`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`202570Orig1s000
`
`STATISTICAL REVIEW(S)
`
`
`
`
`
`
`
`

`

`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Sciences
`Office of Biostatistics
`
`
`S TAT I S T I C A L R E V I E W A N D E VA L U AT I O N
`CLINICAL STUDIES - STATISTICAL REVIEWER’S ADDENDUM
`
`NDA/BLA Serial
`Number:
`Drug Name:
`Indication(s):
`
`Applicant:
`Date(s):
`
`Review Priority:
`
`
`
`NDA 202570
`
`CRIZOTINIB
`Anaplastic lymphoma kinase (ALK)-positive advanced non-small
`cell lung cancer (NSCLC)
`Pfizer, Inc.
`Submission Date: 30 March 2011
`PDUFA Due Date: 30 September 2011
`Priority
`
`Biometrics Division:
`Division of Biometrics 5 (HFD-711)
`Statistical Reviewer:
`Lijun Zhang, Ph.D.
`Concurring Reviewers: Shenghui Tang, Ph.D., Team Leader
`
`Rajeshwari Sridhara, Ph.D., Division Director
`Medical Division:
`Oncology Drug Products (HFD-150)
`Clinical Team:
`Shakun Malik, M.D.
`Virginia Ellen Maher, M.D., Team Leader
`Diane Hanner
`
`
`Project Manager:
`
`Keywords:
`Anaplastic Lymphoma Kinase, Non-Small Cell Lung Cancer, Single-Arm, Objective
`Response Rate, Response Duration
`
`
`
`
`
`
`Reference ID: 3005052
`
`

`

`This is an addendum to Dr. Lijun Zhang’s statistical review (dated August 5, 2011).
`
`ORR Analysis per Investigator Assessments (Primary Efficacy Analysis)
`
`FDA and the applicant reached a consensus that patient 10391003 in Study 1005 was a
`partial responder. Therefore, the total number of responders should be 68 out of 135
`patients evaluable for response. The objective response rate and its corresponding 95%
`confidence interval was calculated as 50% (95% CI: 42%, 59%) for Study 1005.
`
`ORR Analysis per Independent Radiology Reviews
`
`In Study 1005, IRR response rate was 41.9% (95% CI: 32.3%, 51.9%) in IRR response
`evaluable patient population (n=105), and was 32.3% (95% CI: 24.6%, 40.9%) in safety-
`analysis population (n=136).
`
`In Study 1001, IRR response rate was 52.4% (95% CI: 42.4%, 62.2%) in IRR response
`evaluable patient population (n=105), and was 46.2% (95% CI: 37.0%, 55.6%) in safety-
`analysis population (n=119).
`
`On 17 August 2011, the applicant submitted updated objective response data per IRR
`assessments for both studies. Study 1005 had 1 complete response and 62 partial
`responses, with an IRR response rate of 46.3% (95% CI: 37.7%, 55.1%) in safety-
`analysis population (n=136). Study 1001 had 63 partial responses, with an IRR response
`rate of 52.9% (95% CI: 43.6%, 62.2%) in safety-analysis population (n=119).
`
`Reference ID: 3005052
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`LIJUN ZHANG
`08/23/2011
`
`RAJESHWARI SRIDHARA
`08/23/2011
`
`Reference ID: 3005052
`
`

`

`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Sciences
`Office of Biostatistics
`
`
`S TAT I S T I C A L R E V I E W A N D E VA L U AT I O N
`CLINICAL STUDIES-TEAM LEADER’S MEMO
`
`NDA/BLA Serial
`Number:
`Drug Name:
`Indication(s):
`
`Applicant:
`Date(s):
`
`Review Priority:
`
`
`
`NDA 202570
`
`CRIZOTINIB
`Anaplastic lymphoma kinase (ALK)-positive advanced non-small
`cell lung cancer (NSCLC)
`Pfizer, Inc.
`Submission Date: 30 March 2011
`PDUFA Due Date: 30 September 2011
`Review Completion Date: 05 August 2011
`Priority
`
`Biometrics Division:
`Division of Biometrics 5 (HFD-711)
`Primary Reviewer:
`Lijun Zhang, Ph.D.
`Secondary Reviewer:
`Shenghui Tang, Ph.D., Team Leader
`Concurring Reviewer: Rajeshwari Sridhara, Ph.D., Division Director
`Medical Division:
`Oncology Drug Products (HFD-150)
`Clinical Team:
`Shakun Malik, M.D.
`Virginia Ellen Maher, M.D., Team Leader
`Diane Hanner
`
`
`Project Manager:
`
`Keywords:
`Anaplastic Lymphoma Kinase, Non-Small Cell Lung Cancer, Single-Arm, Objective
`Response Rate, Response Duration
`
`
`Reference ID: 2984709
`
`

