55 3.5 Comparators in the Assessment ...... 57 3.5.1 Usage and Results of Selected Key Regimens for the Treatment of R/R DLBCL Patients Ineligible for SCT ...... 58 3.5.1.1 First- and Subsequent-Line Treatment options for SCT Ineligible R/R DLBCL Patients ...... 58 3.5.1.2 Second- and Subsequent-line Treatment Options for SCT Ineligible R/R DLBCL Patients ...... 60
4. CLINICAL EFFECTIVENESS AND SAFETY ...... 62 4.1 Overview of Evidence to Support the Use of Polatuzumab Vedotin in R/R DLBCL ...... 62 4.2 Pivotal Study GO29365 ...... 64 4.2.1 Overall Study Design ...... 65 4.2.2 Discussion of Study Design ...... 73 4.2.2.1 Rationale for the Selection of Comparator for R/R DLBCL Patients (Randomized Phase II) ...... 73 4.2.2.2 PET-defined CR as a Primary Tumor Response Assessment ...... 74 4.2.3 Key Inclusion and Exclusion Criteria ...... 74 4.2.4 Analysis Populations ...... 74 4.2.5 Efficacy Analysis ...... 75 4.2.6 Patient-Reported Outcome (PRO) Analyses ...... 76 4.2.7 Anti-Drug Antibody (ADA) Analysis ...... 76 4.2.8 Safety Reporting and Analysis ...... 76 4.2.8.1 Adverse Events ...... 77 4.3 Results: Study Population ...... 79 4.3.1 Patient Population ...... 79 4.3.2 Patient Disposition ...... 79 4.3.3 Demographics ...... 83 4.3.4 Baseline Disease Characteristics ...... 83 4.3.5 Prior Anti-Lymphoma Therapy ...... 86 4.4 Efficacy Results in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II) ...... 89
3 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4.1 Overview of Efficacy in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II) ...... 90 4.4.2 Efficacy in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 98 4.4.2.1 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 98 Updated Overall Survival...... 100 Totality of R/R DLBCL Data from Study GO29365 (Phase Ib/Phase II Randomized) ...... 102 4.4.2.2 Progression-Free Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 103 Updated Progression-Free Survival as Assessed by the Investigator ...... 112 4.4.2.3 CR Rate (IRC-assessed) by PET-CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 115 4.4.2.4 Response Rates (CR and OR) at the Primary Response Assessment Measured With or Without PET-CT as Assessed by IRC and INV ...... 116 4.4.2.4.1 Response Rates at the Primary Response Assessment Based on PET-CT ...... 117 4.4.2.4.2 Response Rates at the Primary Response Assessment Based on CT Only ...... 118 4.4.2.4.3 Concordance Between IRC and INV Assessments at the PRA ...... 119 4.4.2.5 Best Overall Response (CR/PR) While on Study as Assessed by the IRC and INV ...... 119 4.4.2.6 DOR as Assessed by IRC and INV ...... 121 4.4.2.7 Sensitivity Analyses – PFS by IRC ...... 123 4.4.2.8 Multiple-Cox Regression ...... 124 4.4.3 Subgroup Analyses ...... 124 4.4.3.1 Subgroup Analysis at the Primary Response Assessment by Primary Reasons for Stem Cell Transplant Ineligibility ...... 124
4 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4.3.2 Subgroup Analyses by Number of Lines of Prior Treatment ...... 127 4.4.4 Patient-Reported Outcomes ...... 130 4.4.4.1 Therapy-Induced Neuropathy Assessment Score (TINAS) ...... 130 4.4.4.2 Supportive Data from the ROMULUS Study ...... 131 4.5 Immunogenicity ...... 134 4.6 Safety Results in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; Safety Population) ...... 135 4.6.1 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 135 4.6.2 Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 139 4.6.2.1 Extent of Exposure to Study Treatment in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)...... 139 4.6.2.2 Common Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 141 4.6.2.3 Adverse Events by Severity in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 144 4.6.2.4 Grade 3 Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 145 4.6.2.5 Deaths in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 147 4.6.2.6 Serious Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 150 4.6.2.7 Adverse Events Leading to Withdrawal of Study Treatment in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 151
5 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.8 Adverse Events that Led to Treatment Modification in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 152 4.6.2.9 Selected Adverse Events/Adverse Events to Monitor in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 154 Peripheral Neuropathy ...... 156 Neutropenia Including Febrile Neutropenia ...... 157 Infections and Infestations...... 159 Hepatic Toxicity ...... 160 Anemia ...... 161 Thrombocytopenia ...... 162 Diarrhea ...... 162 4.6.3 Updated Safety Results in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 163 4.6.4 Safety in Special Groups ...... 166 4.7 Additional Data ...... 167 4.7.1 GO29365 Efficacy Results for Arm G (N 42) ...... 167 4.8 Systematic Literature Review and Indirect Treatment Comparison ...... 167 4.8.1 Systematic Literature Review ...... 168 4.8.1.1 Methodology ...... 168 4.8.1.2 Results of the Systematic Literature Review ...... 172 4.8.1.3 Study Quality Assessment ...... 175 4.8.2 Indirect Treatment Comparison (ITC) Feasibility Assessment ...... 177 4.8.2.1 Identification and Selection of Relevant Studies ...... 179 4.8.3 Matching-adjusted Indirect Comparisons (MAICs) ...... 180 4.8.3.1 Introduction to MAICs ...... Fout! Bladwijzer niet gedefinieerd. 4.8.3.2 Comparisons of OS and PFS Endpoints . Fout! Bladwijzer niet gedefinieerd. 4.8.3.3 MAIC Results ...... Fout! Bladwijzer niet gedefinieerd.
6 PTJA06 – Submission Dossier by Roche Registration GmbH
4.8.3.3.1 GO29365 Pola BR Arm Versus CAR-Ts Fout! Bladwijzer niet gedefinieerd. 4.8.3.3.2 GO29365 Pola BR Arm Versus R-GemOx ...... Fout! Bladwijzer niet gedefinieerd. 4.8.3.3.3 GO29365 Pola BR Arm Versus Pixantrone ...... Fout! Bladwijzer niet gedefinieerd. 4.9 Real-World Data Evidence ...... 211 4.9.1 Analysis of the Christie NHS Trust ...... 211 4.9.2 Flatiron Health ...... 216 4.9.3 Veterans Affairs ...... 217
5. DISCUSSION AND CONCLUSIONS FOR THE ASSESSMENT ...... 219
6. REFERENCES ...... 223
7 PTJA06 – Submission Dossier by Roche Registration GmbH
1. List of Tables
Table 1 Scope of the Submission ...... 22 Table 2 Features of Polatuzumab Vedotin (POLIVY®) ...... 27 Table 3 Administration and Dosing of Polatuzumab Vedotin (POLIVY®) in combination with Bendamustine and Rituximab ...... 28 Table 4 Regulatory Status of Polatuzumab Vedotin (POLIVY) ...... 36 Table 5 Ann Arbor Staging Classification ...... 40 Table 6 Lugano Staging Classification ...... 41 Table 7 International Prognostic Index ...... 41 Table 8 DLBCL Treatment Patterns Across Developed Countries ...... 43 Table 9 Relevant Guidelines for Diagnosis and Management of R/R DLBCL Patients Ineligible for SCT ...... 50 Table 10 Key Comparators Recommended by the EUnetHTA in the Proposed PICO ...... 58 Table 11 Overview of Ongoing Key Studies of Interest for Polatuzumab Vedotin ...... 63 Table 12 Characteristics of the Studies ...... 69 Table 13 GO29365: Key Inclusion and Exclusion Criteria ...... 74 Table 14 GO29365: Efficacy Outcome Measures and Analysis Methodology ...... 75 Table 15 GO29365: Disposition of Patients in the Phase Ib Safety Run-In at the Clinical Cut-off Date (30 April 2018) ...... 80 Table 16 Disposition of Patients with R/R DLBCL in the Phase II Randomized Portion of the GO29365 Study at the Clinical Cut- off Date (30 April 2018) ...... 81 Table 17 Duration of Follow-Up (Phase Ib and Randomized Phase II; ITT Population)* ...... 83 Table 18 GO29365: Key Demographic, Baseline Disease Characteristics and Prognostic Factors for Patients with R/R DLBCL Treated with Pola + BR or BR (Phase Ib and Phase II Randomized; ITT Population) ...... 85 Table 19 Previous Anti-Lymphoma Treatment and Response in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib and Phase II Randomized; ITT Population) ...... 88 Table 20 GO29365: Overview of Key Efficacy Outcomes in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 92 Table 21 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 99
8 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 22 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 101 Table 23 Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 104 Table 24 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 107 Table 25 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 113 Table 26 CR Rate (IRC-assessed) by PET-CT at Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 116 Table 27 Objective Response (CR/PR) Rates (IRC-assessed) by PET- CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 117 Table 28 Complete Response (CR) and Objective Response (CR/PR) Rates (INV-assessed) by PET-CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 118 Table 29 Best Overall Response Rate (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 120 Table 30 Best Overall Response Rate (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 121 Table 31 Duration of Response (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 122 Table 32 Summary of Overall Survival by Primary Reason for Stem Cell Transplant Ineligibility ...... 125 Table 33 Summary of Progression-Free Survival by Primary Reason for Stem Cell Transplant Ineligibility ...... 126 Table 34 Summary of PET-CT Response at the Primary Response Assessment by Primary Reasons for Stem Cell Transplant Ineligibility ...... 127 Table 35 Summary of Overall Survival by Prior Line of Therapy (1, 2, 3 or more) ...... 128
9 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 36 Summary of Progression-Free Survival by Prior Line of Therapy (1, 2, 3 or more) ...... 129 Table 37 Summary of PET-CT Complete Response at the Primary Response Assessment by Prior Line of Therapy (1, 2, 3 or more) ...... 130 Table 38 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 138 Table 39 Summary of Extent of Exposure to Pola BR or BR in Patients with R/R DLBCL (Phase Ib/Phase II Randomized; Safety Population) ...... 140 Table 40 Most Frequently Reported Adverse Events (10%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 142 Table 41 Summary of Adverse Events by Highest NCI CTCAE Grade in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 144 Table 42 Most Frequently Reported Grade 35 Adverse Events (5%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 146 Table 43 Deaths Due to Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 148 Table 44 Serious Adverse Events Reported in More than One R/R DLBCL Patient Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 151 Table 45 Incidence of Adverse Events Leading to Discontinuation, Drug Interruption or Dose Modification of Study Drugs in Patients with R/R DLBCL (Phase Ib/Phase II Randomized; Safety Population) ...... 153 Table 46 Selected Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 155 Table 47 Summary of Peripheral Neuropathy in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 157 Table 48 Summary of Neutropenia Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 159
10 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 49 Summary of Infections in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) ...... 160 Table 50 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) (CCOD: 11 October 2018) ...... 166 Table 51 Eligibility Criteria Used in the Initial Screening to Identify Studies Relating to Any Pharmacological Treatment for Patients with R/R DLBCL ...... 169 Table 52 Risk of Bias Across Studies ...... 176 Table 53 List of All Relevant Studies ...... 180 Table 54 Summary of Baseline Characteristics of Mounier et al. (2013) and Matched GO29365 Populations ...... 184 Table 55 Summary of Matching Pre-Selected Prognostic Factors for Pooled Pola BR (GO29365) Group Versus R-GemOx (Mounier et al. 2013) ...... 185 Table 56 Unweighted Complete Response Rates for Pola BR (GO29365) and R-GemOx (Mounier et al. 2013) ...... 185 Table 57 Weighted Complete Response Rates for Pola BR (GO29365) and R-GemOx (Mounier et al. 2013) ...... 186 Table 58 Separate MAICs by Individual Pre-Selected Prognostic Factor: Pola BR (GO29365) Versus R-GemOx (Mounier et al. 2013) ...... 187 Table 59 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Pixantrone (PIX301) ...... 189 Table 60 Unweighted Complete Response Rates for Pola BR (GO29365) and Pixantrone (PIX301) ...... 190 Table 61 Individual MAICS for Complete Response Rates: Pola BR (GO29365) Versus Pixantrone (PIX301) ...... 190 Table 62 Individual MAICs for Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301) ...... 192 Table 63 GO29365: Landmark Progression-Free Survival (Investigator-assessed; CCOD: 30 April 2018) ...... 196 Table 64 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Kymriah (JULIET) ...... 198 Table 65 Unweighted Complete Response Rates for Pola BR (GO29365) and Kymriah (JULIET) ...... 198 Table 66 Weighted Complete Response Rates for Pola BR (GO29365) and Kymriah (JULIET) ...... 199 Table 67 Sensitivity Analysis: Pola BR (GO29365) Versus Kymriah (JULIET) ...... 200
11 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 68 Unweighted Median Overall Survival for Pola BR (GO29365) and Kymriah (JULIET) ...... 201 Table 69 Weighted Median Overall Survival for Pola BR (GO29365) and Kymriah (JULIET) ...... 202 Table 70 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Yescarta (ZUMA-1) ...... 204 Table 71 Unweighted Complete Response Rates for Pola BR (GO29365) and Yescarta (ZUMA-1) ...... 204 Table 72 Weighted Complete Response Rates for Pola BR (GO29365) and Yescarta (ZUMA-1)...... 205 Table 73 Sensitivity Analysis: Pola BR (GO29365) Versus Yescarta (ZUMA-1) ...... 205 Table 74 Unweighted Overall Survival for Pola BR (GO29365) and Yescarta (ZUMA-1) ...... 206 Table 75 Weighted Overall Survival for Pola BR (GO29365) and Yescarta (ZUMA-1) ...... 207 Table 76 Cohort Attrition ...... 212 Table 77 Treatment Regimens Recommended for Transplant Ineligible Patients ...... 213 Table 78 Treatment Regimens Recommended for Transplant Eligible and Ineligible Patients ...... 213 Table 79 Patient Characteristics...... 214 Table 80 Treatments Received by Line of Therapy ...... 215 Table 81 BR Utilization Among Transplant Eligible and Ineligible R/R DLBCL Patients ...... 216 Table 82 Treatment Regimens Received by Second-Line Patients from the Veterans Affairs Database ...... 218 Table 83 Selected Baseline Characteristics for Patients Treated with either BR or R-GemOx from the Veterans Affairs Database ...... 219
12 PTJA06 – Submission Dossier by Roche Registration GmbH
2. List of Figures
Figure 1 Estimated Number of R/R DLBCL Patients Treated in Each Line in the EU ...... 44 Figure 2 Estimated Trend for DLBCL Incidence and Treated R/R Population in the EU28 ...... 45 Figure 3 Highly Fragmented Treatment Landscape for Transplant Ineligible Patients with R/R DLBCL in the EU ...... 48 Figure 4 Treatment Pathway for First-line and R/R Patients with DLBCL ...... 54 Figure 5 Broad Treatment Flow and Estimated Number of R/R DLBCL Patients to Receive Treatment with Polatuzumab Vedotin in the EU28 in 2019 ...... 56 Figure 6 Trend for DLBCL Incidence, R/R DLBCL and Pool of Patients to Receive Treatment with Polatuzumab Vedotin in the EU28 ...... 57 Figure 7 GO29365: Overview of Study Design ...... 67 Figure 8 GO29365: KM Plot of Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 94 Figure 9 GO29365: Forest Plot of Hazard Ratio for Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population) ...... 95 Figure 10 GO29365: KM Plot of Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 96 Figure 11 GO29365: Forest Plot of Hazard Ratio for Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population) ...... 97 Figure 12 KM Plot of Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) ...... 102 Figure 13 KM Overall Survival Estimates for Pola BR (Phase Ib/Phase II Randomized) ...... 103 Figure 14 KM Plot of Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)...... 109 Figure 15 Forest Plot of Hazard Ratio for Progression-Free Survival (INV- assessed) in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population) ...... 111 Figure 16 KM Plot of Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)...... 114
13 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 17 KM Progression-Free Survival (INV-assessed) Estimates for Pola BR (Phase Ib/Phase II Randomized) ...... 115 Figure 18 Mean Change (SE) in MDASI Scores from Baseline by Symptom (DLBCL R Pola [2.4 mg/kg]) ...... 133 Figure 19 Mean Change (SE) in MDASI Scores from Baseline – Severity and Interference Scales (DLBCL R Pola [2.4 mg/kg]) ...... 133 Figure 20 PRISMA Flow Diagram of the Literature Search ...... 174 Figure 21 Assessment Criteria Used to Assess Complete Response (INV-assessed) in the Included Studies ...... 175 Figure 22 Risk of Bias Across Categories ...... 176 Figure 23 Disconnected Network of Evidence ...... 178 Figure 24 Unweighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301) ...... 191 Figure 25 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301) ...... 193 Figure 26 JULIET Study Schema ...... 194 Figure 27 ZUMA-1 Study Schema ...... 195 Figure 28 Unweighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Kymriah (JULIET) ...... 201 Figure 29 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Kymriah (JULIET) ...... 202 Figure 30 Unweighted KM Plot of Overall Survival: Pola BR (GO29365) Versus Yescarta (ZUMA-1) ...... 206 Figure 31 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Yescarta (ZUMA-1) ...... 207
14 PTJA06 – Submission Dossier by Roche Registration GmbH
Abbreviations
AA Ann Arbor aaIPI Age-adjusted international prognostic index acMMAE Antibody‐conjugated monomethyl auristatin E ADA Anti-drug antibody ADC Antibody-drug conjugate AE Adverse event AESI Adverse events of special interest ALT Alanine aminotransferase ASCT Autologous stem cell transplantation ASH American Society of Hematology AST Aspartate aminotransferase AUC Area under the curve BLA Biologics license application BOR Best objective response BR Bendamustine plus rituximab BSC Best supportive care BTD Breakthrough Therapy Designation CAR-T Chimeric antigen receptor T-cell CCOD Clinical cut-off date CD19 Cluster of differentiation 19 CD20 Cluster of differentiation 20 CD79b Cluster of differentiation 79b CHMP Committee for Medicinal Products for Human Use CHOP Cyclophosphamide, doxorubicin, vincristine, and prednisone CI Confidence interval CMH Cochran-Mantel-Haenszel CMV Cytomegalovirus CR Complete response CRF Case report form CRS Cytokine release syndrome CSR Clinical study report CT Computed tomography CYP3A Cytochrome P450 3A
15 PTJA06 – Submission Dossier by Roche Registration GmbH
DILI Drug-induced liver injury DLBCL Diffuse large B-cell lymphoma DoR Duration of response EC European Commission eCRF Electronic case report form EFS Event-free survival EMA European Medicines Agency EOT End of treatment Etoposide, prednisone, vincristine (Oncovin), cyclophosphamide, EPOCH hydroxydaunorubicin ePRO Electronic patient-reported outcome device ESMO European Society for Medical Oncology ESS Effective sample size EU European Union EUnetHTA European Network for Health Technology Assessment FDA Food and Drug Administration FDG-PET 18Fluorodeoxyglucose positron emission tomography FL Follicular lymphoma G-CSF Granulocyte colony-stimulating factor HMRN Haematological Malignancy Research Network HR Hazard ratio HRQoL Health-related quality of life HSCT Hematopoietic stem cell transplantation IAP Independent Assessment Panel IC Immune cells ICD International Classification of Diseases ICH International Conference of Harmonization INV Investigator INV-PFS Investigator-assessed progression-free survival IPI International Prognostic Index IRC Independent Review Committee ITT Intention-to-treat IV Intravenous KM Kaplan-Meier LOQ List of questions
16 PTJA06 – Submission Dossier by Roche Registration GmbH
MAA Marketing Authorization Application MAIC Matching-adjusted indirect comparison MDASI M.D. Anderson Symptom Inventory MedDRA Medical Dictionary for Regulatory Activities MMAE Monomethyl auristatin E MoA Mechanism of action mOS Median overall survival mPFS Median progression-free survival NALT New anti-lymphoma therapy NCCN National Comprehensive Cancer Network National Cancer Institute - Common Terminology Criteria for Adverse NCI CTCAE Events NE Not evaluable NHL Non-Hodgkin’s lymphoma NICE National Institute for Health and Care Excellence NMA Network meta-analysis ODD Orphan Drug Designation OR Objective response ORR Objective response rate OS Overall survival PD Progressive disease PET Positron emission tomography PET-CR Positron emission tomography-complete response PET-CT Positron emission tomography-computed tomography PFS Progression-free survival PICOS Population, interventions, comparisons, outcomes, and Study Design PK Pharmacokinetic PML Progressive multifocal leukoencephalopathy Pola + BG Polatuzumab vedotin plus bendamustine and obinutuzumab Pola + BR Polatuzumab vedotin plus bendamustine and rituximab PRA Primary response assessment PRIME PRIority MEdicines PRO Patient-reported outcome R Rituximab R/R Relapsed/refractory
17 PTJA06 – Submission Dossier by Roche Registration GmbH
R/R DLBCL Relapsed/refractory diffuse large B-cell lymphoma R/R FL Relapsed/refractory follicular lymphoma R-CHOP Rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone R-DHAP Rituximab, dexamethasone, cytarabine, and cisplatin R-ESHAP Rituximab, etoposide, methylprednisolone, cytarabine, and cisplatin R-GDP Rituximab, gemcitabine, cisplatin, and dexamethasone R-GemOx Rituximab, gemcitabine, and oxaliplatin R-InO Inotuzumab ozogamicin plus rituximab R-ICE Rituximab, ifosfamide, carboplatin, and etoposide R + pola Rituximab + polatuzumab vedotin RBC Red blood cell RCT Randomized clinical trial RP2D Recommended Phase II dose SAE Serious adverse event SCT Stem cell transplantation SD Stable disease SE Standard error SLR Systematic literature review SMQ Standardised MedDRA Query SoC Standard of care SOC System organ class TINAS Therapy-induced Neuropathy Assessment Score TLS Tumor lysis syndrome UK United Kingdom US United States WHO World Health Organization
18 PTJA06 – Submission Dossier by Roche Registration GmbH
Using this evidence submission template
This evidence submission template contains suggestions to companies about what information to include, highlighted in blue, which agencies can adapt as necessary. There are also ‘form fields’ that prompt companies for their response, for example [add details here]. To insert a response, a company should click once anywhere within the highlighted text and then type in their response. This overwrites the section that was highlighted. To delete a form field, click anywhere within the highlighted text and press DELETE.
1. SUMMARY 1.1 HEALTH PROBLEM AND UNMET MEDICAL NEED Diffuse large B-cell lymphoma (DLBCL) is the most common histologic subtype of non-Hodgkin’s lymphoma (NHL), accounting for 30% to 48% of all NHL cases. Diffuse large B-cell lymphoma arises from mature B-cells that undergo a neoplastic transformation. Approximately 60% of newly diagnosed patients may be cured with the current standard of care (SoC) that consists of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP). Nevertheless, despite the improvements in overall survival (OS) of patients with DLBCL, one-third of patients present with either a primary refractoriness to the first-line treatment used (mainly R-CHOP) or relapse after reaching a complete response (CR). Relapsing or being refractory to first-line treatment remains the major cause of morbidity and mortality of this disease. Relapsed/refractory diffuse large B-cell lymphoma (R/R DLBCL) patients who are ineligible for stem cell transplant (SCT) have limited treatment options and a poor prognosis with a median overall survival (mOS) of approximately 6 months.
In the European Union (EU), the treatment landscape for patients with R/R DLBCL who are ineligible for transplant or who relapse after transplant is highly fragmented. There are no universally established SoC or regulatory-approved therapies upon first relapse, as no prior randomized trials have established the superiority of one regimen over another for this population. The most commonly prescribed regimens, used off-label, are rituximab in combination with gemcitabine and/or platinum-based therapies or rituximab in combination with bendamustine.
To date, only a few treatments are approved for use in the third-line and beyond setting. Pixantrone (Pixuvri®) monotherapy is approved in the EU for the treatment of adult patients with multiply relapsed/refractory (R/R) aggressive NHL based on the efficacy and safety results from the randomized Phase III, PIX301 study. Pixantrone was further investigated in the pivotal randomized Phase III, PIX306 study, the results of which were recently presented at the American Society of Hematology (ASH) Annual Meeting in
19 PTJA06 – Submission Dossier by Roche Registration GmbH
2018. While the primary endpoint of the PIX306 study was not met, subgroup analysis of progression-free survival (PFS) and OS results in third-line and beyond were comparable when indirectly compared to the pivotal PIX301 study. More recently, in 2018, two cluster of differentiation 19 (CD19)-directed chimeric antigen receptor T-cell (CAR-T) therapies were approved in the EU for the treatment of R/R DLBCL patients who have had two or more lines of prior systemic therapies. While CAR-T therapies have shown efficacy with durable CRs, their use may be limited for the general R/R DLBCL population due to the toxicity profile which requires carefully-selected patients and treatment centers with specially-trained staff. In addition, the waiting period associated with CAR-T manufacture, which is in the range of several weeks, may be prohibitive in patients with rapidly progressing disease.
As all these treatments are indicated for third-line and later, there remains a high unmet medical need for patients with R/R DLBCL after first relapse, due to poor life expectancy and no universally established SoC or regulatory-approved therapies available in the EU, particularly for those who are ineligible for transplant. Newer approaches, for use after first relapse, that incorporate agents with novel mechanisms that offer patients a more effective treatment with improved survival and with acceptable safety profiles are urgently needed.
Polatuzumab vedotin is the first in-class and first antibody-drug conjugate (ADC) in DLBCL. Polatuzumab vedotin has been developed specifically for B-cell malignancies that express the cluster of differentiation 79b (CD79b). Diffuse large B-cell lymphoma expresses this marker ubiquitously making it a promising target for the development of new therapies. Polatuzumab vedotin binds specifically to CD79b which results in ADC internalization and subsequent release, or the potent microtubule inhibitor monomethyl auristatin E (MMAE) inside the targeted neoplastic B-cells. This process is thought to maximize tumor cell death while potentially minimizing the effects on normal healthy cells.
1.2 BASIS FOR THE ASSESSMENT OF POLATUZUMAB VEDOTIN (POLIVY) IN COMBINATION WITH BENDAMUSTINE AND RITUXIMAB Stem cell transplant ineligibility represents the major therapeutic challenge. Regardless of treatment line, patients with R/R DLBCL who are ineligible for SCT treatment are treated without curative intent. There are only a few therapeutic options for patients with R/R DLBCL who are ineligible for SCT due to age, comorbidities or lack of response to salvage treatment or have relapsed after SCT. There are no universally-established therapies or regulatory-approved treatments in this setting. Therefore, the goal for these patients is to achieve a prolonged response and to ultimately extend survival. Due to the heterogeneity of the patient population with overlapping disease characteristics and limited treatment options, patients on first-line and later line relapse, are described as a patient segment with a uniformly high unmet need due to little chance at prolonged
20 PTJA06 – Submission Dossier by Roche Registration GmbH
control of disease with a dismal survival outcome. Primary market research has demonstrated that the treatment landscape is highly fragmented with no clear SoC and that physicians use the same treatment regimens regardless of therapy line. The most recent primary market research data indicate that both bendamustine plus rituximab (BR) and the combination of rituximab, gemcitabine, and oxaliplatin (R-GemOx) are the two most frequently used treatment regimens in transplant ineligible patients in both second-line (BR: 18% and R-GemOx: 20%) and third-line and beyond (BR: 19% and R-GemOx: 17%). Additionally, there is no clear consensus in clinical guidelines, such as those put forth by the European Society for Medical Oncology (ESMO) and the National Comprehensive Cancer Network (NCCN) with regards to management by treatment-line for R/R DLBCL patients who are ineligible for SCT. The unmet need for these patients is reflected in the ESMO guidelines that recommend non-curative platinum- and/or gemcitabine- based regimens and enrolment into clinical trials investigating novel drugs due to the absence of effective therapies. Thus, our position and the focus of this assessment is that R/R DLBCL patients who are ineligible for SCT should be treated as one single population.
The evidence in this submission demonstrates that the addition of the novel ADC, polatuzumab vedotin (POLIVY) to the combination of BR in patients with R/R DLBCL who are ineligible for SCT, resulted in higher CR and objective response rates (ORRs) and longer OS, PFS, best overall response (BOR), and duration of response (DOR) compared to BR alone. In addition, the aim of this submission is to provide the safety and tolerability assessment of the polatuzumab vedotin in combination with BR treatment when administrated to patients with R/R DLBCL and assess the immunogenicity of polatuzumab vedotin. The assessment is based on the Population, Interventions, Comparisons, Outcomes, and Study Design (PICOS) framework (Table 1).
21 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 1 Scope of the Submission
Description Submission Scope Population Adult patients with R/R DLBCL who are not candidates for hematopoietic stem cell transplant Intervention Polatuzumab vedotin in combination with bendamustine and rituximab (BR) Comparators Key comparators of interest include: BR R-GemOx R-ICE R-DHAP Pixantrone Axicabtagene ciloleucel Tisagenlecleucel Outcomes Mortality: Overall survival
Morbidity: Progression-free survival (Independent Review Committee [IRC]- and Investigator [INV]- assessed) by PET-CT or CT alone. Complete response rate using modified Lugano Response Criteria as measured at the primary response assessment (PRA; 68 weeks after Cycle 6 Day 1 or last dose of study medication) as measured by positron emission tomography (PET) – computed tomography (CT) scan and as determined by an IRC Response rates (IRC- and INV-assessed) at PRA by PET or CT alone Best overall response (IRC- and INV-assessed) Duration of response (IRC- and INV-assessed) by PET-CT or CT alone
Other Outcomes: Health-related Quality of Life (HRQoL) Number of patients with polatuzumab vedotin antibodies
Safety (overall and by preferred term and system organ class): Any adverse events (AEs) AEs Grade 34 AEs Grade 5 (AEs leading to death) Any serious adverse events (SAEs) AEs leading to discontinuation AEs of special interest (AESI) Study Design Direct evidence for the clinical efficacy and safety for polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL who are not candidates for hematopoietic stem cell transplant (HSCT) is derived from the Phase Ib safety run-in and Phase II randomized portions of the pivotal GO29365 study (see Section 4).
22 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 1 Scope of the Submission (cont.)
AE: adverse event; AESI: adverse events of special interest; BOR: best objective response; BR: bendamustine and rituximab; CR: complete response; CT: computed tomography; DOR: duration of response; HRQoL: health-related quality of life; HSCT: hematopoietic stem cell transplantation; INV-PFS: investigator-assessed progression-free survival; IRC: Independent Review Committee; IRC-PFS: Independent Review Committee-assessed progression-free survival; OS: overall survival; PET: positron emission tomography; PET-CT: positron emission tomography-computed tomography; PRA: primary response assessment; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-GemOx: rituximab, gemcitabine, and oxaliplatin; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide; SAE: serious adverse event.
1.3 CLINICAL EFFICACY AND SAFETY The efficacy analyses are based on the Phase II randomized portion of the GO29365 study and on an updated analysis conducted one year after all patients had completed their primary response assessment (PRA) (clinical cut-off date [CCOD]: 30 April 2018). Results obtained upon median follow-up of 22.3 months in both the polatuzumab vedotin in combination with bendamustine and rituximab (Pola + BR) arm and BR treatment arm are summarized below for patients with R/R DLBCL. In the randomized comparison, Pola BR showed significantly improved OS, PFS, CR rates and ORR compared to BR alone.
Mortality Importantly, in the randomized Phase II portion, the risk of death was reduced by 58% in patients treated with Pola BR compared to BR (stratified hazard ratio [HR] 0.42; 95% confidence interval [CI]: [0.24, 0.75]; p 0.0023). Median OS was more than 2.5 times longer in patients receiving Pola BR compared to BR alone (12.4 months vs. 4.7 months).
Morbidity In the Phase II randomized portion, the risk of progressive disease (PD) or death as determined by the Independent Review Committee (IRC) was reduced by 64% in patients treated with Pola BR (stratified HR 0.36; 95% CI: [0.21, 0.63]; p 0.0002) with median progression-free survival( mPFS) being over two-fold higher (9.5 months [95% CI: 6.2, 13.9] vs. 3.7 months [95% CI: 2.1, 4.5]). The risk of PD or death as determined by the investigator (INV) was reduced by 66% in patients treated with Pola BR (stratified HR: 0.34; 95% CI: [0.20, 0.57]; p0.0001) with median PFS being over three-fold the duration in patients treated with Pola BR (7.6 months [95% CI: 6.0, 17.0]) compared to BR (2 months [95% CI: 1.5, 3.7]).
The CR rate at the PRA based on positron emission tomography-computed tomography (PET-CT), as determined by the IRC, was higher in the Pola BR treatment arm (40.0% [16/40 patients]; 95% CI: [24.9%, 56.7%]) than the BR arm (17.5% [7/40 patients]; 95% CI: [7.3%, 32.8%]). The difference in CR rates between arms was
23 PTJA06 – Submission Dossier by Roche Registration GmbH
significant (22.5% in favor of Pola BR; 95% CI: [2.6%, 40.2%]; p 0.0261; Cochran Mantel-Haenszel [CMH] chi-square). The ORR (CR or partial response [PR]) at the PRA based on PET-CT as determined by the IRC was 45.0% in the Pola BR arm versus 17.5% in the BR arm (p 0.0069). Investigator-assessment of response by PET-CT at the PRA were highly consistent with the IRC-assessment. At the PRA, the INV-assessed CR rates were 42.5% in the Pola BR arm and 15.0% in the BR arm (p 0.0061).The ORRs, driven by CRs, were 47.5% and 17.5%, respectively. Computed tomography (CT)-based response assessments by the INV and by the IRC at the PRA also showed improved responses with Pola BR compared to BR alone. Concordance rates were high between the INV- and IRC-assessed OR, CR, and PR rates at the PRA.
In addition, IRC- and INV-assessed best overall response (BOR) of patients who were treated with Pola BR were higher compared with those who were treated with BR alone. A best response of CR or PR as assessed by the IRC was achieved by 62.5% of patients in the Pola BR arm and 25.0% of patients in the BR arm while on study. The proportion achieving a best response of CR was higher in the Pola BR arm (50.0% vs. 22.5%). A best response of CR or PR, as assessed by the INV, was achieved by 70.0% of patients in the Pola BR arm and 32.5% of patients in the BR arm while on study. The proportion achieving a best response of CR was higher in the Pola BR arm (57.5% vs. 20.0%).
The median IRC-assessed DOR was 12.6 months (95% CI 7.2, not evaluable [NE]) in the Pola BR arm compared to 7.7 months (95% CI 3.0, 18.9) in the BR arm. The median INV-assessed DOR was over two-fold the duration in the Pola BR arm (10.3 months; 95% CI: [5.6 months, NE]) compared to the BR arm (4.1 months; 95% CI: [2.6 months, 12.7 months]).
Safety In the GO29365 study, the combination of polatuzumab vedotin with BR was associated with additional toxicity as expected when an additional therapeutic agent is added to the BR combination. The safety profile of Pola BR was clinically manageable and in keeping with the known toxicities of the individual agents. A higher rate of Grade 34 cytopenias was observed with Pola BR compared with BR, but this did not result in a higher risk of infection or need for transfusion. This higher rate of AEs was likely contributed to by increased susceptibility in these heavily pre-treated patients, as well as disease progression and subsequent anti-lymphoma therapy in this high risk population of patients with R/R DLBCL. Peripheral neuropathy is a recognized toxicity associated with MMAE based ADCs, and was closely monitored during the GO29365 study. The majority of peripheral neuropathy observed was low grade and reversible, and led to few patients experiencing dose reduction or delay.
24 PTJA06 – Submission Dossier by Roche Registration GmbH
1.4 CONCLUSIONS In conclusion, polatuzumab vedotin in combination with BR showed compelling clinical benefit to patients compared to BR alone as observed by improved OS, PFS, and CR. The safety and tolerability of the Pola BR regimen was consistent with the known safety profiles of polatuzumab vedotin or BR, and in line with what might be expected in this heavily pre-treated patient population. Polatuzumab vedotin in combination with BR not only provides a new treatment option for patients with R/R DLBCL who are not candidates for SCT, but it also offers a new readily available treatment (as combination partners are easily accessible in general oncology practices) for patients with a high unmet medical need, who are rapidly progressing and in urgent need of R/R DLBCL control.
25 PTJA06 – Submission Dossier by Roche Registration GmbH
2. DESCRIPTION AND TECHNICAL CHARACTERISTICS OF THE TECHNOLOGY
Summary of the Characteristics of the Technology
In no more than 6 bullet points describe key statements about the technology and its regulatory status. For example, include statements that describe the key features of the technology and its authorization status.
Polatuzumab vedotin is the first in-class and first ADC in DLBCL. Polatuzumab vedotin is a CD79b-targeted ADC that delivers a potent anti-mitotic agent (MMAE) to B-cells, which results in anti-cancer activity against B-cell malignancies. Upon binding CD79b, polatuzumab vedotin is rapidly internalized and the linker is cleaved by lysosomal proteases, leading to the intracellular release of MMAE. The released MMAE binds to microtubules and kills dividing cells by inhibiting cell division and inducing apoptosis.
The recommended dose of polatuzumab vedotin is 1.8 mg/kg, given as an intravenous (IV) infusion every 21 days in combination with bendamustine (90 mg/m2/day on Day 1 and Day 2 of each cycle) and rituximab (375 mg/m2 on Day 1 of each cycle) for 6 cycles.
Polatuzumab vedotin was granted PRIority MEdicines (PRIME) designation for the treatment of R/R DLBCL on 23 June 2017 and Orphan Drug Designation (ODD) for the treatment of DLBCL on 16 April 2018 by the European Medicines Agency (EMA).
On 13 December 2018, accelerated assessment was agreed by the Committee for Medicinal Products for Human Use (CHMP) for the Marketing Authorization Application (MAA) for polatuzumab vedotin in combination with bendamustine and rituximab for the treatment of adult patients with R/R DLBCL who are not candidates for hematopoietic stem cell transplantation (HSCT). On 20 December 2018, Roche submitted the initial MAA for polatuzumab vedotin in the above-mentioned indication. On 10 June 2019, the United States (U.S) Food and Drug Administration (FDA) granted accelerated approval to polatuzumab vedotin in combination with BR for treatment of adult patients with R/R DLBCL, not otherwise specified, after at least two prior therapies.
On 14 November 2019, the CHMP adopted a positive opinion, recommending the granting of conditional marketing authorization for polatuzumab vedotin in combination with bendamustine and rituximab for the treatment of R/R DLBCL who are not candidates for HSCT. The EC decision is expected in January 2020.
26 PTJA06 – Submission Dossier by Roche Registration GmbH
2.1 CHARACTERISTICS OF THE TECHNOLOGY
In Table 2 provide an overview of the technology.
Table 2 Features of Polatuzumab Vedotin (POLIVY®)
Non-Proprietary Name Polatuzumab vedotin
Proprietary Name POLIVY® Marketing Roche Registration GmbH Authorization Holder Class CD79b-targeted ADC
Active Substance(s) Polatuzumab vedotin Pharmaceutical Powder for concentrate for solution for infusion. Formulation(s) White to greyish-white lyophilized cake. ATC Code L01XC37 Polatuzumab vedotin is a CD79b-targeted ADC that contains a humanized IgG1 anti-human CD79b mAb and a potent anti-mitotic agent, MMAE, linked through a cleavable linker, mc-vc-PAB.
Polatuzumab vedotin specifically binds human CD79b, a signaling component of the B-cell receptor located on the surface of B-cells. As such, CD79b expression is restricted to normal cells within the B-cell Mechanism of Action lineage (with the exception of plasma cells) and malignant B-cells; it is expressed in >95% of DLBCLs.
Upon binding CD79b, polatuzumab vedotin is rapidly internalized and the linker is cleaved by lysosomal proteases to enable intracellular release of MMAE. The released MMAE binds to microtubules and kills dividing cells by inhibiting cell division and inducing apoptosis. ADC: antibody-drug conjugate; ATC: Anatomical Therapeutic Chemical; CD79b: cluster of differentiation 79b; DLBCL: diffuse large B-cell lymphoma; IgG1: immunoglobulin G1; mAb: monoclonal antibody; mc-vc-PAB: maleimidocaproyl-valine-citrulline-ρ-aminobenzyloxycarbonyl; MMAE: monomethyl auristatin E. Source: (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019).
27 PTJA06 – Submission Dossier by Roche Registration GmbH
In Table 3 summarize the information about administration and dosing of the technology.
Table 3 Administration and Dosing of Polatuzumab Vedotin (POLIVY®) in combination with Bendamustine and Rituximab
Polatuzumab vedotin must be reconstituted and diluted using aseptic technique under the supervision of a healthcare Method of Administration professional. Polatuzumab vedotin should be administered as an IV infusion. BR should be administrated as an IV infusion Polatuzumab vedotin: 1.8 mg/kg Doses Bendamustine: 90 mg/m2 Rituximab: 375 mg/m2
Polatuzumab vedotin: Day 1 of each cycle (once every 21 days)
Dosing Frequency Bendamustine: Days 1 and 2 of each cycle
Rituximab: Day 1 of each cycle
Maximum Length of a Course of Six cycles Treatment Anticipated Average Interval None Between Courses of Treatments Anticipated Number of Repeat None Courses of Treatments The infusion rate of polatuzumab vedotin should be slowed or interrupted if the patient develops an infusion-related reaction. Polatuzumab vedotin should be discontinued immediately and permanently if the patient experiences a life-threatening reaction.
Peripheral Neuropathy (Grade 2-3): Hold polatuzumab vedotin dosing until improvement to ≤ Grade 1. If recovered to Grade ≤1 on or before Day 14, restart polatuzumab Dose Adjustments vedotin at a permanently reduced dose of 1.4 mg/kg. If a prior dose reduction to 1.4 mg/kg has occurred, discontinue polatuzumab vedotin. If not recovered to Grade ≤1 on or before Day 14, discontinue polatuzumab vedotin.
Peripheral Neuropathy (Grade 4): Discontinue polatuzumab vedotin.
28 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 3 Administration and Dosing of Polatuzumab Vedotin (POLIVY®) in combination with Bendamustine and Rituximab (cont.)
Myelosuppression Grade 3-4 Neutropenia: Hold all treatment until ANC recovers to >1000 μL. If ANC recovers to >1000 μL on or before Day 7, resume all treatment without any additional dose reductions. If ANC recovers to >1000 μL after Day 7: restart all treatment with a dose reduction of bendamustine from 90 mg/m2 to 70 mg/m2 or 70 mg/m2 to 50 mg/m2 if a bendamustine dose reduction to 50 mg/m2 has already occurred, discontinue all treatment.
Special Populations Hepatic Impairment: The administration and dosing of polatuzumab vedotin in patients Dose Adjustments with moderate or severe hepatic impairment (bilirubin greater than 1.5 ULN) should be avoided. No adjustment in the starting dose is required when administering polatuzumab vedotin to patients with mild hepatic impairment (bilirubin greater than ULN to less than or equal to 1.5 ULN or AST greater than ULN).
No dose adjustments of polatuzumab vedotin are required in the following special populations: Elderly patients ≥65 years Patients with CrCL ≥30 mL/min*
Pediatric Population: The efficacy and safety in children and adolescents <18 years have not been established. No data are available. ANC: absolute neutrophil count; AST: aspartate aminotransferase; BR: bendamustine plus rituximab; CrCL: creatinine clearance; IV: intravenous; ULN: upper limit of normal. * A recommended dose has not been determined for patients with CrCL <30 mL/min due to limited data. Source: (Levact - Article 29 referral 2011; POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019; Rituximab SmPC 2019).
29 PTJA06 – Submission Dossier by Roche Registration GmbH
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).
Polatuzumab vedotin in combination with BR must only be administered under the supervision of a healthcare professional experienced in the diagnosis and treatment of cancer patients with procedures for proper handling and disposal of antineoplastic medicines considered (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019). The care setting in the respective country will be chosen based on the ability to meet these requirements (usually hospitals and/or other outpatient settings).
Polatuzumab vedotin in combination with BR is a treatment regimen readily available and easily accessible in general oncology practices that may provide significant benefit for patients who have rapidly progressing disease and therefore, need urgent access to treatment.
Describe the equipment required to use the technology.
If all equipment is described in response to question 1, state here that there are no additional requirements.
Describe the supplies required to use the technology.
For example, syringes, needles, pharmaceuticals and contrast agents, fluids, bandages.
Polatuzumab vedotin must be reconstituted and diluted under the supervision of a healthcare professional. Polatuzumab vedotin must be reconstituted using sterile water for injection and diluted into an IV infusion bag containing sodium chloride 9 mg/ml (0.9%), sodium chloride 4.5 mg/ml (0.45%), or 5% glucose prior to administration. Polatuzumab vedotin must be administered as an IV infusion with use of syringes, a dedicated infusion line equipped with a sterile, non-pyrogenic, low-protein binding in-line or add-on filter (0.2 or 0.22 micrometer pore size) and catheter. Polatuzumab vedotin must not be administered as an IV push or bolus (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019). If not already pre-medicated, premedication with an antihistamine and anti-pyretic should be administered to patients prior to polatuzumab vedotin (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019).
State the claimed benefits of the technology, including whether the technology should be considered innovative.
For example, whether the technology has increased safety, health benefits, compliance and improved features of administration compared with existing technologies.
Polatuzumab vedotin is the first in-class and first ADC in DLBCL that has been developed specifically for B-cell malignancies that express CD79b. Diffuse large B-cell lymphoma expresses this marker ubiquitously making it a promising target for the
30 PTJA06 – Submission Dossier by Roche Registration GmbH
development of new therapies (Dornan et al. 2009; Pfeifer et al. 2015). Polatuzumab vedotin binds specifically to CD79b which results in ADC internalization and subsequent release of the potent microtubule inhibitor MMAE inside the targeted neoplastic B-cells. This process is thought to maximize tumor cell death while potentially minimizing the effects on normal healthy cells (Ducry et al. 2010). Antibody-based therapy has proven to be effective in DLBCL. A prominent example of this is rituximab, which has been incorporated into DLBCL SoC therapy (i.e. cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP]) and has both improved response rates and increased OS (Coiffier et al. 2002; Pfreundschuh et al. 2006).
Polatuzumab vedotin in combination with BR has distinct advantages over recently approved cell-based therapies. Polatuzumab vedotin provides a readily available “off-the-shelf” treatment option for patients as it is administered as an IV infusion and its combination partners are easily accessible in general oncology practices. Furthermore, no additional requirements are needed for its administration other than those already required for the administration of other conventional cancer treatments.
Polatuzumab vedotin in combination with BR provides a clinically-relevant advantage, with a highly improved efficacy without compromising safety over other available chemotherapy regimens. Based on data from the randomized Phase II GO29365 (NCT02257567) study, polatuzumab vedotin in combination with BR increased survival outcomes by 58% and extended the mOS to 12.4 months (95% CI: 9.0, NE) from 4.7 months (95% CI: 3.7, 8.3) with BR.
The potential benefit of polatuzumab vedotin in combination with BR as an innovative therapy option for a disease with limited treatment options, and thus a considerable unmet need has been recognized by both the EMA and the FDA. Polatuzumab vedotin has received ODD from both the EMA and FDA. The EMA has awarded PRIME designation and accelerated assessment. Similarly, the FDA granted polatuzumab vedotin breakthrough therapy designation (BTD) and a Priority Review. Both the FDA and the EMA acknowledge polatuzumab vedotin as a promising medicine that may offer a major therapeutic advantage over existing treatments and that targets an unmet medical need.
2.2 INVESTMENTS AND TOOLS REQUIRED Summary of Issues Relating to the Investments and Tools Required to Introduce the Technology
In no more than 6 bullet points describe key statements about the investments and tools required to use the technology.
For example, include statements about the equipment and resources required to use the technology and how this differs from the comparators.
31 PTJA06 – Submission Dossier by Roche Registration GmbH
For example, include statements about any new equipment, premises and personnel that will be required if the technology is introduced, or equipment, premises and personnel that will no longer be required.
No additional requirements are needed for the administration of polatuzumab vedotin in combination with BR other than those already required for the administration of other conventional cancer treatments. Procedures for proper handling and disposal of antineoplastic medicines should be considered and utilized.
2.2.1 Requirements to Use the Technology If any special conditions are attached to the regulatory authorization 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.
Interaction with other medicinal products and other forms of interaction
State any contraindications or groups for whom the technology is not recommended.
No additional requirements with regards to the settings for use are needed for the administration of polatuzumab vedotin other than those already required for the administration of other conventional cancer treatments.
Polatuzumab vedotin in combination with BR must only be administered under the supervision of a healthcare professional experienced in the diagnosis and treatment of cancer patients with procedures for proper handling and disposal of antineoplastic medicines considered (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019).
A summary of special warnings and precautions for use as well as patient monitoring for symptoms summarized below and full details are provided in the (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019):
Serious and severe neutropenia and febrile neutropenia have been reported in patients treated with polatuzumab vedotin as early as the first cycle of treatment. Prophylactic granulocyte colony-stimulating factor (G-CSF) administration should be considered. Grade 3 or 4 thrombocytopenia or anemia can also occur with
32 PTJA06 – Submission Dossier by Roche Registration GmbH
polatuzumab vedotin. Complete blood counts should be monitored prior to each dose of polatuzumab vedotin. More frequent lab monitoring and polatuzumab vedotin delays or discontinuation should be considered for patients with Grade 3 or Grade 4 neutropenia and thrombocytopenia (see Table 3).
Patients should be monitored for symptoms of peripheral neuropathy, such as hypoesthesia, hyperesthesia, paresthesia, dysesthesia, neuropathic pain, burning sensation, weakness, or gait disturbance. Patients experiencing new or worsening peripheral neuropathy may require a delay, dose reduction, or discontinuation of polatuzumab vedotin (see Table 3).
Patients should be closely monitored during treatment for signs of bacterial, fungal, or viral infections such as pneumonia and sepsis. Anti-infective prophylaxis should be considered. Polatuzumab vedotin and any concomitant chemotherapy should be discontinued in patients who develop serious infections.
Patients should be monitored closely for new or worsening neurological, cognitive, or behavioral changes suggestive of progressive multifocal leukoencephalopathy (PML). Polatuzumab vedotin and any concomitant chemotherapy should be held if PML is suspected and permanently discontinued if the diagnosis is confirmed.
Patients with high tumor burden and a rapidly proliferative tumor may be at increased risk of tumor lysis syndrome (TLS). Appropriate measures in accordance with local guidelines should be taken prior to treatment with polatuzumab vedotin. Patients should be monitored closely for TLS during treatment with polatuzumab vedotin.
Based on the mechanism of action (MoA) and nonclinical studies, polatuzumab vedotin can be harmful to the fetus when administered to a pregnant woman. Pregnant women should be advised regarding risk to the fetus. Female patients of reproductive potential should be advised to use effective contraception during treatment with polatuzumab vedotin and for at least 9 months after the last dose. Male patients with female partners of reproductive potential should be advised to use effective contraception during treatment with polatuzumab vedotin and for at least 6 months after the last dose.
Serious cases of hepatic toxicity that were consistent with hepatocellular injury, including elevations of transaminases and/or bilirubin, have occurred in patients treated with polatuzumab vedotin. Pre-existing liver disease, elevated baseline liver enzymes, and concomitant medicinal products may increase the risk. Liver enzymes and bilirubin levels should be monitored.
33 PTJA06 – Submission Dossier by Roche Registration GmbH
Based on physiological-based pharmacokinetic (PK) model simulations of MMAE released from polatuzumab vedotin, strong cytochrome P450 3A (CYP3A) inhibitors (e.g., ketoconazole) and inducers may alter the area under the curve (AUC) concentration time. Patients receiving concomitant strong CYP3A inhibitors should be monitored more closely for signs of toxicities. Unconjugated MMAE is not predicted to alter the AUC of concomitant medicines that are CYP3A substrates (e.g. midazolam).
