CENTER FOR DRUG EVALUATION AND RESEARCH

APPLICATION NUMBER:

213736Orig1s000

MULTI-DISCIPLINE REVIEW Summary Review Office Director Cross Discipline Team Leader Review Clinical Review Non-Clinical Review Statistical Review Clinical Pharmacology Review NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

NDA/BLA Multi-Disciplinary Review and Evaluation Application Type NME Application Number(s) NDA 213736 Priority or Standard Priority Submit Date(s) September 30, 2019 Received Date(s) September 30, 2019 PDUFA Goal Date May 30, 2020 Division/Office Division of Oncology 3 Review Completion Date April 17, 2020 Established/Proper Name pemigatinib (Proposed) Trade Name Pemazyre Pharmacologic Class Kinase Inhibitor Code name INCB054828 Applicant Incyte Corporation Dosage form tablet Applicant proposed Dosing 13.5 mg orally once daily for 14 days followed by 7 days off Regimen therapy. Continue treatment until disease progression or unacceptable toxicity occurs. Applicant Proposed Treatment of adults with previously treated, locally advanced Indication(s)/Population(s) or metastatic cholangiocarcinoma with a fibroblast growth factor receptor 2 (FGFR2) fusion or rearrangement as detected by an FDA-approved test. This indication is approved under accelerated approval based on overall response rate and duration of response. Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s). Applicant Proposed SNOMED Cholangiocarcinoma of biliary tract (disorder) CT Indication Disease Term Recommendation on Accelerated Approval Regulatory Action Recommended Treatment of adults with previously treated, unresectable Indication(s)/Population(s) locally advanced or metastatic cholangiocarcinoma with a fibroblast growth factor receptor 2 (FGFR2) fusion or other rearrangement as detected by an FDA-approved test. This indication is approved under accelerated approval based on overall response rate and duration of response. Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s).

1 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Recommended SNOMED CT 312104005/Cholangiocarcinoma of biliary tract (disorder) Indication Disease Term Recommended Dosing 13.5 mg orally once daily for 14 consecutive days followed by Regimen 7 days off therapy, in 21-day cycles

2 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table of Contents Table of Tables ...... 6 Table of Figures ...... 8 Reviewers of Multi-Disciplinary Review and Evaluation ...... 9 Glossary ...... 11 1 Executive Summary ...... 13 Product Introduction ...... 13 Conclusions on the Substantial Evidence of Effectiveness ...... 14 Benefit-Risk Assessment ...... 16 Patient Experience Data ...... 24 2 Therapeutic Context ...... 25 Analysis of Condition ...... 25 Analysis of Current Treatment Options ...... 25 3 Regulatory Background ...... 27 U.S. Regulatory Actions and Marketing History ...... 27 Summary of Presubmission/Submission Regulatory Activity ...... 27 4 Significant Issues from Other Review Disciplines Pertinent to Clinical Conclusions on Efficacy and Safety ...... 30 Office of Scientific Investigations (OSI) ...... 30 Product Quality ...... 30 Clinical Microbiology ...... 31 Devices and Companion Diagnostic Issues ...... 31 5 Nonclinical Pharmacology/Toxicology...... 32 Executive Summary ...... 32 Referenced NDAs, BLAs, DMFs ...... 35 Pharmacology ...... 36 ADME/PK ...... 42 Toxicology ...... 45 General Toxicology ...... 45 Genetic Toxicology ...... 58 Carcinogenicity ...... 59 Reproductive and Developmental Toxicology ...... 59 Other Toxicology Studies ...... 62 6 Clinical Pharmacology ...... 64

3 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Executive Summary ...... 64 Summary of Clinical Pharmacology Assessment ...... 65 Pharmacology and Clinical Pharmacokinetics ...... 65 General Dosing and Therapeutic Individualization ...... 66 Comprehensive Clinical Pharmacology Review ...... 66 General Pharmacology and Pharmacokinetic Characteristics ...... 66 Insert Clinical Pharmacology Questions ...... 68 7 Sources of Clinical Data and Review Strategy ...... 77 Table of Clinical Studies ...... 77 Review Strategy ...... 80 8 Statistical and Clinical and Evaluation ...... 81 Review of Relevant Individual Trials Used to Support Efficacy ...... 81 INCB 54828-202 (FIGHT-202) ...... 81 Study Results ...... 85 8.1.3 Integrated Review of Effectiveness ...... 95 8.1.4 Integrated Assessment of Effectiveness ...... 95 8.2 Review of Safety ...... 96 8.2.1 Safety Review Approach ...... 96 8.2.2 Review of the Safety Database ...... 97 8.2.3 Adequacy of Applicant’s Clinical Safety Assessments ...... 100 8.2.4 Safety Results ...... 104 8.2.5 Analysis of Submission-Specific Safety Issues ...... 147 8.2.6 Clinical Outcome Assessment (COA) Analyses Informing Safety/Tolerability ...... 147 8.2.7 Safety Analyses by Demographic Subgroups ...... 147 8.2.8 Specific Safety Studies/Clinical Trials ...... 148 8.2.9 Additional Safety Explorations ...... 148 8.2.10 Safety in the Postmarket Setting ...... 149 8.2.11 Integrated Assessment of Safety ...... 149 8.3 Statistical Issues ...... 149 8.4 Conclusions and Recommendations ...... 149 9 Advisory Committee Meeting and Other External Consultations ...... 152 10 Pediatrics ...... 153 11 Labeling Recommendations ...... 154 11.1 Prescription Drug Labeling ...... 154 12 Risk Evaluation and Mitigation Strategies (REMS) ...... 160 4 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

13 Postmarketing Requirements and Commitment ...... 161 14 Division Director (DHOT) ...... 162 15 Division Director (OCP) ...... 162 16 Division Director (OB) Comments ...... 162 17 Division Director (Clinical) Comments ...... 162 18 Office Director (or designated signatory authority) Comments ...... 164 19 Appendices ...... 165 19.1 References ...... 165 19.2 Financial Disclosure ...... 167 19.3 Nonclinical Pharmacology/Toxicology...... 168 19.4 OCP Appendices (Technical documents supporting OCP recommendations) ...... 168 19.4.1 Appendix 1. Summary of Bioanalytical Method Validation and Performance .... 168 19.4.2 Appendix 2. Physiologically based pharmacokinetic (PBPK) analyses review ..... 175 19.4.3 Appendix 3. Population PK Analysis...... 184 19.4.4 Appendix 4. Exposure-Response Analysis ...... 191 19.5 Additional Clinical Outcome Assessment Analyses ...... 196

5 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table of Tables

Table 1. Regulatory Activity for IND 138179 ...... 27 Table 2. NDA Submission Regulatory History ...... 29 Table 3. Clinical Sites for Inspection ...... 30 Table 4: In Vitro Inhibition of Human and Rat FGFR Kinases and VEGFR2 ...... 36 Table 5: Effect of Pemigatinib on the In Vitro Viability of Cells with or without FGFR Alterations ...... 37 Table 6: Summary of In Vivo Anti-Tumor Activity in Mice Bearing FGFR2 Amplified KATOIII Xenografts ...... 38 Table 7: Selected Histopathology Findings (3-month Study; Rats) ...... 52 Table 8: Summary of General Pharmacology and Pharmacokinetic Characteristics of Pemigatinib ...... 66 Table 9. Effect of mild or moderate renal impairment on pemigatinib apparent clearance...... 71 Table 10. Effect of mild or moderate hepatic impairment on apparent pemigatinib clearance. 72 Table 11. Listing of Clinical Trials with Pemigatinib ...... 78 Table 12. Dose Reductions for Toxicity in Study INCB 54828-202 ...... 82 Table 13. Schedule of Laboratory Assessments in Study INCB 54828-202 ...... 83 Table 14: Patient Disposition in Cohort A of Study INCB 54828-202 ...... 86 Table 15. Summary of Protocol Deviations in Cohort A of Study INCB 54828-202 ...... 87 Table 16: Demographic Characteristics of Patients in Cohort A of Study INCB 54828-202 ...... 88 Table 17: Baseline Disease Characteristics of Patients in Cohort A ...... 88 Table 18: ORR per RECIST 1.1 as assessed by IRC ...... 91 Table 19: Subgroup Analysis of Cohort A Responses by Type of FGFR2 Abnormality and Fusion Partner1 ...... 91 Table 20: ORR for Demographic Subgroups of Cohort A by IRC per RECIST 1.1 ...... 92 Table 21: Data cut off dates for Pivotal Studies submitted for the Pemigatinib NDA ...... 97 Table 22: Summary for Exposure of Pemigatinib ...... 98 Table 23: Narrative Description of Selected Deaths ...... 105 Table 24: Serious Adverse Events by Preferred Term ≥ 1% incidence in any safety population 109 Table 25: Summary of Patient Narratives for Serious Adverse Events ...... 111 Table 26: Adverse Events Leading to Drug Discontinuation in ≥1% Incidence in any Safety Population ...... 114 Table 27: Treatment-Emergent Adverse Events leading to Treatment Interruption ≥ 1% Incidence in Any Safety Population ...... 118 Table 28: Treatment-Emergent Adverse Events leading to dose reduction ≥ 1% Incidence in any Safety Population ...... 120 Table 29: Incyte Review of Reversibility of Ocular Toxicity for Pemigatinib ...... 125 Table 30: Nail Toxicity Adverse Events (occurring at any rate) ...... 128 Table 31: Bone-related Adverse Events in Patients with Cholangiocarcinoma...... 131 Table 32: Narrative Summary of Pathologic Fractures ...... 133 Table 33: Overview of Safety Profile in Integrated Safety Populations ...... 135 6 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 34: Adverse Reactions (≥ 15% Incidence) in Patients Receiving Pemigatinib in Study-202 ...... 136 Table 35: Overview of Safety Profile by Age (Geriatrics) ...... 147 Table 36. Summary of Method Performance (Incyte assay)...... 169 (b) (4) Table 37. Summary of Method Performance ( assay)...... 175 Table 38. Summary of input parameters for pemigatinib PBPK model...... 178 Table 39. Updated input parameters for pemigatinib model ...... 179 Table 40. Summary of predicted vs. observed pemigatinib PK parameters following single dose or multiple dose oral administration ...... 181 Table 41 Summary of predicted and observed effects of itraconazole (a strong CYP3A inhibitor) and rifampin (a strong CYP3A inducer) on the PK of pemigatinib ...... 181 Table 42. Summary of predicted effects of moderate CYP3A inhibitors on the PK of pemigatinib following single dose or multiple dose administration ...... 182 Table 43. Summary of simulated pemigatinib PK parameters at steady state when it is co­ administered with CYP3A inhibitors ...... 183 Table 44. Overview of studies included in the population PK analysis...... 185 Table 45. Summary of PK sampling times ...... 185 Table 46. Demographics and baseline characteristics of pooled study data ...... 186 Table 47. Parameter estimates of final population PK model ...... 187 Table 48. Summary of ORR by Cycle 1 or During treatment hyperphosphatemia status ...... 193

7 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table of Figures

Figure 1: Effect of Pemigatinib on In Vivo Anti-Tumor Activity in Mice Bearing KATOIII FGFR2 Amplified Xenografts ...... 38 Figure 2: Effect of Single Oral Administration of Pemigatinib on FGFR2 Phosphorylation in KATO III Xenografts ...... 39 Figure 3: Effect of Pemigatinib on CTG-0997 Xenograft Growth in Nude Mice ...... 40 Figure 4: Effect of Single Oral Administration of Pemigatinib on Serum Phosphate in Mice ...... 41 Figure 5. Relationship between pemigatinib exposure and probability of hyperphosphatemia. 69 Figure 6: Swimmer’s plot for responders per RECIST 1.1 assessed by IRC ...... 94 Figure 7: Mean (± SE) Change from Baseline in Serum Creatinine Levels Over Time in Study INCB54828-202 ...... 140 Figure 8. Simulated and observed pemigatinib PK profiles following single dose or multiple dose oral administration ...... 180 Figure 9. Diagnostic plots for the final population PK model ...... 188 Figure 10. Visual predictive check stratified by visit and dose for final population PK model .. 189 Figure 11. Covariate Evaluation of Pemigatinib Oral Clearance Using Final Population PK Model ...... 190 Figure 12. Comparison of Patient Predicted Parameters and Exposures between Applicant’s and Reviewer’s models...... 190 Figure 13. Model-predicted vs. observed relationship of serum phosphate concentration change from baseline and pemigatinib AUCss...... 191 Figure 14. Model-predicted vs. observed relationship of ORR and serum phosphate concentration change from baseline or pemigatinib exposure (Cmax, ss) ...... 192 Figure 15. Model-predicted vs. observed relationship of ORR and pemigatinib exposure (Cmax, ss) ...... 192 Figure 16. PFS Vs. change in serum phosphate concentration from baseline or AUCss following once daily dosing of 13.5 mg pemigatinib ...... 193 Figure 17. Probability of Hyperphosphatemia Vs. Pemigatinib Cmax, ss ...... 195

8 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Reviewers of Multi-Disciplinary Review and Evaluation

Regulatory Project Manager Stacie Woods Nonclinical Reviewer Emily Wearne Nonclinical Team Leader Whitney Helms Office of Clinical Pharmacology Reviewer(s) Robert Schuck, Yangbing Li, Xinyuan Zhang Office of Clinical Pharmacology Team Leader(s) Jeanne Fourie-Zirkelbach, Rosane Charlab Orbach, Jiang Liu, Yuching Yang Clinical Reviewer Naomi Horiba (efficacy), Leigh Marcus (safety) Clinical Team Leader Martha Donoghue Statistical Reviewer Somak Chatterjee Statistical Team Leader Pallavi Mishra-Kalyani Associate Director of Labeling (acting) Ann Marie Trentacosti Cross-Disciplinary Team Leader Martha Donoghue Division Director (DHOT) John Leighton Division Director (OCP) Atik Rahman Division Director (OB) Shenghui Tang Deputy Division Director (OOD/DO3) Lola Fashoyin-Aje Office Director (or designated signatory authority) Marc Theoret

Additional Reviewers of Application OPQ ATL: Xing Wang, Drug Substance: Raymond Frankewich/Su Tran TL; Drug Product: Olen Stephens/Anamitro Banerjee; Facilities/Process: Sridhar Thumma/Bogdan Kurtyka; Biopharmaceutics: Mei Ou/Banu Zolnik; Environmental assessment: James Laurenson Microbiology Sridhar Thumma/ Bogdan Kurtyka TL OPDP Emily Dvorsky/Susannah O’Donnell TL OSI Yang-Min (Max) Ning/Aisha Johnson TL/Kassa Ayalew OSE/DEPI Richard Swain OSE/DMEPA Janine Stewart/Alice Tu TL OSE/DRISK Mei-Yean Chen/Naomi Boston TL Other (CDRH) Bowen Cui, Dun Liang/Donna Roscoe TL Other (ophthalmology consult) Wiley Chambers CDRH= Center for Devices and Radiological Health DEPI= Division of Epidemiology DMEPA=Division of Medication Error Prevention and Analysis DO3= Division of Oncology 3 DRISK=Division of Risk Management 9 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

OOD=Office of Oncologic Diseases OPQ=Office of Pharmaceutical Quality OPDP=Office of Prescription Drug Promotion OSI=Office of Scientific Investigations OSE= Office of Surveillance and Epidemiology

10 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Glossary

AC advisory committee ADME absorption, distribution, metabolism, excretion AE adverse event AR adverse reaction BLA biologics license application BPCA Best Pharmaceuticals for Children Act BRF Benefit Risk Framework CBER Center for Biologics Evaluation and Research CDER Center for Drug Evaluation and Research CDRH Center for Devices and Radiological Health CDTL Cross-Discipline Team Leader CFR Code of Federal Regulations CMC chemistry, manufacturing, and controls COSTART Coding Symbols for Thesaurus of Adverse Reaction Terms CRF case report form CRO contract research organization CRT clinical review template CSR clinical study report CSS Controlled Substance Staff DHOT Division of Hematology Oncology Toxicology DMC data monitoring committee ECG electrocardiogram eCTD electronic common technical document ETASU elements to assure safe use FDA Food and Drug Administration FDAAA Food and Drug Administration Amendments Act of 2007 FDASIA Food and Drug Administration Safety and Innovation Act FGFR Fibroblast growth factor receptor GCP good clinical practice GRMP good review management practice ICH International Conference on Harmonisation IND Investigational New Drug ISE integrated summary of effectiveness ISS integrated summary of safety ITT intent to treat MedDRA Medical Dictionary for Regulatory Activities mITT modified intent to treat NCI-CTCAE National Cancer Institute-Common Terminology Criteria for Adverse Event NDA new drug application NME new molecular entity 11 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

OCS Office of Computational Science OPQ Office of Pharmaceutical Quality OSE Office of Surveillance and Epidemiology OSI Office of Scientific Investigation PBRER Periodic Benefit-Risk Evaluation Report PD pharmacodynamics PI prescribing information PK pharmacokinetics PMC postmarketing commitment PMR postmarketing requirement PP per protocol PPI patient package insert (also known as Patient Information) PREA Pediatric Research Equity Act PRO patient reported outcome PSUR Periodic Safety Update report PT Preferred Term REMS risk evaluation and mitigation strategy SAE serious adverse event SAP statistical analysis plan SGE special government employee SOC standard of care TEAE treatment emergent adverse event

12 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

1 Executive Summary

Product Introduction

On September 30, 2019, Incyte Corporation (Incyte) submitted a New Drug Application (NDA) 213736 for pemigatinib tablets under Section 505(b)(1) of the Federal Food, Drug, and Cosmetic Act (FDCA) seeking accelerated approval of Pemazyre (pemigatinib) for the treatment of adults with previously treated, locally advanced or metastatic cholangiocarcinoma with a fibroblast growth factor receptor 2 (FGFR2) fusion or rearrangement as detected by an FDA-approved test.

Pemigatinib (previously called INCB054828) is a small molecule kinase inhibitor that inhibits the kinase activities of the fibroblast growth factor receptor (FGFR)1 (FGFR1), FGFR2, and FGFR3. Pemigatinib is the active pharmaceutical ingredient in Pemazyre tablets.

The molecular formula for pemigatinib is C24H27F2N5O4 and the molecular mass is 487.5 g/mole. The chemical name is name 3-(2,6-difluoro-3,5-dimethoxyphenyl)-1-ethyl-8­ (morpholin-4-ylmethyl)-1,3,4,7-tetrahydro-2H-pyrrolo[3',2':5,6]pyrido[4,3-d]pyrimidin-2-one. Pemigatinib has the following chemical structure:

O F O O O N N F N

N N H

Pemazyre (pemigatinib) is supplied as uncoated tablets for oral administration that contain 4.5 mg, 9 mg, or 13.5 mg of the pemigatinib active ingredient. The inactive ingredients include magnesium stearate, microcrystalline cellulose, and sodium starch glycolate.

The proposed recommended dosage regimen for pemigatinib is 13.5 mg orally once daily for 14 consecutive days followed by 7 days off therapy in 21-day cycles.

Pemazyre (pemigatinib) is a new molecular entity and has not been previously marketed.

13 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Conclusions on the Substantial Evidence of Effectiveness

The clinical, clinical pharmacology, nonclinical, and statistical review teams unanimously agree that the NDA for Pemazyre (pemigatinib) tablets meets the statutory standards for approval under 21 CFR 314, Subpart H (accelerated approval) for the following indication:

PEMAZYRE is indicated for the treatment of adults with previously treated, unresectable locally advanced or metastatic cholangiocarcinoma with a fibroblast growth factor receptor 2 (FGFR2) fusion or other rearrangement as detected by an FDA-approved test.

This indication is approved under accelerated approval based on overall response rate and duration of response. Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s).

This approval recommendation is primarily based on safety and efficacy results of a single trial, Study INCB 54828-202 (also referred to as Study -202, or FIGHT-202; NCT02924376). Study INCB 54828-202 is an open label, non-randomized, multi-cohort trial that evaluated the efficacy of pemigatinib in 107 patients with locally advanced unresectable or metastatic cholangiocarcinoma harboring an FGFR2 gene fusion or rearrangement, whose disease had progressed on or after at least 1 prior therapy (Cohort A). The safety of pemigatinib was evaluated in 146 patients with previously treated, locally advanced or metastatic cholangiocarcinoma who received at least one dose of pemigatinib in Study INCB 54828-202 (107 patients with tumors harboring a FGFR2 gene fusion or rearrangement enrolled in Cohort A and 39 additional patients whose tumors did not harbor an FGFR2 gene fusion or rearrangement). The safety of pemigatinib is also supported by data from an additional 320 patients treated with pemigatinib as a single agent in other clinical trials.

Study INCB 54828-202 demonstrated a clinically meaningful and durable overall response rate (ORR) in patients with previously treated, locally advanced or metastatic cholangiocarcinoma with a FGFR2 gene fusion or rearrangement, a serious and life-threatening disease. In the first 107 patients with FGFR2 gene fusion/rearrangement--positive solid tumors who received at least one dose of pemigatinib, the estimated ORR was 36% (95% confidence interval [CI]: 27%, 45%). The median duration of response (DOR) was 9.1 months; 24 of the 38 (63%) responders had a DOR lasting at least 6 months and 7 (18%) responders had a DOR lasting at least 12 months at the time of the analysis. There are no available FDA-approved drugs for the second- line treatment of patients with unresectable or metastatic cholangiocarcinoma, and no approved-treatments specifically for the subset of patients with cholangiocarcinoma whose tumors harbor an FGFR2 gene fusion or rearrangement. Taken together, the ORR and DOR results in in Study INCB 54828-202 are reasonably likely to predict an effect on irreversible morbidity or mortality or other clinical benefit, and represent a meaningful advantage over existing treatments for the proposed indication, and satisfy the requirements for accelerated approval. Concurrently with the approval of Pemazyre, FDA will approve the Foundation Medicine companion diagnostic assay (Foundation Medicine CDx) supplemental Premarket 14 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Approval (sPMA) application to select patients with cholangiocarcinoma harboring an FGFR2 gene fusion or other select rearrangement, for treatment with Pemazyre. Incyte has agreed to a postmarketing requirement to conduct a randomized clinical trial demonstrating improvement of progression-free survival or overall survival in patients with unresectable locally advanced or metastatic cholangiocarcinoma with an FGFR2 gene fusion or rearrangement, to confirm the clinical benefit of pemigatinib.

15 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Benefit-Risk Assessment

Benefit-Risk Summary and Assessment

Cholangiocarcinoma (CCA) is a rare cancer arising from epithelial cells of bile ducts that is grouped into anatomic subtypes based on location of origin in the biliary tract: intrahepatic (iCCA), perihilar, or extrahepatic. CCA accounts for approximately 3% of all gastrointestinal cancers and 10-25% of primary hepatic malignancies worldwide and the incidence appears to be rising (Rizvi et al. 2013). In the U.S. in 2014, the annual incidence of iCCA was 1.49 per 100,000, representing a two-fold increase over the past four decades (Saha et al. 2016). The median age at presentation is 65 years in Western industrialized nations, and most patients develop CCA in the absence of identifiable risk factors (Blechacz 2017). Although surgery is the preferred treatment option, most patients have advanced-stage disease at the time of diagnosis, and only 35% are eligible for surgical resection with curative intent. For patients with advanced-stage or unresectable CCA, the median overall survival with standard of care chemotherapy is less than one year (Valle et al. 2010).

Fibroblast growth factor/fibroblast growth factor receptor (FGFR) fusions are a reported genetic modification in iCCA and have been identified as an early driver of oncogenic events in iCCA (Nakamura et al. 2015). FGFR2 fusions are present in an estimated 10-20% of patients with iCCA (Krook et al. 2020). Patients with FGFR rearrangements appear to have a longer median overall survival compared to those with iCCA lacking a FGFR rearrangement. A retrospective review of 377 patients with CCA reported a median OS of 37 months in patients with FGFR gene fusion- positive CCA compared to 20 months in the unselected CCA population (Jain et al. 2018). The baseline demographic and disease characteristics of patients with FGFR rearrangements also appear to be prognostically favorable, including diagnosis at a younger age and an earlier stage of disease. Patients with FGFR fusion-positive CCA are also predominantly female (63%) and usually have iCCA (87%).

There are no FDA-approved drugs for the second-line treatment of patients with unresectable or metastatic cholangiocarcinoma, and no treatments approved specifically for the subset of patients with cholangiocarcinoma with an FGFR2 gene fusion or rearrangement, irrespective of line of treatment.

Study INCB 54828-202 (also referred to as Study -202, or FIGHT-202; NCT02924376) provided the evidence to support the effectiveness of pemigatinib for the proposed indication. Study INCB 54828-202 is an open label, non-randomized, multicohort trial that evaluated pemigatinib in 107 patients with locally advanced unresectable or metastatic cholangiocarcinoma harboring an FGFR2 gene fusion or non-fusion rearrangement, whose disease had progressed on or after at least 1 prior therapy (Cohort A). Qualifying in-frame fusions and other rearrangements were predicted to have a breakpoint within intron17/exon 18 of the FGFR2 gene and leave the FGFR2 kinase domain intact. 16 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Patients received pemigatinib in 21-day cycles at a dosage of 13.5 mg orally once daily for 14 consecutive days, followed by 7 days off therapy. pemigatinib was administered until disease progression or unacceptable toxicity. The major efficacy outcome measures were overall response rate (ORR) and duration of response (DoR) as determined by an independent review committee (IRC) according to RECIST v1.1. The median age was 56 years (range: 26 to 77 years), 61% were female, 74% were White, and 95% had a baseline Eastern Cooperative Oncology Group (ECOG) performance status of less than 2 (ECOG 0: 42%; ECOG 1: 53%). Ninety-eight percent of patients had iCCA. Eighty-six percent of patients had in-frame FGFR2 gene fusions. Among the patients with in-frame FGFR2 gene fusions, the most common FGFR2 fusion identified was FGFR2-BICC1 (34%%). Fourteen percent of patients had other FGFR2 rearrangements that could not be confidently predicted to be in-frame fusions, including rearrangements without an identifiable partner gene. All patients had received at least 1 prior line of systemic therapy, 27% had 2 prior lines of therapy, and 12% had 3 or more prior lines of therapy. Ninety-six percent of patients had received prior platinum-based therapy including 76% with prior gemcitabine/cisplatin.

Study INCB 54828-202 demonstrated a clinically meaningful and durable overall response rate (ORR) in patients with previously treated, locally advanced or metastatic cholangiocarcinoma with a FGFR2 gene fusion or other rearrangement, a serious and life-threatening disease. In the first 107 patients with FGFR2 gene fusion/rearrangement-positive cholangiocarcinoma who received at least one dose of pemigatinib, the estimated overall response rate (ORR) was 36% (95% confidence interval [CI]: 27%, 45%). At the time of the analysis, the median duration of response (DOR) was 9.1 months; 24 of the 38 (63%) responders had a DOR lasting at least 6 months and 7 (18%) responders had DOR lasting at least 12 months .

The primary data supporting the safety of pemigatinib for the proposed indication was derived from a pooled population of 146 patients with previously treated, locally advanced or metastatic cholangiocarcinoma who received at least one dose of pemigatinib in Study INCB 54828-202 (107 patients with tumors harboring a FGFR2 gene fusion or rearrangement enrolled in Cohort A, and 39 additional patients whose tumors did not harbor an FGFR2 gene fusion or rearrangement). The safety evaluation of pemigatinib was also supported by data from an additional 320 patients with a variety of cancers treated with pemigatinib as a single agent in 4 other multicenter single arm trials (Study INCB 54828-101, Study INCB 54828-102, Study INCB 54828-201, and Study INCB 54828-203).

Although assessment of a causal relationship between pemigatinib and treatment-emergent reactions was limited in the context of the single arm design of the trials providing safety data, the adverse reactions observed in patients treated with pemigatinib were largely expected given the mechanism of action and the toxicity profile observed in preclinical studies. Among the 146 patients with cholangiocarcinoma enrolled in Study INCB 54828-202, the most common adverse reactions to pemigatinib (incidence ≥ 20%) are hyperphosphatemia, alopecia, diarrhea, nail toxicity, fatigue, dysgeusia, nausea, constipation, stomatitis, dry eye, dry mouth, decreased appetite, vomiting, arthralgia, abdominal pain,

17 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

hypophosphatemia, back pain, and dry skin. Clinically relevant adverse reactions occurring in ≤ 10% of patients included fractures (2.1%). In all patients treated with pemigatinib, 1.3% experienced pathologic fractures (which included patients with and without cholangiocarcinoma).

The most important serious risks of pemigatinib are ocular toxicity and hyperphosphatemia; these adverse reactions appear manageable and largely reversible with dosage modification and supportive care. Among the 466 patients who received pemigatinib across clinical trials, retinal pigment epithelial detachment (RPED), which can cause symptoms such as blurred vision, visual floaters, or photopsia, occurred in 6% of patients, including Grade 3-4 events in 0.6% of patients. The median time to first onset of RPED was 62 days. RPED led to dose interruption of pemigatinib in 1.7% of patients, and dose reduction and permanent discontinuation in 0.4% and in 0.4% of patients, respectively. RPED resolved or improved to Grade 1 in 87.5% of patients who required dosage modification of pemigatinib for RPED. In order to mitigate the risk of severe RPED, comprehensive ophthalmologic monitoring including ocular coherence tomography is recommended prior to initiation of pemigatinib, every 2 months for the first 6 months of treatment, and every 3 months thereafter during treatment. Increases in phosphate levels are a pharmacodynamic effect of pemigatinib. Among 466 patients who received pemigatinib across clinical trials, hyperphosphatemia was reported in 92% of patients based on laboratory values above the upper limit of normal. The median time to onset of hyperphosphatemia was 8 days (range 1-169). Phosphate lowering therapy was required in 29% of patients receiving pemigatinib.

Overall, the toxicity profile of pemigatinib is considered acceptable when considering the anti-tumor effects (i.e., durable responses) in patients with cholangiocarcinoma harboring a FGFR2 fusion or other rearrangement, who have a poor life expectancy and limited treatment options. The major safety risks of pemigatinib are toxicities that oncologists are well-trained to manage and are acceptable for a population with a serious and life-threatening condition.

Taken together, the ORR and DOR results in Study INCB 54828-202 are reasonably likely to predict an effect on irreversible morbidity or mortality or other clinical benefit, and represent a meaningful advantage over existing treatments for the proposed indication, meeting the requirements for accelerated approval. Concurrently with the approval of pemigatinib, FDA will approve the supplemental Premarket Approval (sPMA) application for the Foundation Medicine companion diagnostic assay (Foundation Medicine CDx), for the selection of patients with FGFR2 fusions or other select rearrangements for treatment with Pemazyre. Incyte has agreed to a postmarketing requirement to conduct a randomized clinical trial demonstrating improvement of progression-free survival or overall survival in patients with unresectable locally advanced or metastatic cholangiocarcinoma with an FGFR2 gene fusion or rearrangement, to confirm the clinical benefit of pemigatinib.

18 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dimension Evidence and Uncertainties Conclusions and Reasons • Cholangiocarcinoma (CCA) is a rare cancer arising from epithelial Cholangiocarcinoma is a serious and life- cells of bile ducts that is grouped into anatomic subtypes based on threatening illness and there is no satisfactory location of origin in the biliary tract: intrahepatic (iCCA), perihilar, or available therapy for the treatment of extrahepatic. CCA accounts for approximately 3% of all cholangiocarcinoma with a FGFR2 gene fusion gastrointestinal cancers and 10-25% of primary hepatic or other rearrangement that has received at malignancies worldwide and the incidence appears to be rising least one prior line of treatment. (Rizvi et al. 2013). In the U.S. in 2014, the annual incidence of iCCA was 1.49 per 100,000, representing a two-fold increase over the past four decades (Saha et al. 2016).

• The median age at presentation is 65 years in Western industrialized nations, and most patients develop CCA in the absence of identifiable risk factors (Blechacz 2017). Analysis of Condition • Although surgery is the preferred treatment option, most patients have advanced-stage disease at the time of diagnosis, and only 35% are eligible for surgical resection with curative intent. For patients with advanced-stage or unresectable CCA, the median overall survival with standard of care chemotherapy is less than one year (Valle et al 2010).

• Fibroblast growth factor/fibroblast growth factor receptor (FGFR) fusions are a reported genetic modification in iCCA and have been identified as an early driver of oncogenic events in iCCA (Nakamura et al. 2015). FGFR2 fusions are present in an estimated 10-20% of patients with iCCA (Krook et al. 2020). Patients with FGFR rearrangements appear to have a longer median overall survival compared to those with iCCA lacking a FGFR rearrangement. A

19 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dimension Evidence and Uncertainties Conclusions and Reasons retrospective review of 377 patients with CCA reported a median OS of 37 months in patients with FGFR gene fusion-positive CCA compared to 20 months in the unselected CCA population (Jain et al. 2018).

• The baseline demographic and disease characteristics of patients with FGFR rearrangements also appear to be prognostically favorable, including diagnosis at a younger age and an earlier stage of disease. Patients with FGFR fusion-positive CCA are also predominantly female (63%) and usually have iCCA (87%).

• Current treatment options for patients with CCA are limited, and There is an unmet medical need for new there are no approved treatments for the treatment of patients effective treatments for patients with with CCA in the second-line setting, irrespective of whether the cholangiocarcinoma with a FGFR2 gene tumor harbors an FGFR2 gene fusion or rearrangement. fusion or other rearrangement who have received at least one prior line of • Entrectinib and larotrectinib are approved for the treatment of treatment. Current treatment options patients with solid tumors that harbor NTRK gene fusions and who for patients with CCA are limited, and Current have no satisfactory alternative treatment options; however, NTRK there are no approved treatments for Treatment gene fusions occur rarely in CCA. NTRK fusions are estimated to the treatment of patients with CCA in Options occur 1.3% of patients with CCA (Kheder et al 2018). the second-line setting, irrespective of whether the tumor harbors an FGFR2 • Pembrolizumab is approved for the treatment of patients with gene fusion or rearrangement. microsatellite-high/mismatch repair deficient (MSI-high/dMMR) solid tumors that have progressed following prior treatment and who have no satisfactory treatment options; however, only approximately 4% of CCA are MSI-H/dMMR (Goeppert et al. 2019).

20 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dimension Evidence and Uncertainties Conclusions and Reasons • The only standard of care second-line chemotherapy option for patients with CCA supported by randomized data is 5­ fluorouracil/leucovorin in combination with oxaliplatin (FOLFOX). In Study ABC-06, FOLFOX conferred a median OS benefit of 6.2 months as compared to 5.3 months with best supportive care (hazard ratio [HR] 0.69; 95% CI: 0.50-0.97; p = 0.031; Lamarca et al. 2019). Other regimens that are considered acceptable include irinotecan in combination with a fluoropyrimidine, a platinum plus fluoropyrimidine, gemcitabine in combination with a platinum or fluoropyrimidine, or regorafenib(NCCN 2020).

• Study INCB 54828-202 (also referred to as Study -202, or FIGHT-202; The magnitude and duration of NCT02924376) provides the evidence of effectiveness for responses observed in patients with pemigatinib for the proposed indication. Study INCB 54828-202 is cholangiocarcinoma with a FGFR2 gene an open label, non-randomized, multicohort trial that evaluated fusion who received prior treatment was pemigatinib in 107 patients with locally advanced unresectable or large and clinically meaningful. The metastatic cholangiocarcinoma harboring an FGFR2 gene fusion or submitted evidence meets the statutory non-fusion rearrangement, whose disease had progressed on or evidentiary standard for accelerated after at least 1 prior therapy (Cohort A). approval. Benefit • Patients received pemigatinib in 21-day cycles at a dosage of Incyte has agreed to a postmarketing 13.5 mg orally once daily for 14 consecutive days, followed by requirement to submit data from a 7 days off therapy. pemigatinib was administered until disease randomized trial to verify and confirm progression or unacceptable toxicity. The major efficacy outcome the clinical benefit of pemigatinib in measures were overall response rate (ORR) and duration of patients with FGFR2 response (DOR) as determined by an independent review fusion/rearrangement- positive committee (IRC) according to RECIST v1.1. cholangiocarcinoma.

21 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dimension Evidence and Uncertainties Conclusions and Reasons • Study INCB 54828-202 demonstrated a clinically meaningful and durable ORR in patients with previously treated, locally advanced or metastatic cholangiocarcinoma with a FGFR2 gene fusion or other rearrangement. These responses were also durable. In the first 107 patients with FGFR2 gene fusion/rearrangement--positive cholangiocarcinoma who received at least one dose of pemigatinib, the estimated ORR was 36% (95% confidence interval [CI]: 27%, 45%). At the time of the analysis, the median DOR was 9.1 months; 24 of the 38 (63%) responders had a DOR lasting at least 6 months and 7 (18%) responders had a DOR of at least 12 months. • The primary data supporting the safety of pemigatinib for the The observed safety profile is acceptable proposed indication was provided from data derived from 146 when assessed in the context of the patients with previously treated, locally advanced or metastatic treatment of a life-threatening disease. cholangiocarcinoma who received at least one dose of pemigatinib Most of the adverse reactions to in Study INCB 54828-202 (107 patients with tumors harboring a pemigatinib were manageable with FGFR2 gene fusion or rearrangement enrolled in Cohort A and 39 supportive care and dose modification additional patients whose tumors did not harbor an FGFR2 gene as needed. The significant and fusion or rearrangement). potentially serious adverse reactions of Risk and Risk hyperphosphatemia and ocular toxicity Management • The safety evaluation of pemigatinib was also supported by data are adequately addressed in the from an additional 320 patients with a variety of cancers treated Warnings and Precautions section and with pemigatinib as a single agent in 4 other multicenter single arm the dose modification recommendations trials (Study INCB 54828-101, Study INCB 54828-102, Study INCB included in product labeling. 54828-201, and Study INCB 54828-203). There were no significant safety • Although assessment of a causal relationship between pemigatinib concerns identified during the review of and treatment-emergent adverse reactions was limited in the the application requiring risk context of the single arm design of the trials providing safety data, management beyond labeling or

22 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dimension Evidence and Uncertainties Conclusions and Reasons the adverse reactions observed in patients treated with pemigatinib warranting consideration for a Risk were largely expected given with the mechanism of action and the Evaluation and Mitigation toxicity profile observed in preclinical studies. Strategy(REMS).

• Among the 146 patients with cholangiocarcinoma enrolled in Study INCB 54828-202, the most common adverse reactions to pemigatinib (incidence ≥ 20%) are hyperphosphatemia, alopecia, diarrhea, nail toxicity, fatigue, dysgeusia, nausea, constipation, stomatitis, dry eye, dry mouth, decreased appetite, vomiting, arthralgia, abdominal pain, hypophosphatemia, back pain, and dry skin. Clinically relevant adverse reactions occurring in ≤ 10% of patients included fractures (2.1%). In all patients treated with pemigatinib, 1.3% experienced pathologic fractures (which included patients with and without cholangiocarcinoma).

23 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Patient Experience Data

Patient Experience Data Relevant to this Application (check all that apply) X The patient experience data that were submitted as part of the Section of review where application include: discussed, if applicable [e.g., Section 6.1 Study endpoints] X Clinical outcome assessment (COA) data, such as X Patient reported outcome (PRO) 8.1.2 Study Results □ Observer reported outcome (ObsRO) □ Clinician reported outcome (ClinRO) □ Performance outcome (PerfO) □ Qualitative studies (e.g., individual patient/caregiver interviews, focus group interviews, expert interviews, Delphi Panel, etc.) □ Patient-focused drug development or other stakeholder meeting summary reports □ Observational survey studies designed to capture patient experience data Natural history studies □ □ Patient preference studies (e.g., submitted studies or scientific publications) □ Other: (Please specify): □ Patient experience data that were not submitted in the application, but were considered in this review: □ Input informed from participation in meetings with patient stakeholders □ Patient-focused drug development or other stakeholder meeting summary reports □ Observational survey studies designed to capture patient experience data □ Other: (Please specify): □ Patient experience data was not submitted as part of this application.

X Martha Donoghue, MD Cross Discipline Team Leader

24 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

2 Therapeutic Context

Analysis of Condition

Cholangiocarcinoma (CCA) is a rare malignancy arising from epithelial cells of bile ducts that is grouped into anatomic subtypes based on location of origin in the biliary tract: intrahepatic (iCCA), perihilar, or extrahepatic. CCA accounts for approximately 3% of all gastrointestinal cancers and 10-25% of primary hepatic malignancies worldwide and the incidence appears to be rising (Rizvi et al. 2013). In the U.S. in 2014, the annual incidence of iCCA was 1.49 per 100,000, representing a two-fold increase over the past four decades (Saha et al. 2016). The median age at presentation is 65 years in Western industrialized nations, and most patients develop CCA in the absence of identifiable risk factors (Blechacz 2017).

Although surgery is the preferred treatment option, most patients have advanced-stage disease at the time of diagnosis, and only 35% are eligible for surgical resection with curative intent. For patients with advanced-stage or unresectable CCA, the median overall survival with standard of care chemotherapy is less than one year (Valle et al 2010).

Fibroblast growth factor/fibroblast growth factor receptor (FGFR) fusions are a reported genetic modification in iCCA and have been identified as an early driver of oncogenic events in iCCA (Nakamura et al. 2015). FGFR2 fusions are present in an estimated 10-20% of patients with iCCA (Krook et al. 2020). Patients with FGFR rearrangements appear to have a longer median overall survival compared to those with iCCA lacking a FGFR rearrangement. A retrospective review of 377 patients with CCA reported a median OS of 37 months in patients with FGFR gene fusion- positive CCA compared to 20 months in the unselected CCA population (Jain et al. 2018). The baseline demographic and disease characteristics of patients with FGFR rearrangements also appear to be prognostically favorable, including diagnosis at a younger age and an earlier stage of disease. Patients with FGFR fusion-positive CCA are also predominantly female (63%) and the location is intrahepatic for the majority of patients (87%).

Analysis of Current Treatment Options

After administration of first-line standard of care chemotherapy with gemcitabine and cisplatin in advanced or metastatic CCA, which confers a 3.6 month median OS benefit compared to gemcitabine alone (Valle et al. 2010), current treatment options for patients with CCA are limited, and there are no approved treatments for the treatment of patients with CCA in the second-line setting. Entrectinib and larotrectinib, and pembrolizumab, respectively, are approved for the treatment of patients with solid tumors with no satisfactory alternative treatment options that harbor NTRK gene fusions or are microsatellite-high/mismatch repair deficient (MSI-high/dMMR), respectively; however, these molecular aberrations occur rarely in

25 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

CCA. NTRK fusions are estimated to occur 1.3% of patients with CCA and approximately 4% of CCA are MSI-H/dMMR (Goeppert et al. 2019 and Kheder et al 2018).

Currently, the only standard of care second-line chemotherapy option for patients with CCA supported by randomized data is 5-fluorouracil/leucovorin in combination with oxaliplatin (FOLFOX). In Study ABC-06, FOLFOX conferred a median OS benefit of 6.2 months as compared to 5.3 months with best supportive care (hazard ratio [HR] 0.69; 95% CI: 0.50-0.97; p = 0.031; Lamarca et al. 2019) in patients with locally advanced or metastatic biliary tract cancers previously treated with gemcitabine and cisplatin. Fatigue and neutropenia were more common in the treatment arm, and the rate of Grade 3-4 toxicity was higher than in the control arm (59% vs. 39%). Other regimens that are considered acceptable include irinotecan in combination with a fluoropyrimidine, a platinum plus fluoropyrimidine, gemcitabine in combination with a platinum or fluoropyrimidine, or regorafenib (NCCN 2020).

Limited data exist regarding outcomes for the population of patients with FGFR genetic fusions specifically. One retrospective study of 45 patients with FGFR2 fusions treated with gemcitabine in combination with a platinum showed no significant differences in progression-free survival (PFS; 0.5 vs 0.5 years, HR 1.19, p = 0.36) (Almquist 2019).

26 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

3 Regulatory Background

U.S. Regulatory Actions and Marketing History

Pemigatinib is not approved for marketing in the United States.

Summary of Presubmission/Submission Regulatory Activity

Table 1. Regulatory Activity for IND 138179

Date Event

The initial IND for INCB054828 (IND 124358) was submitted, containing the clinical protocol for Study INCB 54828-101, entitled “A October 27, 2014 Phase 1, Open-Label, Dose-Escalation, Safety and Tolerability Study of INCB054828 in Subjects with Advanced Malignancies”. On November 26, 2014, the study was allowed to proceed.

Incyte submitted original IND 138179 for pemigatinib containing Study INCB 54828-202, entitled “A Phase 2, Open-Label, Single-Arm, Multicenter Study to Evaluate the Safety of INCB054828 in Subjects January 30, 2018 with Advanced/Metastatic or Surgically Unresectable Cholangiocarcinoma Including FGFR2 Translocations Who Failed Previous Therapy.”

FDA granted orphan drug designation to INCB054828 for the March 12, 2018 treatment of cholangiocarcinoma.

FDA issued preliminary comments in preparation for a Type B End of Phase 2 meeting addressing the design of the proposed trial intended March 8, 2018 to provide the primary evidence to support regular approval, Study INCB 54828-302. Incyte cancelled the meeting on May 10, 2018, following review of the preliminary responses.

FDA held a Type B End of Phase 2 CMC only meeting to discuss the April 5, 2018 CMC development plan for INCB054828.

FDA issued written responses to Incyte regarding the adequacy of a proposed clinical data package to support an NDA comprising results May 21, 2018 from the unplanned post-hoc interim analysis of Study INCB-202, reflecting data from 45 patients enrolled in Cohort A, stating that it would not be sufficient to adequately characterize the safety and 27 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Date Event effectiveness of INCB054828 for a marketing application. FDA recommended a minimum sample size of 100 patients with sufficient follow-up to adequately characterize the safety and effectiveness of pemigatinib in patients with previously treated, unresectable cholangiocarcinoma with an FGFR2 fusion. FDA stated that in general, an ORR with a lower bound of the 95% confidence interval (CI) that is greater than 15%, accompanied by durations of response of sufficient magnitude to be considered clinically meaningful (e.g., at least 6 months in the majority of responders), and a favorable risk/benefit profile may provide sufficient evidence to support a marketing application seeking accelerated approval in the proposed patient population. FDA clarified that ORR should be calculated using a denominator comprising all patients who received at least one dose of pemigatinib. FDA recommended that data be sufficiently mature to assess duration of response for a minimum of 12 months form the time of onset of response for each responder. Incyte submitted Protocol INCB 24828-302, entitled “A Phase 3, Open- Label, Randomized, Active-Controlled, Multicenter Study to Evaluate the Efficacy and Safety of INCB054828 Versus Gemcitabine Plus August 14, 2018 Cisplatin Chemotherapy in First-Line Treatment of Participants With Unresectable or Metastatic Cholangiocarcinoma With FGFR2 Rearrangement”, to IND 138179. FDA granted breakthrough designation to pemigatinib for the treatment of patients with previously treated advanced/metastatic or February 13, 2019 unresectable cholangiocarcinoma with an FGFR2 fusion based on results of an interim analysis of Study INCB 54828-202.

A meeting between Incyte and FDA was held to discuss the content June 12, 2019 and format of an anticipated NDA for pemigatinib based on data derived from Study INCB 54828-202.