`

`This is an original New Drug Application (NDA) submission seeking an accelerated
`approval of crizotinib for the treatment of anaplastic lymphoma kinase (ALK) positive
`advanced non-small cell lung cancer (NSCLC). The applicant has submitted results from
`two single-arm studies: A8081005 (phase 2, second-line therapy) as a pivotal study and
`A8081001 (phase 1 expansion cohort) as a supportive study. The primary efficacy
`endpoint was the objective response rate (ORR) in both studies. Per FDA analyses, the
`ORR was 49.6% (95% CI: 40.9%, 58.4%) in Study A8081005 and 61.2% (95% CI:
`51.7%, 70.1%) in Study A8081001, based on the investigator tumor assessments. The
`medians of response duration were 41.9 and 48.1 weeks in studies A8081005 and
`A8081001, respectively.
`
`For further details regarding the designs, data analyses, and results of both Study
`A8081005 and Study A8081001, please refer to the statistical review by Dr. Lijun Zhang,
`(August 5, 2011).
`
`
`This team leader concurs with the recommendations and conclusions of the statistical
`reviewer (Dr. Lijun Zhang) of this application. The efficacy conclusions should rely on
`clinical judgment, since there were no comparators in these two single-arm studies.
`Whether the endpoint and the size of the effect on this endpoint are adequate for
`accelerated approval is a clinical decision.
`
`
`Reference ID: 2984709
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`SHENGHUI TANG
`08/05/2011
`
`RAJESHWARI SRIDHARA
`08/05/2011
`
`Reference ID: 2984709
`
`

`

`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Sciences
`Office of Biostatistics
`
`
`S TAT I S T I C A L R E V I E W A N D E VA L U AT I O N
`CLINICAL STUDIES
`
`NDA/BLA Serial
`Number:
`
`Drug Name:
`Indication(s):
`
`Applicant:
`Date(s):
`
`Review Priority:
`
`
`
`NDA 202570
`
`CRIZOTINIB
`Anaplastic lymphoma kinase (ALK)-positive advanced non-small
`cell lung cancer (NSCLC)
`Pfizer, Inc.
`Submission Date: 30 March 2011
`PDUFA Due Date: 30 September 2011
`Review Completion Date: 04 August 2011
`Priority
`
`Biometrics Division:
`Division of Biometrics 5 (HFD-711)
`Statistical Reviewer:
`Lijun Zhang, Ph.D.
`Concurring Reviewers: Shenghui Tang, Ph.D., Team Leader
`
`Rajeshwari Sridhara, Ph.D., Division Director
`Medical Division:
`Oncology Drug Products (HFD-150)
`Clinical Team:
`Shakun Malik, M.D.
`Virginia Ellen Maher, M.D., Team Leader
`Diane Hanner
`
`
`Project Manager:
`
`Keywords:
`Anaplastic Lymphoma Kinase, Non-Small Cell Lung Cancer, Single-Arm, Objective
`Response Rate, Response Duration
`
`
`
`Reference ID: 2984676
`
`