In addition, the pharmacokinetics of rituximab and bendamustine are not affected by co-administration with polatuzumab vedotin. Concomitant rituximab is associated with increased antibody conjugated monomethyl auristatin E (acMMAE) plasma AUC by 24% and decreased unconjugated MMAE plasma AUC by 37%, based on PK population analysis. No dose adjustment is required. Bendamustine does not affect acMMAE and unconjugated MMAE plasma AUC.
Hypersensitivity to the active substance or to any of the excipients listed below are provided in the (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019):
Succinic acid Sodium hydroxide Sucrose Polysorbate 20
2.3 REGULATORY STATUS OF THE TECHNOLOGY If the technology is not approved include the information that is expected to be approved.
Complete Table 4 with the marketing authorization status of the technology.
State any other indications not included in the assessment for which the technology has marketing authorization.
List the other countries in which the technology has marketing authorization.
2.3.1 European Union Polatuzumab vedotin was granted a PRIME designation (EMA/668842/2016) for the treatment of R/R DLBCL on 23 June 2017 by the EMA. It was awarded PRIME designation based on its potential to offer a major therapeutic advantage over existing treatments, or benefit patients without treatment options (EMA/191104/2015). Polatuzumab vedotin was granted an ODD (ODD; EU/3/18/2013) by the EMA, for the treatment of DLBCL.
34 PTJA06 – Submission Dossier by Roche Registration GmbH
On 13 December 2018, accelerated assessment was agreed to by the CHMP for the MAA for polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL who are not candidates for HSCT.
On 20 December 2018, Roche submitted the initial MAA to the EMA for polatuzumab vedotin in the above-mentioned indication.
On 14 November 2019, the CHMP adopted a positive opinion, recommending the granting of conditional marketing authorization for polatuzumab vedotin in combination with bendamustine and rituximab for the treatment of R/R DLBCL who are not candidates for HSCT (CHMP Summary of Positive Opinion for POLIVY; EMA/CHMP/402512/2019). The EC decision is expected in January 2020.
2.3.2 United States On 12 December 2016, polatuzumab vedotin for the treatment of DLBCL was granted the US ODD by the FDA.
On 12 September 2017, the FDA granted a BTD for polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL who are not candidates for HSCT.
Roche initiated a rolling submission of the Biologics License Application (BLA) for polatuzumab vedotin in combination BR for the treatment of adult patients with DLBCL, not otherwise specified, who have received at least one prior therapy to the FDA with Part 1 (of 2) submitted on 20 November 2018. Part 2 of the rolling submission was submitted to the FDA on 19 December 2018.
On 15 February 2019, the FDA granted Priority Review and subsequently on 10 June 2019, granted accelerated approval for polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL, not otherwise specified, after at least two prior therapies (POLIVY™ US PI 2019).
35 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 4 Regulatory Status of Polatuzumab Vedotin (POLIVY)
Launched (yes/no). Organization (Expected) Date of Verbatim Wording of the Proposed Indication(s) If no, include proposed Issuing Approval approval date of launch
No. POLIVY® in combination with bendamustine and rituximab is Anticipated launch in first EMA indicated for the treatment of adult patients with relapsed/refractory January 2020 EU countries: January DLBCL who are not candidates for HSCT*. 2020
POLIVY is a CD79b-directed antibody–drug conjugate indicated in combination with bendamustine and a rituximab product for the FDA treatment of adult patients with relapsed or refractory diffuse large June 2019 Yes B-cell lymphoma, not otherwise specified, after at least two prior therapies
ADC: antibody-drug conjugate; BR: bendamustine and rituximab; CD79b: cluster of differentiation 79b; DLBCL: diffuse large B-cell lymphoma; EMA: European Medicines Agency; EU: European Union; FDA: Food and Drug Administration; HSCT: hematopoietic stem cell transplantation; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; US: United States * At the request of the CHMP during the List of Questions (LoQs), the wording of the EU proposed indication was revised to state “relapsed/refractory” instead of “previously treated” to better clarify the intended population to be treated with polatuzumab vedotin in combination with BR. Source: (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019; POLIVY™ US PI 2019).
36 PTJA06 – Submission Dossier by Roche Registration GmbH
2.4 CURRENT USE OF THE TECHNOLOGY In no more than 6 bullet points describe key statements about the current use of the technology.
For example, include statements about the availability and reimbursement status of the technology in other countries, the populations in which the technology is currently used (if available).
Complete only if the technology is available in one or more European countries.
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) authorization.
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.
Polatuzumab vedotin in combination with BR is approved for marketing in the U.S.
As of June 2019, polatuzumab vedotin is not approved for marketing, for any indication, in the EU.
3. HEALTH PROBLEM AND CURRENT CLINICAL PRACTICE
Summary of Issues Relating to the Health Problem and Current Clinical Practice
In no more than 6 bullet points describe key statements about the health problem and current clinical practice.
For example, include statements about the proposed use or target population, unmet needs of treatment and how the technology may address these.
There are no universally established SoC regimens or regulatory-approved therapies upon first relapse, as no prior randomized trials have established the superiority of one regimen over another for the treatment of R/R DLBCL patients. Thus, the treatment landscape for R/R DLBCL patients who are not candidates for SCT is highly fragmented, with no preferred treatment. It is important to note that a number of treatments utilized in the real-world setting are not specifically indicated for R/R DLBCL treatment due to the lack of appropriate evidence.
The greatest unmet need in DLBCL is the effective management of patients with R/R DLBCL ineligible for SCT, as mOS is approximately 6 months despite salvage chemotherapy.
37 PTJA06 – Submission Dossier by Roche Registration GmbH
3.1 OVERVIEW OF THE DISEASE OR HEALTH CONDITION 3.1.1 Disease Overview Define the disease or health condition in the scope of this assessment.
If available include a standardized code such as the ICD code or the DSM code (and the version of the code).
Non-Hodgkin’s lymphoma is the most prevalent hematological malignancy and consists of a heterogeneous group of lymphoproliferative disorders that arise from the lymphoid system (Galaznik et al. 2018; Nogai et al. 2011). Approximately 85% to 90% of NHLs emerge from B-cells, with the remainder stemming from T-cells and natural killer cells (Galaznik et al. 2018; Armitage et al. 2017; Swerdlow et al. 2016). More than 60 different NHL subtypes have been identified and classified by the World Health Organization (WHO) (Swerdlow et al. 2016). Non-Hodgkin’s lymphoma subtypes are categorized by the characteristics of the lymphoma cells, including their morphology, immune-phenotype and their genetic features (Swerdlow et al. 2016).
Diffuse large B-cell lymphoma (International Classification of Diseases (ICD) 10 code: C83.3) is the most common histologic subtype of NHL, accounting for 30% to 48% of all NHL cases and 60% of aggressive lymphomas (Al-Hamadani et al. 2015; Leukemia & Lymphoma Society 2019; Smith et al. 2015; The Non-Hodgkin’s Lymphoma Classification Project 1997). The natural behavior of the aggressive lymphomas, such as DLBCL, is characterized by faster progression and reduced survival compared to indolent NHL (van de Schans et al. 2014).
While DLBCL typically occurs in those over 60 years of age, it can occur in patients of any age, including young adults and children (Le Guyader-Peyrou et al. 2017). The incidence of DLBCL increases with age, with a median age at diagnosis between 64 to 74 years (Armitage et al. 1998; Smith et al. 2015). Elderly DLBCL patients have a poorer prognosis and inferior outcomes than younger DLBCL patients, even with comparable treatment (Kuhnl et al. 2017).
3.1.2 Symptoms and Burden of the Disease Describe the symptoms and burden of the disease or health condition for patients.
Include aspects such as pain, disability, psychosocial issues, or other determinants of morbidity and quality of life from a patient perspective.
Although DLBCL is often asymptomatic, it may be associated with constitutional symptoms, such as fever, recurrent night sweats, weight loss (known as B-symptoms) in approximately 30% of patients, and/or local effects of lymph node enlargement, as well as those of bone marrow failure (Armitage et al. 1998). It is marked by rapidly growing tumors in the lymph nodes, spleen, liver, bone marrow, or other organs. The majority of
38 PTJA06 – Submission Dossier by Roche Registration GmbH
cases (60%) originate in the lymph nodes with the remaining 40% presenting in extranodal sites (Moller et al. 2004). The most common extranodal site is the gastrointestinal tract, but DLBCL can occur in any organ, including the skin, central nervous system, bone marrow, salivary gland, lung, kidney, and liver (Harris et al. 1994; van Baarlen et al. 1988). Bone marrow involvement is found in 11% to 27% of all cases. However, DLBCL rarely infiltrates the peripheral blood (Gouveia et al. 2012).
Tumors typically arise as primary, de novo, or may result from a transformation of an indolent lymphoma (Amin et al. 2017; Gouveia et al. 2012). Hereditary and acquired immunodeficiencies, occupational exposures, and pharmacological immunosuppression in the setting of transplantation or treatment of autoimmune diseases have been identified as factors thought to potentially confer an increased risk of developing DLBCL (Friedberg et al. 2008).
For the vast majority of patients, the etiology of DLBCL is unknown. As for other cancers, DLBCL involves patterns of multifactorial pathogenesis with environmental factors, as well as genetic, occupational and dietary factors, contributing to its development (Gouveia et al. 2012).
The health-related quality of life (HRQoL) for DLBCL patients is impacted by treatment-related AEs. Recent analyses indicate a substantial impact on HRQoL following SCT when approximately 35% of patients returned to work, 14% retired, and 23% died. Currently, there are no HRQoL instruments specific to DLBCL (Lenz et al. 2018).
3.1.3 Classification of the Disease and Prognosis If relevant describe the main subtypes and/or stages of the disease or health condition.
Include any prognostic factors that may affect the course of the disease or health condition.
Diffuse large B-cell lymphoma can be classified by one of four disease stages according to the Ann Arbor (AA) and/or Lugano Staging Classification (see Table 5 and Table 6) (Cheson et al. 2014; Lister et al. 1989; Tilly et al. 2015). The AA staging was the first system widely utilized to stage NHL. It is based on the distribution and number of nodes or organs involved within the lymphoma, the presence or absence of extranodal involvement, and constitutional symptoms. The AA classification system is used in clinical practice to summarize a patient’s clinical stage and to guide treatment (Carbone et al. 1971). In accordance with the Lugano classification, the ESMO recommends that patients be staged by this tool based on results from the 18fluorodeoxyglucose positron emission tomography (FDG-PET)/CT, and considers this type of procedure as the standard for staging DLBCL patients (Tilly et al. 2015). However, the availability of an increasing number of biologic agents with distinct MoAs requires flexibility in
39 PTJA06 – Submission Dossier by Roche Registration GmbH
interpretation of the recommendations to account for these agents’ biologic or immunomodulatory properties (Cheson et al. 2016).
In addition to staging classifications, the International Prognostic Index (IPI) and the age-adjusted International Prognostic Index (aaIPI), an updated version of IPI scoring, are used for determining initial prognosis among patients with DLBCL at the time of diagnosis. The IPI and aaIPI consists of five clinical predictors (Table 7) that are used to assign patients to one of four different risk categories (Tilly et al. 2015). Patients with high IPI scores (45) at the initial DLBCL diagnosis are considered as high-risk patients and have lower survival rates compared to patients with a low IPI score (01) (3-year OS rate; IPI-high score 59% [95% CI: 49% to 69%], a IPI low score 91% [95% CI; 89% to 94%]) (Ziepert et al. 2010). At relapse, and in the case of refractory disease, routine biological tests and evaluation of comorbidities are once again conducted to classify patients according to the IPI and whether they are eligible for SCT (Gisselbrecht et al. 2018). This secondary IPI represents the main survival prognostic factor in R/R DLBCL as patients with a high IPI score of 2 to 3 have a lower survival rate compared with patients with a low secondary IPI score of 0 to 1 (4-year OS; IPI high 45% versus IPI low 72%; p<0.001) (Gisselbrecht et al. 2012). The utility and value of the IPI score in the R/R transplant ineligible DLBCL patient setting and/or beyond the second-line setting, is not clear due to scarcity of data for this subpopulation.
Table 5 Ann Arbor Staging Classification
Stage Involvement of a single lymphatic region (I) or localized involvement of single I extralymphatic organ or site (IE) Involvement of two or more lymphatic regions on the same side of the diaphragm (II) II or localized involvement of a single extralymphatic organ or site and of one or more lymphatic regions on the same side of diaphragm (IIE) III Involvement of lymphatic regions on both sides of the diaphragm Diffuse or disseminated involvement of one or more extralymphatic organs with or IV without lymphatic involvement Source: (Tilly et al. 2015).
40 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 6 Lugano Staging Classification
Stage Involvement Extranodal Status Limited One node or a group of adjacent Single extranodal lesions without nodal Stage I nodes involvement Two or more nodal groups on the Stage I or II by nodal extent with limited Stage II same side of the diaphragm contiguous extranodal involvement Stage II bulky* II as above with “bulky” disease Not applicable Advanced Nodes on both sides of the Stage III diaphragm; nodes above the Not applicable diaphragm with spleen involvement Additional noncontiguous Stage IV Not applicable extralymphatic involvement Note: Extent of disease is determined by positron emission tomography-computed tomography for avid lymphomas and computed tomography for non-avid histologies. Tonsils, Waldeyer’s ring, and spleen are considered nodal tissue. * Whether Stage II “bulky” disease is treated as limited or advanced disease may be determined by histology and a number of prognostic factors. Source: (Cheson et al. 2014).
Table 7 International Prognostic Index
International Prognostic Index (IPI) Risk factors Age 60 years Serum LDH normal Stage IIIIV Performance status 24 Extranodal involvement 1 site Risk categories Low 01 Lowintermediate 2 Highintermediate 3 High 45 Age-adjusted International Prognostic Index (aaIPI) in Patients 60 years Risk factors Serum LDH normal Stage IIIIV Performance status 24 Risk Categories Low 0 Lowintermediate 1 Highintermediate 2 High 3 LDH: lactate dehydrogenase Source: (Tilly et al. 2015).
41 PTJA06 – Submission Dossier by Roche Registration GmbH
3.1.4 Epidemiology Present an estimate of prevalence and/or incidence for the disease or health condition including recent trends.
The age-adjusted incidence rate of DLBCL in Europe and the US is between 3.1 and 5.7 cases per 100,000 population per year, respectively (Cerhan et al. 2014). Given that the estimated number of new cases of NHL worldwide in 2018 was approximately 510,000 (Bray et al. 2018; IARC 2018), and approximately 30% to 48% of NHL cases are DLBCL (Al-Hamadani et al. 2015; Smith et al. 2015; The Non-Hodgkin’s Lymphoma Classification Project 1997), one can reasonably assume that DLBCL will account for approximately 153,000 to 245,000 of these cases. In the EU (28 countries), it is estimated that in 2018 there were 97,000 new cases of NHL (European Cancer Information System; Ferlay et al. 2018), of which it can be estimated that approximately 29,000 to 46,000 of these cases were DLBCL.
The epidemiology of R/R DLBCL remains poorly defined. Historically, a common approach for estimation has been to rely on clinical experience and inferred impacts based on the PARMA and the Collaborative Trial in Relapsed Aggressive Lymphoma (CORAL) clinical trials (Friedberg 2011). However, there has been a concerted effort to model the likely number of patients based on population-based registries. In the United Kingdom (UK), the Haematological Malignancy Research Network (HMRN) is considered the gold standard for data collection and registration for hematological disorders (Smith et al. 2011). In a recent cost-effectiveness analysis, treatment patterns for DLBCL were estimated using HMRN complete clinical trial and treatment registry data for patients diagnosed with DLBCL in 2007 (Wang et al. 2017). Of the 4,880 expected patient diagnoses in 2014, 80% (n 3,892) were estimated to be treated at first-line. This is broadly consistent with an observational study conducted using the Swedish Lymphoma Registry for patients diagnosed between 2007 to 2013, in which 83% of diagnosed patients were treated at first-line (72% receiving treatment with “curative” intent and 11% receiving treatment with “palliative” intent) (Wasterlid et al. 2019). Of the HMRN’s first-line treated patients, 15% (n 577) were estimated to receive second-line treatment (71% non-autologous stem cell transplantation [ASCT]-treated versus 29% ASCT-treated), while 18% (n 106) of second-line patients treated were estimated to receive third-line treatment. The estimated proportion of second-line cases receiving treatment is similar to the 23% (244 out of 1,039) observed among patients diagnosed between 2002 to 2012 at the University of Iowa Mayo Clinic in the U.S (Farooq et al. 2017), in which 33% were estimated to receive SCT (see Table 8).
42 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 8 DLBCL Treatment Patterns Across Developed Countries
UK Sweden US US EU5, US Roche Internal (Qunaj et al. (Friedberg 2011) Source (Wang et al. 2017) (Wasterlid et al. 2019) (Farooq et al. 2017) Estimates 2018) Review (modelled), 2019 Treatment line First-line 80%a 83% 81% 81% Second-line: First-lineb 15% 23% 33% 21% SCT 29% 33% 25% 25% Non-SCT 71% 67% 75% 75% Third-line: Second-linec 18% 33% DLBCL: diffuse large B-cell lymphoma; EU5: France, Germany, Italy, Spain, the UK; SCT: stem cell transplantation UK: United Kingdom; U.S: United States a Curative intent. b Ratio of second-line treated to first-line treated. c Ratio of third-line treated to second-line treated.
43 PTJA06 – Submission Dossier by Roche Registration GmbH
Based on the broad and varied pattern of care received for DLBCL described in Table 8, Roche developed a calibrated incidence-based model to estimate the epidemiology of patients who receive treatment after first-line who are refractory to or relapsed after treatment. The model structure and parameter estimates were based on a literature review and an amalgamation of observed treatment patterns from population-based registries, along with clinical trial data. In 2019, it is estimated that there are approximately 504 million people in the EU28. Of these, it is estimated that approximately 43,000 will be newly diagnosed with DLBCL. Of these, 35,000 patients receive an intervention with ‘curative intent’. The remaining 8,000 patients who receive ‘non-curative treatment’ are likely to be treated with rituximab alone, monochemotherapy, or best supportive care (BSC). The prognosis for these patients is extremely poor and is assumed to reflect palliative intent. After receipt of first-line treatment with “curative intent” (n 35,000), it is estimated that the number of patients treated at second-line in 2019 is approximately 7,300 patients. The estimated number of patients treated at third-line in the same year is estimated to be approximately 2,400 patients, while the number of patients treated at fourth-line or later is approximately 780 patients. This gives an overall estimate of 10,500 patients who could receive treatment for R/R DLBCL in 2019 (see Figure 1).
Figure 1 Estimated Number of R/R DLBCL Patients Treated in Each Line in the EU
1L: first-line; 2L: second-line; 3L: third-line; 4L: fourth-line; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. * Non-curative treated patients received palliative interventions.
Based on the current patterns of care, and with the predicted changes in the demographics of the general population (including increasing age), the estimated number of R/R DLBCL patients treated in 2025 in the EU is expected to be approximately 11,800 (see Figure 2).
44 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 2 Estimated Trend for DLBCL Incidence and Treated R/R Population in the EU28
Incidence RR DLBCL 60.000
50.000
40.000
30.000
20.000 Frequency Frequency (n)
10.000
- 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 Year
RR DLBCL: relapsed/refractory diffuse large B-cell lymphoma.
3.2 CLINICAL MANAGEMENT OF THE DISEASE OR HEALTH CONDITION Describe the clinical pathway of care for different stages and/or subtypes of the disease being considered in the assessment.
The clinical management of DLBCL is stage dependent and relies on several diagnostic procedures. Treatment strategies are stratified according to age, IPI score, and feasibility of dose-intensified approaches (Tilly et al. 2015).
3.2.1 First-line DLBCL Treatment The recommended first-line treatment for patients with DLBCL is typically an anthracycline-based immuno-chemotherapy typically consisting of 6 to 8 cycles of R-CHOP (Amin et al. 2017; Coiffier et al. 2010; Galaznik et al. 2018; Tilly et al. 2015; Farooq et al. 2017). The R-CHOP regimen has improved DLBCL patient survival over the last decade (Le Guyader-Peyrou et al. 2017), increasing patient survival by approximately 15% (Galaznik et al. 2018). Recently however, multiple clinical trials failed to provide further survival benefit for first-line DLBCL treatment (Robust (NCT02285062), Phoenix (NCT01855750), Remarc (NCT01122472), Goya (NCT01287741), REMODL-B (NCT01324596), Alliance/CALGB 50303 (NCT00118209). Consequently, there have been no new approvals for first-line DLBCL since the introduction of the R-CHOP regimen (Amin et al. 2017; Galaznik et al. 2018).
45 PTJA06 – Submission Dossier by Roche Registration GmbH
Patients with DLBCL who were event-free 24 months after therapy induction are considered long-term survivors (NICE 2018a; Maurer et al. 2014; NICE 2018b). Approximately 60% of newly diagnosed patients may be cured with R-CHOP (Vitolo et al. 2017; Sehn et al. 2015). Nevertheless, despite improvements in the survival of DLBCL patients with the R-CHOP regimen, approximately one-third of patients present with either a primary refractoriness to the first-line treatment used (mainly R-CHOP) or relapse after reaching a CR which remains a major cause of morbidity and mortality of this disease (Friedberg 2011).
3.2.2 Treatment of R/R DLBCL Relapsed DLBCL is characterized by the appearance of any new lesion after a documented history of response, while refractory DLBCL is characterized with PD or no response from the start of previous treatment (Gisselbrecht et al. 2018).
Treatment options for patients with R/R DLBCL include chemotherapy/immunotherapy and SCT. Based on the general fitness of the patient, patients are either subjected to non-aggressive or aggressive salvage treatment. There is limited clinical evidence on the real-world outcomes of these treatments (Galaznik et al. 2018). Available data indicate that the majority of patients (59%) do not proceed to SCT and these patients are characterized with poor 2-year OS compared to patients undergoing SCT (Farooq et al. 2017).
3.2.2.1 Treatment of R/R DLBCL Patients Eligible for SCT High-dose chemotherapy followed by SCT is the only potentially curative option, but is only available for younger, fit patients with chemosensitive disease (Tilly et al. 2015). The decision to recommend and proceed with SCT is complex and multifaceted and the relative contributions of these characteristics (which may overlap) to SCT success has not been clearly defined (Tay et al. 2019). Improvement of the response rates to salvage chemotherapy may, therefore, increase the number of patients who can proceed to SCT. Patients who are eligible for SCT and have chemosensitive disease, are treated with chemotherapy regimens, such as rituximab, dexamethasone, cytarabine, cisplatin (R- DHAP), rituximab, ifosfamide, carboplatin, and etoposide (R-ICE) or rituximab, gemcitabine, cisplatin, and dexamethasone (R-GDP) followed by high-dose chemotherapy and SCT (Raut et al. 2014; Tilly et al. 2015). Wide variations in OS outcomes for R/R DLBCL patients are observed, dependent on patient characteristics and line of treatment. In studies, including patients eligible for transplantation, mOS was 9.9 to 44.0 months (Lenz et al. 2018). However, not all patients are fit or eligible for this therapeutic option mostly due to poor performance status, age or organ dysfunction (Tilly et al. 2015).
3.2.2.2 Treatment of R/R DLBCL Patients Ineligible for SCT Among patients treated with the intent to proceed to SCT, less than 50% will proceed to SCT, mainly because of an insufficient response to second-line treatment and/or overall
46 PTJA06 – Submission Dossier by Roche Registration GmbH
unfit or frail performance status after salvage (Gisselbrecht et al. 2018). These patients are subsequently considered as ineligible for SCT.
Patients who are ineligible for SCT due to age or the presence of co-morbidities have little chance at prolonged control of disease with a dismal survival outcome (Galaznik et al. 2018). The treatment options for these patients are limited with the EMSO Guidelines recommending participation in clinical trials with novel agents or treatment with platinum- and/or gemcitabine-based regimens with or without rituximab (Tilly et al. 2015).
From a clinical perspective, SCT ineligibility represents a major therapeutic challenge due to the limited treatment options and lack of opportunity to treat R/R DLBCL SCT ineligible patients with curative intent (Gisselbrecht et al. 2018; Tilly et al. 2015). Current treatment guidelines recommend chemotherapy-based regimens and enrollment in clinical trials with novel MoAs. Consequently, the treatment landscape for R/R DLBCL patients who are ineligible for SCT is highly fragmented regardless of line of therapy (Figure 3). For those patients, the goal of the therapy is to achieve a prolonged response and ultimately extend survival. Due to overlapping disease characteristics and limited treatment options, patients upon first- and later-relapse are described as a patient segment with a uniformly high unmet need due to little chance at prolonged control of the disease with a dismal survival outcome (Galaznik et al. 2018; Gisselbrecht et al. 2018).
Patients with PD after three successive treatment regimens also have a poor prognosis and are unlikely to derive any further benefit from currently available therapies (Galaznik et al. 2018) and are recommended to enroll in clinical trials mostly due to the lack of established benefit of current treatments (Tilly et al. 2015).
Treatment with salvage regimens is recommended for those who had adequate treatment with rituximab and anthracycline-containing first-line therapies. Multiple treatment options, including R-GemOx, BR, R-DHAP, R-ICE, and R-GDP are all used off-label (see Section 3.5.1) (Gisselbrecht et al. 2018; Tilly et al. 2015). Albeit off-label, both R-GemOx and BR are frequently used in the real-world setting (see Figure 3) and are recommended in the ESMO and/or NCCN Clinical Guidelines (see Table 9) for the treatment of R/R DLBCL patients who are ineligible for SCT. In addition to initial prognostic factors, the nature of previous treatments and time from initial treatment are of utmost importance for the therapeutic management of R/R DLBCL (Tilly et al. 2015).
In terms of usage, primary market research conducted by Roche (Roche Data on File 2017) identified R-GemOx and BR as the most frequently used treatment options for R/R DLBCL patients ineligible for SCT after first relapse.
47 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 3 Highly Fragmented Treatment Landscape for Transplant Ineligible Patients with R/R DLBCL in the EU 2L 3L
2L: second-line; 3L: third-line; CD20: cluster of differentiation 20; CHOP: cyclophosphamide, doxorubicin, vincristine, and prednisone CHP: cyclophosphamide, doxorubicin, and prednisone; GemOx: gemcitabine and oxaliplatin; DHAP: dexamethasone, cytarabine, cisplatin; DSHAP: Division of State HIV/AIDS Programs; EPOCH: etoposide, prednisolone, oncovin, cyclophosphamide, hydroxydaunorubicin Hyper-CVAD: hyperfractionated cyclophosphamide, vincristine, doxorubicin, and prednisolone; ICE: Ifosfamide, carboplatin, and etoposide * In combination with anti-CD20 therapy Source: (Roche Data on File 2017).
In conclusion, poor outcomes are observed among patients who are ineligible for SCT (mOS: 6 to 11 months) or have refractory disease (mOS: 6.1 to 7.1 months) after any line of treatment (Arcari et al. 2016; Crump et al. 2017; Czuczman et al. 2017). For patients receiving third-line salvage therapy, OS was also shorter compared with those receiving second-line salvage therapy, ranging from 5.9 to 8.0 months versus 11 to 17.2 months (Lenz et al. 2018).
More recently, in 2018, two CD19s directed CAR-T therapies were approved (based on single-arm studies) in the EU for the treatment of patients who have had two or more lines of prior therapies based on single-arm studies. While these therapies have shown efficacy with potentially durable CRs, their use may be limited for the general R/R DLBCL population due to the toxicity profile which requires carefully selected patients and treatment centers with specially trained staff. Manufacturing potential in certain settings and curative potential are uncertain since long-term follow-up data are not yet available. The results from the ZUMA-1 and JULIET studies demonstrated that approximately half of patients achieve CR in the CAR-T-infused population (ZUMA-1:
48 PTJA06 – Submission Dossier by Roche Registration GmbH
51% CR population at 12 months [mITT population, defined as all patients who received YESCARTA®] (Yescarta SmPC) and JULIET: 40% CR at 3 months or more analysis [Efficacy Analysis set, defined as all patients who received Kymriah®] (Schuster et al. 2019). Consistency, quality, and dose of autologous cell-based therapies remain non-trivial and potentially limit the broad adoption of CAR-T therapies beyond specialized centers. In addition, the processing time and need for additional safety management measures to minimize significant toxicities, such as Cytokine Release Syndrome (CRS) and encephalopathy that are associated with this therapy, may be prohibitive to a substantial proportion of patients with rapidly progressive disease who may not have time to wait for infusion (leukaphersis) scheduling or manufacturing time of CAR-T which is in the range of several weeks (Kymriah SmPC; Yescarta SmPC).
Pixantrone was approved for the treatment of adult patients with multiply R/R aggressive Non-Hodgkin’s B-cell lymphomas (Pixuvri SmPC). Pixantrone initially received conditional approval for multiply relapsed/refractory aggressive NHL, which was subsequently converted to full marketing approval in March 2019 based on the efficacy and safety results from the randomized Phase III, PIX301 study. The benefit of pixantrone treatment has not been established in patients when used as a fifth-line treatment or beyond in patients who are refractory to last therapy. Pixantrone data from the pivotal randomized Phase III, PIX306 study were recently presented at the ASH Annual Meeting in 2018. While the primary endpoint of the PIX306 study was not met, subgroup analysis of PFS and OS results in third-line and beyond were comparable when indirectly compared to PIX301 (Pixuvri Press Release 2019). Pixantrone use is not widespread due to insufficient evidence for efficacy and lack of benefit (Tilly et al. 2015; G-BA 2013; HAS 2016).
Single-agent chemotherapies, such as bendamustine or pixantrone may be considered for elderly transplant ineligible patients diagnosed with R/R DLBCL. In addition, dose-attenuated rituximab-based regimens or less intense regimens are recommended (bendamustine, R-GemOx, pixantrone or dose-attenuated R-DHAP, rituximab, etoposide, methylprednisolone, cytarabine, cisplatin [R-ESHAP], R-ICE) (Buske et al. 2018).
Include a list of relevant guidelines. Table 9 provides a suggested presentation when there are multiple relevant guidelines.
Include a diagram of the care pathway. When there are significant variations in care, more than one diagram may be required.
49 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 9 Relevant Guidelines for Diagnosis and Management of R/R DLBCL Patients Ineligible for SCT
Date of Country(ies) Name of Summary of Recommendations* Issue or to which Society/Organization Last Guideline Issuing Guidelines (Level of Evidence/Grade of Recommendation for the Indication Under Assessment) Update Applies
For a first relapse or progression, patients eligible for a transplant are recommended for treatment with platinum-based chemotherapy regimens (i.e. R-DHAP, R-ICE, R-GDP, mostly I A) as salvage treatment. If patients are chemosensitive, R-HDCT with ASCT is recommended as remission consolidation (II A). Allogeneic transplantation should be considered in patients who have relapsed after R-HDCT with ASCT or in patients with poor risk factors at relapse (III B). Patients not suitable for high-dose therapy may be treated with the same or other salvage regimens as R-GemOx [III, B]. Pixantrone, a new anthracycline- like drug with reduced cardiotoxicity, demonstrated some efficacy in heavily treated patients ESMO 2015 EU [II, C]. However, these patients should be enrolled in clinical trials testing the activity of other novel drugs. The recommended treatment options for SCT ineligible R/R patients encompass participation in clinical trials with novel agents or treatment with platinum and/or gemcitabine-based regimens with or without rituximab.
Patients who have two or more relapses and are eligible for a transplant should be considered for allogeneic transplantation or clinical trials with novel drugs. Patients who are ineligible for a transplant should be considered for clinical trials with novel drugs or palliative care (Tilly et al. 2015).
In R/R disease, the NCCN Guidelines differentiate between patients who have the intention to proceed to high-dose therapy and those who are not suitable for high-dose therapy. For patients who are candidates for high-dose therapy, as in the European guidelines, high-dose therapies include DHAP, ICE, and GDP, whereas therapies that are suited to patients who NCCN 2019 US are ineligible for high-dose treatments, include bendamustine with the option of rituximab, GemOx with the option of rituximab, or rituximab monotherapy (NCCN 2019).
Provided recommendations are category 2A: Evidence from at least one large randomized, controlled trial of good methodological quality (low potential for bias)
50 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 9 Relevant Guidelines for Diagnosis and Management of R/R DLBCL Patients Ineligible for SCT (cont.)
Date of Country(ies) Name of Summary of Recommendations* Issue or to which Society/Organization Last Guideline Issuing Guidelines (Level of Evidence/Grade of Recommendation for the Indication Under Assessment) Update Applies
For patients with R/R DLBCL, single-agent therapy and inclusion in clinical trials are recommended for patients ineligible for high-dose chemotherapy and/or ineligible for SCT. Linea Guida 2018 Italy Combination regimens include R-GemOx or bendamustine +/-rituximab and CVP +/- rituximab. lenalidomide +/- rituximab in in non-GCB patients may be considered
Sociedad Española For the R/R DLBCL patients ineligible for SCT, the following treatment options are de Hematologia y recommended: 2016 Spain Hemoterapia R-GemOx; GELTAMO Guidelines BR treatment option with acceptable safety profile and relatively satisfactory efficacy
Dutch Association for The For the R/R DLBCL patients ineligible for SCT, R-PECC (prednisone, etoposide, 2018 Hematology Netherlands chlorambucil and lomustine) therapy is recommended- draft version of guidelines
For patients who are not candidates for autologous or allogeneic transplant, a curative Deutsche concept is reasonable if the interval between primary diagnosis and relapse is long and the Gesellschaft für disease responds to salvage chemoimmunotherapy. In addition to R-DHAP and R-ICE, the Hämatologie und 2018 Germany R-GemOx and R-ESHAP/ASHAP regimens can be considered. If the time between Medizinische diagnosis and relapse is short or the patient has chemorefractory disease, a palliative Onkologie concept should be preferred. For patients with multiple relapses, pixantrone is available. There is no restriction regarding patient age for CAR-Ts
51 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 9 Relevant Guidelines for Diagnosis and Management of R/R DLBCL Patients Ineligible for SCT (cont.)
Date of Country(ies) Name of Summary of Recommendations* Issue or to which Society/Organization Last Guideline Issuing Guidelines (Level of Evidence/Grade of Recommendation for the Indication Under Assessment) Update Applies Salvage therapy with multi-agent immunochemotherapy is recommended to R/R DLBCL fit enough to tolerate intensive therapy. The goal of this approach is to obtain sufficient response to allow consolidation with SCT but is also beneficial even if not followed by transplantation. In addition, consider R-GDP immunochemotherapy, which is as effective as other commonly used salvage regimens and less toxic Kymriah, tisagenlecleucel therapy is recommended for use within the Cancer Drugs Fund as National Institute for an option for treating R/R DLBCL in adults after two or more systemic therapies Health and Care 2018 UK Yescarta®, axicabtagene ciloleucel therapy is recommended for use within the Cancer Drugs Excellence Fund as an option for treating R/R DLBCL or primary mediastinal large B-cell lymphoma in adults after two or more systemic therapies Pixantrone monotherapy is recommended as an option for treating adults with multiple relapsed or refractory aggressive non-Hodgkin's B-cell lymphoma only if: the patient has previously been treated with rituximab and the patient is receiving third- or fourth-line treatment
The majority of favored salvage regimens in first relapse include either one or both of a platinum agent or ifosfamide, and there is no clearly superior regimen. Transplant-eligible patients should receive intensive salvage chemotherapy with a non-cross resistant regimen followed by SCT consolidation in those achieving CR. In those achieving a PR, second-line salvage chemotherapy can be given followed by ASCT if CR is achieved, or consolidation by ASCT in PR can be considered
NHL 2016 UK Elderly, frail or unfit patient with DLBCL The side effect profile of R-CHOP worsens with increasing comorbidity, functional disability and advancing age. Thus, in practice, only a minority of patients are treated with full dose R- CHOP. Modified R-CHOP with dosage and/or individual drug adjustments should be considered for those unfit for standard treatment. There is no uniformly accepted age cut-off to define an ‘elderly’ patient as it is applied variably to those aged >60, >65 or >70 years in different studies, with patients >80 years being classified as ‘very elderly’ thus the treatment option depends on performance status.
52 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 9 Relevant Guidelines for Diagnosis and Management of R/R DLBCL Patients Ineligible for SCT (cont.)
ESMO levels of evidence and grades of recommendation
Levels of evidence I Evidence from at least one large randomized, controlled trial of good methodological quality (low potential for bias) or meta-analyses of well- conducted randomized trials without heterogeneity II Small randomized trials or large randomized trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or of trials with demonstrated heterogeneity III Prospective cohort studies IV Retrospective cohort studies or case-control studies V Studies without control groups, case reports, expert opinions
Grades of recommendation A Strong evidence for efficacy with a substantial clinical benefit, strongly recommended B Strong or moderate evidence for efficacy but with a limited clinical benefit, generally recommended C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs), optional D Moderate evidence against efficacy or for adverse outcome, generally not recommended E Strong evidence against efficacy or for adverse outcome, never recommended ASCT: autologous stem cell transplantation; CAR-T: chimeric antigen receptor T-cell; CR: complete response; CVP: cyclophosphamide, vincristine and prednisone; DHAP: dexamethasone, cytarabine, and cisplatin; DLBCL: diffuse large B-cell lymphoma; ESMO: European Society for Medical Oncology; EU: European Union; GCB: germinal centre B-cell GDP: gemcitabine, cisplatin, and dexamethasone; GemOx: gemcitabine and oxaliplatin; HD ASCT: high-dose autologous stem cell transplantation; ICE: Ifosfamide, carboplatin, and etoposide; NCCN: National Comprehensive Cancer Network; NHL: Non-Hodgkin’s lymphoma; PR: partial response; BR: bendamustine plus rituximab; R-CHOP: rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-ESHAP/ASHAP: rituximab, etoposide, methylprednisolone, cytarabine, cisplatin/doxorubicin, solumedrol, high-dose cytarabine, and cisplatin; R-GDP: rituximab plus gemcitabine, cisplatin, and dexamethasone; R-GemOx: rituximab, gemcitabine, and oxaliplatin; R-HDCT: rituximab plus high-dose chemotherapy; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide; R-PECC: prednisone, etoposide, chlorambucil and lomustine; R/R: relapsed/refractory; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; SCT: stem cell transplantation; UK: United Kingdom; U.S: United States * In certain markets national guidelines do not describe R/R DLBCL therapeutic management.
53 PTJA06 – Submission Dossier by Roche Registration GmbH
Based on current treatment guidelines, the treatment pathway for patients with DLBCL and the treatment pathway for patients with R/R DLBCL are presented in Figure 4.
Figure 4 Treatment Pathway for First-line and R/R Patients with DLBCL
BSC: best supportive care; CR: complete response; DLBCL: diffuse large B-cell lymphoma; PR: partial response; R: rituximab; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-GDP: rituximab plus gemcitabine, cisplatin, and dexamethasone; R-HDCT: rituximab plus high-dose chemotherapy; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide; SCT: stem cell transplantation Source: Adapted according to ESMO Clinical Practice Guidelines (Tilly et al. 2015) and recent market authorizations in EU.
3.3 UNMET MEDICAL NEED There are only a few therapeutic options for patients with DLBCL who are transplant ineligible due to age, comorbidities, lack of response to salvage treatment, or have relapsed after SCT. There are no universally established therapies or regulatory-approved treatments in this setting. Thus, a high unmet medical need continues to exist since the regimens currently used to treat these patients are limited in efficacy, leading to a mOS of approximately 6 months (Crump et al. 2017; Czuczman et al. 2017). The unmet medical need for R/R DLBCL patients who are ineligible for SCT is reflected in the current ESMO Guidelines that recommend enrollment into clinical trials investigating novel drugs due to the absence of effective therapies (Tilly et al. 2015).
54 PTJA06 – Submission Dossier by Roche Registration GmbH
Therefore, more effective therapies for these patients are urgently needed and their development remains an important objective (Galaznik et al. 2018).
3.4 TARGET POPULATION AND THE PROPOSED POSITIONING OF THE TECHNOLOGY IN THE PATIENT PATHWAY OF CARE The target population may be the population identified in the marketing authorization or a target group of patients using the technology for which the company wants reimbursement.
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.
Describe the target population and the proposed position of the target population in the patient pathway of care.
Provide a justification for the proposed positioning of the technology and the definition of the target population.
Current salvage therapies lead to long-term disease-free survival in only ~10% of R/R DLBCL patients demonstrating an urgent need of better therapeutic options (Amin et al. 2017). As only a subset of R/R DLBCL patients are eligible for SCT, there exists a high unmet need, and the development of more effective therapies for these patients remains an important objective (Galaznik et al. 2018). A lack of viable treatment options with long-term results for these patients currently represents a significant gap in care (Galaznik et al. 2018).
Polatuzumab vedotin in combination with BR showed compelling clinical benefit to patients compared to BR alone as observed by improved OS, PFS, CR rates, and a manageable safety profile, thus providing a new treatment option for R/R DLBCL patients ineligible for SCT, a patient segment characterized with a high unmet medical need. Polatuzumab vedotin in combination with BR provides the potential to offer a major therapeutic advantage over existing treatments.
Polatuzumab vedotin in combination with BR, is proposed to be indicated for the treatment of adult patients with R/R DLBCL who are not candidates for hematopoietic SCT (POLIVY® Proposed/Updated SmPC after CHMP LOQ 2019).
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.
Based on a thorough literature review, and an amalgamation of observed treatment patterns from population-based registries, along with clinical trial data, Roche modelled estimates that in the EU28, the number of treated R/R DLBCL patients eligible to receive polatuzumab vedotin in a year to be approximately 7,800 in 2019 (see Figure 5). By 2025, it is estimated that 8,600 patients would be eligible for treatment with polatuzumab vedotin (see Figure 6).
55 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 5 Broad Treatment Flow and Estimated Number of R/R DLBCL Patients to Receive Treatment with Polatuzumab Vedotin in the EU28 in 2019
1L: first-line; 2L: second-line; 3L: third-line; 4L: fourth-line; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; SCT: stem cell transplantation Yellow numbers indicate the number of patients in the EU28 eligible for treatment with polatuzumab vedotin in 2019. * Non-curative treated patients received palliative interventions.
56 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 6 Trend for DLBCL Incidence, R/R DLBCL and Pool of Patients to Receive Treatment with Polatuzumab Vedotin in the EU28
EU: European Union; Pola: polatuzumab vedotin; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma.
3.5 COMPARATORS IN THE ASSESSMENT
On the basis of the alternatives presented, identify the technologies to be used as comparator(s) for the assessment.
Comparators can differ from the technology in their mechanism of action (whether physical, chemical or mechanical).
If the comparators are different from the technologies identified as alternatives to the intervention or the technologies to which the intervention will be added, provide a justification for the differences.
As seen in clinical guidelines, R/R DLBCL is managed according to whether patients are candidates for SCT (see Section 3.2). The salvage therapies recommended for SCT-ineligible patients mostly involve rituximab in combination with standard anti-neoplastic agents, with the majority of these used off-label (Gisselbrecht et al. 2018; Tilly et al. 2015). Comparator treatments considered relevant to this assessment were agreed with the European Network for Health Technology Assessment (EUnetHTA) Team during the scoping face-to-face meeting in April 2019 and are provided in Table 10. Where possible, Roche agreed to compare against as many of these comparators as possible and to provide justification if a comparison was not feasible; the results are provided in Section 4.8.3.
57 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 10 Key Comparators Recommended by the EUnetHTA in the Proposed PICO
EU ESMO Guideline NCCN Guideline Treatment Regulatory Comment(s) Recommendation Recommendation Status
R-GemOx Not approved Recommended Recommended Off-label use
R-ICE Not approved Recommended Recommended Off-label use
R-DHAP Not approved Recommended Recommended Off-label use
BR Not approved Not described Recommended Off-label use
Yescarta® Approved Not described1 Recommended CAR-T therapy
Kymriah® Approved Not described1 Recommended CAR-T therapy
Pixuvri® Approved Recommended Not recommended Multiply relapsed
BR: bendamustine plus rituximab; CAR-T: chimeric antigen receptor T-cell; ESMO: European Society for Medical Oncology EU: European Union; EUnetHTA: European Network for Health Technology Assessment; NCCN: National Comprehensive Cancer Network; PICO: patient, intervention, comparison, outcome; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-GemOx: rituximab, gemcitabine, and oxaliplatin; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide. 1 The latest ESMO DLBCL Treatment Guidelines were published before Yescarta and Kymriah marketing approval in 2018 in Europe. Source: (Kymriah SmPC; Levact - Article 29 referral 2011; NCCN 2019; Pixuvri SmPC; Rituximab SmPC 2019; Tilly et al. 2015; Yescarta SmPC).
3.5.1 Usage and Results of Selected Key Regimens for the Treatment of R/R DLBCL Patients Ineligible for SCT
3.5.1.1 First- and Subsequent-Line Treatment options for SCT Ineligible R/R DLBCL Patients There are two large prospective Phase II single-arm studies which evaluated BR as a second-line therapy option in DLBCL. The study by Vacirca et al. 2014 included 95% of patients who had previously received rituximab therapy. The CR rate with BR as second-line and onward treatment was only 15.3% (Vacirca et al. 2014) with a mPFS of 3.6 months. Better responses were reported for the BR combination in Japanese patients with relapsed (but not refractory) DLBCL in the study by Ohmachi et al. 2013 with a CR of 37% and mPFS of 6.7 months. Ninety-seven percent of patients had previously received rituximab (Ohmachi et al. 2013). This study excluded patients who were refractory to any prior line of therapy, limited the number of prior therapies to three,
58 PTJA06 – Submission Dossier by Roche Registration GmbH
and limited the performance status to 0 to 1. These factors may account for the higher response rates. However, even within this study, for patients who had relapsed disease within 12 months, the mPFS was poor (mPFS 3.4 months).
In a retrospective multicenter observational study of transplant ineligible R/R DLBCL patients treated with BR, mPFS and mOS were reported to be 3.9 months and 6.7 months, respectively (Hong et al. 2018).
The BR regimen has also been used as a comparator in randomized clinical trials. One of the largest studies to date, a randomized Phase III study of inotuzumab ozogamicin plus rituximab (R-InO) compared with INV’s choice (BR or gemcitabine + rituximab) that included 306 patients with R/R DLBCL, were negative, and failed to show a benefit in terms of objective response rate (ORR; 41% and 44% in the R-InO and immune cells [IC] arms, respectively), PFS (mPFS was 3.7 months and 3.5 months in the R-InO and IC arms, respectively), and OS (mOS was 9.5 months in both the R-InO and IC treatment arms) between patients in the experimental and control arms (Dang et al. 2018).
The BR regimen is a relevant and robust comparator. Based on clinical guidelines, BR is listed as a recommended option for patients with transplant ineligible R/R DLBCL. Based on primary market research data, BR was one the two most frequently prescribed regimens regardless of treatment line and indicates that physicians are choosing to treat their patients with this regimen (Roche Data on File 2017). Furthermore, a thorough literature review of both ongoing and published studies, has shown that the BR regimen continues to be selected as a relevant comparator in clinical trials evaluating potential new treatment options (see Section 4.8.1). This further indicates that the BR regimen is considered an acceptable and relevant comparator in clinical practice.
Similarly, with gemcitabine/platinum-based therapies, outcomes are poor for the transplant ineligible patients, especially if they relapsed early (i.e. ≤12 months) or had prior rituximab exposure. In the Phase II observational study, R-GemOx in the first or second relapse in transplant ineligible patients showed a mPFS and mOS of 11 months and 27 months, respectively, in those who relapsed within a year of their prior therapy and a mPFS and mOS of 4 months and 8 months, respectively, in those who were treated with prior rituximab (Mounier et al. 2013). Similar to BR, R-GemOx has been recognized as a relevant and robust comparator for the treatment of R/R DLBCL patients who are ineligible for SCT.
The limited data from a recent analysis of treatment outcomes in R/R DLBCL patients in the real-world setting from a US database, demonstrated similarity in outcomes between BR and R-GemOx (Ionescu-Ittu et al. 2018). Additionally, the National Institute for Health and Care Excellence (NICE) acknowledges that clinicians consider the various salvage chemotherapy regimens equally effective and GemOx (either with rituximab or without)
59 PTJA06 – Submission Dossier by Roche Registration GmbH
is mainly used with a palliative intent (NICE 2018b). This makes the salvage chemotherapy regimens, in general, but in particular, BR, a valid choice for a comparator in adult patients with R/R DLBCL who are candidates for SCT.
In a retrospective analysis, R-GDP, when used in those who did not ultimately go to transplant, also showed poor outcomes with 2-year PFS and OS of 9% and 11%, respectively (Moccia et al. 2017). Similarly, in the randomized CORAL study (R-ICE/R-DHAP), in patients who had received prior rituximab and relapsed within 12 months and did not go to transplant, 3-year PFS was 14% (Gisselbrecht et al. 2010). Regimens such as R-ICE, R-DHAP and R-GDP are treatment options most frequently used to treat R/R DLBCL patients who are eligible for transplant and are thus not considered relevant comparators for this assessment and will not be discussed further.
3.5.1.2 Second- and Subsequent-line Treatment Options for SCT Ineligible R/R DLBCL Patients Pixantrone is indicated as monotherapy for the treatment of adult patients in the ≥third-line therapy, multiply relapsed or refractory aggressive NHL (Pixuvri SmPC). The benefit of pixantrone treatment has not been established in patients when used as fifth-line chemotherapy in patients who are refractory to last therapy. Pixantrone treatment is approved in the EU for R/R DLBCL treatment (Tilly et al. 2015). However, uncertainties remain regarding pixantrone usage for the treatment of R/R DLBCL patients:
The randomized Phase III, PIX306 study presented at the ASH Annual Meeting in 2018, did not meet its primary endpoint (Salles et al. 2018).
Current ESMO Guidelines suggest that patients eligible for this treatment should be included in clinical trials (Tilly et al. 2015)
Clinical experts explained that pixantrone is rarely used in clinical practice, has poor efficacy and should not be considered (NICE 2018b).
Based on the points listed above, pixantrone is not considered a relevant comparator for this assessment and will not be discussed further
Recently, two CD19-directed CAR-T therapies, axicabtagene ciloleucel (Yescarta®, Gilead) or tisagenlecleucel (Kymriah®, Novartis), have received EMA approval (August 2018 and May 2018, respectively) to treat R/R DLBCL after two or more lines of systemic therapy. While these therapies have shown efficacy with durable CRs, their use may be limited for the general R/R DLBCL population due to the toxicity profile which requires carefully selected patients and treatment in centers with specially trained staff. In addition, the high costs associated with CAR-T therapy represent a significant treatment obstacle. Moreover, the waiting period associated with CAR-T manufacture may be prohibitive in patients with rapidly progressing disease. The NCCN Guidelines
60 PTJA06 – Submission Dossier by Roche Registration GmbH
recommend them as an option after ≥2 prior chemoimmunotherapy regimens (NCCN 2018). Despite the limitation and assuming the ESMO will publish similar recommendations, NCCN CAR-T therapies are considered adequate comparators for polatuzumab vedotin in the third and later treatment lines (Kymriah SmPC).
Best supportive care refers to the therapy that provides the patient with the best possible individually-optimized supportive treatment to alleviate symptoms and improve quality of life. Therefore, palliative care or BSC would not be considered as an appropriate comparator to the chemotherapy-based treatment regimen, polatuzumab vedotin in combination with BR, for the treatment of R/R DLBCL SCT ineligible patients.