A pre-NDA meeting was held to discuss a planned NDA submission for accelerated approval of pemigatinib, primarily based upon top-line results from Study INCB 54828-202, for the proposed indication ”for the treatment of adult patients with previously treated, August 8, 2019 advanced/metastatic or surgically unresectable cholangiocarcinoma with an FGFR2 rearrangement or fusion”. Top-line results included data from 107 patients enrolled in Cohort A of Study INCB 54828-202 based on a data cutoff date of March 22, 2019. Incyte proposed to provide a minimum of 6 months follow-up from the time of initial

28 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Date Event response for 35/38 (92%) of the currently known confirmed responders, and additional follow-up data for duration of response at the 4-month safety update to provide a minimum of 12 months of follow-up for all patients in the efficacy analysis set including a minimum of 6 months’ follow-up from the time of initial response for all confirmed responders. FDA agreed that the proposed data package could support filing of an NDA under the provisions of accelerated approval for the proposed indication. Source: reviewer table

Table 2. NDA Submission Regulatory History

Date Event

Incyte submitted original NDA 213736. This NDA included a request for categorical exclusion from environmental assessment, and a September 30, 2019 request for waiver of in vivo bioavailability studies for pemigatinib 9 mg and 13.5 mg tablets.

Incyte submitted a request for Proprietary Name Review for the October 1, 2019 proposed name Pemazyre.

October 1, 2019 Incyte submitted a Request for Priority Review.

November 26, 2019 FDA granted priority review.

Division of Medication Error Prevention and Analysis (DMEPA) issued December 17, 2019 a Proprietary Name Request, Conditionally Acceptable letter.

January 8, 2020 A mid-cycle communication meeting was held with Incyte.

A mid-cycle communication Advice-Information Request letter was January 17, 2020 issued to Incyte.

29 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

4 Significant Issues from Other Review Disciplines Pertinent to Clinical Conclusions on Efficacy and Safety

Office of Scientific Investigations (OSI)

Site inspections were conducted for Incyte Corporation and 3 clinical sites (Table 3. Diagnostic testing performed centrally to determine cohort assignment and reported tumor response data were verifiable with information and documents available at the sponsor’s site. No major deficiencies were identified and the Office of Scientific Investigations concluded that the clinical data generated from the three investigator sites and submitted to the NDA by Incyte appear to be reliable and supportive of this NDA and the proposed indication for pemigatinib.

Table 3. Clinical Sites for Inspection

Safety Site/Investigator Reason for Site Selection Population Site 012 Vaibhav Sahai Site 012 was selected due to enrollment of large University of Michigan 8 numbers of patients. Rogel Cancer Center Ann Arbor, MI Site 018 Raed Al-Rajabi Site 018 was selected due to enrollment of large (b) (4) University of Kansas 7 numbers of patients and financial disclosure Medical Center . Westwood, KS Site 020 David Gallinson Site 020 was selected due to high treatment effect 4 (b) (4) Summit Medical Group and financial disclosure . Florham, NJ

Product Quality

The Office of Pharmaceutical Quality (OPQ) did not identify any product quality issues that would preclude approval of pemigatinib capsules under this NDA. Please refer to the OPQ review of this application for additional information. No safety or efficacy concerns were identified during this review that related to Chemistry, Manufacturing, and Controls (CMC).

Pursuant to 21 CFR 25.31(b), Incyte submitted a Request for Categorical Exclusion from the preparation of an environmental assessment for pemigatinib and submitted environmental assessment data to support this request. OPQ determined that this request could be granted, as the estimated concentration of the drug substance at the point of entry into the aquatic

30 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (4) environment (EIC) is projected to be less than the 1 ppb limit that allows an exclusion to be granted.

Clinical Microbiology

This NDA was reviewed by OPQ’s Division of Microbiology Assessment. The microbiology reviewers did not identify any issues that would preclude approval of pemigatinib tablets. Pemigatinib should be stored at room temperature 20°C - 25°C (68°F - 77°F); excursions permitted to 15°C - 30°C (59°F - 86°F).

Devices and Companion Diagnostic Issues

Foundation Medicine submitted a supplemental Premarket Approval (sPMA) application to Center of Devices and Radiological health for a companion diagnostic test, Foundation Medicine CDx, for the selection of patients with FGFR2 fusions or select rearrangements for treatment with pemigatinib. The device will receive marketing approval concurrent with approval of this NDA.

31 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

5 Nonclinical Pharmacology/Toxicology

Executive Summary

Pemigatinib (INCB054828) is a kinase inhibitor that has activity against fibroblast growth factor receptor 1 (FGFR1), FGFR2, and FGFR3 at concentrations (0.39-1.2 nM) that have been achieved at the 13.5 mg dose level used in clinical trials conducted to support the approval of the drug (free Cmax of approximately 22 nM based on ~91% protein binding). Pemigatinib also inhibited the kinase activity of FGFR4 at an IC50 of 30 nM. FGFR is a receptor tyrosine kinase involved in angiogenesis, cell proliferation, survival, and migration. Aberrant FGFR signaling resulting from gene amplification or mutation, chromosomal rearrangements, or ligand-dependent receptor activation has been demonstrated in numerous human cancers. While pemigatinib showed biochemical inhibition of vascular endothelial growth factor receptor 2 (VEGFR2; IC50 = 71 nM) at relatively low concentrations, follow-up studies in cell lines did not confirm this activity and pemigatinib did not exhibit substantial off-target activity against other targets in a kinase screen or in in vitro secondary pharmacology screens.

Treatment with pemigatinib inhibited in vitro and in vivo FGFR phosphorylation in FGFR2­ amplified human gastric cancer cells and inhibited in vitro FGFR phosphorylation in Ba/F3 cells stably expressing TEL-FGFR1 or TEL-FGFR3 fusion proteins and in FGFR2-amplified cells spiked with human whole blood. Pemigatinib also reduced the phosphorylation of FGFR1 and the downstream signaling proteins ERK1/2 and STAT5 in cells expressing the constitutively active FGFR1OP2-FGFR1 fusion protein. Consistent with these findings, incubation with pemigatinib inhibited the in vitro viability of cancer cell lines with FGFR1 or FGFR2 amplification; FGFR1, FGFR2, or FGFR3 fusions; and FGFR3 translocations at clinically relevant concentrations but did not substantially inhibit the viability of cells without FGFR alterations, including human T cells, suggesting that the effects of pemigatinib on cell growth are selective for cells dependent on FGFR pathway signaling. Notably, pemigatinib inhibited the in vitro viability of two Ba/F3 cell lines expressing fusions (FGFR2-CCDC6 and FGFR2-AHCYL) identified in responsive patients with cholangiocarcinoma enrolled in pemigatinib clinical trials. Pemigatinib exhibited in vivo anti- tumor activity in mice bearing subcutaneous (SC) human xenografts with FGFR1, FGFR2, or FGFR3 activation, including patient-derived cholangiocarcinoma xenografts expressing a FGFR2­ TRA2B fusion and FGFR2-amplified human gastric cancer xenografts. Pemigatinib doses ≥3 mg/kg were generally not tolerated in mice due to ≥20% body weight loss. Consistent with its mechanism of action, single oral administration of pemigatinib resulted in a dose-dependent increase in serum phosphate levels in mice.

Pemigatinib inhibited the hERG potassium current with an IC50 >8 µM, which is not clinically achievable at the recommended dose of 13.5 mg. Consistent with this in vitro finding, pemigatinib did not induce QTc prolongation in a single-dose safety pharmacology study in cynomolgus monkeys or in the 13-week repeat-dose study in monkeys. Single oral 32 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

administration of 10 mg/kg pemigatinib resulted in piloerection, increased touch response, and transient effects on respiratory frequency in Sprague-Dawley rats, but did not adversely affect central nervous system (CNS) or respiratory parameters.

There were no unique human metabolites detected in vitro in hepatic microsomes in the presence or absence of metabolic activation and no major metabolites detected in humans at levels ≥10% of the parent that might require further toxicological assessment. Following single oral administration, 14C-pemigatinib-derived radioactivity distributed widely to rat tissues especially excretory, endocrine, alimentary canal, and pigmented tissues, but only minimally in CNS tissues.

The Applicant evaluated the safety of pemigatinib in Sprague-Dawley rats and cynomolgus monkeys in GLP-compliant repeat-dose toxicology studies of up to 3 months duration. Pemigatinib induced multi-organ histologic mononuclear cell infiltration in both species. Many of the pemigatinib-induced findings (e.g. hyperphosphatemia, mineralization, physeal dysplasia, and bone/teeth findings) were attributable to the primary pharmacology of FGFR inhibition. For example, consistent with the role of FGF/FGFR signaling in bone development/diseases and the maintenance of adult bone homeostasis, pemigatinib induced bone toxicity in rats and monkeys at exposures approximately 0.4 and 0.04 times, respectively, those achieved clinically at the recommended human dose of 13.5 mg. Pemigatinib induced physeal and cartilage dysplasia in the sternum, femur, and/or vertebral bone in both species. Additional bone findings in rats included increased remodeling, hyperostosis, and necrosis in the femur; cartilage degeneration, hyperostosis, inflammation, and necrosis in the vertebral bone; and teeth findings including dysplasia (complete loss of ameloblasts), increased bone remodeling, and missing upper incisors. Physeal/cartilage dysplasia in the femur and/or sternum and histologic findings in the teeth did not show evidence of complete recovery. Additional tooth-related findings developed in rats during the recovery period including malaligned, whitened, broken, and trimmed/thinned incisors. The Applicant has not reported drug-related fractures or tooth abnormalities in clinical trials and pemigatinib-induced physeal dysplasia and teeth findings are unlikely to be relevant to human adults with closed growth plates; however, these findings may be relevant in pediatric patients and are included in Section 8.4 of the label.

In the 3-month toxicology study in rats, administration of pemigatinib at the high dose of 1.05 mg/kg (which resulted in exposures approximately 1.5 times the human exposure based on steady-state AUC0-24h of 2620 nM*hour at the clinical dose of 13.5 mg once daily) resulted in mortality due to systemic drug-related toxicity and poor clinical condition (reduced body weight/food consumption and clinical signs including dermal atonia, thinness, unkempt appearance, soft feces/diarrhea, dried and/or wet red/yellow/clear material on various body surfaces, pale extremities, and/or impaired use of hindlimb). The cause of death in one preterm decedent was marked chronic active inflammation in the kidneys likely secondary to uroliths in the urinary bladder; tubule dilation was also present in animals at lower dose levels. Histologic findings in high dose preterm decedents were generally similar to those seen in high dose rats that survived to scheduled necropsy. Another high dose rat exhibited intussusception in the

33 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

colon and jejunum and a distended/gas-filled cecum. Two early mortalities occurred at the low (0.27 mg/kg) dose level; the cause of death was undetermined, but one of the early decedents exhibited impaired use of hindlimb(s) and yellow material on various body surfaces. Treatment with ≥0.27 mg/kg pemigatinib (approximately 0.4 times the exposure based on AUC at the human dose of 13.5 mg) induced hyperphosphatemia in rats, which correlated with firm/white areas in the aorta and vena cava and multi-organ histologic soft tissue and vascular mineralization. Mineralization was still present following 42 days of recovery and is consistent with the role of FGFR in phosphorous metabolism. Increased calcium was also present at the 1.05 mg/kg dose level. Pemigatinib induced hyperphosphatemia/hypophosphatemia and dysregulated calcium in the clinic. Other target organs of pemigatinib toxicity in rats included the eye (bilateral corneal crystals), heart (cardiomyopathy and cartilaginous metaplasia), liver (increased alkaline phosphatase, congestion, and hepatocellular necrosis/vacuolation), lungs (alveolar macrophages), bone marrow in the femur (fibrosis and decreased cellularity), and spinal cord (increased severity of axonal degeneration).

Ocular toxicity, primarily dry eye and serous retinal detachment, has occurred in patients who received pemigatinib. While the Applicant did not record serous retinal detachment in animal studies, there are literature reports supporting a role for basic fibroblast growth factor and FGFR/MAPK signaling in protecting/maintaining retinal pigment epithelial cells (Faktorovich, Steinberg, et al. 1990 and 1992; Guillonneau, Bryckaert, et al. 1998; Rosenthal, Malek, et al. 2005; and Van der Noll, Leijen, et al. 2013). FGF and/or FGFR play a key role in lens, corneal, and retina development and adult function (Rosenthal, Malek, et al. 2005; Zhang, Upadhya, et al. 2015; and Robinson 2006). All four FGFR genes are expressed in the lens; FGFR1 and FGFR2 are also expressed in the retina and cornea. Ophthalmologic findings in the 28-day monkey study included moderate lens opacities and slight attenuation of retinal vessels at pemigatinib dose levels ≥0.33 mg/kg and 1 mg/kg, respectively. Intussusception was not seen in clinical trials.

In the 13-week toxicology study in cynomolgus monkeys, there was no mortality at doses up to the high dose level, 1 mg/kg. Treatment with ≥0.1 mg/kg pemigatinib resulted in dry skin, which was also observed in clinical studies. Treatment with 1 mg/kg pemigatinib (approximately 0.5 times the exposure based on AUC at the human dose of 13.5 mg) resulted in increased phosphorous and calcium, which correlated histologically with mineralization in the kidney and showed evidence of recovery. Increased ALT/AST occurred at the high dose level, but did not correlate with any histologic findings in the liver. Pemigatinib also induced mild histologic edema in the epididymis. In a non-GLP 10-day repeat-dose toxicology study in cynomolgus monkeys, at a dose level of 3 mg/kg (approximately 3.8 times the exposure based on AUC at the human dose of 13.5 mg) animals died beginning on Day 7. The preterm decedents exhibited body weight loss, hyperthermia, facial swelling, and decreased activity. The cause of death was not identified; however, pemigatinib induced multi-organ mineralization in these animals and histologic target organs included the kidney, pancreas, stomach, bone marrow, thymus, lymph node, and lung.

34 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The Applicant did not conduct carcinogenicity or fertility studies, and these studies are not needed to support a marketing application for a drug intended to treat patients with advanced cancer. Dose-related findings in reproductive organs in the general toxicology studies included inflammation in the prostate (rats) and edema in the epididymis (monkeys). Mineralization was also observed in the ovaries of rats and monkeys but was not clearly dose-related except for in the 10-day toxicity study in monkeys. Overall, data from the general toxicology studies do not suggest that pemigatinib will substantially impair male or female fertility. Pemigatinib was not mutagenic in an in vitro bacterial reverse mutation (Ames) assay or clastogenic in either the in vitro chromosomal aberration assay in human peripheral blood lymphocytes or the in vivo bone marrow micronucleus assay in rats. In addition, pemigatinib did not demonstrate phototoxic potential in an in vitro neutral red uptake phototoxicity assay.

The Applicant conducted a non-GLP dose range-finding embryo-fetal development study in Sprague-Dawley rats. Because this study demonstrated clear findings at exposure levels below the clinical exposure at the 13.5 mg human dose, this study was sufficient to characterize the potential developmental risk. Once daily oral administration of pemigatinib to pregnant rats during the period of organogenesis (gestation Days 6 to 17) resulted in reduced maternal body weight and food consumption at dose levels ≥0.3 mg/kg (approximately 0.6 times the exposure based on AUC at the human dose of 13.5 mg) accompanied by 100% post-implantation loss (early resorptions). The reduced maternal body weight appeared to be a result of fetal loss rather than maternal toxicity. At doses ≥0.1 mg/kg (approximately 0.2 times the exposure based on AUC at the human dose of 13.5 mg) there was reduced gravid uterine weight as well as a decrease in mean fetal body weight and an increase in fetal visceral (retroesophageal aortic arch) and skeletal (vertebral anomaly, only 12 pairs of ribs present) malformations, major blood vessel variations, and skeletal variations (reduced ossifications, 27 presacral vertebrae, 7th cervical rib, bent ribs). Thus, pemigatinib was teratogenic at exposures lower than those achieved in humans at the recommended dose. As a result, the pharmacology/toxicology review team recommends a warning for embryo-fetal toxicity in the US Prescribing Information (USPI) for PEMAZYRE. Consistent with the recommendations for teratogenic or embryo-lethal but non-genotoxic drugs described in the FDA Guidance for Industry Oncology Pharmaceuticals: Reproductive Toxicity Testing and Labeling Recommendations, and considering the pemigatinib half-life of 15.4 hours, FDA recommends advising males and females of reproductive potential to use effective contraception during treatment with PEMAZYRE and for at least one week after the final dose. The Applicant did not evaluate the presence of pemigatinib in milk; however, because of potential adverse effects of PEMAZYRE on a breastfeeding child, the review team recommends advising patients not to breastfeed during treatment with PEMAZYRE and for 1 week after the final dose. There are no outstanding issues from a pharmacology/toxicology perspective that would prevent the approval of PEMAZYRE for the treatment of patients with previously treated, locally advanced or metastatic cholangiocarcinoma with a FGFR2 fusion or rearrangement as detected by an FDA-approved test. Referenced NDAs, BLAs, DMFs

None 35 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Pharmacology

Primary pharmacology

A. In Vitro Studies

The Applicant investigated whether 100 nM pemigatinib (INCB054828) inhibits 192 kinases in (b) (4) an in vitro Discovery Services Kinase Assay screen (Study #T13-05-06). Incubation with 100 nM pemigatinib inhibited FGFR1, FGFR2, FGFR2 N549H, FGFR3, FGFR3 K650E, and FGFR4 by 99%, 98%, 90%, 98%, 84% and 77%, respectively, but did not inhibit any of the other kinases tested by ≥50%. Of the non-FGFR kinases examined, pemigatinib showed the greatest inhibition of KDR (VEGFR2) and FLT4 (VEGFR3), with but with only 45% and 38% inhibition, respectively. In a second set of in vitro kinase profiling assays (Study #ATG-14.03.1) investigating 56 non-FGFR kinases, pemigatinib did not inhibit the majority of the kinases tested (IC50s ≥ 11201 nM). Pemigatinib inhibited VEGFR2, KIT, and PDGFRβ with IC50 values of 182, 266, and 1787 nM, respectively, suggesting some specificity for FGFR (see Table 4 for FGFR IC50 values).

The Applicant characterized the in vitro activity and selectivity of pemigatinib in biochemical and cellular assays (Study #IN VITRO-14.04.2). In an enzyme activity assay, pemigatinib inhibited the kinase activity of recombinant human FGFR1, FGFR2, and FGFR3 at IC50 concentrations (Table 4) lower than those achieved clinically at the 13.5 mg dose level of pemigatinib (free Cmax at the 13.5 mg clinical dose is approximately 22 nM based on ~90.6% protein binding), but inhibited recombinant human FGFR4 and VEGFR2 with IC50 values ≥30 nM. Pemigatinib inhibited recombinant human and rat FGFR1 and FGFR2 with similar affinity. Additional kinetic studies suggested that pemigatinib is an ATP competitive and reversible inhibitor of FGFR1 (data not shown).

Table 4: In Vitro Inhibition of Human and Rat FGFR Kinases and VEGFR2

(Applicant Table reproduced from Study #IN VITRO-14.04.2)

The Applicant evaluated the effects of pemigatinib on FGFR signaling in cellular assays. As assessed by enzyme-linked immunoassay (ELISA), incubation with pemigatinib for 4 hours inhibited FGFR2 phosphorylation in FGFR2-amplified KATOIII human gastric cancer cells with an

36 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

IC50 of 3.1 nM. As assessed by a bead based immunoassay, incubation with pemigatinib for 1 hour inhibited FGFR phosphorylation in Ba/F3 cells stably expressing TEL-FGFR1 or TEL-FGFR3 fusion proteins with IC50 values of 3.2 nM and 3.7 nM, respectively. Further, as assessed by immunoblotting, incubation with pemigatinib for 2 hours reduced the phosphorylation of FGFR1 (Y653/Y654) and the downstream signaling proteins ERK1/2 and STAT5 with an IC50 of ~2 nM in KG-1A cells expressing the constitutively active FGFR1OP2-FGFR1 fusion protein, indicating that pemigatinib inhibits FGFR1 signaling in vitro (data not shown). To evaluate the effect of pemigatinib on in vitro cell proliferation, investigators incubated cancer cells expressing activating FGFR alterations with pemigatinib for 3 days and then measured cell (b) (4) viability using either an MTS assay or an ATP assay (b) (4) As shown in Table 5, pemigatinib inhibited the in vitro viability of cancer cells with FGFR1 or FGFR2 amplification, FGFR1 or FGFR3 fusions, FGFR3 translocations, or FGFR2 N310R/N549K mutations at clinically relevant concentrations, but did not substantially inhibit the viability of cells without FGFR alterations including human T cells (IC50 values >1500 nM). Pemigatinib also inhibited the in vitro viability of Ba/F3 cells expressing the FGFR2-CCDC6 and FGFR2-AHCYL fusions identified in patients with cholangiocarcinoma with mean IC50 values of ~1.2 nM (data not shown in table).

Table 5: Effect of Pemigatinib on the In Vitro Viability of Cells with or without FGFR Alterations

(Applicant Table reproduced from Study #IN VITRO-14.04.2)

37 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The Applicant next evaluated the selectivity of pemigatinib for FGFR versus VEGFR in human umbilical vein endothelial cells (HUVEC) incubated for 3 days with 25 ng/mL recombinant (b) (4) human VEGF121 or 0.8 ng/mL recombinant human basic FGF (b-FGF). As assessed by (b) (4) , pemigatinib inhibited VEGF- and FGF-induced HUVEC proliferation with IC50 values of 801 nM and 7.9 nM, respectively, indicating that pemigatinib was ~100-fold more selective for FGFR than VEGFR. In addition, pemigatinib inhibited the phosphorylation of FGFR2 in KATOIII cells spiked with human whole blood with an IC50 value of 10.9 ± 3.6 nM, suggesting pemigatinib inhibits FGFR2 in the presence of human serum proteins.

B. In Vivo Studies The Applicant investigated the in vivo anti-tumor activity of pemigatinib in female SCID mice (8/group) subcutaneously (SC) implanted with FGFR2-amplified KATO III human gastric cancer xenografts (Study #PRECLIN-14.06.1). When the mean tumor volume was ~206-333 mm3, mice began receiving vehicle or pemigatinib. Across four experiments, once daily oral administration of ≥0.3 mg/kg pemigatinib resulted in ≥50% tumor growth inhibition (TGI) compared to vehicle control. Pemigatinib induced >20% body weight loss at dose levels ≥10 mg/kg; however, pemigatinib was generally tolerated at dose levels ≤3 mg/kg except for one mouse at the 3 mg/kg dose level. Maximal in vivo anti-tumor activity was seen when pemigatinib concentrations covered the IC50 for inhibition of FGFR2 phosphorylation in KATOIII cells spiked with human whole blood (10.9 nM) for at least 8 hours.

Figure 1: Effect of Pemigatinib on In Vivo Anti-Tumor Activity in Mice Bearing KATOIII FGFR2 Amplified Xenografts

APPEARS THIS WAY ON ORIGINAL

(Representative Applicant Figure reproduced from Study #PRECLIN-14.06.1; Experiment #YZ.CAN070913B)

Table 6: Summary of In Vivo Anti-Tumor Activity in Mice Bearing FGFR2 Amplified KATOIII Xenografts

Substudy Route and frequency Dose %TGI Effect on Body Weight of administration (mg/kg) YZ.CAN053113 Oral once daily for 7-8 1 mg/kg 72% >20% body weight loss in 4/8 days 3 mg/kg 72% mice dosed with 30 mg/kg 10 mg/kg 81% 30 mg/kg NT YZ.CAN061113 Oral once daily for 7-10 0.3 mg/kg 63% >20% body weight loss in days 1 mg/kg 72% most mice dosed with 10 10 mg/kg NT mg/kg

38 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

YZ.CAN070913B Oral once daily for 11 0.03 mg/kg 26% >20% body weight loss in 1/8 days 0.1 mg/kg 44% mice dosed with 3 mg/kg, but 0.3 mg/kg 50% dose level was considered 1 mg/kg 56% tolerated 3 mg/kg 59% TGI = Tumor growth inhibition; NT = Not tolerated; (Reviewer generated table based on results from Study # PRECLIN-14.06.1)

Further, single oral administration of pemigatinib resulted in a dose-dependent inhibition of FGFR2 phosphorylation in KATOIII xenografts compared to vehicle beginning at 4 hours post- dose. The mean IC50 (n=4) for inhibition of in vivo FGFR2 phosphorylation was 22.6 nM.

Figure 2: Effect of Single Oral Administration of Pemigatinib on FGFR2 Phosphorylation in KATO III Xenografts

(Applicant Figure reproduced from Study #PRECLIN-14.06.1; Data from two representative experiments)

The Applicant investigated the in vivo anti-tumor activity of pemigatinib in 12 female nude mice/group bearing established SC low-passage patient-derived cholangiocarcinoma xenografts (CTG-0997 and CTG-0011; Study #PRECLIN-17.18.1). CTG-0997 xenografts express a FGFR2­ TRA2B fusion, whereas CTG-0011 xenografts do not express an FGFR2 mutation or fusion and were included as a negative control. As expected, once daily oral administration of 0.3 or 1 mg/kg pemigatinib for 29 days did not substantially inhibit CTG-0011 xenograft growth (data not shown). In contrast, once daily oral administration of 0.3 and 1 mg/kg pemigatinib for 42 days resulted in 88% and 153% inhibition of CTG-0997 xenograft growth compared to vehicle, respectively, without inducing significant toxicity based on mortality and body weight loss.

39 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 3: Effect of Pemigatinib on CTG-0997 Xenograft Growth in Nude Mice

TGI at 1 mg/kg was statistically significant compared to vehicle; (Applicant Figure reproduced from Study #PRECLIN-17.18.1)

The Applicant investigated the in vivo anti-tumor activity of pemigatinib in 8 female RNU rats/group subcutaneously implanted with RT112 human bladder cancer xenografts expressing the FGFR3-TACC3 translocation (Study #PRECLIN-14.07.1). Once daily oral administration of 0.3 and 1 mg/kg pemigatinib resulted in 62% and 66% TGI compared to vehicle, respectively, without inducing significant body weight loss (data not shown). Pemigatinib induced ≥20% body weight loss at dose levels ≥3 mg/kg, so TGI was not calculated for these dose levels. Combined exposure/activity analysis for RT112 (FGFR3-TACC3), H1581 (FGFR1 amplification), and KATOIII (FGFR2 amplification) xenografts demonstrated that maximal anti-tumor activity was achieved when pemigatinib concentrations were ≥10.9 nM for at least 12 hours (data not shown). Consistent with inhibition of FGFR1 in in vitro studies, pemigatinib (0.3 mg/kg/day) also exhibited in vivo anti-tumor activity against established SC KG1 human erythroleukemia AML xenografts expressing a FGFROP2-FGFR1 fusion in female humanized NOD scid gamma (NSG) mice (Study #PRECLIN-18.29.1; data not shown).

Since FGFR inhibition results in increased circulating phosphate levels by inhibiting FGF23 signaling in the kidney (Wohrle, Bonny, et al. 2011), the Applicant assessed the effects of pemigatinib on inorganic phosphate levels in 4 female C57BL/6 mice/group (Study #PRECLIN­ 14.08.1). Single oral administration of pemigatinib resulted in a dose-dependent increase in serum phosphate levels up to 98% compared to vehicle at 24 hours post-dose, which reached statistical significance at dose levels ≥0.37 mg/kg (Figure 4). Similar results were seen in a second experiment (data not shown). The mean ED50 across two experiments was 2.95 mg/kg pemigatinib.

40 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 4: Effect of Single Oral Administration of Pemigatinib on Serum Phosphate in Mice

(Applicant Figure reproduced from representative experiment from Study #PRECLIN-14.08.1; ED50 = 3.8 mg/kg)

Secondary Pharmacology The Applicant evaluated off-target activity of pemigatinib on 70 receptors, enzymes, and ion channels in Study #T13-05-07 using in vitro receptor binding and enzyme assays. Incubation with 0.1 or 1 µM pemigatinib did not result in >50% inhibition of any of the 70 receptors, enzymes, or ion channels tested.

Safety Pharmacology In GLP-compliant Study # T14-04-09, human embryonic kidney (HEK293) cells stably expressing the human hERG potassium channel were incubated with pemigatinib (3 and 8 µM; higher concentrations not tested due to solubility limitations), HEPES-buffered physiological saline + 0.3% DMSO (negative control), or 60 nM terfenadine (positive control) followed by measurement of potassium current using a whole cell patch-clamp technique. Negative and positive controls behaved as expected. Incubation with 3 µM and 8 µM pemigatinib inhibited hERG-mediated potassium current by 5.8% and 14.1%, respectively, with an IC50 value >8 µM (n=3), suggesting minimal risk of QT prolongation through this pathway in humans given pemigatinib.

In GLP-compliant Study #T16-02-07, four conscious radiotelemetry-instrumented male cynomolgus monkeys received single oral doses of pemigatinib at dose levels of 0, 0.33, 1, or 5 mg/kg using a latin square design with a 7 day washout period between doses; heart rate, arterial blood pressure (systolic, diastolic, calculated mean, and pulse pressure), body temperature, and ECG parameters (PR, QRS, RR, QT, and QTcB), were measured continuously for approximately 1 hour pre-dosing through 24 hours post-dosing. Consistent with the in vitro hERG data, treatment with pemigatinib at dose levels up to 5 mg/kg did not result in QT or QTcB prolongation in monkeys. Single oral administration of pemigatinib at dose levels up to 5 mg/kg did not result in drug-related effects on cardiovascular parameters.

41 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

In order to assess the effects of pemigatinib on the respiratory system, 8 Sprague-Dawley rats/sex/group placed in single-chamber, head-out, neck-sealed plethysmographs received single oral doses of pemigatinib at dose levels of 0, 0.5, 1.5, or 10 mg/kg; respiratory frequency, tidal volume, and minute volume were measured for at least 60 minutes pre-dosing and continuously for 6 hours (360 minutes) post-dose (Study # T16-02-06; GLP-compliant). There was a transient statistically significant 10.6% decrease in mean respiratory frequency at 0-15 minutes post-dose in rats dosed with ≥1.5 mg/kg pemigatinib compared to controls, followed by a statistically significant 14.6% increase in respiratory frequency at 316-330 minutes post- dose in high dose rats. Single oral administration of pemigatinib at doses up to 10 mg/kg did not significantly affect mean tidal volume or minute volume. Pulmonary toxicity has not been observed clinically with pemigatinib.

In GLP-compliant Study #T16-02-05, 6 Sprague-Dawley rats/sex/group received single oral doses of pemigatinib at dose levels of 0, 0.5, 1.5, or 10 mg/kg; observations for a modified Irwin test were performed pre-dosing and approximately 60, 120, 240, and 360 minutes post-dose to evaluate the effects of pemigatinib on the CNS. Investigators assessed gross behavior as well as physiological (including body temperature) and neurological states. Piloerection occurred in one high dose male rat and increased touch response occurred in two high dose rats. Single oral administration of pemigatinib at dose levels up to 10 mg/kg did not adversely affect any other CNS parameters.

ADME/PK

Type of Study Major Findings Absorption The Pharmacokinetics of INCB054828 in Male Sprague-Dawley (SD) rats received a single oral (2 Rats / DMB-14.63.1 mg/kg) or intravenous (IV; 1 mg/kg) dose of the free base of pemigatinib. Pemigatinib demonstrated >100% oral bioavailability.

Oral IV 2 mg/kg 1 mg/kg Cmax (µM) 2.26 NC Tmax (hr) 0.56 NC AUC0-∞ (µM·hr) 7.67 2.03 Half-life (hr) 2.2 4 %F >100% NC CL (L/hr/kg) NC 1.03 Renal CL (L/hr/kg) 0.0875 Dose excreted in NC 9.02 urine (%) Vss (L/kg) NC 1.85 NC = Not calculated; F = Bioavailability; hr = Hour; CL = Clearance; Vss = Volume of distribution

42 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Type of Study Major Findings The Intravenous and Oral Male cynomolgus monkeys received a single oral (2 mg/kg) Pharmacokinetics of INCB054828 in or IV (1 mg/kg) dose of the free base of pemigatinib. There Cynomolgus Monkeys / DMB-14.66.1 was moderate renal clearance.

Oral IV 2 mg/kg 1 mg/kg Cmax (µM) 0.766 NC Tmax (hr) 2 NC AUC0-∞ (µM·hr) 6.19 10.6 Half-life (hr) 5.3 10.3 %F 29 NC CL (L/hr/kg) NC 0.198 Renal CL (L/hr/kg) NC 0.0736 Dose excreted in NC 36 urine (%) Vss (L/kg) NC 0.584 NC = Not calculated; F = Bioavailability; hr = Hour; CL = Clearance; Vss = Volume of distribution

Distribution In Vitro and Ex Vivo Protein Binding of Pemigatinib was 92.3%, 96.8%, and 89.6% protein bound to INCB054828 / DMB-14.71.1 human, SD rat, and cynomolgus monkey serum in vitro, respectively. Pemigatinib was 88.8%, 96.8%, 91.1% protein bound to human, SD rat, and cynomolgus monkey plasma in vitro, respectively. Taking into account plasma and serum data, pemigatinib was ~90.6% bound to human protein in vitro. Using plasma obtained from rats and monkeys, pemigatinib was 96% and 91.4% bound to rat and monkey protein ex vivo, respectively. INCB054828 Brain and Cerebrospinal Four hours after a single IV infusion of 2 mg/kg pemigatinib Fluid Concentrations in Rats / DMB­ to male SD, the total brain concentration was 9% of the 14.64.1 corresponding total plasma concentration, and the cerebrospinal fluid concentration was 13% of the corresponding estimated unbound plasma concentration, suggesting pemigatinib crossed the blood-brain barrier at low levels after IV administration. Distribution of Orally Administered 14C­ Following single oral administration of 10 mg/kg 14C­ INCB054828 in the Urinary Bladder of INCB054828, 14C-pemigatinib-derived radioactivity was Male Sprague-Dawley Rats Using distributed to all tissues of the rat urinary bladder at 1, 2, 4, Microautoradiography and and 8 hrs post-dose, with the highest concentrations in the Quantitative Autoradioluminography / lamina propria and urothelium. DMB-15.106.1

43 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Type of Study Major Findings Metabolism Identification of In Vitro Metabolites of There were no unique human metabolites or any human INCB054828 / DMB-13.49.1 metabolites present at >10% the parent compound.

Relative Abundance of Metabolites in Hepatic Microsomes Metabolite Rat Dog Monkey Human M1 0.3 0.6 3.0 0.1 M2 2.5 1.0 7.6 4.1 M3 0.7 11.3 0.8 0.1 M4 2.5 12.8 23.6 4.1 Pemigatinib 100 100 100 100 M5 0.2 N/F 0.6 0.1 M6 30.3 640.9 12.2 4.3 N/F = Not found; Presented as % of Integrated Parent Peak Area

Relative Abundance of Metabolites in Hepatic S-9 Fraction Metabolite Rat Dog Monkey Human M1 0.1 N/F 0.2 N/F M2 2.6 1.0 3.4 2.2 M3 0.1 0.5 N/F N/F M4 1.1 1.4 2.8 0.7 Pemigatinib 100 100 100 100 M5 0.1 N/F 1.7 0.1 M6 5.7 38.9 1.2 1.7 N/F = Not found; Presented as % of Integrated Parent Peak Area

Incubation with pemigatinib in hepatic microsomes in the presence of glutathione did not reveal any glutathione­ related conjugates in rat or human hepatic microsomes. Identification of the In Vivo Pemigatinib underwent oxidative metabolism following Metabolites of INCB054828 in Male oral administration in rats. M2 (putative O-demethylation and Female Sprague-Dawley Rats product) was present at ≥10% (1.4-18.6%) of parent Following Single and Multiple Oral pemigatinib in plasma. M2, M4, M5, and M6 were present Dosing / DMB-14.75.1 at ≥10% of parent pemigatinib in urine; M6 (putative single hydroxylation of the morpholine moiety of pemigatinib) was the most abundant. Identification of the In Vivo Pemigatinib underwent oxidative metabolism following Metabolites of INCB054828 in Male oral administration in monkeys. M5 (putative bis and Female Cynomolgus Monkeys hydroxylation of the morpholine moiety of pemigatinib) Following Single and Multiple Oral was present at ≥10% (4-11.4%) of parent pemigatinib in Dosing / DMB-14.74.2 plasma. M2, M4, M5, and M6 were present at ≥10% of parent pemigatinib in urine; M5 was the most abundant.

There were no major metabolites detected in human plasma at levels ≥10% of parent pemigatinib. Overall, considering all the species tested, there were no human- specific metabolites and there is adequate toxicological coverage of all metabolites. Excretion

44 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Type of Study Major Findings Quantitative Whole-Body Following a single oral administration of 10 mg/kg (250 Autoradiography and Excretion in Long- µCi/kg) 14C-pemigatinib, 14C-pemigatinib-derived Evans and Sprague-Dawley Rats radioactivity was slightly higher in plasma compared to Following a Single Oral Dose of blood at time points up to 4-8 hrs post-dose, but [14C]INCB054828 / DBM-16.10.1 partitioned into the cellular portion of blood at >8 hrs post- dose. 14C-pemigatinib-derived radioactivity was widely distributed to tissues with maximum concentrations generally observed at 1 hr post-dose. In non-pigmented SD rats, distribution of 14C-pemigatinib was highest in the stomach, liver, adrenal cortex, small intestine, adrenal medulla, Harderian gland, aorta, brown adipose, and kidney cortex/medulla. In pigmented Longs-Evans (LE) rats, distribution of 14C-pemigatinib was highest in the eye uvea, stomach, liver, pigmented skin, adrenal cortex, small intestine, adrenal medulla, Harderian gland, cecum, aorta, kidney cortex/medulla, brown adipose, pituitary gland, and urinary bladder. Tissue distribution was similar in SD and LE rats (excretory, endocrine, and alimentary canal tissues) except 14C-pemigatinib-derived radioactivity associated with melanin in pigmented tissues of LE rats. 14C­ pemigatinib-derived radioactivity was eliminated via biliary and renal excretion in SD and LE rats. Radioactivity was minimally detected in CNS tissues. Identification of the In Vivo Following a single oral administration of 10 mg/kg (250 Metabolites of INCB054828 in Rats µCi/kg) 14C-pemigatinib, parent pemigatinib accounted for Following a Single Oral Dose of 14C­ 37.6-66.4% of the total radioactivity in plasma. M10 INCB054828 / DMB-18.82.1 (hydroxylation of the dihydropyrimidinone moiety of pemigatinib) was present at ≥10% of parent pemigatinib in plasma (8.3-22% of total radioactivity). Approximately 3.76­ 5.98% of the total radioactive dose was recovered in urine. Parent pemigatinib in urine accounted for <1% of the administered dose; M4 and M6 were detected in urine at ≥1% of the administered dose. Feces was the major route of excretion of 14C-pemigatinib-derived radioactivity (~73.76-85.73% of the total radioactive dose). Parent pemigatinib and M2 (INCB056632, O-desmethyl­ INCB054828) in feces accounted for 3.1-5.7% and 16.1­ 21.6% of the administered dose, respectively. M4, M13, and M26 were detected in feces at ≥2% of the administered dose. SD = Sprague-Dawley; Hrs = Hours

Toxicology

General Toxicology

Study title/ number: INCB054828: A 3-Month Oral Gavage Toxicity and Toxicokinetic Study with a 42-Day Recovery Period in Sprague-Dawley Rats / T15-12-01 45 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Key Study Findings • There was mortality at the 1.05 mg/kg dose level potentially due to systemic drug-related toxicity. Preterm decedents exhibited dermal atonia, thinness, unkempt appearance, soft feces/diarrhea, body weight loss, impaired use of hindlimb, piloerection, and/or hypoactivity. • Pemigatinib induced hyperphosphatemia, increased calcium, and multi-organ histologic soft tissue and vascular mineralization • Target organs included the bone (physeal and articular cartilage dysplasia), teeth, eye, GI tract, heart, liver, kidney, lungs, bone marrow in the femur, and spinal cord

(b) (4) Conducting laboratory and location:

GLP compliance: Yes

Methods Dose and frequency of dosing: 0, 0.27, 0.54, and 1.05 mg/kg once daily for up to 92 days Route of administration: Oral gavage Formulation/Vehicle: 0.5% methylcellulose (400 cps) in 50 mM citrate buffer, pH 3.0 ± 0.3 Species/Strain: Rat / Sprague-Dawley Number/Sex/Group: 10/sex/group (main) 5/sex/group (42-day recovery period) Age: ~7 weeks old Satellite groups/ unique design: Toxicokinetic: 6/sex in the control group; 10/sex in treated groups Deviation from study protocol No affecting interpretation of results:

Observations and Results: changes from control

Parameters Major findings Mortality There were 14 mortalities (8 main study rats and 6 TK rats), including 2 at the LD (0.27 mg/kg) (1 main and 1 TK), 1 at the MD (0.54 mg/kg), and 11 at the HD (1.05 mg/kg) (6 main and 5 TK) on Days 23-106. The MD preterm decedent (#9422; Day 46) and one HD preterm decedent (#9308; Day 51) exhibited moderate to marked acute inflammation in the lungs, and their deaths were considered secondary to procedural gavage error. The cause of death of the LD preterm decedents [#9226 (Day 106) and #9399 (Day 74, TK)] was undetermined, although #9399 (euthanized on Day 74) exhibited impaired use of hindlimb(s) and yellow material on various body surfaces. Histologic findings in HD main study preterm decedents were generally similar to those at the Week 8

46 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

necropsy and included physeal dysplasia, bone remodeling in the femur, and multi-organ mineralization. With the exception of rat #9322, the cause of death of the remaining HD preterm decedents was undetermined but likely due to systemic drug-related toxicity. Dried/wet red/yellow/clear material on various body surfaces was also observed in preterm decedents.

1.05 mg/kg M/ 9322 (Found dead on Day 23; 1.05 mg/kg): Thin, dermal atonia; moderate fibrosis in marrow, femur; moderate pelvis dilation, marked chronic active inflammation, moderate infarct, and moderate tubule dilation in the kidney; marked mineralization in glandular stomach; thickened/calculus (I), marked urothelium hyperplasia, and minimal subepithelial mineralization in the urinary bladder; distended ureters with yellow contents. Cause of death was marked chronic active inflammation in the kidneys likely secondary to uroliths in the urinary bladder. M / 9257 (Euthanized on Day 35): Unkempt appearance, thin, soft feces, salivation; distended/gas filled cecum, ileum, jejunum, and stomach; moderate fibrosis in marrow, femur F / 9345 (Euthanized on Day 35): Unkempt appearance, thin; moderate fibrosis in marrow, femur F / 9348 (Euthanized on Day 35): Unkempt appearance, diarrhea, salivation, thin; distended/gas filled colon and ileum F / 9405 (Euthanized on Day 82): Impaired use of right hindlimb, thin

Five HD TK rats (#9312, 9313, 9296, 9340,and 9328) exhibiting hypoactivity, prostration, impaired use of hindlimb(s), piloerection, dermal atonia, thin appearance, increased respiration rate, dilated pupils, body weight loss, red material on various body surfaces, swollen abdomen, unkempt appearance, and/or red material on various body surfaces were euthanized in moribund condition or found dead on Days 37-56. One HD TK preterm decedent (#9313) also exhibited dark red areas on lungs. Toxicity Mitigation Due to mortality and poor clinical condition, dosing was discontinued in the HD group during Week 8. Seven to eight HD rats/sex in the main group and all surviving HD TK rats were euthanized during Week 8 and the remaining 4-5 HD rats/sex were allocated to an early 42-day recovery period.

47 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Clinical Signs See clinical signs in preterm decedents above. There was a dose-related increase in the incidence of thinness, unkempt appearance, and dried and/or wet red/yellow/clear material around the eyes, nose, ventral neck, mouth, forelimb/hindlimb, and/or urogenital/anogenital area at doses ≥0.54 mg/kg.

0.54 mg/kg: Missing upper incisors, broken tail 1.05 mg/kg: Pale extremities, dermal atonia, rectal mucous exudate, dried brown material anogenital area, missing upper incisors

Recovery: Following 42 days of recovery, red/yellow/clear material was still seen around the eyes, nose, mouth, forelimbs, and urogenital area in some rats dosed with ≥0.54 mg/kg. New teeth-related findings developed during the recovery period including malaligned upper incisors at 0.54 mg/kg, whitened lower/upper incisors at dose levels ≥0.27 mg/kg, broken upper incisors at dose levels ≥0.54 mg/kg, and trimmed/thinned lower and upper incisors at dose levels ≥0.54 mg/kg.

Additional findings in the recovery cohort included: 0.54 mg/kg: Tail broken, reddened forelimb/hindlimb, toe nails long/trimmed, thin, unkempt appearance 1.05 mg/kg: Missing upper incisors, toe nails long/trimmed Body Weights and Feed 1.05 mg/kg males: Statistically significant ≤14% decrease in body Consumption weight beginning at Week 4, which correlated with a ≤60% decrease in food consumption. Decreased body weight did not show evidence of recovery and was 24% lower than controls prior to recovery necropsy. 1.05 mg/kg females: Statistically significant ≤10% decrease in body weight during Weeks 8-9, which correlated with a ≤70% decrease in food consumption and showed evidence of recovery Ophthalmoscopy There was an increase in the incidence of bilateral corneal crystals during Week 4 in female rats at the 1.05 mg/kg dose level and during Week 7 in male and female rats at dose levels ≥0.27 mg/kg and 1.05 mg/kg, respectively, compared to controls, which showed evidence of recovery. One 0.54 mg/kg female rat exhibited unilateral conjunctivitis during Week 18.

Hematology Findings included increases in prothrombin time (PT) and neutrophils at dose levels ≥0.54 mg/kg and increased platelets at 1.05 mg/kg. Increased PT did not recover at 1.05 mg/kg.

48 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Hematology: % Change from Concurrent Control (3-month Study; Rats) Dose (mg/kg) 0.27 0.54 1.05 Sex M F M F M^ F^ Prothrombin Time 6%↑* 4%↑ 6%↑ 3%↑ 6%↑R 1%↓R 16%↑R 11%↑R Reticulocytes 22%↓ 24%↓ 14%↓ 17%↓ 24%↑ 78%↑ 7%↓R 1%↑R 17%↓R 19%↓R 26%↓R 19%↓R Neutrophils 30%↑ 17%↓ 35%↑ 82%↑* 52%↑ 151%↑ 40%↑R 17%↓R 30%↑R 169%↑R 4%↓R 0%↑R Platelets 9%↑ 6%↑ 6%↑ 9%↓ 31%↑ 30%↑ 4%↑R 13%↓R 2%↓R 1%↓R 6%↑R 16%↑R Mean platelet 9%↑ 6%↑ volume 10%↑R 12%↑R % relative to controls during Week 8 (HD) or Week 13; *, P ≤ 0.05; R = denotes Recovery group [Week 14 (HD) or Week 19]; ^ = % change compared to control on Week 13 or Recovery Week 19

Clinical Chemistry Findings included increased phosphorous, calcium, triglycerides, ALP, and ALT and decreased albumin, A/G ratio, total protein, glucose, and chloride compared to controls.

