`
`
`
`
`
`
`
`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`202788Orig1s000
`
`
`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
`
`NDA/Serial Number:
`Drug Name:
`Indication(s):
`
`Applicant:
`Date(s):
`Review Priority:
`
`202-788
`
`Subsys (Fentanyl) Sublingual Spray
`
`Management of breakthrough cancer pain in opioid tolerant patients with
`malignancies
`Insys Therapeutics, Inc.
`
`Letter date: March 4, 2011, PDUFA date: January 4, 2012
`
`Standard
`
`
`
`
`
`II
`
`Biometrics Division:
`Statistical Reviewer:
`Yan Zhou, Ph.D.
`Concurring Reviewers: Dionne Price, Ph.D.
`
`
`
`Division of Analgesia, Anesthesia and Addiction Products
`
`
`Medical Division:
`Clinical Team:
`
`Luke Yip, M.D.
`Sharon Hertz, M.D.
`
`
`
`
`
`Keywords: Clinical Studies, cross-over design, ANOVA, permutation test
`
`
`
`
`
`
`
`
`Reference ID: 3051809
`
`
`
`1
`
`

`

`
`Table of Contents
`
`
`
`1. EXECUTIVE SUMMARY .................................................................................................................................4
`2.
`INTRODUCTION ...............................................................................................................................................5
`2.1
`OVERVIEW......................................................................................................................................................5
`2.2
`DATA SOURCES ..............................................................................................................................................7
`3. STATISTICAL EVALUATION ........................................................................................................................7
`3.1
`DATA AND ANALYSIS QUALITY .....................................................................................................................7
`3.2
`EVALUATION OF EFFICACY ............................................................................................................................7
`3.3
`EVALUATION OF SAFETY..............................................................................................................................10
`4.
` FINDINGS IN SPECIAL/SUBGROUP POPULATIONS ............................................................................10
`4.1
`GENDER, RACE, AGE, AND GEOGRAPHIC REGION ........................................................................................10
`4.2 OTHER SPECIAL/SUBGROUP POPULATIONS ..................................................................................................11
`5. SUMMARY AND CONCLUSIONS ................................................................................................................11
`5.1
`STATISTICAL ISSUES AND COLLECTIVE EVIDENCE .......................................................................................11
`5.2
`CONCLUSIONS AND RECOMMENDATIONS .....................................................................................................12
`SIGNATURE/DISTRIBUTION LIST.....................................................................................................................15
`
`
`
`
`Reference ID: 3051809
`
`2
`
`

`

`LIST OF TABLES
`
`Table 1: list of the study included in analyses ...............................................................................................................7
`Table 2: Patient Disposition...........................................................................................................................................8
`Table 3: Baseline Demographic Characteristics for ITT population (N=96).................................................................8
`Table 4: Applicant’s Primary Efficacy Re-Analysis....................................................................................................10
`Table 5: Reviewer’s Primary Efficacy Analysis..........................................................................................................10
`Table 6: Reviewer's Subgroup Analyses for SPID30 ..................................................................................................11
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3051809
`
`3
`
`