Based on all the information presented above, the comparators considered to be the most relevant for this assessment include: BR, R-GemOx, Kymriah, and Yescarta. Furthermore, at the request of the EUnetHTA Assessment Team, a comparison against pixantrone was also requested. The comparisons, where feasible, against each of these treatment regimens is provided in Section 4.8.3.
61 PTJA06 – Submission Dossier by Roche Registration GmbH
4. CLINICAL EFFECTIVENESS AND SAFETY 4.1 OVERVIEW OF EVIDENCE TO SUPPORT THE USE OF POLATUZUMAB VEDOTIN IN R/R DLBCL Direct evidence for the clinical efficacy and safety of Pola BR for the treatment of R/R DLBCL adult patients who are not candidates for HSCT, is derived from one pivotal Phase Ib/II study, GO29365. The study is ongoing, and additional efficacy and safety analyses are planned. An overview of the study design and the objectives for GO29365 are provided in Section 4.2, and the efficacy and safety results are summarized in Section 4.4 and Section 4.6, respectively.
A systematic literature review (SLR) of clinical study evidence on the efficacy, safety and HRQoL of pharmacological interventions for the treatment of R/R DLBCL was performed, and a network meta-analysis (NMA) feasibility assessment was undertaken to explore the plausibility of conducting a formal synthesis of the evidence identified. An overview of the SLR and NMA feasibility assessments are provided in Section 4.8.
In addition, evidence of real-world data is derived from three studies: The Christie National Health Services Foundation Trust, Flatiron Health, and Veteran Affairs. An overview of these three real-world data studies is provided in Section 4.9. These studies describe the characteristics of real-world R/R DLBCL patients and real-world treatment patterns using data from the UK and U.S.
As requested by the EUnetHTA Assessment Team, an overview of ongoing key studies of interest from the polatuzumab vedotin clinical development plan is provided in Table 11.
62 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 11 Overview of Ongoing Key Studies of Interest for Polatuzumab Vedotin
(Estimated) (Estimated) Patient Study Study Number Study Design Intervention Number of Primary Endpoint Population Completion Patients Enrolled Date First-Line
Experimental arm: Phase III, INV-assessed PFS GO39942 Previously Pola R-CHP randomized, using the Lugano (POLARIX) untreated patients (875) (Q2 2021) double-blind, Response Criteria for with DLBCL Control arm: (NCT03274492) placebo-controlled Malignant Lymphoma R-CHOP
Second-Line and Beyond
Experimental arm: Phase III, Pola R-GemOx MO40598 R/R DLBCL randomized, (206) Overall survival (Q4 2023) patients (POLARGO) open-label Control arm: R-GemOx
DLBCL: diffuse large B-cell lymphoma; INV-PFS: investigator-assessed progression-free survival; Pola + R-CHP: polatuzumab vedotin in combination with rituximab plus cyclophosphamide, doxorubicin, and prednisone; Pola + R-GemOx: polatuzumab vedotin in combination with rituximab, gemcitabine, and oxaliplatin; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; R-CHOP: rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone; R-GemOx: rituximab, gemcitabine, and oxaliplatin.
63 PTJA06 – Submission Dossier by Roche Registration GmbH
4.2 PIVOTAL STUDY GO29365 Further information on the presentation of study information can be found in the CONSORT statement for randomized controlled trials, and STROBE guidelines for observational studies.
Include all the studies of the technology relevant to the assessment, as well as studies of comparator technologies (if applicable).
Include ongoing and unpublished studies if these data are available.
In a table, describe the main characteristics of the studies.
For each study provide a flow diagram of the numbers of patients moving through the trial.
Include: patients evaluated for enrolment, those assigned to a treatment category, patients who received treatment as allocated, patients who completed follow-up and patients included in the main analyses.
Tables submitted to the regulatory authorities showing patient flow may be used.
For each study provide a comparison of patients (including demographic, clinical and social information [if applicable]) in treatment arms at baseline.
Tables submitted to the regulatory authorities showing patients’ baseline demographic characteristics may be used if available.
64 PTJA06 – Submission Dossier by Roche Registration GmbH
4.2.1 Overall Study Design Study GO29365 is an ongoing Phase Ib/II, multicenter, open-label study evaluating the safety, tolerability and anti-tumor activity of Pola BR or polatuzumab vedotin plus bendamustine and obinutuzumab (Pola BG) in patients with relapsed/refractory follicular lymphoma (R/R FL) or R/R DLBCL. The overall study design for the GO29365 study is shown in Figure 7 and key study characteristics with respect to the efficacy evaluation are presented in Table 12.
Per the GO29365 protocol (see Appendix 1), efficacy analyses for the DLBCL and FL portions were performed separately as they are two different diseases with different outcomes. Responses to chemoimmunotherapy as well as management strategies differ between the two diseases. While patients with R/R FL were enrolled in the GO29365 study, no claims are being sought at this time for this patient population. Data from the FL and the non-randomized Phase II expansion portions are provided in the GO29365 Interim Clinical Study Report (CSR) for completeness; however, they will not be discussed further in this submission.
The focus of this submission is based only on the Phase II randomized portion of the GO29365 study for patients with R/R DLBCL treated with Pola BR (Arm C) versus BR (Arm D).
The study initially consisted of two stages run sequentially and separately in cohorts of patients with R/R DLBCL. A third cohort (Arm G and Arm H) was later added for R/R DLBCL only:
Phase Ib Safety Run-in Stage: To determine the safety, tolerability and PK of polatuzumab vedotin administered in combination with BR and to identify the recommended Phase II dose (RP2D) of polatuzumab vedotin to be used in combination with BR in the Phase II portion of the study.
Phase II Randomized and Expansion Stages: To evaluate the clinical activity (efficacy), and further evaluate the safety and tolerability, and to characterize the PK of Pola + BR
o The Phase II randomized portion consisted of patients treated with rituximab-containing regimens (Pola BR vs. BR). The Phase II expansion stage was non-randomized and consisted of patients treated with the obinutuzumab-containing regimen, Pola BG.
o The primary objective of the Phase II portion of the study was to evaluate the efficacy of Pola BR compared with BR alone in the randomized portion of the study in patients with or R/R DLBCL as measured by positron emission tomography (PET)-defined CR rate using the modified
65 PTJA06 – Submission Dossier by Roche Registration GmbH
Lugano 2014 Response Criteria, PET-CT criteria) at the primary response assessment.
New formulation single-arm cohort (Arm G and Arm H):
o Arm G was added to evaluate the comparability of the lyophilized formulation of polatuzumab vedotin in combination with BR in R/R DLBCL with the main objective of providing PK/safety data and was subsequently expanded to provide additional data on secondary efficacy outcomes. The full analysis of efficacy and safety will be available in July 2019.
o Following scientific advice received from the CHMP to provide further evaluation of the 140 mg lyophilized formulation of polatuzumab vedotin in combination with BR, a new arm was added to the new formulation cohort (Arm H; N60) in October 2018. For Arm H, the enrollment is ongoing, and the results are expected in Q1 2022.
66 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 7 GO29365: Overview of Study Design
67 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 7 GO29365: Overview of Study Design (cont.)
2L: second-line; acMMAE: antibody-conjugated monomethyl auristatin E BG: bendamustine plus obinutuzumab; BR: bendamustine plus rituximab; D: Day; DLBCL: diffuse large B-cell lymphoma; FL: follicular lymphoma; MMAE: monomethyl auristatin E; PK: pharmacokinetics; RP2D: recommended Phase II dose; R/R: relapsed/refractory; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma R/R FL: relapsed/refractory follicular lymphoma Notes: Rituximab (375 mg/m2) D1 of each cycle, or obinutuzumab (1000 mg) D1, D8, D15 in Cycle 1, then D1 of each subsequent cycle plus bendamustine (90 mg/m2) D2 and D3 in Cycle 1, then D1 and D2 in each subsequent cycle. Polatuzumab vedotin (1.8 mg/kg) D2 in Cycle 1 then D1 of each subsequent cycle. FL: Treatment administered every 28 days x 6 cycles. DLBCL: Treatment administered every 21 days x 6 cycles. a Polatuzumab vedotin (lyophilized) dose 1.8 mg/kg. b Initially, approximately 20 to 30 patients were planned to be enrolled to ensure an adequate number of PK-evaluable patients. A PK-evaluable patient is defined as a patient from whom, at minimum, the PK samplings of all three analyses (i.e. total antibody, acMMAE, and unconjugated MMAE) at Cycle 1 through Cycle 2 Day 1 are obtained. Approximately 10 additional patients with one prior line of therapy (i.e. 2L) will also be enrolled in this arm in order to evaluate the efficacy of Polatuzumab vedotin (lyophilized) in combination with BR in this particular patient subpopulation, bringing the total arm size to approximately 40 patients. c A minimum of approximately 30% of patients who previously received one prior line of therapy (i.e. 2L) will be enrolled in this arm to ensure a sufficient number of efficacy-evaluable patients for this particular subpopulation. Prioritization may be given to ensure such enrollment is achieved in this arm. Source: Section 4.1.1, GO29365 Clinical Overview.
68 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 12 Characteristics of the Studies
Primary Outcome Other Outcome Intervention and Study Measure and Measures and Objective Study Design Eligibility Criteria Comparator (N Reference/ID Follow-Up Time Follow-Up Time enrolled) Point Points Phase Ib Safety Run-in Open-label, Key inclusion A total of 225 patients Safety and RP2D at Pivotal Study Portion: multicenter Phase Ib criteria: were enrolled time of clinical cut-off
GO29365 To determine the safety safety run-in; Signed ICF Randomized Phase II Phase Ib Safety and tolerability, and PK of Age 18 years Pola administered in (Pola BR vs. BR); Run-in Portion: Able to comply combination with BR or Phase II single-arm R/R DLBCL (N 12) with the study BG; expansion Pola + BR (Pola BG); protocol, in the To identify the (N 6) single-arm (Arm G) investigator’s recommended Phase II Pola + BG cohort in R/R DLBCL judgement dose (RP2D) of Pola to be (N 6) (140 mg Lyo DP Histologically used in combination with Pola BR) confirmed FL R/R FL (N 12) BR or BG in patients with (Grade 1, 2, or Pola + BR R/R FL or R/R DLBCL 3a) or DLBCL (N 6) Must have Pola + BG received at least (N 6) one prior therapy for FL or DLBCL. Patients must have relapsed or have become refractory to a prior regimen
69 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 12 Characteristics of the Studies (cont.)
Primary Outcome Other Outcome Intervention and Study Measure and Measures and Objective Study Design Eligibility Criteria Comparator (N Reference/ID Follow-Up Time Follow-Up Time enrolled) Point Points Phase II Randomized Phase II IRC assessed PET OS, Portion: Randomized CR by modified PFS by an IRC To evaluate the efficacy of Portion: Lugano Classification and INV at primary response Pola BR compared with R/R DLBCL (N 80) CR by PET-CT assessment BR alone in the Pola + BR and CT only by (6 to 8 weeks C6D1 randomized portion of the (N 40) investigator study in patients with R/R or last treatment) BR (N 40) OR by PET-CT FL or R/R DLBCL as R/R FL (N 80) and CT only by measured by PET-defined investigator and CR rate using Modified Pola + BR IRC Lugano 2014 Response (N 39) BOR by IRC and Criteria (PET-CT) at the BR (N 41) INV time of primary response assessment as defined by DOR by IRC and an IRC INV Safety PROs Phase II Expansion Portion: Phase II Expansion To evaluate the efficacy, Portion: safety, tolerability and to R/R DLBCL (N 21) characterize the PK of Pola + BG Pola BG R/R FL (N 20) Pola + BG New Formulation Single-Arm New formulation PK and safety Response rates, Cohort (Arm G): Single-Arm Cohort DOR, EFS, PFS, OS To evaluate the PK and safety (Arm G): of a new formulation of Pola R/R DLBCL (N 42) (Lyophilized) BR in R/R Pola (Lyo) BR DLBCL
70 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 12 Characteristics of the Studies (cont.)
Primary Outcome Other Outcome Intervention and Study Measure and Measures and Objective Study Design Eligibility Criteria Comparator (N Reference/ID Follow-Up Time Follow-Up Time enrolled) Point Points New Formulation Single-Arm New Formulation PK and safety Response rates, Cohort (Arm H): Single-Arm Cohort DOR, EFS, PFS, OS Further evaluation of the new (Arm H): formulation of Pola R/R DLBCL (N ~ 60) (Lyophilized) BR in R/R Pola (Lyo) BR DLBCL BOR: best objective response; C: cycle; CR: complete response; CT: computed tomography; D: Day; DOR: duration of response; EFS: Event-free survival; ICF: informed consent form; INV: investigator; IRC: Independent Review Committee; Lyo: lyophilized; Lyo DP: lyophilized drug product; OR: objective response; OS: overall survival; PET: positron emission tomography; PET-CR: positron emission tomography-complete response; PET-CT: positron emission tomography- computed tomography; PFS: progression-free survival; PK: pharmacokinetics; Pola +BR; polatuzumab vedotin in combination with bendamustine plus rituximab; Pola + BG: polatuzumab vedotin in combination with bendamustine plus obinutuzumab; PRO: patient-reported outcome; RP2D: recommended Phase II dose; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; R/R FL: relapsed/refractory follicular lymphoma
71 PTJA06 – Submission Dossier by Roche Registration GmbH
The randomized Phase II portion of the study evaluated Pola (1.8 mg/kg) BR vs. BR and patients were stratified according to the DoR to their last therapy of 12 months versus 12 months. This stratification factor was chosen as it is a reliable prognostic factor in the relapsed and refractory setting (Gisselbrecht et al. 2010; Mounier et al. 2013; Hunter et al. 2017). Patients included in this study must have received at least one prior therapy and must have either relapsed or become refractory to a prior regimen.
The primary objective of the Phase II portion of the study was to evaluate efficacy in terms of positron emission tomography-complete response (PET-CR) rates of Pola BR compared to BR at the time of the PRA (6 to 8 weeks after Cycle 6 Day 1 or the last dose of the study medication) and as assessed by a blinded IRC.
The secondary efficacy objectives of the study are to evaluate the efficacy of Pola BR compared to BR as measured by: CR at PRA based on PET-CT as determined by the INV; OR (CR or PR) at time of the PRA based on PET-CT as determined by the INV and the IRC; CR at the PRA based on CT alone by INV and IRC; OR at the PRA based on CT alone as determined by the INV and the IRC; BOR, CR or PR while in the study, either by PET-CT or CT alone as determined by the INV. Other supportive time-to-event endpoints of median DOR, PFS, EFS, and OS were estimated by the Kaplan-Meier (KM) method.
Response assessments were performed in accordance with modified Lugano 2014 criteria (Cheson et al. 2014). The modification was that a bone marrow biopsy was required to confirm CR by PET-CT criteria if the screening bone marrow showed disease involvement or was not performed. For the IRC, there was an additional modification for an objective assessment of PR by PET-CT which required that the CT scan also show a response of CR or PR (CT criteria) in addition to PET showing a Deauville score of 4 or 5 with a reduced update compared with baseline; otherwise, it was assessed as stable disease (SD). Investigators did not have the CT-based requirement for PR. This discrepancy was due to an inadvertent difference in the response criteria given in the IRC charter and the study protocol for INVs and resulted in minimal impact as it did not affect the primary efficacy endpoint and because most of the responses were CRs. Concordance between the IRC and INV assessments for response rates remained high.
The study is ongoing. The end of the study is scheduled at the time point at which all patients enrolled in the study have either had at least 2 years of follow-up from the time of the treatment completion visit or have discontinued the study (see Section 3.2, GO29365 Study Protocol provided in Appendix 1).
72 PTJA06 – Submission Dossier by Roche Registration GmbH
4.2.2 Discussion of Study Design 4.2.2.1 Rationale for the Selection of Comparator for R/R DLBCL Patients (Randomized Phase II)
Bendamustine and Rituximab For transplant ineligible patients with R/R DLBCL, there is no universally established SoC regimen used as no prior randomized trials have been able to establish the superiority of one regimen over another for this patient population (see Section 3.2). Guidelines, such as those put forth by the NCCN, include BR (see Table 9), which has shown activity in R/R DLBCL (Hong et al. 2018; Ohmachi et al. 2013; Vacirca et al. 2014). In addition, cross-trial comparisons of survival outcomes between different regimens have significant limitations in terms of differences in enrolled or observed populations, such as numbers of refractory versus relapsed (or early relapsed) patients, numbers of prior lines of therapy (e.g. number of patients with one prior line vs. two or more, or if there was a maximum number of prior lines allowed in the trial), and when trials were conducted (some earlier trials occurred at a time when there were still a significant number of patients who were not exposed to rituximab first-line therapy). Thus, it remains difficult to draw definitive conclusions of superiority of one therapy over the other without a randomized trial in a setting where the evidence base is weak.
As one of the first chemotherapy combinations explored in the clinical development plan, BR was selected to partner with polatuzumab vedotin to avoid potential overlapping peripheral neuropathy that may be seen with platinum-based therapies. This concern was due to higher (2.4 mg/kg) and more continuous dosing of polatuzumab vedotin showing higher rates of peripheral neuropathy. With the GO29365 study, a lower dose (1.8 mg/kg) and limiting the dose to cycles, peripheral neuropathy was more tolerable and manageable. sIn order to assess the individual contribution of polatuzumab vedotin in the Pola BR combination, BR was selected as the comparator.
For patients with R/R DLBCL, the dose and schedule for bendamustine used in combination with rituximab in this study (90 mg/m2 administered on two consecutive days for six 21-day cycles) was consistent with the recommendations from an international consensus panel based on data in the relapsed setting at the time the study was initiated (Cheson et al. 2010). Two studies that evaluated the combination of BR in R/R DLBCL used a 120 mg/m2 dose of bendamustine over two consecutive days for up to six cycles (21 days (Ohmachi et al. 2013) or 2 days (Vacirca et al. 2014)) with the standard rituximab dose of 375 mg/m2 on Day 1 of each cycle). Because approximately one-third of patients in these studies required a dose reduction of bendamustine to 90 mg/m2 due to hematological toxicities, the GO29365 study used a starting bendamustine dose of 90 mg/m2 administered over two consecutive days on a 21-day cycle.
73 PTJA06 – Submission Dossier by Roche Registration GmbH
4.2.2.2 PET-defined CR as a Primary Tumor Response Assessment The primary efficacy endpoint is CR as determined by an IRC at the PRA (defined as 6 to 8 weeks after Cycle 6 Day 1 or the last dose of treatment) using a PET-CT scan. PET-CT scanning has been shown in multiple settings to be a more accurate tool for assessing lymphoma activity than CT imaging. In aggressive lymphomas, such as DLBCL, PET-defined CR is a stronger predictor of PFS than response as defined by CT (Casasnovas et al. 2011; Zhu et al. 2013).
4.2.3 Key Inclusion and Exclusion Criteria Full study inclusion and exclusion criteria are described in Section 3.5, GO29365 Interim CSR. Key study inclusion and exclusion criteria for R/R DLBCL patients are summarized in Table 13.
Table 13 GO29365: Key Inclusion and Exclusion Criteria
Key Inclusion Criteria: Signed ICF Age 18 years Able to comply with the study protocol, in the investigator’s judgement Histologically confirmed DLBCL Must have received at least one prior therapy for DLBCL. Patients must have either relapsed or have become refractory to a prior regimen Key Exclusion Criteria: History of severe allergic or anaphylactic reactions to humanized or murine mAbs (or recombinant antibody-related fusion proteins) or known sensitivity or allergy to murine products Contraindication to bendamustine, rituximab, or obinutuzumab Prior use of any mAb, radioimmunoconjugate, or ADC within five half-lives or 4 weeks, whichever is longer before Cycle 1 Day 1 Treatment with radiotherapy, chemotherapy, immunotherapy, immunoresponsive therapy, or any investigational agent for the purposes of treating cancer within 2 weeks prior to Cycle 1 Day 1 Eligibility for autologous stem cell transplant History of transformation of indolent disease to DLBCL
ADC: antibody-drug conjugate; DLBCL: diffuse large B-cell lymphoma; ICF: Informed Consent Form; mAb: monoclonal antibody Source: Section 1.3.1, GO29365 Clinical Overview.
4.2.4 Analysis Populations Efficacy analyses for the randomized Phase II portion of the GO29365 study were based on the intention-to-treat (ITT) population and conducted in accordance with the ITT
74 PTJA06 – Submission Dossier by Roche Registration GmbH
principle (i.e. including all randomized patients irrespective of whether they received study treatment, with patients grouped according to treatment assignment at randomization).
Safety analyses were based on the safety evaluable population which included all treated patients (i.e. patients who received any amount of study medication) according to actual treatment received.
4.2.5 Efficacy Analysis Analysis methodology for key efficacy outcome measures in the GO29365 study are summarized in Table 14.
Table 14 GO29365: Efficacy Outcome Measures and Analysis Methodology
Outcome Measure Analysis Methodology OS Distribution of durations for PFS and OS summarized descriptively using KM INV-PFS methodology (Kaplan et al. 1958) to estimate median (if analytically possible), 1-year, and 2- year PFS and 95% CIs using Greenwood’s formula. There was no pre-specified alpha control plan; p-values are provided for descriptive purposes only. IRC-PFS by PET-CT Patients without a post-baseline tumor assessment were considered non-responders. Analyses of these efficacy response endpoints are identical to those described above for CR rate as measured by PET-CT scan and as determined by an IRC. CR rate using modified Lugano CR rate, defined as the percentage of patients Response Criteria as measured at the with CR, was estimated and the corresponding PRA (6 to 8 weeks after Cycle 6 Day 1 or Clopper-Pearson exact 95% CI was constructed for each treatment arm. last dose of study medication) as The difference in PET-CR rates between Pola measured by PET-CT scan and as BR and BR arms was estimated along with determined by an IRC the corresponding 95% CI on the basis of normal approximation to the binomial distribution. An exploratory comparison of CR rates for the Pola BR and BR regimens was conducted using the CMH 2 square test adjusted for randomization stratification factors (Agresti 2002). BOR (IRC-assessed) by PET-CT Same as above for PFS (IRC-assessed) by PET-CT
75 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 14 GO29365: Efficacy Outcome Measures and Analysis Methodology (cont.)
Outcome Measure Analysis Methodology DOR (IRC-assessed) by PET-CT Same as above for PFS (IRC-assessed) by PET-CT BOR: best overall response; BR: bendamustine plus rituximab; CI: confidence interval; CMH: Cochran-Mantel-Haenszel; CR: complete response; CSR: Clinical Study Report; DOR: duration of response; INV: investigator; IRC: Independent Review Committee; KM: Kaplan-Meier; OS: overall survival; PET-CR: positron emission tomography-complete response; PET-CT: positron emission tomography-computed tomography; PFS: progression-free survival; Pola: polatuzumab vedotin; PRA: primary response assessment Source: Section 3.9.3, GO29365 Interim CSR.
4.2.6 Patient-Reported Outcome (PRO) Analyses The PRO analyses included patients in the ITT population and were analyzed according to assigned treatment.
For the total score and each of the Therapy-Induced Neuropathy Assessment Score (TINAS) single symptom items, descriptive statistics for recorded values at each visit and changes from baseline were calculated. Line charts depicting the mean change over time are provided for the total score and single symptom items. In the event of missing data, total scores could be prorated (estimated) for the total score, as described in Section 3.9.4, GO29365 Interim CSR.
4.2.7 Anti-Drug Antibody (ADA) Analysis The prevalence of ADA at baseline was calculated by dividing the total number of patients in all study groups that tested positive for ADA at baseline by the total number of patients with a valid ADA test result at baseline.
The incidence of ADAs post-baseline in each study group was calculated by dividing the number of patients that developed treatment-induced ADAs (i.e. patients with a negative or missing baseline ADA result(s) and at least one positive post-baseline ADA result), plus the number of patients that had treatment-enhanced ADAs (i.e. the ADA response had increased 0.60 titer units from baseline) in the study by the total number of patients with valid post-baseline results in that study group during the study period.
4.2.8 Safety Reporting and Analysis Safety assessments included monitoring and recording AEs, including serious adverse events (SAEs) and non-serious adverse events of special interest (AESIs), measurement of protocol-specified safety laboratory assessments and vital signs, and other protocol-specified tests that were deemed critical to the safety evaluation of the study.
76 PTJA06 – Submission Dossier by Roche Registration GmbH
The methods used for analyzing and reporting safety endpoints for the GO29365 was pre-specified in the protocol (see Section 6.5, GO29365 Protocol provided in Appendix 1) and were based on the recommendations from the Extension of the CONSORT Statement (Ioannidis et al. 2004) and the International Conference of Harmonization (ICH) E9 (Statistical Principles for Clinical Trials). Typically, the safety and tolerability implication are best addressed using descriptive statistical methods because most trials lack the power to test harm-related hypotheses or have no explicit pre-specified harm-related hypotheses. Additionally, conducting statistical tests over multiple safety outcomes such as AEs causes the multiple testing issue which inflates the Type I error.
4.2.8.1 Adverse Events The standard definition of an AE according to the ICH guideline for GCP is provided in Section 5.2.1 of the GO29365 Protocol (see Appendix 1).
SAEs were defined as any AE that met the criteria listed in Section 5.2.2 of the GO29365 protocol (see Appendix 1) and required expedited reporting (reported by the investigator to the Sponsor immediately (i.e. no more than 24 hours after learning of the event).
Information on AEs was sought by the investigator at each patient contact at scheduled and unscheduled visits. Investigators were responsible for ensuring that all AEs were recorded on the electronic case report form (eCRF) and reported to the Sponsor, and to make an assessment of seriousness, severity and causality in accordance with instructions provided in the protocol.
The severity of AEs was graded by the investigator according to the National Cancer Institute – Common Terminology Criteria for Adverse Events (NCI CTCAE) v4.0 or using Table 9 of the GO29365 Protocol (see Appendix 1) for those events that are not specifically listed in the NCI CTCAE.
Investigators determined whether or not an AE was considered to be related to the study drug following causal attribution guidance provided in Section 5.3.4 of the protocol (see Appendix 1).
For classification purposes, lower level terms were assigned by the Sponsor to the original verbatim descriptions of AE terms entered on the case report form (CRF), using the most up-to-date version of the Medical Dictionary for Regulatory Activities (MedDRA v20.0) terminology for AEs and diseases and the Genentech Drug Thesaurus for medication. Glossaries of adverse events, diseases, and medications are provided in Section 3.9.7.1 of the GO29365 Interim CSR.
77 PTJA06 – Submission Dossier by Roche Registration GmbH
Non-Serious Adverse Events of Special Interest Per protocol, in addition to SAEs, non-serious adverse events of special interest (AESIs) were required to be reported by the investigator to the Sponsor immediately (i.e., no more than 24 hours after learning of the event; see Section 5.4.2 of the protocol in Appendix 1). Non-serious AESIs, are defined as follows:
Potential drug-induced liver injury (DILI) that includes an elevated alanine aminotransferase (ALT) or aspartate aminotransferase (AST) in combination with either an elevated bilirubin or clinical jaundice, as defined by Hy’s Law (see Section 5.3.5.7 of the protocol in Appendix 1)
Suspected transmission of an infectious agent by the study drug, as defined below:
o Any organism, virus, or infectious particle (e.g. prion protein transmitting transmissible spongiform encephalopathy), pathogenic or non-pathogenic, is considered an infectious agent. A transmission of an infectious agent may be suspected from clinical symptoms or laboratory findings that indicate an infection in a patient exposed to a medicinal product. This term applies only when a contamination of the study drug is suspected
o TLS of any Grade (irrespective of causality)
o Second malignancies
Selected Adverse Events/Events to Monitor Selected AEs are identified risks and potential risks of polatuzumab vedotin based on data available at the time of the assessment and include AEs which require further characterization or evaluation through dedicated analyses of these selected AEs used groupings of MedDRA preferred terms as defined in Section 3.9.7.1 of the GO29365 Interim CSR.
All AEs occurring during or after the first treatment are summarized with descriptive statistics (frequency counts and percentages by treatment group) and are presented separately for each treatment regimen (Pola BR and BR), by study phase (Phase Ib safety run-in and Phase II randomized portions) and by NHL histology (DLBCL). Analyses of pooled Phase Ib/Phase II randomized data for Pola BR for DLBCL are also provided.
In summary tables of the overall incidence of AEs, patients who experienced the same event on more than one occasion are counted only once in the calculation of event frequency, and the AE with the most extreme severity was included.
78 PTJA06 – Submission Dossier by Roche Registration GmbH
Full details on the safety reporting and analysis are provided in Section 3.9, GO29365 Interim CSR.
4.3 RESULTS: STUDY POPULATION 4.3.1 Patient Population The GO29365 study included adult patients with R/R DLBCL (see Section 4.2.3 for key inclusion and exclusion criteria) who were not considered as candidates for transplant. The determination of transplant ineligibility is based on clinical experience as there is no established criteria. In general, patients are transplant ineligible typically due to either patient-related characteristics (such as age, comorbidities, performance status) or disease-related characteristics, such as chemo-insensitive/resistant disease as well as relapsing after prior autologous transplant. These two categories comprise the majority of reasons for transplant ineligibility in the two GO29365 randomized study arms and were found, overall, to be generally similar between the two treatment arms. For further details, see Section 4.3.4.
4.3.2 Patient Disposition The disposition of the six patients in the Phase Ib safety run-in and the 80 patients in the Phase II randomized arms for patients with R/R DLBCL are summarized by treatment arm in Table 15 and Table 16, respectively.
79 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 15 GO29365: Disposition of Patients in the Phase Ib Safety Run-In at the Clinical Cut-off Date (30 April 2018)
Study Phase Phase Ib Safety Run-in
Histology DLBCL
Cohort 1a Treatment Pola BR No. of patients enrolled (ITT population) n 6
No. of patients who: Received treatment (Safety population) n 6 Completed all treatments 2 (33.3%) Alive at follow-up 4 (66.7%) Discontinued study due to: Adverse event 0 Death 2 (33.3%) Non-safety reasons 0 CSR: Clinical Study Report; DLBCL: diffuse large B-cell lymphoma; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab. Source: Section 4.1, GO29365 Interim CSR.
80 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 16 Disposition of Patients with R/R DLBCL in the Phase II Randomized Portion of the GO29365 Study at the Clinical Cut-off Date (30 April 2018)
Study Phase Phase II Randomized Portion Histology DLBCL Arm C D Treatment Pola BR BR No. patients enrolled n 40 n 40 (ITT population) No. of patients who: Received treatment n 39 n 39 (Safety population) Completed all 18 (45.0%) 9 (22.5%) treatments Alive at follow-up 11 (27.5%) 4 (10.0%) Discontinued study 29 (72.5%)‡ 36 (90.0%)§ due to: Adverse event 0 0 Death 23 (57.5%) 28 (70.0%) Non-safety 6 (15.0%)‡ 8 (20.0%)§ reasons* BR: bendamustine plus rituximab; DLBCL: diffuse large B-cell lymphoma; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab. * Includes withdrawal by subject, progressive disease, physician decision, lack of efficacy, lost-to-follow-up, and other. Percentages are based on the ITT population. Includes patients who did not receive study treatment due to: ‡ no longer meeting eligibility criteria at time of dosing (1); § screen failure and discontinued of the study due to progressive disease (1). Source: Section 4.1, GO29365 Interim CSR.
A total of 86 patients with R/R DLBCL were included in the Phase Ib safety run-in and in the randomized Phase II portion of the GO29365 study: six patients in the Phase Ib safety run-in assigned to Pola BR [Cohort 1a] and 80 patients in the Phase II randomized portion (40 patients each assigned to Pola BR [Arm C] or BR [Arm D]).
Two patients with R/R DLBCL were enrolled in the study but did not receive any study medication: one in each of the arms of the Phase II randomized portion (BR: one patient reported as a screen failure; Pola BR: one patient no longer met the eligibility criteria at time of dosing).
The 84 patients with R/R DLBCL in the Phase Ib safety run-in and Phase II randomized portion received the following:
81 PTJA06 – Submission Dossier by Roche Registration GmbH
Pola BR: 45 patients (six patients in the Phase Ib safety run-in and 39 patients in the Phase II randomized portion)
BR: 39 patients (all in the Phase II randomized portion)
At the time of clinical cut-off (30 April 2018), the status of the 86 patients with R/R DLBCL in the Phase Ib and Phase II randomized portion (ITT population) was as follows:
19 patients were alive at follow-up (15 patients in the Pola BR arm, and four patients in the BR arm).
53 patients had died (25 patients in the Pola BR arm, and 28 patients in the BR arm).
14 patients withdrew from the study:
o 10 patients withdrew consent (five patients in the Pola BR arm, and five patients in the BR arm)
o One patient in the Pola BR arm withdrew due to “other” reasons: one patient who did not receive the study treatment due to no longer meeting eligibility criteria at the time of dosing.
o Three patients in the BR arm withdrew: two patients due to progressive disease (including the patient who did not receive the study treatment due to screen failure), and one patient due to the physician’s decision
The median duration of follow-up in the randomized Phase II portion of the study was the same for patients with R/R DLBCL in the Pola BR and BR treatment arms (22.3 months in each treatment arm; Table 17).
82 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 17 Duration of Follow-Up (Phase Ib and Randomized Phase II; ITT Population)*
Phase II Phase Ib Randomized Pola BR BR Pola BR Patients with R/R DLBCL Cohort/Arm 1a D C Sample Size, n 6 40 40 Median TTE, mo. 37.6 22.3 22.3 (95% CI) (34.5, 40.1) (20.1, 24.9) (20.5, 23.1) BR: bendamustine plus rituximab; CI: confidence interval; CSR: Clinical Study Report; ITT: intention-to-treat; mo: month; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; TTE: time-to-event * Survival follow-up calculated by the reverse KM method for OS. Source: Section 4.4, GO29365 Interim CSR.
4.3.3 Demographics The key demographics of patients in the Phase Ib safety run-in and Phase II randomized portions of the GO29365 study are summarized in Table 18.
Demographic characteristics were generally well balanced across cohorts of patients treated with Pola BR or BR. Any differences in incidence of demographic characteristics by category observed between Pola BR and BR treatment arms in the Phase II randomized portion was less than 10% (accounted for by four or fewer patients).
Key Demographic characteristics for R/R DLBCL patients treated with Pola BR or BR by line of therapy are provided in Appendix 2.
4.3.4 Baseline Disease Characteristics Most patients with R/R DLBCL in the Phase Ib and Phase II randomized portions (74/86 [86.0%]) had AA Stage III or IV disease at study entry. The majority of patients at baseline (84/86 [97.7%]) were categorized with DLBCL “not otherwise specified” based on the 2016 WHO classification, including 48.8% (42/86 patients) activated B-cell (ABC) type and 38.4% (33/86 patients) germinal center B-cell-like (GCB) type (see Table 18).
In the Phase II randomized portion, a higher proportion of patients with R/R DLBCL in the BR treatment arm, compared to the Pola BR treatment arm had bulky disease (7.5 cm) (BR: 15/40 [37.5%] vs. Pola BR: 10/40 [25.0%]).
Most patients with R/R DLBCL in the Phase II randomized portion had three or more IPI risk factors at enrollment (i.e. high-intermediate or high-risk groups). However, a higher
83 PTJA06 – Submission Dossier by Roche Registration GmbH
proportion of patients in the BR arm were categorized as IPI high risk (4 or 5 risk factors; 42.5%) than patients in the Pola BR (22.5%) arm (see Table 18).
In the Phase II randomized portion, the most common reason for transplant ineligibility fell into two categories: patient characteristics such as age, comorbidity, or inadequate performance status (BR: 22/40 [55.0%]; Pola BR: 14/40 [35.0%]), and disease status, including inadequate response to salvage therapy or relapsing after prior transplant (BR: 15/40 [37.5%], Pola BR: 22/40 [55.0%]) (see Table 18 or Appendix 3).
Key baseline disease characteristics for R/R DLBCL patients treated with Pola BR or BR by line of therapy are provided in Appendix 2.
84 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 18 GO29365: Key Demographic, Baseline Disease Characteristics and Prognostic Factors for Patients with R/R DLBCL Treated with Pola + BR or BR (Phase Ib and Phase II Randomized; ITT Population)
Phase II Phase Ib Phase Ib/II (Randomized) Pola BR Pola BR BR Pola BR (total) N 6 N 40 N 40 N 46 Demographics Age: median, range (years) 65.0 (5879) 71.0 (3084) 67.0 (3386) 66.5 (3386) 65 years 3 (50.0%) 26 (65.0%) 23 (57.5%) 26 (56.5%) Sex, male 4 (66.7%) 25 (62.5%) 28 (70.0%) 32 (69.6%) Race White 5 (83.3%) 31 (77.5%) 26 (65.0%) 31 (67.4%) Asian 1 (16.7%) 4 (10.0%) 6 (15.0%) 7 (15.2%) American Indian or 0 1 (2.5%) 0 0 Alaskan Native Black or African 0 0 3 (7.5%) 3 (6.5%) American Unknown 0 4 (10.0%) 5 (12.5%) 5 (10.9%) Ethnicity Not Hispanic or Latino 6 (100.0%) 36 (90.0%) 35 (87.5%) 41 (89.1%) Hispanic or Latino 0 1 (2.5%) 1 (2.5%) 1 (2.2%) Unknown/not stated 0 3 (7.5%) 4 (10.0%) 4 (8.7%) ECOG PS at Baseline 0 or 1 6 (100.0%) 31 (77.5%) 33 (82.5%) 39(84.7%) 2 0 8 (20.0%) 6 (15.0%) 6 (13.0%) Unknown 0 1 (2.5%) 1 (2.5%) 1 (2.2%) Primary Reason for Stem Cell Transplant Ineligibility Age 1 (16.7%) 19 (47.5%) 13 (32.5%) 14 (30.4) Co-morbidities 0 1 (2.5%) 1 (2.5%) 1 (2.2%) Failed prior transplant 0 6 (15.0%) 10 (25.0%) 10 (21.7%) Insufficient cells collected 0 0 0 0 Insufficient response to 2 (33.3%) 9 (22.5%) 12 (30.0%) 14 (30.4%) salvage therapy Other 1 (16.7%) 1 (2.5%) 2 (5.0%) 3 (6.5%) Patient refused transplant 2 (33.3%) 2 (5.0%) 2 (5.0%) 4 (8.7%) Performance status 0 2 (5.0%) 0 0
85 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 18 GO29365: Key Demographic, Baseline Disease Characteristics and Prognostic Factors for Patients with R/R DLBCL Treated with Pola + BR or BR (Phase Ib and Phase II Randomized; ITT Population) (cont.)
Phase II Phase Ib Phase Ib/II (Randomized) Pola BR Pola BR BR Pola BR (total) N 6 N 40 N 40 N 46 Baseline Disease Characteristics Time since diagnosis at study entry: median, range 0.5 (01) 0.8 (015) 0.7 (020) 0.7 (020) (months) WHO 2016 DLBCL status a DLBCL, NOS: ABC 4 (66.7%) 19 (47.5%) 19 (47.5%) 23 (50.0%) DLBCL, NOS: GCB 1 (16.7%) 17 (42.5%) 15 (37.5%) 16 (34.8%) 45 (high) 0 17 (42.5%) 9 (22.5%) 9 (19.6%) DLBCL, NOS 1 (16.7%) 4 (10.0%) 4 (10.0%) 5 (10.9%) FL 0 0 1 (2.5%) 1 (2.2%) Burkitt lymphoma 0 0 1 (2.5%) 1 (2.2%) Primary mediastinal (thymic) large B-cell - - - - lymphoma Ann Arbor Stage III or IV at 4 (66.7%) 36 (90.0%) 34 (85.0%) 38 (82.6%) study entry Bulky disease (7.5 cm) 1 (16.7%) 15 (37.5%) 10 (25.0%) 11 (23.9%) Extranodal involvement 4 (66.6%) 29 (72.5%) 27 (67.5%) 31 (67.4%) Prognostic Factors IPI score at enrollment 01 (low) 1 (16.7%) 3 (7.5%) 9 (22.5%) 10 (21.7%) 2 (lowintermediate) 3 (50.0%) 8 (20.0%) 9 (22.5%) 12 (26.1%) 3 (highintermediate) 2 (33.3%) 12 (30.0%) 13 (32.5%) 15 (32.6%) 45 (high) 0 17 (42.5%) 9 (22.5%) 9 (19.6%) ABC: activated B-cell type; BR: bendamustine plus rituximab; CSR: Clinical Study Report; ECOG PS: Eastern Cooperative Oncology Group Performance Status; FL: follicular lymphoma; GCB: germinal center B-cell-like; IPI: International Prognostic Index; ITT: intention-to-treat; NOS: not otherwise specified; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; WHO: World Health Organization. a DLBCL status confirmed by central lab (HistoGeneX) review of pathology reports and tissue samples. The diagnosis at study entry was recorded as unknown in cases where a biopsy was not done or was uninformative. Source: Section 4.6.1, GO29365 Interim CSR.
4.3.5 Prior Anti-Lymphoma Therapy Per protocol inclusion criteria (see Section 4.2), all patients had received at least one prior anti-lymphoma therapy for DLBCL.
86 PTJA06 – Submission Dossier by Roche Registration GmbH
Patients in each treatment arm had received a median of two prior lines of anti-lymphoma chemotherapy, and up to a maximum of seven previous lines of treatment (see Table 19).
The full listing of previous systemic anti-lymphoma therapy received by patients with R/R DLBCL are provided by treatment arm (Pola BR vs. BR) in Section 4.6.1.3, GO29365 Interim CSR.
All except one patient (in the randomized Pola BR arm) with R/R DLBCL had received an anti-cluster of differentiation 20 (CD20) agent in a previous line of therapy, and most patients in each treatment cohort had been administered CHOP previously (80%). Prior etoposide, prednisone, vincristine, cyclophosphamide, hydroxydaunorubicin (EPOCH) treatment was received by 3/40 patients (7.5%) in both the Pola BR and BR randomized arms. All except one patient in the Pola BR arm of the Phase II randomized were naïve to bendamustine treatment. The incidence of prior bone-marrow transplantation and cancer radiotherapy are summarized in Table 19.
Most patients were refractory to their last prior anti-lymphoma therapy, defined as no response or progression or relapse within 6 months of the last anti-lymphoma therapy dose (BR: 34/40 patients [85.0%]; Pola BR: 35/46 patients [76.1%]).
The proportion of patients with a DoR to prior therapy of 12 months (a stratification factor for randomization in the randomized Phase II) was similar in the randomized arms (BR: 82.5% vs. Pola BR: 80.0%; see Table 19).
87 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 19 Previous Anti-Lymphoma Treatment and Response in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib and Phase II Randomized; ITT Population)
Phase Ib Phase II Phase Ib/II (Randomized)* Pola BR BR Pola BR Pola BR (total) N 6 N 40 N 40 N 46 Prior anti-lymphoma chemotherapy: 6 (100.0%) 40 (100.0%) 40 (100.0%) 46 (100.0%) Median no. lines (range) 2.0 (12) 2.0 (15) 2.0 (17) 2.0 (17) 1 line 2 (33.3%) 12 (30.0%) 11 (27.5%) 13 (28.3%) 2 lines 4 (66.7%) 9 (22.5%) 11 (27.5%) 15 (32.6%) 3 lines 0 19 (47.5%) 18 (45.0%) 18 (39.1%) Prior treatments: Anti-CD20 6 (100.0%) 40 (100.0%) 39 (97.5%) 45 (97.8%) Bendamustine 0 0 1 (2.5%) 1 (2.2%) Bone-marrow transplant 0 6 (15.0%) 10 (25.0%) 10 (21.7%) Cancer radiotherapy 1 (16.7%) 10 (25.0%) 11 (27.5%) 12 (26.1%) Refractory to last prior anti-CD20 4 (66.7%) 18 (45.0%) 18 (45.0%) 22 (47.8%) agents a Refractory to last prior anti-lymphoma 5 (83.3%) 34 (85.0%) 30 (75.0%) 35 (76.1%) therapy b Time from last anti-lymphoma therapy: 53.0 82.0 131.0 114.0 median, range (days) c (431477) (212948) (1711744) (1711744) Duration of response to prior therapy: d 12 months 5 (83.3%) 33 (82.5%) 32 (80.0%) 37 (80.4%) BR: bendamustine plus rituximab; CD20: cluster of differentiation 20; CRF: case report form; CSR: Clinical Study Report; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma a Defined as no response or progression or relapse within 6 months of last anti-lymphoma therapy end date among patients whose last prior regimen contained anti-CD20. b Defined as no response or progression or relapse within 6 months of last anti-lymphoma therapy end date. c Defined as time from end date of last anti-lymphoma therapy to first dose date. d Duration of response to prior therapy was based on an IxRS for the randomized portion and a CRF for non-randomized portions. Source: Section 4.6.1.3, GO29365 Interim CSR.
88 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4 EFFICACY RESULTS IN PATIENTS WITH R/R DLBCL TREATED WITH POLA BR OR BR (RANDOMIZED PHASE II) Include all the studies of the technology relevant to the assessment, as well as studies of comparator technologies (if applicable).
Include ongoing and unpublished studies if these data are available.
o Describe the relevant endpoints, including the definition of the endpoint, and method of analysis (table 10).
The study results presented should reflect the outcomes relevant to the assessment including, if available, mortality, morbidity, function, (health-related) quality of life and patient satisfaction.
If the endpoint uses a scale, state how it was validated; if this uses responder analyses, state and justify the responder definition.
o Provide a summary of the study results for each relevant comparison and outcome (see example tables 11 and 12).
Data should be presented according to intention-to-treat. Alternative presentations of the data should be justified.
In the case of survival analyses, Kaplan-Meier curves that include the number of patients at risk at various time points should be provided.
If non-comparative data are included in the submission the summary of outcomes should include measures over time (for example baseline, and post-intervention). Estimates should be presented as unadjusted estimates and as estimates adjusted for potential confounders. The confounders used in the adjustment should be stated and their use justified (see example table 13).
89 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4.1 Overview of Efficacy in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II) In no more than 6 bullet points describe key statements relating to the clinical effectiveness of the technology. For example, include statements about the benefit of the technology compared to alternative technologies currently used.
The addition of polatuzumab vedotin to BR (Pola BR; Arm C) resulted in higher response rates and longer OS, PFS, DOR, and BOR compared to BR (Arm D) in the Phase II randomized portion of the GO29365 study. Efficacy results in patients with R/R DLBCL are summarized below and in Table 20. Evaluation of efficacy was not a pre specified objective of the Phase Ib safety run-in portion of the study and hence not included within this section.
The risk of death was reduced by 58% in patients treated with Pola BR compared to BR (stratified HR 0.42; 95% CI: [0.24, 0.75]; p 0.0023; see Figure 8). Median OS was extended to 12.4 months (95% CI: 9.0, NE) in the Pola BR arm, from 4.7 months (95% CI: 3.7, 8.3) in the BR arm. The treatment effect for survival was consistently observed across all subgroups of patients with R/R DLBCL tested (see Figure 9).
Progression-free survival as determined by an IRC was increased in patients treated with Pola BR compared to BR (stratified HR 0.36; 95% CI: [0.21, 0.63]; p 0.0002; see Figure 10) with mPFS being over two-fold higher (9.5 months; 95% CI: [6.2, 13.9] vs. 3.7 months 95% CI: [2.1, 4.5]). The treatment effect for PFS by IRC was consistently observed across all subgroups of patients with R/R DLBCL tested with the exception of AA Stage I/II at study entry (HR 0.53), bulky disease (HR 0.52), and prior bone marrow transplant (HR 0.69) (see Figure 11). Median INV-assessed progression-free survival (INV-PFS) was over three-fold the duration in patients treated with Pola BR (7.6 months; 95% CI: [6.0, 17.0]) compared to BR (2.0 months; 95% CI: [1.5, 3.7]) (stratified HR 0.34; 95% CI: [0.20, 0.57]; p-value 0.0001).
The CR rate at the PRA based on PET-CT, as determined by the IRC, was higher in the Pola BR arm (40.0% [16/40 patients]; 95% CI: [24.9%, 56.7%]) compared with the BR arm (17.5% [7/40 patients]; 95% CI: [7.3%, 32.8%]). The difference in CR rates between arms was significant ( 22.5% in favor of Pola BR; 95% CI: [2.6%, 40.2%]; p 0.0261, CMH chi-square).
Response rates (CR and OR; CR or PR), whether measured with or without PET-CT, as assessed by the IRC or by the INV, remained consistent with the CR rate at the PRA, with a higher proportion of patients with R/R DLBCL achieving CR or OR in the Pola BR arm compared to the BR arm.
90 PTJA06 – Submission Dossier by Roche Registration GmbH
A best response of CR or PR as assessed by the IRC was achieved by 62.5% of patients in the Pola BR arm and 25.0% of patients in the BR arm while on the study. The proportion of patients achieving a best response of CR was 50.0% and 22.5%, respectively. A best response of CR or PR, as assessed by the INV, was achieved by 70.0% of patients in the Pola BR arm and 32.5% of patients in the BR arm while on the study. The proportion of patients achieving a best response of CR was 57.5% and 20.0%, respectively.
Median DoR as determined by the IRC was 12.6 months (95% CI: 7.2, NE) in the Pola BR arm and 7.7 months (95% CI: 4.0, 18.9) in the BR arm (stratified HR 0.47; 95% CI: [0.19, 1.14]). Median DoR by the INV was 10.3 months (95% CI: 5.6, NE) for Pola BR and 4.1 months (95% CI: 2.6, 12.7) for BR (stratified HR 0.44; 95% CI: [0.20, 0.95]).
91 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 20 GO29365: Overview of Key Efficacy Outcomes in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
BR Pola BR
N 40 N 40 Mortality OS Patients with an event, n (%) 28 (70.0%) 23 (57.5%) Median OS (95% CI) 4.7 (3.7, 8.3) 12.4 (9.0, NE) HR (95% CI); stratified p-value (log-rank) 0.42 (0.24, 0.75); p0.0023 Morbidity PFS (IRC) Patients with event, n (%) 32 (80.0%) 25 (62.5%) Median PFS (95% CI) 3.7 (2.1, 4.5) 9.5 (6.2, 13.9) HR (95% CI); stratified p-value (log-rank) 0.36 (0.21, 0.63); p0.0002 PFS (INV) Patients with an event, n (%) 35 (87.5%) 27 (67.5%) Median PFS (95% CI) 2.0 (1.5, 3.7) 7.6 (6.0, 17.0) HR (95% CI); stratified p-value (log-rank) 0.34 (0.20, 0.57); p0.0001 CR† at PRA by PET-CT (IRC) n (%) 7 (17.5%) 16 (40.0%) 95% CI for response rate (Clopper-Pearson) (7.3, 32.8) (24.9, 56.7) (95% CI) (Wilson); p-value (CMH chi-square*) 22.5 (2.62, 40.2); p0.0261 BOR by PET-CT or CT (CR/PR) (IRC) n (%) 10 (25.0%) 25 (62.5%) 95% CI for response rate (Clopper-Pearson) (12.7, 41.2) (45.8, 77.3) (95% CI) (Wilson); p-value (CMH chi-square*) 37.5 (15.8, 54.6); p0.0005 BOR by PET-CT or CT (CR/PR) (INV) n (%) 13 (32.5%) 28 (70.0%) 95% CI for response rate (Clopper-Pearson) (18.6, 49.1) (53.5, 83.4) (95% CI) (Wilson); p-value (CMH chi-square*) 37.5 (15.6, 54.7); p0.0006 DOR (IRC) Patients with an event, n (%) 8/10 (80.0%) 13/25 (52.0%) Median DOR (95% CI) 7.7 (4.0, 18.9) 12.6 (7.2, NE) HR (95% CI); stratified p-value (log-rank) 0.47 (0.19, 1.14); p0.0889 DOR (INV) Patients with an event, n (%) 11/13 (84.6%) 17/28 (60.7%) Median DOR (95% CI) 4.1 (2.6, 12.7) 10.3 (5.6, NE) HR (95% CI); stratified p-value (log-rank) 0.44 (0.20, 0.95); p0.0321
92 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 20 GO29365: Overview of Key Efficacy Outcomes in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.)