`

` Table of Contents
`
`LIST OF TABLES ........................................................................................................................ 3
`
`1. EXECUTIVE SUMMARY.................................................................................................... 4
`
`2. INTRODUCTION.................................................................................................................. 4
`2.1 OVERVIEW......................................................................................................................... 4
`2.2 DATA SOURCES ................................................................................................................. 6
`3. STATISTICAL EVALUATION........................................................................................... 6
`3.1 DATA AND ANALYSIS QUALITY ........................................................................................ 6
`3.2 EVALUATION OF EFFICACY .............................................................................................. 6
`3.3 EVALUATION OF SAFETY................................................................................................ 21
`4.
` FINDINGS IN SPECIAL/SUBGROUP POPULATIONS .............................................. 21
`4.1 GENDER, RACE, AGE, AND GEOGRAPHIC REGION ........................................................ 21
`4.2 OTHER SPECIAL/SUBGROUP POPULATIONS................................................................... 22
`5. SUMMARY AND CONCLUSIONS................................................................................... 23
`5.1 STATISTICAL ISSUES AND COLLECTIVE EVIDENCE ....................................................... 23
`5.2 CONCLUSIONS AND RECOMMENDATIONS ...................................................................... 23
`SIGNATURES/DISTRIBUTION LIST.................................................................................... 24
`
`CHECK LIST.............................................................................................................................. 25
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`2
`
`Reference ID: 2984676
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`LIST OF TABLES
`
`Table 1: Overview of Studies Included in the Submission............................................ 5
`Table 2. Power Calculation for ALK-Negative NSCLC Cohort ................................ 10
`Table 3. Patient Disposition, Safety-Analysis Population ........................................... 11
`Table 4. Treatment Duration, Safety-Analysis population ......................................... 11
`Table 5. Demographics Characteristics, Safety-Analysis Population........................ 12
`Table 6. Baseline Disease Characteristics and Prior Tumor Treatment, Safety-
`Analysis population.......................................................................................... 13
`Table 7. Summary of Best Overall Response based on Investigator Assessments,
`Response-Evaluable Population ..................................................................... 14
`Table 8. Summary of Best Overall Response per Independent Reviewers, IRR
`Response-Evaluable Population ..................................................................... 16
`Table 9. Response Status by Assessment Types ........................................................... 17
`Table 10. Duration of Response..................................................................................... 18
`Table 11. Progression-Free Survival and Overall Survival, Study 1001 Safety-
`Analysis population.......................................................................................... 19
`Table 12. Number of Prior Advanced/Metastatic Regimens (FDA analyses) ........... 20
`Table 13. Summary of Best Overall Response and Response Duration of Study 1005
`(FDA Analyses)................................................................................................. 21
`Table 14. ORR by Gender, Age, Race, and Region, FDA-Response-Evaluable
`Population......................................................................................................... 22
`Table 15. Additional ORR Subgroup Analyses, FDA-Response-Evaluable
`Population......................................................................................................... 23
`
`
`
`3
`
`Reference ID: 2984676
`
`