`

`EXECUTIVE SUMMARY
`
`1.
`
`Insys Therapeutics Inc. submitted a New Drug Application for Subsys (fentanyl) sublingual
` seeking approval for the proposed indication of “management of breakthrough cancer
`pain in patients with malignancies who are already receiving and who are tolerant to around-the-
`clock opioid therapy for their underlying persistent cancer pain.” I conclude that evidence from
`the efficacy study was statistically in favor of Subsys in comparison to placebo as measured by
`the sum of pain intensity difference from baseline to 30 minutes after dosing (SPID30).
`
`The submission contained one efficacy study, INS-05-001, which was a multi-center, crossover
`study to evaluate the safety and efficacy of Subsys. Subjects who successfully titrated during an
`open-label titration phase then entered into a randomized, double-blind, placebo-controlled phase
`of up to 26 days, during which 10 episodes of breakthrough pain were treated with Subsys (7
`episodes) or placebo (3 episodes).
`
`The primary objective of the double-blind phase was to assess the analgesic efficacy of Subsys.
`The primary efficacy endpoint was the summed pain intensity difference at 30 minutes after
`dosing. Secondary endpoints were total pain relief at 30 minutes, subject’s global evaluation of
`study medication at 30 minutes and pain intensity difference (PID) at different time points.
`
`The applicant calculated the mean of the primary efficacy variable SPID30 across episodes for
`each treatment and then analyzed the difference between two averaged values using an analysis
`of covariance (ANCOVA) model with the mean baseline pain intensity across all episodes as a
`covariate. Similar analyses were performed for secondary endpoints. There was a concern that
`the design might not be balanced with respect to the episodes and that the analyses did not
`account for the correlated measurements from each subject. Thus, the agency requested the
`applicant submit analyses of SPID30 using an analysis of variance (ANOVA) model with fixed
`effects for treatment, episode, sequence and a random effect for subject. In response, the
`applicant submitted their re-analyses of SPID30. In addition, the applicant also conducted a
`permutation test to confirm their primary analysis based on the ANOVA model.
`
`The applicant defined the Intent-to-Teat (ITT) population as all randomized subjects who took at
`least one dose of study medication and had at least one pain measurement following
`administration of study medication. The efficacy analysis set included all subjects in ITT
`population who took at least one breakthrough pain episode treated with Subsys and another
`treated with placebo. This analysis set was acceptable because an exclusion of subjects who
`received only one treatment in a crossover design would not lead to the same bias in estimating
`the treatment effect as in a parallel design study. In the study, one subject did not take the
`assigned sequence, but had pain intensity (PI) recorded for all 10 episodes. Two subjects did not
`take their assigned treatments for some episodes (one subject with 5 episodes and another subject
`with 3 episodes). The applicant did not include these episodes in their analyses. To make use of
`all available information, I included these 18 episodes in my analyses.
`
`Based on my review, I conclude that Subsys reduced the pain intensity in patients with
`breakthrough cancer pain when compared to placebo.
`
`
`
`
`Reference ID: 3051809
`
`4
`
`(b) (4)
`
`

`

`2.
`
`INTRODUCTION
`
`2.1 Overview
`
`Fentanyl, the active ingredient in Subsys, is an opioid analgesic. Oral transmucosal fentanyl
`citrate (Actiq), fentanyl citrate buccal tablet Gentora) and fentanyl citrate buccal film (Onsolis)
`have been approved for the management of breakthrough pain in opioid tolerant patients with
`cancer. Subsys is a sublingual (SL) spray formulation of fentanyl citrate designed for oral
`administration. According to the applicant, “Fentanyl SL Spray is expected to provide analgesic
`benefit to patients in an easy-to—use rapid-onset format.” The clinical development program and
`statistical analyses were discussed at several meetings. At the pre—IND meeting on August 25,
`2005 (under IND 72,411), the agency stated that one adequate and well-controlled study would
`be needed to investigate the use of Subsys in patients with breakthrough cancer pain. At the End-
`of-Phase 2 meeting on December 17, 2007, the applicant proposed a linear mixed model with
`fixed effects for treatment and time as the main efficacy analysis method. The agency stated that
`the benefit of including an effect for time was unclear and interest would be in the ANCOVA
`model results. The applicant also stated that the baseline observation canied forward (BOCF)
`method would be used to impute pain intensity at time points after the use of supplemental
`medication. The agency stated that the missing data concern in the crossover study design is not
`the same as in the setting of parallel group chronic pain trials.
`mu)
`
`included in the labeling. The following is quoted from the meeting minutes.
`
`would not be
`
`Question 2a
`Insys proposes, as the main analysis method for the primaljv efiicacy measure and related endpoints, using a
`repeated measures linear mixed model, and treating data at time points after the use ofsupplemental ("rescue ")
`medication as missing. Additionally we will peiform sensitivitv analyses, including those using imputation, to
`assess how conclusions about treatment eflect depend on the handling of data after use of supplemental
`medication. Since we understand,
`in some instances, that the agency has adopted the baseline obsen'ation
`carriedforward (BOCF) approach for such data, we will use BOCF to impute pain intensity at time points after
`the use of supplemental medication, and analyze the within subject treatment summonr using the Wilcoxon
`signed rank test. Does the agencv agree with this statistical approach?
`
`FDA Response
`The Division’s concern regarding missing data has primarily been in the setting of parallel group,
`chronic pain trials. In such trials, patients receive treatment for 12 weeks. Patients may experience some
`reduction in pain intensity, however, they drop out of the study because of intolerable side efl'ects. The
`Division has advocated using missing data strategies that assign a bad score to patients experiencing
`unfavorable outcomes.
`
`You propose a crossover study design where patients assess pain intensity for 30 minutes following each
`treatment administration. The missing data concern is not the same as in the setting of parallel group
`chronic pain trials.
`
`In general, a linear mixed model is an acceptable approach for analyzing the data. Your model will
`include fixed effects for treatment and time. The benefit of including an effect for time is unclear.
`Including terms for sequence and/or period may be more beneficial. Additional comments will be
`provided once the protocol and statistical analysis plan have been submitted.
`
`Reference ID: 3051 809
`
`