BR: bendamustine rituximab; BOR: best overall response; CI: confidence interval; CMH: Cochran Mantel- Haenszel; CR: complete response; CSR: Clinical Study Report; CT: computed tomography; DOR: duration of response; HR: hazard ratio; INV: investigator; IRC: Independent Review Committee; ITT: intention-to- treat; NE: not estimable; OS: overall survival; PET-CT: positron emission tomography-computed tomography; PFS: progression-free survival; PR: partial response; PRA: primary response assessment (68 weeks after Cycle 6 Day 1 or last dose of study medication); Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. * Cochran Mantel-Haenszel 2 test adjusted for randomization stratification factors: DOR to prior therapy (12 months vs. 12 months). † Per modified Lugano 2014 criteria: bone marrow confirmation of PET-CT CR required. PET-CT PR required meeting both PET-CT criteria and CT criteria. Clinical cutoff date (CCOD): 30 April 2018. Source: Section 5.1.1, GO29365 Interim Clinical Study Report (CSR).
93 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 8 GO29365: KM Plot of Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
CCOD: 30 April 2018. BR: bendamustine and rituximab; CCOD: Clinical cut-off date CI: confidence interval; CSR: Clinical Study Report; HR: hazard ratio; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Source: Section 5.1.4.4, GO29365 Interim CSR.
94 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 9 GO29365: Forest Plot of Hazard Ratio for Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI; confidence interval; CSR: Clinical Study Report; DLBCL: diffuse large B-cell lymphoma; ECOG PS: Eastern Cooperative Oncology Group Performance Status; IPI: International Prognostic Index; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; WHO: World Health Organization Source: Section 5.1.4.5, GO29365 Interim CSR.
95 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 10 GO29365: KM Plot of Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI: confidence interval; IRC-PFS: Independent Review Committee- assessed progression-free survival; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Note: The p-value (log-rank) shown is for the unstratified analysis. Source: Section 5.1.3.7, GO29365 Interim CSR.
96 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 11 GO29365: Forest Plot of Hazard Ratio for Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI: confidence interval; DLBCL: diffuse large B-cell lymphoma; ECOG PS: Eastern Cooperative Oncology Group Performance Status; IPI: International Prognostic Index; IRC-PFS: Independent Review Committee-assessed progression-free survival; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Source: Section 5.1.3.7, GO29365 Interim CSR.
97 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4.2 Efficacy in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Efficacy analyses for the randomized Phase II stage (Pola BR vs. BR) of the GO29365 study were based on the ITT population and conducted in accordance with the ITT principle (i.e. including all randomized patients irrespective of whether they received study treatment, with patients grouped according to treatment assignment at randomization).
4.4.2.1 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) At the time of clinical cut-off (30 April 2018), a total of 51 patients with R/R DLBCL in the Phase II randomized portion of the study had died; 28 patients (70.1%) in the BR arm and 23 patients (57.5%) in the Pola BR arm (see Table 21). The main cause of death in both arms was disease progression (17 patients in the BR arm and 14 patients in the Pola BR arm). A further two patients with R/R DLBCL in the Pola BR safety run-in died of progressive disease.
The risk of death was reduced by 58% in patients treated with Pola BR compared to BR (stratified HR 0.42; 95% CI: [0.24, 0.75]; p 0.0023).
Median OS was 12.4 months (95% CI: 9.0, NE) in patients in the Pola BR arm compared to 4.7 months (95% CI: 3.7, 8.3) in the BR arm. A KM plot of OS for patients with R/R DLBCL treated with Pola BR versus BR in the randomized Phase II portion of the GO29365 study is shown in Figure 8.
The duration of survival follow-up for patients with R/R DLBCL as assessed by estimated follow-up methodology was the same in the two treatment arms (median follow-up 22.3 months in both the Pola BR and BR treatment arms). For further details on the duration of survival follow-up, see Table 17.
98 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 21 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_tte_OS_rBRDL_IT Summary of Overall Survival, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 28 (70.0%) 23 (57.5%) Earliest contributing event Death 28 23 Patients without event (%) 12 (30.0%) 17 (42.5%)
Time to event(month) Median 4.731 12.386 95% CI (3.713, 8.312) (9.035, NE) 25% and 75%-ile 3.154, 8.871 6.242, NE Range 0.20* to 25.03 0.30* to 26.09*
Unstratified Analysis p-value (log-rank) 0.0019
Hazard Ratio 0.42 95% CI (0.24, 0.74)
Stratified Analysis p-value (log-rank) 0.0023
Hazard Ratio 0.42 95% CI (0.24, 0.75)
6 months duration Patients remaining at risk 11 30 Event Free Rate (%) 37.38 78.95 95% CI (20.68, 54.07) (65.99, 91.91)
Difference in Event Free Rate -41.57 95% CI (-62.71, -20.43) p-value (Z-test) 0.0001
9 months duration Patients remaining at risk 7 24 Event Free Rate (%) 23.78 65.50 95% CI (8.76, 38.81) (50.29, 80.70)
Difference in Event Free Rate -41.71 95% CI (-63.09, -20.34) p-value (Z-test) 0.0001
12 months duration Patients remaining at risk 7 19 Event Free Rate (%) 23.78 51.85 95% CI (8.76, 38.81) (35.79, 67.92)
Difference in Event Free Rate -28.07 95% CI (-50.07, -6.07) p-value (Z-test) 0.0124
______Stratification factors: Duration of Response to Prior Therapy, IXRS (<=12/>12 months). * Censored at time of data cut off, ^ Censored and event at time of data cut off 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for difference in rates were constructed using the normal approximation to the binomial distribution. Cutoff Date: 30APR2018 Extract Date: 29JUN2018
Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_tte.sas / Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_tte_OS_rBRDL_IT.out 18SEP2018 0:31
BR: bendamustine plus rituximab; CI: confidence interval; CSR: Clinical Study Report; ITT: intention-to- treat; IxRS: interactive voice or Web-based response system; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; NE: Not evaluable; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Source: Section 5.1.4.4, GO29365 Interim CSR.
99 PTJA06 – Submission Dossier by Roche Registration GmbH
Subgroup Analyses: Overall Survival Subgroup analysis of OS in patients with R/R DLBCL by baseline risk factor shows that for all patient subgroups tested, the treatment effect for survival was consistent with and in the same direction (point estimates of HR were generally 0.50) as for the overall R/R DLBCL population (see Figure 9).
Subgroup analyses of OS by primary reason for SCT ineligibility and by prior line of therapy are provided in Section 4.4.3.1 and Section 4.4.3.2, respectively.
Updated Overall Survival The updated efficacy data presented for the GO29365 study is from a CCOD of 11 October 2018. The CCOD date represents approximately 5 months of additional efficacy data from the previous CCOD of 30 April 3018.
The updated OS for the Phase II randomized patients are presented below with an updated clinical cut-off date of 11 October 2018, which provides a median follow-up of 26.5 months.
At the time of the clinical cut-off, one more patient in the Pola BR arm died (60.0%; 24/40 patients) compared to the previous data cut-off on 30 April 2018 (Table 22; Figure 12).
The updated HRs for OS (HR: 0.40; 95% CI: [0.23, 0.70]; see Table 22) remained consistent with the HR for OS (HR: 0.42; 95% CI: [0.24, 0.75]; see Table 21) observed in the 30 April 2018 analysis.
For patients treated with Pola BR, the estimated survival rate at 24 months was 38.2%, compared to 17% in the BR arm (Table 22).
With longer follow-up, the same treatment benefit was observed for Pola BR compared to BR as demonstrated in the pivotal analysis (CCOD: 30 April 2018).
100 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 22 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Summary of Overall Survival, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
______
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 28 (70.0%) 24 (60.0%) Earliest contributing event Death 28 24 Patients without event (%) 12 (30.0%) 16 (40.0%)
Time to event(month) Median 4.731 12.386 95% CI (3.713, 8.312) (9.035, 28.025) 25% and 75%-ile 3.154, 8.871 6.242, NE Range 0.20* to 26.48* 0.30* to 31.08*
Unstratified Analysis p-value (log-rank) 0.0015
Hazard Ratio 0.41 95% CI (0.24, 0.72)
Stratified Analysis p-value (log-rank) 0.0010
Hazard Ratio 0.40 95% CI (0.23, 0.70)
6 months duration Patients remaining at risk 11 30 Event Free Rate (%) 37.38 78.95 95% CI (20.68, 54.07) (65.99, 91.91)
Difference in Event Free Rate -41.57 95% CI (-62.71, -20.43) p-value (Z-test) 0.0001
9 months duration Patients remaining at risk 7 24 Event Free Rate (%) 23.78 65.50 95% CI (8.76, 38.81) (50.29, 80.70)
Difference in Event Free Rate -41.71 95% CI (-63.09, -20.34) p-value (Z-test) 0.0001
12 months duration Patients remaining at risk 7 19 Event Free Rate (%) 23.78 51.85 95% CI (8.76, 38.81) (35.79, 67.92)
Difference in Event Free Rate -28.07 95% CI (-50.07, -6.07) p-value (Z-test) 0.0124
18 months duration Patients remaining at risk 6 15 Event Free Rate (%) 20.39 40.94 95% CI (6.11, 34.67) (25.09, 56.79)
Difference in Event Free Rate -20.55 95% CI (-41.88, 0.79) p-value (Z-test) 0.0590
24 months duration Patients remaining at risk 4 10 Event Free Rate (%) 16.99 38.21 95% CI (3.63, 30.35) (22.54, 53.88)
Difference in Event Free Rate -21.22 95% CI (-41.81, -0.62) p-value (Z-test) 0.0435
______* Censored at time of data cut off, ^ Censored and event at time of data cut off 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for median time were constructed using Brookmeyer and Crowley method.
101 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 22 Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.) Cutoff Date: 11OCT2018 Extract Date: 30NOV2018
Program: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/ema_questions_90dsu/qa/program/t_ef_t te2.sas Output: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/ema_questions_90dsu/qa/output/t_ef_tt e2_OS_rBRDL_IT.out 07MAY2019 17:53 BR: bendamustine plus rituximab; CI: confidence interval; ITT: intention-to-treat; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; NE: Not evaluable; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma.
Figure 12 KM Plot of Overall Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI: confidence interval; DLBCL: diffuse large B-cell lymphoma; ITT: intention-to-treat; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma
Totality of R/R DLBCL Data from Study GO29365 (Phase Ib/Phase II Randomized) To evaluate the long-term survival of Pola BR, an analysis was conducted considering the totality of DLBCL data from the GO29365 study. Patients in the Phase Ib and Phase II randomized portions of the GO29365 study were included. The respective KM estimate for OS is presented in Figure 13.
102 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 13 KM Overall Survival Estimates for Pola BR (Phase Ib/Phase II Randomized)
KM: Kaplan Meier; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab. Source: (Sehn et al. 2019).
4.4.2.2 Progression-Free Survival in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
Progression-Free Survival as Assessed by IRC and Investigator (INV)
IRC-assessed More patients with R/R DLBCL in the BR arm, compared to the Pola BR arm of the Phase II randomized portion of the GO29365 study had a PFS event (PD or death) at the time of the clinical cut-off (80.0% [32/40] vs. 62.5% [25/40]). Deaths accounted for most of the PFS events (19/32) in the BR arm and 13/32 patients in the BR arm had disease progression (see Table 23). The PFS events in the Pola BR included 11 deaths and 14 patients with disease progression as of the CCOD (30 April 2018; see Table 23).
The risk of PD or death was reduced by 64% in patients treated with Pola BR compared to BR (stratified HR 0.36; 95% CI: [0.21, 0.63]; p 0.0002) (see Table 23).
Median PFS was over two-fold the duration in patients treated with Pola BR (9.5 months; 95% CI: [6.2, 13.9]) compared to BR (3.7 months; 95% CI: [2.1, 4.5]) (see Table 23).
103 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 23 Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Summary of Progression-Free Survival by IRCe, RAN BR DLBCL Group, Intent-to-Treatment Patients, April 2019 Efficacy Update Protocol: GO29365
______
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 32 (80.0%) 25 (62.5%) Earliest contributing event Death 19 11 Disease Progression 13 14 Patients without event (%) 8 (20.0%) 15 (37.5%)
Time to event(month) Median 3.713 9.462 95% CI (2.070, 4.534) (6.242, 13.930) 25% and 75%-ile 1.659, 5.060 2.563, NE Range 0.03* to 25.03 0.03* to 24.97*
Unstratified Analysis p-value (log-rank) 0.0004
Hazard Ratio 0.38 95% CI (0.22, 0.66)
Stratified Analysis p-value (log-rank) 0.0002
Hazard Ratio 0.36 95% CI (0.21, 0.63)
6 months duration Patients remaining at risk 5 23 Event Free Rate (%) 15.82 65.59 95% CI (3.16, 28.47) (50.42, 80.76)
Difference in Event Free Rate -49.77 95% CI (-69.52, -30.02) p-value (Z-test) <.0001
9 months duration Patients remaining at risk 5 19 Event Free Rate (%) 15.82 54.18 95% CI (3.16, 28.47) (38.05, 70.31)
Difference in Event Free Rate -38.37 95% CI (-58.87, -17.86) p-value (Z-test) 0.0002
12 months duration Patients remaining at risk 4 15 Event Free Rate (%) 12.65 42.77 95% CI (1.11, 24.19) (26.62, 58.93)
Difference in Event Free Rate -30.12 95% CI (-49.97, -10.27) p-value (Z-test) 0.0029
18 months duration Patients remaining at risk 3 7 Event Free Rate (%) 9.49 31.37 95% CI (0.00, 19.67) (16.14, 46.60)
Difference in Event Free Rate -21.88 95% CI (-40.20, -3.56) p-value (Z-test) 0.0193
24 months duration Patients remaining at risk 1 1 Event Free Rate (%) 4.74 31.37 95% CI (0.00, 13.06) (16.14, 46.60)
Difference in Event Free Rate -26.62 95% CI (-43.97, -9.27) p-value (Z-test) 0.0026
104 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 23 Progression-Free Survival (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.) Stratification factors: Duration of Response to Prior Therapy, IxRS (<=12/>12 months). * Censored at time of data cut-off, ^ Censored and event 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for difference in rates were constructed using the normal approximation to the binomial distribution. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cutoff Date: 30APR2018 Extract Date: 29JUN2018
Program: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/program/t_ef_tte.s as Output: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/output/t_ef_tte_PF SIRC_rBRDL_IT_APR2019.out 17MAY2019 7:25
BR: bendamustine plus rituximab; CI: confidence interval; CT: computed tomography IRC-PFS: Independent Review Committee-assessed progression-free survival; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; NE: Not evaluable; PET-CT: positron emission tomography- computed tomography; Pola + BR; polatuzumab vedotin in combination bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; Source: Section 5.1.3.7, GO29365 Interim CSR.
The KM plot of PFS by IRC for patients with R/R DLBCL treated with Pola BR or BR in the Phase II randomized portion of the GO29365 study is shown in Figure 10. Patients with DLBCL who were event-free 24 months after therapy induction are considered long-term survivors (NICE 2018b; Maurer et al. 2014).
Subgroup Analyses – Progression-Free Survival The forest plot shown in Figure 11 shows the treatment effect for PFS by IRC was consistent and in the same direction as for the overall R/R DLBCL population with point estimates of HR generally 0.50 across all subgroups with the exception of AA Stage I/II at study entry (HR: 0.53), bulky disease (HR: 0.52) and prior bone marrow transplant (HR: 0.69).
INV-assessed More patients with R/R DLBCL in the BR arm, compared to the Pola BR of the Phase II randomized portion of the GO29365 study, had a PFS event (PD or death) at the time of the clinical cut-off (87.5% [35/40 patients] vs. 67.5 [27/40 patients]). Progressive disease accounted for most of the PFS events in both arms (30/33 patients in the BR arm and 21/24 patients in the Pola BR arm). Five patients in the BR arm and six patients in the Pola BR arm died prior to confirmation of PD (see Table 24).
The risk of PD or death was reduced by 66% in patients treated with Pola BR compared to BR (stratified HR 0.34; 95% CI: [0.20, 0.57]; p0.0001; see Table 24).
105 PTJA06 – Submission Dossier by Roche Registration GmbH
Median PFS was over three-fold the duration in patients treated with Pola BR (7.6 months; 95% CI: [6.0, 17.0]) compared to BR (2.0 months; 95% CI: [1.5, 3.7]; see Table 24).
106 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 24 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_tte_PFS_rBRDL_IT Summary of Progression-Free Survival by Investigator, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 35 (87.5%) 27 (67.5%) Earliest contributing event Disease Progression 30 21 Death 5 6 Patients without event (%) 5 (12.5%) 13 (32.5%)
Time to event(month) Median 2.037 7.622 95% CI (1.544, 3.713) (5.979, 16.953) 25% and 75%-ile 0.953, 4.780 2.563, NE Range 0.03* to 21.95* 0.03* to 24.97*
Unstratified Analysis p-value (log-rank) 0.0001
Hazard Ratio 0.37 95% CI (0.22, 0.62)
Stratified Analysis p-value (log-rank) <.0001
Hazard Ratio 0.34 95% CI (0.20, 0.57)
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
6 months duration Patients remaining at risk 6 23 Event Free Rate (%) 16.67 62.95 95% CI (4.56, 28.78) (47.53, 78.37)
Difference in Event Free Rate -46.28 95% CI (-65.89, -26.67) p-value (Z-test) <.0001
9 months duration Patients remaining at risk 6 17 Event Free Rate (%) 16.67 46.53 95% CI (4.56, 28.78) (30.48, 62.57)
Difference in Event Free Rate -29.86 95% CI (-49.96, -9.76) p-value (Z-test) 0.0036
12 months duration Patients remaining at risk 4 14 Event Free Rate (%) 11.11 38.32 95% CI (0.88, 21.34) (22.64, 53.99)
Difference in Event Free Rate -27.20 95% CI (-45.92, -8.48) p-value (Z-test) 0.0044
______Stratification factors: Duration of Response to Prior Therapy, IxRS (<=12/>12 months). * Censored at time of data cut-off, ^ Censored and event 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for difference in rates were constructed using the normal approximation to the binomial distribution. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cutoff Date: 30APR2018 Extract Date: 29JUN2018
Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_tte.sas Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_tte_PFS_rBRDL_IT.out 18SEP2018 0:30
107 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 24 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.)
BR: bendamustine plus rituximab; CI: confidence interval; CSR: Clinical Study Report; CT: computed tomography; DLBCL: diffuse large B-cell lymphoma; INV: investigator; ITT: intent-to-treat; IxRS: interactive voice or Web-based response system; NE: Not evaluable; PET-CT: positron emission tomography- computed tomography; PFS: progression-free survival; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; Source: Section 5.1.4.2, GO29365 Interim CSR.
A KM plot for patients with R/R DLBCL treated with Pola BR or BR in the randomized Phase II portion of the GO29365 study is shown in Figure 14. A clear separation of curves in favor of the Pola BR treatment arm is evident. Any direct comparisons between the two curves beyond 10 months should be interpreted with caution as KM estimates are considered unreliable when there are 20% of patients remaining at risk (Pocock et al. 2002).
108 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 14 KM Plot of Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CSR: Clinical Study Report; INV: investigator; ITT: intent-to-treat; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; PFS: progression-free survival; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab Source: Section 5.1.4.2, GO29365 Interim CSR.
109 PTJA06 – Submission Dossier by Roche Registration GmbH
Subgroup Analyses – Progression-Free Survival Subgroup analysis of PFS in patients with R/R DLBCL evaluated the potential impact of demographic and baseline disease characteristics, and other prognostic factors on treatment effect.
The forest plot shown in Figure 15 shows the treatment effect for PFS by INV was consistent and in the same direction as for the overall R/R DLBCL population with point estimates of HR 0.50 across all except two subgroups. One exception was the subgroup of patients with AA Stage I/II disease at study entry (HR 0.7; 95% CI: [0.1, 4.98]); however, this subgroup was one of the smallest subgroups evaluated with 10 patients in total, and the number of PFS events was small (2/4 patients in the BR arm and 2/6 patients in the Pola BR arm). The other exception was the subgroup of patients with a prior transplant (HR 0.86; 95% CI: [0.26, 2.88]) which was also one of the smaller subgroups evaluated with a total of 16 patients.
110 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 15 Forest Plot of Hazard Ratio for Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR by Subgroup (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI: confidence interval; CSR: Clinical Study Report; CT: computed tomography; ECOG PS: Eastern Cooperative Oncology Group Performance Status; INV-PFS: investigator-assessed progression-free survival; IPI: International Prognostic Index; ITT: intention-to-treat; PET-CT: positron emission tomography-computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; WHO: World Health Organization Source: Section 5.1.4.2, GO29365 Interim CSR.
111 PTJA06 – Submission Dossier by Roche Registration GmbH
Updated Progression-Free Survival as Assessed by the Investigator The updated efficacy data presented for the GO29365 study is from a CCOD of 11 October 2018. The CCOD date represents approximately 5 months of additional efficacy data from the previous CCOD of 30 April 3018.
The updated PFS as assessed by the INV for the Phase II randomized patients are presented below with an updated clinical cut-off date of 11 October 2018, which provides an estimated median follow-up of 26.5 months. Patients with DLBCL who were event-free 24 months after therapy induction are considered long-term survivors (NICE 2018b; Maurer et al. 2014).
Similar to what was seen in the 30 April 2018 cut-off date, more patients in the BR arm, compared to the Pola BR arm had a PFS event (PD or death) (87.5% [35/40 patients] vs. 72.5% [29/40]). Progressive disease accounted for most of the PFS events in both arms (Pola BR: 79.3% [23/29 patients]; BR: 85.7% [30/35 patients]) (see Table 25 and Figure 16).
The updated HR for PFS as assessed by the INV (HR 0.33; 95% CI: [0.20, 0.56]; see Table 25) remained consistent with the HR for PFS as assessed by the INV (HR 0.34; 95% CI: [0.20, 0.57]; see Table 24) observed in the pivotal analysis (CCOD: 30 April 2018).
112 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 25 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Summary of Progression-Free Survival by INV, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
______
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 35 (87.5%) 29 (72.5%) Earliest contributing event Disease Progression 30 23 Death 5 6 Patients without event (%) 5 (12.5%) 11 (27.5%)
Time to event(month) Median 2.037 7.491 95% CI (1.544, 3.713) (4.928, 16.953) 25% and 75%-ile 0.953, 4.780 2.563, 28.025 Range 0.03* to 26.12* 0.03* to 30.95*
Unstratified Analysis p-value (log-rank) <.0001
Hazard Ratio 0.37 95% CI (0.22, 0.62)
Stratified Analysis p-value (log-rank) <.0001
Hazard Ratio 0.33 95% CI (0.20, 0.56)
6 months duration Patients remaining at risk 6 23 Event Free Rate (%) 16.67 60.53 95% CI (4.56, 28.78) (44.99, 76.07)
Difference in Event Free Rate -43.86 95% CI (-63.56, -24.16) p-value (Z-test) <.0001
9 months duration Patients remaining at risk 6 18 Event Free Rate (%) 16.67 47.37 95% CI (4.56, 28.78) (31.49, 63.24)
Difference in Event Free Rate -30.70 95% CI (-50.67, -10.73) p-value (Z-test) 0.0026
12 months duration Patients remaining at risk 4 15 Event Free Rate (%) 11.11 39.47 95% CI (0.88, 21.34) (23.93, 55.01)
Difference in Event Free Rate -28.36 95% CI (-46.97, -9.76) p-value (Z-test) 0.0028
18 months duration Patients remaining at risk 3 11 Event Free Rate (%) 8.33 31.17 95% CI (0.00, 17.34) (16.32, 46.03)
Difference in Event Free Rate -22.84 95% CI (-40.21, -5.47) p-value (Z-test) 0.0100
24 months duration Patients remaining at risk 1 9 Event Free Rate (%) 5.56 28.34 95% CI (0.00, 13.03) (13.84, 42.84)
Difference in Event Free Rate -22.78 95% CI (-39.10, -6.47) p-value (Z-test) 0.0062
______* Censored at time of data cut off, ^ Censored and event at time of data cut off
113 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 25 Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.) 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for median time were constructed using Brookmeyer and Crowley method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off Date: 11OCT2018 Extract Date: 30NOV2018
Program: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/ema_questions_90dsu/qa/program/t_ef_t te2.sas Output: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/ema_questions_90dsu/qa/output/t_ef_tt e2_PFS_rBRDL_IT.out 07MAY2019 18:04
BR: bendamustine plus rituximab; CI: confidence interval; CT: computed tomography; INV: investigator; ITT: intention-to-treat; PET-CT: positron emission tomography-computed tomography; PFS: progression-free survival; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma;
Figure 16 KM Plot of Progression-Free Survival (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population)
BR: bendamustine and rituximab; CI: confidence interval; CT: computed tomography; DLBCL: diffuse large B-cell lymphoma; INV-PFS: investigator-assessed progression-free survival; ITT: intention-to-treat; PET-CT: positron emission tomography-computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma
114 PTJA06 – Submission Dossier by Roche Registration GmbH
Totality of R/R DLBCL Data from Study GO29365 (Phase Ib/Phase II Randomized) To evaluate the long-term PFS of Pola BR, an analysis was conducted considering the totality of DLBCL data from the GO29365 study. When combining data from the Phase Ib safety run-in and data from the randomized Phase II portions of the GO29365 study, an estimated 30% of patients are alive and progression-free 2 years after commencing treatment (Sehn et al. 2019). The KM estimate for PFS (INV-assessed) is presented in Figure 17. Ten (22%) patients treated with Pola BR (3 patients in the Phase Ib safety run-in and 7 patients in the randomized Phase II) remain in complete remission at last follow-up (ongoing duration of response of 20 months) (Sehn et al. 2019).
Figure 17 KM Progression-Free Survival (INV-assessed) Estimates for Pola BR (Phase Ib/Phase II Randomized)
Source: (Sehn et al. 2019).
4.4.2.3 CR Rate (IRC-assessed) by PET-CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) In the Phase II randomized portion of the GO29365 study, the proportion of patients with R/R DLBCL with a CR at the PRA by PET-CT as assessed by IRC was higher in the Pola BR arm (Arm C: 40.0% [16/40 patients]; 95% CI: [24.9%, 56.7%]) compared to patients in the BR arm (Arm D: 17.5% [7/40 patients]; 95% CI: [7.3%, 32.8%]) (see Table 26). The difference in CR rates between arms was statistically significant (22.5%; p 0.0261; CMH chi-square test).
115 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 26 CR Rate (IRC-assessed) by PET-CT at Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_rs_CRIRCPET_rBRDL_IT Summary of CR Rate at Primary Response Assessment with PET - IRC, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Number of Evaluable Patients 40 40 Complete Response (CR) 7 (17.5%) 16 (40.0%)
95% CI for Response Rate (Clopper-Pearson) (7.34, 32.78) (24.86, 56.67)
Difference in Response Rates 22.50 95% CI for Difference in Response Rates (2.62, 40.22) (Wilson) p-Value (Cochran-Mantel-Haenszel) 0.0261
______Stratification factors: Duration of Response to Prior Therapy from IxRS (<=12/>12 months). 95% CI for rates were constructed using Clopper-Pearson method. Cut-off Date: 30APR2018 Extract Date: 29JUN2018
Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_rs.sas / Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_rs_CRIRCPET_rBRDL_IT.out 18SEP2018 0:28 BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; CSR: Clinical Study Report; IRC: Independent Review Committee; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; PET: positron emission tomography; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; PRA: primary response assessment; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; Source: Section 5.1.2, GO29365 Interim CSR.
4.4.2.4 Response Rates (CR and OR) at the Primary Response Assessment Measured With or Without PET-CT as Assessed by IRC and INV Response rates (CR and OR), whether measured with or without PET-CT and assessed by the INV or by the IRC, remained consistent with the CR rate at the PRA, with a higher proportion of patients with R/R DLBCL achieving CR or OR in the Pola BR arm compared with the BR arm. The concordance between the IRC and the INV in the response assessment at the PRA in patients with R/R DLBCL in the randomized Phase II portion of the study was high, particularly for achievement of a CR or PR (90% in each arm), and balanced in the Pola BR and BR arms.
116 PTJA06 – Submission Dossier by Roche Registration GmbH
4.4.2.4.1 Response Rates at the Primary Response Assessment Based on PET-CT IRC-assessed The OR (CR or PR) rate at the time of the PRA based on PET-CT as determined by the IRC was 45.0% (18/40 patients) in the Pola BR versus 17.5% (7/40 patients) in the BR arm, a difference of 27.5% (p 0.0069, CMH chi-square test) (see Table 27). Most patients who had an OR at the PRA in each treatment arm achieved a CR (BR arm: 17.5% CRs, 0 PRs vs. Pola BR: 40.0% CRs, 5.0% PRs).
Table 27 Objective Response (CR/PR) Rates (IRC-assessed) by PET-CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_rs_ORIRCPET_rBRDL_IT Summary of OR Rate at Primary Response Assessment with PET - IRC, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Number of Evaluable Patients 40 40 Overall Response (CR/PR) 7 (17.5%) 18 (45.0%)
95% CI for Response Rate (Clopper-Pearson) (7.34, 32.78) (29.26, 61.51)
Difference in Response Rates 27.50 95% CI for Difference in Response Rates (7.17, 45.02) (Wilson) p-Value (Cochran-Mantel-Haenszel) 0.0069
Complete Response (CR) 7 (17.5%) 16 (40.0%) 95% CI (7.34, 32.78) (24.86, 56.67)
Partial Response (PR) 0 2 (5.0%) 95% CI (0.00, 8.81) (0.61, 16.92)
______Stratification factors: Duration of Response to Prior Therapy from IxRS (<=12/>12 months). 95% CI for rates were constructed using Clopper-Pearson method. Cutoff Date: 30APR2018 Extract Date: 29JUN2018
Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_rs.sas / Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_rs_ORIRCPET_rBRDL_IT.out 18SEP2018 0:28 BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; CSR: Clinical Study Report; DLBCL: diffuse large B-cell lymphoma; IRC: Independent Review Committee; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; OR: objective response; PET; positron emission tomography; PET-CT: positron emission tomography-computed tomography; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; PR: partial response; PRA: primary response assessment; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Source: Section 5.1.3.1, GO29365 Interim CSR.
INV-assessed The INV assessment of response by PET-CT at the PRA was highly consistent with the IRC assessment. Concordance between the two assessment modalities is discussed in Section 4.4.2.4.3.
The CR rates were 42.5% (17/40 patients) in the Pola BR arm and 15.0% (6/40 patients) in the BR arm, a difference of 27.5% (p 0.0061, CMH chi-square test).
117 PTJA06 – Submission Dossier by Roche Registration GmbH
The OR rates, driven by CRs, were 47.5% (19/40 patients) and 17.5% (7/40 patients), respectively (30.0%; p 0.0036, CMH chi-square test) (see Table 28).
Table 28 Complete Response (CR) and Objective Response (CR/PR) Rates (INV-assessed) by PET-CT at the Primary Response Assessment in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_rs_ORPET_rBRDL_IT Summary of OR Rate at Primary Response Assessment with PET, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Number of Evaluable Patients 40 40 Overall Response (CR/PR) 7 (17.5%) 19 (47.5%)
95% CI for Response Rate (Clopper-Pearson) (7.34, 32.78) (31.51, 63.87)
Difference in Response Rates 30.00 95% CI for Difference in Response Rates (9.48, 47.37) (Wilson) p-Value (Cochran-Mantel-Haenszel) 0.0036
Complete Response (CR) 6 (15.0%) 17 (42.5%) 95% CI (5.71, 29.84) (27.04, 59.11)
Partial Response (PR) 1 (2.5%) 2 (5.0%) 95% CI (0.06, 13.16) (0.61, 16.92)
______Stratification factors: Duration of Response to Prior Therapy from IxRS (<=12/>12 months). 95% CI for rates were constructed using Clopper-Pearson method. Cut-off Date: 30APR2018 Extract Date: 29JUN2018
Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_rs.sas / Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_rs_ORPET_rBRDL_IT.out 17SEP2018 20:27 BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; CSR: Clinical Study Report; DLBCL: diffuse large B-cell lymphoma; INV: investigator; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; OR: objective response; PET; positron emission tomography; PET CT: positron emission tomography-computed tomography; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; PR: partial response; PRA: primary response assessment; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Source: Section 5.1.3.1, GO29365 Interim CSR.
4.4.2.4.2 Response Rates at the Primary Response Assessment Based on CT Only IRC-assessed The CR rates for patients with R/R DLBCL, measured by CT only and assessed by an IRC, were 22.5% (9/40 patients) in the Pola BR arm and 2.5% (1/40 patients) in the BR arm.
OR rates were 42.5% and 15.0%, respectively and compared with ORs by PET-CT, were made up of proportionately more PRs (BR arm: 2.5% CRs, 12.5% PRs vs. Pola BR: 22.5% CRs, 20.0% PRs). Full results are provided in Section 5.1.3.2, GO29365 Interim CSR.
118 PTJA06 – Submission Dossier by Roche Registration GmbH
INV-assessed The INV assessment of response by CT at the PRA was similar to the IRC assessments. The proportions of patients with CR and OR were 20.0% (8/40 patients) in the Pola BR arm versus 5.0% (2/40 patients) in the BR arm, and 45.0% (18/40 patients) versus 15.0% (6/40 patients), respectively. Full results are provided in Section 5.1.3.2, GO29365 Interim CSR.
4.4.2.4.3 Concordance Between IRC and INV Assessments at the PRA The concordance between the IRC and the INV in the response assessment at the PRA in patients with R/R DLBCL in the Phase II randomized portion of the GO29365 study was high, particularly for achievement of a CR or PR (90% in each arm), and balanced in the Pola BR and BR arms. The highest discordance was for SD (in all discordant cases, determined by the IRC, but not by the INV), PD (in all discordant cases, determined by the INV, but not by the IRC) and not evaluable in the BR arm (in most discordant cases, determined by the IRC, but not by the INV).
4.4.2.5 Best Overall Response (CR/PR) While on Study as Assessed by the IRC and INV For analyses of BOR, the tumor assessment result was based on PET-CT results.
IRC-assessed A best response of CR or PR was assessed by an IRC and was achieved by 62.5% (25/40) of patients in the Pola BR arm and 25.0% (10/40) of patients in the BR arm while on the study. The proportion of patients achieving a best response of CR was 50.0% (20/40) and 22.5% (9/40), respectively (see Table 29).
119 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 29 Best Overall Response Rate (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Summary of Best Overall Response Rate, RAN BR DLBCL Group, Intent-to-Treatment Patients, April 2019 Efficacy Update Protocol: GO29365
______
BR (Ph II) Pola+BR (PhII) (N=40) (N=40)
______
Number of Evaluable Patients 40 40 Best Overall Response (CR/PR) 10 (25.0%) 25 (62.5%)
95% CI for Response Rate (Clopper-Pearson) (12.69, 41.20) (45.80, 77.27)
Difference in Response Rates 37.50 95% CI for Difference in Response Rates (15.82, 54.62) (Wilson) p-Value (Cochran-Mantel-Haenszel) 0.0005
Complete Response (CR) 9 (22.5%) 20 (50.0%) 95% CI (10.84, 38.45) (33.80, 66.20)
Partial Response (PR) 1 (2.5%) 5 (12.5%) 95% CI (0.06, 13.16) (4.19, 26.80)
Stable Disease (SD) 9 (22.5%) 5 (12.5%) 95% CI (10.84, 38.45) (4.19, 26.80)
Progressive Disease (PD) 8 (20.0%) 6 (15.0%) 95% CI (9.05, 35.65) (5.71, 29.84)
Missing or unevaluable 13 (32.5%) 4 (10.0%) 95% CI (18.57, 49.13) (2.79, 23.66)
______Stratification factors: Duration of Response to Prior Therapy from IxRS (<=12/>12 months). 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off Date: 30APR2018 Extract Date: 29JUN2018
Program: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/program/t_ef_rs.sa s Output: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/output/t_ef_rs_BOR IRC_rBRDL_IT_APR2019.out 17MAY2019 7:18 BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; CT: computed tomography; DLBCL: diffuse large B-cell lymphoma; IRC: Independent Review Committee; ITT: intention- to-treat; IxRS: interactive voice or Web-based response system; PD: progressive disease; PET-CT: positron emission tomography-computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; PR: partial response; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; SD: stable disease Source: Section 5.1.3.5, GO29365 Interim CSR.
INV-assessed A best response of CR or PR as assessed by the INV was achieved by 70.0% (28/40) of patients in the Pola BR arm and 32.5% (13/40) of patients in the BR arm while on study. The proportion of patients achieving a best response of CR was 57.5% (23/40) and 20.0% (8/40), respectively (see Table 30).
120 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 30 Best Overall Response Rate (INV-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) t_ef_rs_BOR_rBRDL_IT Summary of Best Overall Response Rate, RAN BR DLBCL Group, Intent-to-Treatment Patients Protocol: GO29365 BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) Number of Evaluable Patients 40 40 Best Overall Response (CR/PR) 13 (32.5%) 28 (70.0%)
95% CI for Response Rate (Clopper-Pearson) (18.57, 49.13) (53.47, 83.44)
Difference in Response Rates 37.50 95% CI for Difference in Response Rates (15.64, 54.71) (Wilson) p-Value (Cochran-Mantel-Haenszel) 0.0006
Complete Response (CR) 8 (20.0%) 23 (57.5%) 95% CI (9.05, 35.65) (40.89, 72.96)
Partial Response (PR) 5 (12.5%) 5 (12.5%) 95% CI (4.19, 26.80) (4.19, 26.80)
Stable Disease (SD) 2 (5.0%) 1 (2.5%) 95% CI (0.61, 16.92) (0.06, 13.16)
Progressive Disease (PD) 22 (55.0%) 7 (17.5%) 95% CI (38.49, 70.74) (7.34, 32.78)
Missing or unevaluable 3 (7.5%) 4 (10.0%) 95% CI (1.57, 20.39) (2.79, 23.66) Stratification factors: Duration of Response to Prior Therapy from IxRS (<=12/>12 months). 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off Date: 30APR2018 Extract Date: 29JUN2018 Program: /opt/BIOSTAT/prod/cdpt7898/s29365a/t_ef_rs.sas / Output: /opt/BIOSTAT/prod/cdpt7898/i29365h/reports/t_ef_rs_BOR_rBRDL_IT.out 18SEP2018 0:37 Page 1 of 1 BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; CSR: Clinical Study Report; CT: computed tomography; DLBCL: diffuse large B-cell lymphoma; INV: investigator; ITT: intention- to-treat; IxRS: interactive voice or Web-based response system; PD: progressive disease; PET-CT: positron emission tomography-computed tomography; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; PR: partial response; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma; SD: stable disease Source: Section 5.1.3.4, GO29365 Interim CSR.
4.4.2.6 DOR as Assessed by IRC and INV IRC-assessed Among patients who achieved an OR (CR/PR) at any time (Pola BR arm: 25 patients and BR arm: 10 patients), 13 patients in the Pola BR arm (52.0% of the responders) and eight patients in the BR arm (80.0% of the responders) subsequently had PD or died (see Table 31).
The median IRC-assessed DOR was 12.6 months in the Pola BR arm (95% CI: [7.2 months, NE]) compared to 7.7 months in the BR arm (95% CI: [4.0 months, 18.9 months]) (see Table 31).
The risk of responders progressing or dying was reduced by 53% in patients treated with Pola BR compared to BR (stratified HR 0.47; 95% CI: [0.19, 1.14]) (see Table 31).
121 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 31 Duration of Response (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) Summary of Duration of Response by IRC, RAN BR DLBCL Group, Intent-to-Treatment Patients, April 2019 Efficacy Update Protocol: GO29365
______
BR (Ph II) Pola+BR (Ph II) (N=40) (N=40) ______
Patients with event (%) 8 (80.0%) 13 (52.0%) Earliest contributing event Death 5 6 Disease Progression 3 7 Patients without event (%) 2 (20.0%) 12 (48.0%)
Time to event(month) Median 7.655 12.616 95% CI (3.975, 18.891) (7.162, NE) 25% and 75%-ile 3.975, 18.891 5.684, NE Range 0.03* to 19.65 0.59 to 22.54*
Unstratified Analysis p-value (log-rank) 0.1133
Hazard Ratio 0.50 95% CI (0.21, 1.20)
Stratified Analysis p-value (log-rank) 0.0889
Hazard Ratio 0.47 95% CI (0.19, 1.14)
6 months duration Patients remaining at risk 5 17 Event Free Rate (%) 55.56 71.24 95% CI (23.09, 88.02) (53.22, 89.26)
Difference in Event Free Rate -15.68 95% CI (-52.81, 21.45) p-value (Z-test) 0.4078
9 months duration Patients remaining at risk 4 14 Event Free Rate (%) 44.44 58.67 95% CI (11.98, 76.91) (39.00, 78.34)
Difference in Event Free Rate -14.22 95% CI (-52.18, 23.74) p-value (Z-test) 0.4627
12 months duration Patients remaining at risk 4 12 Event Free Rate (%) 44.44 50.29 95% CI (11.98, 76.91) (30.29, 70.28)
Difference in Event Free Rate -5.84 95% CI (-43.97, 32.29) p-value (Z-test) 0.7640
18 months duration Patients remaining at risk 2 7 Event Free Rate (%) 33.33 46.10 95% CI (2.54, 64.13) (26.15, 66.04)
Difference in Event Free Rate -12.76 95% CI (-49.45, 23.93) p-value (Z-test) 0.4954
24 months duration Patients remaining at risk NE NE Event Free Rate (%) NE NE 95% CI NE NE
Difference in Event Free Rate NE 95% CI NE p-value (Z-test) NE
122 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 31 Duration of Response (IRC-assessed) in Patients with R/R DLBCL Treated with Pola BR or BR (Randomized Phase II; ITT Population) (cont.) Stratification factors: Duration of Response to Prior Therapy, IxRS (<=12/>12 months). * Censored at time of data cut off, ^ Censored and event 95% CI for rates were constructed using Clopper-Pearson method. 95% CI for difference in rates were constructed using the normal approximation to the binomial distribution. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off Date: 30APR2018 Extract Date: 29JUN2018
Program: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/program/t_ef_tte.s as Output: root/clinical_studies/RO5541077/CDPT7898/GO29365/data_analysis/efficacy_update/qa/output/t_ef_tte_DO RIRC_rBRDL_IT_APR2019.out 17MAY2019 7:21 BR: bendamustine plus rituximab; CI: confidence interval; CT: computed tomography; DLBCL: diffuse large B-cell lymphoma; DOR: duration of response; IRC: Independent Review Committee; ITT: intention-to-treat; IxRS: interactive voice or Web-based response system; NE: not evaluable; PET-CT: positron emission tomography-computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma Source: Section 5.1.3.6, GO29365 Interim CSR.
INV-assessed Among patients who achieved an OR (CR/PR) at any time (Pola BR arm: 28 patients and BR arm: 13 patients), 17 patients in the Pola BR arm (60.7% of the responders) and 11 patients in the BR arm (84.6% of the responders) subsequently had PD or died.
The median DOR was over two-fold the duration in the Pola BR arm (10.3 months; 95% CI: [5.6 months, NE]) compared to the BR arm (4.1 months; 95% CI: [2.6 months, 12.7 months]).
The risk of responders progressing or dying after response to treatment was reduced by 56% in patients treated with Pola BR compared to BR (stratified HR 0.44; 95% CI: [0.2, 0.95]).
Full results including a KM plot of DOR for Pola BR and BR arms for patients with R/R DLBCL in the randomized Phase II portion of the GO29365 study are provided in Section 5.1.4.1, GO29365 Interim CSR.
4.4.2.7 Sensitivity Analyses – PFS by IRC To assess the robustness of the results of PFS by IRC in the randomized Phase II portion of the GO29365 study, two additional censoring rules were applied.
For the patients who had missed one or more assessments before their recorded event of disease progression or death, the data were censored at the date of the last non-missing disease assessment prior to the events. For patients who started new anti-lymphoma therapies (NALTs) prior to disease progression, the data were censored at the date of the last non-missing disease assessments before the NALT. Estimated
123 PTJA06 – Submission Dossier by Roche Registration GmbH
PFS by an IRC using these additional censoring rules are consistent with the results of those using the standard PFS censoring rule. Further details are provided in Section 5.1.5, GO29365 Interim CSR.
4.4.2.8 Multiple-Cox Regression To further assess the robustness of the treatment effects observed in the Pola BR arm compared to the BR arm in the randomized Phase II portion of the GO29365 study, multiple-Cox regression analyses were conducted for PFS and OS. The results from the stratified multiple-Cox regression analysis show that after adjusting for the potential prognostic factors and baseline characteristics, the treatment effect on PFS and OS between Pola BR versus BR remains robust. Further details are provided in Section 5.1.6, GO29365 Interim CSR.
4.4.3 Subgroup Analyses 4.4.3.1 Subgroup Analysis at the Primary Response Assessment by Primary Reasons for Stem Cell Transplant Ineligibility Subgroup analysis by primary reason of transplant ineligibility (grouped by patient characteristics and disease characteristics) for OS, PFS, and response at the PRA, are presented in Table 32 to Table 34. For patients who were transplant ineligible due to patient characteristics treated with Pola BR compared to BR, the IRC-assessed CR rate was 57.1% versus 18.2%; median INV-assessed PFS was 14.8 months versus 2.5 months, IRC-assessed PFS was not reached versus 4.5 months, and mOS was not reached versus 5.1 months, respectively. For patients who were transplant ineligible due to disease characteristics and treated with Pola BR compared to BR, the IRC-assessed CR rates were 31.8% versus 13.3%, median INV-assessed PFS was 6.0 months versus 1.9 months, IRC-PFS was 7.4 months versus 2.8 months, and mOS was 11.5 months versus 3.5 months.
In conclusion, improved OS, PFS and CR rates are seen in the Pola BR arm compared to BR in all patients regardless of whether the primary reason for transplant ineligibility was related to patient characteristics or disease characteristics. This further supports the clinical benefit of the Pola BR regimen for R/R DLBCL patients who are not candidates for SCT.
124 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 32 Summary of Overall Survival by Primary Reason for Stem Cell Transplant Ineligibility
Overall Survival BR Pola BR Primary Reason – Age, comorbidity, performance status No. of patients n 22 n 14 mOS, mo. 5.1 NE (95% CI) (3.9, 8.3) (7.7, NE) Hazard Ratio (95% CI) 0.27 (0.10, 0.72) 12-month OS rate 20.6% 64.3% 18-month OS rate 13.7% 57.1% Primary Reason – Insufficient response to salvage therapy, failed prior transplant No. of patients n 15 n 22 Median OS, mo. 3.5 11.5 (95% CI) (0.9, 5.3) (6.2, 16.8) Hazard Ratio (95% CI) 0.17 (0.06, 0.44) 12-month OS rate 16.3% 43.8% 18-month OS rate 16.3% 29.2% BR: bendamustine plus rituximab; CI: confidence interval; CT: computed tomography; mo. month; mOS: median overall survival; NE: not estimable; OS: overall survival; PET-CT: positron emission tomography-computed tomography; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; SCT: stem cell transplantation. 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Source: t_ef_tte_SCTINEL_OS_rBRDL_IT (see Appendix 4).
125 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 33 Summary of Progression-Free Survival by Primary Reason for Stem Cell Transplant Ineligibility
Progression Free Survival INV Assessment IRC Assessment BR Pola BR BR Pola BR Primary Reason – Age, comorbidity, performance status No. of patients n 22 n 14 n 22 n 14 mPFS, mo. 2.5 14.8 4.5 NE (95% CI) (1.5, 5.0) (7.7, NE) (2.1, 5.1) (4.7, NE) Hazard Ratio (95% CI) 0.15 (0.05, 0.45) 0.19 (0.06, 0.59) 12-month event-free rate 5.2% 57.1% 11.3% 57.1% 18-month event-free rate NE 41.7% 5.7% 50.0% Primary Reason – Insufficient response to salvage therapy, failed prior transplant No. of patients n 15 n 22 n 15 n 22 mPFS, mo. 1.9 6.0 2.8 7.4 (95% CI) (0.4, 3.1) (3.5, 7.6) (0.9, 3.9) (4.5, 13.4) Hazard Ratio (95% CI) 0.18 (0.07, 0.46) 0.20 (0.07, 0.54) 12-month event-free rate 14.3% 24.5% 9.2% 31.7% 18-month event-free rate 14.3% 19.6% 9.2% 21.1% BR: bendamustine plus rituximab; CI: confidence interval; CT: computed tomography; INV: investigator; IRC: Independent Review Committee; mo: month; mPFS: median progression-free survival; NE: not estimable; PET-CT: positron emission tomography-computed tomography; PFS: progression-free survival; Pola + BR; polatuzumab vedotin in combination with bendamustine plus rituximab; SCT; stem cell transplantation 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Source: t_ef_tte_SCTINEL_PFSIRC3_rBRDL_IT; t_ef_tte_SCTINEL_PFS_rBRDL_IT (see Appendix 5).
126 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 34 Summary of PET-CT Response at the Primary Response Assessment by Primary Reasons for Stem Cell Transplant Ineligibility
Parameter INV Assessment IRC Assessment Response Rate at PRA BR Pola BR BR Pola BR Primary Reason– Age, comorbidity, performance status No. evaluable patients n 22 n 14 n 22 n 14 OR (CR/PR) 4 (18.2%) 10 (71.4%) 4 (18.2%) 10 (71.4%) 95% CI for Response Rate (5.2, 40.3) (41.9, 91.6) (5.2, 40.3) (41.9, 91.6) (Clopper Pearson) CR 3 (13.6%) 8 (57.1%) 4 (18.2%) 8 (57.1%) (95% CI) (2.9, 34.9) (28.9, 82.3) (5.2, 40.3) (28.9, 82.3) Primary Reason– Insufficient response to salvage therapy, failed prior transplant No. evaluable patients n 15 n 22 n 15 n 22 OR (CR/PR) 2 (13.3%) 8 (36.4%) 2 (13.3%) 7 (31.8%) 95% CI for Response Rate (1.7, 40.5) (17.2, 59.3) (1.7, 40.5) (13.9, 54.9) (Clopper Pearson) CR 2 (13.3%) 8 (36.4%) 2 (13.3%) 7 (31.8%) (95% CI) (1.7, 40.5) (17.2, 59.3) (1.7, 40.5) (13.9, 54.9) BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; INV: investigator; IRC: Independent Review Committee; OR: overall response; PET-CT: positron emission tomography- computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; PR: partial response; PRA: Primary Response Assessment; SCT: stem cell transplantation 95% CI for rates were constructed using Clopper-Pearson method. Cutoff date: 30 April 2018. Extract date: 29 June 2018. Source: t_ef_rs_SCTINEL_PRAIRCPET_rBRDL_IT; t_ef_rs_SCTINEL_PRAPET_rBRDL_IT (see Appendix 6).