Clinical Chemistry: % Change from Concurrent Control (3-month Study; Rats) Dose (mg/kg) 0.27 0.54 1.05 Sex M F M F M^ F^ Phosphorous W4 6%↑ 9%↑* 17%↑** 27%↑** 38%↑** W7 3%↑ 5%↑ 10%↑* 16%↑** 19%↑** 43%↑** W13/14 5%↑ 18%↑* 12%↑** 22%↑* 5%↓R 15%↑R W19 18%↓R* 10%↑R 11%↓R 46%↑R Albumin W4 5%↓* 7%↓** 9%↓** W7 4%↓ 5%↓** 9%↓** 5%↓** 16%↓** W13/14 5%↓ 3%↓ 7%↓* 2%↓R 6%↓R W19 6%↓R 12%↓R** 4%↓R Calcium W4 4%↑* 4%↑** W7 1%↑ 2%↑ W13/14 1%↑R 4%↑R W19 5%↓R* 6%↓R** Triglycerides W4 23%↑ 18%↑ 11%↑ 3%↑ 55%↑** 18%↑ W7 8%↑ 7%↑ 15%↑ 4%↑ 77%↑** 59%↑** W13/14 2%↑ 9%↑ 17%↑ 15%↑ 43%↓R 8%↓R W19 24%↓R 0%↑R 31%↓R 6%↑R A/G ratio W4 9%↓** W7 4%↓ 6%↓* 6%↓* 15%↓** W13/14 2%↓ 10%↓* 1%↑R 8%↓R W19 10%↓R** 1%↓R Glucose W4 6%↓* W7 8%↓ 8%↓ 10%↓* 15%↓** W13/14 6%↓ 6%↓ 5%↓ 19%↑R 5%↑R W19 2%↑R 1%↓R 7%↑R

49 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dose (mg/kg) 0.27 0.54 1.05 Sex M F M F M^ F^ Chloride W4 3%↓** W7 2%↓** W13/14 3%↓* 2%↓ 1%↑R W19 0%↓R 1%↓R ALP W4 8%↑ 25%↑* 13%↑ 18%↑ 29%↑** 38%↑** W7 7%↓ 17%↑ 0.5%↑ 6%↑ 9%↑ 21%↑ W13/14 6%↑ 8%↑ 8%↑ 5%↑ 17%↑R 33%↑R W19 0%↑R 6%↑R 1%↓R 24%↓R % relative to controls during Week 8 (HD) or Week 13; *, P ≤ 0.05; **, P ≤ 0.01; W = Week; R = denotes Recovery group [Week 14 (HD) or Week 19]; ^ = % change for Week 14 compared to control on Week 19

Urinalysis Increase in mean urine volume up to 136% during Week 8 at the 1.05 mg/kg dose level, which showed evidence of recovery Gross Pathology Firm/white area(s) in the aorta and vena cava correlated histologically with mineralization. Soft, nodules and/or white discoloration in the bone correlated histologically with physeal dysplasia. Thin/white discoloration in the teeth correlated histologically with dysplasia.

Selected Gross Pathology Findings (3-month Study; Rats) Dose (mg/kg) 0 0.27 0.54 1.05 Sex M F M F M F M F Aorta Firm 1 White area(s) 1 Cecum Distended 2 3 Colon Intussusception 1 Duodenum Distended 1 1 Ileum Distended 2 1 Jejunum Distended 2 1 Intussusception 1 Liver Dark red discoloration 1 Pale 1 Vena cava Firm 1 White area(s) 1 Bone Nodule(s) 2 Soft 3 Thick white contents 3 White discoloration 1 Mandibular Enlarged 4,2R lymph node Dark red discoloration 1R 1R 1R Mediastinal Enlarged 1 lymph node Dark red discoloration 1R Axillary Dark red discoloration 1R lymph node Lung White area 1 Adhesions 1R Skin Red matting 1R 2 Brown matting 1 Teeth- White discoloration 1R 2R 4R 4R 1,4R incisors Thin 1R 3R 2R 4R Malaligned 1R Missing 1R Uterus Clear fluid contents 1 1 1

50 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dose (mg/kg) 0 0.27 0.54 1.05 Sex M F M F M F M F Iliac lymph Enlarged 1 node Urinary Thickened 1 bladder Calculus(I) 1 Distended 1 Heart adhesions 1 Pituitary Enlarged 1R

Organ Weights Unremarkable Histopathology See Table 7. In addition to the findings in Table 7, there was an Adequate battery: Yes; all increased incidence and/or severity of minimal to marked soft tissue tissues were examined from preterm and/or vascular mineralization at dose levels ≥0.27 mg/kg in the decedents, control, MD and HD rats. aorta, bone (cervical and/or thoracic vertebrae; meninges and Target organs and gross lesions were vascular), colon (tunica and vascular), duodenum (vascular), eyes examined from LD rats. (conjunctiva and vascular), femur, heart (myocardium and vascular), kidney (including vascular), larynx, mesenteric lymph node (vascular and germinal), lungs (including vascular), ovaries, pancreas (vascular), pharynx, mandibular salivary gland (vascular), skeletal muscle (myofiber and vascular), spinal cord (meninges), glandular stomach (including vascular), tongue (vascular), and trachea. Following 42 days of recovery, minimal to moderate mineralization was still present in the heart, kidney, larynx, mesenteric lymph node, glandular stomach, and tongue. There was also an increased incidence and/or severity (up to moderate) of multi-organ (including the brain) mononuclear cell infiltration at dose levels ≥0.27 mg/kg, which generally showed evidence of recovery. Toxicokinetics T1/2: 1.82-4.07 hours; Tmax: 1-2 hours Dose proportionality: Cmax and AUC0-24 hr generally increased dose proportionally Accumulation: Yes, ≤2.5-fold based on Cmax on Day 90/61 compared to Day 0 Sex differences: Cmax and AUC0-24 hr were ≤2.2-fold higher in females compared to males Summary of Toxicokinetics (3-month Study; Rats)

Day Dose Cmax AUC0-24 hr T1/2 Tmax (mg/kg) (µM) (µM·hr) (hr) (hr) M F M F M F M F 0 0.27 0.0934 0.129 0.513 0.964 2.66 4.07 1 1 0.54 0.161 0.275 0.956 1.85 1.82 3.02 1 1 1.05 0.317 0.676 2.31 5.04 2.32 2.74 2 2 21 0.27 0.135 0.217 0.743 1.08 3.72 3.36 1 1 0.54 0.185 0.363 1.18 2.00 3.24 3.86 1 1 1.05 0.572 0.698 3.19 4.45 2.97 3.48 1 1 90/61^ 0.27 0.143 0.298 0.743 1.21 2.59 2.63 2 1 0.54 0.283 0.418 1.23 2.77 2.45 3.14 1 2 1.05 0.808 0.894 3.58 4.44 2.7 3.15 1 2 Hr = hours; ^ = Day 90 for LD and MD and Day 61 for HD -: indicates reduction in parameters compared to control. TK: Toxicokinetic; HD: High dose (1.05 mg/kg/day); MD: Mid dose (0.54 mg/kg); LD: Low dose (0.27 mg/kg)

51 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 7: Selected Histopathology Findings (3-month Study; Rats)

Dose (mg/kg) 0 0.27 0.54 1.05 Sex M F M F M F M F # of rats examined 10,5R 10,5R 0,4R 0,5R 10,5R 10,4R 7,5R 8,4R Bone (cervical and/or # of rats examined: 0 0 0 0 0 0 0 5 thoracic vertebra)* Degeneration, cartilage Min 1 Moderate 2 Physeal dysplasia Mild 1 Moderate 2 Marked 2 Hyperostosis, Moderate 2 Inflammation, granulomatous 1 Moderate 1 Marked 1 Necrosis, Marked Bone (cervical vertebra) # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 0,5R 0,4R Dysplasia, physeal, Mild 1 1R Degeneration, cartilage, 1 Min Inflammation, 1R granulomatous Mild Colon Intussusception 1 Esophagus Degeneration, muscle Min 3 Moderate 1 Femur # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 7,5R 8,4R Dysplasia, articular cartilage 1 4,1R 4,1R 5,2R 5,3R Min 1,1R Mild Dysplasia, physeal 1,2R 2,2R 4,1R 3,1R 2 Min 1,1R 1,2R 1,1R 4,1R Mild 1R 1,2R 2,3R Moderate 1 Marked Increased remodeling 3 2 Min 2 2 Mild 2 2 Moderate Necrosis, subphyseal 1 bone Moderate 1 Hyperostosis-periosteal new bone, Min 1R Increased thickness, cortical bone, diaphysis, Mild

52 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dose (mg/kg) 0 0.27 0.54 1.05 Sex M F M F M F M F # of rats examined 10,5R 10,5R 0,4R 0,5R 10,5R 10,4R 7,5R 8,4R Heart # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 7,5R 8,4R Cardiomyopathy Min 6,3R 7,2R 1 5,3R 3,2R 3 Mild 1R 1R 2 3 1 Metaplasia, cartilaginous Min 1 Moderate 1 Kidney # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 7,5R 8,4R Nephropathy, chronic progressive Min 6,4R 2 4,2R 2 5,2R 1 1R 3 Mild 1,1R 1,1R 3,1R 2 Dilation, tubule Min 1 1 Mild 1 1R 1,1R Hyperplasia, transitional epithelium, Mild 1 1R Inflammation, chronic- active 1 Moderate Liver Congestion, Min 2 Necrosis, hepatocellular, 1 1 Min Vacuolation, 1 hepatocellular 2 Min 1R Mild Lipidosis, tension, Min Axillary lymph node # of rats examined: 10 10 0 0 10 10,1R 7 8 Hemorrhage, acute, Mild 1R Mandibular lymph node # of rats examined: 10,1R 10 0 0 10 10,1R 7,1R 8,2R Hemorrhage, acute, Mild 1R 1R Lungs # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 7,5R 8,4R Eosinophil infiltrate, perivascular, Min 1 2 Macrophages, alveolar Min 1 3 2 4,1R 2 3,2R 4,1R Mild 1 3,1R 3 4 Inflammation, subacute Min 2R 1R 4 1R 3R 2R Mild 1,1R Inflammation, acute, Min 1R Mild 1 Inflammation, chronic active, pleura Mild 1 Moderate 1R Marrow, femur # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 7,5R 8,4R Decreased cellularity Min 1,1R 2 1R 3,3R Mild 2 1 Moderate 1 Fibrosis

53 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Dose (mg/kg) 0 0.27 0.54 1.05 Sex M F M F M F M F # of rats examined 10,5R 10,5R 0,4R 0,5R 10,5R 10,4R 7,5R 8,4R Min 1 1 Mild 1 Pancreas Focus, basophilic hypertrophic, Min 1 Inflammation, chronic- active 1 Moderate Atrophy, acinar 1 2 Min 1 Mild Prostate Inflammation, subacute Min 1 2 Mild 1 Sublingual salivary gland Acinary atrophy, Min 1 Ductal epithelium hyperplasia 1 Min Spinal cord Degeneration, axonal Min 3 2 1 2 Moderate 2 Sternum # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,5R 10,4R 7,5R 8,4R Dysplasia, physeal Min 3 4,1R Mild 1 Teeth-incisors # of rats examined: 10,5R 10,5R 10,4R 10,5R 10,4R 10,4R 7,5R 8,4R Dysplasia Mild 1R Moderate 1,1R 2 4,3R 7 1,1R 1 Marked 1 2,4R 6,2R 7,4R Increased bone remodeling 4,1R 3,1R Min 1 1 1 3,2R Mild 1 1,1R Moderate Thyroid glands Cyst, ultimobranchial 1 Urinary bladder # of rats examined: 10 10 1 0 10 10 7 8 Hyperplasia, urothelium Moderate 1 Uterus # of rats examined: 0 10 0 1 0 10 0 8 Dilation, luminal Min 1 Mild 3 1 3 1 Moderate 1 Groups 1-3 were necropsied at Week 13; Group 4 was necropsied at Week 8. R = Recovery cohort (Groups 1-3 necropsied at Week 19 and Group 4 necropsied at Week 14) * collected from female rats with vertebral gross observations; not examined in HD males because it was added to the protocol tissue list after they were necropsied

54 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study title/ number: A 13-Week Study of INCB054828 by Oral Gavage in Cynomolgus Monkeys with a 6-Week Recovery Period / T15-12-03

Key Study Findings • There was no mortality • Pemigatinib induced dry skin, increased phosphorous and calcium, and histologic mineralization in the kidney and adrenal gland • Target organs included the bone (physeal dysplasia in the femur and cartilage dysplasia in the sternum), kidney, and epididymis

(b) (4) Conducting laboratory and location: GLP compliance: Yes

Methods Dose and frequency of dosing: 0, 0.1, 0.33, and 1 mg/kg once daily for 13 weeks Route of administration: Oral gavage Formulation/Vehicle: 0.5% methylcellulose (400 cps) in citrate buffer, pH 3.0 ± 0.3 Species/Strain: Monkey / Cynomolgus Number/Sex/Group: 4/sex/group (main) 2/sex/group (6-week recovery period) Age: 2.3-4 years old Satellite groups/ unique design: None / None Deviation from study protocol No affecting interpretation of results:

Observations and Results: changes from control

Parameters Major findings Mortality None Clinical Signs 0.1 mg/kg: Dry brown material, dry skin, and black skin (nose) 0.33 mg/kg: Dry brown material, dry skin, and tremors (1 monkey on Day 40) 1 mg/kg: Dry white material and dry skin Findings showed evidence of recovery. Body Weights and Feed Unremarkable Consumption Ophthalmoscopy One 0.1 mg/kg female monkey (#2504) exhibited slight focal corneal opacity indicative of a faint central superficial corneal scar in the left eye on Days 36 and 89, which was considered secondary to incidental corneal trauma that had healed. There were no findings at higher dose levels. ECG Unremarkable Hematology Unremarkable Clinical Chemistry Findings showed evidence of recovery compared to control and/or baseline

55 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Clinical Chemistry: % Change from Concurrent Control (13-week Study; Monkeys) Dose (mg/kg) 0.1 0.33 1 Sex M F M F M F ALT D 40 224%↑* D 86 8%↑ 9%↑ 62%↑* 193%↑* D 133 10%↑R 8%↑R 3%↑R 28%↓R AST D 40 74%↑* D 86 5%↓ D 133 42%↓R Phosphorous D 40 11%↑ 6%↑ D 86 18%↑* 7%↑ D 133 19%↓R 24%↓R Calcium D 86 6%↑* D 133 2%↑R Globulin D 40 6%↑ 13%↑ D 86 9%↑ 19%↑* D 133 6%↓R 15%↑R Albumin/globulin D 40 5%↓ 11%↓ ratio D 86 12%↓ 17%↓* D 133 3%↑R 18%↓R % relative to controls on Day 40 or 86; *, P ≤ 0.05; D = Day; R = denotes Recovery group (Day 133)

Urinalysis Unremarkable Gross Pathology Dark red focus in the colon and stomach correlated histologically with mild congestion. Thickness in the colon correlated histologically with mild hypertrophy. Gross Pathology Findings (13-week Study; Monkeys) Dose (mg/kg) 0 0.1 0.33 1 Sex M F M F M F M F Kidney Abnormal appearance 1 Colon Dark red focus 1R 1 Thick 1R Stomach Dark red linear focus, 1 1 antrum, mucosa Ovary Cyst 1R R = denotes Recovery group

Organ Weights Unremarkable Histopathology See Table below. In addition to these findings, there was an increased Adequate battery: Yes incidence and/or severity of multi-organ (including the sciatic nerve) mononuclear cell infiltration up to mild at dose levels ≥0.1 mg/kg, which trended towards recovery in most tissues.

56 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Histopathology Findings (13-week Study; Monkeys) Dose (mg/kg) 0 0.1 0.33 1 Sex M F M F M F M F # of monkeys examined 4,2R 4,2R 4,2R 4,2R 4,2R 4,2R 4,2R 4,2R Bone, femur Physeal dysplasia Min 1R 1,1R Mild 3 4 1 Bone, sternum Dysplasia, cartilage Min 1 1 1 Mild 1,1R 2,2R 3,2R Cervix Neutrophilic infiltration, lumen, Mild 1 Epididymis Edema, Mild 1 Adrenal gland Mineralization, Min 1 Parathyroid gland Cyst 1R Thyroid gland Cyst Mild 1 1 Moderate 1 Kidney Mineralization, Min 1 Regeneration, tubular, Min 1 Cyst, Moderate 1 Colon Congestion, Mild 1R 1 Hypertrophy, smooth muscle Mild 1R Spleen Hypertrophy, follicle, Mild 1R Stomach Congestion, Mild 1 1 R = denotes Recovery group

Toxicokinetics T1/2: 2.04-4.61 hours; Tmax: 1-1.67 hours Dose proportionality: Cmax and AUC0-24 hr generally increased dose proportionally Accumulation: None Sex differences: Cmax was ≤2-fold higher in females compared to males on Day 30; there were no substantial differences on Day 1 or 86 Summary of Toxicokinetics (13-week Study; Monkeys)

Day Dose Cmax AUC0-24 hr T1/2 Tmax (mg/kg) (µM) (µM·hr) (hr) (hr) M F M F M F M F 1 0.1 0.0253 0.0383 0.119 0.13 2.89 2.08 1.33 1 0.33 0.082 0.114 0.3 0.403 2.77 2.38 1.17 1.67 1 0.367 0.366 1.44 1.31 3.3 4 1.17 1.33 30 0.1 0.0232 0.0366 0.103 0.133 3.12 2.6 1.17 1 0.33 0.0682 0.125 0.286 0.417 4.11 4.04 1.17 1 1 0.224 0.463 1.21 1.33 4.61 4.58 1.33 1 86 0.1 0.0414 0.0333 0.122 0.112 2.04 2.49 1 1.17 0.33 0.11 0.124 0.317 0.358 2.38 4.21 1 1 1 0.379 0.403 1.51 1.26 3.92 3.89 1.5 1 Hr = hours

General toxicology; additional studies

57 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The Applicant conducted GLP-compliant 28-day repeat-dose toxicity studies in Sprague Dawley rats (Study # T14-02-06) and cynomolgus monkeys (Study # T14-04-01) investigating once daily oral administration of pemigatinib at the same dose levels evaluated in the 3-month toxicology studies. Rats were dosed for 14 days on, 7 days off, 14 days on in the 28-day study. Similar target organs of toxicity were generally observed in the 28-day and 13-week toxicology studies, although mortality was not observed in the 28-day studies.

In the 28-day study, one monkey dosed with 0.1 mg/kg pemigatinib had a chronic fracture of the growth plate that was spanned by a band of fibrous tissue and well-formed bone spicules. A relationship to the drug is uncertain because the bone formation within the gap was completely remodeled, which likely required more time than the dosing phase of this study. Additional ophthalmologic findings in the 28-day monkey study included moderate lens opacities and slight attenuation of retinal vessels at pemigatinib dose levels ≥0.33 mg/kg and 1 mg/kg, respectively. In a non-GLP 10-day repeat-dose toxicology study in cynomolgus monkeys (Study #T13-10-13), once daily oral administration of pemigatinib resulted in 33% mortality at the 3 mg/kg dose level (approximately 3.8 times the exposure based on AUC at the human dose of 13.5 mg) beginning on Day 7. The preterm decedents exhibited body weight loss, hyperthermia, facial swelling, and decreased activity. The cause of death was not identified; however, histologic findings included tubular hyperplasia/hypertrophy, congestion, tubular degeneration, and mineralization in the kidney; vacuolar degeneration in the pancreas; degeneration, ulcer, and necrosis in the stomach; hypercellularity and inflammation in the bone marrow; hypocellularity in the thymus and lymph node; bronchiolitis in the lung; and multi-organ mineralization.

Genetic Toxicology

In Vitro Reverse Mutation Assay in Bacterial Cells (Ames) Study title/ number: INCB054828: Bacterial Reverse Mutation Assay / T14-05-03 Key Study Findings: • Pemigatinib was not mutagenic in the strains tested in the presence or absence of S9 • Standard positive controls confirmed the sensitivity and validity of the assay GLP compliance: Yes Test system: Salmonella typhimurium strains TA98, TA100, TA1535, and TA1537 and Escherichia coli strain WP2 uvrA; up to 5000 µg/plate; +/- S9 Study is valid: Yes

In Vitro Assays in Mammalian Cells Study title/ number: INCB054828: In Vitro Mammalian Chromosomal Aberration Assay in Human Peripheral Blood Lymphocytes (HPBL) / T16-01-06 Key Study Findings: • Pemigatinib did not induce significant or dose-dependent structural or numerical chromosomal aberrations in the presence or absence of S9

58 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

• Standard positive controls confirmed the sensitivity and validity of the assay GLP compliance: Yes Test system: Human peripheral blood lymphocytes (HPBL); up to 488 µg/mL in preliminary toxicity assay and up to 40 µg/mL for chromosome aberrations; 4 hours +/- S9 and 20 hours -S9 Study is valid: Yes

In Vivo Clastogenicity Assay in Rodent (Micronucleus Assay) Study title/ number: INCB054828: In Vivo Mammalian Erythrocyte Micronucleus Assay in Rats / T16-01-05 Key Study Findings: • Pemigatinib was negative for the induction of micronucleated polychromatic erythrocytes in rat bone marrow • Standard positive controls confirmed the sensitivity and validity of the assay GLP compliance: Yes Test system: Sprague-Dawley rat bone marrow; single oral doses up to 2000 mg/kg (dose range- finding assay) and 100 mg/kg (definitive assay) Study is valid: Yes

Other Genetic Toxicity Studies None

Carcinogenicity

Carcinogenicity studies were not conducted or needed to support the use of pemigatinib in the currently proposed indication per ICH S9.

Reproductive and Developmental Toxicology

Fertility and Early Embryonic Development No study or assessment conducted.

Embryo-Fetal Development

The Applicant did not submit GLP-compliant embryo-fetal development studies in two species. Rather, since a preliminary dose range-finding embryofetal development study in rats demonstrated clear positive results at exposure multiples below the human exposure at the clinical dose of 13.5 mg, FDA agreed that this study was sufficient to convey the developmental risk to patients.

59 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study title/ number: An Oral (Gavage) Dose Range-Finding Study of the Effects of INCB054828 on Embryo/Fetal Development in Rats / T18-02-04 Key Study Findings • Pemigatinib reduced maternal body weight/food consumption and gravid uterine weight at dose levels ≥0.3 mg/kg and ≥0.1 mg/kg, respectively • Pemigatinib induced 100% embryo-fetal mortality due to post-implantation loss (early resorptions) at dose levels ≥0.3 mg/kg (approximately 0.6 times the exposure based on AUC at the human dose of 13.5 mg) • Pemigatinib reduced fetal body weight and was teratogenic at the 0.1 mg/kg dose level (approximately 0.2 times the exposure based on AUC at the human dose of 13.5 mg)

(b) (4) Conducting laboratory and location:

GLP compliance: No

Methods Dose and frequency of dosing: 0, 0.1, 0.3, and 1 mg/kg once daily from Gestation Day (GD) 6 to 17 Route of administration: Oral gavage Formulation/Vehicle: 0.5% methylcellulose (400 cps) in 50 mM citrate buffer, pH 3.0 ± 0.3 Species/Strain: Rat / Sprague-Dawley Number/Sex/Group: 8 females/group Satellite groups: Toxicokinetic: 8 females/group Study design: Time-mated female rats (mating confirmed on GD 0) were received on GD2 and administered pemigatinib once daily on GD 6-17. Rats were euthanized on GD 21. Deviation from study protocol No affecting interpretation of results:

Observations and Results

Parameters Major findings Mortality None Clinical Signs Four HD rats exhibited red vaginal discharge beginning on GD 19; two of these HD rats also exhibited red material on the anogenital area beginning on GD 16. These clinical signs correlated with fetal resorptions. One MD rat exhibited brown vaginal discharge. Body Weights There was a ≤29% decrease in mean body weight in MD and HD rats on GD 16-21 and GD 15-21, respectively, which correlated with decreased litter size and did not appear to be a clear sign of maternal toxicity

60 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Gravid Uterine Weights Statistically significant 12%, 99%, and 99% decrease in mean gravid uterine weight in LD, MD, and HD rats, respectively Necropsy findings 100% embryo-fetal mortality occurred at the MD and HD due to Cesarean Section Data post-implantation loss (early resorptions). Decreased fetal body weight was observed at the LD. Cesarean Section Findings (Embryo-fetal development Study; Rats) Dose (mg/kg) 0 0.1 0.3 1 # Mated females 8 8 8 8 # Pregnant (%) 8 (100%) 8 (100%) 8 (100%) 8 (100%) # Dams with live fetuses 8 (100%) 8 (100%) 0 (0%) 0 (0%) Mean # corpora lutea 14.1 13.5 13.6 15 Mean # implantation sites 13.3 13.4 12.6 12.3 Mean pre-implantation loss (%) 0.9 (5.5%) 0.1 (1%) 1 (6.6%) 2.8 (16.2%) Mean post-implantation loss 0.8 (5.3%) 0.8 (5.8%) 12.6 (100%)** 12.3 (100%)** Mean # early resorptions (%) 0.8 (5.3%) 0.8 (5.8%) 12.6 (100%)** 12.3 (100%)** Mean # late resorptions 0 0 0 0 Mean # total resorptions 5.3% 5.8% 100%** 100%** Mean # dead fetuses 0 0 0 0 Mean # live fetuses (%) 12.5 (94.7%) 12.6 (94.2%) 0** (0%)** 0** (0%)** Mean fetal body weight (g) 6.2 5.3 NA NA % Difference from control NA 14.5%↓** Mean fetal sex ratio (% males) 55.1% 56.3% NA NA NA = Not applicable; **, P ≤ 0.01 vs. controls; Mean % was calculated on a litter basis

Necropsy findings Evaluation of fetal morphology at the MD and HD was Offspring precluded by 100% post-implantation loss. There were no external malformations or variations.

Fetal Malformations and Variations (Embryo-fetal development Study; Rats) Dose (mg/kg) 0 0.1 0.3 1 # Fetuses examined 100 101 N N # of Litters examined 8 8 N N Total % per litter with malformations 0 37.5% NA NA Visceral malformations: # of fetuses affected (% per litter) Retroesophageal aortic arch 0 1 (12.5%) NA NA Total % per litter with visceral 0% 12.5% NA NA malformations Skeletal malformations: # of fetuses affected (% per litter) Vertebral anomaly with or without 0 4 (37.5%) NA NA associated rib anomaly Only 12 pairs of ribs present 0 1 (12.5%) NA NA Total % per litter with skeletal 0% 37.5% NA NA malformations Visceral variations: # of fetuses affected (% per litter) Renal papilla(e) not developed and/or 16 (50%) 13 (50%) NA NA distended ureters Major blood vessel variation^ 0 (0%) 11 (50%) NA NA Total % per litter with visceral variations 50% 50% NA NA Skeletal variations: # of fetuses affected (% per litter) 14th rudimentary rib(s) 23 (75%) 14 (50%) NA NA

61 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Sternebra(e) malaligned (slight or 2 (25%) 6 (37.5%) NA NA moderate) 7th cervical rib(s) 0 (0%) 3 (37.5%) NA NA Reduced ossification of the 13th rib(s) 0 (0%) 19 (75%) NA NA Hyoid unossified 0 (0%) 1 (12.5%) NA NA 25 presacral vertebrae 1 (12.5%) 1 (12.5%) NA NA Bent rib(s) 0 (0%) 1 (12.5%) NA NA Sternebra(e) #5 and/or #6 unossified 1 (12.5%) 1 (12.5%) NA NA Vertebral centra not fully ossified 3 (25%) 3 (25%) NA NA 27 presacral vertebrae 0 (0%) 6 (50%) NA NA Reduced ossification of the vertebral 0 (0%) 1 (12.5%) NA NA arches Total % per litter with skeletal variations 87.5% 100% NA NA N = No viable fetuses/litters available for evaluation; % per litter = (Total # of litters with malformation or variation / Total # of litters) x 100; NA = Not applicable; ^ = right carotid and subclavian arteries arose independently from the aortic arch; no brachiocephalic trunk Toxicokinetics T1/2: 2-4.2 hours; Tmax: 1-2 hours Dose proportionality: Cmax and AUC0-24 hr generally increased dose proportionally Accumulation: None Summary of Toxicokinetics (Embryo-fetal development Study; Rats)

Day Dose Cmax AUC0-24 hr T1/2 (mg/kg) (µM) (µM·hr) (hr) GD 6 0.1 0.062 0.416 3.4 0.3 0.181 1.34 NC 1 0.695 6.22 2.0 GD 17 0.1 0.0637 0.558 4.2 0.3 0.188 1.54 2.5 1 0.744 6.25 3.0 Hr = hours; NC = Not calculable LD: low dose (0.1 mg/kg); MD: mid dose (0.3 mg/kg); HD: high dose (1 mg/kg)

Prenatal and Postnatal Development

No study or assessment conducted.

Other Toxicology Studies

ICH M7 In Silico Genotoxicity Assessment of Potential Impurities / TOX-IS-19-05-01 (b) (4) (b) (4) The Applicant evaluated 26 starting materials, intermediates, or potential impurities for potential mutagenicity using Derek Nexus (Deductive Estimation of Risk on Existing Knowledge) Version 6.01, a rules-based expert system, and Sarah Nexus Version 3.0.0, a statistical based methodology for the prediction of mutagenicity. A combined assessment was performed to assign each compound a genotoxicity classification of 1-5 according to ICH M7. Eighteen of the 26 structures were assigned to ICH M7 Class 5 and treated as non-mutagenic (b) (4) impurities. has not been tested for mutagenicity and is considered a genotoxic (b) (4) impurity. The assessment of five compounds

62 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (4) was inconclusive using Sarah Nexus, but additional evaluation (b) (4) resulted in an ICH M7 classification of 4/5 (non-mutagenic). Compound (b) (4) had an unclassified feature in Derek and was Out of Domain in Sarah due to the BH3 complex. Since this complex is expected to dissociate in vivo, it was classified as Class 4/5 (non­ (b) (4) (b) (4) mutagenic) consistent with . Starting material was assigned to ICH M7 (b) (4) (b) Class 3 based on identification of alerting structures. was not detected (< (4)ppm) in three batches of pemigatinib drug substance (CA17-1186, CA18-0351 and CA19-0466) selected from each manufacturing campaign. Compound (b) (4) was assigned to ICH M7 Class 5 by Nexus, but contained a misclassified feature within Derek, so it was reclassified as Class 3 and treated as a potentially genotoxic impurity. (b) (4) was not tested for mutagenicity and was (b) (b) (4) not detected (< (4) ppm) in three batches of product from the validation campaign (CA19­ 0374, CA-190375 and CA19-0376).

Neutral Red Uptake Phototoxicity Assay of INCB054828 in BALB/c 3T3 Mouse Fibroblasts / T16-09-08 In this GLP-compliant assay, the Applicant evaluated the phototoxic potential of pemigatinib by measuring the relative reduction in viability of BALB/c 3T3 mouse fibroblasts exposed to pemigatinib and ultraviolet radiation compared to the viability of fibroblasts exposed to pemigatinib in the absence of ultraviolet radiation. Pemigatinib did not demonstrate phototoxic potential [Photoirritancy Factor (PIF) = 1] compared to the positive control promethazine (PIF = 16.569) under the conditions of this in vitro assay. The criterion for “phototoxic” is PIF >5.

X Emily F. Wearne, PhD X Whitney S. Helms, PhD Primary Reviewer Team Leader

63 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

6 Clinical Pharmacology

Executive Summary

The Applicant seeks approval for the treatment of adults with previously treated, locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or rearrangement. The proposed dosage is 13.5 mg orally once daily, with or without food for 14 days followed by 7 days off therapy. The primary evidence of efficacy and safety was demonstrated by the open- label, single arm Study INCB 54828-202 in patients with advanced or metastatic cholangiocarcinoma, with FGFR2 fusion or rearrangement, whose disease had progressed on or after at least one prior therapy. The dose escalation Study INCB 54828-101 provides additional data to support the safety of pemigatinib . The most frequently reported treatment emergent adverse event (TEAE) was hyperphosphatemia, which occurred in 88 (60.2%) of patients. Based on the clinical pharmacology review, the Applicant’s proposed dose reduction scheme in labeling adequately mitigates the risk of serious adverse events related to hyperphosphatemia.

Pemigatinib is a CYP3A4 substrate, and moderate to strong CYP3A4 inducers should be avoided during pemigatinib therapy; if a moderate to strong CYP3A4 inhibitor is administered concomitantly with pemigatinib, the pemigatinib dose should be reduced.

Recommendations The Office of Clinical Pharmacology has reviewed the information contained in NDA 213736. This NDA is approvable from a Clinical Pharmacology perspective. The key review issues with specific recommendations/comments are summarized below.

Review Issue Recommendations/Comments Pivotal or supportive The primary evidence supporting pemigatinib effectiveness for treatment evidence of effectiveness of cholangiocarcinoma with a FGFR2 fusion or rearrangement is derived from INCB 54828-202, a multicenter, open label, single arm study. General dosing instructions The proposed dosing regimen is 13.5 mg orally once daily for 14 days followed by 7 days off therapy. Dosing in patient subgroups • Moderate to strong CYP3A4 inducers: avoid concomitant use with (intrinsic and extrinsic pemigatinib factors) • Moderate to strong CYP3A4 inhibitors: Avoid use if possible. If a moderate to strong CYP3A4 inhibitor is administered concomitantly with pemigatinib, pemigatinib dose should be reduced: o From 13.5 mg orally once daily to 9 mg orally once daily o From 9 mg orally once daily to 4.5 mg orally once daily • The pharmacokinetics (PK) of pemigatinib in patients with severe renal insufficiency (RI) and hepatic impairment (HI) is unknown. See Post- Marketing Requirements and Commitments for studies to evaluate HI and RI. Labeling Generally acceptable upon the applicant’s agreement with the reviewer’s specific content and format recommendations.

64 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Post-Marketing Requirement (PMR) or Commitment (PMC)

PMRs: 1. Submit the analysis and datasets with the final study report for an ongoing hepatic impairment clinical trial to evaluate the pharmacokinetics and safety of pemigatinib in patients with normal hepatic function and patients with hepatic impairment. 2. Submit the analysis and datasets with the final study report for an ongoing renal impairment clinical trial to evaluate the pharmacokinetics and safety of pemigatinib in patients with normal renal function and patients with renal impairment.

Summary of Clinical Pharmacology Assessment

Pharmacology and Clinical Pharmacokinetics

The geometric mean steady-state pemigatinib AUC0-24h was 2620 nM·h (54% CV) and Cmax was 236 nM (56% CV) for 13.5 mg orally once daily. Steady state pemigatinib concentrations increased proportionally over the dose range of 1 to 20 mg (0.07 to 1.5 times the recommended dose). Steady-state was achieved within 4 days following repeated once daily dosing. With repeated once daily dosing, pemigatinib accumulated with a median accumulation ratio of 1.63 (range 0.63 to 3.28).

Absorption The median time to achieve peak pemigatinib plasma concentration (Tmax) was 1.13 (range 0.50-6.00) hours. Effect of Food: Administration of PEMAZYRE with a high-fat and high-calorie meal (approximately 1000 calories with 150 calories from protein, 250 calories from carbohydrate, and 500-600 calories from fat) had no clinically meaningful effect on pemigatinib pharmacokinetics.

Distribution The estimated apparent volume of distribution was 235 L (60.8%) following a 13.5 mg oral dose. In vitro, pemigatinib was 90.6% bound to human plasma proteins at concentrations ranging from 1 to 10 µM.

Elimination The geometric mean elimination half-life (t½) of pemigatinib was 15.4 (51.6% CV) hours and the geometric mean apparent clearance (CL/F) was 10.6 L/h (54% CV). Metabolism: Pemigatinib is predominantly metabolized by CYP3A4 in vitro. The major drug- related moiety in plasma was unchanged pemigatinib in a human [14C] mass balance study. Excretion: Following a single oral 13 mg dose of radiolabeled pemigatinib, 82.4% of the dose was recovered in feces (1.4% as unchanged) and 12.6% in urine (1% as unchanged).

65 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

General Dosing and Therapeutic Individualization

General Dosing

The proposed dosing regimen for pemigatinib is 13.5 mg taken orally once daily without regard to food for 14 days followed by 7 days off therapy. INCB 54828-202, a single arm study in patients with advanced or metastatic cholangiocarcinoma with FGFR2 fusion or rearrangement whose disease had progressed on or after at least one prior therapy; Study INCB 54828-101 provided additional safety data . The most frequently reported TEAE was hyperphosphatemia, which occurred in 88 (60.2%) patients. The proposed dosage is effective and has a manageable safety profile.

Therapeutic Individualization

The population PK (PopPK) analysis indicated that the impact of age (21-79 years), body weight (39.8-156 kg), race/ethnicity (Hispanic, Japanese), mild to moderate renal impairment, or mild to moderate hepatic impairment did not result in clinically meaningful differences in systemic exposure of pemigatinib.

From a mechanistic perspective, the proposed indication in FGFR2 rearrangements or fusions is acceptable based on the available clinical and nonclinical evidence.

Outstanding Issues

A PMR will be issued to submit datasets and final study reports for studies to assess the effect of renal impairment and the effect of hepatic impairment on the PK of pemigatinib. Refer to Post-Marketing Requirements and Commitments for details.

Comprehensive Clinical Pharmacology Review

General Pharmacology and Pharmacokinetic Characteristics

Table 8: Summary of General Pharmacology and Pharmacokinetic Characteristics of Pemigatinib Pharmacology Pemigatinib is a small molecule kinase inhibitor of FGFR1-4. Pemigatinib inhibits Mechanism of Action FGFR phosphorylation and signaling and decreases cell viability in cancer cell lines with activating FGFR amplifications and fusions that result in constitutive activation of FGFR signaling. PK/ECG data from the dose escalation study INCB 54828-101 in patients with advanced malignancies characterized the ability of pemigatinib to prolong the QT Prolongation QT interval over a dose range of 1 to 20 mg daily. No large mean increase in QTc (i.e. >20 ms) was detected at the observed steady state geometric mean maximum concentration (Cmax) of 421 nM. General Information

66 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Validated bioanalytical assays (LC-MS/MS) were developed to measure Bioanalytical assay pemigatinib concentrations in clinical studies included in the NDA (see Appendix 1). Oral clearance of pemigatinib is approximately 50% higher in healthy individuals compared to patients with advanced malignancies. Inter-individual variability in Healthy vs. Patients oral clearance was higher in patients with advanced malignancies (40-55%) compared to healthy individuals (28-36%). Single Dose PK

Geometric mean Cmax: -161 nM (CV%=59.6%) following a single 13.5 mg dose

Geometric Mean AUC0-24: Drug Exposure at Steady State -1610 nM*h (CV%=52.9%) following a single 13.5 mg dose Following the Therapeutic Dosing Regimen Steady State PK

Geometric mean Cmax,ss: - 236 nM (CV% =56.4%) for 13.5 mg QD

Geometric Mean AUCss,0-24: - 2620 nM*h (CV% = 54.1%) for 13.5 mg QD

Dose Proportionality Pemigatinib showed dose-proportional increase in both Cmax and AUC across the dose range of 1 to 20 mg at steady state. Following oral dosing of pemigatinib 13.5 mg QD, the geometric mean Accumulation accumulation ratio for AUC0-24 was 1.61. Steady state is projected to be reached following 4 days of pemigatinib dosing at 13.5 mg QD. Absorption Bioavailability The Applicant determined that pemigatinib has low solubility and high permeability (BCS Class II compound). Absolute oral bioavailability has not been determined. Study INCB 54828-202, the 13.5 mg daily dose was administered as three 4.5 mg tablets that are identical in formulation to the to-be-marketed 4.5 mg tablets. The to-be-marketed 9 mg and 13.5 mg tablets are quantitatively proportional to the 4.5 mg tablet formulation. A biowaiver was granted for in vivo bioequivalence for pemigatinib 9 mg and 13.5 mg tablets. Tmax 1.13 (range 0.50-6.00) hours. Distribution Volume of Distribution Vd in the terminal elimination phase is 235 L (CV%=60.8%) following a 13.5 mg dose. Plasma Protein Binding Pemigatinib was 90.6% bound to human plasma proteins at concentrations ranging from 1 to 10 µM. Elimination

Half-life The geometric mean plasma elimination half-life (t½) is 15.4 hours (CV% = 51.6%). Clearance The geometric mean apparent clearance (CL/F) is 10.6 L/h (CV%=54%). Metabolism

Primary Metabolic Pathway(s) Primarily metabolized by CYP3A4. No major metabolites (≥ 10% of total compound- related material) were detected in plasma.

67 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Inhibition/Induction of An in vitro study demonstrated that pemigatinib was not a potent inhibitor or Metabolism and Transporter inducer of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, or CYP3A4. Systems. Pemigatinib is an inhibitor of P-gp, OCT2, and MATE1. Excretion Pemigatinib is primarily eliminated in feces (82.4%) and to a lesser extent in urine (12.6%). In feces, 44.4% of the dose was recovered as M2 (INCB056632, O­ Primary Excretion Pathways desmethyl-INCB054828), while the intact parent pemigatinib accounted for 1.4% of the administered dose. Phase II metabolites, M7 and M9, were the only metabolites detected in urine at levels ≥ 1% of the administered dose (4.4% and 2.1%, respectively).

Insert Clinical Pharmacology Questions

Does the clinical pharmacology program provide supportive evidence of effectiveness?

Yes. The recommended dose of pemigatinib 13.5 mg taken orally once daily for 14 days followed by 7 days off therapy was evaluated in Study INCB 54828-202, which demonstrated an acceptable efficacy and toxicity profile. In addition, the proposed dose appears to provide sufficient exposure to achieve adequate target engagement.

The first-in-human dose-finding studies of pemigatinib utilized the 14 days on followed by 7 days off dose schedule based the preclinical toxicology profile of pemigatinib, and the on-target hyperphosphatemia associated with pemigatinib. The interruption of dosing for 7 days within the dosing regimen allowed for recovery from hyperphosphatemia.

Efficacy of pemigatinib for the treatment of patients with cholangiocarcinoma with a FGFR2 fusion or rearrangement was evaluated in Study INCB 54828-202. In this study, 107 patients with FGFR2 rearrangements or fusions were enrolled in Cohort A and assigned to receive pemigatinib 13.5 mg once daily treatment on a 14 days on 7 days off schedule, taken after a 2­ hour fast. No other dosing regimens were investigated in the study. Dose interruptions and reductions were recommended for management of hyperphosphatemia. Overall response (ORR) based on Independent Review Committee (IRC)-assessed, confirmed tumor responses was 35.5% (95% CI: 26.50, 45.35). Median DOR was 7.49 months (95% CI: 5.65, 14.49), with a minimum of 6 months of follow-up from initial response in 92% of participants who had a confirmed tumor response.

Exposure-response analyses demonstrate that the probability of hyperphosphatemia, an on- target effect of pemigatinib, increases with increasing pemigatinib exposure (Figure 5). The exposure-response relationship should be interpreted with caution given the limitation of the study design (e.g., small number of patients and evaluation of only one dose level). The relationship between pemigatinib exposure parameters and ORR and progression free survival (PFS) was inconclusive.

68 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 5. Relationship between pemigatinib exposure and probability of hyperphosphatemia.

APPEARS THIS WAY ON ORIGINAL

Source: Applicant’s Study Report DMB-19.120.1.

Is the proposed dosing regimen appropriate for the general patient population for which the indication is being sought?

Yes. The Applicant’s proposed dosing regimen of 13.5 mg taken orally once daily for 14 days followed by 7 days off therapy is acceptable based on the PK, efficacy, and safety data in the proposed indicated population of patients with cholangiocarcinoma with a FGFR2 fusion or rearrangement in study INCB 54828-202. Nonclinical studies and pharmacodynamic data from Study INCB 54828-101 provide additional data supporting the proposed dosing regimen in the general patient population.

Dosing Regimen The recommended dose of pemigatinib is 13.5 mg taken orally once daily for 14 days followed by 7 days off therapy. During pemigatinib therapy phosphate levels are monitored and a dosage modification scheme is provided in labeling for patients who experience hyperphosphatemia, retinal pigment epithelial detachment, hepatotoxicity, and other adverse events. Pemigatinib may be taken without regard to food, which is supported by the lack of a statistically or clinically significant effect of food in the food effect substudy of Study INCB 54828-101.

Dose Selection Rationale In non-clinical studies of KATO III human gastric cancer-bearing SCID mice, maximal tumor growth inhibition was observed at pemigatinib doses that provided >12 hours of coverage above IC50 (Study Report PRECLIN-14.06.1). Subsequently, in Study INCB 54828-101 dosages

69 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

above 6 mg once daily provided >50% inhibition of ex vivo phosphorylated FGFR2α at trough. Dosages up to 20 mg daily were assessed and the maximum tolerated dose (MTD) was not reached. Pemigatinib 13.5 mg daily provided 76% inhibition of ex vivo phosphorylated FGFR2α at trough. The 13.5 mg daily dose was identified as the pharmacologically active dose (defined as the dose at which approximately 67% of participants experienced elevations in serum phosphate and was selected as the recommended phase 2 dose (RP2D) per the protocol, which stated that the RP2D would the MTD or the pharmacologically active dose, whichever was the lowest. In addition, the 50 patients who received the 13.5 mg intermittent daily dosing regimen had generally lower rates of treatment-emergent adverse events (TEAEs) compared to the 19 patients who received 13.5 mg continuous daily dosing. Among the 10 most common TEAEs, occurrence of hyperphosphatemia (76.0% vs. 84.2%), dry mouth (34.0% vs. 52.6%), alopecia (30.0% vs. 42.1%), stomatitis (22.0% vs. 57.9%), diarrhea (24.0% vs. 52.6%), constipation (28.0% vs. 36.8%), nausea (16.0% vs. 42.1%), and decreased appetite (24.0% vs. 31.6%) were numerically lower in patients who received intermittent dosing compared to continuous dosing while rates of fatigue (38.0% vs. 36.8%) were similar and rates of dysgeusia were numerically higher with intermittent dosing (16.0% vs. 21.1%).

Support for the 13.5 mg orally once daily for 14 days followed by 7 days off therapy treatment regimen was provided by the PK, efficacy, and safety profiles observed in the INCB 54828-202 study. In this study, the ORR based on IRC-assessed, confirmed tumor responses was 35.5% (95% CI: 26.50, 45.35). The most commonly reported TEAE was hyperphosphatemia, which occurred in 88 (60.3%) patients; however, no patients had ≥ Grade 3 hyperphosphatemia, indicating that the dose reduction scheme appropriately mitigated the risk of this on-target toxicity. All participants had at least one TEAE and 93 (63.7%) had a Grade≥ 3 TEAE. TEAEs led to dose interruption of pemigatinib in 62 (42.5%) patients, dose reductions in 20 (13.7%) patients, and discontinuations in 13 (8.9%) patients.

QT Prolongation In study INCB 54828-101, a 12-lead ECG was taken on Day 1 and Day 14 of Cycle 1 prior to dosing, then 1, 2, and 4 hours post-dose (approximately 5 minutes before PK blood draws). Doses of pemigatinib up to 20 mg daily did not have a large effect on the QT/QTc interval, defined as exceeding the threshold of concern (i.e., >20 ms). As such, the QT-IRT Review team concluded that the data from INCB 54828-101 suggest a lack of large mean effect on the QTc interval and determined that these data were adequate to characterize the effect of pemigatinib 13.5 mg daily on the QTc interval.

Is an alternative dosing regimen or management strategy required for subpopulations based on intrinsic patient factors?

No. The PopPK analysis (See Pharmacometrics Review; Appendix 3) indicated that age (21-79 years), body weight (39.8-156 kg), race/ethnicity (Hispanic, Japanese), creatinine clearance, mild to moderate renal impairment, or mild to moderate hepatic impairment did not have a clinically meaningful effect on systemic exposure of pemigatinib. In addition, clinical data support targeted activity of pemigatinib in cholangiocarcinoma with FGFR2 fusions or 70 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

rearrangements. As such, no changes in the dosing regimen or management strategy are recommended for subpopulations based on intrinsic factors at this time.

Renal Impairment Participants were classified as having normal renal function, mild renal impairment, or moderate renal impairment using the MDRD equation for estimating glomerular filtration rate (GFR). In the PopPK model, 146 (45.9%) participants had normal renal function, 134 (42.1%) had mild renal impairment, and 38 (11.9%) had moderate renal impairment. There were no significant differences in apparent clearance of pemigatinib in participants with mild or moderate renal impairment compared to participants with normal renal function (Table 9). A dedicated renal impairment study (INCB 54828-108) is currently being conducted to further assesses the impact of renal impairment on pemigatinib pharmacokinetics, including individuals with severe renal impairment; a PMR will be issued to submit the dataset and final study report.