`

`Sponsor Reply (provided prior to Industry Meeting)
`Insys noted FDA’s comment that the “benefit of including an effect for time is unclear.” Insys would like to
`clarify how the time effect is needed to identify the 30-minute time point of our main efficacy endpoint, As noted
`on p. 29 of the briefing document, the primary efficacy endpoint, i.e., the summed Page 7 IND 72,411 Insys
`Therapeutics Inc. EOPII Meeting Minutes Fentanyl Sublingual Spray pain intensity differences at 30 minutes
`[SPID(30)], is defined mathematically as a linear combination of pain intensity (PI) at time points up and
`including 30 minutes.
`Specifically:
`SPID(30) = 30*PI(0) – 5*PI(5) – 5*PI(10) – 5*PI(15) – 15*PI(30).
`
`However, rather than pre-calculating SPID(30) before statistical analysis, which might require imputation for
`missing data, we have chosen to implement the mathematical definition within the modeling and to allow the
`modeling to handle missing data automatically in the normal course of model fitting, without external
`imputation rules.
`
`To see how this might work, consider an implementation of the mixed model using SAS, with PI as dependent
`variable and with the treatment (TRT) and time (TIME) factors as fixed effects. Suppose the levels of TRT are
`coded as 0 = Placebo and 1= Fentanyl SL Spray, and the levels of TIME as 0, 5, 10, 15, 30, 45 and 60
`(minutes). Given the model parameters and SPID as a function of PI, a statement in SAS to assess the treatment
`effect with respect to SPID(30) is:
`
`Contrast "Trt effect SPID(30)" TRT*TIME -30 5 5 5 15 0 0 30 -5 -5 -5 -15 0
`0;
`
`Insys noted the comment that “including terms for sequence and/or period may be beneficial.” In the current
`analysis plan, the period effect is considered random, nested within subject. As a sensitivity analysis we will
`model period as a fixed effect, crossed with the subject effect. Also, there are 29 sequences, i.e., 29 different
`orderings of 3 placebo and 7 Fentanyl SL Spray treatments to which a subject may be randomized; we will
`examine the sequence effect descriptively.
`
`Insys noted the comment that “additional comments will be provided once the protocol and statistical analysis
`plan have been submitted.” Insys submitted the statistical analysis plan at the agency’s request on December 5.
`If any questions or comments remain after our teleconference on December 17, Insys will look forward to
`hearing and discussing them.
`
`Discussion
`Ms. Meaker noted that the Agency’s comment was related to the fact that linear models are often employed for
`longer study timepoints, so the Division was not sure these were the appropriate models to utilize. However,
`from the draft statistical analysis plan (SAP) the firm shared by email, she understands that the Agency will see
`both this analysis and the ANCOVA for the SPID (30) endpoint.
`
`This is acceptable with the understanding that the Agency is interested first in the ANCOVA model results. Ms.
`Meaker stated that it is acceptable for the sponsor to conduct mixed-model imputation as a sensitivity analysis,
`noting that any discrepancies will need to be discussed in the study report.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The sponsor stated that they will amend their SAP based on the comments received and officially submit it to
`the IND.
`
`During the August 17, 2010 pre-NDA meeting, the agency restated that a graphical
`representation of the primary efficacy endpoint
`
` may be included in the labeling.
`
`
`Study INS-05-001 was designed to comply with the agency’s requirement and support the
`application. My statistical review focuses on Study INS-05-001 which was a double-blind,
`randomized, placebo-controlled, multi-center, crossover trial.
`
`
`
`Reference ID: 3051809
`
`6
`
`(b) (4)
`
`