4.4.3.2 Subgroup Analyses by Number of Lines of Prior Treatment Subgroup analysis by number of lines of prior treatment for OS, PFS, and response at the PRA are presented in Table 35 to Table 37. Overall survival, PFS, and PET-CR rates at the PRA (end of treatment) were higher in the Pola BR arm compared to the BR arm regardless of the number of lines of prior treatment. Complete response rates were observed to be higher with fewer lines of therapy.
Efficacy results were consistent across patients after first, second, third or later relapse in terms of showing better efficacy in the Pola BR arm compared to the BR arm. There is no plausible reason to assume that only later lines of treatment would benefit from the addition of polatuzumab vedotin to BR, neither from a mechanistic or biologic point of view.
127 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 35 Summary of Overall Survival by Prior Line of Therapy (1, 2, 3 or more)
BR Pola + BR (N 40) (N 40) All lines, n 40 40 Median OS, mo. (95% CI) 4.7 (3.7, 8.3) 12.4 (9.0, NE) 1 prior line, n 12 11 Median OS, mo. (95% CI) 5.9 (4.7, 8.4) NE (10.5, NE) HR (95% CI) 0.29 (0.05, 1.64) 9-month OS rate 13.9% 80.0% 12-month OS rate 13.9% 70.0% 18-month OS rate 13.9% 60.0% 2 prior lines, n 9 11 Median OS, mo. (95% CI) 6.0 (3.7, 18.3) 8.9 (6.2, NE) HR (95% CI) 0.46 (0.14, 1.47) 9-month OS rate 25.9% 45.5% 12-month OS rate 25.9% 36.4% 18-month OS rate 25.9% 36.4% 3 or more prior lines, n 19 18 Median OS, mo. (95% CI) 3.7 (2.4, 5.3) 12.4 (6.2, NE) HR (95% CI) 0.37 (0.16, 0.85) 9-month rate 26.6% 70.6% 12-month OS rate 26.6% 51.3% 18-month OS rate 19.9% 32.1% BR: bendamustine plus rituximab; CI: confidence interval; HR: hazard ratio; mo. month; NE: not evaluable; OS: overall survival; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off date: 30 April 2018. Extract date: 29 June 2018. Source: t_ef_tte_PRIATYML_OS_rBRDL_IT (see Appendix 7).
128 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 36 Summary of Progression-Free Survival by Prior Line of Therapy (1, 2, 3 or more)
INV Assessment IRC Assessment BR Pola + BR BR Pola + BR (N 40) (N 40) (N 40) (N 40) All lines, n 40 40 40 40 Median PFS, mo. (95% 2.0 (1.5, 3.7) 7.6 (6.0, 17.0) 3.7 (2.4, 4.5) 11.1 (6.2, CI) 13.9) 1 prior line, n 12 11 12 11 Median PFS, mo. (95% 3.9 (1.4, 5.1) NE (10.4, NE) 4.7 (1.9, 5.8) 13.6 (10.4, CI) NE) HR (95% CI) 0.18 (0.04, 0.93) 0.22 (0.04, 1.18) 9-month event-free rate 10.4% 80.0% 10.4% 80.0% 12-month event-free rate NE 60.0% 10.4% 60.0% 18-month event-free rate NE 50.0% 10.4% 40.0% 2 prior lines, n 9 11 9 11 Median PFS, mo. (95% 1.9 (0.5, 2.8) 6.7 (1.4, 19.0) 3.8 (0.4, 6.2 (1.5, NE) CI) 10.0) HR (95% CI) 0.40 (0.14, 1.11) 0.38 (0.12, 1.19) 9-month event-free rate 22.2% 36.4% 25.0% 36.4% 12-month event-free rate 11.1% 36.4% 12.5% 36.4% 18-month event-free rate 11.1% 27.3% 12.5% 36.4% 3 or more prior lines, n 19 18 19 18 Median PFS, mo. (95% 2.0 (1.5, 3.1) 6.3 (2.6, 11.5) 3.2 (0.9, 4.1) 9.0 (2.6, 13.9) CI) HR (95% CI) 0.32 (0.15, 0.70) 0.36 (0.16, 0.80) 9-month event-free rate 16.7% 32.7% 14.4% 50.3% 12-month event-free rate 16.7% 26.1% 14.4% 36.0% 18-month event-free rate 11.1% 19.6% 7.2% 21.6% BR: bendamustine plus rituximab; CI: confidence interval; HR: hazard ratio; INV: investigator; IRC: Independent Review Committee: mo. month; NE: not evaluable; PFS: progression-free survival; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab. 95% CI for rates were constructed using Clopper-Pearson method. Tumor assessment result is based on PET-CT whenever it is available and valid and uses CT only result if PET-CT result is missing. Cut-off date: 30 April 2018. Extract date: 29 June 2018. Source: t_ef_tte_PRIATYML_PFSIRC3_rBRDL_IT; t_ef_tte_PRIATYML_PFS_rBRDL_IT (Appendix 8)
129 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 37 Summary of PET-CT Complete Response at the Primary Response Assessment by Prior Line of Therapy (1, 2, 3 or more)
INV Assessment IRC Assessment BR Pola BR BR Pola BR (N 40) (N 40) (N 40) (N 40) All lines, n 40 40 40 40 CR 6 (15.0%) 17 (42.5%) 7 (17.5%) 16 (40.0%) (95% CI for CR Rate) (5.7, 29.8) (27.0, 59.1) (7.3, 32.8) (24.9, 56.7) 1 prior line, n 12 11 12 11 CR 1 (8.3%) 7 (63.6%) 2 (16.7%) 7 (63.6%) (95% CI for CR Rate) (0.2, 38.5) (30.8, 89.1) (2.1, 48.4) (30.8, 89.1) 2 prior lines, n 9 11 9 11 CR 2 (22.2%) 5 (45.5%) 2 (22.2%) 4 (36.4%) (95% CI for CR Rate) (2.8, 60.0) (16.8, 76.6) (2.8, 60.0) (10.9, 69.2) 3 or more prior lines, n 19 18 19 18 CR 3 (15.8%) 5 (27.8%) 3 (15.8%) 5 (27.8%) (95% CI for CR Rate) (3.4, 39.6) (9.7, 53.5) (3.4, 39.6) (9.7, 53.5) BR: bendamustine plus rituximab; CI: confidence interval; CR: complete response; INV: investigator; IRC: Independent Review Committee: PET-CT: positron emission tomography-computed tomography; Pola + BR: polatuzumab vedotin in combination with bendamustine plus rituximab; PRA: primary response assessment 95% CI for rates were constructed using Clopper-Pearson method. Cut-off date: 30 April 2018. Extract date: 29 June 2018. Source: t_ef_rs_PRIATYML_PRAIRCPET_rBRDL_IT; t_ef_rs_PRIATYML_PRAPET_rBRDL_IT (see Appendix 9).
4.4.4 Patient-Reported Outcomes 4.4.4.1 Therapy-Induced Neuropathy Assessment Score (TINAS) Patients followed the same schedule for PRO assessments during treatment regardless of treatment arm in both the Phase Ib and Phase II portions of the study. Total scores could be prorated (estimated) for the total score, as described in Section 4.2.6, in the event of missing item level data.
The scheduling of PRO using TINAS was changed from daily in the first version of the protocol to weekly in the second and subsequent versions of the protocol to reduce patient burden. As development of neuropathy was not expected to occur early in treatment or vary significantly from day-to-day, weekly assessments were judged to be adequate to capture symptom onset and variability. While reprogramming the electronic patient-reported outcome devices (ePROs) to reflect the scheduling change, the baseline assessment (Cycle 1 Day 1) was inadvertently made inaccessible to patients. This error was not identified and corrected for 9 months, leading to a significant loss of baseline data (20.8% [5/24 patients] and 29.4% [59/201 patients] completed baseline in
130 PTJA06 – Submission Dossier by Roche Registration GmbH
the Phase Ib and Phase II randomized portion, respectively). For the reported analyses, available data were summarized at the assessment point at which they were collected, with no attempt to move the baseline assessment or impute missing baseline data.
Patient-reported outcome completion over the course of the trial was generally limited for both Phase Ib and Phase II randomized portions of the GO29365 study. Due to limited post-treatment follow-up data, results are presented here only up to the end of treatment. Compliance in the BR arm declined more rapidly than in the Pola BR arm.
Summaries of descriptive statistics for mean/median values and changes from baseline of the TINAS items, the total score during each week of treatment and at the end of treatment, are provided in Section 6.1, GO29365 Interim CSR. Available sample sizes for the Pola BR and BR arms varied across assessments but showed steady declines over the course of treatment. Scores after Week 11 and Week 29 were based on fewer than 25% of patients in the ITT population for the BR and Pola BR arms, respectively, and should generally be interpreted with caution. Mean scores for individual TINAS items were low (1.5) at the beginning of treatment in both the Pola BR and BR arms, indicating that patients were experiencing relatively little peripheral neuropathy burden (range is 0 to 10 for each item, see Section 3.8.3.2, GO29365 Interim CSR), and generally stayed low during treatment. By the end of treatment, the severity of peripheral neuropathy symptoms was rated as low, with the highest means observed for numbness/tingling in hands/feet, although still 2.
Trends in mean scores over time were further explored graphically in longitudinal plots for each TINAS item and for the total score. The plot of total score showed relatively little change over time in either the Pola BR arm or the BR arm. Several sensory (i.e. hot/burning sensations in hands/feet, pins/needles in arms/legs, numbness/tingling in hands/feet, cramps in hands/feet) and motor items (i.e. trouble grasping small objects, trouble walking, and difficulty with balance) showed slight elevations during treatment with Pola BR, whereas trajectories for the remaining items were largely flat. The mean scores rarely exceeded 3.0 during treatment, indicating that overall, patients perceived peripheral neuropathy to be mild. Trends for the BR arm across all items were relatively flat. Full results are provided in Section 6.1, GO29365 Interim CSR.
4.4.4.2 Supportive Data from the ROMULUS Study Another data source available for PROs in R/R DLBCL patients treated with polatuzumab vedotin is the ROMULUS study. The ROMULUS study evaluated the efficacy and safety of polatuzumab vedotin at a higher dose (Pola [2.4 mg/kg]) in combination with rituximab (R) in patients with R/R DLBCL.
In the ROMULUS study, the M.D. Anderson Symptom Inventory (MDASI) (Cleeland et al. 2000) was collected pre-dose during the first eight cycles of rituximab polatuzumab vedotin (R Pola [2.4 mg/kg]) treated R/R DLBCL. The MDASI is a multi-symptom self-report measure for clinical and research use. The MDASI’s 13 core-symptom items,
131 PTJA06 – Submission Dossier by Roche Registration GmbH
plus an additional 4 items, for a total of 17 symptom items, include those found to have the highest frequency and/or severity in patients with various cancers and treatment types. These include pain, fatigue, nausea, disturbed sleep, emotional distress, shortness of breath, lack of appetite, drowsiness, dry mouth, sadness, vomiting, constipation, diarrhea, mouth/throat sores, difficulty remembering, weakness in arms/legs, and numbness or tingling. Six additional items focus on the degree of interference of the aforementioned symptoms for a total of 23 items in the questionnaire. Each item is scored on a 0 to 10 numeric rating scale in which 0 is “Not present” and 10 is “As bad as you can imagine”, using a recall period of 24 hours. The symptom and interference items can be averages to create symptom severity and symptom interference scales, respectively, with a resulting score range of 010. The MDASI PRO instrument was supplied in the local language and electronic handheld computers were used for the daily collection of symptoms derived from the MDASI.
Patients reported little symptom burden at baseline with low mean scores observed for the MDASI symptom severity and symptom interference scales in the R Pola treatment arm (n 20; Mean 1.60; SD 1.49, and n 20; Mean 2.90, SD 3.11, respectively).
The mean (SE) change from baseline by assessment visit for select symptoms (i.e. fatigue, numbness/tingling, pain, and weakness in arms/legs) (see Figure 18) and the symptom severity and interference scale (see Figure 19) scores are depicted graphically in longitudinal plots.
132 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 18 Mean Change (SE) in MDASI Scores from Baseline by Symptom (DLBCL R Pola [2.4 mg/kg])
DLBCL: diffuse large B-cell lymphoma; MDASI: M.D. Anderson Symptom Inventory; R + Pola: rituximab + polatuzumab vedotin.
Figure 19 Mean Change (SE) in MDASI Scores from Baseline – Severity and Interference Scales (DLBCL R Pola [2.4 mg/kg])
DLBCL: diffuse large B-cell lymphoma; MDASI: M.D. Anderson Symptom Inventory; R + Pola: rituximab + polatuzumab vedotin.
133 PTJA06 – Submission Dossier by Roche Registration GmbH
The longitudinal plots show relatively little change in individual item scores for fatigue, pain, or in either the symptom severity or interference scales over the course of treatment in either arm. There was a steady increase in numbness/tingling over time. These increases were still relatively small with mean changes of 0.99 at six cycles and 2.27 at eight cycles. Weakness in arms/legs increased slightly in the R Pola (2.4 mg/kg) arm over the course of treatment. The sensitivity analysis using a longitudinal mixed effect model as well as pooling of the data from R Pola treated patients with follicular lymphoma provided similar results. Taken together, these results indicate that symptom burden and interference remained relatively low while on treatment with R Pola (2.4 mg/kg). The greatest change experienced by patients was an increase in numbness/tingling which was reported as mild, on average.
Given that very minimal peripheral neuropathy symptom burden and interference was seen with the R Pola (2.4 mg/kg), one could assume that similar, if not lower (i.e. better) values, would be observed with a lower dose of polatuzumab vedotin, such as 1.8 mg/kg. This was confirmed with the results from the TINAS instrument which showed that patients experienced minimal peripheral neuropathy in the Pola BR arm of the GO29365 study.
4.5 IMMUNOGENICITY The ADAs were measured to assess the immunogenicity of polatuzumab vedotin (see Section 3.8.5, GO29365 Interim CSR). The ADAs against rituximab were not monitored in this study as the immunogenicity of rituximab in patients with NHL has historically been low. Full immunogenicity results are provided in Section 8, GO29365 Interim CSR and a summary is provided below.
The overall incidence of treatment-emergent (sum of treatment-induced and treatment-enhanced) ADAs in all polatuzumab vedotin treatment groups, was 6.0% (8/134 ADA-evaluable patients [patients who had an ADA result from at least one post-baseline sample]). Of the eight ADA-positive patients, four were R/R DLBCL patients. Three out of the four ADA-positive R/R DLBCL patients developed treatment- emergent ADAs to polatuzumab vedotin by Cycle 2 and one patient developed ADA after the treatment period. The ADA titers ranged from 2.05 to 2.71 titer units, with a median time to onset of approximately 2.93 weeks after the first polatuzumab vedotin dose. The one R/R DLBCL patient who presented with ADA at Cycle 2, and was diagnosed with PD at Cycle 3, had a titer of 2.26 titer units at Cycle 2; no additional ADA data were available.
False-negative results due to drug interference are unlikely, as levels of circulating ADC were below levels expected to interfere in the assay, based on the drug tolerance profile of the assay, based on the drug tolerance profile of the assay. The low incidence of ADAs in patients receiving polatuzumab vedotin suggests that the immunogenicity potential is low, and this is not unexpected given that the MoA of polatuzumab vedotin is to target and kill B-cells. There appeared to be no impact of ADAs on polatuzumab
134 PTJA06 – Submission Dossier by Roche Registration GmbH
vedotin PK. In terms of safety, based on the current data, there is no identifiable relationship between ADA-positivity and reported AEs. In addition, the emergence of ADAs to polatuzumab vedotin did not appear to impact efficacy with ongoing long-term responses despite the development of ADAs. The study regimen has a fixed number of cycles therefore, once a patient achieves a response at the end of treatment, duration of efficacy after cessation of treatment should not be impacted by ADA development as maintenance of response is not reliant on further dosing of polatuzumab vedotin. Patients will continue to be closely monitored for any potential immune response to polatuzumab vedotin for all current and future studies.
4.6 SAFETY RESULTS IN PATIENTS WITH R/R DLBCL TREATED WITH POLA BR OR BR (RANDOMIZED PHASE II; SAFETY POPULATION) Include all the studies of the technology relevant to the assessment, as well as studies of comparator technologies (if applicable).
Include ongoing and unpublished studies when these data are available.
o Describe the relevant endpoints, including the definition of the endpoint and methods of analysis (table 14).
The study results presented should reflect the safety outcomes relevant to the assessment. If the endpoint uses a scale, state how it was validated; if this uses responder analyses, state and justify the responder definition.
o 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.
Example table 15 provides an overview of adverse events. Table 16 is given as an example of a more detailed presentation of the data. When presenting data specify: the number of patients, the number of events and the absolute and relative risk (with 95% confidence intervals). Order data by system class and frequency of events. For non-comparative studies complete only the column for the intervention.
Tables from regulatory documents providing the same information may also be used.
Repeat for each study providing safety data.
4.6.1 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) In no more than 6 bullet points describe key statements relating to the safety of the technology. For example, include statements about the relative safety of the technology compared to alternative technologies currently used.
135 PTJA06 – Submission Dossier by Roche Registration GmbH
The safety profile of polatuzumab vedotin in combination with BR in patients with R/R DLBCL, in the context of the clinically significant benefit observed and longer treatment duration, was acceptable and overall was consistent with that observed in the BR arm. An overview of the overall incidence of AEs by category as of the clinical cut-off date (30 April 2018) in patients with R/R DLBCL treated with Pola BR versus BR is presented below and in Table 38.
The duration of exposure to study treatment was longer in the Pola BR arm compared to the BR arm (Pola BR arm: median 5 cycles vs. BR arm: median three cycles). Cumulative exposure was also higher in the Pola BR arm compared with the BR arm, and more patients with R/R DLBCL completed their planned number of treatment cycles (46.2% vs. 23.1%, respectively) because of fewer early discontinuations due to disease progression.
Grade 3 AEs were reported at a higher overall incidence in patients treated with Pola BR (84.6% vs. 74.4%, respectively), with the difference driven mainly by a higher incidence of grade 3 or 4 cytopenias (neutropenia, thrombocytopenia, and anemia) in the Pola BR arm.
Fewer patients died due to disease progression in the Pola BR arm compared to the BR arm (14 patients vs. 17 patients, respectively). Overall, nine patients died due to AEs in the Pola BR arm and 11 patients died due to AEs in the BR arm. Four of the fatal AEs (Grade 5) in the Pola BR arm occurred during follow-up after study treatment completion or discontinuation, of which three Grade 5 AEs occurred following disease progression.
The overall incidence of SAEs was almost similar in the Pola BR and BR arms (64.1% vs. 61.5%, respectively), with no notable differences between the two treatment arms in the most frequently occurring serious toxicities of infections (by system organ class [SOC]) and neutropenia events including febrile neutropenia (by Standardized Medical Dictionary for Regulatory Activity Query [SMQ]).
In the setting of longer treatment exposure in the Pola BR arm, treatment discontinuations due to AEs were more frequent in the Pola BR arm compared to the BR arm (33.3% vs. 15.4%, respectively); the most frequent AEs leading to treatment discontinuation were cytopenias (neutropenia and thrombocytopenia).
Of selected AEs/AEs to monitor defined as identified and potential risks of polatuzumab vedotin:
o Neutropenia, thrombocytopenia, anemia (all Grade and Grade 3) were more frequently reported in the Pola BR arm compared to the BR arm but did not lead to significantly increased infections or red blood cell (RBC) or platelet transfusions.
136 PTJA06 – Submission Dossier by Roche Registration GmbH
o The incidence of infections (all Grades, Grade 3 and serious) was similar between the Pola BR and BR treatment arms, including four fatal infections in each arm. o Peripheral neuropathy events, conservatively assessed using the broader SMQ, including related medical conditions, were, as expected, more frequently reported in the Pola BR arm compared to the BR arm (43.6% vs. 7.7%; all Grade 1 or 2). A single patient discontinued polatuzumab vedotin due to muscle atrophy (Grade 1) and two patients had their polatuzumab vedotin dose reduced due to Grade 2 peripheral neuropathy. o Grade 3 hepatic toxicity events (by broad SMQs) were reported in two patients in the Pola BR arm and one patient in the BR arm. None satisfied the criteria to be considered indicative of drug-induced liver injury (as defined by Hy’s law).
137 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 38 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Phase Ib Phase II Phase Ib/II Study Phase (Randomized) Cohort/Arm 1a D C 1a C Pola BR BR Pola BR Pola BR Treatment (total) Sample size N 6 N 39 N 39 N 45 Total no. deaths 2 (33.3%) 28 (71.8%) 23 (59.0%) 25 (55.6%) Deaths due to PD 2 (100%) 17 (60.7%) 14 (60.9%) 16 (64%) Patients with at least one: Any AE 6 (100.0%) 38 (97.4%) 39 (100.0%) 45 (100.0%) Grade 34 AE 5 (83.3%) 28 (71.8%) 33 (84.6%) 38 (84.4%) Grade 5 0 11 (28.2%) 9 (23.1%) 9 (20.0%) Serious AE 4 (66.7%) 24 (61.5%) 25 (64.1%) 29 (64.4%) AE leading to discontinuation of: Pola 0 NA 12 (30.8%) 12 (26.7%) Any study drug 1 (16.7) 6 (15.4%) 13 (33.3%) 14 (31.1%) AE leading to modification/interruption
of: Pola 2 (33.3%) NA 22 (56.4%) 24 (53.3%) Any study drug 3 (50.0%) 19 (48.7%) 28 (71.8%) 31 (68.9%) AEs to monitor: Grade 2 peripheral neuropathy 0 2 (5.1%) 6 (15.4%) 6 (13.3%) Grade 3 neutropenia 2 (33.3%) 18 (46.2%) 23 (59.0%) 25 (55.6%) Grade 3 hepatotoxicity 0 1 (2.6%) 2 (5.1%) 2 (4.4%) Grade 3 infections and 2 (33.3%) 12 (30.8%) 13 (33.3%) 15 (33.3%) infestations AE: adverse event; BR: bendamustine rituximab; PD: progressive disease; Pola: polatuzumab vedotin. Investigator text for AEs encoded using MedDRA v21.0. Includes AEs with onset from first dose of study drug through 90 days after last dose of study drug Source: Section 9.1.1, GO29365 Interim CSR.
138 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2 Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
4.6.2.1 Extent of Exposure to Study Treatment in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) Planned treatment of patients with R/R DLBCL given Pola BR (Phase Ib safety run-in Cohort 1a and randomized Phase II Arm C) or BR (randomized Phase II Arm D) consisted of 6 21-day cycles during which a total of six doses of polatuzumab vedotin (1.8 mg/kg) (Pola BR), 12 doses of bendamustine (90 mg/m2), and six doses of rituximab (375 mg/m2) were to be administered (see Figure 7).
The extent of exposure of individual components of study treatment for patients with R/R DLBCL treated with the Pola BR and BR treatment regimens by study phase and overall is summarized in Table 39.
139 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 39 Summary of Extent of Exposure to Pola BR or BR in Patients with R/R DLBCL (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase Ib Phase II Total (Randomized) (Ph Ib/Ph II) Treatment Pola + BR BR Pola + BR Pola + BR Sample size N 6 N 39 N 39 N 45 Polatuzumab Vedotin Median treatment duration, months 2.4 (0.66-3.9) NE 3.2 (0.0-5.9) 3.2 (0.0-5.9) (range) Median no. of cycles (range) 4.5 (2-6) NE 5.0 (1-6) 5.0 (1-6) Mean cumulative dose, mg (SD) 628 (306) NE 599 (276) 604 (277) 101.5 97.3 99.5 Dose intensity, median % (range)* NE (87.2-103.7) (58.4-112.7) (58.4-112.7) Bendamustine Median treatment duration, months 2.4 1.4 3.2 3.2 (range) (0.7-3.9) (0.03-4.4) (0.03-5.1) (0.03-5.1) Median no. of cycles (range) 4.5 (2-6) 3.0 (1-6) 5.0 (1-6) 5.0 (1-6) 1426 989 1439.49 1437.71 Mean cumulative dose, mg (SD) (607) (584) (603.56) (597.06) 98.6 95.6 95.4 95.45 Dose intensity, median % (range)* (78.1-101.7) (63.6-103.1) (53.4-102.1) (53.4-102.1) Rituximab Median treatment duration, months 2.4 (0.7-4.0) 1.4 (0.0-4.4) 3.2 (0.0-6.0) 3.2 (0.0-6.0) (range) Median no. of cycles (range) 4.5 (2-6) 3.0 (1-6) 5.0 (1-6) 5.0 (1-6) Mean cumulative dose, mg (SD) 3127 (1382) 2097 (1280) 3099 (1290) 3103 (1287) 98.5 96.7 94.7 92.5 Dose intensity, median % (range)* (70.7-105.2) (49-102.7) (70.7-105.2) (70.7-105.2) * Adjusted for dose modification and delay. Source: Section 9.1.2, GO29365 Interim CSR.
140 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.2 Common Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) All patients with R/R DLBCL treated with polatuzumab vedotin had at least one AE during the study.
Summaries of all reported AEs by preferred term and SOC (see Appendix 10) and of the most common AEs with an incidence of 10% and 10% overall (Phase Ib/II combined; see Table 40 and Appendix 11) are provided. Adverse events reported by 30% of patients with R/R DLBCL treated with Pola BR were neutropenia, thrombocytopenia, fatigue, diarrhea, anemia, and pyrexia.
The most frequently affected SOCs (AEs affecting over 50% of patients) in the Pola BR arm were gastrointestinal disorders (80% [36/45 patients]), blood and lymphatic system disorders (75.6% [34/45 patients]), general disorders and administrative site conditions (68.9% [31/45 patients]), nervous system disorders and infections and infestations (53.3% [24/45 patients] each), and metabolism and nutrition disorders (51.1% [23/45 patients]).
In the randomized Phase II, AEs of any grade were reported in 100% of patients (39/39) in the Pola BR arm and 97.4% of patients (38/39) in the BR arm.
The AEs by preferred term with a 10% higher incidence in the Pola BR arm compared to the BR arm were neutropenia, thrombocytopenia, anemia, lymphopenia, diarrhea, vomiting, upper abdominal pain, pyrexia, chills, pneumonia, hypokalemia, hypoalbuminemia, peripheral neuropathy, dizziness, peripheral sensory neuropathy, and pruritis.
Adverse events in the following SOCs were more frequently reported (10% higher incidence) in the Pola BR arm compared to the BR arm: gastrointestinal disorders (82.1% vs. 64.1%), blood and lymphatic system disorders (82.1% vs. 64.1%), metabolism and nutrition disorders (46.2% vs. 46.2%), nervous system disorders (51.3% vs. 28.2%), and musculoskeletal and connective tissue disorders (41.0% vs. 28.2%).
141 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 40 Most Frequently Reported Adverse Events (10%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase I Phase II Phase Ib/II (randomized) Treatment Pola BR BR Pola BR Pola BR (total) Sample size N 6 N 39 N 39 N 45 Total number of patients with at least 6 (100.0%) 38 (97.4%) 39 (100.0%) 45 (100.0%) one AE: Blood and Lymphatic System Disorders Neutropenia 0 15 (38.5%) 21 (53.8%) 21 (46.7%) Anemia 0 10 (25.6%) 21 (53.8%) 21 (46.7%) Thrombocytopenia 2 (33.3%) 11 (28.2%) 19 (48.7%) 21 (46.7%) Febrile neutropenia 1 (16.7%) 5 (12.8%) 4 (10.3%) 5 (11.1%) Leukopenia 0 5 (12.8%) 5 (12.8%) 5 (11.1%) Lymphopenia 0 0 5 (12.8%) 5 (11.1%) Pancytopenia 1 (16.7%) 0 2 (5.1%) 3 (6.7%) Gastrointestinal Disorders Diarrhea 2 (33.3%) 11 (28.2%) 15 (38.5%) 17 (37.8%) Nausea 3 (50.0%) 16 (41.0%) 12 (30.8%) 15 (33.3%) Constipation 1 (16.7%) 8 (20.5%) 7 (17.9%) 8 (17.8%) Vomiting 1 (16.7%) 5 (12.8%) 7 (17.9%) 8 (17.8%) Abdominal pain 1 (16.7%) 4 (10.3%) 4 (10.3%) 5 (11.1%) Abdominal pain upper 0 2 (5.1%) 5 (12.8%) 5 (11.1%) Abdominal distension 1 (16.7%) 0 1 (2.6%) 2 (4.4%) General Disorders and Administration Site Conditions Fatigue 4 (66.7%) 14 (35.9%) 14 (35.9%) 18 (40.0%) Pyrexia 2 (33.3%) 9 (23.1%) 13 (33.3%) 15 (33.3%) Chills 1 (16.7%) 3 (7.7%) 4 (10.3%) 5 (11.1%) Asthenia 1 (16.7%) 6 (15.4%) 4 (10.3%) 5 (11.1%) Edema, peripheral 0 3 (7.7%) 2 (5.1%) 2 (4.4%) Infections and Infestations Pneumonia 2 (33.3%) 4 (10.3%) 5 (12.8%) 7 (15.6%) Herpes zoster 1 (16.7%) 2 (5.1%) 1 (2.6%) 2 (4.4%) Upper respiratory tract infection 2 (33.3%) 1 (2.6%) 2 (5.1%) 4 (8.9%) Urinary tract infection 1 (16.7%) 2 (5.1%) 1 (2.6%) 2 (4.4%)
142 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 40 Most Frequently Reported Adverse Events (10%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) (cont.)
Pola BR Treatment Pola BR BR Pola BR (total) Investigations Weight decreased 2 (33.3%) 3 (7.7%) 5 (12.8%) 7 (15.6%) Metabolism and Nutrition Disorders Decreased appetite 2 (33.3%) 8 (20.5%) 10 (25.6%) 12 (26.7%) Hypokalemia 3 (50.0%) 3 (7.7%) 4 (10.3%) 7 (15.6%) Hypoalbuminemia 1 (16.7%) 2 (5.1%) 5 (12.8%) 6 (13.3%) Hypocalcemia 2 (33.3%) 1 (2.6%) 3 (7.7%) 5 (11.1%) Hypomagnesemia 2 (33.3%) 4 (10.3%) 1 (2.6%) 3 (6.7%) Dehydration 2 (33.3%) 0 2 (5.1%) 4 (8.9%) Hypophosphatemia 1 (16.7%) 1 (2.6%) 2 (5.1%) 3 (6.7%) Musculoskeletal and Connective Tissue Disorders Back pain 0 4 (10.3%) 2 (5.1%) 2 (4.4%) Bone pain 0 0 0 0 Muscular weakness 0 1 (2.6%) 2 (5.1%) 2 (4.4%) Pain in extremity 0 2 (5.1%) 2 (5.1%) 2 (4.4%) Nervous System Disorders Peripheral neuropathy 0 1 (2.6%) 9 (23.1%) 9 (20.0%) Dizziness 1 (16.7%) 3 (7.7%) 5 (12.8%) 6 (13.3%) Peripheral sensory neuropathy 0 0 6 (15.4%) 6 (13.3%) Headache 1 (16.7%) 2 (5.1%) 3 (7.7%) 4 (8.9%) Paresthesia 0 0 2 (5.1%) 2 (4.4%) Psychiatric Disorders Anxiety 0 2 (5.1%) 3 (7.7%) 3 (6.7%) Respiratory, Thoracic and Mediastinal Disorders Cough 1 (16.7%) 8 (20.5%) 6 (15.4%) 7 (15.6%) Dyspnea 0 2 (5.1%) 3 (7.7%) 3 (6.7%) Pleural effusion 1 (16.7%) 4 (10.3%) 2 (5.1%) 3 (6.7%) Skin and Subcutaneous Tissue Disorders Pruritus 1 (16.7%) 4 (10.3%) 5 (12.8%) 6 (13.3%) Rash 1 (16.7%) 5 (12.8%) 2 (5.1%) 3 (6.7%) Vascular Disorders Hypotension 1 (16.7%) 5 (12.8%) 3 (7.7%) 4 (8.9%)
143 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 40 Most Frequently Reported Adverse Events (10%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) (cont.)
* Table only shows preferred terms reported in 10% of all patients with R/R DLBCL treated with either Pola BR (at least 5/45 patients in Phase Ib/II combined). Source: Section 9.1.3.1, GO29365 Interim CSR.
4.6.2.3 Adverse Events by Severity in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) An overall summary of AE severity by NCI CTCAE grading in patients with R/R DLBCL treated with Pola BR and BR is shown in Table 41.
While more AEs were reported per patient for patients with R/R DLBCL treated with polatuzumab vedotin compared to BR, the distribution of AE severity by Grade was balanced across treatment groups.
The majority of AEs were Grade 1 or 2 in severity (approximately 70% of the total number of reported events for each treatment arm).
The majority of patients had at least one Grade 3 AE (see Section 4.6.2.4). The proportion of patients with Grade 4 events was slightly higher in patients treated with Pola BR compared to BR (37.8% vs. 25.6%; Table 41).
Table 41 Summary of Adverse Events by Highest NCI CTCAE Grade in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
NCI CTCAE Grade Total 1 2 3 4 5 BR (n 39) Pts with at least one 38 (97.4%) 5 (12.8%) 4 (10.3%) 8 (20.5%) 10 (25.6%) 11 (28.2%) AE Total no. AEs 446 188 136 78 33 11 (Phase II) Pola + BR (n 45) Pts with at least one 45 (100%) 1 (2.2%) 6 (13.3%) 12 (26.7%) 17 (37.8%) 9 (20.0%) AE Total no. AEs 97/614 59/264 28/158 8/130 2/53 0/9 (Phase Ib/II) AE: adverse event; NCI CTCAE: National Cancer Institute Common Terminology Criteria for Adverse Events Multiple occurrences of the same AE in one individual are counted once at the maximum intensity. Source: Section 9.1.4, GO29365 Interim CSR.
144 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.4 Grade 3 Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) A total of 202 Grade 35 AEs were reported in 84.4% (38/45) of patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined).
Grade 35 AEs were most frequently reported (20% patients) from the SOCs of blood and lymphatic disorders (64.4%), infection and infestations (28.9%), and gastrointestinal disorders (22.2%).
The most frequently reported Grade 35 AEs by preferred term in patients with R/R DLBCL treated with Pola BR (10% patients in Phase Ib/II combined) were neutropenia, thrombocytopenia, anemia, febrile neutropenia, and lymphopenia (all Grade 3 or 4) and were all from the SOC blood and lymphatic system disorders. Grade 35 AEs reported at a frequency of 5% of patients (in at least 3/45 patients with R/R DLBCL treated with Pola BR in Phase Ib/II combined) are shown in Table 42.
In the randomized Phase II, the incidence of Grade 34 events was higher in the Pola BR arm compared to the BR arm (84.6% vs. 71.8%). Grade 34 AEs by preferred term with a 10% higher incidence in the Pola BR arm compared to the BR arm were neutropenia, thrombocytopenia, anemia, febrile neutropenia, and lymphopenia (all Grade 3 or 4) (see Table 42). Grade 5 (fatal) AEs in patients with R/R DLBCL treated with BR were higher than Pola BR and are described separately in Section 4.6.2.5.
Summaries of the incidence of specifically Grade 3 and 4 AEs by SOC and preferred term are provided in Section 9.1.3.1, GO29365 Interim CSR.
145 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 42 Most Frequently Reported Grade 35 Adverse Events (5%*) in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) Study Phase Phase I Phase II Phase Ib/II (Randomized) Cohort/Arm 1a D C 1a + C Treatment Pola + BR BR Pola + BR Pola + BR (total) Sample size N 6 N 39 N 39 N 45 Total number of patients with at least one AE: Blood & Lymphatic System Disorders Neutropenia 0 13 (33.3%) 18 (46.2%) 18 (40.0%) Thrombocytopenia 1 (16.7%) 9 (23.1%) 16 (41.0%) 17 (40.0%) Anemia 0 7 (17.9%) 11 (28.2%) 11 (24.2%) Febrile neutropenia 1 (16.7%) 5 (12.8%) 4 (10.3%) 5 (11.1%) Lymphopenia 0 0 5 (12.8%) 5 (11.1%) Leukopenia 0 3 (7.7%) 3 (7.7%) 3 (6.7%) Infections & Infestations Pneumonia 1 (16.7%) 1 (2.6%)† 3 (7.7%)† 4 (8.9%) Herpes zoster 0 1 (2.6%) 0 0 Metabolism & Nutritional Disorders Hypokalemia 0 1 (2.6%) 3 (7.7%) 3 (6.7%) Hypophosphatemia 0 1 (2.6%) 1 (2.6%) 1 (2.2%) Skin & Subcutaneous Tissue Disorders Rash 0 3 (7.7%) 0 0 Gastrointestinal Disorders Diarrhea 1 (16.7%) 1 (2.6%) 1 (2.6%) 2 (4.4%) General Disorders & Administration Site Conditions Fatigue 1 (16.7%) 1 (2.6%) 1 (2.6%) 2 (4.4%) Cardiac Disorders Atrial fibrillation 0 1 (2.6%) 0 0 Respiratory, Thoracic & Mediastinal Disorders Hypoxia 0 1 (2.6%) 1 (2.6%) 1 (2.2%) * AE preferred terms reported in 5% of all patients with R/R DLBCL treated with Pola BR (at least 3 patients out of a total of 45 in Phase Ib/II combined). † All AEs shown in this table of most frequently reported Grade 35 AEs (with onset from first dose of study drug through 90 days after last dose of study drug) were Grade 3 or 4 except for 2 fatal (Grade 5) events of pneumonia (one event each in Pola BR arm and BR arm of randomized Phase II (see Table 43). Source: Section 9.1.4.1, GO29365 Interim CSR.
146 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.5 Deaths in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) A total of 54.3% (25/46) of patients with R/R DLBCL in the Pola BR treatment arms had died at the time of the clinical cut-off. The main cause of death was disease progression (in 16/25 deaths [64.0%]). Adverse events accounted for the nine other deaths in patients who were administered Pola BR (9/25 deaths [36.0%]); six deaths occurred during follow-up after study treatment completion or discontinuation, of which, four deaths occurred following disease progression (see below and footnote to Table 43.
In the randomized Phase II, fewer deaths overall had occurred in the Pola BR arm (23 deaths), compared to the BR arm (28 deaths); the difference was due to fewer deaths due to disease progression in the Pola BR arm (14 patients vs. 17 patients). The number of deaths due to adverse events was nine patients in the Pola BR and 11 patients in the BR arm.
Table 43 lists the deaths due to fatal AEs in patients with R/R DLBCL. Pneumonia and sepsis were the cause of death in two patients in the Pola BR and BR arm, respectively. No other fatal AE by preferred term affected more than one patient in either treatment arm.
By SOC infections and infestations accounted for the most fatal AEs (five patients in the BR arm and four patients in the Pola BR arm).
Three deaths due to AEs in the Pola BR arm (pneumonia, hemoptysis, and pulmonary edema), four deaths due to AEs in the BR arm (septic shock and pneumonia) occurred during the study treatment period, after first dose of study treatment and before treatment discontinuation or completion.
For 8/11 fatal AEs in the BR arm and 8/9 fatal AEs in the Pola BR arm, the date of death was 30 days after the last dose of study treatment.
147 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 43 Deaths Due to Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Day of Last Study Day of AE Day of AE MedDRA Preferred Term Study Drug Comments Onset Death Administration BR (Phase II Randomized [N 39]) 1 179‡ Death 109 179 not related to any study drug 2 158‡†¶ Cerebral hemorrhage 86 158 not related to any study drug 3 114* Septic shock 94 117 related to bendamustine and rituximab not related to any study drug 4 9* Pneumonia 3 9 bendamustine and rituximab withdrawn 5 133* Cerebrovascular accident 123 134 not related to any study drug Multiple organ dysfunction 6 70‡† 3 73 not related to any study drug syndrome not related to any study drug 7 52*† Sepsis 22 57 bendamustine and rituximab withdrawn. Multiple organ dysfunction 8 151‡† 114 152 not related to any study drug syndrome 9 451‡ Cardiac failure 130 451 not related to any study drug 10 542‡†¶ Leukoencephalopathy 122 555 not related to any study drug 11 75‡† Sepsis 44 112 not related to any study drug Pola BR (Phase II Randomized [N 39]) related to polatuzumab vedotin, 1 286‡ Meningoencephalitis herpetic 120 317 bendamustine and rituximab related to rituximab 2 71* Pneumonia 44 75 all study drugs withdrawn 3 78‡†¶ Distributive shock 23 78 not related to any study drug
148 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 43 Deaths Due to Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) (cont.)
Day of Last Study Day of AE Day of AE MedDRA Preferred Term Study Drug Comments Onset Death Administration not related to any study drug 4 42* Hemoptysis 3 42 all study drugs withdrawn 5 121‡† Pneumonia 58 133 not related to any study drug 6 422‡†¶ Hemorrhage intracranial 156 422 not related to any study drug 7 337‡†¶ Renal failure 114 358 not related to any study drug related to polatuzumab vedotin, 8 11* Pulmonary edema 3 31 bendamustine and rituximab all study drugs withdrawn 9 379‡¶ Sepsis 59 384 not related to any study drug AE: adverse event; BR: bendamustine plus rituximab; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. * Event occurred during the study treatment period. ‡ Event occurred during the follow-up period, after study treatment discontinuation or completion and greater than 30 days after the last dose of study drug administration. † Event occurred after progressive disease. ¶ Event occurred after NALT. Source: Section 9.1.5, GO29365 Interim CSR.
149 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.6 Serious Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) Serious AEs with fatal outcome are described in the preceding section (see Section 4.6.2.5).
A total of 57 SAEs were reported in 64.4% (29/45) of patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined).
The most common SAEs by preferred term were febrile neutropenia (five patients [11.1%]), pneumonia, and pyrexia (four patients [8.9%] each). Other SAEs reported in more than one patient by preferred term in either treatment arm are shown in Table 44.
Serious AEs that most frequently affected patients (10%) were in the SOCs of infections and infestations (28.9%), blood and lymphatic disorders (15.6%), and gastrointestinal disorders (13.3%).
In the randomized Phase II, the overall incidence of SAEs was similar (Pola BR: 64.1% [25/39 patients] vs. BR: 61.5% [24/39 patients]) between the arms. Serious AEs by preferred term or by SOC were generally balanced with no more than three patients in either arm accounting for any difference observed. The severity of SAEs by NCI CTCAE grading was also balanced between the Pola BR and BR treatment arms. Nine patients in the Pola BR and 11 patients in the BR arm had SAEs with fatal outcome (Grade 5) (see Table 43 and Appendix 12).
150 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 44 Serious Adverse Events Reported in More than One R/R DLBCL Patient Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase I Phase II Phase Ib/II (Randomized) Cohort/Arm 1a D C 1a C Treatment Pola BR BR Pola BR Pola BR (total) Sample size N 6 N 39 N 39 N 45 MedDRA SOC preferred term Total number of patients with at least one AE Infections & Infestations Pneumonia 1 (16.7%) 3 (7.7%) 3 (7.7%) 4 (8.9%) Sepsis 0 2 (5.1%) 2 (5.1%) 2 (4.4%) Blood & Lymphatic System Disorders Febrile Neutropenia 1 (16.7%) 4 (10.3%) 4 (10.3%) 5 (11.1%) Anemia 0 1 (2.6%) 2 (5.1%) 2 (4.4%) Thrombocytopenia 0 1 (2.6%) 2 (5.1%) 2 (4.4%) Neutropenia 0 2 (5.1%) 1 (2.6%) 1 (2.2%) General Disorders & Administrative Site Conditions Pyrexia 0 0 4 (10.3%) 4 (8.9%) BR: bendamustine plus rituximab; Pola: polatuzumab vedotin; SOC: system organ class. Source: Section 9.1.6.1, GO29365 Interim CSR.
4.6.2.7 Adverse Events Leading to Withdrawal of Study Treatment in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) Table 45 summarizes the incidence of AEs leading to study drug discontinuation, dose reduction, or drug interruption (delays or withholding of dose) by individual study drug, and by treatment regimen arm in patients with R/R DLBCL.
Adverse events led to discontinuation of any study drug in 31.3% (14/45) of patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined) (see Table 45).
In most cases (12/14) all drugs (polatuzumab vedotin, bendamustine, and rituximab) were discontinued permanently at the same time (in two patients, only bendamustine was discontinued).
151 PTJA06 – Submission Dossier by Roche Registration GmbH
The most common AEs leading to the discontinuation of any treatment were cytopenias (8/14 discontinuations; predominantly Grade 3), mainly thrombocytopenia, neutropenia, and pancytopenia.
Three patients were withdrawn from all study treatments due to fatal (Grade 5) AEs of hemoptysis, pulmonary edema, and pneumonia (see Table 43).
In the randomized Phase II, a higher proportion of patients withdrew any study treatment due to AEs in the Pola BR arm (13/39 patients [33.3%]) compared with the BR arm (6/39 patients [15.4%]).
Adverse events that led to the discontinuation of polatuzumab vedotin in 26.7% (12/45) of patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined) (see Appendix 13); the most frequent AE by preferred term was thrombocytopenia (four patients), followed by neutropenia (two patients). All other AEs leading to polatuzumab vedotin withdrawal were unique events affecting single patients.
None of the patients with R/R DLBCL who received Pola BR withdrew due to the preferred term peripheral neuropathy (see Section 4.6.2.9).
Summaries of AEs that led to the discontinuation of bendamustine and of rituximab are provided in Section 9.1.7, GO29365 Interim CSR.
4.6.2.8 Adverse Events that Led to Treatment Modification in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) Drug interruption (dose delays/withholding of dose) was the most common action taken in response to AEs. Approximately one half of patients with R/R DLBCL had any drug interrupted (23/45 patients [51.1%]) due to an AE (see Table 45). The most frequent events leading to drug interruption were neutropenia and thrombocytopenia.
The incidence of dose reductions of study treatment due to AEs is summarized by treatment arm and by individual drug in Table 45. Reduction of polatuzumab vedotin dose occurred in two patients with R/R DLBCL receiving Pola BR in the randomized Phase II, both due to Grade 2 peripheral neuropathy. Reduction in the bendamustine dose in patients with R/R DLBCL (six patients receiving Pola BR and four patients in BR), was most frequently due to haematological toxicities (mainly Grade 3 or 4 neutropenia or thrombocytopenia).
152 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 45 Incidence of Adverse Events Leading to Discontinuation, Drug Interruption or Dose Modification of Study Drugs in Patients with R/R DLBCL (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Ph Ib Ph II Total (Randomized) (Ph Ib/II) Cohort/Arm 1a D C 1a C Treatment Pola BR BR Pola BR Pola BR Sample size N 6 N 39 N 39 N 45 Any study drug Patients with AEs leading to any study drug Discontinuation 1 (16.7%) 6 (15.4%) 13 (33.3%) 14 (31.1%) Dose reduction 1 (16.7%) 4 (10.3%) 7 (17.9%) 8 (17.8%) Drug interruption 2 (33.3%) 15 (38.5%) 21 (53.8%) 23 (51.1%) (delay/withholding dose) Polatuzumab vedotin Patients with AEs leading to polatuzumab vedotin Discontinuation 0 NA 12 (30.8%) 12 (26.7%) Dose reduction 0 NA 2 (5.1%) 2 (4.4%) Drug interruption 2 (33.3%) NA 20 (51.3%) 22 (48.9%) (delay/withholding dose) Bendamustine Patients with AEs leading to bendamustine Discontinuation 1 (16.7%) 4 (10.3%) 13 (33.3%) 14 (31.1%) Dose reduction 1 (16.7%) 4 (10.3%) 5 (12.8%) 6 (13.3%) Drug interruption 1 (16.7%) 15 (38.5%) 18 (46.2%) 19 (42.2%) (delay/withholding dose) Rituximab Patients with AEs leading to R Discontinuation 0 6 (15.4%) 12 (30.8%) 12 (26.7%) Drug interruption 2 (33.3%) 16 (41.0%) 20 (51.3%) 22 (48.9%) (delay/withholding dose) AE: adverse event; BR: bendamustine plus rituximab; NA: not applicable; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Source: Section 9.1.8, GO29365 Interim CSR.
153 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.2.9 Selected Adverse Events/Adverse Events to Monitor in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) This section describes the results of the analyses of selected AEs, defined as identified risks and potential risks of polatuzumab vedotin based on data available at the time of the assessment, and includes AEs which require further characterization or evaluation through dedicated analysis, for which additional data may be collected (events to monitor)
An overview of the overall incidence of selected AEs by category with R/R DLBCL is shown in Table 46. Summaries of selected AEs by preferred term under each selected AE category by treatment arm (Pola BR and BR) is provided in Appendix 14. Selected AEs with onset within 30 days of last dose of study drug is also provided in Section 9.1.9, GO29365 Interim CSR.
Most patients with R/R DLBCL treated with Pola BR had at least one selected AE (95.6% [43/45 patients]). A higher incidence of selected AEs were reported in the Pola BR arm compared to the BR arm of the randomized Phase II (97.4% vs. 87.2%); the difference driven mainly by a 10% incidence in the Pola BR arm compared to the BR arm of events in the following selected AE categories: neutropenia (by SMQ: 64.1% vs. 43.6%), peripheral neuropathy (by SMQ: 43.6% vs. 7.7%), anemia (by preferred term: 53.8% vs. 25.6%), thrombocytopenia (by SMQ: 51.3% vs. 33.3%), infusion-related reactions (related AEs during or within 24 hours of infusion: 33.3% vs. 23.1%), gastrointestinal toxicity (SOC: 82.1% vs. 64.1%), and joint pain, arthralgias and skeletal pain (by verbatim and preferred terms: 15.4% vs. 2.6%).
There were no events under the selected AE categories of reproductive toxicity (by SOC pregnancy, puerperium and perinatal conditions), drug-drug interactions (by preferred term), or TLS (preferred term, a protocol-specified AESI) (see Table 46).
154 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 46 Selected Adverse Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase Ib Phase II Phase Ib/II (Randomized) Cohort/Arm 1a D C 1a + C Treatment Pola + BR BR Pola + BR Pola + BR (total) Sample size N 6 N 39 N 39 N 45 Patients with at least one adverse event: 5 (83.3%) 34 (87.2%) 38 (97.4%) 43 (95.6%) Selected AEs/AEs to monitor category Neutropenia 2 (33.3%) 17 (43.6%) 25 (64.1%) 27 (60.0%) Peripheral neuropathy 1 (16.7%) 3 (7.7%) 17 (43.6%) 18 (40.0%) Anemia 0 10 (25.6%) 21 (53.8%) 21 (46.7%) Thrombocytopenia 2 (33.3%) 13 (33.3%) 20 (51.3%) 22 (48.9%) Infections & Infestations 3 (50.0%) 20 (51.3%) 21 (53.8%) 24 (53.3%) IRRs (related AEs during/within 24 hrs of 2 (33.3%) 9 (23.1%) 13 (33.3%) 15 (33.3%) end of infusion) Hepatic toxicity 2 (33.3%) 5 (12.8%) 7 (17.9) 9 (20.0%) Fatigue and asthenia 4 (66.7%) 19 (48.7%) 18 (46.2%) 22 (48.9%) Hyperglycemia 0 1 (2.6%) 1 (2.6%) 1 (2.2%) Renal toxicity 1 (16.7%) 5 (12.8%) 5 (12.8%) 6 (13.3%) Gastrointestinal toxicity 4 (66.7%) 25 (64.1%) 32 (82.1%) 36 (80.0%) Pulmonary toxicity 0 1 (2.6%) 2 (5.1%) 2 (4.4%) Joint pain, arthralgias, and skeletal pain 0 1 (2.6) 6 (15.4%) 6 (13.3%) Alopecia 0 1 (2.6%) 0 0 Cardiac toxicity and arrhythmias 1 (16.7%) 5 (12.8%) 0 1 (2.2%)
Ocular Toxicity 0 1 (2.6%) 0 0 Dysguesia 2 (33.3%) 0 1 (2.6%) 3 (6.7%) Malignancies 1 (16.7%) 2 (5.1%) 1 (2.6%) 2 (4.4%) Myelodysplastic syndrome 0 1 (2.6%) 1 (2.6%) 1 (2.2%) Definitions of grouped terms and MedDRA codes for selected AESIs/AEs to monitor are provided in Section 3.9.7, GO29365 Interim CSR. Note: No selected AEs of reproductive toxicity, drug-drug interaction or tumor lysis syndrome were reported. Source: Section 9.1.9, GO29365 Interim CSR.