Table 9. Effect of mild or moderate renal impairment on pemigatinib apparent clearance.

Renal Function Number of Participants GMR (90% CI)

Normal 146 Reference

Mild Impairment 134 0.991 (0.906, 1.084)

Moderate impairment 38 0.994 (0.877, 1.128)

Source: FDA pharmacometrics review

Pemigatinib is an inhibitor of OCT2 (IC50 = 0.075 μM; DMB-14.62.1) and MATE1 (IC50 = 1.1 μM; DMB-18.141.1). OCT2 is expressed basolaterally in renal proximal tubule cells and transports creatinine into the proximal tubule; MATE1 is expressed apically and effluxes creatinine into the urine. Therefore, inhibitors of OCT2 or MATE1 may cause increases in serum creatinine by decreasing active secretion of creatinine, without causing damage to nephrons or a reduction in GFR. Within the first 21-day cycle of pemigatinib dosing, serum creatinine increased (mean increase of 0.2 mg/dL) and reached steady state by Day 8, and then decreased during the 7 days off therapy. Therefore, use of alternative markers to assess renal function may be considered if persistent elevations in serum creatinine are observed.

Hepatic Impairment In the PopPK model 213 (67%) of participants were classified as having normal hepatic function, while 94 (29.4%) had mild hepatic impairment, and 11 (3.5%) had moderate hepatic impairment based on the NCI hepatic impairment classification. There were no significant differences in apparent clearance of pemigatinib in participants with mild or moderate hepatic impairment compared to participants with normal hepatic function (Table 10). A dedicated hepatic impairment study (INCB 54828-107) is currently being conducted to further assess the 71 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

impact of hepatic impairment on pemigatinib pharmacokinetics, including individuals with severe hepatic impairment; a PMR will be issued for submission of the study report and datasets.

Table 10. Effect of mild or moderate hepatic impairment on apparent pemigatinib clearance.

Hepatic Function Number of Participants GMR (90% CI)

Normal 213 Reference

Mild 94 0.996 (0.915, 1.084)

Moderate 11 1.031 (0.842, 1.263 )

Source: FDA pharmacometrics review

FGF/FGFR Alterations The proposed indication for pemigatinib is for the treatment of adults with previously treated, locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or rearrangement.

FGFR2 rearrangements may arise by different mechanisms and may occur independently of a fusion gene. In the INCB 54828-202 trial, the Applicant classified FGFR2 rearrangements as either a fusion or rearrangement according to the following definitions:

• An FGFR2 rearrangement predicted to be a fusion: Breakpoint is within the FGFR2 intron 17/exon 18 hotspot and the partner gene is known in the literature or is a novel partner that is predicted to be in-frame with FGFR2. • An FGFR2 rearrangement, which cannot be confidently predicted to be a fusion: Breakpoint is within the FGFR2 intron 17/exon 18 hotspot, but the partner gene is novel and out-of-frame or out-of-strand with exon 17 of FGFR2. Alternatively, the downstream end of the breakpoint may be in an intergenic region and not within another gene (designated as partner N/A).

Both FGFR2 fusions and rearrangements (as defined above) are predicted to leave the kinase domain intact, and lack C-terminal residues, thus potentially leading to ligand-independent activation, decreased receptor internalization/degradation, and increased receptor autophosphorylation/activation (Li, et. al. 2019). According to the Applicant, these similarities predict pemigatinib's FGFR inhibitory activity across cholangiocarcinoma patients with FGFR2 fusions or rearrangements within the intron 17/exon 18 hotspot and an intact kinase domain, and support grouping FGFR2 fusions or rearrangements for the purpose of study enrollment.

72 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study INCB 54828-202 included three cohorts of patients with advanced or metastatic cholangiocarcinoma defined by FGF/FGFR alteration status. Cohort A enrolled 107 patients with FGFR2 fusions or rearrangements, Cohort B enrolled 20 patients with other FGF/FGFR alterations (eligible alterations included FGFR point mutations, FGF/FGFR amplifications, FGFR1 fusions, and FGFR3 fusions), and Cohort C enrolled patients negative for FGF/FGFR alterations. Enrollment and initial cohort assignment were permitted based on tumor genomic testing results from a local laboratory; final cohort assignment for statistical analyses was based on tumor FGF/FGFR status as determined by central testing using the DNA-based next generation sequencing Foundation Medicine clinical trial assay (CTA). The primary endpoint of the study was ORR in Cohort A; ORR in Cohorts A + B, ORR in Cohort B and ORR Cohort C were all secondary endpoints.

Cohort A enrolled 107 patients with cholangiocarcinoma with FGFR2 fusions (N=92) or rearrangements (N=15) as defined by the Applicant. There were 56 unique FGFR2 rearrangement or fusion partners; 31 patients had a FGFR2-BICC1 fusion, which was the most frequently identified FGFR2 fusion or rearrangement, no other partner was identified in more than four patients. Five patients had FGFR2 rearrangements denoted as “FGFR2 rearrangement (partner N/A )”. In Cohort A, ORR based on IRC-assessed, confirmed tumor responses was 35.5% (95% CI: 26.5, 45.35), including 3 complete responses (2.8%) and 35 partial responses (32.7%); among the 38 participants in Cohort A with IRC-assessed, confirmed tumor responses, median DOR was 9.1 months (95% CI: 6.0, 14.5).

Efficacy in Cohort A was similar across the evaluable molecular subtypes. The IRC-confirmed ORR in patients with FGFR2 fusions was 34.8% (95% CI: 25.2, 45.4) and the ORR in patients with FGFR2 rearrangements was 40.0% (95% CI: 16.3, 67.7). The ORR in patients with FGFR2-BICC1 fusion was 32.3% (95% CI: 16.7, 51.4), the combined ORR for other fusion or rearrangements was 36.8% (95% CI: 26.1, 48.7). Also, the ORR in patients with partner N/A rearrangements (N=5) was 40%. Please refer to Section 8 for additional information on pemigatinib efficacy across molecular subtypes (Table 19).

Of the 20 patients enrolled in Cohort B, 7 had FRS2 activation, 5 had FGF3, FGF4, FGF19 amplification, 4 had the FGFR2 p.C382R point mutation, and the remaining patients had other FGF/FGFR pathway alterations (one alteration each) identified in the tumor. As defined by the protocol, patients enrolled in Cohort C did not have any FGF/FGFR alterations identified. Among the 20 patients in Cohort B and the 18 patients in Cohort C, there were no IRC-assessed confirmed tumor responses. In Cohort B, ORR based on IRC-assessed, unconfirmed tumor response was 10.0% (95% CI: 1.23, 31.70), with 2 patients having partial responses. In Cohort C, ORR based on IRC-assessed, unconfirmed tumor responses was 5.6% (95% CI: 0.14, 27.29), with 1 patient having a partial response.

Are there clinically relevant food-drug or drug-drug interactions, and what is the appropriate management strategy?

Yes, specific recommendations are provided below. 73 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Effect of CYP3A Inhibitors on Pemigatinib Pemigatinib is metabolized in the liver, predominately by CYP3A4. In Study INCB 54828-104 healthy participants received pemigatinib 4.5 mg as a single dose on Day 1, followed by itraconazole (a strong CYP3A inhibitor) 200 mg once daily on Days 4-7, then pemigatinib 4.5 mg as a single dose with itraconazole 200 mg single dose on Day 8, then itraconazole 200 mg daily Days 9-11. Co-administration of itraconazole significantly increased Cmax (geometric mean ratio (GMR) 117%; 90% CI 107%-129%) and AUC0-∞ (GMR 188%; 90% CI 177%-206%) compared to administration of pemigatinib alone.

Physiologically-based pharmacokinetic (PBPK) analyses were conducted to evaluate the effects of another strong CYP3A inhibitor (clarithromycin), moderate CYP3A inhibitors (erythromycin, diltiazem, and fluconazole), and a weak CYP3A inhibitor (fluvoxamine) on the PK of pemigatinib. When pemigatinib is co-administered with a strong CYP3A inhibitor, the model predicted an approximate 2-fold increase in pemigatinib exposure. The model-predicted geometric mean AUC of pemigatinib was increased by approximately 50% to 90% when pemigatinib was co- administered with a moderate CYP3A inhibitor. The predicted geometric mean AUC of pemigatinib was decreased by more than 50% when it was co-administered with efavirenz (a moderate CYP3A inducer) 600 mg QD for 2 weeks.

The increase in exposure of pemigatinib when administered concomitantly with moderate or strong CYP3A inhibitors may increase the frequency or severity of adverse events. Therefore, if use of moderate to strong CYP3A inhibitors cannot be avoided, a dose reduction from the recommended dose of 13.5 mg orally to pemigatinib 9 mg orally daily is recommended.

Effect of CYP3A Inducers on Pemigatinib In Study INCB 54828-104 healthy participants received pemigatinib 13.5 mg as a single dose on Day 1, followed by rifampin (a strong CYP3A inducer) 600 mg daily on days 4-10, then pemigatinib 13.5 mg single dose and rifampin 600 mg single dose on Day 11, then rifampin 600 mg as a single dose on Day 12. Co-administration of rifampin significantly decreased Cmax (GMR 38.0%; 90% CI 33.2%-43.5%) and AUC0-∞ (GMR 14.9%; 90% CI 13.9%-16.1%) compared to administration of pemigatinib alone.

PBPK analyses were conducted to evaluate the effects of a moderate CYP3A inducer (efavirenz) on the PK of pemigatinib. The predicted geometric mean AUC of pemigatinib was decreased by more than 50% when it was co-administered with efavirenz 600 mg QD for 2 weeks.

The decrease in exposure when pemigatinib is administered concomitantly with moderate to strong CYP3A inducers may result in decreased efficacy of pemigatinib; therefore, moderate to strong CYP3A inducers should be avoided in patients administered pemigatinib.

74 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Gastric pH modifying agents In Study INCB 54828-106, Cohort 1 participants received a single oral 13.5 mg dose of pemigatinib in the fasted state on Day 1, followed by esomeprazole (a proton-pump inhibitor) 40 mg daily in the fed state on Days 3-7, then a single oral dose of pemigatinib 13.5 mg plus esomeprazole 40 mg on Day 8. Cohort 2 participants received a single oral 13.5 mg dose of pemigatinib in the fasted state on Day 1, 150 mg ranitidine (a histamin-2 (H2) blocker) orally every 12 hours in the fed state for Days 3-5, then a single dose of pemigatinib 13.5 mg plus ranitidine 150 mg every 12 hours in the fasted state on Day 6.

Co-administration of esomeprazole with pemigatinib resulted in a statistically significant reduction in Cmax (GMR 65.3%; 90% CI 54.7% - 78.0%) and AUC0-∞ (GMR 92.1%; 90% CI 88.6% ­ 95.8%) and a delay in Tmax of approximately 1 hour compared to administration of pemigatinib alone. These results are consistent with a delay in absorption of pemigatinib secondary to higher gastric pH with concomitant administration of esomeprazole; however, the marginal decrease in exposure is not anticipated to be clinically meaningful. Co-administration of ranitidine with pemigatinib did not have a statistically significant impact on exposure.

Transporters In vitro studies indicate that pemigatinib is a substrate of P-gp and BCRP (study reports DMB­ 14.61.1 and DMB-19.77.1). However, efflux mediated by P-gp and BCRP was saturated at concentrations of 1 and 30 µM and the gut concentration is approximately 111 µM at the recommended dosage (13.5 mg/250 mL). As such, efflux through P-gp and BCRP are not expected to affect absorption of pemigatinib at the recommended clinical dosages.

In vitro studies also indicate that pemigatinib is an inhibitor of P-gp, BCRP, OATP1B3, OCT2, MATE1, and weak inhibitor of OAT3 and MATE2K (study reports DMB-14.61.1, MB-19.77.1, DMB-18.141.1, DMB-14.62.1). Drug interactions with BCRP, OATP1B3, OAT3, and MATE2K substrates are not anticipated to be clinically relevant. A PBPK model was developed to further assess the potential effect of pemigatinib as an inhibitor of P-gp, OCT2, and MATE1. The PBPK analysis was inadequate to evaluate the effect of pemigatinib on the PK of a P-gp, OCT2, or MATE substrate due to the lack of quantitative in vitro to in vivo extrapolation for Ki (the inhibition constant) values (refer to Appendix for details). However, the Applicant claims that pemigatinib is a low solubility, high permeability (BCS Class II) compound that is poorly soluble at pH above 2; therefore, based on this assumption, the gut concentration/IC50 ratio is likely less than 10 for the majority of the gastrointestinal tract, and the inhibitory effect of pemigatinib on P-gp is not likely to cause a clinically meaningful change in exposures of P-gp substrates. The potential interaction between pemigatinib and OCT2 or MATE1 substrates was further assessed in response to an Information Request from OCP. Incyte submitted data indicating that 21 patients in the INCB 54828-101 and INCB 54828-202 studies were administered metformin (an OCT2 and MATE1 substrate) concomitantly with pemigatinib; none of these patients required dose adjustments for metformin and no notable changes in blood glucose concentrations were observed during treatment with pemigatinib compared to

75 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

baseline. As such, inhibition of OCT2 or MATE1 by pemigatinib does not appear to have a clinically meaningful effect on OCT2 or MATE1 substrates.

Effect of Food The effect of food on pemigatinib PK was assessed in INCB 54828-101, an open-label, dose- escalation study of pemigatinib in patients with advanced malignancies, that included a food- effect substudy. The effect of food at steady state was assessed using a standardized high­ fat/high-calorie breakfast (approximately 1000 calories with 150 calories from protein, 250 calories from carbohydrate, and 500-600 calories from fat) in 12 participants who were administered pemigatinib in the fasted (Cycle 1 Day 14) and fed (Cycle 2 Day 14) states. The high-fat, high-calorie breakfast resulted in a statistically significant delay in Tmax from 1.58 hours in the fasted state to 4.02 hours in the fed state, but did not significantly impact any other pemigatinib PK parameters. The effect of food on pemigatinib exposure is not anticipated to be clinically meaningful.

X Robert Schuck X Jeanne Fourie Zirkelbach

Primary Reviewer Team Leader

76 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

7 Sources of Clinical Data and Review Strategy

Table of Clinical Studies

Refer to Table 11 below.

77 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 11. Listing of Clinical Trials with Pemigatinib Study Number of Study Status; Study Design Regimen/Schedule1 Study Population Patients Endpoint/Objective Study Enrolled Report INCB 54828-101 Open-label, Safety, tolerability, PK, Part 1: various Advanced malignancies 160 Ongoing; multicenter, PD Part 2: pemigatinib 9 mg or 13.5 interim results dose-escalation mg on intermittent vs. continuous and expansion schedule study Part 3: pemigatinib 9 mg or 13.5 mg in combination with chemotherapy (various) on intermittent vs. continuous schedule INCB 54828-102 Open-label, Safety, tolerability, PK, Part 1 (dose escalation): Japanese patients with 25 Ongoing; multicenter PD pemigatinib 9 or advanced malignancies interim results dose escalation 13.5 mg daily on an intermittent and expansion schedule study Part 2 (dose expansion): Pemigatinib 13.5 mg daily on an intermittent schedule INCB 54828-104 Open-label Extrinsic factor PK Cohort 1: pemigatinib 4.5 mg Healthy volunteers 36 Completed; drug-drug single dose on Days 1 and 8 final interaction Cohort 2: pemigatinib 13.5 mg on study Days 1 and 11 INCB 54828-105 Open-label PK Single dose of pemigatinib 11 mg Healthy volunteers 7 Completed; AME study final results INCB 54828-106 Open-label Effect of esomeprazole Cohort 1: pemigatinib 13.5 mg Healthy volunteers 35 Completed; drug-drug and ranitidine on single dose on Days 1 and 8 final results interaction pemigatinib PK Cohort 2: pemigatinib 13.5 mg on study Days 1 and 6 INCB 54828-201 Single-arm, Efficacy, safety, Pemigatinib 13.5 mg daily for 14 Patients with metastatic 184 Ongoing; open-label, tolerability days then off for 7 days or surgically unresectable interim results multicenter urothelial carcinoma study harboring FGF/FGFR alterations

78 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study Number of Study Status; Study Design Regimen/Schedule1 Study Population Patients Endpoint/Objective Study Enrolled Report INCB 54828-202 Single-arm, ORR per RECIST 1.1 by Pemigatinib 13.5 mg daily for 14 Patients with 146 Ongoing; open-label, IRC, safety, tolerability days then off for 7 days advanced/metastatic or interim results multi-cohort, surgically unresectable global study cholangiocarcinoma who have progressed after at least 1 previous systemic treatment INCB 54828-203 Single-arm, Efficacy, safety, Pemigatinib 13.5 mg daily for 14 Patients with 15 Ongoing; open-label, tolerability days then off for 7 days myeloid/lymphoid interim results multicenter neoplasms with 8p11 study rearrangement known to lead to FGFR1 activation 1 Pemigatinib was administered orally in all studies. ORR=overall response rate, PK=pharmacokinetics, PD= pharmacodynamics, IRC=independent review committee, AME=absorption, metabolism, excretion

79 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Review Strategy

A single clinical trial, Study INCB 54828-202, supports Incyte’s NDA for pemigatinib for the treatment of adults with previously treated, locally advanced or metastatic cholangiocarcinoma with a fibroblast growth factor receptor 2 (FGFR2) fusion or rearrangement as detected by an FDA-approved test.

Data from studies that evaluated patients with solid tumors, healthy volunteers, and patients from Cohorts B and C of Study INCB 54828-202 were considered supportive for the assessments of safety. Please see Section 8.2.1 Safety Review Approach below for the safety review strategy.

The clinical review of efficacy included primary analyses using the clinical datasets to supplement and/or confirm Incyte’s results as reported in the Clinical Study Reports (CSR) for Study INCB 54828-202 and the Summary of Clinical Efficacy (SCE). As agreed upon during the June 12, 2019 meeting to discuss content and format of an NDA submission, a 4-month update was submitted (data cut-off date: August 30, 2019) during the review of the NDA and is the basis for the final analyses of duration of response.

80 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

8 Statistical and Clinical and Evaluation

Review of Relevant Individual Trials Used to Support Efficacy

INCB 54828-202 (FIGHT-202)

Trial Design

Study INCB 54828-202 (FIGHT-202) is a multi-center, international, open-label, single arm trial evaluating the safety and efficacy of pemigatinib in adult patients with FGFR positive surgically unresectable or advanced/metastatic cholangiocarcinoma who had disease progression following treatment with at least one line of systemic therapy. Patient accrual was conducted at 67 sites across the United States, Europe and Asia. Patients were assigned to one of three cohorts based on tumor FGFR status: - Cohort A: FGFR2 rearrangements or fusions - Cohort B: Other FGF/FGFR alterations - Cohort C: No FGF/FGFR alterations

The key eligibility inclusion criteria are as follows: • Histologically or cytologically confirmed advanced/metastatic or surgically unresectable cholangiocarcinoma. Patients will be assigned to one of 3 cohorts o Cohort A: FGFR2 translocations with a documented fusion partner in central laboratory report o Cohort B: other FGF/FGFR alterations o Cohort C (US only): negative for FGF/FGFR alterations • Radiographically measurable disease per RECIST v1.1 • Documentation of FGF/FGFR gene alteration status • Documented disease progression after at least 1 line of prior systemic therapy • ECOG performance status of 0 to 2 • Archival tumor specimen (more recent preferred, and recommended no older than 2 years) or willingness to undergo a biopsy.

Patients are ineligible for enrollment if they had previously received a selective FGFR inhibitor, have current evidence of clinically significant corneal or retinal disorder confirmed by ophthalmologic examination, or have history of and/or current evidence of ectopic mineralization/calcification.

The dose of pemigatinib self-administered by patients is 13.5 mg orally daily for 2-weeks-on then 1-week-off in cycles of 21 days. Patients were instructed to take pemigatinib immediately upon rising or after a 2-hour fast and to continue fasting for one additional hour after taking study drug. Patients receive pemigatinib until progression of disease or unacceptable toxicity.

Dose modifications for toxicity are shown below with dose-level (DL) reductions to 9 mg (DL -1) 81 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

and 6 mg (DL -2). Dose reductions below 6 mg are not allowed, and the dosing schedule cannot be changed.

Table 12. Dose Reductions for Toxicity in Study INCB 54828-202

Adverse Event Action

AST and/or ALT is > 5.0 × ULN Step 1: Interrupt study drug up to 2 weeks (14 days) until the toxicity has resolved to ≤ Grade 1 except by approval of the medical monitor. Step 2: Restart study drug at same dose. If assessed as related to study drug, restart study drug at next lower dose; monitor as clinically indicated. Any Grade 1 or Grade 2 toxicity Continue study drug treatment and treat the toxicity; monitor. Any Grade 3 toxicity, if clinically Step 1: Interrupt study drug up to 2 weeks significant and not manageable by (14 days), until toxicity resolves to ≤ Grade 1. supportive care Step 2: Restart study drug at same dose. If assessed as related to study drug, restart study drug at next lower dose; monitor as clinically indicated. Any recurrent Grade 3 toxicity after Discontinue study drug administration and 2 dose reductions follow-up per protocol. Any other Grade 4 toxicity. Discontinue study drug administration and follow-up per protocol.

Patients were clinically monitored weekly for the first cycle then every 3 weeks, at the end of treatment, and 30 days following the end of treatment. Imaging with computed tomography (CT) or magnetic resonance imaging (MRI) was performed within 28 days of Day 1 of treatment and every 2 cycles through Cycle 4 then every 3 cycles. A schedule of laboratory assessments is shown below in Table 13.

82 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 13. Schedule of Laboratory Assessments in Study INCB 54828-202

Source: copied from the protocol, Version 6

Tumor FGFR status was centrally confirmed with the Foundation Medicine Clinical Trial Assay (CTA). Enrollment and initial cohort assignment were permitted based on genomic testing results from a local laboratory. Final cohort assignment for statistical analyses was based on results from the central genomics laboratory. The Foundation Medicine CTA distinguished between FGFR2 rearrangements predicted to be fusions and those that could not be confidently predicted to be a fusion according to the following rules: • FGFR2 rearrangement predicted to be a fusion: The breakpoint is within the FGFR2 intron 17/exon 18 hotspot and the partner gene is known in the literature or is a novel partner that is predicted to be in-frame with FGFR2. • FGFR2 rearrangement which cannot be confidently predicted to be a fusion: The breakpoint is within the FGFR2 intron 17/exon 18 hotspot but the partner gene is novel and out-of-frame or out-of-strand with exon 17 of FGFR2. Alternatively, the downstream end of the breakpoint may be in an intergenic region and not within another gene (designated as partner N/A).

Please see Section 8.1.2 below regarding the results of FGFR status and changes to cohort assignments that were made during the conduct of the trial.

The FDA review of efficacy for this application is restricted to data derived from the 107 patients enrolled in Cohort A, the sole cohort pertinent to the proposed indication.

Study Endpoints

The primary endpoint of the study was overall response rate (ORR) per RECIST 1.1 as assessed by an Independent Review Committee (IRC), defined as the proportion of patients who received at least one dose of pemigatinib who exhibited a complete response (CR) or partial response (PR). The IRC was comprised of two primary reviewers, and a third blinded reviewer was the adjudicator for discordant cases of the two primary reviewers.

83 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Secondary endpoints included duration of response (DOR), progression-free survival (PFS) per RECIST 1.1 as assessed by the IRC, and overall survival (OS). Patient reported outcomes based on European Organization for Research and Treatment of Cancer Quality-of-Life Questionnaire (EORTC QLQ-C30) and EORTC QLQ-BIL21 (questionnaire for measuring quality of life in patients with cholangiocarcinoma and gallbladder cancer) were included as exploratory endpoints for this study.

Statistical Analysis Plan

In Cohort A, assuming a response rate of 33% and a drop-out rate of 10%, 100 patients were needed to obtain a probability of more than 95% to rule out a response rate of 15% by the lower limit of the 95% confidence interval (CI). One non-binding interim analysis for futility was planned for Cohort A after 25 patients were enrolled and had at least one tumor assessment (or had permanently discontinued from the study). If there were less than three responders, accrual to Cohort A could be stopped. An interim analysis was conducted after 47 patients were enrolled in Cohort A based on the data cutoff date of November 27, 2017. Eight patients had confirmed partial responses resulting in an ORR of 17% with a 95% CI of (7.6%, 31%).

The primary efficacy analysis for ORR was to be conducted in the population of patients in Cohort A with a confirmed FGFR2 rearrangement or fusion who received at least one dose of pemigatinib. Secondary time-to-event endpoints were described with Kaplan-Meier estimates and Brookmeyer-Crowley 95% CI.

Reviewer Comment: The timing of the final analysis was not pre-specified in the protocol. However, during the pre-NDA meeting dated August 8, 2019 FDA acknowledged that the proposed data cut-off of March 22. 2019 would provide a minimum of 7 months of follow-up for all patients and a minimum of 6 months of follow-up from time of initial response for 92% of responders.

There was no pre-specified analysis plan for the patient-reported outcomes analyses.

Reviewer Comment: In a single arm trial, FDA considers time-to-event endpoints to be uninterpretable, and the results will not be described in this review.

Protocol Amendments

The following protocol amendments instituted changes that materially impacted the conduct or analysis of results for Study INCB 54828-202.

Amendment #4 (version March 21, 2017) The protocol was revised to permit patient enrollment based upon local genomic testing results with final results to be determined by the central genomics laboratory. Incyte specified that final cohort assignment for the statistical analysis would be based upon the central genomics testing results.

84 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Amendment #5 (version October 3, 2017) The protocol was amended to incorporate the following changes: • The total number of patients for all cohorts was increased from 100 to 140; the number of patients to be enrolled into Cohort A was increased from 60 to 100. Assuming an ORR of 33% for pemigatinib, the sample size of 100 patients would provide > 95% probability (previously 80% with a sample size of 60) of showing a lower limit of the 95% confidence of > 15%. • A list of tumor FGF/FGFR alterations that would qualify patients for enrollment and instructions for documenting the fusion partner in the central laboratory report were added.

Amendment #6 (version February 15, 2018) The protocol was amended to incorporate the following changes: • Monitoring was revised to add a fundoscopy with digital imaging as part of the comprehensive eye examination. • The adverse events reporting section was updated to include specific guidance on grading hyperphosphatemia.

Amendment #7 (version April 2, 2020) The protocol was amended to incorporate the following changes: • Guidelines were provided for the management of serous retinal detachment and retinal pigmented epithelium detachment. • An optic coherence tomography (OCT) will be performed at the time of the regularly scheduled eye examination. • Based on results from drug-drug interaction studies, the restricted medications were updated to remove CYP3A4 inducers and proton pump inhibitors and to add OCT2 substrates. Prohibited medications were updated to include moderate CYP3A4 inducers.

Study Results

Compliance with Good Clinical Practices

The NDA included a statement from Incyte indicating that Study INCB 54828-202 was conducted in accordance with the ethical principles of Good Clinical Practice, according to the International Council for Harmonisation Guidelines.

Financial Disclosure

In accordance with 21 CFR 54, Incyte submitted a list of study clinical investigators for Study INCB 54828-202 and Form FDA 3454. According to Incyte, investigators provided certification indicating that they held no financial interests or arrangements that required disclosure per the criteria described on Form 3454, with the following two exceptions:

85 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

• One investigator (Site 018) received more than $25,000 for speaker and consultation fees. • One investigator (Site 020) held equity interest in Incyte Corporation that exceeded $50,000. The investigator also received fees exceeding $25,000 for participation in a (b) (4) speakers bureau .

Reviewer comment: Site 018 enrolled 7 patients with 14% of patients identified as responders, and Site 020 enrolled 5 patients with 40% of patients identified as responders. Both sites were inspected by the Office of Scientific Investigations and no deficiencies were identified. Additionally, use of a blinded independent review committee mitigates the potential for a biased assessment of efficacy in Study INCB 54828-202.

Data Quality and Integrity

No data quality related issues were identified during the review process. All datasets were submitted in electronic common technical document (eCTD) standard format. The submission had all the requisite components of a supplemental application and was reviewable.

Patient Disposition

Table 14 summarizes the patient disposition in Cohort A. As of the data cut-off date of August 30, 2019, 15 patients (14%) were still receiving treatment with pemigatinib, 69 patients (64%) had discontinued trial participation due to disease progression, and 1 patient (1%) had died while under treatment.

Table 14: Patient Disposition in Cohort A of Study INCB 54828-202

Patient Disposition n (%)

Patients treated 107 (100) Treatment ongoing 15 (14) Treatment discontinued 92 (86) Reason for discontinuation n = 92 Disease progression 69 (75) Patient choice 6 (7) Adverse event 6 (7) Physician decision 5 (5.4) Death 1 (1.1) Other 5 (5.4)

86 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Protocol Violations/Deviations

Table 15. Summary of Protocol Deviations in Cohort A of Study INCB 54828-202

Number of Patients (%) Deviation Category n=107 Adverse Event 2 (1.9) Informed Consent 7 (7) Entry Criteria 3 (2.8) Concomitant Medications 0 Non-compliance with Study 29 (27) Treatment Non-compliance with Study Procedure (out of window 53 (50) assessment) Non-compliance with Study Procedure (missed 95 (89) assessment) Other 19 (18)

Three patients in Cohort A were identified with protocol deviations pertaining to entry criteria. (b) (6) • Patient : The ECG did not display PR interval at study entry. The patient was in the subgroup of patients who achieved a partial response. (b) (6) • Patient : The patient had prior receipt of an FGFR inhibitor. The patient was not identified as a responder. (b) (6) • Patient : The patient had missing serology results at the screening visit. The patient was not identified as having a response.

The most common protocol deviations were missed assessments (89%), most of which were missed laboratory assessments.

Reviewer Comment: Given that two of the three patients with protocol deviations pertaining to entry criteria did not achieve at least a partial response and that the violation in the patient who did achieve a partial response was a minor one, this reviewer determined that the protocol violations did not likely impact the efficacy results in favor of the pemigatinib arm.

Table of Demographic Characteristics

The baseline demographic and disease characteristics of patients enrolled in Cohort A are

87 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

presented in Table 16 and Table 17, respectively.

Table 16: Demographic Characteristics of Patients in Cohort A of Study INCB 54828-202

Demographic Characteristics (N=107) n (%)

Sex Male 42 (39) Female 65 (61) Age Mean years (SD) 55 (12) Median (years) 56 Min, max (years) 26, 77 Age Group < 65 years 82 (77) ≥ 65 years 25 (23) Race White 79 (74) Black or African-American 7 (7) Asian 11(10) Other 4 (3.7) Missing 6 (6) Ethnicity Hispanic or Latino 2 (2) Not Hispanic or Latino 87 (81) Region United States 64 (60) Other 43 (40) ECOG Score 0 45 (42) 1-2 62 (58) Data cut-off date: March 22, 2019.

Table 17: Baseline Disease Characteristics of Patients in Cohort A

Disease Characteristics (N=107) n (%)

FGFR2 Alteration Fusion 92 (86) Other rearrangement 15 (14)

88 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Disease Characteristics (N=107) n (%)

Fusion Partner BICC1 fusion partner 31 (29) Other 76 (71) Primary Location Intrahepatic 105 (98) TNM Classification M0 16 (15) M1 88 (82) Prior Lines of Therapy 1 65 (61) 2 29 (27) ≥ 3 13 (12) Type of Prior Therapy Cisplatin-containing 81 (76) Gemcitabine-containing 91 (85) Prior radiotherapy 28 (26) Prior surgery 38 (36) Data cut-off date: March 22, 2019.

Cohort A of INCB 54828-202 enrolled patients who had tumors with in-frame fusions and other rearrangements that were predicted to have a breakpoint within intron17/exon 18 of the FGFR2 gene, leaving the FGFR2 kinase domain intact. A total of 92 (86%) patients enrolled in Cohort A had in-frame fusions and 15 patients (14%) had other FGFR2 rearrangements that could not be confidently predicted to be in-frame fusions, including 5 patients with rearrangements without an identifiable partner gene. These 5 patients had FGFR2 rearrangements denoted either as “FGFR2 rearrangement (N/A partner)” or “FGFR2 rearrangement intron 17” by the central genomics laboratory and were initially assigned to Cohort B of Study INCB 54828-202 based on the study protocol. Incyte determined in a discussion with the central genomics laboratory that the distinction between these two classes of alterations was due to medical reporting nomenclature differences between pre-existing (b) (4) (b) (4) commercial report and clinica reports, and that the two alteration types were identical. Specifically, the central genomics laboratory confirmed that FGFR2 rearrangement (N/A partner) alterations categorized as known or likely cancer-related alterations must have FGFR2 breakpoints within the intron 17 or exon 18 hotspots. Therefore, the 5 patients with FGFR2 rearrangement (N/A partner) were assigned to Cohort A for final statistical analyses prior to database lock.

Reviewer Comment: The demographic and disease characteristics of the patient population enrolled in Cohort A generally appear consistent with the profile of patients described to date in scientific literature, i.e., the younger female-predominant population with cholangiocarcinoma

89 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

arising in the intrahepatic region. The prior therapy received by most patients in Cohort A is consistent with standard of care first line treatment with gemcitabine and cisplatin. Regarding the reassignment of 5 patients from Cohort B to Cohort A, the decision was based on clarification of terminology prior to the database lock and analysis of efficacy, and is not likely to have affected the primary outcome.

Treatment Compliance, Concomitant Medications, and Rescue Medication Use

Treatment compliance was assessed by pill counts performed at the clinical sites. For Cohort A, median compliance was 100% (range 90% - 124%).

Concomitant medications were used by 99% of participants and the most frequent concomitant medication category was natural opium alkaloids, taken by 46% of patients in Cohort A with 30% started after study start. The most frequently used (≥ 20%) concomitant medications were paracetamol (36%), ondansetron (27%), omeprazole (22%), cholecalciferol (22%), sodium chloride solution (21%), and lorazepam (20%). Concomitant phosphate binders were used by 17% of participants. The most frequently reported phosphate binding medication was sevelamer reported as sevelamer (8%), sevelamer carbonate (4.1%), and sevelamer hydrochloride (0.7%). One additional participant (0.7%) used furosemide as a phosphaturic agent.

The most frequently used concomitant medications (in ≥ 7% of patients) that were initiated on or after the date of study start included sodium chloride (25%), paracetamol (21%), colecalciferol (14%), loperamide (12%), ciprofloxacin (11%), ibuprofen (11%), fentanyl (10%), levofloxacin (9%), oxycodone (8%), sevelamer (7%), and piperacillin/tazobactam (7%).

No rescue medications were used.

Reviewer Comment: Treatment compliance was high with at least 90% compliance for each patient. The concomitant medications used by patients were expected for a population of patients with cholangiocarcinoma that may cause hepatic pain requiring analgesics and cholangitis requiring the use of antibiotics. Other concomitant meds may reflect the use of supportive treatments for the toxicities associated with pemigatinib treatment (e.g., hyperphosphatemia, nausea, and diarrhea). The profile of concomitant medications did not raise any additional safety signals.

Efficacy Results – Primary Endpoint

Efficacy results for the primary endpoint of ORR are presented in Table 18, Table 19, and Table 20, including by FGFR2 fusion or rearrangement subgroups. Of the 107 patients enrolled in Cohort A of Study INCB 54828-202, 103 patients had post-baseline target lesion measurements as confirmed by the IRC. Three patients were non-evaluable for ORR: 2 due to study discontinuation before post-baseline assessment and 1 due to discontinuation prior to a second post-baseline assessment (the first assessment for this patient was stable disease).

90 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Nevertheless, the analysis of ORR presented below is based on the overall study population consisting of all patients who received at least one dose of pemigatinib (N=107). In the overall population, there were 3 complete responses (CR, 3%) and 35 partial responses (PR, 33%). Eleven responses (29%) were ongoing at the time of data cut-off.

Table 18: ORR per RECIST 1.1 as assessed by IRC

Efficacy Parameter N = 107

ORR (95% CI) 36% (27, 45) Complete response n=3 2.8% Partial response n=35 33% Median DoR (months) (95% CI)a 9.1 (6.0, 14.5) Patients with DoR ≥ 6 months, n (%) 24 (63%) Patients with DoR ≥ 12 months, n (%) 7 (18%) a The 95% confidence interval (CI) was calculated using the Brookmeyer and Crowley's method. Note: Data are by RECIST v1.1 per IRC, and all responses were confirmed.

Table 19 shows the response rate for the subgroups by all fusions, all rearrangements, and by fusion partners that occurred in ≥ 2 patients. Additional fusion partners occurred in one patient each and are not listed here.

Table 19: Subgroup Analysis of Cohort A Responses by Type of FGFR2 Abnormality and Fusion Partner1

ORR Subgroup N Responses (95% CI) Fusions 92 34.8 (25, 45) Complete response 2 Partial response 30 Rearrangements 15 40 (20, 64) Complete response 1 Partial response 5

Fusion Partner BICC1 31 10 32 (17, 51) N/A 5 2 40 (5, 85) KIAA1217 4 3 75 (19, 99) AHCYL1 3 2 67 (9, 99) AFF4 2 2 100 (16, 100) ARHGAP24 2 0

91 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

ORR Subgroup N Responses (95% CI) CCDC6 2 2 100 (16, 100) MACF1 2 0 NOL4 2 1 50 (1, 99) NRAP 2 0 PAWR 2 1 50 (1, 99) SHROOM3 2 0 SLMAP 2 1 50 (1, 99) TACC1 2 1 50 (1, 99) TRIM8 2 1 50 (1, 99) Data cut-off dates: March 22, 2019 1ORR was based on IRC assessment.

Table 20 provides a subgroup analysis of ORR by key demographic characteristics.

Table 20: ORR for Demographic Subgroups of Cohort A by IRC per RECIST 1.1

Characteristics Responders/n Response Rate (95% CI)

Sex Male 13/42 31 (18, 47) Female 25/65 38 (27, 51) Age Group < 65 years 28/82 34 (24, 45) ≥ 65 years 10/25 40 (21, 61) Race White 27/79 34 (24, 46) All Others 11/28 39 (22, 59) Region United States 23/64 36 (24, 49) Other 15/43 35 (21, 51) ECOG Score 0 16/45 36 (22, 51) 1-2 22/62 35 (24, 49) Prior Lines of Therapy 1 24/65 37 (25, 50) 2 10/29 34 (18, 54) ≥ 3 4/13 31 (9, 61) Data cut-off date: March 22, 2019

92 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Concordance of evaluation of response determination was evaluated among the two radiologists comprising the independent review committee. Among the 38 confirmed responders, 22 patients (58%) were assessed to have a response by both the radiologists, 9 (24%) were determined to be responders by Radiologist 1 but not by Radiologist 2, and 7 (18%) were determined to be responders by Radiologist 2 but not by Radiologist 1.

Reviewer Comment: The observed estimate of ORR and DOR demonstrated in Study INCB 54828-202 are sufficiently large in magnitude to be clinically meaningful in patients with cholangiocarcinoma harboring a FGFR2 fusion or other rearrangement who have received prior treatment. Anti-tumor response was observed in patients with FGFR2 rearrangements and fusions according to Incyte’s molecular classification of FGFR2 aberrations, and also across a variety of different fusion partners, although given the large number of identified fusions, the interpretation of results in these subpopulations should be interpreted with caution given the small sample sizes of most of these subpopulations. Subgroup analyses by key demographic subgroups, although again limited by small patient numbers, appeared to show a consistent effect irrespective of number of prior lines of treatment and other potentially prognostic demographic factors.

Dose/Dose Response

Refer to Section 6 (Clinical Pharmacology) for analysis of dose-response data.

Durability of Response

Kaplan-Meier estimates of duration of response are provided in the Efficacy Results – Primary Endpoint section above. In addition, a swimmer’s plot for duration of response and outcomes of patients with a confirmed response is provided in Figure 6.

93 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 6: Swimmer’s plot for responders per RECIST 1.1 assessed by IRC

Persistence of Effect

See description of durability of response in the above sections.

Efficacy Results – Secondary or exploratory COA (PRO) endpoints

Study INCB 54828-202 evaluated patient reported outcomes as an exploratory endpoint. Relevant quality of life measures were assessed using the EORTC QLQ-C30 and the EORTC QLQ­ BIL21 questionnaires. Responses to the quality of life questionnaires were collected at screening and every 3 weeks starting with Cycle 3 until the end of treatment. Additionally, responses to the QLQ-BIL21 questionnaire were collected for patients in the United States, United Kingdom, Italy, Germany, and South Korea. Of the 107 patients in Cohort A, 102 (95%) had quality of life measurements at baseline as well as follow-up measurements. There were no significant trends in the PRO data.

Reviewer Comment: Due to the lack of a pre-specified analysis plan of the patient-reported outcomes and the open-label single-arm design of this study, the analyses of patient-reported outcomes data from Study INCB 54828-202 are difficult to interpret.

Additional Analyses Conducted on the Individual Trial

None

94 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

8.1.3 Integrated Review of Effectiveness

The efficacy data from other cholangiocarcinoma patients in pemigatinib clinical trials was not pooled for this application per agreement with Incyte due to limited patient numbers. Eight patients with cholangiocarcinoma and an FGFR2 aberration were enrolled to Study INCB 54828-101 and treated with various doses of pemigatinib, with 3 responses observed (38%; 95% CI: 14, 69). The fusion or rearrangements observed in the 3 responders were FGFR2-CCDC, FGFR2-CLIP1, and FGFR2 translocation (intron 17). One additional patient with FGFR2 rearranged cholangiocarcinoma was enrolled to Study INCB 54828-102 and received pemigatinib at 13.5 mg daily on an intermittent schedule but did not achieve a response. These findings are consistent with the efficacy results for Study INCB 54828-202. Due to insufficient numbers of patients in the other studies, no additional analyses were performed.

Reviewer Comment: The response rate in these 8 patients, although based upon a small sample size with a wide range in the confidence interval, appears consistent with the efficacy results for Cohort A of Study INCB 54828-202.

Primary Endpoints

Not applicable – See the above description of responses in patients with cholangiocarcinoma harboring any FGFR rearrangement in studies other than Study INCB 54828-202 within the pemigatinib development program.

Secondary and Other Endpoints

Not applicable.

Subpopulations

Not applicable.

Conducting a pooled subgroup analysis

Not applicable.

Additional Efficacy Considerations

None

8.1.4 Integrated Assessment of Effectiveness

See Section 8.1.3 above.

95 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

8.2 Review of Safety

8.2.1 Safety Review Approach

The clinical assessment of the safety of pemigatinib is based on data from a total of 484 patients enrolled across 5 single-arm trials (Study INCB 54828-101 [n=116], Study INCB 54828­ 102 [n=25], Study INCB 54828-201 [n=182], Study INCB 54828-202 [n=146], and Study INCB 54828-203 [n=15]) and 78 healthy volunteers enrolled across 3 studies (Study INCB 54828-104, Study INCB 54828-105, and Study INCB 54828-106). The definitions of the pooled safety populations were previously agreed upon between FDA and the Applicant, Incyte, during the June 12, 2019 Type B meeting (minutes issued June 14, 2019, Comment 10). The integrated summary of safety (ISS) included pooled data in the “modified safety population”, also referred to as the “all cancer population” in this review, comprising data from all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) or who experienced toxicity at least possibly related to pemigatinib prior to completion of the first cycle (n=466). The pooled population of patients with cholangiocarcinoma is a subset of the modified safety population and includes all patients with cholangiocarcinoma who received pemigatinib as a single agent, irrespective of FGF/FGFR status (“the pooled cholangiocarcinoma population”; n=161). Exposure and safety data from the 3 clinical pharmacology studies involving healthy volunteers were not pooled due to differences in the study populations and the durations of exposure. The safety monitoring period spanned the time of first administration of pemigatinib until 30 days following the last dose of pemigatinib, including all adverse events (AEs) or related serious adverse events (SAEs).

The safety analyses described below were performed in the modified safety population, the pooled cholangiocarcinoma population, and in the population of patients with cholangiocarcinoma enrolled on Study INCB 54828-202 who received at least 1 dose of pemigatinib (n=146) irrespective of duration of treatment.

Eighteen patients with cancer enrolled across the 5 studies were excluded from the modified safety population because they did not complete at least one cycle of treatment or experience a TEAE prior to discontinuation of pemigatinib, per FDA’s recommendations discussed during the June 12, 2019 Type B meeting.

All studies (Study INCB 54828-202, Study INCB 54828-101, Study INCB 54828-102, Study INCB 54828-201, Study INCB 54828-203) were ongoing during the original NDA submission; therefore, interim clinical study reports were reviewed. The data cut-off dates for safety analyses in the studies submitted to the NDA are provided in Table 21:

96 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 21: Data cut off dates for Pivotal Studies submitted for the Pemigatinib NDA

Study Date

Study INCB 54828-202 22 MAR 2019

Study INCB 54828-101 19 FEB 2019

Study INCB 54828-102 18 JAN 2019

Study INCB 54828-201 22 FEB 2019

Study INCB 54828-203 11 JAN 2019

Source: Reviewer generated table based on information provided in the SCS

Data from the 4-month Safety Update included safety data using a data cut-off date of August 30, 2019. Narratives of deaths and SAEs from all 5 studies were submitted for events that occurred through the time of the data cut-off date. The review of safety included review of the interim clinical study reports (CSR), SDTM and analysis datasets, line-listings, case report forms (CRFs), and patient narratives from all 5 trials. The clinical reviewers confirmed Incyte’s safety analyses, conducting analyses of primary data using the MedDRA-based Adverse Event Diagnostics (MAED) tool and JMP programs, and supplementary analyses performed by the OCE safety team.

Reviewer Comment: This reviewer conducted a review of the results of the Applicant’s analyses of safety presented in the 4-month Safety Update, including a spot check verification of the results; no new safety signals were identified.

8.2.2 Review of the Safety Database

Overall Exposure

The pooled safety analyses was conducted based upon data from the modified safety population (also referred to as the “all cancer population”) which included patients with cancer enrolled in Studies INCB 54848-101, -102, -201, -202, and -203 who received at least one 21-day treatment cycle of pemigatinib as a single agent or experienced a toxicity considered at least possibly related to pemigatinib prior to completion of the first cycle.

The proposed dosage regimen of pemigatinib is 13.5 mg daily on a 2-weeks-on/1-week-off schedule. A total of 374 of 466 patients received 13.5 mg pemigatinib orally once daily on an intermittent schedule (80%), and 45 of 466 patients received 13.5 pemigatinib once daily on a continuous schedule (10%). Therefore, a total of 419 of 466 patients (90%) in the modified safety population received the proposed 13.5 mg dose of pemigatinib.

97 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

In the pooled cholangiocarcinoma population, also called the “all cholangiocarcinoma” population, 156 of the 161 patients (97%) received 13.5 mg pemigatinib either on an intermittent (n=154) or continuous dosing schedule (n=2). Therefore, a total of 158 of 161 patients (98%) received the proposed 13.5mg dose of pemigatinib. The median duration of exposure for the pooled cholangiocarcinoma patient population and the modified safety population was 181 and 104 days, respectively (see Table 22). Among patients in Study INCB 54828-202 (n=146), 49% received pemigatinib for > 6 months, and 16% received pemigatinib for > 12 months. For the modified safety population, which included patients treated with pemigatinib over a dose range of 1 to 20 mg daily on an intermittent or continuous schedule, 31% of patients received pemigatinib for > 6 months, and 9% of patients received pemigatinib for > 12 months.