`

` Table 1: list of the study included in analyses
`Study Number
`Number of
`Sample Size
`(Dates Conducted)
`Centers
`(Locations)
`
`35 centers
`
`(All US)
`
`
`INS-05-001
`
`(10/2007 – 02/2010)
`
`
`
`
`Titration:
` n = 130
`
`Randomization:
` Subsys
` n= 98
` Placebo
` n= 98
`
`
`
`2.2 Data Sources
`
`Type of
`Control
`
`
`Placebo
`
`Duration of
`Treatment
`
`
`Titration:
` 26 days
`
`Double-Blind
`Treatment:
` 26 days
`
`Design
`
`
`Randomized,
`Double-blind,
`Cross-over,
`Placebo-
`controlled,
`Multicenter with
`an open-label
`titration phase
`
`
`STATISTICAL EVALUATION
`
`
`The initially submitted data can be found at \\Cdsesub1\evsprod\NDA202788\0003\m5\datasets.
`The applicant submitted datasets containing the raw data from the Case Report Form (CRF) as
`well as the derived analysis datasets. The datasets were not provided in SDTM or AdaM format.
`The applicant didn’t submit analysis-ready datasets initially. On April 8, 2011, we requested the
`applicant submit analysis-ready datasets. The applicant submitted additional datasets per the
`Division’s request, which can be accessed at \\Cdsesub1\evsprod\NDA202788\0007\m5\datasets.
`
`
`3.
`
`
`3.1 Data and Analysis Quality
`
`
`The resubmitted define document for the datasets clearly specified the source or the derivation of
`most variables. I was able to reproduce the secondary variables of interest as well as the primary
`outcome.
`
`
`3.2 Evaluation of Efficacy
`
`
`
`Study Design and Endpoints
`
`
`After identification of an effective Subsys dose in the open-label titration phase, eligible subjects
`entered the double-blind phase. Each subject was given 10 doses of study medication, 7 doses of
`Subsys sublingual spray at the stable dose identified during the titration phase and 3 matching
`placebo doses. The ordering of Subsys and placebo doses was determined at random. There were
`totally 29 possible treatment sequences, and each patient was randomly assigned to one of them.
`The PI was assessed by the subject using a 0-100 mm visual analog scale (VAS), where one
`anchor represented no pain and the other reflected the worst possible pain. The PI was assessed
`at the following times during each breakthrough pain episode: 0, 5, 10, 15, 30, 45 and 60 minutes.
`
`One hundred and sixty-one subjects were enrolled from 35 sites in the United States. One
`hundred and thirty subjects entered into the titration phase. Ninety-eight subjects achieved an
`
`
`
`Reference ID: 3051809
`
`7
`
`

`

`individualized successful dose during the titration phase and were randomized to one of 29
`sequences with 10 treatment episodes.
`
`The primary objective of the study was to demonstrate the superiority in analgesic efficacy of
`Subsys sublingual spray compared to placebo. The primary efficacy endpoint was the summed
`pain intensity difference from baseline to 30 minutes after dosing. Secondary endpoints included
`total pain relief at 30 minutes, subject’s global evaluation of study medication at 30 minutes and
`pain intensity difference at different time points.
`
`
`Patient Disposition, Demographic and Baseline Characteristics
`
`Number (%) of Patients
`161
`130
` 98 (100)
`96 (98)
`95 (97)
`79 (81)
`3 (3)
`1 (1)
`1 (1)
`1 (1)
`
`
`The disposition of subjects is shown in Table 2. A total 161 subjects were enrolled into the study.
`Among 98 randomized subjects, 3 subjects discontinued prior to completing the study, 1 due to
`adverse events (AE), 1 due to subject’s decision and 1 did not comply with the protocol.
`
` Table 2: Patient Disposition
`
`Screened
`Titrated
`Randomization
`ITT*
`Completed
` Completed 10 episodes
`Discontinued
` Adverse events
` not complied with protocol
` Subject’s decision
` Source: Reviewer’s Analyses
` *Two subjects (114001, 119003) have no efficacy data due to an equipment malfunction
`
`The demographic and baseline characteristics for ITT population are shown in Table 3. The
`majority of the subjects were white (91%), and the mean age was 54 years.
`
` Table 3: Baseline Demographic Characteristics for ITT population (N=96)
`Age (years)
`
`Mean (SD)
` 54 (12)
`Range
` 24-85
`
`
`Age Group (years), n (%):
`
`< 65
` 80 (83%)
`>=65
` 16 (17%)
`
`Race, n (%)
`White
` 87 (91%)
`Black or African American
`7 (7%)
`Native Hawaiian or Other Pacific Islander
`1 (1%)
`American Indian or Alaskan Native
`1 (1%)
`Other
` 1 (1%)
`Gender, n (%)
`
`Female
` 52 (54%)
`Male
` 44 (46%)
` Source: Clinical Study Report Table 14.1.4
`
`
`
`
`Reference ID: 3051809
`
`8
`
`