`

`1.
`
`EXECUTIVE SUMMARY
`
`Crizotinib, a new molecular entity (NME), is a small-molecule inhibitor of anaplastic lymphoma
`kinase (ALK) receptor tyrosine kinase. In the current original New Drug Application (NDA)
`submission, the applicant seeks an accelerated approval of crizotinib for the treatment of ALK
`positive advanced non-small cell lung cancer (NSCLC). It is based primarily on two single-arm
`studies: A8081005 (phase 2, second-line therapy) as a pivotal study and A8081001 (phase 1
`expansion cohort) as a supportive study. The primary efficacy endpoint was the objective
`response rate (ORR) in both studies. Per FDA analyses, the ORR was 49.6% (95% CI: 40.9%,
`58.4%) in Study A8081005 and 61.2% (95% CI: 51.7%, 70.1%) in Study A8081001, based on
`the investigator tumor assessments. The medians of response duration were 41.9 and 48.1 weeks
`in studies A8081005 and A8081001, respectively. The efficacy conclusions should rely on
`clinical judgment, since there were no comparators in these two single-arm studies. Whether the
`endpoint and the size of the effect on this endpoint are adequate for accelerated approval is a
`clinical decision.
`
`2.
`
`INTRODUCTION
`
`2.1 Overview
`
`Non-small cell lung cancer (NSCLC) is the most common fatal malignancy in the United States,
`with an ORR of 15% to 32% by first-line treatment and less than 10% by second-line treatment
`in patients with advanced NSCLC. ALK-positive NSCLC accounts for 3% to 5% of all NSCLC.
`
`The proposed indication is for the treatment of ALK-positive advanced NSCLC. There are four
`ongoing clinical studies of crizotinib in ALK-positive advanced NSCLC under IND 73,544:
`A8081001 (phase 1 expansion cohort), A8081005 (phase 2, second-line therapy), A8081007
`(phase 3, second-line therapy), and A8081014 (phase 3, first-line therapy). This NDA is based
`primarily on two single-arm studies: A8081005 as a pivotal study and A8081001 as a supportive
`study. For simplicity, the last 4 characters of each study ID will be used to represent each study
`hereafter.
`
`Study 1005 was entitled “Phase 2, Open-Label, Single Arm study of the Efficacy and Safety of
`PF-02341066 in Patients with Advanced Non-Small Cell Lung Cancer Harboring a
`Translocation or Inversion Involving the Anaplastic Lymphoma Kinase (ALK) Gene Locus”.
`The original protocol of Study 1005 was dated 24 June 2009 and amended 9 times thereafter.
`Following implementation of Amendment 1 (dated 12 August 2009), patients who were
`ineligible to enroll in Study 1007 could enroll in this study. In Amendment 2 (dated 27 August
`2009), the RECIST version was modified to Version 1.1.
`
`Enrollment of Study 1005 was ongoing as of the study cutoff date (15 September 2010) in the
`original NDA submission, and 148 patients have been enrolled from 66 study sites in North
`America, Europe, Asia, and Australia. A total of 136 patients have received at least one dose of
`study treatment. The 60-day clinical data update (cutoff date: 01 February 2011) provided further
`efficacy data for these 136 patients, though 265 patients have been enrolled at this time.
`
`
`
`Reference ID: 2984676
`
`4
`
`

`

`Study 1001 was entitled “A phase 1 Safety, Pharmcokinetic and Pharmacodynamic Study of PF-
`02341066, a c-Met/HGFR selective Tyrosine Kinase Inhibitor, Administered Orally to Patients
`with Advanced Cancer”. The original protocol of Study 1001 was dated 05 December 2005 and
`amended 15 times thereafter. The study was initially designed as a phase 1 dose-escalation study
`followed by a recommended phase 2 dose (RP2D) expansion cohort to further evaluate the safety
`and PK of the MTS of crizotinib. In Amendment 4, EML4-ALK-positive NSCLC patients were
`allowed to enter. In Amendment 12, a cohort consisting of ALK FISH negative NSCLC patients
`was added (n=25-40) to assist in validating the companion FISH diagnostic. In Amendment 14,
`retrospective evaluation of all tumor scans from ALK-positive and ALK-negative NSCLC
`patients by independent radiology group (IRR) was added.
`
`As of the clinical data cutoff date (15 September 2010), Study 1001 enrolled 38 patients in the
`dose-escalation cohort, 119 patients in the RP2D ALK-positive NSCLC cohort, 5 patients in the
`RP2D ALK-negative NSCLC cohort, and 50 in the RP2D other cohort, from 8 sites in the United
`States, Korea, and Australia. The study accrual is ongoing. On June 10, 2011, the applicant
`submitted an updated report for the ALK-negative NSCLC cohort per the Agency’s request. A
`total of 23 ALK-negative NSCLC patients were enrolled and treated as of 27 May 2011.
`
`The primary efficacy endpoint in both studies was objective response rate (CR + PR) based on
`the investigator tumor assessments.. No statistical inference on comparison was conducted in
`both single-arm, open-label studies. Two retrospective analyses, i.e., a covariate-matched
`analysis and a covariate-adjusted modeling analysis, were performed to support the primary
`findings from the single-arm study, 1001.
`
`In Study 1005, ALK-positive NSCLC was identified using the Vysis ALK Break-Apart FISH
`Probe Kit assay which is under FDA review for marketing. In Study 1001, ALK-positive
`NSCLC was identified using a number of local clinical trial assays.
`
`The randomized studies 1007 and 1014 are both open-label studies with PFS as the primary
`endpoint, and Study 1007 is powered for overall survival as well. The accrual for both studies is
`ongoing.
`
`Table 1: Overview of Studies Included in the Submission
`Population, Phase,
`Treatment
`Follow-Up
`Number of
`and Study Design
`Period
`Period
`Enrolled
`Patientsa
`ALK-positive:
`148
`
`
`
`
`
`
`
`
`
`
`Treated until
`PD,
`unacceptable
`toxicity, consent
`withdrawal, or
`protocol
`noncompliance
`
`
`
`
`
`Follow-up for
`survival every
`2 months until
`death, or until
`the last patient
`discontinued
`crizotinib
`treatment,
`whichever
`came first
`
`
`Study No.
`
`1005
`(pivotal)
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 2984676
`
`Phase 2, Open-
`Label, Single Arm
`study for Efficacy
`and Safety in
`Patients with ALK-
`positive Advanced
`NSCLC
`
`
`
`
`
`Efficacy
`Endpoints
`
`Primary:
`ORR
`Secondary:
`DR, DCR,
`OS, PFS
`
`
`
`
`
`5
`
`