155 PTJA06 – Submission Dossier by Roche Registration GmbH
Peripheral Neuropathy Peripheral neuropathy (sensory and/or motor), both serious and non-serious is an identified risk for polatuzumab vedotin.
The overall incidence of peripheral neuropathy in patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined) was 40.0% (18/45 patients with 29 peripheral neuropathy events) (see Table 47). The most frequently reported peripheral neuropathy event by preferred term in patients treated with Pola BR was neuropathy peripheral (nine patients) followed by peripheral sensory neuropathy (six patients).
Six of the 18 patients with R/R DLBCL with peripheral neuropathy events had Grade 2 events (preferred terms: neuropathy peripheral [five patients], peripheral sensory neuropathy and muscular weakness (one patient each). No Grade 3 or serious events of peripheral neuropathy were reported in patients with R/R DLBCL treated with Pola BR.
Ten of the 18 patients with R/R DLBCL who had a peripheral neuropathy event during treatment with Pola BR had a history of prior peripheral neuropathy, and of these, eight patients had ongoing Grade 1 peripheral neuropathy at baseline.
One patient discontinued all study treatments due to loss of muscle mass, reported as the preferred term muscle atrophy (Grade 1), which was included under the conservative analysis of peripheral neuropathy using the wide SMQ for peripheral neuropathy.
The polatuzumab vedotin dose was reduced in two patients, in both cases due to Grade 2 peripheral neuropathy. In both cases, the event had resolved by the time of the clinical cut-off (see also Section 4.6.2.9).
Overall, the peripheral neuropathy events had resolved in 10/18 patients with peripheral neuropathy events without the need, in most cases, for delays or any change in dose of study treatment.
The median time to onset of peripheral neuropathy in patients with R/R DLBCL treated with Pola BR was 1.8 months (0.04.9 months), and the median time to resolution was 1.2 months (0.022.4 months).
In the randomized Phase II, the incidence of all grade peripheral neuropathy events was higher in the Pola BR arm compared to the BR arm (43.6% [17/39 patients with 26 events] vs. 7.7% [3/39 patients with 4 events]). The peripheral neuropathy events reported in the BR arm were Grade 2 neuralgia, peripheral neuropathy, muscular weakness, and decreased vibratory sense.
156 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 47 Summary of Peripheral Neuropathy in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase Ib Phase II Phase Ib/II (Randomized) Cohort/Arm 1a D C 1a + C Treatment Pola + BR BR Pola + BR Pola + BR (total) Sample size N 6 N 39 N 39 N 45 Total no. of PN events 3 4 26 29 Patients with at least one: Any grade PN event 1 (16.7%) 3 (7.7%) 17 (43.6%) 18 (40.0%) Grade 2 PN event* 0 2 (5.1%) 6 (15.4%) 6 (15.4%) Serious PN event 0 0 0 0 PN event leading to any study drug 0 0 1 (2.6%) 1 (2.2%) discontinuation PN event leading to any study drug 0 0 2 (5.1%)a 2 (4.4%)a reduction PN event leading to any study drug 0 0 2 (5.1%)a 2 (4.4%)a interruption BR: bendamustine plus rituximab; PN: peripheral neuropathy; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Analyses of peripheral neuropathy events were based on all preferred terms captured under the standardized MedDRA query (SMQ) “Peripheral Neuropathy [wide]” (20000034w). * All peripheral neuropathy events reported are Grade 2 only. a Polatuzumab vedotin dose was reduced while bendamustine and rituximab was delayed in two patients. Source: Section 9.1.9.1, GO29365 Interim CSR.
Neutropenia Including Febrile Neutropenia A total of 100 neutropenia events (by SMQ) were reported in 60.0% (27/45) of patients with R/R DLBCL treated with Pola BR. Over one-half of patients had Grade 3 or 4 events (55.6% [25/45 patients]). None were fatal (Grade 5) (see Table 48).
After neutropenia (21 patients), the most frequently reported preferred term under the SMQ used for the analysis were lymphopenia (six patients), febrile neutropenia, and leukopenia (five patients each) (see Appendix 14).
In the randomized Phase II, a higher incidence of neutropenia and related events were reported in the Pola BR arm compared to the BR arm (all Grade: 64.1% vs. 53.8%, respectively; Grade 3 or 4: 59.0% vs. 46.2%). Serious neutropenia events affected more patients in the BR arm than in the Pola BR arm (six patients [15.4%] vs. five patients [12.8%], respectively) (see Table 48).
157 PTJA06 – Submission Dossier by Roche Registration GmbH
Neutropenia events led to discontinuation of any study drug in five patients in the Pola BR arm (in three cases, all three study treatments were withdrawn at the same time) compared to one patient in the BR arm (see Table 48). More patients in the Pola BR compared to the BR arm had delays in study drug administration due to neutropenia (13 patients [33.3%] vs. seven patients [17.9%]). Two patients in each treatment arm had their bendamustine dose reduced due to neutropenia events.
Most neutropenia events were reversible. In 15/27 patients with R/R DLBCL treated with Pola BR with reported neutropenia events, all events had resolved by the time of the clinical cut-off.
The proportion of patients with R/R DLBCL receiving G-CSF support was comparable between the two arms, ranging from 36% to 57% across the treatment cycles for Pola BR and 33% to 55% for BR.
The median time to onset for a Grade 3 neutropenia event was 1.2 months in the Pola BR arm compared to 1.4 months in the BR arm.
158 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 48 Summary of Neutropenia Events in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase Ib Phase II Phase Ib/II (randomized) Cohort/Arm 1a D C 1a + C Treatment Pola + BR BR Pola + BR Pola + BR (total) Sample size N 6 N 39 N 39 N 45 Total no. of neutropenia events 4 56 96 100 Patients with at least one: Any grade neutropenia event 2 (33.3%) 21 (53.8%) 25 (64.1%) 27 (60.0%) Grade 34 neutropenia event 2 (33.3%) 18 (46.2%) 23 (59.0%) 25 (55.6%) Grade 5 neutropenia event 0 0 0 0 Serious neutropenia event 1 (16.7%) 6 (15.4%) 5 (12.8%) 6 (13.3%) Neutropenia event leading to any study 0 1 (2.6%) 5 (12.8%) 5 (11.1%) drug discontinuation Neutropenia event leading to any study 0 2 (5.1%)a 2 (5.1%)a 2 (4.4%) drug reduction Neutropenia event leading to any study 0 6 (15.4%) 13 (33.3%) 13 (28.9%) drug interruption BR: bendamustine plus rituximab; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Analyses of neutropenia events were based on all preferred terms captured under the standardized MedDRA query (SMQ) “Haematopoietic leukopenia [narrow]” (20000030n). a Only bendamustine dose was reduced. Source: Section 9.1.9.2, GO29365 Interim CSR.
Infections and Infestations The overall incidence of infection in patients with R/R DLBCL treated with Pola BR (Phase Ib/II combined) was 53.3% (24/45 patients). Most were Grade 1 or 2.
In the randomized Phase II, a similar incidence of infections in patients with R/R DLBCL was reported in the Pola BR and BR arms, including all grade (53.8% vs. 51.3%, respectively), Grade 3 events (23.1% vs. 20.5%), fatal events (four patients [10.3%] in each arm), and serious events (28.2% vs. 30.8%).
The four fatal infections in the Pola BR arm were pneumonia (two patients), sepsis, and herpetic meningoencephalitis; while in the BR arm, they were pneumonia, sepsis (two patients), and septic shock (see also Table 43 and Section 4.6.2.5 for more information).
159 PTJA06 – Submission Dossier by Roche Registration GmbH
Regarding the nature of infections, reported preferred terms included bacterial, viral and fungal types (see Appendix 14). In the Pola BR arm, six opportunistic infection events were reported in four patients: herpetic encephalitis, cerebral toxoplasmosis and cytomegalovirus (CMV) pneumonia (in the same patient), CMV infection (reported twice in the same patient), and Pneumocystis jiroveci pneumonia (PJP). In the BR arm, opportunistic infection events were reported in 2 patients: CMV infection and pulmonary mycosis.
The median time to onset for an infection event was 1.4 months (range: 0.39.1 months) in the Pola BR arm and 0.9 months (range: 0.17.2 months) in the BR arm.
Table 49 Summary of Infections in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population)
Study Phase Phase Ib Phase II Phase Ib/II (Randomized) Cohort/Arm 1a D C 1a + C Treatment Pola + BR BR Pola + BR Pola + BR (total) Sample size N 6 N 39 N 39 N 45 Total no. of infections 8 28 42 50 Patients with at least one: Any grade infection 3 (50.0%) 20 (51.3%) 21 (53.8%) 24 (53.3%) Grade 34 infection 2 (33.3%) 8 (20.5%) 9 (23.1%) 11 (24.4%) Grade 5 infection 0 4 (10.3%) 4 (10.3%) 4 (8.9%) Serious infection 2 (33.3%) 12 (30.8%) 11 (28.2%) 13 (28.9%) Infection leading to any study drug 1 (16.7%) 2 (5.1%) 1 (2.6%) 2 (4.4%) discontinuation Infection leading to any study drug 0 1 (2.6%) 1 (2.6%) 1 (2.2%) reduction Infection leading to study drug 0 5 (12.8%) 4 (10.3%) 4 (8.9%) interruption Opportunistic infection 0 2 (5.1%) 4 (10.3%) 4 (8.9%) BR: bendamustine plus rituximab; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Analyses of infection events were based on all preferred terms captured under System Organ Class (SOC) “Infections and Infestations”. Source: Section 9.1.9.3, GO29365 Interim CSR.
Hepatic Toxicity Hepatic toxicity events (15 in total) were reported in 20.0% (9/45) of patients with R/R DLBCL treated with Pola BR. Most of the events were Grade 1 or 2 except for two events of Grade 34 (one event of Grade 3 hypoalbuminemia, which had resolved with
160 PTJA06 – Submission Dossier by Roche Registration GmbH
concomitant treatment with albumin at the time of clinical cut-off and another event was Grade 4 transaminase increased, which had recovered and resolved without any concomitant treatment)
The most common hepatic toxicity event reported in patients with R/R DLBCL treated with Pola BR by preferred term was hypoalbuminemia (six patients), followed by increased aspartate aminotransferase (two patients); all other events were unique events affecting single patients. There were no cases of drug-induced liver injury that included elevated ALT or AST in combination with either elevated bilirubin or clinical jaundice as defined by Hy’s Law.
In the randomized Phase II, hepatic toxicity events were reported in seven patients (17.9%) with R/R DLBCL in the Pola BR arm, and five patients (12.8%) in the BR arm (see Appendix 14).
No hepatic toxicity events in any Pola BR or BR treatment arm led to study drug discontinuation, reduction, or dose delays.
Anemia The overall frequency of anemia events reported as the preferred term “anaemia” in patients with R/R DLBCL treated with Pola BR was 46.7% (21/45 patients). Eleven patients had Grade 3 or 4 anemia events and 2 patients had events reported as serious (see Appendix 14).
In the randomized Phase II, anemia was reported more frequently in the Pola BR arm compared with the BR arms (all Grade: 53.8% vs. 25.6%, respectively; Grade 3 or 4: 28.2% vs. 17.9%). None of the events were fatal (Grade 5) AEs.
Treatment-emergent laboratory decrease in hemoglobin (shifts from baseline to worse NCI CTCAE grade for low hemoglobin) were also more frequently detected in the Pola BR arm compared with the BR arm: any Grade worsening, 84.6% (33/39 patients) versus 69.2% (27/38 patients); worsening to Grade 3 or 4: 17.9% (7/39 patients) versus 10.3% (4/39 patients). Full results are provided in Section 9.1.9.5, GO29365 Interim CSR.
Despite the difference in reported incidence of anemia in the Pola BR and BR arms, the proportion of patients who received RBC transfusions in each arm was similar (Pola BR: 23.1% [9/39 patients] vs. BR: 17.9% [7/39 patients]).
None of the events of anemia in patients with R/R DLBCL led to discontinuation or reduction in study treatment dose. At the time of the clinical cut-off, the event had resolved in 6/17 patients with reported anemia in the Pola BR arm.
161 PTJA06 – Submission Dossier by Roche Registration GmbH
Thrombocytopenia Thrombocytopenia events were reported in approximately one half of all patients with R/R DLBCL treated with Pola BR (48.9% [22/45 patients]). Most were Grade 3 or 4 events (affecting 40.0% [18/45 patients]) and were reversible. None of the thrombocytopenia events were fatal. Two patients experienced serious events (both Grade 4) (see Appendix 14).
Thrombocytopenia events (all Grade and Grade 3 or 4) were reported more frequently in patients receiving Pola BR than BR in the randomized Phase II (all Grade: 51.3% vs. 33.3%, respectively; Grade 3 or 4: 43.6% vs. 25.6%, respectively). The number of patients with thrombocytopenia events leading to any study drug discontinuation (5 vs. 2), or study drug reduction (4 vs. 1; in all cases only bendamustine was reduced), or requiring dosing delays (6 vs. 1) was numerically higher in the Pola BR arm compared to the BR arm. Full results are provided in Section 9.1.9.6, GO29365 Interim CSR.
Similar rates of platelet transfusions were reported in both arms (Pola BR: 10.3% vs. BR: 12.8%).
One patient with Grade 2 thrombocytopenia had a concurrent bleeding event (Grade 5 hemoptysis) in the setting of anticoagulation therapy due to thromboembolic event (pulmonary embolism and deep vein thrombosis). In addition, the patient had an extranodal lesion in lung at baseline. For further information, see Section 9.1.9.6, GO29365 Interim CSR.
Treatment-emergent laboratory decreases from baseline in platelets (shifts from baseline to worse NCI CTCAE grade for low platelets) were more frequently detected in the Pola BR arm compared to the BR arm: any Grade worsening: 71.8% (28/39 patients) versus 60.5% (23/38 patients); worsening to Grade 3 or 4: 28.2% (11/39 patients) versus 15.8% (6/38 patients). Further information is provided in Section 9.5.1, GO29365 Interim CSR.
Diarrhea Diarrhea is a potential risk (as part of gastrointestinal toxicities) for polatuzumab vedotin.
Diarrhea events were reported in 40.0% (18/45) of patients with R/R DLBCL treated with Pola BR. All were Grade 1 or 2 in severity, with the exception of one Grade 3 (and serious) event in one patient which led to their study treatments being delayed. This and most other diarrhea events were reversible and none led to any drug discontinuation or reduction.
Diarrhea was reported more frequently in the Pola BR arm compared with the BR arm in the randomized Phase II (40.0% vs. 20.5%).
162 PTJA06 – Submission Dossier by Roche Registration GmbH
4.6.3 Updated Safety Results in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) The updated safety data presented for the GO29365 is from a CCOD of 11 October 2018. The CCOD represents approximately 5 months of additional safety data from the previous CCOD of 30 April 2018.
The duration of exposure to study treatment was higher in the Pola BR arm (median 5 cycles) compared to BR arm (median 3 cycles). An overview of the overall incidence of AEs by category as of the data cut-off date of 11 October 2018 in patients with R/R DLBCL treated with Pola BR is presented in Table 50. The key safety findings in patients with R/R DLBCL treated with Pola BR arm summarized below:
The safety profile of polatuzumab vedotin when combined with BR in patients with R/R DLBCL, in the context of the clinically significant benefit observed and longer treatment duration, was acceptable and overall was consistent with that observed in the BR group.
The incidence of AEs was comparable between the Pola BR and BR treatment arms (100.0% vs. 97.4%). The most frequently reported AEs (occurring in 20% of patients) in the Pola BR arm were neutropenia, thrombocytopenia, anemia, fatigue, diarrhea, nausea, pyrexia, decreased appetite, and peripheral neuropathy. In comparison to the BR arm, the events of neutropenia, thrombocytopenia, anemia, diarrhea, pyrexia, lymphopenia, and peripheral neuropathy events were reported at a higher incidence (10% difference) in the Pola BR arm.
The incidence of Grade 3 or 4 AEs was higher in the Pola BR arm compared with the BR arm (86.7% vs. 71.8%); this difference was driven mainly by a higher incidence of Grade 3 or 4 neutropenia, thrombocytopenia, lymphopenia, and anemia in the Pola BR arm.
The overall incidence of deaths was lower in the Pola BR arm compared with the BR arm (57.7% vs. 71.8%). Most deaths in both treatment arms were due to disease progression. Nine (20.0%) deaths in the Pola BR arm and 10 (25.6%) deaths in the BR group were due to AEs.
The incidence of SAEs was comparable between the Pola BR and BR arms (66.7% vs. 61.5%), with no notable differences between the two groups in the most frequently occurring SAEs, which were infections (by System Organ Class [SOC]) and neutropenia events including febrile neutropenia (by SMQ).
163 PTJA06 – Submission Dossier by Roche Registration GmbH
In the Pola BR arm, the incidence of AEs that led to discontinuation of polatuzumab vedotin was 26.7%, and the most frequently reported events were thrombocytopenia and neutropenia.
In the Pola BR arm, the incidence of AEs that led to dose interruption of polatuzumab vedotin was 48.9% and the most frequently reported events were neutropenia, thrombocytopenia, and febrile neutropenia.
In the Pola BR arm, 2 (4.4%) patients had dose reduction of polatuzumab vedotin due to Grade 2 peripheral neuropathy. Permanent dose reduction of polatuzumab vedotin (from 1.8 mg/kg to 1.4 mg/kg) was mandatory for Grade 2 or 3 peripheral neuropathy events per protocol.
Neutropenia: The incidence of neutropenia events was comparable between the Pola BR and BR arms (57.8% vs. 53.8%). The majority of the events were Grade 3 or 4 in severity. The incidence of serious neutropenia events was low in both the Pola BR and BR arms (11.1% vs. 15.4%). The incidence of febrile neutropenia was also comparable between the Pola BR and BR arms (8.9% vs. 12.8%). The proportion of patients receiving G-CSF support was comparable between the Pola BR and BR arms (64.4% vs. 61.5%). The laboratory decreases in total neutrophil (shifts from baseline to worse NCI CTCAE Grade 3 or 4 post-baseline for neutrophil hypo) were reported in 51.1% and 33.3% of patients in the Pola BR arm and BR arm, respectively. In the Pola BR arm, neutropenia events led to discontinuation of polatuzumab vedotin in 8.9% of patients, and dose delay or interruption of polatuzumab vedotin in 28.9% of patients. No neutropenia events led to dose reduction of polatuzumab vedotin.
Peripheral Neuropathy: The incidence of peripheral neuropathy events was higher in the Pola BR arm compared with the BR arm (40.0% vs. 7.7%), and all the events were Grade 1 or 2 in severity for both treatment arms. In the Pola BR arm 8/18 (44.4%) patients who had peripheral neuropathy events during treatment had ongoing peripheral neuropathy events at baseline. In the Pola BR arm, 1 (2.2%) patient discontinued all study treatments due to Grade 1 muscular atrophy. Two (4.4%) patients had polatuzumab vedotin dose reduced due to Grade 2 peripheral neuropathy. No peripheral neuropathy events led to only dose delay or interruption of polatuzumab vedotin. At the CCOD, most patients (61.1%; 11/18) who experienced peripheral neuropathy events in the Pola BR arm had peripheral neuropathy events resolved; however, no patients in the BR arm had peripheral neuropathy events resolved.
Infections: The incidence of infections events was comparable between the Pola BR and BR arms (53.3% vs. 51.3%). Most events were Grade 3 in both treatment arms, and 4 patients from each group (Pola BR: 8.9% vs. BR: 10.3%) experienced Grade 5 infections events. The incidence of serious
164 PTJA06 – Submission Dossier by Roche Registration GmbH
infections events was also comparable between the Pola BR and BR arms (31.1% vs. 30.8%). In the Pola BR arm, infections events led to discontinuation of polatuzumab vedotin in 1 (2.2%) patient, dose delay or interruption of polatuzumab vedotin in 4 (8.9%) patients. No infections events led to dose reduction of polatuzumab vedotin.
Hepatic Toxicity: The incidence of hepatic toxicity events was 20.0% in the Pola BR arm and 10.3% in the BR arm. The majority of events were Grade 1 or 2 in severity in both treatment arms. The most common event in the Pola BR arm was hypoalbuminemia. In the Pola BR arm, 2 (4.4%) patients experienced Grade 3 hypoalbuminemia and Grade 4 transaminase increased. This Grade 4 event of transaminase increased was serious, however, it occurred during follow-up in the setting of disease progression leading to initiation of new anti-lymphoma treatment (NALT). In the BR arm, 1 (2.6%) patient experienced Grade 3 blood alkaline phosphates increased, and was serious. No hepatic toxicity events led to discontinuation, dose delay, interruption, or dose reduction of polatuzumab vedotin in the Pola BR arm.
Anemia: The incidence of anemia events was higher in the Pola BR arm compared with the BR arm (46.7% vs. 28.2%). Most events were Grade 3 or 4 in severity in both treatment arms. Two (4.4%) patients in the Pola BR arm and 1 (2.6%) patient in the BR arm experienced serious anemia. The laboratory decrease in hemoglobin (shifts from baseline to worse NCI CTCAE Grade 3 or 4 post-baseline for hemoglobin hypo) was reported in 17.8% and 12.8% of patients for the Pola BR and BR arms, respectively. The proportion of patients receiving RBC transfusion was comparable between the Pola BR and BR arms (20.0% vs. 17.9%). In the Pola BR arm, 1 (2.2%) patient experienced Grade 3 anemia that led to dose interruption of polatuzumab vedotin. No anemia led to discontinuation or dose reduction of polatuzumab vedotin.
Thrombocytopenia: The incidence of thrombocytopenia events was higher in the Pola BR arm compared with the BR arm (48.9% vs. 33.3%). The majority of events were Grade 3 or 4 in severity in both treatment arms. Two (4.4%) patients in the Pola BR arm and 1 (2.6%) patient in the BR arm experienced serious thrombocytopenia. In the Pola BR arm, 1 (2.2%) patient with Grade 2 thrombocytopenia had a concurrent bleeding event. The proportion of patients who received platelet transfusion was comparable between the Pola BR and BR arms (8.9% vs. 12.8%). The laboratory decrease in platelets (shifts from baseline to worse NCI CTCAE Grade 3 or 4 post-baseline for platelet hypo) was reported in 33.3% and 25.6% of patients for the Pola BR and BR arms, respectively. In the Pola BR arm, thrombocytopenia events led to discontinuation of polatuzumab vedotin in 5 (11.1%) patients, and dose delay or interruption of polatuzumab vedotin in 6 (13.3%) patients. No thrombocytopenia events led to dose reduction of polatuzumab vedotin.
165 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 50 Overview of Safety in Patients with R/R DLBCL Treated with Pola BR or BR (Phase Ib/Phase II Randomized; Safety Population) (CCOD: 11 October 2018)
Study Phase Phase II Phase Ib/II Cohort/Arm D 1a C BR Pola BR Treatment (total) Sample size N 39 N 45 Patients with at least one: Any AE 38 (97.4%) 45 (100.0%) Grade 34 AE 28 (71.8%) 39 (86.7%) Grade 5 10 9 (25.6%) 9 (20.0%) Serious AE 24 (61.5%) 30 (66.7%) AE leading to discontinuation of: Pola NA 12 (26.7%) Any study drug 5 (12.8%) 14 (31.1%) AE leading to modification/interruption of: Pola NA 24 (53.3%) Any study drug 19 (48.7%) 31 (68.9%) AEs to monitor: Any Grade peripheral neuropathy (by 3 (7.7%) 18 (40.0%) SMQ) Any Grade neutropenia (by SMQ) 21 (53.8%) 26 (57.8%) Any Grade hepatotoxicity (by SMQ) 4 (10.3%) 9 (20.0%) Any Grade infections (by SOC) 20 (51.3%) 24 (53.3%) Any Grade thrombocytopenia (by SMQ) 13 (33.3%) 22 (48.9%) Any Grade anemia (by SMQ) 11 (28.2%) 21 (46.7%) AE: adverse event; BR: bendamustine plus rituximab; Pola BR: polatuzumab vedotin in combination with bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma. Investigator text for AEs encoded using MedDRA v21.0. Includes AEs with onset from first dose of study drug through 90 days after last dose of study drug Source: Section 2, 90-day Safety Update Report (see Appendix 15).
4.6.4 Safety in Special Groups Results of the safety analysis by baseline demographic factors (age, sex, and race) were generally consistent. Safety data for patients with organ (liver and kidney) impairment is limited and no overt safety concerns for these groups were identified. Further information is provided in Section 5, GO29365 Summary of Clinical Safety.
166 PTJA06 – Submission Dossier by Roche Registration GmbH
4.7 ADDITIONAL DATA 4.7.1 GO29365 Efficacy Results for Arm G (N 42) Study GO29365 was amended (protocol version 5) in order to include Arm G with the primary objective to provide clinical experience in terms of PK and safety utilizing the lyophilized formulation of polatuzumab vedotin in combination with BR. The results, taken together with the totality of available bioanalytical and non-clinical data have demonstrated material comparability between the formulation used in the Phase II randomized portion of the GO29365 study and the lyophilized material intended for commercialization.
Furthermore, as Arm G has recently completed enrollment and the last patient has reached the PRA the full analysis of efficacy and safety; see Supplementary Results Report for Study GO29365 – Arm G.
4.7.2 GO29365 Pooled Pola BR Efficacy Results (N 88) All Pola BR patients (n 88) from the Phase Ib (n 6), randomized Phase II (n 40), and Arm G (n 42) were pooled in order to increase the sample size. A clinical cut-off date of 30 March 2019 was used in order to include Arm G patients to ensure that all the available Pola BR data were utilized. In addition to CR comparisons, ORR and time-to-event comparisons are also provided as supportive data (see Appendix 16). It is acknowledged that there is a much shorter duration of follow-up in Arm G (8 months) compared to the randomized (30 months) and Phase Ib (46 months) cohorts.
4.8 SYSTEMATIC LITERATURE REVIEW AND INDIRECT TREATMENT COMPARISON This section should describe how relevant studies were identified:
Studies of the technology in the indication under assessment
Studies of the comparators (if applicable).
State the databases and trial registries searched and, when relevant, the platforms used to do this.
State the date the searches were done and any limits (for example date, language) placed on the searches.
Include as an appendix the search terms and strategies used to interrogate each database or registry.
For bibliographic databases: Include the complete search strategies (with the names of the interfaces), the years covered by the search, the date of the last search and the number of hits per line.
167 PTJA06 – Submission Dossier by Roche Registration GmbH
For study registries: Include the search terms, the input interface (for example, basic search or advanced search), and the number of hits retrieved.
If a search filter is used (that is, a predefined combination of search terms to filter references with a specific content), provide a reference to the filter used.
In a table, state the inclusion and exclusion criteria used to select studies and justify these.
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.
To support this submission for polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL who are not candidates for HSCT, a systematic literature review (SLR) of clinical study evidence on the efficacy, safety, and HRQoL of pharmacological interventions for the treatment of R/R DLBCL was performed. The methodology and results of the SLR and meta-analysis feasibility assessment is presented in the following sections.
4.8.1 Systematic Literature Review 4.8.1.1 Methodology The SLR was performed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Guidelines on 4 September 2018, covering Embase, MEDLINE®, and the Cochrane Library. These electronic databases were used to identify relevant publications using a pre-defined search string (see Appendix 17). The methodology used for electronic database searching, screening of identified studies and hand-searching, as well as the selection criteria applied are detailed in full in the Clinical SLR and Feasibility Report (see Appendix 18).
The eligibility criteria used to determine relevance for inclusion of publications in the SLR were in the form of the patient, intervention, comparison, outcome (PICO) model, and are detailed in Table 51.
168 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 51 Eligibility Criteria Used in the Initial Screening to Identify Studies Relating to Any Pharmacological Treatment for Patients with R/R DLBCL
Clinical Inclusion Criteria Exclusion Criteria Effectiveness Adult patients (18 years) with R/R Animal/in vitro studies Population DLBCL who are receiving second- or third-line (or beyond) treatment
Subgroups of interest include: SCT ineligible Failed prior autologous transplant DOR to prior therapy (12 months vs. 12 months) Age (60 vs. 60) Stage of disease (III vs. IIIIV) Prior systemic therapy Refractory vs. relapse Extranodal site involvement (01 vs. 24) ECOG PS Polatuzumab vedotin in Intervention combination with bendamustine and rituximab
169 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 51 Eligibility Criteria Used in the Initial Screening to Identify Studies Relating to Any Pharmacological Treatment for Patients with R/R DLBCL (cont.)
Clinical Inclusion Criteria Exclusion Criteria Effectiveness
Licensed or pharmaceutical First-line treatments Comparators treatments available for R/R DLBCL Non-pharmacological treatments patients: Bendamustine rituximab CEPP rituximab CEOP rituximab DA-EPOCH rituximab GDP rituximab Gemcitabine, carboplatin, dexamethasone, rituximab GemOx rituximab Gemcitabine vinorelbine rituximab Lenalidomide rituximab Rituximab Ibrutinib Pixantrone CAR-T (axicabtagene ciloleucel or tisagenlecleucel) Venetoclax Apatinib DHAP rituximab ICE rituximab MINE rituximab ESHAP rituximab IME rituximab IVE rituximab R-PECC Note: Any study which evaluates only one arm of interest was taken forward into the initial phase of the ITC feasibility assessment as it may serve as a bridging study to connect two comparators which are relevant to the decision problem.
170 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 51 Eligibility Criteria Used in the Initial Screening to Identify Studies Relating to Any Pharmacological Treatment for Patients with R/R DLBCL (cont.)
Clinical Inclusion Criteria Exclusion Criteria Effectiveness
Efficacy: Outcome(s) not listed Outcomes o OS o PFS o TTP o EFS o DOR o Response rates (CR, PR, SD) o Any response rates reported as PET-CR (i.e. metabolic CR) or using older criteria (e.g. CRu), or a mixture of various different criteria (Cheson et al. 2007); Lugano (Cheson et al. 1999); modified Lugano (Cheson et al. 2014). o ORR o DCR o DoT o DoT beyond progression Safety: o All grade treatment-related AEs o Treatment-related Grade 3 or 4 AEs o Treatment-related SAEs o Treatment-related Grade 5 AEs o Tolerability (dose reductions and interruptions, discontinuation [any reason], discontinuation [due to AEs]) HRQoL and PRO measures (e.g. EORTC QLQ-C30) RCTs, any duration (irrespective Reviews/editorials, case Study Design of blinding) reports/case series Prospective single-arm studies Retrospective single-arm studies Comparative observation studies
171 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 51 Eligibility Criteria Used in the Initial Screening to Identify Studies Relating to Any Pharmacological Treatment for Patients with R/R DLBCL (cont.)
Clinical Inclusion Criteria Exclusion Criteria Effectiveness
English language publications Non-English language Language publications without an English Restrictions abstract
No restriction Date of Publication
No restriction Countries
AE: adverse event; CAR-T: chimeric antigen receptor T-cell; CEOP: cyclophosphamide, etoposide, vincristine and prednisone; CEPP: cyclophosphamide, etoposide, procarbazine and prednisone; CR: complete response; CRu: complete response, unconfirmed; DA-EPOCH: dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin; DCR: disease control rate; DHAP: dexamethasone, cytarabine and cisplatin; DOR: duration of response; DoT: duration of treatment; ECOG PS: Eastern Cooperative Oncology Group Performance Status; EFS: event-free survival; EORTC QLQ-C30: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ESHAP: etoposide, methylprednisolone, cytarabine, cisplatin; GDP: gemcitabine, cisplatin, and dexamethasone; GemOx: gemcitabine and oxaliplatin; HRQoL: health-related quality of life; ICE: ifosfamide, carboplatin and etoposide; IME: ifosfamide, mitoxantrone, etoposide; ITC: indirect treatment comparison; IVE: ifosfamide, epirubicin and etoposide; MINE: mesna, ifosfamide, mitoxantrone, etoposide; ORR: objective response rate; OS: overall survival; PET-CR: positron emission tomography-complete response; PFS: progression-free survival; PR: partial response; PRO: patient-related outcome; R-PECC: rituximab, lomustine, etoposide, chlorambucil, prednisone; RCT: randomized clinical trial; R/R DLBCL: relapsed/refractory diffuse large B- cell lymphoma; SAE: serious adverse event; SCT: stem cell transplantation; SD: stable disease; SLR: systematic literature review; TTP: time-to-progression. Source: Section 6, Clinical SLR and Feasibility Report (see Appendix 18).
4.8.1.2 Results of the Systematic Literature Review The process of study selection is depicted in the PRISMA flow diagram (Figure 20).
In total, 36 unique studies from 90 publications (nine RCTs and 27 single-arm trials) were identified that met the inclusion criteria. Of these 36 studies, 19 unique studies from 53 publications were considered for extraction. The studies were selected for extraction if the study included transplant ineligible patients, as this patient population is comparable with the patients enrolled in the GO29365 trial. A list of trials which were included and extracted is presented in Appendix 19.
A list of trials which were included but not extracted, as they either enrolled transplant eligible patients or transplantation status was unknown, is presented in Appendix 20.
172 PTJA06 – Submission Dossier by Roche Registration GmbH
All the included studies were English language publications. There were two studies that were only published as conference abstracts (Schuster et al. 2018; Sehn et al. 2018). In terms of study design, six studies were published as RCTs (Aribi et al. 2010; Aviles et al. 2010; Dang et al. 2018; Czuczman et al. 2017; Pettengell et al. 2012; Sehn et al. 2018a) and thirteen studies were prospective observational/single-arm trials (El Gnaoui et al. 2007; Lakshmaiah et al. 2015; Lopez et al. 2008; Mounier et al. 2013; Neelapu et al. 2017a; Ohmachi et al. 2013; Papageorgiou et al. 2005; Schuster et al. 2018; Schuster et al. 2017; Vacirca et al. 2014; Wiernik et al. 2008; Zinzani et al. 2011; Witzig et al. 2011). The 19 trials included in the review provided data on 11 different interventions.
All the included studies recruited patients with R/R DLBCL. Detailed study eligibility criteria are provided in the (Roche Data on File 2019). Of the 19 studies, 14 studies included transplant ineligible patients, three studies included transplantation or chemotherapy relapsed patients, while two studies included elderly, frail patients who could be considered as transplant ineligible (Aribi et al. 2010; Aviles et al. 2010). Furthermore, the sample size of the included studies varied from 14 patients (Schuster et al. 2017) to 338 patients (Kuruvilla et al. 2015; Dang et al. 2014), and the median follow-up time varied from 4.7 months (Ohmachi et al. 2013) to 65 months (Mounier et al. 2013).
Across the included studies, ORR was the most commonly reported outcome. All 19 studies reported ORR, and 12 of these measured ORR as the primary outcome (El Gnaoui et al. 2007; Lakshmaiah et al. 2015; Mounier et al. 2013; Neelapu et al. 2017a; Ohmachi et al. 2013; Schuster et al. 2018; Schuster et al. 2017; Vacirca et al. 2014; Wiernik et al. 2008; Zinzani et al. 2011). The second most commonly reported outcome was CR (see Figure 21). Secondary outcomes included PFS, OS, EFS, DOR, safety, and discontinuations.
This SLR provides a comprehensive summary of the currently published evidence of treatments for patients with R/R DLBCL. The findings from the SLR highlight that there is a paucity of published RCTs to establish the comparative efficacy/effectiveness and safety of treatments for DLBCL, with only six RCTs meeting the pre-defined inclusion criteria and reporting on R/R DLBCL. Furthermore, it is difficult to summarize the relative efficacy and safety of the different treatment options due to a lack of robust published evidence.
173 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 20 PRISMA Flow Diagram of the Literature Search
SCT: stem cell transplantation Source: Section 8.1, Clinical SLR and Feasibility Report (see Appendix 18).
174 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 21 Assessment Criteria Used to Assess Complete Response (INV-assessed) in the Included Studies
* Includes patients with complete response unconfirmed † Includes patients who were rituximab naive, and rituximab pre-treated (54% of the pixantrone group, and 56% of the comparator group received rituximab). Source: (Sehn et al. 2019).
4.8.1.3 Study Quality Assessment A quality (risk of bias) assessment of the RCTs was conducted using the Cochrane Collaboration tool. This approach was based on the guidance provided by the Centre for Reviews and Disseminations for assessing the quality of studies included in SLRs and assess the likelihood of selection, performance, attrition and detection bias.
A formal quality assessment was performed to assess the risk of potential bias in the studies, which met the inclusion criteria for the SLR. Quality assessment was performed for RCTs independently by two reviewers using the Cochrane critical appraisal tool (see Appendix 10 of Clinical SLR and Feasibility Report provided in Appendix 18). Figure 22 displays the distribution of risk of bias across categories and Table 52 presents the risk of bias per study.
175 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 22 Risk of Bias Across Categories
Risk 1: Random sequence generation (selection bias); Risk 2: Allocation concealment (selection bias); Risk 3: Blinding of participants and personnel (performance bias); Risk 4: Blinding of outcome assessment (detection bias); Risk 5: Incomplete outcome data (attrition bias); Risk 6: Selective reporting (reporting bias); Risk 7: Other bias. Source: Section 9, Clinical SLR and Feasibility Report (see Appendix 18).
Table 52 Risk of Bias Across Studies
Risk 1 Risk 2 Risk 3 Risk 4 Risk 5 Risk 6 Risk 7 Aribi et al. Unclear Unclear Yes No Unclear Unclear Unclear Avilés et al. Unclear Unclear Unclear Unclear Unclear Unclear Unclear Dang et al. Unclear Unclear No Unclear Yes Unclear Unclear Sehn et al. Yes Yes No Yes Yes Yes Yes Pettengell et al. Yes Yes No Yes Unclear Unclear Yes Czuczman et al. Unclear Unclear No Yes Unclear Unclear Unclear
Risk 1: Random sequence generation (selection bias); Risk 2: Allocation concealment (selection bias); Risk 3: Blinding of participants and personnel (performance bias); Risk 4: Blinding of outcome assessment (detection bias); Risk 5: Incomplete outcome data (attrition bias); Risk 6: Selective reporting (reporting bias); Risk 7: Other bias. Source: Section 9, Clinical SLR and Feasibility Report (see Appendix 18).
The quality assessment of included observational studies (including prospective interventional studies) were conducted using the Quality Assessment Tool for Quantitative Studies produced as part of the Effective Public Health Practice Project (EPHPP). For every study, each component was rated as: strong, moderate, or weak, which was used to assign an overall rating for the study (see Appendix 10 of Clinical SLR and Feasibility Report provided in Appendix 18).
176 PTJA06 – Submission Dossier by Roche Registration GmbH
4.8.2 Indirect Treatment Comparison (ITC) Feasibility Assessment Based on the results of the SLR, a feasibility assessment for an ITC was undertaken to explore the plausibility of conducting a formal synthesis of evidence identified. However, it was not possible to construct a connected network of evidence using the six RCTs identified in the SLR (Figure 23).
A supplementary process was then conducted, whereby studies not meeting the SLR protocol criteria were re-evaluated in order to see whether they could be brought in to help construct a connected network of evidence and fill data gaps, to provide links between comparators to form a network. Despite this broader inclusion, it was still not possible to construct a single network of evidence, and thus, conventional network meta-analyses methods were deemed unfeasible at the outset. Therefore, only studies meeting the SLR criteria were taken forward into the ITC feasibility assessment and included an evaluation of data for the following key comparators of interest in addition to polatuzumab vedotin in combination with BR:
Bendamustine plus rituximab (BR)
Rituximab, gemcitabine and oxaliplatin (R-GemOx)
Gemcitabine, cisplatin and dexamethasone (GDP)
Pixantrone (Pixuvri)
CAR-Ts:
o axicabtagene ciloleucel (Yescarta)
o tisagenlecleucel (Kymriah)
177 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 23 Disconnected Network of Evidence
BR: bendamustine plus rituximab; DHAP: dexamethasone, cytarabine, cisplatin; ESHAP: etoposide, methylprednisolone, cytarabine, cisplatin; GDP: gemcitabine, cisplatin and dexamethasone; R: rituximab; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-ESHAP: rituximab, etoposide, methylprednisolone, cytarabine, cisplatin; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide; SCT: stem cell transplantation; SLR: systematic literature review Note Black circles represent SCT ineligible; White circles represent SCT eligible Source: Appendix 11, Clinical SLR and Feasibility Report (see Appendix 18).
178 PTJA06 – Submission Dossier by Roche Registration GmbH
4.8.2.1 Identification and Selection of Relevant Studies Studies should be ordered by study design (RCT and non-RCT [if applicable]) and status (complete and ongoing).
Include all the studies of the technology relevant to the assessment, as well as studies of comparator technologies (if applicable).
For assessments where there are a lot of studies, more than one table may be needed; for example, dividing the evidence by complete and ongoing studies, randomized and non- randomized evidence or evidence for the technology versus evidence for the comparator(s).
In a table provide a list of the relevant studies identified.
As no ITC could be performed based upon randomized evidence, demonstration of the effectiveness of polatuzumab vedotin in combination with BR for the treatment of adult patients with R/R DLBCL who are not candidates for HSCT, will be based solely on the polatuzumab vedotin development program.
One study from the polatuzumab vedotin development program has been identified in the SLR (see Table 53):
GO29365 (NCT02257567)
The quality rating, using the Cochrane Collaboration tool for assessing risk of bias, results in a rating of “low bias” for GO29365.
179 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 53 List of All Relevant Studies
Status Study Available Documentation* (Ongoing**/ Reference/ID Completed) Randomized Controlled Trials Interim Clinical Study Report (CCOD: 30 April 2018) Summary of Clinical Efficacy (CCOD: 30 April 2018) Summary of Clinical Safety (CCOD: 30 April 2018) Clinical Overview (CCOD: 30 April 2018) BLA 90-day Safety Update Report (see GO29365 Appendix 15) Study Protocol (Version 7, last amendment: Ongoing Clinicaltrials.gov: 2 October 2018) (see Appendix 1) (estimated study (NCT02257567) Statistical Analysis Plan (included in Section 6 of the Study Protocol provided in Appendix 1) completion date: Q1 EudraCT Number: Publications: 2022) 2014-001361-28 o (Herrera et al. 2016) o (Matasar et al. 2017) o (Sehn et al. 2017a) o (Sehn et al. 2018a) o (Sehn et al. 2018b) o (Sehn et al. 2017b) BLA: Biologics License Application; CCOD: clinical cut-off date; * Include references to all linked documents and indicate the expected date of publication for any unpublished clinical studies. ** Include expected date of completion.
4.8.3 Matching-Adjusted Indirect Comparisons (MAICs)
4.8.3.1 Overview of Matching-Adjusted Indirect Comparison Methodology Given the absence of head-to-head randomized data, indirect comparisons were used to compare Pola BR to other therapies using results reported in single-arm trials. As an alternative, more focused ITC approaches may be explored to estimate the comparative efficacy and safety of Pola BR versus other key comparators of interest. However, cross-trial comparisons can be biased by differences in patient populations, differences in the definitions of outcome measures, and limitation in modeling due to variant sample sizes and the availability of baseline covariates (Signorovitch et al. 2012). Matching-adjusted indirect comparisons (MAICs) can be conducted in this setting to try to reduce cross-trial differences and provide meaningful comparative evidence. The individual patient characteristics from patients in the GO29365 study were used to match (as best as possible) the baseline summary statistics reported in the other trials. After matching the populations, weighting was performed and the outcomes are then reported for more balanced populations. Hidden bias due to unknown or unreported covariates
180 PTJA06 – Submission Dossier by Roche Registration GmbH
may still remain; however, in the absence of head-to-head randomized comparisons, this is the best possible approach.
MAICs were used to establish links to interventions not connected to the network through RCTs. MAICs were used to compare Pola BR to four treatments: R-GemOx (Mounier et al. 2013), pixantrone (Pettengell et al. 2012), Kymriah (JULIET) (Schuster et al. 2019) and Yescarta (ZUMA-1) (Neelapu et al. 2017a). For BR, the data presented in Section 4.4.2 demonstrates the superiority of Pola BR compared to BR in a randomized head-to-head comparison (GO29365 study); therefore, a MAIC to BR literature data was not performed.
The MAICs were conducted in line with the methodology described in the NICE Decision Support Unit (DSU) Technical Support Document 18 (Phillippo et al. 2016). Study-specific weights were derived to match the R/R DLBCL patients of the GO29365 study who received Pola BR with those populations reported in the single-arm studies separately. The weights then allowed for the derivation of adjusted CR rates and KM plots for OS for each of the single-arm studies.
For the MAICs, the pooled Pola BR population consisting of 88 patients (Phase Ib: n 6, randomized Phase II: n 40, and Arm G: n 42) was used in order to increase the Pola BR sample size. This is the ITT population of all patients assigned to Pola BR.
It is important to note that there are major caveats in the comparison to CAR-T studies, which cannot be addressed by MAICs and put Pola BR data from the GO29365 study at a disadvantage in the comparison. Unlike typical randomized trials with conventional oncology treatments, the two pivotal CAR-T trials are both single-arm studies and reported results primarily in a modified intent-to-treat (mITT) patient population, which only considers those patients who received treatment. However, the ITT population was used for the MAICs in order to avoid inherent bias in results that contain only treated patients. In the JULIET study, 165 patients were enrolled but only 115 patients had received Kymriah treatment at the time of the latest data cut (Schuster et al. 2019). Similarly, in the ZUMA-1 study, 81 patients were enrolled but only 77 patients received Yescarta treatment (Kite G-BA Assessment 2018b). Additionally, in the Kymriah European Assessment Report (EPAR), the CHMP states “there is a concern that the use of the efficacy analysis set (EAS) for all efficacy outcomes ignores the impact of waiting time and bridging therapy thus leading to an enrichment of the patient population in the EAS for patients having a better prognosis, and an overestimation of efficacy for Kymriah” (Kymriah EPAR 2018).
Covariates were selected based on factors that could impact outcomes. For example, patient characteristics, age (Hedström et al. 2015; Smith et al. 2015), and performance status (Smith et al. 2015; Sehn et al. 2017a) are known to impact outcomes. In terms of disease characteristics, number of prior lines of treatment, and refractory disease or
181 PTJA06 – Submission Dossier by Roche Registration GmbH
early relapse (Gisselbrecht et al. 2010; Van Den Neste et al. 2016; Crump et al. 2017) can impact efficacy (Vardhana et al. 2017) and thus these were selected as well. Finally, in the third-line (and beyond) population, patients may have failed prior autologous transplant and such patients do poorly (Van Den Neste et al. 2017; Crump et al. 2017). Pre-selected prognostic factors in the order of importance were:
Age at baseline (mean, SD where available) Eastern Cooperative Oncology Group Performance Status (ECOG PS) score at baseline Proportion of refractory patients Number of prior treatment lines (1 prior treatment line) Prior SCT
Endpoints Compared CR is the primary focus of the comparison.
Reported (non-adjusted) CR rates were taken from the relevant published literature (Mounier et al. 2013; Kymriah EPAR 2018; Kite G-BA Assessment 2018b; Kite G-BA Addendum 2018a; Pettengell et al. 2012).
For the unweighted rates, 95% CI by Clopper Pearson exact method were used, and for the unweighted rate differences, 95% CI by Wilson were used to be consistent with the GO29365 study. For the weighted rates and rate differences, empirical bootstrapped CIs were used. For the unweighted HRs the 95% Wald CIs were reported. Furthermore, for the weighted HR, 95% Wald CIs were constructed using robust variance estimators.
4.8.3.2 Comparisons to R-GemOx, Pixantrone, Kymriah, and Yescarta
4.8.3.2.1 Comparison to R-GemOx The Mounier et al. (2013) study was a single-arm multicenter Phase II study that evaluated R-GemOx in 49 R/R DLBCL transplant-ineligible patients. Patients were eligible if they were in first or second relapse and had been previously treated with a standard chemotherapy regimen that included an anthracycline with or without rituximab. Only efficacy results that included patients who were treated with prior rituximab are provided below, as rituximab in the first-line setting is standard of care. From the GO29365 pooled Pola BR population, only patients who had received 12 prior lines of treatment were selected, to more closely match the population in the Mounier et al. (2013) study. Responses were evaluated according to Cheson 1997, with CT after 4 and 8 cycles given every 2 weeks. Mounier et al. (2013) reported responses after 4 cycles (induction) whereas in the GO29365 study reported responses are from the end of treatment (i.e. PRA performed 68 weeks after Cycle 6 Day 1 or last treatment). The CR/Cru rate reported by Mounier et al. (2013) after induction was 44% and after 4 additional cycles (maintenance) it was slightly lower at 38%. Thus a comparison CR
182 PTJA06 – Submission Dossier by Roche Registration GmbH
rate at PRA of GO29365 to the CR rate in Mounier et al. (2013) after 4 cycles would not be biased in favor of GO29365. Investigator-reported responses were used from the GO29365 study, as IRC was not used in Mounier et al. (2013).
Endpoints In addition to CR comparisons and median OS comparisons are also provided.
Methodology for Matching-Adjusted Indirect Comparison with R-GemOx based on the Mounier et al. (2013) study Patient baseline characteristics from the GO29365 study were matched with the patient baseline characteristics from the Mounier et al. (2013) study based on the following pre-selected prognostic factors: age at baseline 60 years, ECOG PS score 2, refractory to last line of treatment, number of prior treatment lines, and prior SCT. A total of 65 (out of 88 in the pooled Pola BR group) R/R DLBCL patients from the GO29365 study had a complete set of pre-selected prognostic factor values available and were thus eligible for inclusion in the comparison. Patients who received one or two prior lines of treatment were included in the comparison since the Mounier et al. (2013) study also reported data for this patient population. Furthermore, patients who had received three prior lines of treatment were also included in the comparison provided that their third-line treatment was a bone marrow transplant used for consolidation (i.e., patient entered study at first or second relapse) since the Mounier et al. (2013) study also reported data for this patient population. From the Mounier et al. (2013) study aggregate level pre-selected prognostic factor values and CR data are available for 49 patients. The observed CR rates for patients who received prior rituximab was 42% and was 48% for patients without prior rituximab treatment. The corresponding median OS was 8 months and 27 months for patients who received prior rituximab and for patients without prior rituximab treatment, respectively. Therefore, since adjustment for prior rituximab treatment could not be performed, it was assumed that for the CR comparison the use of the overall (ITT) CR rate was acceptable. The median OS of 8 months reported among the patients with prior rituximab treatment was considered to be comparable with the median OS in the GO29365 study and hence the populations are comparable. Since there were no KM plots available for OS for the prior rituximab treated population it was therefore not possible to estimate the HR. The pre-selected prognostic factors for patients included from the GO29365 and the Mounier et al. (2013) studies are summarized in Table 54 and Table 55. Any discrepancies between the above listed pre-selected prognostic factors from the two studies will be addressed by the MAIC. Median survival time estimates with bootstrapped CIs are reported for each intervention and the rate difference with the corresponding CIs.