Table 22: Summary for Exposure of Pemigatinib

Parameter All Cancer# Population All Cholangiocarcinoma$ INCB54828-202 (N=466) Population Population n (%) (N=161) (N=146) n (%) n (%) Median treatment 3.4 5.9 5.9 duration (Months) Range: 0.1-24.1 Range 0.2-24.1 Range: 0.2–24 >6 months 143 (31) 79 (49) 71 (49) >12 months 40 (9) 26 (16) 23 (16) <=1 month 46 (10) 12 (7) 13 (9) >1 - 3 months 173 (37) 42 (26) 38 (26) >3 - 6 months 104 (22) 28 (17) 24 (16) >6 - 9 months 67 (14) 31 (19) 29 (20) >9 - 12 months 36 (8) 22 (14) 19 (13) >12 - 15 months 20 (4.3) 11 (7) 10 (7) >15 - 18 months 6 (1.3) 5 (3.1) 5 (3.4) >18 - 21 months 9 (1.9) 6 (3.7) 5 (3.4) >21 - 24 months 4 (0.9) 3 (1.9) 3 (2.1) >24 months 1 (0.2) 1 (0.6) 0 Source: adex.xpt #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib

FDA’s primary analysis of the safety of the proposed dosage regimen of pemigatinib, 13.5 mg daily on a 2-weeks-on/1-week-off schedule, was based on results from Study INCB 54828-202 (n=146). Safety results for the pooled cholangiocarcinoma population (n=161) included 143 patients from Study INCB 54828-202 (3 patients were excluded due to not completing the 21­ (b) (6) day treatment cycle: Patient only had 7 days of treatment, experienced a small bowel (b) (6) obstruction twice and withdrew from the study; Patient died from progressive disease (b) (6) [PD] on Day 14, and Patient died from PD on Day 7) and an additional 11 patients treated with the proposed dosage regimen from Studies INCB 54828-101 and -102. Seven participants who received starting doses/schedules other than 13.5 mg daily on a 2-weeks­ on/1-week-off (i.e., pemigatinib 9 mg daily on an intermittent schedule, pemigatinib 20 mg

98 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

daily on an intermittent schedule, pemigatinib 13.5 mg daily on a continuous schedule, and pemigatinib 20 mg daily on a continuous schedule) were considered supportive, for the total of n=161.

Reviewer Comment: Notwithstanding the limited size of the safety population, the clinical review team considered the exposure to pemigatinib adequate to conduct a risk:benefit assessment. The majority of patients (90%) across the modified safety population received pemigatinib at the RP2D dose of 13.5 mg once daily. FDA notes that the pooled cholangiocarcinoma population had longer exposures to pemigatinib compared to the modified safety population. This is likely due to response to pemigatinib (i.e., a greater proportion of patients in the pooled cholangiocarcinoma population experienced responses to pemigatinib and therefore continued treatment whereas a greater proportion of patients in the modified safety population discontinued therapy due to progressive disease) and supports mechanism of action and efficacy of pemigatinib in the cholangiocarcinoma population with FGFR2 fusions or other rearrangements, as only 2 patients in the pooled cholangiocarcinoma population had tumors without an FGFR2 alteration.

Adequacy of the safety database:

Given the rarity of patients with cholangiocarcinoma harboring a FGFR2 fusion or rearrangement, the degree of pemigatinib exposure achieved in the safety population, and the observed adverse reaction profile, which is consistent with the mechanism of action of pemigatinib, FDA considered the safety database adequate to characterize risks of pemigatinib in the post market setting in patients with cholangiocarcinoma who have received prior therapy and whose tumors harbor FGFR2 fusions or other rearrangements.

Subgroup safety analyses to evaluate the impact of age on the frequency of AEs in each pooled safety population, and additional exploratory analyses were conducted based on gender, race, and performance score. Limitations in interpreting the results of these subgroup analyses include the small sample sizes and the lack of internal control in all studies.

The NDA submission included safety data from 484 patients with advanced malignancies who received at least 1 dose of pemigatinib administered as monotherapy, including 146 participants in Study INCB 54828-202. A total of 466 of these patients met the criterion for inclusion in the modified safety population (defined above); 161 of these patients were included in the pooled cholangiocarcinoma patient population. The primary analysis of the safety for the proposed dosage regimen of pemigatinib 13.5 mg daily on a 2-weeks-on/1-week­ off schedule, in patients with previously treated advanced/metastatic or surgically unresectable cholangiocarcinoma with a FGFR2 fusion or rearrangement, is based on results from Study INCB 54828-202. Safety results for the pooled cholangiocarcinoma patient population and the modified safety population are supportive of the safety data from Study INCB 54828-202 and provide a broader patient experience for identifying less common but potentially important adverse reactions. 99 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Safety monitoring in the pemigatinib studies consisted of collection of adverse events (AEs), serious adverse events (SAEs), laboratory tests (standard hematology and blood chemistries), physical observations/measurements (vital signs, electrocardiograms [ECGs], Eastern Cooperative Oncology Group [ECOG] status, eye exams), and pregnancy testing in female patients of childbearing potential.

To monitor for potential retinal-related visual disturbances during treatment with pemigatinib, eye examinations were required at screening, during treatment every 3 cycles (± 14 days), at the end of treatment, and as clinically indicated. Comprehensive eye examinations included visual acuity tests, slit-lamp examination, and fundoscopy with digital imaging. Additional ophthalmologic assessments e.g. optical coherence tomography (OCT) were to be performed if clinically relevant retinal findings were observed on ophthalmologic examinations, and in patients with reported visual AEs or change in visual acuity if the events or changes were suspected to be of retinal origin.

Reviewer Comment: The safety monitoring in Study INCB 54828-202 was generally adequate, with the exception of monitoring for ocular toxicity as described in the corresponding section below.

8.2.3 Adequacy of Applicant’s Clinical Safety Assessments

Issues Regarding Data Integrity and Submission Quality

The data submitted was organized and were generally of adequate quality to perform a comprehensive review of the safety of pemigatinib. Several information requests were sent to Incyte during the review of safety to confirm data, request additional data, request alternative presentations of safety data, or clarify minor discrepancies. For example, toxicity grades were missing for 15 AEs in 13 patients across the trials. Incyte provided timely and adequate responses, including additional analyses and clarifications as required. Data was verified and characterized (see final prescribing information for pemigatinib).

Reviewer Comment: Several adverse event preferred terms in the adverse event dataset are not included in the CTCAE version 4.03, which was used for grading of AEs in the studies submitted to the NDA. Specifically, Vitamin D decreased, disease progression (listed as such even though this should not be considered an AE), general physical health deterioration, hydrocholecystis, and hyperphosphatemia were considered “non-graded” events. For seven of the AEs with missing AE severity grades (gallbladder distension, nausea, vomiting, hiccups, pleural effusion, periodontal disease, hyperphosphatemia), the Applicant and FDA performed additional analyses and assigned a severity grade based upon their assessment. In some instances, AEs such as “general physical health deterioration” or “disease progression” were graded as 5

100 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(fatal); FDA review of the deaths of these patients is described in the Deaths subsection of Section 8.2.4 of this review. Four AEs of hyperphosphatemia were missing a toxicity grade (discussed in the Laboratory Findings subsection of Section 8.2.4 below). The remainder of the AEs with missing toxicity grades are either discussed in other sections of the safety review or were not relevant to product labelling.

Categorization of Adverse Events

Incyte coded verbatim AE terms using The Medical Dictionary for Regulatory Activities (MedDRA) v21.1. According to the Summary of Clinical Safety (SCS) (in Module 2.7.4 of the NDA), treatment-emergent adverse events (TEAEs) were defined as all AEs occurring from the time of initiation of pemigatinib through 30 days after the last dose. Information regarding deaths occurring within 30 days of discontinuation of pemigatinib was also collected.

The clinical reviewer assessed the adequacy of Incyte’s mapping of AE verbatim terms to MedDRA preferred terms (PTs) for 100% of the ISS datasets. The majority of nonidentical terms were due to spelling differences (e.g., anemia versus anaemia), use of abbreviations instead of full text (e.g., ALT increase versus alanine aminotransaminase increased), and verbatim terms that included descriptors (e.g., abdominal cramping versus abdominal pain). Overall, the MedDRA PTs listed in the safety datasets adequately represented the verbatim terms in the CRFs.

FDA’s assessment of the safety and tolerability of pemigatinib was based on the review of the cause and incidence of deaths, AEs of any grade, severe AEs, non-fatal serious adverse events (SAEs), AEs leading to discontinuation, AEs leading to dose reduction or interruption, select “clinically notable” AEs “, clinical laboratory assessments (hematology, serum chemistry, and liver and thyroid function tests), and vital sign measurements. Multiple occurrences of the same event for a patient were counted once at the maximum severity reported. All events were followed until they resolved, the Investigator assessed them as stable or were otherwise explained, or the patient was lost to follow-up, whichever occurred first.

Safety data was available for treatment-emergent AE (TEAE), defined as any event arising or worsening after the start of study drug administration until 30 days after the last administration of pemigatinib. For the purpose of the AE tables, an event was considered related to pemigatinib if it was possibly, probably, or definitely related in the AE electronic CRF (eCRF) form for the individual study. If the investigator did not specify the relationship of the AE to the study drug, then the AE was considered to be treatment-related.

Specifically, regarding ocular toxicity, FDA identified several PTs that were include in the Applicant’s composite term for “Serous Retinal Detachment” that were assessed to be inappropriately included; FDA’s analysis of events meeting the criteria to be considered serous retinal detachment excluded the following PTs: • chorioretinal folds • chorioretinal scar

101 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

• macular edema • retinal edema • retinal exudates • retinal pigmentation.

Safety data including case narratives were available for treatment-related SAEs, “clinically notable TEAEs”, and all listings of AEs include all events that occurred during or after the first study drug treatment up to the data cutoff date. Clinically notable AEs identified by the Incyte included: hyperphosphatemia, hypophosphatemia, serious retinal detachment, and nail toxicity. Customized aggregates of MedDRA PTs to describe TEAEs for which there is a specific clinical interest in connection with the pemigatinib (i.e., clinically notable TEAEs) or that are similar in nature but not identical were summarized by aggregate category and composite terms (i.e., hyperphosphatemia, hypophosphatemia, serous retinal detachment, and nail toxicity) and MedDRA PT. FDA reviewed each clinically notable AE (see Section 8.2.4 for the reviewers conclusions).

An SAE was defined as an event that meets at least 1 of the following criteria: 1. Congenital anomaly or birth defect 2. Significant disability 3. Death 4. Hospitalization 5. Life threatening 6. Other medically important event.

Deaths reported during the study treatment period and those reported during the follow-up period after treatment completion/discontinuation and causes of death were summarized.

For laboratory test results, standard normal ranges were used by Incyte to identify values outside the normal ranges. Abnormal laboratory results were graded using NCI CTCAE v4.03, when possible. Laboratory abnormalities and TEAEs that did not have specific grading outlined in CTCAE v.4.03 were graded on a scale of 1 to 4 as follows:

• Grade 1: mild; asymptomatic or mild symptoms; clinical or diagnostic observations only; intervention not indicated • Grade 2: moderate; minimal, local, or noninvasive intervention indicated; limiting age-appropriate activities of daily living • Grade 3: severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting selfcare activities of daily living • Grade 4: life-threatening consequences; urgent intervention indicated

The NDA included shift analyses of treatment-emergent laboratory abnormalities by baseline grade and maximum post-baseline CTCAE grade. For each laboratory parameter, the baseline

102 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

laboratory value was defined as the last laboratory value collected on or prior to the date of first dose of pemigatinib. Demographics and selected laboratory parameters were also analyzed by hepatic impairment subgroup and renal impairment subgroup.

For the reporting of adverse events for hyperphosphatemia/elevated serum phosphate, sites were instructed to report an adverse event of hyperphosphatemia (greater than 5.5 mg/dL). The threshold of 5.5 mg/dL was a) above the highest upper limit of normal amongst the local lab parameters and b) per hyperphosphatemia management guidelines, when the patients should start supportive care, which included diet modifications. However, many sites did not report an adverse event of hyperphosphatemia for levels above the normal range or even if above 5.5mg/dL if it was not considered clinically significant by the investigator. Therefore, the assessment of hyperphosphatemia included analyses of adverse events as well as elevated phosphate levels recorded as a laboratory abnormality.

Incyte used CTCAE 4.03 to categorize the severity of laboratory abnormalities except for elevations in creatinine. Incyte used three methods to grade the severity of creatinine elevations; the first method assigned grades using the CTCAE v4.03 grading for acute kidney injury. The second method assigned grades based on the baseline value and upper normal limit (ULN) of creatinine. The third method graded for creatinine and assigns grades based only on ULN of creatinine.

Incyte submitted ECG analyses based on data from Study INCB 54828-101.

Incyte presented summaries of clinical laboratory test and vital signs data that relied heavily on Study INCB 54828-202 as compared with the pooled populations.

Reviewer Comment: FDA’s safety analysis generally did not take into account investigator assessment of causality since in open label single arm trials, all TEAEs are considered to be adverse reactions unless clearly due to an alternative cause. ECG data was not submitted from the pivotal study, Study -202; however, there were no cardiac safety signals identified in this application (see detailed review of safety, below). Otherwise, Incyte’s approach to collection, categorization, and analysis of safety data in this application is acceptable, even will the assessment of the AEs that were not assigned a grade initially, and the grading of hyperphosphatemia in laboratory values according to CTCAE 5.0, and creatinine according to ULN.

The discussion of laboratory abnormalities and vital sign data was more comprehensive in the CSR for Study INCB 54828-202 than in the CSRs for the other clinical trials, which is acceptable as Study INCB 54828-202 provided the primary basis to support the safety and effectiveness of pemigatinib for the proposed indication and data from this trial is therefore most reflective of patients that will be treated with pemigatinib in the post-marketing setting.

103 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

During the safety review of this application, FDA submitted information requests (IR) to obtain additional information, clarify the safety data collection procedures used by Incyte, or provide advice regarding the approach to summarizing safety information in product labeling. Incyte and FDA agreed to composite terms (for example, PTs to be grouped for composite terms in ocular toxicity, which is discussed further in “Ocular toxicity” below) which are reflected in patient labelling.

Routine Clinical Tests

Safety evaluations in each of the clinical studies in patients with advanced malignancies included AE and concomitant medication monitoring, clinical laboratory tests, vital signs measurements, physical examinations, ECGs, and eye examinations. Chemistry, hematology, coagulation panel, lipid panel, serology, endocrine function, and urinalysis were assessed locally at screening, Day 1 of every cycle before dose administration, and at the end of therapy. Vital sign measurements include blood pressure, pulse, respiratory rate, and body temperature. Blood pressure and pulse were measured with the subject in the recumbent, semi recumbent, or sitting position after approximately 5 minutes of rest.

Eye examinations included visual acuity tests, slit-lamp examination, and fundoscopy with digital imaging once every 3 cycles (± 14 days), at EOT and as clinically indicated. Additional ophthalmologic assessments (e.g. optical coherence tomography) were to be performed if clinically relevant retinal findings were observed and in participants with reported visual AEs or change in visual acuity if the events or changes were suspected to be of retinal origin.

ECGs were not pooled for the cholangiocarcinoma or the modified safety populations, because the primary evaluation of potential effects of pemigatinib on cardiac function is based on data from Study INCB 54828-101. As stated above, summaries of clinical laboratory test and vital signs data were more comprehensive for Study INCB 54828-202 as compared with the pooled populations.

8.2.4 Safety Results

Deaths

There were 36 patients (8%) in the all cancer population, 7 patients (4.3%) in the all cholangiocarcinoma population, and 6 patients (4.1%) in the Study INCB 54828-202 population who died within 30 days of the last dose of pemigatinib and whose death was possibly related to a TEAE. Causes of death that occurred > 30 days after the last dose of pemigatinib were also reviewed. The clinical reviewer reviewed the narrative summaries and AEs to verify the cause of death described by Incyte for all deaths. A summary of patient narratives for patients whose assigned cause of death was not reported as a preferred term related to underlying disease (such as “general physical health deterioration”) or disease progression is presented in tabular

104 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

form in Table 23.

Table 23: Narrative Description of Selected Deaths

Day of Patient ID/Study/Dose Death AE PT Narrative Death (b) (6) /-202/13.5mg Failure to Thrive 60 63-yr-old female with intrahepatic intermittent cholangiocarcinoma in liver and lymph nodes and prior systemic therapy with cisplatin, fluorouracil, folinic acid, gemcitabine, and oxaliplatin. On Day 43, had CT c/w PD, taken off study, and hospice recommended. On Day 54, the patient had an outpatient paracentesis and requested admission to the hospital for comfort care. (b) (6) /-202/13.5mg Failure to Thrive 125 59-yr-old female with cholangiocarcinoma with intermittent liver, omentum, and peritoneum involvement. Prior systemic therapy with capecitabine, carboplatin, fluorouracil, folinic acid, oxaliplatin, paclitaxel, and pazopanib. On Day 58, pemigatinib was interrupted due to multiple AEs (Grade 3 hypokalemia, Grade 2 proteinuria, Grade 2 knee skin sensitivity, Grade 2 alkaline phosphatase increased, and Grade 3 weight loss), and on Day 93, pemigatinib was restarted at a reduced dose of 9 mg. The patient experienced FTT and PD on Day 114 and pemigatinib was interrupted, On Day 120 the patient experienced constant nausea, vomiting, no oral intake, was admitted to the hospital, required hospice orders on Day 124, and died on Day 125. (b) (6) /-202/13.5mg Pleural effusion 59 66-yr-old female with extrahepatic intermittent cholangiocarcinoma including involvement of the liver, lung, and lymph nodes. Prior systemic therapy with cisplatin and gemcitabine. On Day 35, the patient was admitted to the hospital for increasing abdominal pain, hyponatremia, and ileus. On Day 38, the patient had PD by CT. On Day 44, the patient had a pleural effusion considered due to PD and was placed on hospice care on Day 45. On Day 59, the patient died from PD. (b) (6) /-202/13.5mg Sepsis 64 70-yr-old male with extrahepatic intermittent cholangiocarcinoma with involvement of the liver, pancreas, lymph nodes, gall bladder, and peritoneum. Prior therapy included gemcitabine and insertion of a biliary prosthesis. Admitted to hospital on Day 30 with E. Coli sepsis. On Day 40, PD was confirmed by CT. On Day 41, blood cultures were also positive for Klebsiella, and on

105 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Day of Patient ID/Study/Dose Death AE PT Narrative Death Day 50, blood cultures were positive for multiple organisms. From Day 57 to Day 64, the patient had progressive clinical deterioration with persistent fever, anasarca associated with alteration of the state of consciousness in regard to hepatic insufficiency in the context of disease progression, and bile duct infection. The patient died on Day 64. (b) (6) /-201/13.5mg Cerebrovascular 64 64-yr-old female with urothelial cancer and intermittent accident medical history of obesity, hypothyroidism, and patent ovale foramen. Prior systemic therapy for cholangiocarcinoma included atezolizumab, cisplatin, gemcitabine, and pembrolizumab. On an unspecified date, the patient developed left sided limb paresthesia and later had left sided limb weakness. On Day 59, the patient presented with acute left sided weakness and had difficulty walking. MRI of the brain documented a CVA and showed multiple areas of stroke consistent with a bilateral embolic event. Lipid panel test results included high TGs and drug was withdrawn on Day 62. On Day 64, the patient died. Death was attributed to cancer and CVA. No autopsy was performed. (b) (6) /-201/13.5mg Intestinal infarction 69 71-yr-old male with metastatic urothelial cancer intermittent Peritonitis with involvement of lymph nodes, skin or subcutaneous tissue lung, peritoneum. Prior systemic therapy included cisplatin, gemcitabine, and pembrolizumab. On Day 59 the patient had nausea, vomiting, and abdominal pain. A CT of the abdomen revealed perforation of the cecum with ischemia in the ascending colon and the patient underwent laparotomy with right hemicolectomy, and intestinal infarct diagnosed on Day 61. Small intestine perforation and serorupture were sutured. On Day 65, bronchoscopy revealed mucus, acute inflammation. An ultrasound of the abdomen revealed intra-abdominal collection, right cholecystitis. On Day 68, pemigatinib was permanently withdrawn. On Day 69, the patient died. Autopsy showed macroscopically and microscopically advanced, metastasized disease involving the lungs, colon, liver, prevertebral lymph nodes, paravertebral muscles, and invasion into the aortic wall. (b) (6) /­ Cardiac failure 41 77-yr-old male with urothelial cancer and history 201/continuous of aortic valve disease, cardiac pacemaker insertion, hypertension, knee arthroplasty, osteoarthritis, prostate cancer, and

106 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Day of Patient ID/Study/Dose Death AE PT Narrative Death prostatectomy. Investigators attributed the patient’s Grade 4 cardiac insufficiency to the medical history of valve prosthesis. (b) (6) /-201/13.5mg Cognitive Disorder 8 88-yr-old male with urothelial cancer in lymph intermittent Increased hepatic nodes, bladder, lung, liver, adrenals, omentum, enzyme peritoneum and prior therapy with cisplatin, and gemcitabine. Patient was admitted to the hospital on Day 5 for cognitive deterioration and suspected sepsis as WBC was 20K/µL and AST 262IU/l. Patient died on Day 8 of treatment. (b) (6) /-201/13.5mg Nosocomial 103 75-yr-old male with urothelial cancer and bony intermittent Pneumonia metastases, multiple prior systemic therapies. On Day 84, the patient was admitted to the hospital for hypercalcemia experienced keratitis, and pemigatinib was discontinued. The patient developed a pneumonia during the hospitalization (on an unknown day) and Died on Day 103. (b) (6) -201/13.5mg AKI 33 57-yr-old male with urothelial cancer and intermittent metastases to lymph nodes, bladder, and liver, and multiple prior systemic therapies. On Day 24 experienced ARF, which presented with 48 hours of altered general status. A bladder scan revealed PD and the investigator reported the event as ARF due to PD. Pemigatinib was discontinued on Day 25 and the patient died on Day 33. (b) (6) /-203/13.5mg AKI 104 78-yr-old female with AML, history of cystitis and intermittent increased BUN, and prior therapy with cyclophosphamide, daunorubicin, fludarabine, mitoxantrone, and cytarabine. On Day 104, admitted to the hospital for ARF and bladder obstruction; creatinine was 8.56 mg/dL. The renal failure was postrenal, due to bladder tamponade; a mass was removed with histology consistent with PD. On Day 129 the patient experienced ARF again and on Day 131 the patient died. No autopsy was performed. Source: Reviewed generated tables based on patients narratives in each corresponding CSR. AE=adverse event; AKI=acute kidney injury; AML=acute myeloid leukemia; ARF=acute renal failure; BUN=blood urea nitrogen; CVA=cerebrovascular accident; CT=computed topography; FTT=failure to thrive; MRI=Magnetic Resonance Imaging; TG=triglycerides; WBC=white blood cells, PD=progressive disease.

The causes of death in Study INCB 54828-202 included failure to thrive in 2 patients and bile duct obstruction, cholangitis, sepsis, and pleural effusion each in a single patient. Across the all cancer population only 1 event was considered related to pemigatinib by the investigator: (b) (6) cerebrovascular accident in patient on Study-201 in the all cancer population. Causality assessment for the event was confounded by a concurrent cardiovascular condition (patent foramen ovale), obesity, and hypothyroidism (see narrative in Table 23). Acute respiratory

107 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (6) failure was the only other fatal event in a patient ( on Study 202, see Table 23) with cholangiocarcinoma in a patient in the all cholangiocarcinoma population.

(b) (6) Regarding Patient who developed a cerebrovascular accident (CVA) 59 days after starting pemigatinib, the investigator considered the CVA to be serious (fatal, life threatening, hospitalization) and assessed that there was a reasonable possibility that pemigatinib caused the CVA. The investigator also attributed the death to underlying urothelial cancer in addition to the CVA. In response to an FDA IR sent on February 4, 2020, Incyte clarified that triglyceride levels were not routinely assessed in Study INCB 54828-201, but for this patient, baseline triglycerides were 121 mg/dL, and triglycerides prior to the event were 153 mg/dL (Grade 1). Based on the medical history obtained at the time of enrollment, this patient did not have a prior history of hyperlipidemia, thrombosis, stroke, hypertension, or diabetes. Potential predisposing factors for thrombosis/CVA include obesity (171cm/5’7” and weighed 102Kg/230lbs), hypothyroidism, metastatic cancer, and patent foramen ovale.

(b) (6) Regarding Patient , who developed cognitive changes and increase AST, the investigator reported that the final diagnosis was no other than “cognitive deterioration”. Alternative causality was progression of study indication, specifically PD. The investigator provided rationale for causality assessment to study treatment medication as not suspected relation to study medication, suspected sepsis due to PD.

There were patients in Study -201 and -203 with acute kidney injury (AKI) and acute renal failure injury listed as cause of death, in the context of urothelial cancer (or AML) and progressive disease (see also Table 23, and Table 26). For a full discussion of AKI, see section below on “Acute Kidney Injury”.

(b) (6) FDA reviewed three autopsy reports (patients ) submitted to the NDA; there were no autopsy signs of ectopic mineralization in the three patients for whom autopsy information was available (see Hyperphosphatemia in Laboratory Findings).

Reviewer Comment: Based on review of the patient narratives, the clinical reviewer agrees that the majority of patient deaths were likely related, at least in part, to underlying cancer. However, given the limitations of assessing causality in single arm trials, FDA recommended that the fatal TEAEs of failure to thrive, bile duct obstruction, cholangitis, sepsis, and pleural effusion occurring in Study 202 be included as adverse reactions to pemigatinib in product labeling. The reviewer was unable to make a conclusive assessment regarding whether pemigatinib had a causal role in the death of Patient (b) (6) due to CVA; however, based on the temporal relationship to initiation of pemigatinib and acknowledging the limitations of causality assessments in single arm trials, FDA conservatively attributes this death to pemigatinib. The case of cardiac failure in Patient (b) (6) is confounded by the patient’s history of valve

108 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

prothesis. Because the AEs CVA and MI occurred in a single patient each, occurred outside of Study 202, and are confounded by the patient’s underlying medical condition, FDA decided that these potential safety signals will not be described in product labeling but will be monitored in ongoing clinical trials of pemigatinib and through postmarketing pharmacovigilance.

The overall incidence of TEAEs with a fatal outcome in the all cancer population was higher than the overall incidences for Study INCB 54828-202 and the all cholangiocarcinoma population, which is likely a reflection of treatment effect of pemigatinib in patients with FGFR2-fusion positive cholangiocarcinoma.

Serious Adverse Events

Every SAE, regardless of suspected causality (e.g., relationship to study drug(s) or study procedure or disease progression), occurring after the patient has signed the ICF through the last study visit (or 30 days after the last dose of study drug, whichever is later) was reported.

The analyses of SAEs presented by Incyte are based on the adverse event dataset in the modified safety population using the data included in the NDA submission for patients with a TEAE that met any of the SAE criteria, irrespective of whether the AE was judged to be related to study drug.

SAEs occurred in 42% patients in the all cancer population, and in 45% of patients in Study-INCB 54828-202. The most frequently reported SAEs by preferred term (PT).

Table 24 provides a summary of the per-patient incidence of SAEs regardless of causality by PT in order of decreasing frequency.

Table 24: Serious Adverse Events by Preferred Term ≥ 1% incidence in any safety population

All All cancer# INCB54828-202 cholangiocarcinoma$ N = 466 N = 146 PT N = 161 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 n (%) n (%) n (%) n (%) n (%) n (%) Any SAE 194 (42) 181 (39) 67 (42) 59 (37) 65 (45) 57 (39) Lung Infection1 22 (4.7) 20 (4.3) 8 (5) 7 (4.3) 9 (6) 8 (6) Urinary tract infection 15 (3.2) 13 (2.8) 3 (1.9) 2 (1.2) 3 (2.1) 2 (1.4) Acute kidney injury 14 (3) 12 (2.6) 2 (1.2) 2 (1.2) 3 (2.1) 3 (2.1) Abdominal pain 13 (2.8) 11 (2.4) 7 (4.3) 6 (3.7) 7 (4.8) 6 (4.1) Pyrexia 13 (2.8) 4 (0.9) 7 (4.3) 1 (0.6) 7 (4.8) 1 (0.7) Pain2 9 (1.9) 9 (1.9) 3 (1.9) 3 (1.9) 2 (1.4) 2 (1.4) General physical health deterioration 7 (1.5) 7 (1.5) 0 0 0 0

109 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

All All cancer# INCB54828-202 cholangiocarcinoma$ N = 466 N = 146 PT N = 161 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 n (%) n (%) n (%) n (%) n (%) n (%) Hematuria 7 (1.5) 6 (1.3) 0 0 0 0 Cholangitis 7 (1.5) 6 (1.3) 5 (3.1) 4 (2.5) 5 (3.4) 4 (2.7) Fatigue/Asthenia 7 (1.5) 6 (1.3) 3 (1.9) 2 (1.2) 2 (1.4) 2 (1.4) Constipation 6 (1.3) 2 (0.4) 0 0 0 0 Small intestinal obstruction 6 (1.3) 5 (1.1) 2 (1.2) 2 (1.2) 3 (2.1) 3 (2.1) Dehydration 6 (1.3) 3 (0.6) 3 (1.9) 1 (0.6) 2 (1.4) 1 (0.7) Nausea 5 (1.1) 3 (0.6) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Vomiting 5 (1.1) 4 (0.9) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Urosepsis 5 (1.1) 5 (1.1) 0 0 0 0 Intestinal obstruction 4 (0.9) 4 (0.9) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Hypotension 4 (0.9) 4 (0.9) 2 (1.2) 2 (1.2) 1 (0.7) 1 (0.7) Esophageal varices hemorrhage 3 (0.6) 3 (0.6) 2 (1.2) 2 (1.2) 1 (0.7) 1 (0.7) Bacteremia 3 (0.6) 2 (0.4) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Cholangitis infective 3 (0.6) 2 (0.4) 3 (1.9) 2 (1.2) 3 (2.1) 2 (1.4) Sepsis 3 (0.6) 3 (0.6) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Failure to thrive 3 (0.6) 3 (0.6) 3 (1.9) 3 (1.9) 3 (2.1) 3 (2.1) Ascites 2 (0.4) 2 (0.4) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Gastrointestinal hemorrhage 2 (0.4) 2 (0.4) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Chills 2 (0.4) 1 (0.2) 2 (1.2) 1 (0.6) 2 (1.4) 1 (0.7) Bile duct obstruction 2 (0.4) 2 (0.4) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Device occlusion 2 (0.4) 1 (0.2) 2 (1.2) 1 (0.6) 2 (1.4) 1 (0.7) Source: adae.xpt. Variables used: USUBJID, TRT01A, AEDECOD, AEBODSYS, AETOXGR, AESER. #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib 1Lung infection includes: Pleural effusion, Pneumonia, Pneumonia aspiration, and Pneumonitis 2Pain includes: Back pain, and Pain in extremity.

The FDA review included assessment of all SAEs for each safety population. Table 25 provides a summary of patient narratives for select SAEs with PTs of interest that were reflected in patient labelling (excluding patient deaths, which are summarized in Table 23, and Grade ≥ 3 AEs, which are summarized in Table 33).

110 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 25: Summary of Patient Narratives for Serious Adverse Events

Study Day Patient ID/Study/Dose SAE PT Narrative of Onset 64-yr-old male with urothelial cancer and nephrostomy tubes, metastases to the bone, bladder, lymph nodes, lung, and liver and prior systemic therapy with carboplatin and gemcitabine. On Day 54 the patient presented with bilateral trichiasis. On Day 61 he was admitted to the hospital with fatigue and diagnosed with bacteremia and a UTI, (b) (6) - AKI 89 hyperkalemia and hyperphosphatemia. On 201/intermittent/13.5mg Optic neuropathy 112 Day 89 routine bloodwork showed increased creatinine 6.7mg/dL (normal range not provided) and the patient was diagnosed with chronic renal failure. On Day 112 the patient experienced bilateral optic neuropathy and vision loss, considered by the investigator to be related to pemigatinib and the event improved after study drug discontinuation. The patient discontinued the trial on Day 211 due to death. 70-yr-old male with urothelial cancer and metastases to the lymph nodes, corpus cavernosum, and pleura. Prior systemic therapy included carboplatin, cisplatin, gemcitabine, paclitaxel, and surgery. On Day 35 the patient (b) (6) - experienced blurred vision and was diagnosed RPED 35 201/intermittent/13.5mg with central serous chorioretinopathy; pemigatinib was withdrawn on Day 45. No interventions or medications were administered and on Day 58, the RPED resolved (dechallenge). The patient discontinued from the trial on Day 49. 54-yr-old male with intrahepatic cholangiocarcinoma and liver, lung, and lymph nodes metastases. Prior systemic therapy included fluorouracil, folinic acid, irinotecan, and paclitaxel albumin. On Day 125 and Day 244 the patient had grade 3 abdominal pain. The last dose of pemigatinib was on Day (b) (6) - 343; the patient discontinued study treatment Increased Cr 358 202/intermittent/13.5mg due to PD. On Day 358 routine bloodwork revealed increased creatinine (highest level was 4.4mg/dL [reference range 0.7-1.3 mg/dL]) and was diagnosed with AKI of unknown etiology. He was admitted to the hospital for “further management”. On Day 362 serum creatinine was 2.7 mg/dL and continued to fall to within normal limits. (b) (6) - AKI 60-yr-old female with extrahepatic 60 202/intermittent/13.5mg Increased Cr cholangiocarcinoma and liver, lymph node and

111 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study Day Patient ID/Study/Dose SAE PT Narrative of Onset perihepatic metastases, and renal impairment at baseline. Baseline creatine was 0.9mg/dL. On Day 8 the patient had a creatinine level of 3.3 mg/dL and hyperkalemia and was hospitalized. Pemigatinib was withdrawn and the patient’s serum creatinine returned to baseline after being off pemigatinib for 7 days on Day 15. The patient discontinued study on Day 21 due to “rapid clinical progression” and died on Day 46 due to PD. The patient was not rechallenged. The investigator reported that there is not a reasonable possibility that pemigatinib caused the AKI and considered the AKI to be due to PD. 73-yr-old female with intrahepatic cholangiocarcinoma and liver, lung, lymph nodes, portacaval, and peritoneal metastases. On Day 290, the subject developed retinal (b) (6) detachment of her right eye, and pemigatinib - Retinal detachment 290 was withheld. On Day 294, corrective surgery 202/intermittent 13.5mg was performed. On Day 295, slit lamp and fundus exams were normal. On Day 314, pemigatinib was restarted. On Day 316, slit lamp and fundus exams were normal, and retina was attached. 63-yr-old female with intrahepatic cholangiocarcinoma and metastases to bone liver and lung. On Day 56, the patient developed left acute partial vision loss. A fundus examination performed on that day (b) (6) - Optic Ischemic revealed a swollen superior aspect of the left 56 202/intermittent 13.5mg Neuropathy optic nerve consistent with ischemic optic neuropathy. The outcome of the left non arteritic ischemic optic neuropathy was reported as not recovered/not resolved as of Day 465. The action taken with pemigatinib was reported as continued unchanged. 55-yr-old female with intrahepatic cholangiocarcinoma and metastases to bone, liver, lung, lymph nodes, ascites, pleural effusion, and adrenal glands. Prior therapy was with cisplatin, fluorouracil, gemcitabine, and oxaliplatin. She had recurrent malignant ascites (b) (6) - UTI 3 at baseline which was drained every five days at 202/intermittent 13.5mg AKI 34 the time of the report and on Day 3 she complained of burning and difficulty with urination and had urine > 100K CFU Escherichia coli. and a blood culture positive for micrococcus. On Day 34 the patient had a creatinine of 1.95mg/dL and was treated with Bactrim for presumed UTI. Pemigatinib was

112 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Study Day Patient ID/Study/Dose SAE PT Narrative of Onset held on Day 35 due to AKI. On Day 39 creatinine was 2.28 mg/dL patient was started on ciprofloxacin due to urine culture result indicating organism was Bactrim-resistant. Day 42 creatinine was 2.42 mg/dL, and pemigatinib was withheld. Day 46 creatinine was 3.47 mg/dL and renal ultrasound revealed right renal cyst and ascites. On Day 47, paracentesis was performed and 5 liters of fluid were removed. On Day 56 the patient died. 59-yr-old female with intrahepatic cholangiocarcinoma and metastases to liver, lung and lymph nodes. The patient had ongoing thrombotic events during treatment with pemigatinib, including an incidental pulmonary embolism with onset on Day 41 and a deep vein thrombosis with onset on Day 85 in addition to low PTT at baseline and on Day 43. On Day 168, a routine eye examination revealed a central retinal artery occlusion with emboli of the right eye. Visual acuities were 20/20 (left eye) and not assessed (right eye; baseline visual acuities were 20/20 bilaterally), and pemigatinib was withheld. Treatment of the central retinal Retinal Artery artery occlusion with 0.5% Glaucon Ophthalmic (b) (6) Occlusion 167 (timolol maleate) at 0.05 ml, BID was initiated -202/intermittent Embolic Cerebral 174 for a week.. On Day 174 the event was not 13.5mg Infarction 22 resolved. On Day 174, she had acute alteration of consciousness and choked on water. On Day 182, an examination showed macular ischemia of the right eye, with visual acuities of 20/30 (left eye) and not assessed (right eye). On Day 182, a CT of the chest and whole abdomen revealed a new lesion in the spleen and pemigatinib was permanently discontinued. A brain CT revealed evidence of cerebral infarction in the left parietal lobe, left external capsule and left corona radiata. On Day 212, vision had not recovered and the patient died on Day 266 due to PD. The investigator considered these SAEs unrelated to pemigatinib and likely due to emboli caused by underlying cancer. Source: reviewer generated table from CSR -201, -202, -203. PD=progressive disease, AKI=acute kidney injury, ARF=acute renal failure, mets=metastases, RPED=retinal pigment epithelial detachment; UTI=urinary tract infection

(b) (6) One patient with an indwelling Foley catheter was diagnosed with a UTI while on vacation but did not fill the prescription for Bactrim. This patient subsequently developed a “kidney infection”, which was reported as SAE. This SAE was considered not related to 113 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

pemigatinib and was attributed to the patient’s underlying predisposition to development of a UTI due to the Foley catheter and delay in treatment of diagnosed cystitis.

Reviewer Comment: In general, the FDA reviewer considers the investigator attribution of SAEs to be generally accurate.. The majority of SAEs appear to be either wholly or in part related to underlying cancer and disease progression (e.g., sepsis from cholangitis in patients with underlying cholangiocarcinoma). Based upon the temporal relationship between onset of the SAE and the mechanism of action of pemigatinib, there is a reasonable possibility that pemigatinib played a causal role in development of some SAEs, such as those related to ocular toxicity. This reviewer therefore does not agree with investigator assessment that retinal detachment was not related to pemigatinib in Patient (b) (6) . This reviewer generally agrees with the investigator assessment that AKI was unrelated to pemigatinib for Patient (b) (6) (b) (6) and Patient (b) (6) but there was no rechallenge data to assess further. Because attribution may be unreliable in single arm trials, these SAEs are described as adverse reactions in the package insert for pemigatinib.

Dropouts and/or Discontinuations Due to Adverse Effects

In the all cancer population (N=466), of the 386 patients (83%) off study treatment at the time of database cutoff (multiple studies with multiple data cut off dates and the most recent, Study INCB54828-202 one was 22 Mar 2019), the most common reason for discontinuation was disease progression (n=272, 58%) with 31 (7%) of patients discontinuing due to an AE. In Study INCB54828-202, of the N=146 patients in the safety database, 115 (79%) patients discontinued treatment by the database cutoff (22 Mar 2019); Of these 115 patients, 13 (9%) discontinued due to AE and 85 (58%) discontinued due to PD.

Patient narratives and AEs resulting in discontinuation of pemigatinib were reviewed. Notable adverse reactions requiring permanent discontinuation in ≥1% of patients who received pemigatinib included acute kidney injury and intestinal obstruction (Table 26).

Table 26: Adverse Events Leading to Drug Discontinuation in ≥1% Incidence in any Safety Population

Trial arm All All cancer# INCB54828-202 cholangiocarcinoma@ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Any TEAE 43 (9) 32 (7) 13 (8) 12 (8) 13 (9) 11 (8) Acute kidney injury 4 (0.9) 4 (0.9) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Intestinal obstruction 2 (0.4) 2 (0.4) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4)

114 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Trial arm All All cancer# INCB54828-202 cholangiocarcinoma@ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Small intestinal obstruction 2 (0.4) 1 (0.2) 0 0 0 0 Dehydration 2 (0.4) 0 1 (0.6) 0 0 0 Pneumonia 2 (0.4) 2 (0.4) 0 0 0 0 Lung Infection& 2 (0.4) 2 (0.4) 0 0 0 0 RPED^ 4 (0.9) 3 (0.6) 0 0 0 0 Dysesthesia% 2 (0.4) 1 (0.2) 0 0 0 0 Ascites 1 (0.2) 0 0 0 0 0 Diarrhea 1 (0.2) 1 (0.2) 0 0 0 0 Gastrointestinal hemorrhage 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Intestinal infarction 1 (0.2) 1 (0.2) 0 0 0 0 Obstruction gastric 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Stomatitis 1 (0.2) 0 0 0 0 0 Cerebrovascular accident 1 (0.2) 1 (0.2) 0 0 0 0 Embolic cerebral infarction 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Hyperesthesia 1 (0.2) 0 0 0 0 0 Paraplegia 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Spinal cord compression 1 (0.2) 1 (0.2) 0 0 0 0 Ulcerative keratitis 1 (0.2) 1 (0.2) 0 0 0 0 Calciphylaxis 1 (0.2) 0 0 0 0 0 Hyponatremia* 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 0 0 Starvation 1 (0.2) 1 (0.2) 0 0 0 0 Amylase increased* 1 (0.2) 0 0 0 0 0 Blood alkaline phosphatase increased* 1 (0.2) 1 (0.2) 0 0 0 0 Blood creatinine increased* 1 (0.2) 1 (0.2) 0 0 0 0 Glomerular filtration rate decreased 1 (0.2) 0 0 0 0 0 Lipase increased* 1 (0.2) 0 0 0 0 0 Disease progression 1 (0.2) 1 (0.2) 0 0 0 0 General physical health deterioration 1 (0.2) 1 (0.2) 0 0 0 0 Multiple organ dysfunction syndrome 1 (0.2) 1 (0.2) 0 0 0 0

115 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Trial arm All All cancer# INCB54828-202 cholangiocarcinoma@ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Bile duct obstruction 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Cholangitis 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Hyperbilirubinemia* 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Brain cancer metastatic 1 (0.2) 1 (0.2) 0 0 0 0 Malignant ascites 1 (0.2) 0 1 (0.6) 0 1 (0.7) 0 Malignant neoplasm progression 1 (0.2) 1 (0.2) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Anemia* 1 (0.2) 0 0 0 0 0 Fatigue/Asthenia# 1 (0.2) 0 1 (0.6) 0 0 0 Dry Eye$ 1 (0.2) 0 0 0 0 0 Palmar-plantar erythrodysesthesia 1 (0.2) 1 (0.2) 0 0 0 0 syndrome Respiratory failure 1 (0.2) 1 (0.2) 0 0 0 0 Eyelash Changes% 1 (0.2) 0 0 0 0 0 Performance status decreased 0 0 0 0 1 (0.7) 0

Source: adae.xpt, adsl.xpt. Variables used: USUBJID, TRT01A, AEDECOD, AEBODSYS, AETOXGR, AEACN. #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) @subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib &Lung infection includes: Pneumonia ^RPED (Retinal Pigment Epithelial Detachment) includes: chorioretinopathy, detachment of retinal pigment epithelium, maculopathy, retinopathy, retinal detachment or disorder, retinal thickening, subretinal fluid %Dysesthesia includes: Hyperaesthesia, and Palmar-plantar erythrodysaesthesia syndrome #Fatigue/Asthenia includes: Fatigue. $Dry eye includes: Dry eye, keratitis, lacrimation increased, meibomian gland dysfunction, pinguecula, and punctate keratitis %Eyelash changes includes: eyelash changes, Trichiasis, trichomegaly *Should be interpreted in context of ADLB dataset, see “Laboratory Findings”

In the all cancer population (N=466), 4 patients (0.9%) discontinued pemigatinib due to AKI, defined as a creatinine clearance ≤ 30 mL/min based on Cockcroft-Gault, including 2 patients (1.4%) in Study INCB54828-202. The patients who discontinued pemigatinib due to AKI are described below.

(b) (6) • Patient : a 60-year-old female with metastatic extrahepatic cholangiocarcinoma involving the liver, peri-hepatic area, and lymph nodes and renal impairment at baseline (creatinine 0.9mg/dL). On Day 8 the patient had a creatinine level of 3.3 mg/dL (normal range: 0.6-1.1 mg/dL), elevated BUN (76 mg/dL), and hyperkalemia, and was admitted to the hospital with AKI. Pemigatinib was withdrawn.

116 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The SAEs of AKI and elevated creatinine resolved on Day 13, and on Day 15 her serum creatinine level was back to baseline at 0.9 mg/dL, and BUN was 14 mg/dL. The patient was taken off drug on Day 22 due to “rapid clinical progression” and died on Day 46 due to PD; therefore she was not rechallenged. The investigator reported that there is not a reasonable possibility that pemigatinib caused the elevated creatinine and AKI, and attributed both to PD (see Table 25, the discussion under Acute Kidney Injury in Significant Adverse Events, and Laboratory Findings for additional discussion of this case).

(b) (6) • Patient : a 55-year-old female with intrahepatic cholangiocarcinoma and metastases to bone, liver, lung, lymph nodes, ascites, pleural effusion, and adrenal glands. Prior therapy included cisplatin, fluorouracil, gemcitabine, and oxaliplatin and the episode of AKI occurred in the context of an untreated UTI (see Table 25 and the discussion under Acute Kidney Injury in Significant Adverse Events for additional discussion of this case).

(b) (6) • Patient : a 57-year-old male with urothelial cancer and metastases to lymph nodes, bladder, and liver, who had received multiple prior systemic therapies. On Day 24 the patient experienced ARF, which presented with 48 hours of altered “general status” associated with somnolence. Pemigatinib was discontinued on Day 25. A bladder scan revealed PD. The investigator attributed the ARF to PD. The patient died on Day 33 (Table 23).

(b) (6) • Patient : a 78-year-old female with AML, history of cystitis and increased BUN, and prior therapy with cyclophosphamide, daunorubicin, fludarabine, mitoxantrone, and cytarabine. On Day 104, the patient was admitted to the hospital for ARF and bladder obstruction; his serum creatinine was 8.6 mg/dL. The reason for renal failure was reported as postrenal due to bladder tamponade, and subsequently a mass was removed with histology consistent with PD. On Day 129 the patient experienced ARF again and on Day 131 the patient died. No autopsy was performed (Table 23). Related cases that were not reported as AKI but have similar clinical significance are described below.

(b) (6) • Patient : 54-year-old male with intrahepatic cholangiocarcinoma and liver, lung, and lymph nodes metastases. Prior systemic therapy included fluorouracil, folinic acid, irinotecan, and paclitaxel albumin. He discontinued pemigatinib on Day 343 due to PD. On Day 358, the patient had an elevated creatinine of 4.4mg/dL and was admitted to the hospital. On Day 362, the creatinine was 2.7 mg/dL, and ultimately returned to a normal level.

(b) (6) • Patient had urothelial cancer and a history of chronic kidney disease. This patient discontinued pemigatinib due to a grade 2 TEAE of decreased glomerular filtration rate.