`

`Statistical Methodologies
`
`
`The applicant’s analyses for the primary efficacy endpoint, SPID30, were based on the mean of
`SPID30 across episodes for each treatment. For each subject, SPID30 of the 7 episodes treated
`with Subsys were averaged into a single value and SPID30 of the 3 episodes treated with placebo
`were averaged into a single value and the difference of these two averaged values was then
`analyzed by using an ANCOVA model with the mean baseline pain intensity across all episodes
`as a covariate. During the review, I requested the applicant submit analyses of SPID30 using an
`ANOVA model with fixed effects for treatment, episode, sequence and a random effect for
`subject. The requested analysis appropriately accounts for the correlation that arises from
`multiple measurements coming from each subject. Responding to the information request, the
`applicant submitted their re-analyses of SPID30. During the review, a permutation test on the
`primary comparison was also requested due to the possibility of confounding with an unbalanced
`randomization scheme. The applicant responded to the request and performed the permutation
`test. The test confirmed their primary results based on the ANOVA model.
`
`The ITT population was defined as all randomized subjects who took at least one dose of study
`medication and had at least one pain measurement following administration of study medication.
`All efficacy analyses were based on the efficacy analysis set which included all subjects in ITT
`population who took at least one breakthrough pain episode treated with Subsys and another
`treated with placebo. Among 96 subjects in the ITT population, 4 subjects only took episode(s)
`of one treatment. The applicant’s efficacy analysis set did not include these 4 subjects. In the
`review, I found among these 4 subjects, 1 subject did not take the assigned sequence, but had PI
`recorded for all 10 episodes. My analysis set included this subject. In addition, the applicant did
`not include 8 episodes in which subjects did not take their assigned treatments. To make use of
`all available information, I included these 8 episodes in my analyses.
`
`Efficacy data recorded after rescue medication was taken for an episode were disregarded, and
`the missing values were imputed using the last observation carried forward method (LOCF) for
`that episode. For subjects that discontinued from the study during an episode, LOCF was used
`for that episode. Missing values in episodes after dropout were not imputed at all and subsequent
`episodes were excluded from the analyses.
`
`Results and Conclusions
`
` I
`
` replicated the applicant’s primary analysis. In both the applicant’s analysis (Table 4) and my
`analysis (Table 5), Subsys sublingual spray was statistically significantly different from and
`superior to placebo in terms of the primary efficacy variable SPID30. The secondary endpoints
`were also favorable for Subsys sublingual spray. In the study, 57 subjects took rescue medication.
`Only 14 subjects took rescue medication during the initial 30 minutes.
`
`
`
`
`
`
`
`
`
`Reference ID: 3051809
`
`9
`
`