`

`
`
`1001
`(supportive)
`
`
`
`Treated until
`PD,
`unacceptable
`toxicity, consent
`withdrawal,
`non-protocol
`anti-cancer
`therapy, or
`investigator’s
`decision
`
`
`
`Follow-up for
`survival every
`3 months for a
`minimum of 1
`year after last
`dose of
`crizotinib
`
`
`
`
`Phase 1 Safety,
`Pharmcokinetic and
`Pharmacodynamic
`Study, in Patients
`with Advanced
`Cancer
`
`
`
`
`
` a Enrollment number is based on the data from the original NDA submission.
` b Study 1001 ALK-negative cohort has enrolled 25 patients in the 10 June 2011 update, and 23 patients were
`confirmed as ALK-negative.
`
`
`
`
`Primary:
`ORR
`Secondary:
`DR, DCR,
`OS, PFS
`
`
`
`Dose escalation
`cohort: 38
`
`ALK-positive
`cohort: 119
`
`ALK-negative
`cohort: 5b
`
`Other: 50
`
`
`2.2 Data Sources
`
`Electronic submission including protocols, SAPs, study reports, and analysis datasets for the
`original NDA submission is located on network with network path:
`\\CDSESUB1\EVSPROD\NDA202570\0002\. The 60-day clinical data update for studies 1005
`and 1001 is located at: \\CDSESUB1\EVSPROD\NDA202570\0009\. On June 13, 2011, the
`applicant submitted an update on ALK-negative cohort of Study 1001, and the network path is
`\\CDSESUB1\EVSPROD\NDA202570\0016\
`
`3.
`
`STATISTICAL EVALUATION
`
`3.1 Data and Analysis Quality
`
`There are no major issues identified on data quality from the statistical perspective, except for
`some minor issues, such as inadequate comments in the data define file which was then revised
`according to the FDA’s request.
`
`3.2 Evaluation of Efficacy
`
`The efficacy of this application was based on two single-arm studies, study 1005 and study 1001,
`for the indication of ALK-positive advanced NSCLC treatment. For Study 1005, this efficacy
`review is based on the 60-day update data; for Study 1001, this review focuses on ALK-positive
`NSCLC cohort.
`
`3.2.1 Overall Study Design
`
`Study 1005
`
`Study 1005 is an ongoing, open-label, single–arm phase 2 study of oral crizotinib in patients with
`ALK-positive advanced NSCLC who have been treated with at least one prior chemotherapy
`
`
`
`Reference ID: 2984676
`
`6
`
`