117 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The 2 cases of intestinal obstruction were reviewed; in both cases, the cause of intestinal (b) (6) obstruction was likely due to intra-abdominal PD and its sequelae (for example, patient (b) (6) developed malignant ascites). (b) (6) A single patient (Patient ) discontinued pemigatinib due to hyperbilirubinemia. Patient (b) (6) was a 42-year-old female with intrahepatic cholangiocarcinoma with metastases to liver and lymph nodes. On Day 147, she had a nonserious AE of Grade 3 hyperbilirubinemia. Laboratory values on that day included bilirubin 63 μmol/L (normal range: 1.7-20.5 μmol/L), direct bilirubin 23.9 μmol/L (normal range: 0-6.8 μmol/L), AST 46 U/L (normal range: 0-32 U/L), ALT 41 U/L (normal range: 0-31 U/L), and alkaline phosphatase 463 U/L (normal range: 30-120 U/L). Pemigatinib was interrupted on Day 293, and eventually permanently discontinued on Day 329 as a result of the event of hyperbilirubinemia, which remained ongoing in the absence of drug rechallenge. The patient died on Day 497 (this case is further described in the discussion of hepatotoxicity in the Laboratory Findings section of this review). There were 2 patients (0.4%) in the all cancer safety population who discontinued pemigatinib due to retinal pigment epithelial detachment (RPED). One patient (0.2%) each discontinued for keratitis (composite term is dry eye), and trichiasis (composite term: eyelash changes). No patients with cholangiocarcinoma discontinued pemigatinib due to ocular toxicity. Notable patients with ocular toxicities are described in the Significant Adverse Events: Ocular Toxicity section of this review. In Study INCB54828-202, dosage interruptions due to an adverse reaction occurred in 43% of patients. Adverse reactions requiring dosage interruption in ≥ 1% of patients who received pemigatinib are listed in Table 27. Note that drug interruptions due laboratory values (i.e. “hypercalcemia” and “hyperphosphatemia” will be discussed in the Laboratory Findings section of this review, are not discussed in the setting of AE PTs). Table 27: Treatment-Emergent Adverse Events leading to Treatment Interruption ≥ 1% Incidence in Any Safety Population

Safety population All All cancer# INCB54828-202 cholangiocarcinoma$ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Any TEAE 201 (43) 129 (28) 68 (42) 48 (30) 62 (43) 46 (32) Stomatitis 23 (4.9) 16 (3.4) 11 (7) 5 (3.1) 11 (8) 5 (3.4) Palmar-plantar 16 (3.4) 8 (1.7) 10 (6.2) 6 (3.7) 8 (6) 4 (2.7) erythrodysaesthesia syndrome Fatigue 14 (3) 8 (1.7) 6 (3.7) 2 (1.2) 6 (4.1) 2 (1.4) Abdominal pain 10 (2.1) 6 (1.3) 4 (2.5) 4 (2.5) 4 (2.7) 4 (2.7) Decreased appetite 10 (2.1) 3 (0.6) 3 (1.9) 1 (0.6) 2 (1.4) 1 (0.7)

118 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Safety population All All cancer# INCB54828-202 cholangiocarcinoma$ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Aspartate aminotransferase 10 (2.1) 7 (1.5) 3 (1.9) 3 (1.9) 3 (2.1) 3 (2.1) increased* Acute kidney injury 10 (2.1) 5 (1.1) 1 (0.6) 0 2 (1.4) 1 (0.7) Arthralgia 9 (1.9) 8 (1.7) 7 (4.3) 7 (4.3) 7 (4.8) 7 (4.8) Onycholysis 8 (1.7) 4 (0.9) 1 (0.6) 1 (0.6) 0 0 Diarrhea 7 (1.5) 1 (0.2) 2 (1.2) 1 (0.6) 2 (1.4) 1 (0.7) Urinary tract infection 7 (1.5) 5 (1.1) 0 0 0 0 Asthenia 7 (1.5) 2 (0.4) 2 (1.2) 0 3 (2.1) 0 Pyrexia 6 (1.3) 2 (0.4) 3 (1.9) 0 3 (2.1) 0 Alanine aminotransferase 6 (1.3) 5 (1.1) 3 (1.9) 2 (1.2) 3 (2.1) 2 (1.4) increased* Constipation 5 (1.1) 2 (0.4) 0 0 0 0 Vomiting 5 (1.1) 0 1 (0.6) 0 1 (0.7) 0 Onychomadesis 5 (1.1) 0 3 (1.9) 0 2 (1.4) 0 Back pain 5 (1.1) 4 (0.9) 3 (1.9) 2 (1.2) 2 (1.4) 1 (0.7) Nausea 4 (0.9) 1 (0.2) 1 (0.6) 0 1 (0.7) 0 Dehydration 4 (0.9) 2 (0.4) 2 (1.2) 1 (0.6) 2 (1.4) 1 (0.7) Hypercalcemia* 4 (0.9) 1 (0.2) 1 (0.6) 0 2 (1.4) 0 Paronychia 4 (0.9) 3 (0.6) 2 (1.2) 2 (1.2) 1 (0.7) 1 (0.7) Pneumonia 4 (0.9) 3 (0.6) 1 (0.6) 1 (0.6) 1 (0.7) 1 (0.7) Cholangitis 4 (0.9) 3 (0.6) 3 (1.9) 2 (1.2) 3 (2.1) 2 (1.4) Hypotension 4 (0.9) 3 (0.6) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Small intestinal obstruction 3 (0.6) 3 (0.6) 2 (1.2) 2 (1.2) 3 (2.1) 3 (2.1) Blood alkaline phosphatase 3 (0.6) 3 (0.6) 3 (1.9) 3 (1.9) 3 (2.1) 3 (2.1) increased* Electrocardiogram QT prolonged 3 (0.6) 0 2 (1.2) 0 2 (1.4) 0 Pain in extremity 3 (0.6) 2 (0.4) 3 (1.9) 2 (1.2) 2 (1.4) 1 (0.7) Syncope 3 (0.6) 3 (0.6) 2 (1.2) 2 (1.2) 2 (1.4) 2 (1.4) Source: safety review team; adae.xpt, adsl.xpt. Variables used: USUBJID, TRT01A, AEDECOD, AEBODSYS, AETOXGR, AEACN. #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib

119 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

*are to be assessed in the ADLB dataset, see “Laboratory Findings”

Dose reductions due to an adverse reaction in Study INCB54828-202 occurred in 14% of patients. Table 28 provides a list of TEAE leading to dose reduction in each safety population.

Table 28: Treatment-Emergent Adverse Events leading to dose reduction ≥ 1% Incidence in any Safety Population

Safety Population All All cancer# INCB54828-202 cholangiocarcinoma$ TEAE N = 466 N = 146 N = 161 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 Grade 1-5 Grade 3-4 n (%) n (%) n (%) n (%) n (%) n (%) Any TEAE 70 (15) 21 (4.5) 23 (14) 6 (3.7) 20 (14) 5 (3.4) Stomatitis 15 (3.2) 6 (1.3) 5 (3.1) 1 (0.6) 5 (3.4) 1 (0.7) Palmar-plantar erythrodysaesthesia 7 (1.5) 3 (0.6) 6 (3.7) 2 (1.2) 5 (3.4) 1 (0.7) syndrome Fatigue 6 (1.3) 2 (0.4) 1 (0.6) 0 0 0 Arthralgia 6 (1.3) 1 (0.2) 5 (3.1) 1 (0.6) 5 (3.4) 1 (0.7) Diarrhea 5 (1.1) 1 (0.2) 0 0 0 0 Dry skin 4 (0.9) 0 1 (0.6) 0 1 (0.7) 0 Onychomadesis 4 (0.9) 1 (0.2) 2 (1.2) 0 2 (1.4) 0 Dry mouth 3 (0.6) 0 0 0 0 0 Alopecia 3 (0.6) 0 1 (0.6) 0 0 0 Nail toxicity 3 (0.6) 1 (0.2) 1 (0.6) 0 1 (0.7) 0 Onycholysis 3 (0.6) 0 1 (0.6) 0 0 0 Asthenia 3 (0.6) 0 2 (1.2) 0 2 (1.4) 0 Decreased appetite 3 (0.6) 1 (0.2) 1 (0.6) 0 1 (0.7) 0 Source: safety team; adae.xpt, adsl.xpt. Variables used: USUBJID, TRT01A, AEDECOD, AEBODSYS, AETOXGR, AEACN. #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib

Reviewer Comment: The incidences of drug discontinuations, interruptions and reductions are relatively low and are supportive of the safety of the proposed dosage regimen in patients with refractory cholangiocarcinoma. It is difficult to reliably assess whether pemigatinib had a causal role for some TEAEs given the single arm nature of the trials providing safety data. Oncologists are familiar with instituting dosage modifications for treatment-related toxicities dosage modification instructions are provided in the pemigatinib package insert for the most common and serious adverse reactions to pemigatinib (Section 2). Note that the PT listed above are from AE datasets and not the laboratory value dataset (ADLB). For the review the dose interruptions 120 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

due to laboratory values (specifically due to hyperphosphatemia and hepatotoxicity), see the Laboratory Findings section of this review.

Significant Adverse Events

The ICH E3 guidance recommends that marked laboratory abnormalities not meeting the definition of SAEs also be considered significant AEs. These laboratory abnormalities are described in the Laboratory Findings section of this review.

In addition, the ICH E3 guidance considers other potentially important abnormalities, such as severe AEs (i.e., adverse events of ≥ Grade 3 severity as graded by the NCI CTCAE criteria that do not meet the definition of a serious AE) as potentially significant.

Hyperphosphatemia and Hypophosphatemia See the Laboratory Findings of this review for discussion of hyperphosphatemia and hypophosphatemia.

Ocular Toxicity Serous retinal detachment denotes an accumulation of fluid within a virtual space known as the subretinal space, which is located between the retinal photoreceptors and the apical retinal pigment epithelium, leading to separation of the photoreceptor outer segment from the apical retinal pigment epithelium without signs of inflammation (Nti et al 2019, van der Noll et al 2013). The mechanism of serous retinal detachment usually involves an incompetent inner and/or outer retinal barrier, or a defect at the level of the retinal pigment epithelium leading to altered homeostasis within the subretinal space. The prototypical idiopathic serous retinal detachment is central serous chorioretinopathy. In central serous chorioretinopathy, the location and magnitude of the abnormalities determine whether they are subclinical or, more frequently, present as painless blurred vision, distortion of images, and/or the emergence of a visual field defect. Serous retinal detachment/central serous chorioretinopathy is frequently reported for inhibitors of the mitogen-activated protein kinase (MEK/MAPK) pathway and is sometimes also referred as MEK inhibitor–associated retinopathy. The mechanism(s) by which inhibition of MEK and MAPK pathways produces subretinal fluid and how MEK inhibitor– associated retinopathy is different from central serous chorioretinopathy are not fully understood (Francis et al 2017). There is evidence supporting that inhibition of the MAPK pathway interferes with the maintenance of the outer retinal barrier and/or phagocytic, immunologic and/or pump function of the retinal pigment epithelium (Nti et al 2019). Inhibition of FGFR may lead to a similar interference with the integrity of the outer retinal barrier and/or function of the retinal pigment epithelium, as FGFR2 is expressed in retina and is important in the function of retinal pigmented epithelial cells (Rosenthal et al 2005).

The Division of Anti-infective and Ophthalmology Drug Products was consulted to assist with the evaluation of ocular toxicity and to provide advice regarding the description of ocular toxicity in the Warnings and Precautions section and dosage modification instructions for ocular toxicity in the Dosage and Administration section of pemigatinib product labeling. The 121 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

ophthalmology consultant recommended reclassification of PTs to more accurately reflect the ocular class effects of FGFR inhibitors. The PTs and composite terms used for the evaluation of ocular toxicities are described below.

• The composite term for Retinal Pigment Epithelial Detachment (RPED) includes the following PTs: • chorioretinopathy • detachment of retinal pigment epithelium • maculopathy • retinopathy • retinal detachment or disorder • retinal thickening • serous retinal detachment • subretinal fluid • The “Dry eye” composite term includes: Dry eye, meibomian gland dysfunction and lacrimation increased, keratitis and punctate keratitis, any terms related to the growth of pinguecula • Conjunctivitis allergic and eye pruritis • Eye pain and eye irritation • Eyelash changes, growth of eyelashes, trichiasis and trichomegaly • Vision blurred, visual impairment and visual acuity reduced • Vitreous detachment and vitreous floaters

No patients discontinued pemigatinib due to ocular toxicity in Study INCB 54828-202; however, 4 patients (0.9%) discontinued pemigatinib in the all cancer safety population due to ocular toxicity. Two patients (0.4%) discontinued in the all cancer population due to RPED, specifically, 1 patient each for detachment of retinal pigment epithelium (0.2%), and subretinal fluid (0.2%). One patient (0.2%) in the all cancer population discontinued due to keratitis (composite term “dry eye”) and 1 patient (0.2%) discontinued due to trichiasis (composite term “eyelash changes”). In Study -202, the median time of onset for dry eyes was 53 days (7.6 weeks) and for RPED was 62 days. One patient in Study INCB 54828-202 had a dose reduction while the same 0.4% patients with discontinuations in the all cancer population had also had a dose reduction.

Notable cases of ocular toxicity are summarized below.

(b) (6) • Patient : 83-year-old male with urothelial cancer on pemigatinib intermittent dosing experienced Grade 1 non serous subretinal fluid, macular degeneration, and retinal disorder (RPED) on Day 82. Pemigatinib was interrupted and finally discontinued on Day 106 due to the subretinal fluid.

(b) (6) • Patient : 70-year-old male with urothelial cancer and metastases to the lymph nodes, corpus cavernosum, and pleura and prior systemic therapy with

122 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

carboplatin, cisplatin, gemcitabine, paclitaxel, and surgery who experienced blurred vision and was diagnosed with central serous chorioretinopathy (RPED) on Day 35. Pemigatinib was withdrawn on Day 45. No interventions or medications were administered and on Day 58, RPED resolved (positive dechallenge). The patient discontinued the trial on Day 49 due to ocular toxicity. Further details of this SAE are provided in Table 25.

(b) (6) • Patient : 75-year-old male with urothelial cancer who was hospitalized for hypercalcemia and keratitis on Day 84 and pemigatinib was discontinued due to keratitis. He received ophthalmic treatment with fluorometholone and eye lubricants as needed Day 1 to Day 84 to treat the keratitis. There was no rechallenge.

(b) (6) • Patient : 63-year-old female with intrahepatic cholangiocarcinoma and metastases to bone liver and lung who developed left acute partial vision loss on Day 56. A fundus examination that day revealed a swollen superior aspect of the left optic nerve consistent with ischemic optic neuropathy. The outcome of the left non arteritic ischemic optic neuropathy was reported as not recovered/not resolved as of Day 465. The action taken with study drug was reported as continued unchanged. Further details of this SAE are provided in Table 25.

(b) (6) • Patient : 73-year-old female with intrahepatic cholangiocarcinoma and liver, lung, lymph nodes, portacaval, and peritoneal metastases, and medical history of age- related macular degeneration for 10 years who developed a Grade 3 retinal detachment of her right eye on Day 290. Pemigatinib was interrupted. On Day 294, corrective surgery was performed; additional treatment included prednisolone and tobramycin/dexamethasone. On Day 295, slit lamp and fundus exams were normal. On Day 314, pemigatinib was restarted. On Day 316, slit lamp and fundus exams were normal, and retina was attached. The investigator did not consider the ocular toxicity related to pemigatinib and gave no other possible etiology for the ocular toxicity. Additional details are provided in Table 25.

(b) (6) • Patient : 59-year-old female with intrahepatic cholangiocarcinoma and metastases to liver, lung and lymph nodes, ongoing medical history of age-related macular degeneration and concurrent unrelated events of pulmonary embolism who was diagnosed with central retinal artery occlusion with emboli of the right eye on Day 168 on a routine eye examination. Visual acuities were 20/20 (left eye) and not assessed (right eye; baseline visual acuities were 20/20 bilaterally). Of note, this subject had ongoing thrombotic events, including an incidental pulmonary embolism with onset on Day 41 and a deep vein thrombosis with onset on Day 85 in addition to low PTT at baseline and Day 43. Treatment of the central retinal artery occlusion was initiated with 0.5% Glaucon Ophthalmic (timolol maleate) and pemigatinib was held. On Day 174 the event had not resolved, and on Day 182, an examination showed macular ischemia of the right eye, with visual acuities of 20/30 (left eye) and not assessed (right eye) and a

123 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

CT revealed a new lesion in the spleen, resulting in permanent discontinuation of pemigatinib. A brain CT revealed evidence of cerebral infarction in the left parietal lobe, left external capsule and left corona radiata. On Day 212, vision had not recovered. The investigator considered these events to be unrelated to pemigatinib and related to emboli that were likely to be caused by the patient’s underlying cancer. The patient died on Day 266 due to PD.

(b) (6) • Patient : required 2 dose reductions of pemigatinib due to 2 separate events of Grade 2 detachment of retinal pigment epithelium. The first event occurred approximately 3 months after the patient’s first dose of pemigatinib and resolved approximately two weeks later following pemigatinib dose interruption. The patient resumed pemigatinib at a reduced dose of 9 mg. A second event of RPED occurred approximately 1.5 months later and resolved after interruption of pemigatinib for approximately 1-week. The patient resumed pemigatinib at a reduced dose of 6 mg but was discontinued due to PD 20 days later. An eye examination at the end-of-treatment visit showed worsening cataracts as compared with baseline; however, the patient’s fundus examination was otherwise normal.

(b) (6) • Patient : 64-year-old male with bladder cancer and nephrostomy tubes, metastases to the bone, lymph nodes, lung, and liver and prior systemic therapy with carboplatin and gemcitabine and multiple comorbidities. Visual acuities were 6/6 in meters (left eye) and 6/15 in meters (right eye) at screening. On Day 54, the patient presented with bilateral trichiasis. On Day 112, the patient experienced bilateral optic neuropathy and vision loss. On Day 113 his symptoms were worse and an ophthalmologic exam revealed bilateral papilledema and signs of bilateral optic neuropathy but no evidence of retinal detachment. The examination also showed dilated pupils with minimal response, indicative of a positive relative afferent pupillary defect (RAPD) on the left. The visit summary described bilateral neuropathy with reduced visual field, bilateral disk edema with positive RAPD on the left side, revealed undocumented new edema with significant damage of visual field more on the left side. A CT scan of his head was normal. On Day 114, a brain MRI did not show any signs of leptomeningeal involvement or parenchymal CNS involvement and was normal except for showing evidence of papilledema. On Day 117, pemigatinib was interrupted but the patient did not receive other medical treatment. On Day 122, an eye examination revealed bilateral papilledema and signs of bilateral optic neuropathy. As of Day, 124, there was no improvement and on Day 142 an eye examination revealed visual acuities of 6/6 (left eye) and 6/20 (right eye). Pemigatinib was discontinued on Day 152 due to PD. On Day 197 improvement in his vision was improved. The patient died on Day 211.

(b) (6) • Patient : diagnosed with Grade 4 retinal detachment of the right eye on Day 78 that was categorized as nonserious. No action was taken with the study drug due to this event, which was considered related to pemigatinib by the investigator. Visual

124 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

acuities were 20/20 (left eye) and 20/30 (right eye) on Day 43 and 20/20 (left eye) and 20/400 (right eye) at the end-of-treatment (EOT) eye examination on Day 87.

(b) (6) • Patient : experienced Grade 2 events of eye swelling and blurred vision and discontinued study drug due PD on the same day. The patient died 4 days later.

(b) (6) • Patient : 76-year-old male patient with medical history significant for bilateral childhood glaucoma. Baseline eye exam was normal. The first abnormality of asymptomatic grade 2 pseudovitelliform macular dystrophy was reported on Day 50, for which dose was interrupted for 2 weeks and restarted at a reduced dose of 9mg once it improved to grade 1. The maculopathy remained stable after dose reduction, until it was no longer reported in eye exams (reported as “same as baseline”) on Day 353). The event was ongoing at the time of the data cutoff date.

Incyte conducted an assessment of the reversibility of ocular toxicity due to pemigatinib submitted as a response to an IR sent 18 Dec 2019. The majority of patients with ocular toxicity due to pemigatinib was reversible. RPED resolved or improved to Grade 1 levels in 88% of patients who required dosage modification (Table 29).

Table 29: Incyte Review of Reversibility of Ocular Toxicity for Pemigatinib

Source: Table submitted in response to FDA IR dated 18 Dec 2019 CCA=cholangiocarcinoma population All cancer: irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule)

Reviewer Comment: Following discussions with the Division of Anti-infective and Ophthalmology Drug Products and the OOD safety monitoring team, and for consistency with recent labelling practices in OOD, the clinical team recommended use of the composite term “Retinal Pigment Epithelial Detachment (RPED)” in pemigatinib product labeling to encompass the PTs listed above.

Evidence of RPED is typically not detected by visual acuity examination, slit-lamp examination or fundoscopy until the serous detachment involves the fovea and can be detected earlier by visual field testing or optical coherence tomography (OCT). Although OCT evaluation was conducted in 125 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

patients with signs or symptoms related to visual toxicity in the pemigatinib clinical trials, the frequency of asymptomatic serous retinal detachments/RPED is not well characterized because routine periodic OCT monitoring was not employed in the pemigatinib clinical trials. The lack of periodic monitoring potentially significantly decreased the frequency of reported RPED; for example Patient (b) (6) had RPED on exam at the end of treatment exam but no symptoms. The true incidence of RPED is likely to be higher than that reported in the clinical database for pemigatinib because routine OCT testing was not performed in asymptomatic patients in pemigatinib clinical trials. Additionally, related preferred terms for ocular adverse events were not aggregated by Incyte, and therefore the incidence of ocular toxicity initially reported by Incyte was artificially low and did not reflect the true scope of ocular toxicity in the clinical database. During labeling negotiations, Incyte agreed to use the composite term of RPED to more accurately communicate the incidence of ocular toxicity and include instructions for periodic OCT assessment in patients treated with pemigatinib. FDA also recommended that Incyte amend ongoing trials to include OCT assessment at baseline and periodically in all patients (in addition to as needed) and that this monitoring also occur in all future trials.

The role of FGFR in maintenance of ocular health, non-clinical data, and review of cases of ocular toxicity in this NDA (with the exception of cases of age-related toxicity such as macular degeneration and occasional cases that were confounded by comorbid or predisposing conditions) support a causal role for pemigatinib in ocular toxicity. Most ocular toxicity was reversible with discontinuation of pemigatinib. Therefore, the risk of ocular toxicity is described in the Warnings and Precautions section of product labeling for pemigatinib and dosage modification instructions for ocular toxicity are included in the Dosage and Administration section.

Acute Kidney Injury Acute kidney injury (AKI) was documented as an AE in 37 patients (8%) in the all cancer population and 11 patients (8%) enrolled on Study INCB 54828-202; AKI was severe (Grade ≥ 3) in 2 patients (1.4%) in Study-202. The all cancer incidence of AKI as an SAE was 3%, and in Study INCB 54828-202 was 2.1%. In Study INCB 54828-202, only 9 patients of the 60 patients with increased creatinine (15%) had AKI; thus, in 85% of patients, AKI did not appear to be associated with increased creatinine.

In Study INCB54828-202, 2 (1.4%) patients discontinued pemigatinib due to AKI (refer to the Laboratory Findings section of this review for a discussion of increased creatinine and Table 26). There were no deaths attributable to AKI.

Some patients with SAEs of AKI had predisposing factors such as urothelial cancer (Table 23 and Table 25). Cases of AKI for which a causal role for pemigatinib cannot be ruled out are summarized below:

(b) (6) • Patient : 60-yr-old female with extrahepatic cholangiocarcinoma with liver, lymph node and perihepatic metastases, and renal impairment at baseline with

126 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

creatinine 0.9mg/dL. On Day 8 the patient had a creatinine level of 3.3 mg/dL and hyperkalemia and was hospitalized. Pemigatinib was withdrawn and the patient’s serum creatinine returned to baseline after being off pemigatinib for 7 days. The patient was not rechallenged with pemigatinib and died on Day 46 due to PD. The investigator did not attribute the AKI to pemigatinib and considered the AKI to be due to PD.

(b) (6) • Patient : 55-yr-old female with intrahepatic cholangiocarcinoma and metastases to bone, liver, lung, lymph nodes, ascites, pleural effusion, and adrenal glands. Prior therapy was with cisplatin, fluorouracil, gemcitabine, and oxaliplatin. The patient had recurrent malignant ascites at baseline. On Day 3 she complained of burning and difficulty with urination and had urine > 100K CFU Escherichia coli. and a blood culture positive for micrococcus. On Day 34 the patient had a creatinine of 1.95mg/dL and was treated with Bactrim for presumed UTI. On Day 35, pemigatinib was held. On Day 39 creatinine was 2.28 mg/dL patient was started on ciprofloxacin due to urine culture result indicating organism was Bactrim-resistant. Day 42 creatinine was 2.42 mg/dL, and pemigatinib was withheld. Day 46 creatinine was 3.47 mg/dL and renal ultrasound revealed right renal cyst and ascites. On Day 47, paracentesis was performed and 5 liters of fluid were removed. On Day 54, the study was discontinued due to AKI. On Day 56 the patient died due to PD. The investigator assessed that there was a reasonable possibility that study procedure/conduct contributed to the event.

Reviewer Comment: This reviewer generally agrees with the investigator that it is unlikely that pemigatinib caused a major role in the development of AKI in patient (b) (6) , but given the positive dechallenge and lack of rechallenge information, a causal role for pemigatinib cannot be ruled out. Regarding patient (b) (6) , it is possible that the UTI caused AKI, but due to the single arm nature of the trial, it is difficult to exclude a causal role for pemigatinib. Acute kidney injury occurred most often in participants with urothelial carcinoma and was clearly unrelated to pemigatinib with the exception of the 2 SAEs in Study INCB 54828-202 described above. There were no deaths due to AKI and the incidence of discontinuation due to AKI was relatively low: 2 patients (1.4% [see Dropouts and/or Discontinuations Due to Adverse Effects]). Although the risk of AKI with pemigatinib appears low, it is described as an adverse reaction in Section 6 of product labeling due to its clinical significance and the temporal relationship between onset of AKI and pemigatinib in some cases for which a causal role for pemigatinib could not be excluded in INCB 54828-202.

Nail toxicity To evaluate manifestations of FGFR inhibition related to nail toxicity, a customized aggregate of PTs was developed including the following PTs: fungal paronychia, nail bed bleeding, nail bed tenderness, nail discoloration, nail discomfort, nail disorder, nail dystrophy, nail hypertrophy, nail infection, nail ridging, onychalgia, onychoclasis, onycholysis, onychomadesis, onychomycosis, and paronychia. Nail toxicity events, most frequently reported as events of onycholysis, nail discoloration, and onychomadesis, were common in Study INCB 54828-202

127 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(per-patient incidence of 41%), the all cholangiocarcinoma population (45%), and the all cancer population (33%). AEs relating to nail toxicity are summarized in Table 30. Reviewer Note: product labeling uses Incyte’s composite term for nail toxicity (including the PTs of nail toxicity, nail disorder, nail discoloration, nail dystrophy, nail hypertrophy, nail ridging, nail infection, onychalgia, onychoclasis, onycholysis, onychomadesis, onychomycosis, and paronychia) and therefore the incidence of nail toxicity listed in Section 6 of product labeling is 43%.

Nail toxicity was generally Grade 1 or 2 in severity (Table 30), and no nail toxicity event was serious or led to pemigatinib discontinuation (Table 24,Table 26). Nail toxicity events (nail disorder, nail dystrophy, onychalgia, onycholysis, onychomadesis, and paronychia) leading to pemigatinib dose interruption occurred in 3.4%, 4.3%, and 4.5% of participants in Study INCB 54828-202, the Cholangiocarcinoma Population, and the All Cancer Population, respectively, and events leading to dose reduction occurred in 2.7%, 3.1%, and 2.6%, respectively.

Table 30: Nail Toxicity Adverse Events (occurring at any rate)

Safety Population All All cancer# INCB54828-202 cholangiocarcinoma$ Grouped Term N = 466 N = 146 N = 161 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 n (%) n (%) n (%) n (%) n (%) n (%) Nail Toxicity (composite term) 155 (33) 11 (2.4) 71 (44) 4 (2.5) 60 (41) 3 (2.1) Onycholysis 45 4 18 1 13 0 Nail discoloration 42 1 18 1 14 1 Onychomadesis 32 1 16 0 14 0 Paronychia 30 4 14 2 10 1 Nail dystrophy 23 0 11 0 11 0 Nail disorder 16 3 5 1 5 1 Onychoclasis 12 1 10 1 9 1 Onychalgia 9 3 1 0 1 0 Nail ridging 7 0 3 0 3 0 Onychomycosis 7 0 5 0 4 0 Nail infection 4 0 2 0 1 0 Fungal paronychia 1 0 0 0 0 0 Nail bed bleeding 1 0 0 0 0 0 Nail bed tenderness 1 0 0 0 0 0 Nail discomfort 1 0 0 0 0 0

128 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Safety Population All All cancer# INCB54828-202 cholangiocarcinoma$ Grouped Term N = 466 N = 146 N = 161 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 Grade 1-5 Grade 3-5 n (%) n (%) n (%) n (%) n (%) n (%) Nail hypertrophy 1 0 1 0 1 0 Source: safety team; adae.xpt. Variables used: USUBJID, TRT01A, AEDECOD, AEBODSYS, AETOXGR. #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib

Reviewer Comment: Nail toxicities were common in patients treated with pemigatinib and are most likely related to its mechanism of action. Nail toxicity was mild and is considered acceptable in the context of an oncology population. Nail toxicity is described in Section 6 and in the “Patient Information” section of product labeling for pemigatinib.

Fractures There is a mechanistic plausibility that pemigatinib may predispose patients to fractures and teeth abnormalities due to effects of FGFR inhibition on calcium/phosphate metabolism and bone formation, remodeling, and mineralization. Therefore, FDA conducted an analysis to assess the risk of bone and teeth toxicity with pemigatinib.

Nonclinical data: Physeal and cartilage dysplasia have previously been reported following administration of FGFR tyrosine kinase inhibitors (Brown et al 2005). According to Incyte, cartilage (monkey-only) and physeal dysplasia observed in rat and monkey safety studies (28 days and 3-months) were attributed to the intended pharmacology of pemigatinib (FGFR inhibition at subtherapeutic doses). Rats used in toxicology studies were still rapidly growing at the time of initiation of these studies and cynomolgus monkeys in toxicology studies were sexually immature; therefore, physeal findings observed in both rats and non- human primates are considered unlikely to occur in adult patients, who have closed growth plates. Similarly, the tooth findings in rats are considered not relevant to adult patients, as incisors grow continuously in rats (Lin et al 2009).

A single monkey (No. 2003) dosed at 0.1 mg/kg/day in the 28-day study had a chronic fracture of the growth plate that was spanned by a band of fibrous tissue and well-formed bone spicules. The growth plate fracture was considered unlikely to be related to pemigatinib because the bone formation within the gap was completely remodeled, a change that would have likely required more than the length of the dosing period. No other animals in this dose group had alterations of the growth plate noted histologically. Additionally, no other animals in the 28-day study and no animals in the 3-month study (0.1, 0.33, or 1 mg/kg) had bone fractures.

Clinical data: Based upon review of a response from Incyte to an IR sent December 6, 2019 and

129 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

additional discussion with Incyte, FDA identified treatment-emergent fractures in 3 patients (2.1%) in Study INCB 54828-202, 6 patients (3.7%) in the all cholangiocarcinoma population, (b) (6) and 11 patients (2.4%) in the all cancer population (with inclusion of patient with a SAE of femur fracture in addition to Incyte’s submission). Most of the patients who experienced fractures had proceeding comorbidities such as osteoporosis, osteopenia, degenerative arthritis, bony metastases and pathological fractures. Patients with cholangiocarcinoma and AEs of related to bone toxicity are described in Table 31.

130 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 31: Bone-related Adverse Events in Patients with Cholangiocarcinoma

Patient/ Significant Comments and Lab Normal Dose Adverse Study Gender/ Grade Medical Baseline Labs Labs at AE Limits Event Age History

(b) (6) 101 13.5mg CD Spinal 2 Osteopenia, vit Vit D: 37 ng/mL Vit D: 24 ng/mL Vit D: 30­ Female compress D deficiency; Ph: 4.6 mg/dL Ph: 2.4mg/dL 100 ng/mL Ph: 2.5-4.5 mg/dL 67 years ion bone mets Ca: 9.5 mg/dL Ca: 8.8 mg/dL Ca: 8.4-10.2 mg/dL fracture (b) (6) 13.5mg ID Tooth 1 Other AE of Vit D: ND Vit D: 37ng/mL Vit D: 30-100 ng/mL Female fracture teeth grinding Phos: 3.4 mg/dL Ph: 1.9 mg/dL Phos: 2.3-4.5 mg/dL 52 years (hairline at same time Calcium: 9.1 Ca: 8.3 mg/dL Ca: 8.7-10.2 mg/dL (b) (6) mg/dL 202 13.5mg ID Displaced i 3 Osteopenia; Vit D: 17 ng/mL Vit D: 19ng/mL Vit D: 30-80ng/mL Female fracture of osteoporosis Phos: 4.9 mg/dL Ph: 2.3 mg/dL Phos: 2.2-4.5 mg/dL Calcium: 8.4­ 10.5 mg/dL 63 years big toe Calcium: 9.3 Ca: 9.6 mg/dL (left foot) mg/dL (b) (6) 13.5mg ID Osteoporo 2 Hypocalcemia Vit D: 25 ng/mL Vit D: ND Vit D: ≥ 20 ng/mL Phos: 2.5­ Female tic Phos: 4.6 mg/dL Ph: 7.4 mg/dL 4.5 mg/dL 78 years* compressi Calcium: Ca: 9.7 mg/dL Ca: 8.4-10.6 mg/dL on 9.5 mg/dL Labs are 2 weeks after event fracture (b) (6) 13.5mg ID Compression 3 Lymphoma; Vit D: 16 ng/mL Vit D: 46 ng/mL Vit D: 30-100 Female Tooth fracture 2 liver neoplasm; Phos: 3.4 mg/dL Ph: 2.4 mg/dL ng/mL 47 years bariatric surgery Calcium: Ca: 9.1 mg/dL Ph: 2.3- 4.5 mg/dL 8.5 mg/dL Ca: 8.5-10.5 mg/dL Both events occurred at the same time

131 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Patient/ Significant Comments and Lab Normal Dose Adverse Study Gender/ Grade Medical Baseline Labs Labs at AE Limits Event Age History

(b) (6) 13.5mg Upper limb 2 Atrial fib, Vit D: Vit D: Fracture is ID fracture hypertension 26 nmol/L 46 nmol/L secondary to fall Female (shoulder) Phos: Phos: Vit D: ≥ 50 nmol/L 58 1.14 mmol/L 0.88 mmol/L Ph: 0.8-1.5 nmol/L Years Calcium: Calcium: Ca: 2.2-2.7 nmol/L 2.33 mmol/L 2.38 mmol/L Source: Copy of Incyte’s Table 6 submitted in response to IR dated 6 Dec 2019; ID=intermittent dose, CD=continuous dose, fib=fibrillation * Per Incyte’s label dated 23 March 2020, Patient (b) (6) experienced a fracture prior to dosing with pemigatinib

132 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

FDA reviewed narratives for all patients with fractures. Patients with pathological fractures, across the all cancer safety population are described in the narratives in Table 32.

Table 32: Narrative Summary of Pathologic Fractures

Study Number Patient Preferred Term Narrative INCB 54828­ (b) (6) Femur fracture 69-year-old male with testicular cancer who received 101 pemigatinib in combination with pembrolizumab. Past medical history included hypertension, hypercalcemia, cancer pain, hyperphosphatemia, and metastatic neoplasm of the left tibia, knee and femur with left above knee amputation. On Day 10 he lost his balance and twisted his left lower extremity prosthesis, at which time he heard a "snap" that was followed by pain and an inability to bear weight. At the ER he had unspecified imaging which showed a left subtrochanteric pathologic femur fracture. He subsequently underwent a left hip disarticulation.

(b) (6) Spinal compression 70-year-old female with esophageal cancer and a history of fracture back pain. On an unspecified date, the patient presented to the ER after a trip and fall out of a chair landing on her buttock. On Day 48, an X-ray of the lumbar and thoracic spine disclosed an acute biconcave compression fracture at T12 and the patient was admitted to the hospital. MRI of the lumbar spine confirmed the compression fracture of T12 with 50% loss of height and an old compression fracture of L1. The patient underwent percutaneous open balloon tamponade followed by internal stabilization with methyl methacrylate cement at the T12, L1 and L3 levels. Her symptoms improved and X-rays showed no fracture. Pathology results of bone sample indicated sclerotic trabecular bone. (b) (6) Spinal fracture 54-year-old female with breast cancer and history of cauda equina syndrome, hypercalcemia, and hyperphosphatemia requiring phosphate binders. On Day 29 the patient experienced bilateral leg pain and back pain. MRI thoracic and lumbar spine, with/without contrast-extensive osseous metastases throughout the spine has slightly progressed; new compression fracture of the superior endplate of T4 vertebral body; old compression fracture of T7 vertebral body, no compression of the thoracic spinal cord; postoperative changes of T11-L3 fusion; persistent soft tissue tumor at the right aspect of the pathologically compressed L1 vertebral body measuring approximate 2.5 cm. The same day, the participant experienced pathological fracture of L1 and T4 vertebrae.

133 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

INCB 54828­ (b) (6) Fractured sacrum 71-year-old female with metastatic urothelial cancer and 201 history of constipation, pain (cancer), osteopenia, lumbar degenerative joint disease, bone metastasis, and impaired functional mobility, balance, gait and endurance. During Cycle 4 (estimate Day 112), the patient complained of left hip and spine pain. On Day 158 experienced left parafalcine subdural hematoma after she hit her head from a fall in the shower. On Day 161 the patient was dragging her foot and CT of the chest, abdomen, pelvis was completed and per the radiologist, the patient might have a sacral fracture. (b) (6) Spinal fracture 71-year-old male with metastatic urothelial cancer with a past medical history of back pain, T8 vertebral body/pathological fracture, L4 compression fracture and painful L4 vertebral body lesion. kyphoplasty for a T8 vertebral body/pathological fracture. On (b) (6) the participant received palliative radiation of 8 Gy in single fraction to T7-T9. On (b) (6) , the participant underwent kyphoplasty for a L4 compression fracture. On (b) (6) , the patient received palliative radiation due to a painful L4 vertebral body lesion. On (b) (6) (b) (6) , the participant presented to a local emergency department (ED) due to the worsening back pain and radiculopathy of T8. On the same day MRI of the thoracic spine was completed at the outside hospital (results were not provided), and oral dexamethasone 40 mg daily was initiated for pain management. On Day 48 the patient experienced back pain from PD in T8 and admitted, and a CT of the chest and a CT of the abdomen/pelvis demonstrated PD. (b) (6) Thoracic vertebral 64-year-old male with metastatic urothelial cancer who had fracture spinal radiation 6 months into treatment with pemigatinib. After that, he felt weakness in legs and difficulty in walking. CT revealed a pathological fracture in T4. A laminectomy, excision of extradural tumor, complete posterior fusion were Source: Reviewer generated table from IR response 6 Dec 2019 (received 11 Dec 2019) and ADSL dataset. PD=progressive disease, MRI=magnetic resonance imaging, CT=computed topography.

Reviewer Comment: FDA’s assessment of causality was challenging because the cases of fracture described in the table above are confounded by multiple factors such as pre-existing osteopenia and minor trauma, and because of the single arm design of the trials included in the safety database and the small sample size of the safety database itself. Background fracture incidence rate in adults with solid tumors is estimated to be as high as 18% according to data analyzed from 2013-17 using (b) (6) Data (FDA clinical review for NDA 212725); compared against this background rate of fractures, the 2.1% rate of bone AE in Study INCB 54828-202 does not appear to pose a clinically significant risk. However, mechanistically there is a biologic plausibility for a causal relationship between pemigatinib and fractures characterized by the clinical effects of pemigatinib on serum phosphate levels observed in clinical trials of pemigatinib .Therefore, fractures are included as an adverse reaction to pemigatinib in Section 6 of product labeling.

134 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Treatment Emergent Adverse Events and Adverse Reactions

The tables in this section summarize the incidence of treatment-emergent adverse events (TEAEs), defined as AEs that occurred from the time of the first dose until 30 days after the last dose of pemigatinib. FDA reviewers analyzed common TEAEs in the safety analysis dataset submitted in the original NDA based upon system organ class (SOC), high-level term (HLT), high- level group term (HLGT), and referred term (PT) levels of the MedDRA hierarchy.

An overview of the safety profile in patients treated with pemigatinib in the prespecified safety populations (all cancer n=466; all cholangiocarcinoma n=161; cholangiocarcinoma on Study INCB 54828-202 n=146) is provided below in Table 33.

Table 33: Overview of Safety Profile in Integrated Safety Populations

All cancer# All Study INCB N= 466 cholangiocarcinoma$ 54828-202 n (%) N= 161 n = 146 n (%) n (%)

Any AE 465 (99.8) 161 (100) 146 (100)

SAE 194 (42) 67 (42) 65 (46) > Grade 3 AE 284 (61) 100 (62) 93 (64)

Grades 3-4 AE 284 (61) 100 (62) 93 (64)

AE leading to discontinuation 45 (10) 13 (8) 13 (9) AE leading to dose reduction 70 (15) 23 (14) 20 (14)

AE leading to drug interruption 202 (43) 68 (42) 62 (43)

Outcome of AE is fatal 36 (8) 7 (4.3) 6 (4.1) Source: ADAE dataset; Variables used: USUBJID, SAFFL, TRTEMFL, TRT01A, AEDECOD, AETOXGR, AESER, AEACN, AEOUT=fatal, AESDTH, DTHFL #all patients with cancer, irrespective of tumor FGF/FGFR status, who received a minimum of one cycle (21 days) of pemigatinib as a single agent (either on an intermittent or continuous daily dosing schedule) $subset of the modified safety population and includes all patients with cholangiocarcinoma who received at least 1 dose of pemigatinib

In the all cancer safety population, almost all patients (99.8%) experienced at least one TEAE, and Grade 3-4 events were experienced in 61% of patients. In Study INCB 54828-202, the most frequently reported AEs (incidence ≥15%) were hyperphosphatemia, alopecia, diarrhea, nail toxicity fatigue, nail toxicity, dysgeusia, nausea, constipation, stomatitis, dry eye, dry mouth, decreased appetite, vomiting, dry eye, arthralgia, abdominal pain, hypophosphatemia, back pain, and dry skin. Table 34 provides a summary of the per-patient incidence (PPI) of AEs by SOC for AE with a PPI of ≥ 20% (see also Section 6 of product labeling for pemigatinib).

135 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 34: Adverse Reactions (≥ 15% Incidence) in Patients Receiving Pemigatinib in Study-202

Pemigatinib N=146 All Gradesa Grades ≥ 3* Adverse Reaction (%) (%) Metabolism and nutrition disorders Hyperphosphatemiab 60 0 Decreased appetite 33 1.4 Hypophosphatemiac 23 12 Dehydration 15 3.4 Skin and subcutaneous tissue disorders Alopecia 49 0 Nail toxicityd 43 2.1 Dry skin 20 0.7 Palmar-plantar erythrodysesthesia syndrome 15 4.1 Gastrointestinal disorders Diarrhea 47 2.7 Nausea 40 2.1 Constipation 35 0.7 Stomatitis 35 5 Dry mouth 34 0 Vomiting 27 1.4 Abdominal pain 23 4.8 General disorders Fatigue 42 4.8 Edema peripheral 18 0.7 Nervous system disorders Dysgeusia 40 0 Headache 16 0 Eye disorders

e Dry eye 35 0.7

136 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Pemigatinib N=146 All Gradesa Grades ≥ 3* Adverse Reaction (%) (%) Musculoskeletal and connective tissue disorders Arthralgia 25 6 Back pain 20 2.7 Pain in extremity 19 2.1 Infections and infestations Urinary tract infection 16 2.7 Investigations Weight loss 16 2.1 Source: US package insert *Only Grades 3 – 4 were identified. a Graded per NCI CTCAE 4.03 b Includes hyperphosphatemia and blood phosphorous increased; graded based on clinical severity and medical interventions taken according to the "investigations-other, specify" category in NCI CTCAE v4.03. c Includes hypophosphatemia and blood phosphorous decreased dIncludes nail toxicity, nail disorder, nail discoloration, nail dystrophy, nail hypertrophy, nail ridging, nail infection, onychalgia, onychoclasis, onycholysis, onychomadesis, onychomycosis, and paronychia. e Includes dry eye, keratitis, lacrimation increased, pinguecula, and punctate keratitis.

Reviewer Comment: composite terms that were negotiated with Incyte are defined in the footnotes to the above table.

Laboratory Findings

Hyperphosphatemia and Hypophosphatemia Hyperphosphatemia is an on-target effect of FGFR inhibition and was expected with pemigatinib administration. The study protocols included recommendations for management of hyperphosphatemia based on serum phosphate levels (> 5.5 and ≤ 7 mg/dL, > 7 and ≤ 10 mg/dL, or > 10 mg/dL). Dietary phosphate restriction, administration of phosphate-binding therapy, and increased phosphate monitoring were recommended initially. Addition of a phosphaturic agent was also permitted. If these interventions were insufficient to manage serum phosphate levels while taking pemigatinib, then dose modification (interruption and/or dose reduction) and finally permanent discontinuation of pemigatinib were recommended.

Hyperphosphatemia generally does not cause symptoms unless there is precipitation of calcium-phosphate crystals leading to hypocalcemia; soft tissue mineralization, tetany, seizure activity, QT interval prolongation, and arrhythmias may result from effects on calcium, extra skeletal deposition of calcium phosphate crystals, and electrical hyperexcitability (Peppers et al 1991). Potential clinical sequelae of hyperphosphatemia were identified in 2 of the 5 patients who had serious and/or ≥ Grade 3 hyperphosphatemia across the all cancer population. Muscle

137 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

spasms (Grade 1) and corneal opacity (Grade 1), both considered related to pemigatinib, occurred in a patient with ongoing punctate keratitis, vitreous detachment, and intraocular lens implant that were present at baseline (6 days before the first dose of pemigatinib). No other potential sequelae were identified among patients with serious and/or ≥ Grade 3 hyperphosphatemia, and review of all cases of seizure and cases of ECG QT prolonged in patients in the all cancer safety population did not suggest a relationship to pemigatinib.

TEAEs of calciphylaxis (with concurrent peripheral venous disease) and calcinosis cutis occurred (b) (6) (b) (6) in a single patient each (Patient and Patient ; cases further described below). These TEAE were mild in severity, non-serious, and occurred in the context of disease that could predispose to these events. (b) (6) • Patient was enrolled in Study -101 for the treatment of Wilms tumor and had a medical history of radical nephrectomy and inferior vena cava thrombectomy. This patient developed a confirmed case of calcinosis cutis reported as rash on Day 23. Phosphate levels ranged from 5.7-7.9 mg/dL (ULN 4.5mg/dL) and the investigator considered the AE not related to PD.

(b) (6) • Patient was enrolled in Study -201 for the treatment of urothelial carcinoma and had a medical history of left bundle branch block, myocardial infarction, gout, increased blood creatinine, chronic kidney disease, and radical cystoprostatectomy. Phosphate levels ranged from 4.6-7.9 mg/dL on treatment. Pemigatinib was held on Day 316 and ultimately permanently discontinued due to bilateral lower extremity pain and swelling consistent with grade 1 calciphylaxis (which led to discontinuation of pemigatinib). The patient was still alive at the time of data cutoff.