183 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 54 Summary of Baseline Characteristics of Mounier et al. (2013) and Matched GO29365 Populations
GO29365* Mounier et al. (2013) Pooled Pola BR Phase II BR N 49 N 65 N 24 Median age, years 70 (2794) 72 (4283) 69 (4177)a (range) 60: 80% 60: 83% 60: 69% Sex, Male 66.2% 58.3% 55% ECOG PS at baseline 01 89.2% 83.3% 78% 24 10.8% 16.7% 22% Lines of prior treatment 1 43.1% 50.0% 86% 2 56.9%b 50.0% b 14% Median time from last treatment, 4.0 4.9 14 months (range) (0.6391.5) (0.698.3) (1130) DOR to last treatment 1 year 81.5% 75.0% 54% 1 year 18.5% 25.0% 46% Primary refractory 24.6%d 41.7%d 12%c First relapse 18.5% 8.3% 74% Secondary refractory 52.3% 37.5% 0 Second relapse 4.6% 12.5% 14% Prior bone marrow transplant 15.4% 12.5% 35% Prior anti-CD20 98.5% 100% 63% Ann Arbor Stage III/IV at 76.9% 87.5% 88% enrollment Saa-IPIe IPI IPI 01: 21.5% 01: 8.3% 01: 31% 2: 21.5% 2: 12.5% 23: 69% IPI score at enrollment 3: 35.4% 3: 33.3% Saa-IPI 4: 20.0% 4: 33.3% 01: 31% 5: 1.5% 5: 12.5% 23: 69% DOR: duration of response; ECOG: Eastern Cooperative Oncology Group; IPI: International Prognostic Index; PS: performance status; Saa-IPI: second-line age-adjusted International Prognostic Index * GO29365 clinical cut-off date: 30 April 2018. a Enrollment included only patients aged 18 to 75 years b High-dose chemotherapy and transplant were considered as a separate line from salvage therapy in the GO29365 study. This figure includes patients that had 3 prior lines of treatment with transplants captured as third-line in the GO29365 study (and no further lines of treatment).
184 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 54 Summary of Baseline Characteristics of Mounier et al. (2013) and Matched GO29365 Populations (cont.) c Six patients enrolled in Mounier et al. (2013) had primary refractory disease. d Defined as no response or progression or relapse within 6 months of last anti-lymphoma treatment end date e Reported two discrepant set of numbers.
Table 55 Summary of Matching Pre-Selected Prognostic Factors for Pooled Pola BR (GO29365) Group Versus R-GemOx (Mounier et al. 2013)
Pola BR Mounier et al. (2013) GO29365 N 49 N 65 Age at baseline 60 years 80% 69% ECOG PS score 2 10.8% 22% Refractory patients 76.9% 12% Number of prior treatment lines 56.9% 14% 1 Prior SCT 15.4% 35% ECOG: Eastern Cooperative Oncology Group; PS: performance status; SCT: stem cell transplant.
MAIC Results: Pola BR (GO29365) vs. R-GemOx (Mounier et al. 2013) The estimated unweighted investigator assessed CR rate for patients in the pooled Pola BR group was 44.6% (95% CI: [32.3, 57.5]) compared with 42.9% (95% CI: [28.8, 57.8]) for R-GemOx in the Mounier et al. (2013) study. The estimated weighted investigator assessed CR rate difference is 1.8% (95% CI: [20.0, 16.9]). These results are summarized below in Table 56.
Table 56 Unweighted Complete Response Rates for Pola BR (GO29365) and R-GemOx (Mounier et al. 2013)
Pola BR Mounier et al. (2013) GO29365 N 49 N 65 Unweighted CR rate, n 44.6% 42.9% (95% CI) (32.3, 57.5) (28.8, 57.8) Difference in CR rates 1.8 (95% CI) (20.0, 16.9) CI: confidence interval; CR: complete response
The weighted ESS after matching with Mounier et al. (2013) patient baseline characteristics was 9.5. The estimated weighted investigator assessed CR rate for patients in the pooled Pola BR group was 80% (95% CI: [60.7, 94.1]) compared with
185 PTJA06 – Submission Dossier by Roche Registration GmbH
42.9% (95% CI: [28.8, 57.8]) for R-GemOx in the Mounier et al. (2013) study. The estimated weighted investigator assessed CR rate difference is 36.2% (95% CI: [19.2, 72.8]; see Table 57).
Table 57 Weighted Complete Response Rates for Pola BR (GO29365) and R-GemOx (Mounier et al. 2013)
Pola BR Mounier et al. (2013) GO29365 N 49 ESS 9.8 CR Rate 80.1% 42.9% (95% CI) (63.9, 100) (28.8, 57.8) Difference in CR rates 37.2% (95% CI) (15.9, 76.1) CI: confidence interval; CR: complete response; ESS: effective sample size.
A sensitivity analysis was performed by conducting separate MAICs for each of the individual pre-selected prognostic factors mentioned above. The results of the sensitivity analysis are presented in Table 58. The weighted ESS after matching ranged from 19.4 to 60.7 and the weighted investigator assessed CR rate difference ranged from 31.5% to 0%.
186 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 58 Separate MAICs by Individual Pre-Selected Prognostic Factor: Pola BR (GO29365) Versus R-GemOx (Mounier et al. 2013)
Difference in CR CR Rate Rate to Mounier Median OS ESS (95% CI) et al. (2013) (months) (95% CI) 44.6% 1.8% 16.8 Unweighted comparison 56 (32.2, 57.5) (20.0, 16.9) (9.2, NR) 80.1% 37.2% 18.4 Full MAIC* 9.7 (63.9, 100) (15.9, 76.1) (12.7, NR) Separate MAICs by Individual Pre-Selected Prognostic Factor Age at baseline 60 42.8% 0 16.8 60.7 years (30.1, 54.3) (18.4, 19.3) (8.9, NR) 46.2% 4.3% 12.6 ECOG PS score 2 56.9 (32.5, 59.5) (15.7, 21.5) (8.9, NR) 74.4% 31.5% NR Refractory patients 19.4 (57.0, 94.5) (10.0, 55.7) (12.6, NR) Number of prior 48.6% 5.7% 18.4 37.3 treatment lines 2 (32.6, 65.1) (15.0, 27.4) (9.2, NR) 45.8% 3.0% 16.8 Prior SCT 50.5 (31.9, 60.4) (17.2, 22.9) (9.2, NR) CI: confidence interval; CR: complete response; ECOG: Eastern Cooperative Oncology Group; MAIC: matching-adjusted indirect comparison; OS: overall survival; PS: performance status; SCT: stem cell transplant * Full MAIC, defined as comparing all the pre-selected prognostic factors (age at baseline, ECOG PS, refractory status, number of prior lines of treatment, and prior SCT) simultaneously.
The estimated unweighted median OS for patients in the pooled Pola BR group was 16.8 months compared with 8 months for patients treated with R-GemOx and who received prior rituximab treatment. The estimated weighted median OS for patients in the pooled Pola BR group based on the full MAIC was 18.4 months and the lower level of the 95% CI is 12.7.
The estimated weighted median OS for patients in the pooled Pola BR group in each of the separate MAICs conducted for each individual pre-selected prognostic factor ranged from 12.6 months to median not reached.
The results of the MAICs reflect the large imbalances in the level of pre-selected prognostic factors, especially in the rate of refractory patients 12.2% (Mounier et al. 2013) versus 76.9% (GO29365 comparable study population). Applying the full set of pre-selected prognostic factors (age at baseline, ECOG PS, refractory to last line of treatment, number of prior lines of treatment, and prior SCT) reduced the ESS from 65 to 9.7, whereas adjusting for refractoriness alone reduced the ESS to 15.9. Applying a subset of pre-selected prognostic factors such as refractory to last line of treatment,
187 PTJA06 – Submission Dossier by Roche Registration GmbH
number of prior lines of treatment, and prior SCT produced results which were almost identical to those observed with the full MAIC (defined as comparing all the pre-selected prognostic factors simultaneously). However, adjusting for number of prior lines of treatment or prior SCT did not seem to have an impact on the results. When taking into account the R/R status of the patients, as expected a change in CR and median OS occurs in favor of Pola BR.
Conclusions from Comparison of Pola BR (GO29365) to R-GemOx (Mounier et al. 2013) In a side-by-side comparison, Pola BR showed higher CR and ORR compared to R-GemOx when GO29365 patients were selected to match as best as possible the patient population in Mounier et al. (2013). MAICs were also performed and further supported the improved efficacy that Pola BR demonstrates over R-GemOx in terms of response rates. After accounting for the imbalances in refractory status, Pola BR shows a clear improvement in efficacy compared to R-GemOx.
4.8.3.2.2 Comparison to Pixantrone
Methodology for Matching-Adjusted Indirect Comparison with Pixantrone An indirect comparison was made to pixantrone monotherapy using the pivotal PIX301 study (Pixuvri SmPC; Pettengell et al. 2012). In the PIX301 study, the CR rate (CR/Cru) at EOT as assessed by the Independent Assessment Panel (IAP) was 20% (14/70) with a median PFS (IAP) of 5.3 months (95% CI: [2.3 months, 6.2 months]). Median OS was reported to be 10.2 months (95% CI: [6.4 months, 15.7 months]).
Differences in eligibility criteria may make cross-trial comparisons challenging. The PIX301 study enrolled patients with multiply R/R disease (i.e. two or more prior lines of treatment) whereas the GO29365 study excluded patients with only one relapse. The PIX301 study excluded anthracycline-refractory patients (patients had to have responded to a prior standard anthracycline-containing regimen for at least 24 weeks) which would mean that those who were primary refractory would be excluded as standard first-line treatments for DLBCL contain anthracyclines. The GO29365 study included primary refractory patients; therefore, patients from the GO29365 study were selected if they were non-primary refractory to match the population of the PIX301 study. In addition, only patients who had received at least two prior lines of treatment were selected for this analysis since the population in the PIX301 study only include patients with at least two prior lines. With the resulting small sample size, conducting a full MAIC (defined as adjusting for all pre-selected prognostic factors simultaneously) was not feasible; therefore, separate MAICs for each individual pre-selected prognostic factor (age at baseline, ECOG PS score at baseline, proportion of refractory patients, number of prior lines of treatment, and prior SCT) were conducted. In the below section, results from the unweighted comparison as well as results from the separate MAICs for each of the individual pre-selected prognostic factors listed above are presented.
188 PTJA06 – Submission Dossier by Roche Registration GmbH
Patient baseline characteristics from the GO29365 study were matched with the patient baseline characteristics from the PIX301 study based on the above-mentioned pre-selected prognostic factors. A total of 21 (out of 88 in the pooled Pola BR group) R/R DLBCL patients from the GO29365 study had a complete set of pre-selected prognostic factor values available and were thus eligible for inclusion in the comparison. Patients with ECOG PS 2 who were non-primary refractory were included in the comparison (naive unweighted and individually weighted) since the PIX301 study also reported data for this patient population. Patients in the GO29365 study who had received one prior line of treatment were excluded since these patients were not included in the PIX301 study. From the PIX301 study aggregate level pre-selected prognostic factor values and CR and OS data are available for 70 patients (Pettengell et al. 2012). The pre-selected prognostic factors for patients included from both the GO29365 and PIX301 studies are summarized in Table 59. Any discrepancies between the above listed pre-selected prognostic factors from the two studies will be addressed by factor in separate MAICs.
Table 59 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Pixantrone (PIX301)
Pola BR Pixantrone GO29365 PIX301 (N 21) (N 70) Age 60 years* 86% 51% ECOG PS 12 62% 63% Refractory patients 90% 57% Number of prior treatment lines 3 3 (median)† (25) (29) Prior SCT 38% 16% ECOG: Eastern Cooperative Oncology Group; PS: performance status; SCT: stem cell transplant * The PIX301 study only reported median (range) for age, no mean and standard deviation values were available. † Patients who had received one prior line of treatment were already excluded from the GO29365 R/R DLBCL patient subpopulation used for the comparison therefore, no further adjustment was needed. Source: Pettengell et al. 2012.
MAIC Results: Pola BR (GO29365) vs. Pixantrone (PIX301) The estimated unweighted centrally assessed CR rate at EOT for patients treated with Pola BR was 66.7% (95% CI: [43.0, 81.9]) compared with 20.0% (95% CI: [11.4, 31.3]) for pixantrone. The estimated unweighted CR rate difference is 46.7% (95% CI: [22.7, 67.9]). These results are presented in Table 60.
189 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 60 Unweighted Complete Response Rates for Pola BR (GO29365) and Pixantrone (PIX301)
Pola BR Pixantrone G029365 PIX301 N 21 N 70 Unweighted CR rate, n 66.7% 20.0% (95% CI) (43.0, 85.4) (11.4, 31.3) Difference in CR rates 46.7% (95% CI) (22.7, 67.9) CI: confidence interval; CR: complete response. Source: Pettengell et al. 2012.
Separate MAICs for each of the individual pre-selected prognostic factors mentioned above were conducted and the results are presented in Table 61. The weighted effective sample size (ESS) after matching ranged from 9.1 to 21.0 and the weighted CR rate difference ranged from 46.5% to 60.2% with all the lower limits of the confidence intervals above 0. The results of the individual weighted MAICs suggest that Pola BR performs better than pixantrone (PIX301).
Table 61 Individual MAICS for Complete Response Rates: Pola BR (GO29365) Versus Pixantrone (PIX301)
CR Rate Difference for PIX301 ESS (95% CI) 46.7% Unweighted comparison 21 (22.7, 67.9) Separate MAICs by Individual Pre-Selected Prognostic Factor 60.2% Age 60 years 10.6 (45.3, 79.5) 46.5% ECOG PS 21.0 (26.0, 70.1) 59.0% Refractory patients 9.1 (43.0, 82.8) 52.6% Prior SCT 17.4 (30.9, 76.8) CI: confidence interval; CR: complete response; ECOG: Eastern Cooperative Oncology Group: EOT: end of treatment; ESS: effective sample size; MAICs: matching-adjusted indirect comparisons; PS: performance status; SCT: stem cell transplant PIX301: CR rate at EOT was 20.0% (95% CI: [11.4, 31.3]) Note: Number of prior treatment lines was not adjusted for as patients who had received one prior line of treatment were already excluded during patient selection.
The estimated unweighted median OS for patients treated with Pola BR was 32.0 months compared with 10.2 months for patients treated with pixantrone. The unweighted Cox Model based HR (Pola BR vs. pixantrone [PIX301]) was 0.30
190 PTJA06 – Submission Dossier by Roche Registration GmbH
(95% CI: [0.12, 0.76]). The unweighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot of pixantrone is presented in Figure 24.
Figure 24 Unweighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301)
Separate MAICs for each of the individual pre-selected prognostic factors mentioned above were conducted for OS and the results are presented in Table 62. The estimated weighted median OS for patients treated with Pola BR was 32 months for all factors evaluated and the HRs ranged from 0.18 to 0.31. For each separate comparison, the upper confidence interval was below 1. The weighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot of pixantrone is presented in Figure 25.
191 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 62 Individual MAICs for Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301)
OS HR Median OS ESS (95% CI) (95% CI) Unweighted 0.30 32.0 21.0 comparison (0.12, 0.76) (16.8, NR) Separate MAICs by Individual Pre-Selected Prognostic Factor 0.18 32.0 Age 60 years 10.6 (0.07, 0.52) (16.8, NR) 0.31 32.0 ECOG PS 21.0 (0.13, 0.75) (12.4, NR) 0.18 32.0 Refractory patients 9.1 (0.06, 0.54) (16.8, NR) 0.23 32.0 Prior SCT 17.4 (0.08, 0.69) (28.0, NR) CI: confidence interval; ECOG: Eastern Cooperative Oncology Group; ESS: effective sample size; HR: hazard ratio; MAICs: matching-adjusted indirect comparisons; OS: overall survival; PS: performance status; SCT: stem cell transplant. PIX301: Median OS was 10.2 months (95% CI: [6.4, 15.7]) Note: Number of prior lines of treatment was not adjusted for as patients who had received one prior line of treatment were already excluded during patient selection.
192 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 25 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Pixantrone (PIX301)
Conclusions from Comparison of Pola BR (GO29365) to Pixantrone (PIX301) Pola BR efficacy in terms of CR and OS were better when compared to pixantrone after both unweighted and weighted analyses. Overall, the PIX301 study included patients who had a better disease prognosis compared to the patients included in the GO29365 study since the PIX301 study excluded primary refractory patients. One limitation with these comparisons is that after matching, the ESS for Pola BR decreases especially when primary refractory patients are removed from the GO29365 R/R DLBCL patient population in order to match the population in the PIX301 study. The higher CR rates seen in the matched GO29365 R/R DLBCL patient population clearly surpass those observed in the PIX301 study. Furthermore, as additional support, the OS observed from the matched GO29365 R/R DLBCL patient population also appears better compared to pixantrone.
4.8.3.2.3 Comparison to CAR-T Cell Therapies CAR-T cell therapies are a unique modality of treatment making comparisons to Pola BR or even other chemo-immunotherapy regimens difficult. As current CAR-Ts are manufactured on an individualized basis and can only be administered at select specialist centers, patients who require rapid disease control are unlikely to be suitable candidates. Patients eligible for treatment with CAR-T therapy must have disease kinetics that enable them to be able to wait out the prolonged manufacturing periods
193 PTJA06 – Submission Dossier by Roche Registration GmbH
following apheresis as well as waiting for an appointment to be seen at the specialist center, as well as the time required to schedule apheresis. Furthermore, an additional potential logistical barrier for some patients is the requirement to relocate to within one-hour of the CAR-T specialist center for a duration of 4 weeks. The two studies that served as the pivotal trials for the CAR-Ts, JULIET for Kymriah and ZUMA-1 for Yescarta differed in their allowance of bridging chemotherapy during the period between apheresis and infusion as well as their estimates on planned manufacturing timelines which could impact patients selected to be screened for the trials and subsequently enrolled and treated. The two CAR-T cell therapies will be compared to Pola BR using the ITT population from these trials.
The first comparison will be with Kymriah using the pivotal study, JULIET, in which the treatment paradigm included bridging therapy followed by lymphodepletion and CAR-T cell infusion. As shown in Figure 26, apheresis occurs during the screening period and the estimated time given from apheresis to CAR-T cell infusion was around 16 weeks, during which bridging therapy was permitted.
Figure 26 JULIET Study Schema
Source: Kymriah EPAR 2018.
The second comparison will be with Yescarta using the pivotal study, ZUMA-1, which did not allow bridging therapy. The manufacturing period was estimated to be around 710 days with infusion following approximately 5 days later, a much shorter time period than in the JULIET study. As shown in Figure 27, patients were screened prior to leukapheresis in the ZUMA-1 study whereas in the JULIET study, screening and apheresis occurred during the same period of time.
194 PTJA06 – Submission Dossier by Roche Registration GmbH
Figure 27 ZUMA-1 Study Schema
Source: Yescarta EPAR 2018.
The use of bridging therapy differs between the two studies and may in part explain the differences in efficacy when examining the studies from an ITT perspective as well as having to wait different time periods (longer time to infusion in the JULIET study).
These wait times are not insignificant. For Kymriah, the median time from enrollment to infusion was 54 days (range: 30357 days). A recent publication by Schuster et al. (2019) details the reasons why 50 patients discontinued from the JULIET study before receiving CAR-T treatment and the reasons were as follows: failure of CAR-T manufacture (n 12), death prior to infusion (n 16), treating physician decided against further participation (n 16), patient decided against further participation (n 2), protocol deviation (n 1), and adverse events (n 3). The main reason for discontinuations such as death, physician’s decision, and patient’s decision was due to disease progression. Most of the patients who were unable to be re-infused were refractory patients at the time of enrollment.
For Yescarta in the ZUMA-1 study, the median time from leukapheresis to infusion was 24 days (range: 1673 days), much shorter than that for Kymriah. Out of 111 leukapheresed patients, 101 patients were infused (Yescarta SmPC).
In order to be able to compare Pola BR with CAR-T cell therapies, the focus has been on indirect comparisons of clinical trial data based on patient eligibility for the different trials. Previously the extended manufacturing period for CAR-T cell therapies has been discussed under ‘convenience’. However, the issue of manufacturing time should also be considered as part of the efficacy comparison. Third-line (and beyond) R/R DLBCL
195 PTJA06 – Submission Dossier by Roche Registration GmbH
patients face the risk of early progression and death during the CAR-T cell manufacturing period which has been evident in clinical practice and in CAR-T cell trials.
Given the median waiting period for CAR-T is about 3060 days, we examined the impact of Pola BR on early progression and death by looking at the PFS rate within the first 30 and 60 days, which reflect the manufacturing tomes for CAR-T cells in clinical trials and practice (Kymriah EPAR 2018; Yescarta EPAR 2018; Paillassa et al. 2019). The GO29365 study population is particularly relevant to investigate treatment effect on rapid progression given the high proportion of refractory patients enrolled. As shown in Table 63, the rate of progression or death was reduced as early as 30 days compared to the control arm.
This demonstration of efficacy in a proportion of the target patient population who are inadequately served by CAR-T cell therapy provides additional evidence for the benefit of Pola BR.
Table 63 GO29365: Landmark Progression-Free Survival (Investigator-assessed; CCOD: 30 April 2018)
Pola BR BR All Patients N 40 N 40 30 days 100% 74% 60 days 84% 53% 3L+ Patients N 29 N 28 30 days 100% 67% 60 days 64% 44%
4.8.3.2.3.1 Comparison to CAR-T Cell Therapies: Kymriah The pivotal JULIET study that led to the approval of Kymriah was an open-label multicenter single-arm trial for patients with R/R DLBCL (including transformed follicular lymphoma) who had received at least two prior lines of chemotherapy or whom had relapsed following autologous HSCT. Exclusions included ECOG PS 2, creatinine clearance 60, cardiac ejection fraction 45%, and absolute lymphocyte count of 300/µL. Bridging therapy was permitted between leukapheresis and pre-CAR-T lymphodepletion chemotherapy to allow control of disease burden.
The JULIET study enrolled 165 patients; however, not all patients were able to be infused with CAR-T. Of the 92 patients who received Kymriah (Kymriah SmPC), 90% received physician’s choice of bridging chemotherapy. The median number of bridging regimens was 1 (range: 15) and 83% received 2 regimens.
To avoid selection bias, the ITT population is used in order to facilitate a comparison between Pola BR with the treatment paradigm outlined in the JULIET study (bridging therapy as an option followed by leukapheresis and CAR-T infusion).
196 PTJA06 – Submission Dossier by Roche Registration GmbH
In the JULIET study, the ORR (central review) in all enrolled patients (ITT population; N 165) was 34% (56/165) was 24% (40/165) CR. The median OS and PFS observed for the ITT population treated with Kymriah was 8.2 months (95% CI: [5.8, 11.7]) and 4.4 months (95% CI: [3.6, 5.1]) (Kymriah SmPC 2018).
Methodology for Matching-Adjusted Indirect Comparison with Kymriah based on the JULIET Study Patient baseline characteristics from the GO29365 study were matched with the patient baseline characteristics from the ITT population of the JULIET study based on the following pre-selected prognostic factors: age at baseline, ECOG PS, refractory to last line of treatment, number of prior treatment lines, and prior stem cell transplant. Of the 88 pooled Pola BR patients, 78 patients from the GO29365 study had a complete set of pre-selected prognostic factor values available and were thus eligible for the comparison. To optimize the use of study data, patients from the GO29365 study who had received one prior line of treatment were included since the JULIET study also reported data for patients who had received one prior line of treatment. Patients with ECOG PS 1 were excluded from the GO29365 study population since the JULIET study did not enroll patients with ECOG PS 1 as part of the ITT population. From the JULIET study, aggregate level patient baseline characteristics and CR and OS data were available for 165 patients comprising the full ITT population based on the data cut-off of 8 December 2017 (Kymriah SmPC 2018). The pre-selected prognostic factors for patients included for the comparison from both the GO29365 and JULIET studies are summarized in Table 64. Any discrepancies between the above-mentioned pre-selected prognostic factors from the two studies will be addressed by the MAIC. Plots of the matching-adjusted weighted Pola BR curves together with the digitized KM OS curves were obtained from Schuster et al. (2019). Methodology for median OS estimates and HRs are the same as those described in the pixantrone MAIC methodology section above.
197 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 64 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Kymriah (JULIET)
Pola BR Kymriah GO29365 JULIET N 78* N 165 Age at baseline, mean (SD) 65.1 (12.1) 55.9 (12.9) ECOG PS Score 1 66.7% 53.3% Refractory patients 82.1% 58.2% Number of prior treatment lines 69.2% 96.4% 2 Prior SCT 19.2% 43.6% ECOG: Eastern Cooperative Oncology Group; PS: performance status; SCT: stem cell transplant; SD: standard deviation * In order to adjust for the Kymriah population, the Pola BR patient population selected (N 78) excludes patients with ECOG PS 1 or unknown. Source: Kymriah EPAR 2018.
MAIC Results: Pola BR (GO29365) vs. Kymriah (JULIET) The estimated unweighted centrally assessed best CR rate for patients treated with Pola BR was 47.4% (95% CI: [36.0, 59.1]) compared with 24.2% (95% CI: [17.9, 31.5]) for apheresed patients in the JULIET study (data cut-off date: 8 December 2017). The estimated unweighted naive CR rate difference is 23.2% (95% CI: [9.8, 36.0]; see Table 65).
Table 65 Unweighted Complete Response Rates for Pola BR (GO29365) and Kymriah (JULIET)
Pola BR Kymriah GO29365 JULIET N 78 N 165 Unweighted CR rate 47.4% 24.2% (95% CI) (36.0, 59.1) (17.9, 31.5) Difference in CR rate 23.2% (95% CI) (9.8, 36.0) CI: confidence interval; CR: complete response Source: Kymriah EPAR 2018.
The weighted ESS after matching with JULIET patient baseline characteristics was 14.6. The time-to-event median follow-up time for R/R DLBCL patients in the GO29365 study treated with Pola BR was 19.6 months compared with 19 months for patients treated with Kymriah (Kymriah EPAR 2018). The maximum follow-up times were 49.7 months and 20 months, respectively. The estimated weighted centrally assessed best CR rate for patients treated with Pola BR was 54.3% (95% CI: [35.4, 67.0]) compared with
198 PTJA06 – Submission Dossier by Roche Registration GmbH
24.2% (95% CI: [17.9, 31.5]) for apheresed patients in the JULIET study. The estimated weighted CR rate difference is 30.0% (95% CI: [10.3, 45.2]; see Table 66).
Table 66 Weighted Complete Response Rates for Pola BR (GO29365) and Kymriah (JULIET)
Pola BR Kymriah GO29365 JULIET ESS 14.6 N 165 CR rate 54.3% 24.2% (95% CI) (35.4, 67.0) (17.9, 31.5) Difference in CR rate 30.0% (95% CI) (10.3, 45.2) CI: confidence interval; CR: complete response; ESS: effective sample size
A sensitivity analysis was performed by conducting separate MAICs for each of the individual pre-selected prognostic factors mentioned above. The results of the sensitivity analysis are presented in Table 67. The ESS ranged from 48.2 to 72.4 and the CR rate difference ranged from 12.1 to 38.5. The results of the sensitivity analysis were supportive of the results from the full MAIC.
199 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 67 Sensitivity Analysis: Pola BR (GO29365) Versus Kymriah (JULIET)
Difference in CR CR Rate HR ESS Rate to Kymriah (95% CI) (95% CI) (95% CI) 47.4% 23.2% 0.76 Non-adjusted 78 (36.0, 59.1) (9.8, 36.0) (0.52, 1.1) 54.3% 30.0% 0.54 Full MAIC* 14.6 (35.4, 67.0) (10.3, 45.2) (0.25, 1.18) Sensitivity Analysis Age at baseline, mean 36.4% 12.1% 0.96 48.2 (SD) (24.5, 47.6) (1.7, 26.0) (0.64, 1.42) 48.7% 24.5% 0.68 ECOG PS 1 72.4 (37.0, 60.8) (11.8, 38.9) (0.47, 1.00) 62.7% 38.5% 0.57 Refractory patients 56.3 (55.6, 69.3) (29.3, 47.6) (0.37, 0.88) Number of prior treatment 38.3% 14.0% 0.88 57.9 lines 2 (25.3, 49.9) (0.7, 27.3) (0.60, 1.30) 49.2% 25.0% 0.72 Prior SCT 56.4 (35.7, 63.5) (10.1, 40.9) (0.50, 1.06) CI: confidence interval; CR: complete response; ECOG: Eastern Cooperative Oncology Group; HR: hazard ratio; MAIC: matching-adjusted indirect comparison; PS: performance status; SCT: stem cell transplant; SD: standard deviation * Full MAIC, defined as comparing all the pre-selected prognostic factors (age at baseline, ECOG PS, refractory status, number of prior lines of treatment, and prior SCT) simultaneously.
The estimated unweighted median OS for patients treated with Pola BR was 11.8 months (95% CI: [9.0, 32.0]) compared with 8.2 months (95% CI: [5.8, 11.7]) for patients treated with Kymriah in the JULIET study. The unweighted Cox Model based HR (Pola BR [GO29365] vs. Kymriah [JULIET]) was 0.76 (95% CI: [0.52, 1.11]; see Table 68). The unweighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot of Kymriah from the JULIET study is presented in Figure 28.
200 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 68 Unweighted Median Overall Survival for Pola BR (GO29365) and Kymriah (JULIET)
Pola BR Kymriah GO29365 JULIET N 78 N 165 Median OS, months 11.8 8.2 (95% CI) (9.0, 32.0) (5.8, 11.7) HR 0.76 (95% CI) (0.52, 1.11) CI: confidence interval; HR: hazard ratio; OS: overall survival
Figure 28 Unweighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Kymriah (JULIET)
The estimated weighted median OS for patients treated with Pola BR was not reached (95% CI: [5.3, NR]) compared with 8.2 months (95% CI: [5.8, 11.7]) for patients treated with Kymriah in the JULIET study. The weighted Cox Model based HR (Pola BR [GO29365] vs. Kymriah [JULIET]) was 0.54 (95% CI: [0.25, 1.18]; see Table 69). The weighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot of Kymriah is presented in Figure 29. The results of the full MAIC suggest that Pola BR performs better than Kymriah.
201 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 69 Weighted Median Overall Survival for Pola BR (GO29365) and Kymriah (JULIET)
Pola BR Kymriah GO29365 JULIET ESS 14.6 N 165 Median OS, months NR 8.2 (95% CI) (5.3, NR) (5.8, 11.7) HR 0.54 (95% CI) (0.25, 1.18) CI: confidence interval; ESS: effective sample size; HR: hazard ratio; NR: not reached; OS: overall survival
Figure 29 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Kymriah (JULIET)
Conclusions from Comparison of Pola BR (GO29365) to Kymriah (JULIET) Overall, the analysis supports that Pola BR compares favorably in terms of CR rates as well as OS. Adjusting for the differences in baseline characteristics however, results in a small ESS for the Pola BR arm for the full MAIC which is a limitation. Furthermore, another limitation is that there may be inherent differences in the patient populations that are not accounted for in the baseline characteristics listed. There may be patient selection bias in that Kymriah is a novel modality of treatment and carries with it the risk
202 PTJA06 – Submission Dossier by Roche Registration GmbH
of CRS, which may influence physicians’ decisions on which patients to consider for the JULIET trial at a time when management and outcomes of CRS were less well defined.
4.8.3.2.3.2 Comparison to CAR-T Cell Therapies: Yescarta
Methodology for Matching-Adjusted Indirect Comparison with Yescarta based on the ZUMA-1 Study Patient baseline characteristics from the GO29365 study were matched with the patient baseline characteristics from the DLBCL ITT population of the ZUMA-1 study based on the following pre-selected prognostic factors: age at baseline, ECOG PS, refractory to last line of treatment, number of prior treatment lines, and prior SCT. Of the 88 pooled Pola BR patients, 66 patients from the GO29365 study had data for all the above mentioned pre-selected prognostic factors and were thus eligible for inclusion in the comparison. To optimize the use of the study data, patients from the GO29365 study who had received one prior line of treatment were also included in the comparison since the ZUMA-1 study also included these patients. Additionally, patients with ECOG PS 2 and who were refractory to their last line of treatment or those of whom had relapsed after SCT from the GO29365 study were included in the comparison (naive unweighted and weighted [MAIC]) since this patient population was also included in the ZUMA-1 study. From the ZUMA-1 study, aggregate level baseline characteristics and CR and OS data were available for 81 patients comprising the DLBCL ITT population (Kite G-BA Assessment 2018b; Kite G-BA Addendum 2018a). The pre-selected prognostic factors for patients included in the comparison from the GO2936 and ZUMA-1 studies are summarized in Table 70. Any discrepancies between the above-mentioned pre-selected prognostic factors from the two studies will be addressed by the MAIC. Plots of the matching-adjusted weighted Pola BR curves together with the digitized KM OS curves were obtained for Yescarta. Methodology for median OS estimates and HRs are the same as those described above in the pixantrone MAIC methodology section.
203 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 70 Summary of Matching Pre-Selected Prognostic Factors for Pola BR (GO29365) Versus Yescarta (ZUMA-1)
Pola BR Yescarta GO29365 ZUMA-1 N 66 N 81 Age at baseline, mean 63.2 57.1 (SD) (11.8) (10.6) ECOG PS score 1 68.2% 64% Refractory patients 97.0% 78% Number of prior treatment lines 80.3% 98% 2 Prior SCT 22.7% 23% ECOG: Eastern Cooperative Oncology Group; PS: performance status; SCT: stem cell transplant; SD: standard deviation Source: (Kite G-BA Assessment 2018b; Kite G-BA Addendum 2018a).
MAIC Results: Pola BR (GO29365) vs. Yescarta (ZUMA-1) The estimated unweighted centrally assessed best CR rate for patients treated with Pola BR was 37.9% (95% CI: [26.2, 50.7]) compared with 45.7% (95% CI: [34.6, 57.1]) for apheresed patients in the ZUMA-1 study. The estimated unweighted naive CR rate difference is 7.8% (95% CI: [23.0, 8.5]; see Table 71).
Table 71 Unweighted Complete Response Rates for Pola BR (GO29365) and Yescarta (ZUMA-1)
Pola BR Yescarta GO29365 ZUMA-1 N 66 N 81 Unweighted CR rate 37.9% 45.7% (95% CI) (26.2, 50.7) (34.6, 57.1) Difference in CR rate 7.8% (95% CI) (23.8, 8.5) CI: confidence interval; CR: complete response
The weighted effective sample size after matching with ZUMA-1 patient baseline characteristics was 18.6. The time-to-event median follow-up time for R/R DLBCL patients in the GO29365 study treated with Pola BR was 18.9 months compared with ~20 months for patients treated with Yescarta. The maximum follow-up times were 49.7 months and ~20 months, respectively.
The estimated weighted centrally assessed best CR rate for patients treated with Pola BR was 39.2% (95% CI: [24.6, 56.1]) compared with 45.7% (95% CI: [34.6, 57.1]) for apheresed patients in the ZUMA-1 study. The estimated weighted CR rate difference is 6.5% (95% CI: [25.5, 13.5]; see Table 72).
204 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 72 Weighted Complete Response Rates for Pola BR (GO29365) and Yescarta (ZUMA-1)
Pola BR Yescarta GO29365 ZUMA-1 ESS 18.6 N 81 CR rate 39.3% 45.7% (95% CI) (24.6, 56.1) (34.6, 57.1) Difference in CR rate 6.5% (95% CI) (25.5, 13.5) CI: confidence interval; CR: complete response; ESS: effective sample size
A sensitivity analysis was performed by conducting separate MAICs for each of the individual pre-selected prognostic factors mentioned above. The results of the sensitivity analysis are presented in Table 73. The ESS ranged from 29.7 to 66 and the CR rate difference ranged from 36.0 to 50.0. The results of the sensitivity analysis were supportive of the results from the full MAIC.
Table 73 Sensitivity Analysis: Pola BR (GO29365) Versus Yescarta (ZUMA-1)
Difference in CR CR Rate HR ESS Rate to Yescarta (95% CI) (95% CI) (95% CI) 37.9% 7.8% 1.46 Non-adjusted 66.0 (26.2, 50.7) (23.8, 8.5) (0.94, 2.27) 39.2% 6.5% 1.38 Full MAIC* 18.6 (24.6, 56.1) (25.5, 13.5) (0.57, 3.31) Sensitivity Analysis Age at baseline, mean 36.0% 9.6% 1.67 44.4 (SD) (22.4, 49.6) (28.1, 8.4) (1.02, 2.73) 38.2% 7.5% 1.41 ECOG PS 1 65.5 (26.3, 50.3) (23.3, 7.8) (0.91, 2.20) 50.0% 4.4% 1.03 Refractory patients 29.7 (40.3, 68.2) (9.7, 26.3) (0.54, 1.96) Number of prior treatment 36.1% 9.6% 1.49 55.1 lines 2 (23.3, 49.3) (26.6, 8.5) (0.94, 2.35) 37.9% 7.7% 1.46 Prior SCT 66.0 (25.7, 48.7) (24.2, 9.0) (0.94, 2.26) CI: confidence interval; CR: complete response; ECOG: Eastern Cooperative Oncology Group: ESS: effective sample size; HR: hazard ratio; MAIC: matching-adjusted indirect comparison; PS: performance status; SCT: stem cell transplant; SD: standard deviation * Full MAIC, defined as comparing all the pre-selected prognostic factors (age at baseline, ECOG PS, refractory to last line of treatment, number of prior lines of treatment, and prior SCT) simultaneously.
205 PTJA06 – Submission Dossier by Roche Registration GmbH
The estimated unweighted median OS for patients treated with Pola BR was 9.2 months (95% CI: [7.2, 28.0]) compared with 15.4 months (95% CI: [11.1, NR]) for patients treated with Yescarta in the ZUMA-1 study. The unweighted Cox Model based HR (Pola BR [GO29365] vs. Yescarta [ZUMA-1]) was 1.46 (95% CI: [0.94, 2.27]; see Table 74). The unweighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot for Yescarta from the ZUMA-1 study is presented in Figure 30.
Table 74 Unweighted Overall Survival for Pola BR (GO29365) and Yescarta (ZUMA-1)
Pola BR Yescarta GO29365 ZUMA-1 N 66 N 81 Unweighted median OS, months 9.2 15.4 (95% CI) (7.2, 28.0) (11.1, NR) HR 1.46 (95% CI) (0.94, 2.27) CI: confidence interval; HR: hazard ratio; OS: overall survival
Figure 30 Unweighted KM Plot of Overall Survival: Pola BR (GO29365) Versus Yescarta (ZUMA-1)
206 PTJA06 – Submission Dossier by Roche Registration GmbH
The estimated weighted median OS for patients in the pooled Pola BR group was 9.0 months (95% CI: [4.4, NR]) compared with 15.4 months (95% CI: [11.1, NR]) for patients treated with Yescarta in the ZUMA-1 study. The weighted Cox Model based HR (Pola BR [GO29365] vs. Yescarta [ZUMA-1]) was 1.38 (95% CI: [0.57, 3.31]; see Table 75). The weighted KM plot for Pola BR treated R/R DLBCL patients from the GO29365 study versus the KM plot for Yescarta is presented in Figure 31. The results of the full MAIC suggest that Pola BR potentially performs as well as Yescarta.
Table 75 Weighted Overall Survival for Pola BR (GO29365) and Yescarta (ZUMA-1)
Pola BR Yescarta GO29365 ZUMA-1 ESS 18.6 N 81 Median OS, months 9.0 15.4 (95% CI) (4.4, NR) (11.1, NR HR 1.38 (95% CI) (0.57, 3.31) CI: confidence interval; ESS: effective sample size; HR: hazard ratio; NR: not reached; OS: overall survival
Figure 31 Weighted KM Plots of Overall Survival: Pola BR (GO29365) Versus Yescarta (ZUMA-1)
207 PTJA06 – Submission Dossier by Roche Registration GmbH
Conclusions from Comparison of Pola BR (GO29365) to Yescarta (ZUMA-1) The efficacy comparison between Pola BR and Yescarta is limited by trial design which may impact patient selection into the ZUMA-1 study as the study did not allow for bridging therapy. It is possible that patients in whom physicians felt could wait out the manufacturing time would be selected for enrollment into this study. These types of patient differences are not possible to account for in a cross-trial comparison and should be taken into account when interpreting the data. The weighted CR rate from the MAIC for Pola BR was 39.2% which was 6.5% lower than the CR rate of Yescarta (45.7%) observed in the ZUMA-1 study. The differences in CR rates for Pola BR compared to Yescarta from the separate MAICs adjusted for age at baseline, ECOG PS, refractory status, number of prior lines of treatment, and prior SCT ranged from 9.6% to 4.3%. Altogether, the MAICs suggest that Pola BR could perform as well as Yescarta.
4.8.3.2.3.3 Safety of Pola BR (GO29365) Versus CAR-T Cell Therapies The clinical safety profile of Pola BR offers an advantage over CAR-T cell therapy. Treatment-emergent life-threatening of fatal cytokine-release syndrome (CRS) and neurologic toxicities have occurred following CAR-T cell therapy. CRS was associated with end organ dysfunction resulting in SAEs such as acute kidney injury, hypoxia, hypotension, disseminated intravascular aoagulopathy, cardiac arrest, cardiac failure, hemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH/MAS). The worsening of underlying organ pathologies has also been observed in the setting of CRS. Eleven percent (11%) of DLBCL patients treated with Yescarta (Yescarta SmPC) and 22% of DLBCL patients treated with Kymriah (Kymriah SmPC 2018) experienced Grade 3 or higher (severe, life-threatening, and fatal) CRS. The median time to onset was 2 days for Yescarta and 3 days for Kymriah. The median duration was 7 days for both Yescarta and Kymriah.
In addition, following CAR-T cell therapy, fatal or life-threatening neurologic toxicities have been reported, including seizures, cerebral edema, and encephalopathies. The neurologic events are often observed in association with or after development of CRS. Factors associated with the risk of neurotoxicity include higher tumor burden and pre-existing neurologic comorbidities (Hay et al. 2017; Wang et al. 2018). The conditioning regimen given prior to CAR-T infusion may also contribute to the risk of neurotoxicity. Thirty-two percent (32%) of patients treated with Yescarta (Yescarta SmPC) and 12% of DLBCL patients treated with Kymriah (Kymriah SmPC 2018) experienced Grade 3 or higher neurotoxicity events. The median time to onset was 5 days for Yescarta and 7 days for Kymriah. The median duration was 13 days for Yescarta and 12 days for Kymriah.
Due to these identified risks, conditions of restriction and use have been implemented for the CAR-T cell therapies with risk minimization measures in the labeling document (i.e. SmPC). Patients are required to be monitored daily for the first 10 days following infusion. Hospitalization should be considered for the first 10 days post infusion or at the first signs/symptoms of CRS and/or neurologic toxicity. Furthermore, the patients that
208 PTJA06 – Submission Dossier by Roche Registration GmbH
develop these AEs will require hospitalization, with some patients being admitted to the intensive care unit. This imposes a high burden on the healthcare system. As CRS is associated with end organ dysfunction and a worsening of underlying organ pathologies, the use of CAR-Ts may be limited in patients who have pre-existing organ dysfunction. Additional risk minimization measures are included in their EU Risk Management Plans that mandate availability of tocilizumab and site qualification in accordance with the agreed control distribution program, as well as educational programs for both healthcare providers and patients, so that prescribers are aware on how to mitigate and manage risks associated with CRS and neurological toxicities.
In comparison, the combination of Pola BR was associated with additional toxicity, but for the most part was clinically manageable and in keeping with the known toxicities of the individual agents. There have been no reports of CRS in patients treated with Pola BR. Neurologic toxicities following treated with Pola BR have been limited to peripheral neuropathy, most were low grade and non-serious, rather than any CNS toxicity. Overall, Pola BR offers an advantage over CAR-T cell therapies in that it does not carry the risk of CRS or central neurotoxicity which in turn may limit the patients that can receive CAR-T cell therapies.
Advantage in Terms of Convenience Unlike chemo-immunotherapeutic agents, CAR-T cell therapies are manufactured on a per-patient basis. The time from study entry to treatment was used as a surrogate for the rapidity at which patients could be treated. In the clinical trial setting, where treatment is limited to selected centers and patients are often referred from outside clinics, the time from diagnosis to treatment was inherently longer than it would be in standard clinical practice. In the GO29365 study, the median duration of time from diagnosis at study entry to treatment was similar for both randomized arms (0.8 months and 0.7 months for BR and Pola BR, respectively). It is expected that in a real-world setting, this timeline could be even shorter, since polatuzumab vedotin is available as an ‘off-the-shelf’ product and would not require a trip to a center where a clinical trial is held. In the real-world setting, any required testing or evaluations needed prior to patient treatment can usually be done faster.
In contrast, patients eligible for treatment with CAR-Ts must be able to wait out the prolonged manufacturing periods following apheresis. For example, in the JULIET study, at the time of the primary analysis, the median time from enrollment to infusion was 54 days (range: 30357 days) with a median time from screening to infusion of 119 days (range: 49396 days). From the 165 patients enrolled, 111 patients were infused. A schematic of the study periods for the JULIET study is shown in Figure 26.
In the ZUMA-1 study, the median time from leukapheresis to product delivery was 17 days (range: 1451 days) and the median time from leukapheresis to infusion was 23 days (range: 1572 days). A schematic of the study periods for the ZUMA-1 study is shown in Figure 27. The number of screen failures is not available in the publication of
209 PTJA06 – Submission Dossier by Roche Registration GmbH
ZUMA-1. Of the 111 enrolled and leukapheresed patients, 101 patients received the IMP. Of the 10 patients who failed to receive the product, one was due to manufacturing failure. The further 9 patients were not treated due to progressive disease, serious adverse reactions following leukapheresis or chemotherapy or due to undetectable disease.
This considerable time span from screening and enrollment to infusion is of significant importance given that these therapeutic agents are intended for the treatment of patients with an advanced disease, expected to progress rapidly.
It is also worth considering that only certain centers are qualified to administer CAR-T cell therapy. Patients may not live near such centers or be able to reach such centers to even be considered for CAR-T cell therapy either due to personal reasons or else due to disease reasons. Travel may be prohibitive if the patient is very symptomatic from disease. In such cases, rapid response to an effective treatment is needed and Pola BR can provide such an option.
Patients eligible for treatment with CAR-T cell therapies are required to wait out the prolonged manufacturing periods following apheresis. There is limited real-world evidence regarding the impact of the time needed for the manufacture of the CAR-Ts on patient outcomes.
While CAR-Ts produce clinically meaningful responses in R/R DLBCL patients, these studies have demonstrated that from the time of leukapheresis, between 9% and 30% of patients did not receive CAR-T product. While this represents a proportion of patients for whom CAR-T was not suitable, this group of patients could potentially be greater given that some patients may not reach the step of leukapheresis. While there might be diverse reasons for this group of patients not receiving CAR-Ts, the most commonly reported reasons were disease progression, non-conforming CAR-T product, and development of patient condition that prohibited CAR-T infusion (e.g. infection). These collective issues would ostensibly not be of issue with an immediately available treatment such as polatuzumab vedotin. Given the heterogeneous nature of reasons why patients are not able to receive CAR-Ts after leukapheresis, and also the heterogeneous reasons listed in the warnings and precautions of the CAR-T SmPCs that may lead to patients entirely not being considered for CAR-T cell therapy, there exists a subset of R/R DLBCL patients who are not able to benefit from CAR-Ts. Given the relative equipoise in clinical efficacy when comparing an ITT population receiving CAR-T to Pola BR in the GO29365 study, the broader availability and broader applicability of a regimen such as Pola BR is in itself a favorable quality over that of CAR-T cell therapy.
210 PTJA06 – Submission Dossier by Roche Registration GmbH
4.9 REAL-WORLD DATA EVIDENCE The data on R/R DLBCL patients treated in different settings in the UK (The Christie NHS Trust) and the US (Flatiron and the Veteran Affairs Administration) were analyzed. Collectively, they demonstrate the heterogeneity of treatment options available to treat R/R DLBCL patients.
4.9.1 Analysis of the Christie NHS Trust The Christie National Health Services Foundation Trust is the largest single-site cancer center in Europe and is a referral center for cancer treatment in the UK, treating more than 44,000 patients per year. Data was entered into an electronic form by NHS staff, as databases within the Christie are not linked and some data does not exist in the electronic record. Patients included in this study were treated at The Christie between 1 January 2011 and 31 December 2017 for R/R DLBCL. Patients were included in the cohort if they were at least 18 years old; histologically diagnosed with DLBCL, and no history of transformed lymphoma. Patients with primary mediastinal B-cell lymphoma, Burkitt-like lymphoma, central nervous system (CNS) lymphoma, or leptomeningeal infiltration were excluded.
The goal of this analysis was to describe patient characteristics and treatment patterns amongst real-world R/R DLBCL patients. Patients were identified as being either candidates or non-candidates for a transplant based on the second-line treatment regimen they received. Roche analyzed the final cohort of 90 patients delivered by The Christie, who were treated for R/R DLBCL between 2011 and 2017. Roche further excluded four patients who had radiotherapy recorded as their second-line therapy (3 patients) or had no therapy listed after first-line treatment with R-CHOP. Table 76 shows attrition for patients that were analyzed.
211 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 76 Cohort Attrition
Criteria N All R/R DLBCL patients in The Christie 90 2L+ patients 86 Transplant ineligible patients 47 BR-treated transplant ineligible patients 12 2L+: second-line and beyond; BR: bendamustine and rituximab; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma.
The analysis describes the outcomes for all transplant ineligible patients, thus including all available treatments received by these patients. In this study, patients were deemed transplant ineligible if they did not receive any of the treatment regimens listed in Table 77 in second-line or if they received any regimen listed in Table 78 but did not receive a transplant in second-line or within 15 days of the start of third-line treatment.
Table 77 and Table 78 reflect the recommendations in the 2017 NCCN Guidelines for the management of DLBCL. Gemcitabine, cisplatin, and dexamethasone, and GemOx, are listed separately because NCCN Guidelines recommended them for both transplant eligible and ineligible patients.
212 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 77 Treatment Regimens Recommended for Transplant Ineligible Patients
Treatment Regimen DHAP (dexamethasone, cytarabine, cisplatin) rituximab ESHAP (etoposide, methylprednisolone, cytarabine, cisplatin) rituximab ICE (ifosfamide, carboplatin, etoposide) rituximab MINE (mesna, ifosfamide, mitoxantrone, etoposide) rituximab
Table 78 Treatment Regimens Recommended for Transplant Eligible and Ineligible Patients
Treatment Regimen (gemcitabine, cisplatin, dexamethasone) rituximab or gemcitabine, GDP dexamethasone, carboplatin) +/- rituximab
GemOx (gemcitabine and oxaliplatin) rituximab
Roche identified 47 transplant ineligible patients in the database whose characteristics at the time of abstraction are provided in Table 79 below.