`

`regimen. Patients were treated with 250 mg BID oral crizotinib continuously in 3-week cycles
`until the occurrence of PD or clinical deterioration, unacceptable toxicity, patient’s withdrawal of
`consent, or protocol noncompliance. However, crizotinib treatment could be continued beyond
`RECIST-defined PD, if, in the judgment of the investigator, there was evidence of clinical
`benefit. Disease assessment was to have included a CT or MRI scan at baseline, and every 6
`weeks from the date of first dose of crizotinib. The evaluation of antitumor efficacy for ALK-
`positive NSCLC was based on investigator-assessed ORR by RECIST (version 1.1).
`
`Study 1001
`
`Study 1001 is an ongoing phase 1 study of oral crizotinib to identify the MTD in patients with
`advanced cancers and evaluate the efficacy in RP2D enriched cohorts.
`
`The study includes 3 parts:
`
`1. Dose escalation for determination of MTD for twice daily (BID) dosing.
`
`2. RP2D cohorts
`a. ALK-positive NSCLC
`b. ALK-negative NSCLC, and
`c. Other – ALK-dependent tumors other than NSCLC and c-Met-dependent tumors
`
`3. Dose escalation for determination of MTD for once daily (QD) dosing. This cohort was
`initiated shortly before the database cut-off date for this NDA submission and was not
`included in the clinical study report.
`
`In the ALK-positive NSCLC cohort, patients were treated with 250 mg BID oral crizotinib
`continuously in 4-week cycles, until the occurrence of progressive disease (PD) or clinical
`deterioration, unacceptable toxicity that did not improve with dosing interruption, dose
`reduction, and/or standard medical therapy, patient’s withdrawal of consent, investigator’s
`determination that it was in the patient’s best interest to discontinue therapy, or initiation of
`treatment with another anticancer therapy. Crizotinib treatment could be interrupted to allow
`surgery and/or palliative radiation therapy to localized sites of disease progression. Crizotinib
`treatment could have been continued beyond RECIST-defined PD if, in the opinion of the
`investigator, the benefit/risk assessment justified continuation of treatment.
`
`Disease assessment at baseline (screening) was to include a CT or MRI scan of the chest,
`abdomen, and pelvis; brain and bone scans were to be performed if disease at these sites was
`suspected. Scans were to be repeated at all sites of known disease every 2 cycles (i.e., every 8
`weeks unless treatment delayed). The antitumor efficacy evaluation was based on investigator-
`assessed ORR according to RECIST (version 1.0).
`
`3.2.1.2 Efficacy Endpoints
`
`For both Study 1005 and Study 1001, the primary efficacy endpoint was overall confirmed ORR,
`defined as the proportion of patients with confirmed complete response (CR) or confirmed
`partial response (PR) according to RECIST (RECIST v1.1 and RECIST v1.0 in 1005 and 1001,
`
`
`7
`
`Reference ID: 2984676
`
`

`

`respectively), relative to the response-evaluable population (defined in Section 3.2.1.4).
`Confirmed responses were those that persist on repeat imaging study ≥4 weeks after initial
`documentation of response. In both studies, the primary analyses of ORR used the investigator’s
`recorded measurements and assessments for target, non-target, and new lesions to
`programmatically evaluate response using rules based on RECIST. In addition, all available
`scans were retrospectively reviewed by IRR.
`
`The secondary efficacy endpoints included duration of response (DR), time to tumor response
`(TTR), disease control rate (DCR), progression-free survival (PFS), and overall survival (OS).
`
`Time to Response was defined as the time in weeks from the date of the first dose of crizotinib
`to first documentation of objective tumor response (CR or PR) that was subsequently confirmed.
`TTR was only calculated for patients in the response-evaluable population who had a confirmed
`objective tumor response.
`
`Duration of Response was defined as the time in weeks from the first documentation of
`objective tumor response (CR or PR) that was subsequently confirmed to the first documentation
`of objective disease progression or death on study due to any cause, whichever occurred first.
`DR was only calculated for patients in the response-evaluable population who had a confirmed
`objective tumor response.
`
`Disease Control Rate at 6 weeks, 8 weeks, and 16 weeks were defined as the percent of
`patients in the response-evaluable population with a confirmed CR, confirmed PR, or SD
`according to RECIST-defined tumor response assessments at 6 and 12 weeks (Study 1005), or 8
`and 16 weeks (Study 1001), respectively, after the first dose of crizotinib.
`
`Progression-Free Survival was defined in months as the time from the date of the first dose of
`crizotinib to the date of first documentation of objective tumor progression or death due to any
`cause, whichever occurred first, in the safety-analysis population. Only deaths that occurred
`within 2 assessment intervals (~ 16 weeks) after the last dose of crizotinib were included in the
`PFS analysis.
`
`Overall Survival was defined in months as the time from the date of the first dose of crizotinib
`to the date of death due to any cause in the safety-analysis population. All deaths were included
`in the analysis.
`
`3.2.1.4 Efficacy Analysis Population
`
`The safety-analysis (SA) population included all enrolled patients who received at least one dose
`of crizotinib staring on Cycle 1 Day 1. The safety-analysis population was the primary
`population for evaluating patient characteristics, treatment administration, and safety endpoints.
`
`The response-evaluable (RE) population was defined as all patients in the safety-analysis
`population who had an adequate baseline disease assessment. The response-evaluable population
`was used in the primary efficacy analyses.
`
`
`
`Reference ID: 2984676
`
`8
`
`