Blood samples to measure clinical chemistry parameters, including phosphate, calcium, vitamin D (1,25-dihydroxyvitamin D and 25-hydroxyvitamin D), and parathyroid hormone, were collected at the following protocol-defined timepoints in Study INCB 54828-202: baseline, Days 8 and 15 of Cycle 1, Day 1 of every subsequent cycle, at the EOT visit, and at the safety follow- up visit. Concurrent changes in mean calcium levels were not evident in patients who developed phosphate laboratory abnormalities.

Grading of hyperphosphatemia was based on CTCAE 5.0 because CTCAE 4.03 does not provide grading for hyerphosphatemia . Among 466 patients who received pemigatinib in the all cancer population, hyperphosphatemia as a laboratory abnormality was reported in 94% of patients. The median time to onset of hyperphosphatemia was 8 days (range 1-169). Phosphate lowering therapy was required in 29% of patients receiving pemigatinib.

In Study INCB 54828-202, the incidence of hyperphosphatemia as a laboratory abnormality was 94%, which is comparable to the incidence of hyperphosphatemia in the all cholangiocarcinoma population (94%) and in the all cancer safety population (92%). In Study -202, none of the hyperphosphatemia laboratory abnormalities or TEAEs were severe. Hyperphosphatemia was managed with diet, phosphate-lowering therapy, and pemigatinib dose modification, as

138 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

needed. In the all-cancer population, 8 (1.7%) patients required interruption of pemigatinib due to phosphate levels of > 7mg/dL for greater than 14 days; these patients were enrolled on INCB (b) (6) 54828-101, and included 2 patients with cholangiocarcinoma ( ).

In Study INCB 54828-202, hypophosphatemia was reported as a laboratory abnormality in 68% of patients (Grade ≥ 3 in 38%) and as a TEAE in 23% of patients (12% ≥ Grade 3); however, no TEAEs of hypophosphatemia were serious or led to dose modification or discontinuation of pemigatinib. Severe (Grades ≥ 3) hypophosphatemia was mainly a laboratory finding and was generally not associated with clinically significant signs or symptoms.

Reviewer Comment: Hyperphosphatemia occurred in almost all patients (94%) treated with pemigatinib in Study INCB 54828-202 and in the all cancer population, and is an expected adverse reaction to pemigatinib based on its mechanism of action. No patients were discontinued from pemigatinib for hyperphosphatemia, and hyperphosphatemia was successfully managed with dietary changes and institution of phosphate binders, with drug interruption and dosage reduction when necessary. There were no patients on Study -202 with serum phosphate >7mg/dL for >10 days and no patients on any study had serum phosphate > 10 mg/dL for >7 days. Of the 2 patients with cholangiocarcinoma with elevated serum phosphate of >7mg/dL for > 14 days, one had a prior history of hyperphosphatemia (Patient (b) (6) ) due to prior treatment with TAS-120, an investigational FGFR inhibitor; the other patient (Patient (b) (6) ) was on pemigatinib continuous dosing, which may have contributed to the development of hyperphosphatemia.

Incyte was also asked to submit autopsy reports, if applicable, to assess for ectopic mineralization due to imbalances in calcium/phosphate metabolism. Only 3 autopsy reports were available, and the review of these autopsy reports did not yield any instances of ectopic mineralization. Although two cases potentially indicative of ectopic mineralization (calcinosis cutis in Patient (b) (6) and calciphylaxis in Patient (b) (6) described above ) were identified in the clinical database, these cases were mild, were confounded my other factors, and occurred outside of Study 202; these events are therefore not described in product labeling. Continued pharmacovigilance through evaluation of postmarketing safety reports and safety data from ongoing and future clinical trials of pemigatinib is warranted and will be helpful to further characterize this potential safety signal.

Hyperphosphatemia did not appear to lead to clinical sequelae such as seizures or prolonged QT prolongation and was manageable with supportive care and dosage modification as needed. . Due to the prevalence of hyperphosphatemia with pemigatinib and the need for early identification and intervention to avoid potential clinical sequelae, hyperphosphatemia is described in the Warnings and Precautions section of product labeling for pemigatinib and management guidelines are included in the dosage modification table in Section 2. Hypophosphatemia generally occurred following institution of phosphate binder treatment (i.e., was due to unintended overtreatment of hyperphosphatemia).

139 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Increased creatinine Increases in creatinine were observed in patients receiving pemigatinib. In Study INCB54828­ 202, an initial increase from baseline in mean creatinine levels of approximately 16 μmol/L occurred on Cycle 1 Day 8 and 15. Although mean creatinine levels generally decreased on Cycle 2 Day 1, ,mean creatinine levels remained elevated above baseline throughout the study. Elevated creatinine levels (above the ULN) were present in 13% of patients at baseline and were reported in 34% of patients on Cycle 1 Day 15 and in approximately 20% to 30% of patients on Day 1 of subsequent cycles. All shifts in creatinine over baseline levels were to CTC Grade 1 or 2 levels, with the exception of 2 patients who had a shift to Grade 3 creatinine levels (see Figure 7). The observed pattern of creatinine elevation is consistent with in vitro studies demonstrating that pemigatinib is an inhibitor of the renal transporters OCT2 and MATE1. As described in Section 12.3 (Pharmacokinetics) of pemigatinib product labeling, pemigatinib may increase serum creatinine by decreasing renal tubular secretion of creatinine; this may occur due to inhibition of OCT2 and MATE1 and may not affect glomerular function. Figure 7: Mean (± SE) Change from Baseline in Serum Creatinine Levels Over Time in Study INCB54828-202

Source: Copied from Study 202 interim CSR submitted in Module 5.3.5.3 Serious ≥ Grade 3 AEs of blood creatinine increased and/or acute kidney injury occurred in 4 patients (see Table 25), including the 2 patients with CTC Grade 3 creatinine elevation described above. These patients all had renal impairment at baseline. Notable patients from Study INCB54828-202 with increased creatinine that have already been discussed in Dropouts and/or Discontinuations Due to Adverse Effects are described again below.

(b) (6) • Patient is a 60-year-old female with metastatic extrahepatic cholangiocarcinoma involving the liver, peri-hepatic area, and lymph nodes and renal impairment, with a baseline creatinine of 0.9 mg/dL. On Day 8 the patient had a

140 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

creatinine level of 3.3 mg/dL (normal range: 0.6-1.1 mg/dL), elevated BUN (76 mg/dL),and hyperkalemia, and was admitted to the hospital with AKI. Pemigatinib was withdrawn on Day 9. The SAEs of AKI and elevated creatinine resolved on Day 13, and on Day 15 her serum creatinine level was 0.9 mg/dL, and BUN was 14 mg/dL. The patient discontinued pemigatinib on Day 22 and died on Day 46 due to PD. There was no rechallenge. The investigator reported that there is not a reasonable possibility that pemigatinib caused the elevated creatinine and AKI, and attributed both to PD.

(b) (6) • Patient is a 54-yr male with intrahepatic cholangiocarcinoma and liver, lung, and lymph nodes metastases. Prior systemic therapy was with fluorouracil, folinic acid, irinotecan, and paclitaxel albumin. He also had mild renal impairment at baseline and had a creatinine level of 4.4 mg/dL (normal range: 0.7-1.3 mg/dL) on Day 358, which was reported as a SAE of blood creatinine increased that was unrelated to pemigatinib. Blood urea nitrogen on that day was 33 mg/dL (normal range: 8-20 mg/dL). The patient had discontinued pemigatinib on Day 343 due to PD. On Day 362, the creatinine level was 2.7 mg/dL and BUN was 25 mg/dL, and the SAE of blood creatinine increased was considered resolved.

Reviewer Comment: In most patients, elevations in creatinine values appeared to be isolated laboratory findings that were not accompanied by clinical symptoms (e.g., concurrent kidney injury), a conclusion that is further supported by the absence of discernable trends in BUN and the underlying inhibition of the OCT2 and MATE1 renal transporters by pemigatinib. Increases in creatinine are described in Section 6 of the label. Per the FDA Clinical Pharmacology review and as described in product labeling for pemigatinib, within the first 21-day cycle of pemigatinib dosing, serum creatinine increased (mean increase of 0.2 mg/dL) and reached steady state by Day 8, and then decreased during the 7 days off therapy. Alternative markers of renal function should be considered if persistent elevations in serum creatinine are observed (see Section 6 [Adverse Reactions] and Section 12.3 [Clinical Pharmacology] of pemigatinib product labeling). Although the cases of AKI described above are confounded, given the temporal relationship (b) (6) between initiation of pemigatinib in these cases and the onset of AKI in Patient and the limitations associated with assessing causality of AE in single arm trials, AKI will be included as a rare adverse reaction to pemigatinib in Section 6 of product labeling.

Hepatotoxicity Study INCB 54828-202 Nineteen percent of the study population had a medical history of a hepatobiliary disorder. Approximately 5% of the population had a prior history of increased ALT, and 8% had a prior history of increased AST. Approximately 49% of patients in Study INCB 54828-202 had hepatic impairment at baseline.

Increased AST and ALT of any grade occurred as a TEAE in 43% of patients each. Grade 3–4 increased AST or ALT occurred in 6% and 4.1% of patients, respectively. The median time to onset of increased AST was 3.1 weeks (range: 1-80). The median time to onset of increased ALT was 1.29 weeks (range: 1-52). Increased AST or ALT leading to dose interruption occurred in

141 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

2.1% of patients, each, and events leading to dose reductions occurred in 0.7% of patients each. No patients discontinued pemigatinib due to increased AST or ALT.

Increased bilirubin of any grade occurred in 26% of patients. Grade 3– 4 increased bilirubin occurred in 6% of patients. The median time to onset of increased bilirubin was 10 weeks (range: 1.1-70). Increased bilirubin leading to dose interruptions occurred in 1.4% of patients, and no patients had a dose reduction due to increased bilirubin. Pemigatinib was discontinued due to increased bilirubin in 0.7% of patients.

Notable patient narratives for patients enrolled on Study -202 with elevations in ALT, AST or bilirubin to greater than 3 times the upper limit of normal are summarized below. None of liver test abnormalities met Hy’s law criteria for drug-induced liver injury.

(b) (6) • Patient is a 67-year-old male with stage 2 cholangiocarcinoma (no FGF/FGFR alteration) who underwent 3 prior lines of treatment before enrolling into Cohort C and received 13.5 mg pemigatinib administered as intermittent dosing (2 weeks on followed by 1 week off). Baseline AST was 21 IU/L, ALT was 20 IU/L and total bilirubin was 0.5 mg/dL. On Cycle 1 Day 8, AST was 548 IU/L, ALT was 606 IU/L and total bilirubin was 1.8 mg/dL. The adverse events of increased AST and ALT were reported as grade 3 adverse events, and resolved approximately one week following study drug interruption.

(b) (6) • Patient is a 69-year-old male, diagnosed with stage 4 cholangiocarcinoma (FGFR2 TTC28 fusion). The patient underwent 2 prior lines of treatment before enrolling into Cohort A to receive 13.5 mg pemigatinib. At baseline, AST was 75 U/L (high), ALT was 32 U/L and total bilirubin was 1.7 mg/dL (high). Total bilirubin levels oscillated around baseline values (with ranges between 1.0mg/dL and 1.8 mg/dL) up until Cycle 9 Day 1 when total bilirubin levels increased to 2.4 mg/dL. However, it returned to baseline (Cycle 10 Day 1 at 1.4 mg/dL) after a planned dose interruption and a dose interruption due to grade 2 hand-foot syndrome. Between Cycle 11 Day 1 and Cycle 23 Day 1, total bilirubin ranged from 2.0 mg/dL to 3.1 mg/dL; the patient had an interruption for a week and then pemigatinib was restarted at a reduced dose of 9 mg daily. On Cycle 24 Day 1, pemigatinib was again interrupted due to Grade 3 increased bilirubin at 7.3 mg/dL “secondary to another medical condition”. The patient tolerated subsequent drug rechallenge, with bilirubin ranging from 1.1-1.8 mg/dL until Cycle 34, when the patient had PD and discontinued pemigatinib.

(b) (6) • Patient is a 59-year-old female diagnosed with stage 4 cholangiocarcinoma (FGFR2 BICC1). The patient underwent 3 prior lines of treatment before enrolling into Cohort A and received the 13.5 mg intermittent dosing regimen. Baseline AST was 156 U/L, ALT was 50 U/L and bilirubin was 1.3 mg/dL. At Cycle 16 Day 1, the patient’s AST was Grade 2 at 190 U/L, ALT was Grade 1 at 153 U/L, and bilirubin was 0.6 mg/dL. Study drug was interrupted for approximately 3 weeks, and both AST/ALT increases were considered related to pemigatinib. Pemigatinib was restarted during Cycle 17 at a

142 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

reduced dose of 9 mg. Two weeks later, AST was 158 U/L (Grade 1) and the ALT was 76 U/L (Grade 1). The patient remained on treatment until the end of Cycle 19.

(b) (6) • Patient is a 45-year-old female diagnosed with stage 4 cholangiocarcinoma (FGFR2 BICC1). The patient underwent surgical resection and received 3 prior therapies before enrolling into Cohort A to receive the 13.5 mg intermittent dosing regimen. At baseline, AST was Grade 1 at 101 U/L, ALT was 109 U/L (high) and bilirubin was 0.6 mg/dL. On Day 7 the AST was Grade 3 at 371 U/L, ALT was Grade 3 at 471 U/L and bilirubin was 0.7 mg/dL, and pemigatinib was interrupted for a week; the elevations were considered related to pemigatinib. Pemigatinib was restarted at a reduced dose of 9 mg (intermittent dosing) but was interrupted again one week later on Day 22 due to Grade 3 AEs of worsening ALT at 289 U/L and AST at 318 U/L (considered related to study drug). Pemigatinib was restarted a week later at the 9 mg dose (intermittent dosing) and the patient continued to have intermittent Grade 2-3 elevation in ALT and AST requiring dose interruption, followed by recommencement of pemigatinib at the 9 mg dose level. These elevations were accompanied by concomitant increases in bilirubin to Grade 2 (range 1.1-2.0 mg/dL), which was 2xULN. On Day 244, bilirubin was Grade 3 at 3.9 mg/dL (normal range: 0.0-1.0 mg/dL), the patient was hospitalized for biliary obstruction, and pemigatinib was withdrawn. This patient’s hyperbilirubinemia was attributed to biliary obstruction; two days prior to this admission, the patient had undergone an ERCP (Endoscopic Retrograde Cholangiopancreatography) which showed disease progression involving the liver. Therefore Hy’s law was not met.

(b) (6) • Patient is a 42-year-old Caucasian female diagnosed with stage 2 cholangiocarcinoma (FGFR2-BICC1 fusion) in the liver and lymph nodes. The patient received radiation and 3 lines of chemotherapy before enrolling into Cohort A to receive the 13.5 mg daily intermittent dosing schedule of pemigatinib. Baseline AST was 62 U/L (high), ALT was 34 U/L (high) and bilirubin was 2.0 mg/dL (34 μmol/L; high). On Cycle 7 Day 14, the patient had a nonserious AE of Grade 3 hyperbilirubinemia: bilirubin 63 μmol/L (normal range: 1.7-20.5 μmol/L), direct bilirubin 23.9 μmol/L (normal range: 0­ 6.8 μmol/L), AST 46 U/L (normal range: 0-32 U/L), ALT 41 U/L (normal range: 0-31 U/L), and alkaline phosphatase 463 U/L (normal range: 30-120 U/L). The patient was rechallenged at a dose of 13.5 mg and continued on therapy until Cycle 16 Day 4. Hyperbilirubinemia was ongoing throughout treatment (end of follow up visit bilirubin was 15 mg/dL or 256.5 μmol/L). The investigator considered the AE to be related to study drug while Incyte assigned causality as “not suspected”, indicating that the case is highly confounded by the patient’s prior medical history of hyperbilirubinemia, the underlying severe intrahepatic cholangiocarcinoma, and the concurrent increase of alkaline phosphatase to almost four times the upper limit of normal which supports an obstructive etiology. The patient discontinued the study on Cycle 17 due to death from PD (also see Dropouts and/or Discontinuations Due to Adverse Effects and Table 26).

143 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Cholangiocarcinoma Population Increased AST and ALT of any grade occurred in 44% and 43% of patients, respectively. Grade 3 – 4 increased AST or ALT occurred in 6% and 3.7% of patients, respectively. The median time to onset of increased AST was 4 weeks (range: 1-80). The median time to onset of increased ALT was 1.3 weeks (range: 0.3-61). Increased AST or ALT leading to dose interruptions occurred in 1.9% and 1.9% each, and events leading to dose reductions occurred in 0.6% of patients each. No patients discontinued pemigatinib due to increased AST or ALT. Increased bilirubin of any grade occurred in 26% of patients. Grade 3 – 4 increased bilirubin occurred in 4.3% of patients. The median time to onset of increased bilirubin was 12.4 weeks (range: 1.1-103). Increased bilirubin leading to dose interruptions occurred in 1.2% of patients, and no patients had a dose reduction due to increased bilirubin. Pemigatinib was discontinued due to increased bilirubin in 0.6% of patients.

All Cancer Population Increased AST and ALT of any grade occurred in 35% and 33% of patients, respectively. Grade 3 – 4 increased AST or ALT occurred in 3.9% and 2.6% of patients, respectively. The median time to onset of increased AST was 3.3 weeks (range: 0.1-80). The median time to onset of increased ALT was 2.1 weeks (range: 0.1-61). Increased AST or ALT leading to dose interruptions occurred in 2.1% and 1.3% of patients, respectively, and increased AST or ALT led to dose reduction in 0.2% of patients each. No patients discontinued pemigatinib due to increased AST or ALT. Increased bilirubin of any grade occurred in 15% of patients. Grade 3 – 4 increased bilirubin occurred in 2.2% of patients. The median time to onset of increased bilirubin was 7.2 (range: 0.3 to 103 weeks). Increased bilirubin leading to dose interruptions occurred in 0.6% of patients, and no patient had a dose reduction due to increased bilirubin. Pemigatinib was discontinued due to increased bilirubin in 0.2% of patients.

A careful review of the safety data included in this application did not reveal any cases of Hy’s law. As reported in the 4-month safety update submitted to the application, one patient in (b) (6) Study INCB 54828-202 (Patient , described below) met the laboratory criteria for a potential Hy’s law case at a single time point (during an early termination visit). FDA clinical review excluded drug-induced liver injury in this patient because the elevated AST, ALT, and bilirubin were likely to be due to malignant obstruction of the biliary tract.

(b) (6) Patient had Stage 4 intrahepatic cholangiocarcinoma 1.5 years prior to study entry. At study entry, the patient had disease in liver and lungs and had Grade 1 elevations in ALT, AST, and bilirubin, which continued throughout study participation. The patient developed cholangitis during Cycle 3 which resolved prior to starting Cycle 4 and continued on study (b) (6) through Cycle 12 until PD (new lesion in liver- Segment III). The investigator assessed the elevations in liver function tests as not related to pemigatinib and related to PD.

(b) (6) Reviewer Comment: For Patient with increases in liver enzymes, and Grade 2 hyperbilirubinemia (but did not meet Hy’s criteria on the basis of laboratory abnormalities alone as bilirubin was 2xULN): the patient had evidence of biliary obstruction, so there was an

144 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

alternate cause of the underlying hyperbilirubinemia. Given the single arm nature of the clinical trials, underlying cancer typically involving the liver and biliary tract, and baseline elevation of AST, ALT, and/or bilirubin in many patients, assessment of causality was sometimes difficult. However, overall, increases in AST, ALT, and bilirubin observed in clinical trials of pemigatinib were not accompanied by clinical sequalae and were manageable with dosage modification. No cases of drug-induced liver injury were evident in the safety database. Nevertheless, Section 6 of pemigatinib product labeling informs healthcare providers of the risk of elevations of AST, ALT, and bilirubinemia and Section 2 provides dosage modification instructions based upon severity and duration of these elevations.

Vital Signs

The vital signs dataset included data on weight, body mass index, body surface area, height, heart rate, and blood pressure. Most patients in Study INCB 54828-202 had normal vital signs at baseline and at all timepoints assessed. Changes in these parameters were generally small, and no clinically meaningful trends were observed. Measured values that were outside the normal range with a > 25% change from baseline were considered to be “alert vital signs”. Vital sign values meeting alert criteria occurred infrequently among patients in Study INCB 54828-202, with increases in pulse being most common. These increases were generally transient and not clinically meaningful.

Although > 25% changes from baseline occurred somewhat frequently for all parameters, the only parameter for which > 20% of participants had an abnormal value in either population was pulse > 100 bpm (34% and 33% of participants in the Cholangiocarcinoma Population and the All Cancer Population, respectively).

Electrocardiograms (ECGs)

In Study -202, 12-lead ECGs were performed at baseline, Cycle 1 Day 1 and Day 15, and then on Day 1 of each subsequent cycle. There was a single patient with prolonged QT who had a remote history of hyperphosphatemia (see the discussion of hyperphosphatemia above). The events of ECG QT prolongation had onset 27 and 53 days after the Grade 3 AE of hyperphosphatemia resolved and were not considered related to pemigatinib.

Per Incyte, the ECG results available as of the data cutoff dates suggest no meaningful effect of pemigatinib on ECG parameters. An analysis of AEs within the Torsade de pointes/QT prolongation SMQ and any events of seizure identified in the clinical database for the all cancer safety population supported this conclusion.

Clinical reviewer Comment: there does not appear to be a correlation between ECG abnormalities and pemigatinib.

QT

145 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

A cardiac safety analysis was performed using 12-lead ECG data from 116 patients treated with pemigatinib monotherapy at doses of 1 to 20 mg daily in Study INCB 54828-101. In study INCB 54828-101, a 12-lead ECG was performed on Day 1 and Day 14 of Cycle 1 prior to dosing, then 1, 2, and 4 hours post-dose. Timed 12-lead ECGs were performed before but within 15 minutes of the PK blood draw at the corresponding timepoint. All 12-lead ECGs were recorded by centrally provided equipment and analyzed at the central ECG laboratory using a semi­ automated technique (i.e., over-readings of ECG intervals). Results of central tendency, outlier, and concentration-QT analyses are detailed in the cardiac safety report submitted by Incyte and reviewed by FDA. Observed changes in ECG parameters in patients treated with pemigatinib monotherapy in Study INCB 54828-101 were generally small, and the categorical analysis did not reveal dose dependence in the incidences of QTcF outliers. The categorical analysis did not reveal dose dependence in the incidences of QTcF outliers, and no patients had a QTcF > 500 ms or ΔQTcF > 60 ms. For the 3 highest doses studied (9 mg daily [N = 21], 13.5 mg daily [N = 67], and 20 mg daily [N = 18]), the largest least squares means of ΔQTcF on Cycle 1 Day 1 were all at 1-hour post dose: 8.9 ms (90% CI: 4.69, 13.06), 7.8 ms (90% CI: 5.42, 10.15), and 0 ms (90% CI: -4.49, 4.58), respectively. At steady state, the largest least squares means of ΔQTcF were 11.8 ms (90% CI: 7.29, 16.40) at 1-hour post dose, 4.6 ms (90% CI: 2.02, 7.12) at 1-hour post dose, and 3.8 ms (90% CI: -1.20, 8.80) at 2 hours post dose, respectively. In the concentration-QTcF analysis, a prespecified linear mixed-effects model was determined to be an appropriate fit to the data and was used to establish the relationship between plasma pemigatinib concentration and ΔQTcF. The estimated slope of the concentration-QTcF relationship was shallow: 0.00391 ms per nM (95% CI: -0.01244, 0.02026), which was not statistically significantly different from 0. Using this concentration-QTcF model, the QT effect (ΔQTcF) can be predicted to be 4.18 ms (90% CI: 2.13, 6.24) at the therapeutic dose level (13.5 mg QD, observed Cmax,ss = 236 nM [56.4% CV]), and 4.91 ms (90% CI: 0.60, 9.22) at the highest dose level studied (20 mg QD, observed Cmax,ss = 421 nM [38.7% CV]). A large QT/QTc effect exceeding the threshold of concern (20 ms) was excluded within the observed range of pemigatinib plasma concentrations. The categorical analysis of PR/ΔPR and QRS/ΔQRS did not suggest dose-dependence in the outlier incidences. The least squares mean of ΔPR and ΔQRS were small, suggesting that pemigatinib at the studied doses did not have a relevant effect on cardiac conduction (i.e., the PR and QRS intervals).

Clinical reviewer Comment: No large effects on prolongation of the QT interval were identified during review of the data summarized above. Doses of pemigatinib up to 20 mg daily did not have a large effect on the QT/QTc interval, defined as exceeding the threshold of concern (i.e., >20 ms). As such, the FDA QT-IRT Review team concluded that the data from INCB 54828- 101 suggest a lack of large mean effect on the QTc interval and determined that these data were adequate to characterize the effect of pemigatinib 13.5 mg daily on the QTc interval.

146 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Immunogenicity

No safety issues related to immunogenicity were identified for pemigatinib.

8.2.5 Analysis of Submission-Specific Safety Issues

Not applicable

8.2.6 Clinical Outcome Assessment (COA) Analyses Informing Safety/Tolerability

Not applicable. This application did not include COA analyses informing the safety or tolerability of pemigatinib.

8.2.7 Safety Analyses by Demographic Subgroups

Geriatric Patients TEAE were assessed based on age (< 65 years old and ≥ 75 years old). In Study-202, 32% of patients were 65 years and older, and 8% of patients were 75 years and older. No overall differences in safety were observed between older and younger patients (Table 35).

Table 35: Overview of Safety Profile by Age (Geriatrics)

All cancer All cholangiocarcinoma Study INCB54828-202 N Age (%) < 65 >= 65 < 75 >= 75 < 65 >= 65 < 75 >= 75 < 65 >= 65 < 75 >= 75 N = 262 N = 204 N = 405 N = 61 N = 110 N = 51 N = 150 N = 11 N = 100 N = 46 N = 135 N = 11 261 204 404 61 110 51 150 11 100 46 135 11 All-Grade AEs (99.6) (100) (99.8) (100) (100) (100) (100) (100) (100) (100) (100) (100) 129 246 38 37 35 Grade 3-5 AEs 155 (59) 63 (57) 94 (63) 6 (55) 58 (58) 86 (64) 7 (64) (63) (61) (62) (73) (76) 109 218 30 34 33 Grade 3-4 AEs 139 (53) 59 (54) 87 (58) 6 (55) 54 (54) 80 (59) 7 (64) (53) (54) (49) (67) (72) 2 Grade 5 16 (6) 20 (10) 28 (7) 8 (13) 4 (3.6) 3 (6) 7 (4.7) 0 4 (4) 6 (4.4) 0 (4.3) 165 29 27 26 SAEs 105 (40) 89 (44) 40 (36) 62 (41) 5 (46) 39 (39) 59 (44) 6 (55) (41) (48) (53) (57) Treatment 2 discontinuation 27 (10) 18 (9) 39 (10) 6 (10) 11 (10) 12 (8) 1 (9) 10 (10) 3 (7) 11 (8) 2 (18) (3.9) due to AEs

147 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

All cancer All cholangiocarcinoma Study INCB54828-202 N Age (%) < 65 >= 65 < 75 >= 75 < 65 >= 65 < 75 >= 75 < 65 >= 65 < 75 >= 75 N = 262 N = 204 N = 405 N = 61 N = 110 N = 51 N = 150 N = 11 N = 100 N = 46 N = 135 N = 11 Dose modifications 114 193 31 30 28 (interruption 110 (42) 43 (39) 69 (46) 4 (36) 39 (39) 62 (46) 5 (46) (56) (48) (51) (59) (61) and reduction) due to AEs Dose 102 175 27 29 27 interruption 100 (38) 39 (36) 64 (43) 4 (36) 35 (35) 57 (4) 5 (46) (50) (43) (44) (57) (59) due to AEs Dose reduction 16 28 (11) 42 (21) 54 (13) 14 (13) 9 (18) 23 (15) 0 12 (12) 8 (17) 20 (15) 0 due to AEs (26) Source: adae.xpt, adsl.xpt. Variables used: USUBJID, SAFFL, TRTEMFL, TRT01A, AEDECOD, AETOXGR, AESER, AEACN, AESDTH, DTHFL, AGE.

Reviewer Comment: A review of specific AEs by PT occurring in ≥ 10% of the safety population by age (in patients aged 65+ and 75+ versus younger patients) did not reveal any clear differences in the type or incidence of TEAEs in older patients compared to younger patients.

8.2.8 Specific Safety Studies/Clinical Trials

There were no additional studies performed to evaluate any specific safety concerns.

8.2.9 Additional Safety Explorations

Human Carcinogenicity or Tumor Development

Carcinogenicity studies were not conducted or required to support the use of pemigatinib in the proposed indication.

Human Reproduction and Pregnancy

No pregnancies occurred during the development program and there was no known drug exposure of breastfed infants of lactating women. In the protocols, females of reproductive potential were advised to use effective contraception during treatment and for 1 week after the final dose of pemigatinib. Based on its mechanism of action and findings in animals, pemigatinib can cause fetal harm when administered to pregnant women and is in patient labelling (Section 8.3). There are no data on the presence of pemigatinib or its metabolites in human milk, the effects of pemigatinib on the breastfed child, or on milk production.

Pediatrics and Assessment of Effects on Growth

There were no pediatric patients enrolled on trials with pemigatinib or submitted to the NDA.

Overdose, Drug Abuse Potential, Withdrawal, and Rebound

148 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The highest total daily dose of pemigatinib tested in clinical studies was 20 mg daily in Study INCB 54828-101. No DLTs occurred at this dose level. Overdose was not defined for each of the clinical studies in this application. However, 1 patient had a Grade 1 AE of accidental (study drug) overdose on Days 16 to 20. The patient in Study INCB 54828-201 had a concurrent Grade 1 AE of blood phosphorus increased, which resolved, and no other AEs at the time of the event. Treatment of suspected overdose with pemigatinib should consist of general supportive measures. There is no known potential for abuse of pemigatinib. No withdrawal and rebound effects of pemigatinib were identified.

8.2.10 Safety in the Postmarket Setting

Safety Concerns Identified Through Postmarket Experience

Pemigatinib has not been approved in any country at the time of this review. Incyte has applied for European marketing authorization for pemigatinib.

Expectations on Safety in the Postmarket Setting

The review teams determined that a REMS is not required to ensure safe and effective use of pemigatinib. Pemigatinib will be prescribed by oncologists who are trained how to monitor, diagnose, and manage serious adverse reactions caused by anti-neoplastic drugs in accordance with FDA-approved labeling. Additionally, standard practice in oncology dictates informed consent prior to prescribing or administering anti-neoplastic drugs.

8.2.11 Integrated Assessment of Safety

The above safety assessment incorporates data from multiple trials and is therefore integrated, in the “all cancer” and “all cholangiocarcinoma” patient populations.

8.3 Statistical Issues

There were no major statistical issues in the review of this NDA. However, the single-arm design of the main trial supporting efficacy limits the interpretation of time-to-event endpoints, such as PFS and OS, given the absence of an active control or comparison to the natural history of disease.

8.4 Conclusions and Recommendations

The primary efficacy results of Study INCB 54828-202, based upon data from 107 patients with cholangiocarcinoma harboring a FGFR2 gene fusion or other rearrangement (Cohort A), demonstrate that treatment with pemigatinib results in clinically meaningful responses that are durable in patients with previously treated unresectable locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or other rearrangement, a serious and life threatening disease. The response rate is sufficiently large in magnitude (36% with a lower

149 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

bound of the 95% confidence interval of 27%) and durability (DoR ≥ 6 months in 63% of responders and ≥ 12 months in 18% of responders) to be clinically meaningful to patients with cholangiocarcinoma who have received prior treatment, given that the life expectancy in this setting is 5-6 months. There are no available FDA-approved drugs for the second-line treatment of patients with unresectable or metastatic cholangiocarcinoma, and no approved treatments specifically for the subset of patients with cholangiocarcinoma with an FGFR2 gene fusion or other rearrangement.

The safety review consisted of data from 466 adult patients from 5 single-arm trials: Study INCB 54828-202, Study INCB 54828-101, Study INCB 54828-102, Study INCB 54828-201, and Study INCB 54828-203. In general, the adverse reactions observed with pemigatinib were consistent with the mechanism of action and toxicity observed in preclinical studies of pemigatinib. The primary risks related to pemigatinib are hyperphosphatemia and ocular toxicities including RPED. These serious risks are adequately addressed in the Warnings and Precautions and Dosage Modifications sections of pemigatinib product labeling. The rate of permanent discontinuation of pemigatinib due to TEAEs was 9% in Study INCB 54828-202 and most discontinuations were attributed to disease progression. The major safety risks of pemigatinib are toxicities that oncologists frequently manage and the toxicity profile of pemigatinib is considered acceptable when considering the anti-tumor effects observed (e.g., durable responses) in a patient population with a serious and life-threatening condition.

Taken together with the safety results of pemigatinib, the results demonstrated in Study INCB 54828-202 are reasonably likely to predict an effect on irreversible morbidity or mortality or other clinical benefit, and represent a meaningful advantage over existing treatments for the proposed indication, meeting the requirements for accelerated approval.

150 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

X Somak Chatterjee, PhD X Pallavi Mishra-Kalyani, PhD Somak Chatterjee Pallavi Mishra-Kalyani Primary Statistical Reviewer Statistical Team Leader

X Naomi Horiba, MD, MPH X Martha Donoghue, MD Naomi Horiba Martha Donoghue Primary Clinical Reviewer Clinical Team Leader

151 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

9 Advisory Committee Meeting and Other External Consultations

The Division did not obtain the advice of the Oncologic Drug Advisory Committee (ODAC) for this NDA because no review issues were identified that raised significant public health questions regarding the risk:benefit assessment of pemigatinib for the proposed indication.

Two special government employees (SGEs), an adult oncologist with expertise in cholangiocarcinoma and a patient representative with a prior history of cholangiocarcinoma who has expertise in issues relating to the development of treatments for cancer, were cleared by the Advisory Committee Oversight and Management Staff (ACOMS) to participate in a divisional assignment to consult on this NDA. A separate teleconference was held with each SGE to discuss the application. During these teleconferences, the SGEs indicated that they considered the risk:benefit assessment favorable for use of pemigatinib in patients with cholangiocarcinoma harboring a FGFR2 gene fusion or other rearrangement who had received prior treatment. The SGEs also stated that they did not have concerns regarding approval of pemigatinib for treatment of this patient population.

152 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

10 Pediatrics

Pemigatinib was granted Orphan Designation on March 12, 2018 for the treatment of cholangiocarcinoma and in accordance with 21 CFR 314.55(5)(d), orphan drugs are exempt from the pediatric study requirements under 21 CFR 314.55.

153 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

11 Labeling Recommendations

11.1 Prescription Drug Labeling

The table below (Table 21 ) summarizes changes to the proposed prescribing information (PI) made by FDA. See the final approved prescribing information for PEMAZYRE (pemigatinib tablets) accompanying the approval letter for more information.

Table 21. Summary of Significant Labeling Changes to PEMAZYRE Prescribing Information Section/Subsection Proposed Labeling Approved Labeling INDICATIONS AND USAGE Text describing accelerated Accelerated approval text was approval stated that the revised to reflect the indication is approved terminology used in CLINICAL under accelerated approval STUDIES to describe the based on “ (b) (4) primary endpoint, “overall response rate”. response rate”. The indication statement was modified to add the word “unresectable” before locally advanced cholangiocarcinoma and to clarify that FGFR2 gene fusions are a subset of FGFR2 rearrangements through addition of the word “other” before “rearrangement” in the indication. DOSAGE AND ADMINISTRATION PEMAZYRE dosage PEMAZYRE dose reductions modifications for the and dosage modifications for adverse reactions adverse reactions were revised hyperphosphatemia and into a tabular presentation; serous retinal detachment dosage modifications for were described. “other adverse reactions” were added to the table. In addition, the adverse reaction, (b) (4) (b) (4) was revised to “retinal pigment epithelial detachment (RPED)” and PEMAZYRE dosage modification recommendations for this adverse reaction were revised. The dosage modification The recommendation for recommendations for PEMAZYRE dosage strong CYP3A inhibitors and modifications for concomitant inducers subsection use with moderate CYP3A 154 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

provided dosage inhibitors was added to the adjustments for recommendations concerning concomitant use of concomitant use with strong PEMAZYRE with strong and CYP3A inhibitors; (b) (4) moderate CYP3A inhibitors (b) (4) and a recommendation to avoid use of PEMAZYRE with CYP3A inducers

WARNINGS AND PRECAUTIONS 5.1 (b) (4) Revised to: “5.1 Ocular (b) (4) Toxicity” because the body of the warning and precaution was revised to discuss not only retinal pigment epithelial detachment or RPED (revised (b) (4) (b) (4) to more adequately describe the adverse reaction) but also dry eye as well.

Although the incidence of RPED observed across clinical trials was noted, a statement was also including stating that clinical trials of PEMAZYRE did not conduct routine monitoring including optical coherence tomography (OCT) to detect asymptomatic RPED; therefore, the incidence of asymptomatic RPED with PEMAZYRE is unknown. In addition, the risk mitigation strategy for this adverse reaction was revised to specifically recommend OCT examinations for monitoring.

5.3 Embryo-Fetal Toxicity 5.3 Embryo-Fetal Toxicity Included a Revised the contraception recommendation for duration recommendation to 1 females of reproductive week after the final dose to potential to use effective reflect the current contraception during recommendations for treatment and for (b) (4) nongenotoxic pharmaceuticals

155 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

after the last dose. with embryo-fetal lethality [5 x T1/2 (5 x 15.4 hours) ADVERSE REACTIONS/Clinical Trials Added additional demographic Experience information about the safety population (age, sex, and race); added incidence/listing of serious and fatal adverse reactions; and added adverse reactions leading to dosage interruptions or dose reductions. The common adverse reactions table was revised to reflect the incidence only for each adverse reaction (b) (4) (b) (4) . In addition, the table was revised to include adverse reactions grades ≥ 3 as opposed to just grades 3 and 4.

A commentary regarding grading for the laboratory abnormality of creatinine using ULN and hyperphosphatemia to CTCAE 5.0 was added. DRUG INTERACTIONS Drug interactions were reordered based on clinical importance.

Provided dose adjustment The recommendation for recommendations PEMAZYRE dosage concerning concomitant modifications for concomitant use of PEMAZYRE with use of PEMAZYRE with CYP3A strong CYP3A4 inhibitors inhibitors was revised to recommend a PEMAZYRE dose modification for both strong or moderate CYP3A inhibitors, if concomitant use could not be avoided.

(b) (4)

156 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (4)

USE IN SPECIFIC Advised women not to Revised the recommendations POPULATIONS/Lactation breastfeed during to not breastfeed during treatment with PEMAZYRE treatment with PEMAZYRE and and for one month after for one week after the final the final dose. dose ( (b) (4) (b) (4)

USE IN SPECIFIC Contraception POPULATIONS/Females and Males of Revised the recommendation Reproductive Potential concerning the duration of use of effective contraception for females (see Warnings and Precautions (5.3) above).

(b) (4)

CLINICAL Cardiac Electrophysiology PHARMACOLOGY/Pharmacodynamics Revised to include dose coverage of 1.5-fold and indicate that a large effect was excluded rather than a clinically meaningful effect. Serum Phosphate Provided additional exposure/response information concerning increased serum phosphate levels CLINICAL Revised to include additional PHARMACOLOGY/Pharmacokinetics steady-state, absorption, and distribution information Drug Interaction Studies Drug Interaction Studies Clinical Studies and Model- Clinical Studies and Model- Based Approaches Based Approaches

157 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Described the effect of Revised the drug interaction CYP3A inhibitors, CYP3A data to reflect the percent inducers, and acid-lowering change in Cmax and AUC. agents on the Cmax and AUC of pemigatinib in terms of geometric mean ratios. Add a statement that no clinically significant differences in metformin glucose levels were observed with co- administration of pemigatinib. Drug Interaction Studies Drug Interaction Studies In Vitro Studies In Vitro Studies (b) (4)

CARCINOGENESIS, MUTATION, Statement added that oral IMPAIRMENT OF FERTILITY administration of pemigatinib did not result in any dose- related findings likely to result in impaired fertility in male and female reproductive organs.

158 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

CLINICAL STUDIES Revised the clinical study design description to include the name of the trial (Fight­ 202) and NCT number. Added text to provide information regarding FGFR2 “in-frame fusions and other rearrangements” that qualified patients for inclusion into Cohort A (those predicted to have a breakpoint within intron17/exon 18 of the FGFR2 gene leaving the FGFR2 kinase domain intact).

The term for the major efficacy outcome was changed (b) (4) (b) (4) to “overall response rate” (consistent with other recently approved oncology labeling)

CLINICAL STUDIES Efficacy Results: (b) (4) (b) (4)

In describing the efficacy results, the efficacy parameters greater than 5% were rounded to the nearest digit and estimates of mean and 95% CI rounded to one place after the decimal. The duration of response was revised to show the 6 and 12 month observed duration of response rates (b) (4) (b) (4)

(b) (4) PATIENT COUNSELING INFORMATION 159 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (4)

In addition, the information was reordered to reflect clinical importance and a heading concerning drug interactions was added.

12 Risk Evaluation and Mitigation Strategies (REMS)

[The risks of pemigatinib are acceptable in the indicated patient population with a serious and life-threatening condition; the safe use of pemigatinib can be adequately implemented in the post-marketing setting through product labeling. No additional risk management strategies are recommended.

160 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

13 Postmarketing Requirements and Commitment

Incyte agreed to the following clinical postmarketing requirement (PMR):

“Submit the results of a randomized trial demonstrating improvement of progression-free survival or overall survival in patients with unresectable locally advanced or metastatic cholangiocarcinoma with a fibroblast growth receptor 2 (FGFR2) gene fusion or rearrangement, to confirm clinical benefit of pemigatinib.“

Schedule milestone dates are as follows: Final Protocol Submission: 11/2018 (completed) Trial Completion: 06/2026 Final Report Submission: 12/2026

161 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

14 Division Director (DHOT)

X

15 Division Director (OCP)

X

16 Division Director (OB) Comments

X

17 Division Director (Clinical) Comments

I concur with the review team’s conclusion that the primary efficacy results of Study INCB 54828-202 (Cohort A) demonstrate that treatment with pemigatinib results in durable and clinically meaningful responses in patients with previously treated unresectable locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or other rearrangement. The adverse reactions observed with pemigatinib were expected given the mechanism of action of pemigatinib, and toxicity profile observed in the preclinical evaluation of pemigatinib, and acceptable in the context of a serious and life threatening disease. Hyperphosphatemia and ocular toxicities including RPED are the primary risks of pemigatinib and these and other pemigatinib-related toxicities are manageable by oncologists following guidelines described in product labeling and with use of supportive measures.

I agree that the benefit:risk assessment of pemigatinib is favorable in the indicated population and that the results of Study INCB 54828-202 characterized by a durable and clinically meaningful ORR, are reasonably likely to predict an effect on irreversible morbidity or mortality or other clinical benefit, and represent a meaningful advantage over existing treatments for

162 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

patients with previously treated unresectable locally advanced or metastatic cholangiocarcinoma with an FGFR2 fusion or other rearrangement.

X Lola A. Fashoyin-Aje, MD, MPH

163 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

18 Office Director (or designated signatory authority) Comments

This application was reviewed by the Oncology Center of Excellence (OCE) per the OCE Intercenter Agreement. My signature below represents an approval recommendation for the clinical portion of this application under the OCE.

X

164 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

19 Appendices

19.1 References

Almquist DR, Javle M, Ciombor KK, et al. FGFR2 fusions and its effect of patient outcomes in intrahepatic cholangiocarcinoma. Annals of Oncology 2019. 30 suppl. 5: v279.

Blechacz B. Cholangiocarcinoma: Current Knowledge and New Developments. Gut Liver 2017; 11(1):13-26.

Brown AP, Courtney CL, King Lm, Groom SC, and Graziano MJ. Cartilage Dysplasia and Tissue Mineralization in the Rat Following Administration of a FGF Receptor Tyrosine Kinase Inhibitor. Toxicologic Pathology 2005; 33:449–455.

Entrectinib review: https://www.accessdata.fda.gov/drugsatfda docs/nda/2019/212725Orig1s000,%20212726Orig 1s000MultidisciplineR.pdf

Faktorovich EG, RH Steinberg, D Yasumura, et al. Photoreceptor degeneration in inherited retinal dystrophy delayed by basic fibroblast growth factor. Nature 1990; 347(6288): 83-86.

Faktorovich EG, RH Steinberg, D Yasumura, et al. Basic fibroblast growth factor and local injury protect photoreceptors from light damage in the rat. J Neurosci 1992; 12(9): 3554-3567.

Francis JH, Habib LA, Abramson DH, Yannuzzi LA, Heinemann M, Gounder MM, Grisham RN, Postow MA, Shoushtari AN, Chi P, Segal NH, Yaeger R, Ho AL, Chapman PB, Catalanotti F. Clinical and Morphologic Characteristics of MEK Inhibitors Associated Retinopathy Differences from Central Serous Chorioretinopathy. American Academy of Ophthalmology, 2017: 1788-98.

Goeppert B, Roessler S, Renner M, et al. Mismatch repair deficiency is a rare but putative therapeutically relevant finding in non-liver fluke associated cholangiocarcinoma. British Journal of Cancer 2019; 120: 109-14.

Guillonneau X, M Bryckaert, C Launay-Long, et al. Endogenous FGF1-induced activation and synthesis of extracellular signal-regulated kinase 2 reduce cell apoptosis in retinal-pigmented epithelial cells. J Biol Chem 1998; 273(35)L 22367-22373.

Kheder D and Hong D. Emerging targeted therapy for tumors with NTRK fusion proteins. Clinical Cancer Research 2019; 24(23): 5807-5814.

Krook MA, Lenyo A, Wilberding M, et al. Efficacy of FGFR inhibitors and combination therapies for acquired resistance in FGFR2-fusion cholangiocarcinoma. Molecular Cancer Therapeutics

165 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

2020; 19: 847-57.

Lamarca A, Palmer DH, Easan HS, et al. A randomised phase III, multi-centre, open-label study of active symptom control (ASC) alone or ASC with oxaliplatin / 5-FU chemotherapy (ASC+mFOLFOX) for patients (pts) with locally advanced / metastatic biliary tract cancers (ABC) previously-treated with cisplatin/gemcitabine (CisGem) chemotherapy (ABC-06). Journal of Clinical Oncology 2019; 37(15) suppl: Abst. 4003.

Li F, Peiris MN, Donoghue DJ. Functions of FGFR2 corrupted by translocations in intrahepatic cholangiocarcinoma. Cytokine Growth Factor Rev 2019; https://doi.org/10.1016/j.cytogfr.2019.12.005.

Lin Y, Cheng YSL, Qin C, D'Souza R, Wang F. FGFR2 in the dental epithelium is essential for development and maintenance of the maxillary cervical loop, a stem cell niche in mouse incisors. Dev Dyn 2009;238:324-330.

Nakamura H, Arai Y, Totoki Y, et al. Genomic spectra of biliary tract cancer. Nature Genetics 2015; 47: 1003-10.

National Comprehensive Cancer Network. Hepatobiliary Cancers (Version 1.2020). https://www.nccn.org/professionals/physician gls/pdf/hepatobiliary.pdf. Accessed April 8, 2020.