213 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 79 Patient Characteristics
Characteristic n (%) Age at diagnosis, median (IQR) 68 (57 – 74) Gender Male 28 (59.6) Female 19 (40.4) ECOG at index 0 7 (14.9) 1 16 (34.0) 2 6 (12.8) 3 3 (6.4) 4 3. (6.4) Not documented 12 (25.5) AA stage at diagnosis I 5 (10.6) II 6 (12.8) III 8 (17.0) IV 28 (59.6) Not documented Bulky disease 35 (74.5) Number of prior lines of therapy 1 20 (42.6) 2 17 (36.2) 3 8 (17.0) 4 0 (0) 5 2 (4.3) Refractory to last treatment 33 (70.2) AA: Ann Arbor; ECOG: Eastern Cooperative Oncology Group; IQR: interquartile range
Patients in this cohort were treated with the following regimens across their various lines of therapy.
214 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 80 Treatments Received by Line of Therapy
2L 3L 4L 5L 6L
(N 47) (N 27) (N 10) (N 2) (N 2) R-GDP 16 (34.0) - Bendamustine R 4 (8.5) 8 (29.6) 4 (40) 1 (50) R-GCVP 7 (14.9) - R-CHOP 4 (8.5) - Gemcitabine 5 (10.6) 2 (7.4) Rituximab 1 (2.1) - 1 (10) Rituximab inotuzumab 1 (2.1) - R-Gem-P 2 (4.2) - Rituximab, gemcitabine, 1 (2.1) - lenalidomide DA-EPOCH-R 1 (2.1) - Etoposide 1 (2.1) 2 (7.4) Lenalidomide 1 (2.1) 2 (7.4) 1 (10) Gem-P 1 (2.1) - Coltuximab ravtansine (SAR3419) 1 (2.1) - Rituximab/gemcitabine/ 1 (2.1) - methylprednisolone/lenalidomide R-DHAP - 2 (7.4) Pixantrone - 2 (7.4) Polatuzumab vedotin bendamustine + - 1 (3.7) obinituzumab Ofatumumab - 2 (7.4) R-ICE - 3 (11.1) Tazemetostat - 1 (3.7) Vistusertib - 1 (3.7) Methotrexate, cytarabine, - 1 (3.7) thiotepa, rituximab (MatRix) Durvalumab bendaustine - - 1 (10) rituximab ONO/GS-4059 idelalisib - - 1 (10) Rituximab AZD2014 1 (10) Sotrastaurin (AEB071) 1 (10) Lilotumab betalutin 1 (50) ADCT-402 2 (100) 2L: second-line; 3L: third-line; 4L: fourth-line; 5L: fifth-line; 6L: sixth-line; DA-EPOCH-R: dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin-rituximab; Gem-P: gemcitabine, cisplatin and methylprednisolone; MatRix: methotrexate, cytarabine, thiotepa, rituximab; N: number of patients; R-CHOP: rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone; R-DHAP: rituximab plus dexamethasone, cytarabine, cisplatin; R-GCVP: rituximab, gemcitabine, cyclophosphamide, vincristine, and prednisolone; R-GDP: rituximab plus gemcitabine, cisplatin, and dexamethasone; R-Gem-P: rituximab, gemcitabine, cisplatin and methylprednisolone; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide.
215 PTJA06 – Submission Dossier by Roche Registration GmbH
This study shows the diversity of regimens used to treat R/R DLBCL among transplant ineligible patients in the UK and the similarity of regimens used across the second- and third-line patients.
4.9.2 Flatiron Health The objective of this study was to describe R/R DLBCL treatment patterns and BR utilization. This study utilized Flatiron Health’s longitudinal, demographically and geographically diverse database containing electronic health record (EHR) data from over 265 cancer clinics (approximately 800 sites of care), including more than 2 million active US patients with cancer who are available for analysis. The de-identified patient-level data in the EHRs includes structured data (e.g. laboratory values and prescribed drugs) in addition to unstructured data collected via technology-enabled chart abstraction from physician’s notes and other unstructured documents (e.g. detailed biomarkers). We conducted a retrospective cohort study using Flatiron Health’s data on patients with DLBCL diagnosed from 2011 through 2018 at any of the 265 cancer clinics in the US. We included patients with at least one prior line of therapy who initiated treatment within 90 days of their diagnosis. We identified 993 patients receiving second-line treatment or beyond. The BR treatment was received by 97 (9.8%) patients in the overall cohort. The proportion of patients receiving BR ranged from 7.8% in second-line to 1.4% in fifth-line.
Table 81 BR Utilization Among Transplant Eligible and Ineligible R/R DLBCL Patients
Line of therapy
Treatment received 2L 3L 4L 5L (N 993) (N 386) (N 166) (N 70) BR, n (%) 77 (7.8) 14 (3.6) 5 (3.0) 1 (1.4) Other (most common)
regimens n (%) 22 (2.2%) 3 (0.8%) 1 (0.6%)
CODOX-M R 21 (2.1%) 6 (1.6%) 2 (2.9%)
DA-EPOCH R 62 (6.2%) 56 (14.5%) 19 (11.4%) 3 (4.3%)
GemOx R 17 (1.7%) 11 (2.8%) 2 (1.2%) 2 (2.9%)
R Monotherapy 68 (6.8%) 9 (2.3%) 2 (1.2%)
R-CHOP 227 (22.9%) 29 (7.5%) 4 (2.4%) 3 (4.3%)
R-ICE 227 (22.9%) 29 (7.5%) 4 (2.4%) 3 (4.3%)
216 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 81 BR Utilization Among Transplant Eligible and Ineligible R/R DLBCL Patients (cont.)
2L: second-line; 3L: third-line; 4L: fourth-line; 5L: fifth-line; CODOX-M R: cyclophosphamide, cytarabine, vincristine, doxorubicin, methotrexate rituximab; DA-EPOCH R: etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin rituximab; GemOx: gemcitabine, oxaliplatin rituximab; N: number of patients; R: rituximab; R-CHOP: rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone; R-ICE: rituximab, ifosfamide, carboplatin, etoposide; R/R DLBCL: relapsed/refractory diffuse large B-cell lymphoma.
The median age of patients receiving BR was 75 years (interquartile [IQR] range 68–79); 90% of patients were 60 years or older. The BR treatment was the second-most common regimen in second-line and the third-most common in third-line. The most common regimen in second-line (and second-most common in third-line) was R-ICE (typically given to prepare patients for transplant).
4.9.3 Veterans Affairs The US Department of Veterans Affairs (VA) database contains electronic medical records (EMRs) of the largest integrated healthcare system in the US, with over 1,500 sites of care, serving approximately 8.76 million veterans each year. The VA Health Administration Corporate Data Warehouse (CDW) includes all medical encounter information in the VA, which includes 152 medical centers, and nearly 1,400 community-based outpatient clinics, community-living centers, veteran centers, and domiciliaries. Patients who are followed in the VA healthcare system include military employees who served 24 continuous months or the full period for which they were called to active duty. These patients are mostly males and are eligible for VA insurance over their lifetime. The objective of this analysis was to describe the treatment patterns for second-line R/R DLBCL patients.
We identified 702 DLBCL patients who initiated second-line therapy between 2004 and 2016. A total of 206 regimens were identified. More than half of the patients treated (52) used one of 14 regimens, with the remaining 192 regimens accounting for the therapies received by the remaining patients. The 14 commonly used regimens were listed in the NCCN Guidelines. Table 82 shows the regimens received by the second-line patients.
217 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 82 Treatment Regimens Received by Second-Line Patients from the Veterans Affairs Database
Regimen n (%)
R-ICE 77 (11.0%)
R-CHOP 75 (10.7%)
R-Monotherapy 34 (4.8%)
BR 32 (4.6%)
Methotrexate 24 (3.4%)
R-ESHAP 23 (3.3%)
R-DHAP/R-EPOCH/R-GDP 18 (2.6%)
Cyclophosphamide, doxorubicin, rituximab, vinblastine, 14 (2.0%) vincristine
R-CVP 11 (1.6%)
Cyclophosphamide, etoposide, rituximab, vincristine 11 (1.6%)
R-GemOx 10 (1.4%)
Other regimens, all agent(s) included in NCCN guidelines 267 (38.0%)
Otherregimens, not all agent(s) included in NCCN guidelines 106 (15.1%) BR: bendamustine plus rituximab; NCCN: National Comprehensive Cancer Network; R-CHOP: rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone; R-CVP: rituximab plus cyclophosphamide, vincristine and prednisone; R-DHAP/R-EPOCH/R-GDP: rituximab plus dexamethasone, cytarabine, cisplatin/rituximab plus etoposide, prednisolone, oncovin, cyclophosphamide, hydroxydaunorubicin/rituximab plus gemcitabine, cisplatin, and dexamethasone; R-ESHAP: rituximab, etoposide, methylprednisolone, cytarabine, cisplatin; R-GemOx: rituximab, gemcitabine, and oxaliplatin; R-ICE: rituximab plus ifosfamide, carboplatin, and etoposide; R-monotherapy: rituximab monotherapy.
This study did not analyze transplant eligible and ineligible patients separately. However, BR and R-GemOx are two of the more commonly used regimens among transplant ineligible patients. A descriptive comparison of selected baseline characteristics for patients treated with BR and R-GemOx are shown in Table 83
218 PTJA06 – Submission Dossier by Roche Registration GmbH
Table 83 Selected Baseline Characteristics for Patients Treated with either BR or R-GemOx from the Veterans Affairs Database
R-GemOx Treated BR Treated Patients Patients (N 32) (N 10)
Age (years), mean (SD) 65 (10) 66 (6) Median (IQR) 63 (6069) 64 (6270)
Male 31 (96.9) 10 (100.0)
Stage: I 1 (3.1) 1 (10.0) II 5 (15.6) 1 (10.0) II 8 (25.0) 2 (20.0) IV 14 (43.8) 4 (40.0) Unknown/missing 4 (12.5) 2 (20.0)
Charlson comorbidity score 3.9 (2.2) 3.2 (1.9)
Refractory to last line of therapy 13 (40.6) 4 (40.0)
Number of Lymph nodes involved: 0 21 (65.6) 5 (50.0) ≥1 10 (31.3) 2 (20.0) Unknown/missing 1 (3.1) 3 (30.0)
BR: bendamustine plus rituximab; IQR: interquartile; R-GemOx: rituximab, gemcitabine, oxaliplatin; SD: stable disease.
This study highlights the diversity of regimens used to treated DLBCL patients in second-line. The BR treatment is one of the more commonly used regimens in this population. The baseline characteristics of patients who receive two of the more common regimens (BR and R-GemOx) for patients who are not candidates for transplant are similar.
5. DISCUSSION AND CONCLUSIONS FOR THE ASSESSMENT
Provide a general interpretation of the evidence base considering the benefits associated with the technology relative to those of the comparators.
Summarize the internal validity of the evidence base, taking into account 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.
219 PTJA06 – Submission Dossier by Roche Registration GmbH
Provide a brief statement of the relevance of the evidence base to the scope of the assessment.
Diffuse large B-cell lymphoma is the most common histologic subtype of NHL, accounting for 30% to 48% of all NHL cases and 60% of aggressive lymphomas (Al- Hamadani et al. 2015; Leukemia & Lymphoma Society 2019; Smith et al. 2015; The Non-Hodgkin’s Lymphoma Classification Project 1997). Approximately 60% of newly diagnosed patients may be cured with first-line treatment with R-CHOP (Sehn et al. 2015; Vitolo et al. 2017). Nevertheless, despite improvements in the survival of DLBCL patients with this regimen, approximately one-third of patients present with either a primary refractoriness to the first-line treatment used (mainly R-CHOP) or relapse after reaching a CR which remains the major cause of morbidity and mortality of this disease (Friedberg 2011). Relapsed/refractory DLBCL patients who are ineligible for SCT have limited treatment options and have a poor prognosis with a mOS of approximately 6 months (Crump et al. 2017; Czuczman et al. 2017) with survival times decreasing with additional treatment lines.
In the EU, the treatment landscape for patients with R/R DLBCL who are ineligible for SCT or who relapse after transplant, is highly fragmented. There is no universally established standard of care or regulatory-approved therapies for use upon first relapse, as no prior randomized trials have established the superiority of one regimen over another for this patient population. The most commonly prescribed regimens, used off-label, are rituximab in combination with gemcitabine and/or platinum-based therapies or rituximab in combination with bendamustine (BR). Therefore, newer approaches, that incorporate agents with novel mechanisms, with acceptable safety profiles, and are readily available for immediate use, are clearly needed. Polatuzumab vedotin (POLIVY) has the potential to fill this need.
The evidence that is currently available is considered by clinical guidelines as weak, is based on single-arm studies, and is consistent with the lack of regulatory-approved therapies for R/R DLBCL patients at first relapse. This was also evident in the SLR which was conducted to summarize the currently published evidence regarding treatments for R/R DLBCL (including patients eligible and ineligible for SCT) where only a few studies (19) were identified. The majority of trials (13) were prospective observational studies/single-arm trials leaving only 6 high-quality RCTs as the overall evidence base in R/R DLBCL, thus preventing implementation of an ITC or NMA. Only, 2 of the 6 RCTs that met the inclusion criteria had positive outcomes. Of these 2 RCTs, one was in the second- and subsequent-line setting and the other was in the third- and subsequent-line setting. Despite imaging assessment criteria becoming more stringent over time, the GO29365 study was amongst the highest with regards to CR rates in patients with R/R DLBCL using the strictest response criteria (modified Lugano Response Criteria). These findings highlight the paucity of published RCTs to establish the comparative efficacy of treatments for R/R DLBCL, and demonstrate a fragmented treatment landscape as well as lack of standard of care in this setting (Thuresson et al.
220 PTJA06 – Submission Dossier by Roche Registration GmbH
2019). As a high-quality RCT, the GO29365 study determines a robust evidence base within the context of transplant ineligible R/R DLBCL patients and this assessment.
As described above, there was insufficient evidence to compare the clinical efficacy of Pola BR to other salvage therapies (other than BR), R-GemOx, pixantrone, or CAR-T therapies for R/R DLBCL using ITCs. As there was no randomized evidence available, comparisons were made through MAICs. The results of the MAICs suggest that the OS demonstrated by R/R DLBCL patients treated with Pola BR, could be similar to that of CAR-T therapies, and better than the OS observed in patients treated with pixantrone.
Furthermore, RWD on R/R DLBCL patients treated in different settings in the UK (The Christie NHS Trust), and the U.S (Flatiron and the Veteran Affairs Administration) were analyzed. Collectively, they demonstrate the heterogeneity of treatment options available to treat R/R DLBCL patients. This along with the broadly similar effectiveness of standard salvage therapies in R/R DLBCL and the frequent use of the BR combination in this population makes BR a valid comparator to prove the benefits of polatuzumab vedotin (POLIVY) in R/R DLBCL.
In the GO29365 study, the combination of polatuzumab vedotin with BR was associated with additional toxicity as expected when an additional therapeutic agent is added to the BR combination. The safety profile of Pola BR was clinically manageable and in keeping with the known toxicities of the individual agents. A higher rate of Grade 34 cytopenias was observed with Pola BR compared with BR, but this did not result in a higher risk of infection or need for transfusion. This higher rate of AEs was likely contributed to by increased susceptibility in these heavily pre-treated patients, as well as disease progression and subsequent anti-lymphoma therapy in this high risk population of patients with R/R DLBCL. Peripheral neuropathy is a recognized toxicity associated with MMAE based ADCs, and was closely monitored during the GO29365 study. The majority of peripheral neuropathy observed was low grade and reversible, and led to few patients experiencing dose reduction or delay.
Study GO20365 demonstrated a clinically meaningful benefit with the addition of the novel ADC, polatuzumab vedotin (POLIVY), to BR compared to BR alone. In the randomized portion of the study, Pola BR showed significantly improved OS, PFS and CR rates compared to BR alone. The risk of death was reduced by 58% in patients treated with Pola BR compared to BR (stratified HR 0.42; 95% CI: [0.24, 0.75]; p 0.0023). Median OS was more than 2.5 times longer in patients receiving Pola BR compared to BR alone (12.4 months vs. 4.7 months). PFS as determined by IRC was increased in patients treated with Pola BR compared to BR (stratified HR 0.36; 95% CI: [0.21, 0.63]; p 0.0002) with median PFS being over 2-fold higher with Pola BR compared to BR (9.5 months vs. 3.7 months). Similarly PFS as determined by the INV was increased in patients treated with Pola BR compared to BR (stratified HR 0.34; 95% CI: [0.20, 0.57]; p0.0001) with median PFS being over 3-fold higher with Pola BR compared to BR (7.6 months vs. 2.0 months). The CR rate at the PRA based
221 PTJA06 – Submission Dossier by Roche Registration GmbH
on PET-CT, as determined by the IRC, was higher in the Pola BR arm (40%; 95% CI: [24.9%, 56.7%]) compared with the BR arm (17.5%; 95% CI: [7.3%, 32.8%]). The difference in CR rates between arms was significant ( 22.5% in favor of Pola BR; 95% CI: [2.6%, 40.2%]; p 0.0261; CMH chi-square). No previously reported randomized study has demonstrated a survival benefit in this patient population (CTI Biopharma News Release 2018; Czuczman et al. 2017; Dang et al. 2018; Sehn et al. 2019).
When combining data from the Phase Ib safety run-in and data from the randomized Phase II portions of the GO29365 study, an estimated 30% of patients were alive and progression-free 2 years after commencing treatment. Ten (22%) patients treated with Pola BR (3 patients in the Phase Ib safety run-in and 7 patients in the randomized Phase II) remain in complete remission at last follow-up (ongoing duration of response of 20 months) (Sehn et al. 2019). There were no obvious clinical predictors of response, as all subgroups examined appeared to benefit, including patients with refractory disease and patients who received multiple prior lines of therapy.
With longer follow-up (CCOD: 18 October 2018; median follow-up of 27.6 months in the Pola BR and BR treatment arms), the same treatment benefit was observed for Pola BR compared to BR as demonstrated in the pivotal analysis (CCOD: 30 April 2018) (Sehn et al. 2019).
Using data from the GO29365 study, projected clinical benefits of Pola BR versus BR alone include a 4-year increase in life-years (mixture cure model: 4.7 years vs. 0.8 years) and at least a 4-fold increase in the proportion of long-term survivors (i.e., patients who had not died from their disease within ~2 years (38%; 95% CI: [31, 42] vs. 9%; 95% CI: [1, 21]) (Sehn et al. 2019). Sensitivity analysis showed that censoring patients who had received SCT/CAR-T therapy had no impact on OS estimated, suggesting the effects were due to Pola BR rather than subsequent therapy (Sehn et al. 2019).
In conclusion, the combination of Pola BR is readily available (as the combination partners are easily accessible in general oncology practices) which is especially relevant for patients who are progressing rapidly and need urgent disease control, with no additional burden to health systems. Finally, in the context of a randomized study, the Pola BR combination demonstrated a compelling clinical benefit to patients over BR, as observed by significantly improved OS, PFS and CR rates, with a manageable safety profile, thus providing a new therapeutic option for transplant ineligible patients with R/R DLBCL patients, a patient population with a high unmet medical need.
222 PTJA06 – Submission Dossier by Roche Registration GmbH
6. REFERENCES
All references are available upon request
Agresti, A. (2002) Categorical Data Analysis. 2nd Edition, John Wiley & Sons, Inc., New York. 320-332.
Al-Hamadani, M., T. M. Habermann, J. R. Cerhan, et al. (2015) Non-Hodgkin lymphoma subtype distribution, geodemographic patterns, and survival in the US: A longitudinal analysis of the National Cancer Data Base from 1998 to 2011. Am J Hematol, 90, 790-5.
Amin, A. D., T. L. Peters, L. Li, et al. (2017) Diffuse large B-cell lymphoma: can genomics improve treatment options for a curable cancer? Cold Spring Harb Mol Case Stud, 3, a001719.
Arcari, A., A. Chiappella, M. Spina, et al. (2016) Safety and efficacy of rituximab plus bendamustine in relapsed or refractory diffuse large B-cell lymphoma patients: an Italian retrospective multicenter study. Leuk Lymphoma, 57, 1823-30.
Aribi, M., N. Mesli, N. Remla, et al. (2010) Gemcitabine and treatment of diffuse large B- cell lymphoma in relapsed or refractory elderly patients: a prospective randomized trial in Algeria. J Cancer Res Ther, 6, 41-6.
Armitage, J. O., R. D. Gascoyne, M. A. Lunning & F. Cavalli (2017) Non-Hodgkin lymphoma. Lancet, 390, 298-310.
Armitage, J. O. & D. D. Weisenburger (1998) New approach to classifying non-Hodgkin's lymphomas: clinical features of the major histologic subtypes. Non-Hodgkin's Lymphoma Classification Project. J Clin Oncol, 16, 2780-95.
Aviles, A., N. Neri, J. Huerta-Guzman & M. de Jesus Nambo (2010) ESHAP versus rituximab-ESHAP in frail patients with refractory diffuse large B-cell lymphoma. Clin Lymphoma Myeloma Leuk, 10, 125-8.
Bray, F., J. Ferlay, I. Soerjomataram, et al. (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin, 68, 394-424.
Buske, C., M. Hutchings, M. Ladetto, et al. (2018) ESMO Consensus Conference on malignant lymphoma: general perspectives and recommendations for the clinical management of the elderly patient with malignant lymphoma. Ann Oncol, 29, 544-562.
Carbone, P. P., H. S. Kaplan, K. Musshoff, et al. (1971) Report of the Committee on Hodgkin's Disease Staging Classification. Cancer Res, 31, 1860-1.
Casasnovas, R. O., M. Meignan, A. Berriolo-Riedinger, et al. (2011) SUVmax reduction improves early prognosis value of interim positron emission tomography scans in diffuse large B-cell lymphoma. Blood, 118, 37-43.
223 PTJA06 – Submission Dossier by Roche Registration GmbH
Cerhan, J. R., A. Kricker, O. Paltiel, et al. (2014) Medical history, lifestyle, family history, and occupational risk factors for diffuse large B-cell lymphoma: the InterLymph Non-Hodgkin Lymphoma Subtypes Project. J Natl Cancer Inst Monogr, 2014, 15- 25.
Cheson, B. D., S. Ansell, L. Schwartz, et al. (2016) Refinement of the Lugano Classification lymphoma response criteria in the era of immunomodulatory therapy. Blood, 128, 2489-2496.
Cheson, B. D., R. I. Fisher, S. F. Barrington, et al. (2014) Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol, 32, 3059-68.
Cheson, B. D., S. J. Horning, B. Coiffier, et al. (1999) Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol, 17, 1244.
Cheson, B. D., B. Pfistner, M. E. Juweid, et al. (2007) Revised response criteria for malignant lymphoma. J Clin Oncol, 25, 579-86.
Cheson, B. D., C. M. Wendtner, A. Pieper, et al. (2010) Optimal use of bendamustine in chronic lymphocytic leukemia, non-Hodgkin lymphomas, and multiple myeloma: treatment recommendations from an international consensus panel. Clin Lymphoma Myeloma Leuk, 10, 21-7.
Cleeland, C. S., T. R. Mendoza, X. S. Wang, et al. (2000) Assessing symptom distress in cancer patients: the M.D. Anderson Symptom Inventory. Cancer, 89, 1634-46.
Coiffier, B., E. Lepage, J. Briere, et al. (2002) CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med, 346, 235-42.
Coiffier, B., C. Thieblemont, E. Van Den Neste, et al. (2010) Long-term outcome of patients in the LNH-98.5 trial, the first randomized study comparing rituximab- CHOP to standard CHOP chemotherapy in DLBCL patients: a study by the Groupe d'Etudes des Lymphomes de l'Adulte. Blood, 116, 2040-5.
Crump, M., S. S. Neelapu, U. Farooq, et al. (2017) Outcomes in refractory diffuse large B-cell lymphoma: results from the international SCHOLAR-1 study. Blood, 130, 1800-1808.
CTI Biopharma News Release (2018) Results of Phase III (PIX306) Trial Evaluating Progression-Free Survival of PIXUVRI® (pixantrone) Combined with Rituximab in Patients with Aggressive B-cell Non-Hodgkin Lymphoma http://investors.ctibiopharma.com/phoenix.zhtml?c=92775&p=irol- newsArticle&ID=2357334,
Czuczman, M. S., M. Trneny, A. Davies, et al. (2017) A Phase 2/3 Multicenter, Randomized, Open-Label Study to Compare the Efficacy and Safety of Lenalidomide Versus Investigator's Choice in Patients with Relapsed or Refractory Diffuse Large B-Cell Lymphoma. Clin Cancer Res, 23, 4127-4137.
224 PTJA06 – Submission Dossier by Roche Registration GmbH
Dang, N. H., M. Ogura, S. Castaigne, et al. (2014) Randomized, phase 3 trial of inotuzumab ozogamicin plus rituximab (R-InO) versus chemotherapy for relapsed/refractory aggressive B-cell non-Hodgkin lymphoma (B-NHL). Journal of Clinical Oncology, 32, 8529-8529.
Dang, N. H., M. Ogura, S. Castaigne, et al. (2018) Randomized, phase 3 trial of inotuzumab ozogamicin plus rituximab versus chemotherapy plus rituximab for relapsed/refractory aggressive B-cell non-Hodgkin lymphoma. Br J Haematol, 182, 583-586.
Dornan, D., F. Bennett, Y. Chen, et al. (2009) Therapeutic potential of an anti-CD79b antibody-drug conjugate, anti-CD79b-vc-MMAE, for the treatment of non-Hodgkin lymphoma. Blood, 114, 2721-9.
Ducry, L. & B. Stump (2010) Antibody-drug conjugates: linking cytotoxic payloads to monoclonal antibodies. Bioconjug Chem, 21, 5-13.
El Gnaoui, T., J. Dupuis, K. Belhadj, et al. (2007) Rituximab, gemcitabine and oxaliplatin: an effective salvage regimen for patients with relapsed or refractory B-cell lymphoma not candidates for high-dose therapy. Ann Oncol, 18, 1363-8.
European Cancer Information System Estimates of cancer incidence and mortality in 2018, for all countries https://ecis.jrc.ec.europa.eu/explorer.php?$0-0$1-All$2- All$4-1,2$3-50$6-0,14$5-2008,2008$7-7$CEstByCountry$X0_8-3$X0_19- AE28E$X0_20-No$CEstRelative$X1_8-3$X1_9-AE28$X1_19- AE28E$CEstByCountryTable$X2_19-AE28E.
Farooq, U., M. J. Maurer, C. A. Thompson, et al. (2017) Clinical heterogeneity of diffuse large B cell lymphoma following failure of front-line immunochemotherapy. Br J Haematol, 179, 50-60.
Ferlay, J., M. Colombet, I. Soerjomataram, et al. (2018) Cancer incidence and mortality patterns in Europe: Estimates for 40 countries and 25 major cancers in 2018. Eur J Cancer, 103, 356-387.
Friedberg, J. W. (2011) Relapsed/refractory diffuse large B-cell lymphoma. Hematology Am Soc Hematol Educ Program, 2011, 498-505.
Friedberg, J. W. & R. I. Fisher (2008) Diffuse large B-cell lymphoma. Hematol Oncol Clin North Am, 22, 941-52, ix.
G-BA (2013) des Gemeinsamen Bundesausschusses über eine Änderung der Arzneimittel-Richtlinie (AM-RL): Anlage XII - Beschlüsse über die Nutzenbewertung von Arzneimitteln miteuen Wirkstoffen nach § 35a SGB V – Pixantron https://www.g-ba.de/downloads/39-261-1712/2013-05-16_AM-RL- XII_Pixantron_BAnz.pdf
Galaznik, A., R. Huelin, M. Stokes, et al. (2018) Systematic review of therapy used in relapsed or refractory diffuse large B-cell lymphoma and follicular lymphoma. Future Sci OA, 4, Fso322.
225 PTJA06 – Submission Dossier by Roche Registration GmbH
Gisselbrecht, C., B. Glass, N. Mounier, et al. (2010) Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol, 28, 4184-90.
Gisselbrecht, C., N. Schmitz, N. Mounier, et al. (2012) Rituximab maintenance therapy after autologous stem-cell transplantation in patients with relapsed CD20(+) diffuse large B-cell lymphoma: final analysis of the collaborative trial in relapsed aggressive lymphoma. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 30, 4462-4469.
Gisselbrecht, C. & E. Van Den Neste (2018) How I manage patients with relapsed/refractory diffuse large B cell lymphoma. Br J Haematol, 182, 633-643.
Gouveia, G. R., S. A. Siqueira & J. Pereira (2012) Pathophysiology and molecular aspects of diffuse large B-cell lymphoma. Rev Bras Hematol Hemoter, 34, 447- 51.
Harris, N. L., E. S. Jaffe, H. Stein, et al. (1994) A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood, 84, 1361-92.
HAS (2016) Brief Summary of the Transparency Committee Opinion - Pixuvri https://www.has-sante.fr/portail/upload/docs/application/pdf/2017- 05/dir24/pixuvri_summary_ct15478.pdf
Hedström, G., O. Hagberg, M. Jerkeman, et al. (2015) The impact of age on survival of diffuse large B-cell lymphoma - a population-based study. Acta Oncol, 54:91623.
Herrera, A. F., M. J. Matasar, S. Assouline, et al. (2016) Polatuzumab Vedotin Combined with Bendamustine (B) and Rituximab (R) or Obinutuzumab (G) in Patients with Relapsed or Refractory (R/R) Follicular Lymphoma (FL) or Diffuse Large B-Cell Lymphoma (DLBCL): Preliminary Results of a Phase Ib/II Dose- Escalation Study. Blood, 128, 4194-4194.
Hong, J. Y., D. H. Yoon, C. Suh, et al. (2018) Bendamustine plus rituximab for relapsed or refractory diffuse large B cell lymphoma: a multicenter retrospective analysis. Ann Hematol, 97, 1437-1443.
Hunter, B. D., M. Herr, P. J. Meacham, et al. (2017) Late Relapses After High-dose Chemotherapy and Autologous Stem Cell Transplantation in Patients With Diffuse Large B-cell Lymphoma in the Rituximab Era. Clin Lymphoma Myeloma Leuk, 17, 145-151.
IARC (2018) International Agency for Research on Cancer. World Health Organization. Globocan 2018, Cancer Today http://gco.iarc.fr/today/home.
Ioannidis, J. P., S. J. Evans, P. C. Gotzsche, et al. (2004) Better reporting of harms in randomized trials: an extension of the CONSORT statement. Ann Intern Med, 141, 781-8.
226 PTJA06 – Submission Dossier by Roche Registration GmbH
Ionescu-Ittu, R., A. Shang, N. Vander Velde, et al. (2018) Comparable Overall Survival with Rituximab-Bendamustine (R-Benda) and Rituximab-Gemcitabine-Oxaliplatin (R-GemOx) When Used As Second-Line (2L) Treatment for Diffuse Large B-Cell Lymphoma (DLBCL): A Real-World Study Using US Veterans Health Administration Data. Blood, 132, 1711-1711.
Kaplan, E. L. & P. Meier (1958) Nonparametric Estimation from Incomplete Observations. Journal of the American Statistical Association, 53, 457-481.
Kite G-BA Addendum (2018a) Module 4A https://www.g-ba.de/downloads/92-975- 2747/Addendum%20zu%20Modul%204A.pdf
Kite G-BA Assessment (2018b) Module 4A https://www.g-ba.de/downloads/92-975- 2746/2018-10-31_Modul4A_Axicabtagen-Ciloleucel.pdf
Kuhnl, A., D. Cunningham, N. Counsell, et al. (2017) Outcome of elderly patients with diffuse large B-cell lymphoma treated with R-CHOP: results from the UK NCRI R- CHOP14v21 trial with combined analysis of molecular characteristics with the DSHNHL RICOVER-60 trial. Ann Oncol, 28, 1540-1546.
Kuruvilla, J., D. A. MacDonald, C. T. Kouroukis, et al. (2015) Salvage chemotherapy and autologous stem cell transplantation for transformed indolent lymphoma: a subset analysis of NCIC CTG LY12. Blood, 126, 733-8.
Kymriah EPAR (2018) Assessment Report https://www.ema.europa.eu/en/documents/assessment-report/kymriah-epar- public-assessment-report_en.pdf
Kymriah SmPC Summary of Product Characteristics https://www.ema.europa.eu/en/documents/product-information/kymriah-epar- product-information_en.pdf
Lakshmaiah, K. C., K. S. Rachan Shetty, V. Sathyanarayanan, et al. (2015) Lenalidomide in relapsed refractory non-Hodgkin's lymphoma: An Indian perspective. J Cancer Res Ther, 11, 857-61.
Le Guyader-Peyrou, S., S. Orazio, O. Dejardin, et al. (2017) Factors related to the relative survival of patients with diffuse large B-cell lymphoma in a population- based study in France: does socio-economic status have a role? Haematologica, 102, 584-592.
Lenz, G., S. C. Rutherford, A. Davies, et al. (2018) The Burden of Relapsed/Refractory (R/R) Diffuse Large B-Cell Lymphoma (DLBCL): A Systematic Literature Review (SLR). Presented at the ASH Annual Meeting, Dec. 1, 2018 https://ash.confex.com/ash/2018/webprogram/Paper112878.html
Leukemia & Lymphoma Society (2019) News and Updates https://www.lls.org/.
Levact - Article 29 referral (2011) https://www.ema.europa.eu/en/documents/referral/levact-article-29-referral- annex-i-ii-iii-iv_en.pdf
227 PTJA06 – Submission Dossier by Roche Registration GmbH
Lister, T. A., D. Crowther, S. B. Sutcliffe, et al. (1989) Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin's disease: Cotswolds meeting. J Clin Oncol, 7, 1630-6.
Lopez, A., A. Gutierrez, A. Palacios, et al. (2008) GEMOX-R regimen is a highly effective salvage regimen in patients with refractory/relapsing diffuse large-cell lymphoma: a phase II study. Eur J Haematol, 80, 127-32.
Lydersen, S. (2015) Statistical review: frequently given comments. Ann Rheum Dis, 74, 323-5.
Matasar, M., A. F. Herrera, M. Kamdar, et al. (2017) Polatuzumab Vedotin Plus Bendamustine and Rituximab or Obinutuzumab in Relapsed/Refractory FL or DLBCL: Updated Results of a Phase 1b/2 Study. Hematological Oncology, 35, 271-272.
Maurer, M. J., H. Ghesquieres, J. P. Jais, et al. (2014) Event-free survival at 24 months is a robust end point for disease-related outcome in diffuse large B-cell lymphoma treated with immunochemotherapy. J Clin Oncol, 32, 1066-73.
Moccia, A. A., F. Hitz, P. Hoskins, et al. (2017) Gemcitabine, dexamethasone, and cisplatin (GDP) is an effective and well-tolerated salvage therapy for relapsed/refractory diffuse large B-cell lymphoma and Hodgkin lymphoma. Leuk Lymphoma, 58, 324-332.
Moller, M. B., N. T. Pedersen & B. E. Christensen (2004) Diffuse large B-cell lymphoma: clinical implications of extranodal versus nodal presentation--a population-based study of 1575 cases. Br J Haematol, 124, 151-9.
Mounier, N., T. El Gnaoui, H. Tilly, et al. (2013) Rituximab plus gemcitabine and oxaliplatin in patients with refractory/relapsed diffuse large B-cell lymphoma who are not candidates for high-dose therapy. A phase II Lymphoma Study Association trial. Haematologica, 98, 1726-31.
NCCN (2018) Clinical Practice Guidelines in Oncology.
NCCN (2019) Clinical Practice Guidelines in Oncology: B-Cell Lymphomas https://www.nccn.org/professionals/physician_gls/pdf/b-cell.pdf
NCT00118209 Rituximab and Combination Chemotherapy in Treating Patients With Diffuse Large B-Cell Non-Hodgkin's Lymphoma https://clinicaltrials.gov/ct2/show/NCT00118209
NCT01122472 Study of Lenalidomide Maintenance Versus Placebo in Responding Elderly Patients With DLBCL and Treated With R-CHOP (REMARC) https://clinicaltrials.gov/ct2/show/NCT01122472
NCT01287741 A Study of Obinutuzumab in Combination With CHOP Chemotherapy Versus Rituximab With CHOP in Participants With CD20-Positive Diffuse Large B-Cell Lymphoma (GOYA) https://clinicaltrials.gov/ct2/show/NCT01287741
228 PTJA06 – Submission Dossier by Roche Registration GmbH
NCT01324596 A Randomised Evaluation of Molecular Guided Therapy for Diffuse Large B-cell Lymphoma With Bortezomib (REMoDL-B) https://clinicaltrials.gov/ct2/show/NCT01324596
NCT01855750 A Study of the Bruton's Tyrosine Kinase Inhibitor, PCI-32765 (Ibrutinib), in Combination With Rituximab, Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone in Patients With Newly Diagnosed Non-Germinal Center B-Cell Subtype of Diffuse Large B-Cell Lymphoma https://clinicaltrials.gov/ct2/show/NCT01855750
NCT02257567 A Study of Polatuzumab Vedotin (DCDS4501A) in Combination With Rituximab or Obinutuzumab Plus Bendamustine in Participants With Relapsed or Refractory Follicular or Diffuse Large B-Cell Lymphoma https://clinicaltrials.gov/ct2/show/NCT02257567
NCT02285062 Efficacy and Safety Study of Lenalidomide Plus R-CHOP Chemotherapy Versus Placebo Plus R-CHOP Chemotherapy in Untreated ABC Type Diffuse Large B-cell Lymphoma (ROBUST) https://clinicaltrials.gov/ct2/show/NCT02285062
NCT03274492 A Study Comparing the Efficacy and Safety of Polatuzumab Vedotin With Rituximab-Cyclophosphamide, Doxorubicin, and Prednisone (R-CHP) Versus Rituximab-Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone (R- CHOP) in Participants With Diffuse Large B-Cell Lymphoma (POLARIX) https://clinicaltrials.gov/ct2/show/NCT03274492?term=POLARIX&rank=1
Neelapu, S. S., F. L. Locke, N. L. Bartlett, et al. (2017a) Axicabtagene Ciloleucel CAR T- Cell Therapy in Refractory Large B-Cell Lymphoma. N Engl J Med, 377, 2531- 2544.
NICE (2018a) Tisagenlecleucel for treating relapsed or refractory diffuse large B-cell lymphoma after 2 or more systemic therapies. Appraisal consultation document https://www.nice.org.uk/guidance/gid-ta10269/documents/appraisal-consultation- document
NICE (2018b) Tisagenlecleucel for treating relapsed or refractory diffuse large B-cell lymphoma after 2 or more systemic therapies [ID1166] https://www.nice.org.uk/guidance/ta567/evidence/appraisal-consultation- committee-papers-pdf-6718510621
Nogai, H., B. Dorken & G. Lenz (2011) Pathogenesis of non-Hodgkin's lymphoma. J Clin Oncol, 29, 1803-11.
Ohmachi, K., N. Niitsu, T. Uchida, et al. (2013) Multicenter phase II study of bendamustine plus rituximab in patients with relapsed or refractory diffuse large B-cell lymphoma. J Clin Oncol, 31, 2103-9.
Paillassa, J., L. Vercellino, R. Di Blasi, et al. (2019) Impact of Bridging Chemotherapy on Clinical Outcomes on CD19 CAR T Therapy in Relapse/Refractory Diffuse Large B-Cell Lymphoma in Real World Experience. ASH 2019: Abstract 2886.
229 PTJA06 – Submission Dossier by Roche Registration GmbH
Papageorgiou, E. S., P. Tsirigotis, M. Dimopoulos, et al. (2005) Combination chemotherapy with gemcitabine and vinorelbine in the treatment of relapsed or refractory diffuse large B-cell lymphoma: a phase-II trial by the Hellenic Cooperative Oncology Group. Eur J Haematol, 75, 124-9.
Pettengell, R., B. Coiffier, G. Narayanan, et al. (2012) Pixantrone dimaleate versus other chemotherapeutic agents as a single-agent salvage treatment in patients with relapsed or refractory aggressive non-Hodgkin lymphoma: a phase 3, multicentre, open-label, randomised trial. Lancet Oncol, 13, 696-706.
Pfeifer, M., B. Zheng, T. Erdmann, et al. (2015) Anti-CD22 and anti-CD79B antibody drug conjugates are active in different molecular diffuse large B-cell lymphoma subtypes. Leukemia, 29, 1578-86.
Pfreundschuh, M., L. Trumper, A. Osterborg, et al. (2006) CHOP-like chemotherapy plus rituximab versus CHOP-like chemotherapy alone in young patients with good- prognosis diffuse large-B-cell lymphoma: a randomised controlled trial by the MabThera International Trial (MInT) Group. Lancet Oncol, 7, 379-91.
Phillippo, D., T. Ades, S. Dias, et al. 2016. NICE DSU Technical Support Document 18. Decision Support Unit, ScHARR, University of Sheffield: NICE Decision Support Unit.
Pixuvri Press Release (2019) PIXUVRI® (pixantrone) Receives Positive CHMP Opinion to Convert Conditional Approval into Standard Marketing Authorization in Patients with Aggressive non-Hodgkin B-cell Lymphoma https://www.businesswire.com/news/home/20190409005183/en/PIXUVRI%C2% AE-pixantrone-Receives-Positive-CHMP-Opinion-Convert
Pixuvri SmPC Summary of Product Characteristics.
Pocock, S. J., T. C. Clayton & D. G. Altman (2002) Survival plots of time-to-event outcomes in clinical trials: good practice and pitfalls. Lancet, 359, 1686-9.
POLIVY® Proposed/Updated SmPC after CHMP LOQ (2019) Summary of Product Characteristics
POLIVY™ US PI (2019).
Qunaj, L., J. J. Castillo & A. J. Olszewski (2018) Survival of patients with CD20-negative variants of large B-cell lymphoma: an analysis of the National Cancer Data Base. Leuk Lymphoma, 59, 1375-1383.
Raut, L. S. & P. P. Chakrabarti (2014) Management of relapsed-refractory diffuse large B cell lymphoma. South Asian J Cancer, 3, 66-70.
Rituximab SmPC (2019) Summary of Product Characteristics https://www.ema.europa.eu/en/documents/product-information/mabthera-epar- product-information_en.pdf
Roche Data on File (2017) Polatuzumab Vedotin R/R DLBCL TPP Testing: EU3 (Germany, France, UK) Primary Market Research.
230 PTJA06 – Submission Dossier by Roche Registration GmbH
Roche Data on File (2019) Clinical SLR and Feasibility Report.
Salles, G. A., W. Jurczak, D. J. Andorsky, et al. (2018) Results of a Phase 3 Randomised Multicenter Study Comparing Pixantrone + Rituximab with Gemcitabine + Rituximab in Patients with Relapsed Aggressive B-Cell Non- Hodgkin Lymphoma Not Eligible for Stem Cell Transplantation. Blood, 132, 4189- 4189.
Schuster, S. J., M. R. Bishop, C. Tam, et al. (2018) Sustained Disease Control for Adult Patients with Relapsed or Refractory Diffuse Large B-Cell Lymphoma: An Updated Analysis of Juliet, a Global Pivotal Phase 2 Trial of Tisagenlecleucel. Presented at the ASH Annual Meeting Dec .1, 2018. Am. Soc. Hematology.
Schuster, S. J., M. R. Bishop, C. S. Tam, et al. (2019) Tisagenlecleucel in Adult Relapsed or Refractory Diffuse Large B-Cell Lymphoma. New England Journal of Medicine, 380, 45-56.
Schuster, S. J., J. Svoboda, E. A. Chong, et al. (2017) Chimeric Antigen Receptor T Cells in Refractory B-Cell Lymphomas. N Engl J Med, 377, 2545-2554.
Sehn, L. H., C. Flowers, A. McMillan, et al. (2019) Estimation of Long-term Survival with Polatuzumab Vedotin Plus Bendamustine and Rituximab for Patients with Relapsed/Refractory Diffuse Large B-Cell Lymphoma (R/R DLBCL) https://onlinelibrary.wiley.com/doi/epdf/10.1002/hon.66_2630
Sehn, L. H. & R. D. Gascoyne (2015) Diffuse large B-cell lymphoma: optimizing outcome in the context of clinical and biologic heterogeneity. Blood, 125, 22-32.
Sehn, L. H., A. F. Herrera, M. J. Matasar, et al. (2018) Polatuzumab Vedotin (Pola) Plus Bendamustine (B) with Rituximab (R) or Obinutuzumab (G) in Relapsed/Refractory (R/R) Diffuse Large B-Cell Lymphoma (DLBCL): Updated Results of a Phase (Ph) Ib/II Study. In ASH.
Sehn, L. H., A. F. Herrera, M. J. Matasar, et al. (2018b) Polatuzumab Vedotin (Pola) Plus Bendamustine (B) with Rituximab (R) or Obinutuzumab (G) in Relapsed/Refractory (R/R) Diffuse Large B-Cell Lymphoma (DLBCL): Updated Results of a Phase (Ph) Ib/II Study. Abstract 1683 presented at ASH Annual Meeting, Dec. 1, 2018.
Sehn, L. H., A. F. Herrera, M. J. Matasar, et al. (2017a) Addition of Polatuzumab Vedotin to Bendamustine and Rituximab (BR) Improves Outcomes in Transplant- Ineligible Patients with Relapsed/Refractory (R/R) Diffuse Large B-Cell Lymphoma (DLBCL) Versus BR Alone: Results from a Randomized Phase 2 Study. Blood, 130, 2821-2821.
Sehn, L. H., M. Kamdar & et al. (2017b) Adding polatuzumab vedotin (POLA) to bendamustine and rituximab (BR) treatment improves survival in patients with relapsed.refractory DLBCL: results of a phase 2 clinical trial. 23rd Congress of the European Hematology Association Abstract S802 Blood.
Sehn, L. H., M. Kamdar, A. F. Herrera, et al. (2018a) Randomized phase 2 trial of polatuzumab vedotin (pola) with bendamustine and rituximab (BR) in
231 PTJA06 – Submission Dossier by Roche Registration GmbH
relapsed/refractory (r/r) FL and DLBCL. Journal of Clinical Oncology, 36, 7507- 7507.
Signorovitch, J. E., V. Sikirica, M. Haim Erder, et al. (2012) Matching-Adjusted Indirect Comparisons: A New Tool for Timely Comparative Effectiveness Research. Value Health, 15(6):940-7.
Smith, A., S. Crouch, S. Lax, et al. (2015) Lymphoma incidence, survival and prevalence 2004-2014: sub-type analyses from the UK's Haematological Malignancy Research Network. Br J Cancer, 112, 1575-84.
Smith, A., D. Howell, R. Patmore, et al. (2011) Incidence of haematological malignancy by sub-type: a report from the Haematological Malignancy Research Network. Br J Cancer, 105, 1684-92.
Swerdlow, S. H., E. Campo, S. A. Pileri, et al. (2016) The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood, 127, 2375-90.
Tay, J., A. Daly, K. Jamani, et al. (2019) Patient eligibility for hematopoietic stem cell transplantation: a review of patient-associated variables. Bone Marrow Transplantation, 54, 368-382.
The Non-Hodgkin’s Lymphoma Classification Project (1997) A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin's lymphoma. The Non-Hodgkin's Lymphoma Classification Project. Blood, 89, 3909-18.
Thuresson, P. O., N. Vander Velde, R. Harvey, et al. 2019. A Systematic Review of the Clinical Efficacy of Treatments in Relapsed or Refractory Diffuse Large B-Cell Lymphoma (R/R DLBCL): PF310. Presented at the 24trh EHA Congress, 13-16 June 2019, Amsterdam, The Netherlands.
Tilly, H., M. Gomes da Silva, U. Vitolo, et al. (2015) Diffuse large B-cell lymphoma (DLBCL): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow- up. Ann Oncol, 26 Suppl 5, v116-25.
Vacirca, J. L., P. I. Acs, I. A. Tabbara, et al. (2014) Bendamustine combined with rituximab for patients with relapsed or refractory diffuse large B cell lymphoma. Ann Hematol, 93, 403-9. van Baarlen, J., H. J. Schuurman & J. A. van Unnik (1988) Multilobated non-Hodgkin's lymphoma. A clinicopathologic entity. Cancer, 61, 1371-6. van de Schans, S. A. M., L. N. van Steenbergen, J. W. W. Coebergh, et al. (2014) Actual prognosis during follow-up of survivors of B-cell non-Hodgkin lymphoma in the Netherlands. Haematologica, 99, 339-345.
Van Den Neste, E., N. Schmitz, N. Mounier, et al. (2016) Outcome of patients with relapsed diffuse large B-cell lymphoma who fail second-line salvage regimens in the International CORAL study. Bone Marrow Transplant, 51: 517.
Van Den Neste, E., N. Schmitz, N. Mounier, et al. (2017) Outcomes of diffuse large B- cell lymphoma patients relapsing after autologous stem cell transplantation: an
232 PTJA06 – Submission Dossier by Roche Registration GmbH
analysis of patients included in the CORAL study. Bone Marrow Transplant, 52: 21621.
Vardhana, S. A., C. S. Sauter, M. J. Matasar, et al. (2017) Outcomes of primary refractory diffuse large B-cell lymphoma (DLBCL) treated with salvage chemotherapy and intention to transplant in the rituximab era. Br J Haematol, 176:59199.
Vitolo, U., M. Trneny, D. Belada, et al. (2017) Obinutuzumab or Rituximab Plus Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone in Previously Untreated Diffuse Large B-Cell Lymphoma. J Clin Oncol, 35, 3529-3537.
Wang, H. I., A. Smith, E. Aas, et al. (2017) Treatment cost and life expectancy of diffuse large B-cell lymphoma (DLBCL): a discrete event simulation model on a UK population-based observational cohort. Eur J Health Econ, 18, 255-267.
Wasterlid, T., M. Mohammadi, K. E. Smedby, et al. (2019) Impact of comorbidity on disease characteristics, treatment intent and outcome in diffuse large B-cell lymphoma: a Swedish lymphoma register study. J Intern Med, 285, 455-468.
Wiernik, P. H., I. S. Lossos, J. M. Tuscano, et al. (2008) Lenalidomide monotherapy in relapsed or refractory aggressive non-Hodgkin's lymphoma. J Clin Oncol, 26, 4952-7.
Witzig, T. E., J. M. Vose, P. L. Zinzani, et al. (2011) An international phase II trial of single-agent lenalidomide for relapsed or refractory aggressive B-cell non- Hodgkin's lymphoma. Ann Oncol, 22, 1622-7.
Yescarta EPAR (2018) Assessment Report https://www.ema.europa.eu/en/documents/assessment-report/yescarta-epar- public-assessment-report_en.pdf
Yescarta SmPC Summary of Product Characteristics https://www.ema.europa.eu/en/documents/product-information/yescarta-epar- product-information_en.pdf
Zhu, Y., J. Lu, X. Wei, et al. (2013) The predictive value of interim and final [18F] fluorodeoxyglucose positron emission tomography after rituximab-chemotherapy in the treatment of non-Hodgkin's lymphoma: a meta-analysis. Biomed Res Int, 2013, 275805.
Ziepert, M., D. Hasenclever, E. Kuhnt, et al. (2010) Standard International Prognostic Index Remains a Valid Predictor of Outcome for Patients With Aggressive CD20+ B-Cell Lymphoma in the Rituximab Era. Journal of Clinical Oncology, 28, 2373- 2380.
Zinzani, P. L., C. Pellegrini, L. Gandolfi, et al. (2011) Combination of lenalidomide and rituximab in elderly patients with relapsed or refractory diffuse large B-cell lymphoma: a phase 2 trial. Clin Lymphoma Myeloma Leuk, 11, 462-6.
233