`

`In addition, for any preliminary (interim) reporting of the data, patients also needed to meet 1 of
`the following 2 criteria:
`• Had at least one post-baseline disease assessment performed at least 6 weeks after treatment
`start
`• Withdrew from the study or experienced progression/death at any time on study
`
`The independent review response-evaluable (IRR RE) population was defined identically to the
`response-evaluable population, however, the assessments were based on IRR rather than the
`investigator, with the following exceptions: patients were included in the IRR response-evaluable
`population without having baseline tumor evaluation by IRR if the investigator evaluation was
`recorded as an early death (within 42 days from first dose) or withdrawn from the study without
`a tumor assessment.
`
`3.2.1.4 Sample Size Determination
`
`The target sample size of Study 1005 was 250. The sample size was determined based on the
`expected number (n=100) of patients who would cross-over from the chemotherapy comparator
`arm of Study 1007, an ongoing, randomized, phase 3 study of crizotinib in patients with
`previously treated ALK-positive NSCLC, and additional patients (n=150) who would be enrolled
`based on other eligibility criteria. This sample size was also considered adequate to detect
`adverse events of low frequency (≥1%).
`
`The number of patients enrolled in the dose escalation phase of Study 1001 was dependent upon
`the observed safety profile and study objectives, which would determine the number of patients
`per dose level, the number of dose escalations and the number of cohorts. It was anticipated that
`a total of approximately 40 patients would be enrolled in the dose escalation phase.
`
`The RP2D ALK-positive NSCLC cohort in Study 1001 was originally designed to enroll at least
`25 patients. During the study, enrollment was expanded to further explore the safety and efficacy
`of this cohort. There was no specified sample size.
`
`The main objective of the RP2D ALK- negative NSCLC cohort in Study 1001 was to evaluate
`the objective response in this group of patients and to compare with the objective response
`observed from ALK-positive NSCLC patients enrolled in Study 1007 and/or 1005. Response to
`crizotinib among ALK-negative patients was expected to be low, therefore, the ALK-negative
`NSCLC cohort in Study 1001 was first limited to a total of 25 patients. If ≤ 3 objective
`responses (CR or PR) have been observed in the first 25 ALK marker negative patients, no
`additional ALK marker negative NSCLC patients would have been enrolled into this trial. If > 3
`objective responses have been observed among the 25 ALK marker negative patients, additional
`patients would have been enrolled in this trial as noted in Table 2.
`
`
`
`
`
`
`Reference ID: 2984676
`
`9
`
`