Nti AA, Serrano LW, Sandhu HS, Uyhazi KE, Edelstein ID, Zhou EJ, Bowman S, Song, D, Gangadhar TC, Schuchter LM, Mitnick S, Huang A, Nichols CW, Amaravadi RK, Kim BJ, Aleman TS. Frequent Subclinical Macular Changes in Combined Braf/Mek Inhibition with High- Dose Hydroxychloroquine as Treatment for Advanced Metastatic Braf Mutant Melanoma. Preliminary Results From a Phase I/II Clinical Treatment Trial. Retina 2019; 39:502–513.

Peppers MP, Geheb M, Desai T. Hypophosphatemia and Hyperphosphatemia. Critical Care Clinics 1991; 7(1):201-14.

Rizvi S and Borad M. The rise of the FGFR inhibitor in advanced biliary cancer: the next cover of time magazine? J Gastroninest Oncol. 2016; 7(5): 789-96.

Rizvi S and Gores GJ. Pathogenesis, diagnosis, and management of cholangiocarcinoma. Gastroenterology 2013; 145(6):1215-29.

Robinson M. An essential role for FGF Receptor Signaling in Lens Development. Semin Cell Devl Biol 2006; 17(6): 726-740.

Rosenthal R, Malek G, Salomon N, Peill-Meininghaus M, Coeppicus L, Wohlleben H, Wimmers S, Rickman CB, Strauss O. The fibroblast growth factor receptors, FGFR-1 and FGFR-2, mediate two

166 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

independent signalling pathways in human retinal pigment epithelial cells. Biochemical and Biophysical Research Communications 2005; 337: 241-7.

Saha SK, Zhu AX, Fuchs CS, and Brooks GA. Forty-year trends in cholangiocarcinoma incidence in the US: intrahepatic disease on the rise. Oncologist 2016; 21(5):594–9.

Valle J, Wasan H, Palmer DH, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. New England Journal of Medicine 2010; 362(14): 1273-81.

Van der Noll R, S Leijen, GHG Neuteboom, et al. Effect of inhibition of the FGFR-MAPK signaling pathway on the development of ocular toxicities. Cancer Treat Rev 2013; 39(6): 664-672.

Wohrle S, O Bonny, N Beluch, et al. FGF receptors control vitamin D and phosphate homeostasis by mediated renal FGF-23 signaling and regulating FGF-23 expression in bone. J Bone Miner Res 2011; 26(10): 2486-2497.

Zhang J, D Upadhya, L Lu, et al. Fibroblast growth factor receptor 2 (FGFR2) is required for corneal epithelial cell proliferation and differentiation during embryonic development. PLoS One 2015; 10(1): e0117089

19.2 Financial Disclosure

[Insert text here]

Covered Clinical Study (Name and/or Number): Study INCB 54828-202

Was a list of clinical investigators provided: Yes No (Request list from Applicant) Total number of investigators identified: 1,012 Number of investigators who are Sponsor employees (including both full-time and part-time employees): 0

Number of investigators with disclosable financial interests/arrangements (Form FDA 3455): 2 If there are investigators with disclosable financial interests/arrangements, identify the number of investigators with interests/arrangements in each category (as defined in 21 CFR 54.2(a), (b), (c) and (f)):

167 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Compensation to the investigator for conducting the study where the value could be influenced by the outcome of the study: 0 Significant payments of other sorts: 2 Proprietary interest in the product tested held by investigator: 0 Significant equity interest held by investigator: 1 Sponsor of covered study: 0 Is an attachment provided with details Yes No (Request details from of the disclosable financial Applicant) interests/arrangements: Is a description of the steps taken to Yes No (Request information minimize potential bias provided: from Applicant) Number of investigators with certification of due diligence (Form FDA 3454, box 3) 0 Is an attachment provided with the Yes No (Request explanation reason: from Applicant)

19.3 Nonclinical Pharmacology/Toxicology

No additional information.

19.4 OCP Appendices (Technical documents supporting OCP recommendations)

19.4.1 Appendix 1. Summary of Bioanalytical Method Validation and Performance

Two bioanalytical methods were developed and validated to quantify pemigatinib plasma levels in the clinical program; one was developed by Incyte, and the second by (b) (4) .

168 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 36. Summary of Method Performance (Incyte assay).

(b) (4) (b) (4) (b) (4) (b) (4) (b) (4) (b) (4) (b) (4)

169 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

170 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

171 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

172 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

173 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

174 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

(b) (4) Table 37. Summary of Method Performance .

19.4.2 Appendix 2. Physiologically based pharmacokinetic (PBPK) analyses review

Executive Summary The objective of this review is to evaluate the adequacy of the Applicant’s physiologically based pharmacokinetic (PBPK) analyses to evaluate the drug-drug interaction (DDI) potentials.

• The effect of strong, moderate, and weak CYP3A inhibitors on pemigatinib PK • The effect of strong, moderate, and weak CYP3A inducers on pemigatinib PK • The effect of pemigatinib as inhibitor on P-gp, OCT2, or MATE1 substrates The Division of Pharmacometrics (DPM) has review the PBPK analyses report (DMB-19.25.1), response to FDA’s information request (IR) submitted on January 6, January 21, and January 31, 2020, and the modeling supporting files, and concluded the followings.

175 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

• The PBPK analysis is adequate to evaluate the effects of moderate CYP3A inhibitors and inducers on the PK of pemigatinib. The predicted geometric mean AUC of pemigatinib was increased by about 50% to 90% when pemigatinib was co-administered with a moderate CYP3A inhibitor. The predicted geometric mean AUC of pemigatinib was decreased by more than 50% when it was co-administered with efavirenz (a moderate CYP3A inducer) 600 mg QD for 2 weeks. • The PBPK analysis is inadequate to evaluate the effect of pemigatinib on the PK of a P­ gp or OCT2/MATEs substrate due to the lack of quantitative in vitro in vivo extrapolation for Ki (the inhibition constant) values.

Background Following oral administration in the fasted state, pemigatinib was absorbed quickly with median Tmax approximately 1 to 2 hours. Linear PK was observed over a dose range of 1 to 20 mg. The protein binding of pemigatinib was independent of pemigatinib concentrations with an average of 9.4% unbound fraction in human plasma and serum. At the recommended dose of 13.5 mg QD, steady state was achieved by Day 4 with an approximate accumulation ratio of 1.6 for AUC0-24. The geometric mean of steady-state half-life, Cmax and AUC0-24 were 15.4 hours, 236 nM and 2620 h*nM, respectively (Summary of Clinical Pharmacology).

In the human absorption, metabolism, and excretion (AME) study (INCB 54828-105) where single dose of two 4.5 mg tablets and one 2 mg tablet of pemigatinib followed 10 minutes later by an oral dose solution of approximately 2 mg of 14C labeled pemigatinib, about 1.4% of the administered radioactive dose was recovered as unchanged pemigatinib in feces, and about 1.0% of the dose was excreted in urine as unchanged pemigatinib. The study suggested almost complete oral absorption and low renal clearance of pemigatinib. No major metabolites were detected in plasma (Summary of Clinical Pharmacology).

Pemigatinib has low and pH-dependent solubility over the physiological pH range (1-7.4). In vitro transport study (DMB-14.61.1) indicated that pemigatinib was a substrate of both P-gp and BCRP; however, the efflux mediated by P-gp and BCRP was saturated at concentrations of 1 and 30 μM, respectively. The pemigatinib concentrations in the GI lumen were estimated to be higher than 1 μM. Dose proportionally was demonstrated over the dose range of 1 to 20 mg for steady state pemigatinib concentrations. Therefore, P-gp and BCRP were not incorporated into the pemigatinib model. In vitro study (DMB-14.73.1) have indicated that CYP3A4 is the major isozyme responsible for the metabolism of pemigatinib.

In vitro pemigatinib was not a potent inhibitor of human liver microsomal CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4-mediated testosterone hydroxylation activity, or CYP3A4-mediated midazolam hydroxylation activity, with IC50 values greater than 25 μM (DMB-14.72.1). In vitro, pemigatinib was not an inducer of CYP1A2 or CYP3A4. Therefore, these interactions were not incorporated into the PBPK model. Pemigatinib is an inhibitor of P-gp (IC50 = 4.8 μM; DMB-14.61.1), OCT2 (IC50 = 0.075 μM; DMB­ 14.62.1) and MATE1 (IC50 = 1.1 μM; DMB-18.141.1). In addition, although pemigatinib is an

176 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

inhibitor of MATE2K (IC50 = 15.3 μM; DMB-18.141.1), OAT3 (IC50 > 33 μM), and OATP1B3 (IC50 = 3 μM; DMB-14.62.1), no DDIs are expected at clinically relevant exposures.

A clinical DDI study was conducted to assess the effects of itraconazole (a strong CYP3A4 inhibitor), and rifampin (a strong CYP3A4 inducer) on pemigatinib PK (INCB 54828-104; DMB­ 18.64.1). The study showed an approximate 90% increase in pemigatinib AUC with itraconazole coadministration and an 85% decrease in AUC with rifampin coadministration.

The DDI study with gastric pH-modifying agents (INCB 54828-106; DMB-18.66.1) showed that the geometric mean Cmax and AUC of pemigatinib decreased by 35% and 8%, respectively, upon coadministration with proton pump inhibitor, esomeprazole; the geometric mean Cmax and AUC of pemigatinib decreased by 2% and increased by 3%, respectively, upon coadministration with histamine-2 antagonist, ranitidine.

Methods Software

Simcyp Version 17.1 (Certara) was used for PBPK analyses by the Applicant and reviewer. The default Sim-Healthy Volunteer (HV) population and Sim-Cancer population were used to simulate studies in healthy subjects and cancer patients, respectively.

PBPK model development

The pemigatinib substrate model consisted of an Advanced Dissolution, Absorption and Metabolism (ADAM) model for absorption, and a minimal PBPK model for distribution. Intrinsic clearance was calculated using retrograde model and consisted of CYP3A4 (~55%), additional HLM (~44%), and renal clearance (~1%) pathways. The input parameters are summarized in Table 38.

177 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 38. Summary of input parameters for pemigatinib PBPK model

Source: Table 3 of PBPK report DMB-19.25.1

Reviewer’s comments: Initially, the Applicant used ADMET predicted parameters, logP, pKa, B/P as model input while these values can be experimentally measured. In addition, it was found that the input solubility for pH 6.5 and 7.4 was 0.01 mg/mL while the measured solubility was < 0.001 mg (pH 5.3, 6.5, 7.4 aqueous buffer), 0.219 mg/mL (pH 5.0 FeSSIF), 0.008 mg/mL (pH 6.5 FaSSIF), or 0.003 mg/mL (pH 6.8 SIF). The review team sent an information request on December 11, 2019 and asked the Applicant to provide explanation for not using the experimentally measured parameters and inconsistency between the solubility input and experimental data. The Applicant submitted a written response on January 6, 2020 and the modeling supporting files on January 21, 2020. In the response, the Applicant revised model by using measured parameters (Table 39). As these parameters were used for Vss calculation, to match the observed Vss, the Kp scalar was adjusted accordingly. The updated model predicted PK profiles were almost overlap with the original model predicted PK profiles. 178 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Table 39. Updated input parameters for pemigatinib model

Source: Table 1 of the response to 11 December 2019 clinical pharmacology information request submitted on January 6, 2020

Pemigatinib PBPK model validation

Pemigatinib PBPK model predictions were compared with observed data from the PK study INCB 54828-101, and DDI study INCB 54828-104.

Sensitivity analyses

Sensitivity analyses were conducted for fmCYP3A4 (fraction of pemigatinib being metabolized by CYP3A), pemigatinib in vitro inhibition constant (Ki) towards P-gp and OCT2. The pemigatinib Ki values for P-gp and OCT2 were lowered by 10-fold based on in vitro measured Ki values.

Other perpetrator and substrate models

Simcyp (V17) default perpetrator models, itraconazole, erythromycin, diltiazem, fluconazole, fluvoxamine, rifampin, and efavirenz were used for CYP3A4-mediated DDI simulations. Dexamethasone (as a weak CYP3A4 inducer) model was developed based on literature PBPK model (Pilla Reddy et al 2018). Simcyp (V17) default digoxin model and metformin model were used to evaluate P-gp and OCT2/MATEs mediated DDIs, respectively.

Reviewer’s comments: In the December 11, 2019 information request, the reviewer requested the Applicant to submit the dexamethasone model verification files which should at least include the comparison between the predicted and observed dexamethasone PK at steady state, and the effect of dexamethasone on the PK of a sensitive CYP3A substrate. The Applicant responded that ‘Although the rebuilt dexamethasone PBPK model was able to capture observed dexamethasone PK following a single oral dose of 8 mg dexamethasone, the model was not able to reproduce DDI results between dexamethasone and casopitant’. Nevertheless, itraconazole increased pemigatinib exposure by about 90% suggesting that pemigatinib is not a sensitive substrate. Therefore, the effect of a weak CYP3A4 inducer (such as dexamethasone) on pemigatinib is expected to be small.

PBPK model application

179 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The PBPK model was used for the following purposes.

• To predict the effects of strong, moderate, and weak CYP3A inhibitors on pemigatinib PK • To predict the effects of strong, moderate, and weak CYP3A inducers on pemigatinib PK • To predict the effects of pemigatinib as an inhibitor of gut P-gp, OCT2 or MATEs substrate

Results 1. Can PBPK analyses provide a reasonable description of the PK of pemigatinib? Yes. Pemigatinib PBPK model predicted PK profiles and parameters were compared to the observed PK profiles and parameters following single dose (SD) administration of 4.5 mg or 13.5 mg pemigatinib in healthy subjects or following multiple dose (MD) administration of 6 mg, 9 mg, 13.5 mg, or 20 mg pemigatinib in cancer patients.

The Applicant updated input parameters by using measured logP, pKa, B/P, and aqueous solubility in response to our IR. The Applicant showed that the simulated PK profiles following single dose administration of 4.5 mg or 13.5 mg pemigatinib in healthy subjects (HVs) were almost superimposable. The Applicant did not repeat all the simulations conducted in the original PBPK analyses report using the updated model file. The reviewer conducted PK simulations using the updated model files and Simcyp default HV population or cancer populations. The results are summarized in Figure 8 and Table 40. Overall, the predicted PK parameters (Cmax and AUCinf for SD or AUCss,24 for MD, expressed as geometric means) are within 0.5-1.2 of the observed PK parameters (Table 40). The model tends to under-predict the observed PK especially for Cmax. The simulated PK parameters were slightly higher using HVs population compared to those simulated using cancer patient population (Table 40).

Figure 8. Simulated and observed pemigatinib PK profiles following single dose or multiple dose oral administration

180 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Green solid line: simulated mean pemigatinib PK profiles; grey solid lines: 5th and 95th percentiles of simulated pemigatinib PK profiles (N=100); dotted symbols: observed mean pemigatinib PK profiles. Sources: reviewer’s simulation using Applicant’s updated compound file with measured logP, pKa, B/P, and aqueous solubility, and submitted xml files containing observed PK data. Dosing regimen and observed data source: A. 4.5 mg SD, HVs, 54828-104-cohort2-54828­ alone.xml; B. 13.5 mg SD, HVs, 54828-104-cohort1-54828-alone.xml; C. 6 mg QD, cancer patients, 54828-101-6mg.xml; D. 9 mg QD, cancer patients, 54828-101-9mg.xml; E. 13.5 mg QD, cancer patients, 54828-101-135mg.xml; F. 20 mg QD, cancer patients, 54828-101-20mg.xml

Table 40. Summary of predicted vs. observed pemigatinib PK parameters following single dose or multiple dose oral administration

Observed Predicted Predicted/Observed Dose Study Population N Cmax AUC population N Cmax AUC Cmax AUC (mg) (nM) (nM*h) (nM) (nM*h) 4.5 SD 104 HV 18 55 672 HV 100 50 634 0.91 0.94 13.5 SD 104 HV 18 176 1960 HV 100 149 1893 0.85 0.97 13.5 SD 106 HV 17 178 1580 HV 100 149 1893 0.84 1.20 13.5 SD 106 HV 18 128 1560 HV 100 149 1893 1.17 1.21 6 QD 101 Cancer 4 79 1050 Cancer 100 68 785 0.87 0.75 9 QD 101 Cancer 18 162 1670 Cancer 100 102 1175 0.63 0.70 13.5 QD 101 Cancer 57 236 2620 Cancer 100 151 1744 0.64 0.67 20 QD 101 Cancer 13 421 4150 Cancer 100 210 2485 0.50 0.60 6 QD 101 Cancer 4 79 1050 HV 100 84 844 1.06 0.80 9 QD 101 Cancer 18 162 1670 HV 100 125 1265 0.77 0.76 13.5 QD 101 Cancer 57 236 2620 HV 100 187 1892 0.79 0.72 20 QD 101 Cancer 13 421 4150 HV 100 260 2709 0.62 0.65 PK parameters are expressed as geometric mean values. SD, single dose; MD, multiple dose; for SD, Cmax and AUCinf were calculated; for MD, steady state Cmax,ss and AUCss,24 were calculated; Sources: observed PK parameters are from Tables 4-7, and 17 in the Summary of Clinical Pharmacology; simulated PK parameters were based on reviewer’s simulation using Applicant’s updated compound file with measured logP, pKa, B/P, and aqueous solubility.

2. Can PBPK analyses predict the effects of moderate CYP3A inhibitors on the PK of pemigatinib? Yes. A fmCYP3A of 0.55 was obtained by fitting the observed itraconazole effects on pemigatinib PK. About 1% of total clearance was assigned to the renal pathway, and the rest of the clearance (about 44%) was assigned to non-specific metabolic pathway. In the response to FDA’s IR, the Applicant indicated that there is no other single pathway that is responsible for >25% of total clearance. The predicted and the observed CmaxR and AUCR were similar in the presence and absence of itraconazole (200 mg capsules QD under fed condition) (Table 41).

Table 41 Summary of predicted and observed effects of itraconazole (a strong CYP3A inhibitor) and rifampin (a strong CYP3A inducer) on the PK of pemigatinib

Observed Predicted Pred./Obs. Perpetrator CmaxR AUCtR CmaxR AUCtR CmaxR AUCtR Itraconazole 1.17 1.91 1.19 1.96 1.02 1.03 Rifampin 0.38 0.15 0.63 0.32 1.66 2.13

181 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Source: Table 2 of the response to 11 December 2019 clinical pharmacology information request submitted on January 6, 2020

Based on the estimated fmCYP3A, the Applicant conducted DDI simulations to evaluate the effects of other strong CYP3A inhibitors (clarithromycin), moderate CYP3A inhibitors (erythromycin, diltiazem, and fluconazole), and weak CYP3A inhibitor (fluvoxamine) on the PK of pemigatinib. As shown in Table 42, the AUCR (the AUC ratios of pemigatinib in the presence and absence of an inhibitor) values were similar following multiple dose vs. single dose administration of pemigatinib for the same inhibitor. CmaxR values were higher at steady state compared to the single dose administration. In general, the change in AUC is larger than the change Cmax of pemigatinib for the same inhibitor. The AUCR was about 2-fold when pemigatinib was co-administered with a strong CYP3A inhibitor (such as itraconazole and clarithromycin). The predicted geometric mean AUC of pemigatinib was increased by about 56% to 87% when pemigatinib was co-administered with a moderate CYP3A inhibitor (Table 42). A weak CYP3A inhibitor (such as fluvoxamine) does not have much impact on pemigatinib exposure.

Pemigatinib PK profiles were also simulated following 9 mg QD or 13.5 mg QD in the presence of a CYP3A inhibitor (Table 43). Simulated AUCss,24 following 9mg QD of pemigatinib in the presence of a strong CYP3A inhibitor were about 30% higher compared to the observed AUCss,24 following 13.5mg QD pemigatinib alone. Simulated AUCss,24 following 9mg QD of pemigatinib in the presence of a moderate CYP3A inhibitor were similar to the observed AUCss,24 following 13.5mg QD pemigatinib alone. However, it should be notice that the pemigatinib based model under-predicted the observed PK (Table 40). Therefore, dose adjustment should be based on the predicted AUCR where the comparison was based on simulated PK.

Table 42 and Table 43 summarized the simulated DDI results with various CYP3A inhibitors. The itraconazole DDI study was conducted in HVs. The Applicant conducted DDI simulations for itraconazole in HVs and for the inhibitors not being tested in clinical DDI studies in the cancer population. As shown in Table 40, the simulated difference using cancer population vs. HV population was about 20% and 10% in Cmax and AUC, respectively. The reviewer conducted additional DDI simulations using both the healthy volunteers and the cancer population to explore the difference in predicted CmaxR and AUCR between the HV population and the cancer population. In general, the difference in predicted CmaxR and AUCR between the HV population and the cancer population were less than 10% (data not shown).

Table 42. Summary of predicted effects of moderate CYP3A inhibitors on the PK of pemigatinib following single dose or multiple dose administration

Perpetrator and dose CmaxR AUCR Cmax,ssR AUCss,24R Itraconazole 200 mg QD 1.18 2.11 1.68 2.14 Clarithromycin 500 mg BID 1.17 2.06 1.66 2.10 Clarithromycin 250 mg BID 1.14 1.79 1.53 1.85 Erythromycin 500 mg BID 1.12 1.54 1.33 1.57

182 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Erythromycin 500 mg TID 1.15 1.75 1.43 1.76 Diltiazem 60 mg TID 1.12 1.63 1.40 1.65 Fluconazole 400 mg QD 1.15 1.83 1.52 1.87 Fluconazole 200 mg QD 1.13 1.63 1.40 1.66 Fluvoxamine 50 mg QD 1.03 1.05 1.05 1.06 Cmax,ssR and AUCss,24R are the pemigatinib Cmax and AUC24 ratios at steady state with vs. without an inhibitor; QD, once daily; BID, twice daily; TID, four times per day; Sources: Applicant submitted output files, Applicant’s response to information request submitted on January 6, 2020.

Table 43. Summary of simulated pemigatinib PK parameters at steady state when it is co- administered with CYP3A inhibitors

AUC Pemigatinib dose Perpetrator and dose C (nM) ss,24 C (nM) max,ss (nM*hr) min,ss 9 mg QD Itraconazole 200 mg QD 197 3250 97 9 mg QD Clarithromycin 500 mg BID 194 3169 95 9 mg QD Clarithromycin 250 mg BID 178 2795 81 13.5 mg QD 233 3503 97 Erythromycin 500 mg BID 9 mg QD 157 2356 65 13.5 mg QD 233 3512 97 Diltiazem 60 mg TID 9 mg QD 163 2503 71 13.5 mg QD 260 4106 119 Fluconazole 400 mg QD 9 mg QD 176 2768 80 13.5 mg QD 239 3644 102 Fluconazole 200 mg QD 9 mg QD 161 2456 69 13.5 mg QD Fluvoxamine 50 mg QD 180 2367 56 236 2620 56.8 13.5 mg QD Alone (observed) (56.4%) (54.1%) (81.5%) Simulated PK parameters are expressed as geometric means; observed PK parameters following 13.5 mg QD was expressed as geometric means (CV%); simulations were conducted in cancer-population; Sources: Applicant submitted output files, Applicant’s response to information request submitted on January 6 and January 31, 2020.

3. Can PBPK analyses predict the effects of moderate CYP3A inducers on the PK of pemigatinib? Yes. It was noticed that the rifampin DDI model under-predicted the observed effect of rifampin on the PK of pemigatinib (Table 41). The Applicant indicated that could be due to additional DDI effect on absorption of pemigatinib. Another reason could be that rifampin induces multiple enzyme-mediated pathways, however, the model only considered CYP3A­ mediated induction effect. Nevertheless, as the fmCYP3A of pemigatinib was determined via the itraconazole inhibition DDI study, evaluation of the effect of a moderate CYP3A inducer on the PK of pemigatinib relies on the estimate of the moderate inducer’s induction potential. The default efavirenz model (Simcyp V17) was used to evaluate the effect of a moderate CYP3A inducer on the PK of pemigatinib. The efavirenz model was validated against single and multiple dose PK studies. The CYP3A4 induction parameters were based on mRNA levels and were calibrated against rifampicin in vitro data. The fu,inc and fugut values for efavirenz were optimized based on the comparison of the predicted and observed alfentanil i.v. and p.o. AUC ratios co-administration with efavirenz. The CYP3A4 induction effects of efavirenz was validated against the DDI studies with sensitive CYP3A substrates, alfentanil (IV and PO) and maraviroc. Using the efavirenz model, the predicted pemigatinib CmaxR and AUCR were 0.63 and 0.43,

183 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

respectively, when pemigatinib was co-administered with 600 mg efavirenz QD for 2 weeks vs. pemigatinib was administered alone.

4. Can PBPK analyses predict the effects of pemigatinib on the PK of digoxin (a P-gp substrate)? No, the PBPK analyses cannot be used to quantitatively predict the effects of pemigatinib on the PK of digoxin due to the lack of in vitro in vivo extrapolation for Ki. In general, in vitro Ki values for transporter-mediated DDIs under-predict in vivo DDI magnitude. For P-gp mediated DDI, the difference can vary from the lowest reported literature Ki value (e.g. verapamil P-gp inhibition) to a few hundred lower than in vitro Ki value (in house data). Therefore, a 10-fold lower Ki value may not be able to capture the potential pemigatinib P-gp inhibition risk.

5. Can PBPK analyses predict the effects of pemigatinib on the PK of metformin (an OCT2/MATEs substrate)? No, the PBPK analyses cannot be used to quantitatively predict the effects of pemigatinib on the PK of metformin due to the lack of in vitro in vivo extrapolation for Ki. In the metformin model, the OCT2/MATE1 mediated DDI is completely allocated to OCT2 because alternating MATE1 function does not change metformin exposure. That means the model cannot discriminate OCT2 and/or MATE1 mediated DDIs. However, in vitro study suggested that metformin had lower inhibition Ki values for MATEs compared to that for OCT2 (Burt et al. European Journal of Pharmaceutical Sciences 88 (2016) 70–82). In order to recover the observed cimetidine-metformin DDI, the in vitro cimetidine OCT2 Ki value was decreased by almost 500-fold (124 uM 0.25 uM) (Burt et al. 2016). It could be because that both OCT2 and MATE2 effects were allocated to OCT2. Therefore, a 10-fold lower Ki value may not be able to capture the potential pemigatinib OCT2 inhibition risk.

Conclusions For the reasons describe above, the reviewer concluded the followings. • The PBPK analysis is adequate to evaluate the effects of moderate CYP3A inhibitors and inducers on the PK of pemigatinib. The predicted geometric mean AUC of pemigatinib was increased by about 50% to 90% when pemigatinib was co-administered with a moderate CYP3A inhibitor (Table 42). The predicted geometric mean AUC of pemigatinib was decreased by more than 50% when it was co-administered with efavirenz (a moderate CYP3A inducer) 600 mg QD for 2 weeks. • The PBPK analysis is inadequate to evaluate the effect of pemigatinib on the PK of a P­ gp or OCT2/MATEs substrate due to the lack of quantitative in vitro in vivo extrapolation for Ki (the inhibition constant) values.

19.4.3 Appendix 3. Population PK Analysis

The applicant conducted a population PK analysis to characterize the PK of pemigatinib, identify covariate factors that could affect pemigatinib disposition and compare the individual exposure

184 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

estimates for subsequent exposure-response analysis. Data was collected in all tumor types enrolled across the studies (INCB-54828-101, INCB 54828-102 and INCB 54828-202), The studies included in the population PK analysis and the and the PK sampling times are shown in Table 44 and Table 45.

Table 44. Overview of studies included in the population PK analysis.

Source: The Applicant’s population PK report (dmb-19-120-1), Page 48, Table 2.

Table 45. Summary of PK sampling times

Source: The Applicant’s population PK report (dmb-19-120-1), Page 49, Table 3.

The primary population of the subjects were white (68.2%) including 28 Japanese (8.8%). The median age of all subjects was 59 years old (range: 21-79), 45% were males and the median body weight and BMI were 73.3 kg (range: 39.8-156 kg) and 25.8 kg/m2 (range: 14.3- 58.2), respectively. The median value of the calculated creatinine clearance (MDRD GFR) was 85 mL/min/1.73 m2 (range: 33-229). The majority (213; 67%) of the 318 cancer participants were of normal liver function, 94 (29.6%) were of mild hepatic dysfunction, and 11 (3.5%) were of moderate hepatic impairment based on NCI hepatic impairment classification. The majority of

185 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

the 318 cancer participants had normal (146; 45.9%) or mildly impaired (134; 42.1%) renal function, and 38 (11.9%) had moderate renal impairment. No severe renal and hepatic impairment participant was enrolled in any of the 3 studies based on exclusion criteria. Demographic data and laboratory values for subjects in each of the three studies are shown in Table 46.

Table 46. Demographics and baseline characteristics of pooled study data

Source: The Applicant’s population PK report (dmb-19-120-1), Page 53, Table 10.

186 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The population PK analysis was conducted via nonlinear mixed-effects modeling with the NONMEM software, version 7.4.1 using first-order conditional estimation with INTERACTION option (FOCE+I). Concentration collected before the first dose were excluded from the PK analysis as well as post-dose observations that were below the limit of quantification (BLQ). The plasma concentrations of pemigatinib data were described by a two-compartment disposition model with first-order absorption and linear elimination. Log transformed concentration data was used for modeling in applicant’s analysis. Several covariates including renal impairment, hepatic impairment, hyperphosphatemia, gender, race, age, BW, cancer type (cholangiocarcinoma vs. other cancers), BMI, AST, ALT, ALP, ALB, TBIL, MDRD (creatinine clearance), FGFR2 rearrangement or fusion, treatment (monotherapy vs. combination therapy). The concomitant medications explored were CYP3A4 inhibitors/inducers, gastric pH-modifying agents (PPI and H2 antagonist), and phosphate management agents (phosphate binders and diuretics).

The final population PK parameters fir pemigatinib are presented in Table 47. The final PK model was parameterized in terms of ka, CL/F, Vc/F, Q/F, and Vp/F. Estimated fixed and random effect parameters were estimated with good precision (%RSE < 29%). The magnitude of the interindividual variability was moderate for CL/F (43.4 %CV) and Vc/F (35.1 %CV) but high for ka (127 %CV). Residual variability was moderate.

Table 47. Parameter estimates of final population PK model

Source: The Applicant’s population PK report (dmb-19-120-1), Page 58, Table 16.

187 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The diagnostic plots for the final PK model are shown in Figure 9. Diagnostic plots for the final population PK model Figure 10. Visual predictive check stratified by visit and dose for final population PK model. The VPC (visual predictive check) on the final PK model in are stratified by visit and dose. The model described the observed data well and the model predictions were generally within the 90% prediction intervals. No apparent bias was observed in the overall model fit for the data.

Figure 9. Diagnostic plots for the final population PK model

Source: Reviewer’s analysis

188 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 10. Visual predictive check stratified by visit and dose for final population PK model

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 109-110, Figure 18.

Empirical Bayes estimates of the PK parameters were generated with the final PK model for individual participant. The clearances in mild or moderate renal impairment or hepatic impairment participants are not statistically significantly different from the participants with normal renal or hepatic function. The impact of age, BW, race/ethnicity (Hispanic and Japanese), creatinine clearance, clinical labs (ALB, TBIL, AST, ALT, and ALP), concomitant medication (weak/moderate CYP3A4 inhibitors, phosphate binders, diuretics, PPI and H2­ antagonist), renal and hepatic impairment (mild, moderate), and tumor type (cholangiocarcinoma with FGFR2 rearrangement or fusion) on pemigatinib clearance was not statistically significant. The results were shown on Figure 11.

189 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

Figure 11. Covariate Evaluation of Pemigatinib Oral Clearance Using Final Population PK Model

Source: The Applicant’s population PK report (dmb-19-120-1), Page 111-112, Figure 19.

Figure 12. Comparison of Patient Predicted Parameters and Exposures between Applicant’s and Reviewer’s models.

Source: Reviewer’s analysis.

Reviewer’s comments: The population PK model developed by the applicant was verified by the reviewer. The model appears to be reasonable in general because there was a good agreement between observations and predictions. The reviewer also conducted an independent population PK

190 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

analysis with untransformed concentration data to further verify applicant’s model. Similar model structure, individual patient predicted parameters and exposures were obtained which was shown in Figure 12.

Using the final population PK model, individual post hoc clearances (CL/F) were used to estimate the influence of renal impairment and hepatic impairment on PK. The results were verified by the reviewer and acceptable. Evaluation of the influence of CYP3A4 inhibitors on PK with population PK were not appropriate as there was not sufficient information (detailed dose given, the time of drug administration and the time of drug discontinuation) recorded.

19.4.4 Appendix 4. Exposure-Response Analysis

The relationship of pemigatinib and change in serum phosphate concentration from baseline (mean of C1D8 and C1D15) was evaluated with patients across the studies INCB-54828-101, INCB 54828-102 and INCB 54828-202. The correlation of the pemigatinib exposure (AUCss) and change in serum phosphate concentrations was described by Emax model. The increase in serum phosphate was exposure-dependent and followed a sigmoidal relationship. Model- predicted vs. observed relationship of pemigatinib AUCss and serum phosphate concentration change from baseline was shown in Figure 13.

Figure 13. Model-predicted vs. observed relationship of serum phosphate concentration change from baseline and pemigatinib AUCss.

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 113, Figure 20.

The relationship of objective response rate (ORR) and change in serum phosphate concentration from baseline was evaluated with participants in study INCB 54828-202 Cohort A (cholangiocarcinoma with FGFR2 rearrangement or fusion). The relationship followed a bell- shaped curve and a binary logistic regression with quadratic function was used to evaluate the relationship. The model predicted ORR at dose 6, 9, 13.5 and 20 mg suggesting that 13.5 mg is

191 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

an optimal dose for clinical development. Model predicted vs. observed relationship of ORR and change in serum phosphate concentration from baseline was shown Figure 14.

Figure 14. Model-predicted vs. observed relationship of ORR and serum phosphate concentration change from baseline or pemigatinib exposure (Cmax, ss)

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 114, Figure 21.

The exposure-response relationship of ORR and pemigatinib exposure was also evaluated with participants in study INCB 54828-202 Cohort A. The relationship followed a bell-shaped curve and a binary logistic regression with quadratic function was used to evaluate the relationship. Cmax, ss was identified to have the strongest association with ORR based on AIC criteria. The model predicted ORR (42.8%) at dose 13.5 mg is similar as the ORR (44%) predicted from change in serum phosphate concentration from baseline. Model predicted vs. observed relationship of ORR and Cmax, ss was shown in Figure 15.

Figure 15. Model-predicted vs. observed relationship of ORR and pemigatinib exposure (Cmax, ss)

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 115, Figure 22.

Reviewer’s comments:

192 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The result of the binary logistic regression models with quadratic function for ORR vs. serum phosphate concentration change from baseline and ORR vs. pemigatinib exposure were verified by the reviewer. The reason for low response rate in patients with higher steady state exposure was not identified. The relationships of ORR vs. serum phosphate concentration change from baseline and ORR vs. pemigatinib exposure are inconclusive. As the serum phosphate concentration change from baseline seems to associate with ORR, the relationship between ORR and hyperphosphatemia status was also analyzed by the reviewer. The result was shown on Table 48 and no strong relationship was identified. The negative predictive value is 64.5% and hyperphosphatemia status is not a sensitive predictor for the efficacy.

Table 48. Summary of ORR by Cycle 1 or During treatment hyperphosphatemia status

Hyperphosphatemia At Cycle 1 During Treatment Status ORR (n/N) ORR (n/N)

Yes 36.4% (20/55) 33.9% (20/59)

34.6% (18/52, 4 became No hyperphosphatemia in 37.5% (18/48) later cycle)

The relationships of PFS and change in serum phosphate concentration from baseline or PFS and pemigatinib exposure were evaluated with participants in study INCB 54828-202 Cohort A. The log-rank test did not show significant differences in PFS among 4 quartiles (Figure 16).

Figure 16. PFS Vs. change in serum phosphate concentration from baseline or AUCss following APPEARS THIS WAY ON ORIGINAL once daily dosing of 13.5 mg pemigatinib

Source: The Applicant’s population PK report (dmb-19-120-1), Page 116, Figures 23 and 24

Reviewer’s comments:

193 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

The results of survival analysis for PFS were verified by reviewer. As the median PFS in 4th quartile AUCss is higher than the other three quartiles, reviewer combined the survival data from th the first three quartiles AUCss into one group to compare with 4 quantile. The median PFS for st rd th 1 -3 quartiles AUCss is 6.9 (95%CI: 6.1, 9.6) months, while the median PFS for 4 quartile AUCss is 14.4 (95%CI: 4.7, NA) months. No significant difference was identified in these two groups.

The relationship of pemigatinib exposure and increase in serum creatinine concentration (C1D8) was evaluated with participants across the studies INCB-54828-101, INCB 54828-102 and INCB 54828-202. The correlation of the pemigatinib exposure (AUCss) and percent change in serum creatinine concentration from baseline was described by the Emax model. The increase in serum phosphate was exposure-dependent and followed a sigmoidal relationship. Model-predicted vs. observed relationship of pemigatinib AUCss and serum creatinine concentration change from baseline was shown in Figure 17.

Figure 17. Model-predicted vs. observed relationship of percent change in serum creatinine concentration from baseline and pemigatinib AUCss.

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 117, Figure 25.

The relationship of AE response (TEAEs with incidence rate >20% or clinically notable AEs) and pemigatinib exposure at steady state were evaluated with participants in monotherapy treatment across the studies INCB 54828-101, INCB 54828-102, and INCB 54828-202. No statistically significant correction was identified for steady-state pemigatinib exposures and any TEAEs or clinically notable AEs evaluated in this analysis except the significant correlation between pemigatinib exposures (AUCss, Cmax,ss, and Cmin,ss) and hyperphosphatemia. Cmax,ss has shown the strongest association with hyperphosphatemia response by stepwise logistic regression analysis. The odds ratio of hyperphosphatemia incidence rate was increased by 5.8% for every 16.7 nM increase in Cmax,ss. The estimated probability of hyperphosphatemia incidence at dose of 6, 9, 13.5, and 20 mg is 52.1%, 59.0%, 64.9%, and 77.6%, respectively. Model- predicted vs. observed relationship of pemigatinib Cmax, ss and probability of

194 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

hyperphosphatemia response was shown in Figure 17. Probability of Hyperphosphatemia Vs. Pemigatinib Cmax, ss

Figure 17. Probability of Hyperphosphatemia Vs. Pemigatinib Cmax, ss

APPEARS THIS WAY ON ORIGINAL

Source: The Applicant’s population PK report (dmb-19-120-1), Page 117, Figure 26.

Reviewer’s comments: The logistic regression model for hyperphosphatemia and pemigatinib exposures was verified by the reviewer. Positive relationship between the probability of hyperphosphatemia and pemigatinib exposures was identified, which was also confirmed by the model for serum phosphate concentration changing from baseline vs. pemigatinib exposures.

195 Version date: April 2, 2018

Reference ID: 4594191 NDA/BLA Multi-disciplinary Review and Evaluation NDA 213736 PEMAZYRE (pemigatinib)

19.5 Additional Clinical Outcome Assessment Analyses

Not applicable.

196 Version date: April 2, 2018

Reference ID: 4594191 NDA 213736

Signatures

SECTIONS AUTHORED/ DISCIPLINE REVIEWER OFFICE/DIVISION AUTHORED/ APPROVED APPROVED Select one: Nonclinical Sections: 5 and 19.1 X Authored Reviewer Emily Wearne OOD/DHOT Approved

Digitally signed by Emily F. Wearne -S Signature: DN: c=US, o=U.S. Government, ou=HHS, Emily F. ou=FDA, ou=People, 0.9.2342.19200300.100.1.1=2001296929, cn=Emily F. Wearne -S Wearne -S Date: 2020.04.14 08:24:14 -04'00' Select one: Nonclinical Sections: 5 and 19.1 X Authored Team Whitney Helms OOD/DHOT Leader X Approved

Signature: Digitally signed by Whitney S. Helms -S DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, 0.9.2342.19200300.100.1.1=2000585776, Whitney S. Helms -S cn=Whitney S. Helms -S Date: 2020.04.13 17:42:06 -04'00' Select one: Nonclinical Sections: 5 and 19.1 Authored Team John Leighton OOD/DHOT Division X Approved Director Signature: Digitally signed by John K. Leighton -S DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, 0.9.2342.19200300.100.1.1=1300085260, John K. Leighton -S cn=John K. Leighton -S Date: 2020 04 14 08:11:29 -04'00' Select one: Clinical Sections: 6 and 19.5 X Authored Pharmacology Robert Schuck OTS/OCP Reviewer Approved

Signature: Digitally signed by Robert Schuck Robert DN: cn=Robert Schuck, o, ou, [email protected], c=US Schuck Date: 2020 04 14 09:28:55 04'00' Select one: Clinical Sections: 19.5.2 Authored Pharmacology Yuching Yang OTS/OCP/DPM Reviewer X Approved

Signature: Digitally signed by Yuching Yang -S DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Yuching Yang -S cn=Yuching Yang -S, 0.9.2342.19200300.100.1.1=2000846164 Date: 2020.04.13 18:02:50 -04'00'

Reference ID: 4594191 NDA 213736

SECTIONS AUTHORED/ DISCIPLINE REVIEWER OFFICE/DIVISION AUTHORED/ APPROVED APPROVED Select one: Clinical Jeanne Fourie Authored OTS/OCP Sections: 6 and 19.5 Pharmacology Team Zirkelbach Leader X Approved

Dig tally signed by Jeanne Fourie Zirkelbach S Signature: DN: c=US o=U S Government ou=HHS ou=FDA ou=People Jeanne Fourie Zirkelbach -S 0 9 2342 19200300 100 1 1=1300434575 cn=Jeanne Fourie Zirkelbach S Date: 2020 04 13 18:40:32 04'00' Select one: Clinical Sections: 6, 19.5.1, Authored Pharmacology Nam Atiqur Rahman OTS/OCP/DCPII 19.5.2, 19.5.3, 19.5.4, Division Director X Approved

Digitally signed by Nam A. Rahman -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, cn=Nam A. Rahman ­ Nam A. Rahman -S S, 0.9.2342.19200300.100.1.1=1300072597 Date: 2020.04.14 08:51:29 -04'00' Select one: Pharmacometrics Sections:19.5.3 and Authored Reviewer Yangbing Li OTS/OCP/DPM 19.5.4 X Approved

Signature: Digitally signed by Yangbing Li -S (Affiliate) DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Yangbing Li -S (Affiliate) 0.9.2342.19200300.100.1.1=2002587726, cn=Yangbing Li -S (Affiliate) Date: 2020.04.14 11:04:39 -04'00' Pharmacology Select one: Reviewer Xinyuan Zhang OTS/OCP/DPM Sections: 19.5.2 X Authored Approved Signature: Digitally signed by Xinyuan Zhang -S DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, cn=Xinyuan Zhang -S, Xinyuan Zhang -S 0.9.2342.19200300.100.1.1=2000431943 Date: 2020.04.13 15:21:49 -04'00' Genomics Team Select one: Leader Rosane Charlab OTS/OCP Sections: 6 X Authored Orbach X Approved

Digitally signed by Rosane Charlaborbach -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Rosane Charlaborbach -S 0.9.2342.19200300.100.1.1=1300436672, cn=Rosane Charlaborbach -S Date: 2020.04.13 16:29:57 -04'00'

Pharmacometrics Select one: Team Leader Jiang Liu OTS/OCP/DPM Sections: 6 and 19.5 X Authored X Approved

Signature: Digitally signed by Jiang Liu -S DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Jiang Liu -S cn=Jiang Liu -S, 0.9.2342.19200300.100.1.1=2000348510 Date: 2020.04.13 17:04:31 -04'00'

Reference ID: 4594191 NDA 213736

SECTIONS AUTHORED/ DISCIPLINE REVIEWER OFFICE/DIVISION AUTHORED/ APPROVED APPROVED Select one: Associate Director of Ann Marie Sections: 11 x_ Authored Labeling (acting) CDER/OND Trentacosti x Approved

Digitally signed by Ann M. Trentacosti -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Ann M. Trentacosti -S 0.9.2342.19200300.100.1.1=1300216343, cn=Ann M. Trentacosti -S Date: 2020.04.14 15:13:52 -04'00' Select one: Clinical Reviewer Margit (Naomi) X Authored OOD/DO3 Sections: 1.4, 2, 3, 4, 7, Horiba 8, 8.1, 8.4, 9, 19.1, 19.2 Approved

Digitally signed by Margit Horiba -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Margit Horiba -S cn=Margit Horiba -S, 0.9.2342.19200300.100.1.1=2001698250 Date: 2020.04.17 12:07:00 -04'00' Select one: Clinical Reviewer Leigh Marcus OOD/DO3 X Authored Sections: 8.2, 1.3 Approved Digitally signed by Leigh J. Marcus -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Leigh J. Marcus -S 0.9.2342.19200300.100.1.1=0013338033, cn=Leigh J. Marcus -S Date: 2020.04.17 15:05:19 -04'00' Select one: Statistical Reviewer Somak Chatterjee OTS/OB/DBV Sections: 8.1, 8.3 x Authored Approved

Signature: Digitally signed by Somak Chatterjee -S Somak Chatterjee ­ DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, 0.9.2342.19200300.100.1.1=2002732785, cn=Somak Chatterjee -S S Date: 2020.04.13 22:07:11 -04'00' Select one: Statistical Team Leader Pallavi Mishra- Authored OTS/OB/DBV Sections: 8.1, 8.3 Kalyani X Approved

Digitally signed by Pallavi S. Mishra-kalyani -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Pallavi S. Mishra-kalyani -S 0.9.2342.19200300.100.1.1=2001675542, cn=Pallavi S. Mishra-kalyani -S Date: 2020.04.13 15:49:06 -04'00'

Reference ID: 4594191 NDA 213736

SECTIONS AUTHORED/ DISCIPLINE REVIEWER OFFICE/DIVISION AUTHORED/ APPROVED APPROVED Clinical Team Leader/ Select one: Cross-disciplinary Sections: 1 X Authored Team Leader Martha Donoghue OOD/DO3 (CDTL) Sections: (all) X Approved

Digitally signed by Martha Donoghue -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, 0.9.2342.19200300.100.1.1=2000339007, Martha Donoghue -S cn=Martha Donoghue -S Date: 2020.04.17 15:15:43 -04'00' Select one: Division Sections: All X Authored Director Lola Fashoyin-Aje OOD/DO3 (Clinical) X Approved

Digitally signed by Ibilola Fashoyin-aje -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Ibilola Fashoyin-aje -S 0.9.2342.19200300.100.1.1=0014067124, cn=Ibilola Fashoyin-aje -S Date: 2020.04.17 14:09:15 -04'00' Select one: Division Director Sections: 8.1, 8.3, 8.4 x Authored (OB) Shenghui Tang OTS/OB/DBV x Approved

Digitally signed by Shenghui Tang -S Signature: DN: c=US, o=U.S. Government, ou=HHS, ou=FDA, ou=People, Shenghui Tang -S cn=Shenghui Tang -S, 0.9 2342.19200300.100.1.1=1300224175 Date: 2020.04.16 12:20:50 -04'00'

Reference ID: 4594191 Signature Page 1 of 1 ------This is a representation of an electronic record that was signed electronically. Following this are manifestations of any and all electronic signatures for this electronic record. ------/s/ ------

STACIE A WOODS 04/17/2020 05:23:58 PM

MARTHA B DONOGHUE 04/17/2020 05:29:43 PM

IBILOLA A FASHOYIN-AJE 04/17/2020 05:31:39 PM

MARC R THEORET 04/17/2020 05:54:34 PM My signature indicates that I have considered the assessments and recommendations included in this Review in determining the regulatory action

Reference ID: 4594191