`

`Table 2. Power Calculation for ALK-Negative NSCLC Cohort
`Additional
`Total ALK
`Exact 90% CI
`Exact 90% CI Around
`Responses in
`ALK -Patients
`- Patients in
`* Around
`40% ORR Assumed for
`First 25
`to be Enrolled
`this Trial
`ORR (column
`160 ALK + Patients
`ALK -
`1/ column 3 x 100)
`(Protocol 1007/1005)
`Patients
`(5%, 28%)
`(34%, 47%)
`30
`5
`4
`(6%, 28%)
`(34%, 47%)
`35
`10
`5
`(7%, 27%)
`(34%, 47%)
`40
`15
`6
`Note: if ≥ 7 responses have been observed among the first 25 ALK marker negative subjects then no additional
`patients would have been enrolled beyond 40 until read out of study 1007 and/or 1005 study results.
`
`[Source: Study 1001 SAP V3 Table 4]
`
`Reviewer’s Comments:
`
`In Study 1007, patients randomized to the comparator arm (standard chemotherapy with
`docetaxel or pemetrexed) who experienced disease progression, confirmed by the independent
`radiology laboratory used for the study, were permitted to crossover to crizotinib treatment by
`enrolling in Study 1005.
`
`3.2.1.5 Efficacy Analysis Methods
`
`The primary efficacy endpoint was the objective response rate (ORR = CR+ PR) and the primary
`efficacy analysis was based on the response-evaluable population. Exact 2-sided 95% confidence
`intervals based on the F distribution were calculated for all proportion estimates.
`
`Estimates of time-to-event endpoints were obtained using the Kaplan-Meier method. DR and
`TTR were also summarized using descriptive statistics for confirmed objective responders.
`
`3.2.2. Efficacy Results from the Applicant
`
`3.2.2.1 Patient Disposition, Demographic and Baseline Characteristics
`
`Patients Enrollment and Treatment Discontinuation
`
`As of the cutoff date for the original NDA submission, Study 1005 enrolled 148 patients, of
`whom, 136 patients received study treatment (including 13 patients who crossed over from the
`comparator arm of Study 1007) and had documentation of initiating crizotinib treatment. Three
`patients were enrolled in error and did not receive study treatment, and 9 patients did not yet
`have documentation of study treatment administration. Update on these 136 treated patients was
`submitted in the 60-day clinical update, in which, 43 (29.1%) patients discontinued treatment,
`and 93 (62.8%) patients were ongoing with treatment, as shown in Table 3.
`
`Study 1001 enrolled a total of 174 patients to the RP2D cohorts, including 119 patients in the
`ALK-positive NSCLC cohort, 5 patients in the ALK-negative NSCLC cohort, and 50 in the
`RP2D other cohort from 8 sites in the United States, Korea, and Australia. Six patients in the
`RP2D other cohort did not receive study treatment. At the time of study cut-off, the proportions
`
`
`
`Reference ID: 2984676
`
`10
`
`

`

`of patients who discontinued treatment were 35.3% in the ALK-positive NSCLC cohort, 40.0%
`in the ALK-negative NSCLC cohort, and 84.1% in the RP2D other cohort. The most common
`reason for discontinuation was disease progression (Table 3).
`
`Table 3. Patient Disposition, Safety-Analysis Population
`Study 1005
`Study 1001 RP2D Cohorts
`ALK-Positive
`ALK-Positive
`ALK-Negative
`Other
`NSCLC
`NSCLC
`NSCLC
`250 mg
`250 mg BID
`250 mg BID
`250 mg BID
`BID
`136
`119
`5
`44
`93 (68.4)
`77 (64.7)
`3 (60.0)
`7 (15.9)
`
`Number (%) of Patients
`Treated
`Ongoing at data cut-off
`date
`Discontinued
`Adverse event
`Progressive disease
`Patient died
`Patient no longer willing
`to participate in study
`Global deterioration of
`health status
`Lost to Follow-Up
`Other
`
`43 (31.6)
`6 (4.4)
`26 (19.1)
`6 (4.4)
`2 (1.5)
`
`2 (1.5)
`
`42 (35.3)
`3 (2.5)
`25 (21.0)
`8 (6.7)
`1 (0.8)
`
`0
`
`2 (40.0)
`0
`1 (20.0)
`1 (20.0)
`0
`
`0
`
`37 (84.1)
`3 (6.8)
`24 (54.5)
`1 (2.3)
`2 (4.5)
`
`0
`
`1 (0.7)
`0
`
`0
`5 (4.2)
`
`0
`0
`
`0
`7 (15.9)
`
`
` [Source: Study 1005 60-day clinical updates Table 13.1.3.1.1 and Study 1001 CSR Table 13.1.3.1b]
`
`Treatment Exposure
`
`As of the study cut-off date, the median duration of treatment was 22.3 weeks and 31.8 weeks in
`Studies 1005 and 1001 for all treated patients, r