`

` Table 4: Applicant’s Primary Efficacy Re-Analysis
`SPID30
`Subsys Sublingual Spray
`(N of subjects = 92)
`(N of episodes = 620)
`656
`(43)
`
`
`< 0.0001
`
`Placebo
`(N of subjects = 92)
`(N of episodes = 266)
`394
`(47)
`
`
`
`
`
`Placebo
`(N of subjects = 93)
`(N of episodes = 272)
`387
`(45)
`
`
`
`LSMEANS
`(SE)
`Difference from Placebo
`95% CI
`P-value*
` Source: study report: Table 14.2.21
` * P-value based on the ANOVA model with fixed effect treatment, episode, sequence and a random effect
` subject
`
`
` Table 5: Reviewer’s Primary Efficacy Analysis
`SPID30
`Subsys Sublingual Spray
` (N of subjects = 93)
`(N of episodes = 632)
`644
`LSMEANS
`(41)
`(SE)
`257 (29)
`Difference from Placebo (SE)
`(200, 315)
`95% CI
`
`< 0.0001
`P-value*
` * P-value based on the ANOVA model with fixed effect treatment, episode, sequence and a random effect
` subject
`
`
`
`3.3 Evaluation of Safety
`
`
`The evaluation of the safety data was conducted by Dr. Luke Yip. The reader is referred to Dr.
`Yip’s review for information regarding the adverse event profile. Safety risks appear consistent
`for this drug type.
`
`
`4.
`
`My subgroup analyses didn’t reveal any issues that were concerning. The SPID30 was higher for
`Subsys across the subgroups including gender and age.
`
`
` FINDINGS IN SPECIAL/SUBGROUP POPULATIONS
`
`4.1 Gender, Race, Age, and Geographic Region
`
`The applicant performed subgroup analyses for gender (female and male), age (<60 and ≥60
`years, <65 and ≥65 years, and <75 and ≥75 years), race, type of around-the-clock pain
`medication used, type of prior breakthrough pain medication used, and successful dose of Subsys
`in their original analyses but did not perform subgroup analyses using the agency-requested
`analyses. I conducted subgroup analyses for gender (female and male) and age (<65 and >=65).
`Race was not included in the assessment of subgroups because the majority of the study
`population was white. In my analyses, I utilized the same ANOVA model with additional terms
`for each demographic variable and its interaction with treatment.
`
`
`
`
`Reference ID: 3051809
`
`10
`
`

`

`There was no statistically significant interaction between gender and treatment. Although there
`was a statistically significant interaction between age and treatment, for both age groups the
`mean of SPID30 was greater in the Subsys group compared to the placebo group.
`
` Table 6: Reviewer's Subgroup Analyses for SPID30
`Subsys
`Placebo
`
`Endpoint
`n Mean (SD) n Mean (SD)
`SPID30
`
`
`
`
` Gender
`
`
`
`
` Female
`50
`621 (514)
`50
`401 (493)
` Male
`43
`667 (634)
`43
`388 (529)
`
`
`
`
`
` Age (years)
`
`
`
`
` < 65
`77
`667 (594)
`77
`390 (521)
` >= 65
`16
`525 (447)
`16
`415 (453)
`
`
`
`4.2 Other Special/Subgroup Populations
`
`
`No other subgroup analyses were requested by Dr. Yip.
`
`
`5.
`
`
`SUMMARY AND CONCLUSIONS
`
`5.1 Statistical Issues and Collective Evidence
`
`The applicant calculated the mean of the primary efficacy endpoint SPID30 across episodes for
`each treatment and analyzed the difference between two averaged values by using an ANCOVA
`model. Since the study design may not have been balanced with respect to the episode effect, the
`division requested the applicant conduct an additional analysis of SPID30 by using an ANOVA
`model with fixed effects for treatment, episode, sequence and a random effect subject. This
`requested analysis also accounted for the correlation among measurements that is apparent in
`crossover studies. A permutation test was also requested by the division to address the concern
`that the randomization scheme may not be balanced. In response, the applicant re-analyzed
`SPID30 by using the requested ANOVA model and performed the requested permutation test
`which confirmed their primary results based on the ANOVA model.
`
`The applicant did not include 18 episodes in which subjects did not take their assigned
`treatments. I included these 18 episodes in my analyses to make use of all available information.
`
`In the study, dropout was not a concern, and missing data were handled appropriately.
`
`Since the applicant proposed a novel sublingual formulation of fentanyl, a well-known active
`substance for the treatment of pain, the division required demonstration of the efficacy in a single
`adequate and well-controlled clinical trial. The data from Study INS-05-001 provided
`statistically significant evidence of the efficacy of Subsys sublingual spray as a treatment of
`breakthrough pain in cancer patients.
`
`
`
`
`
`Reference ID: 3051809
`
`11
`
`

`

`5.2 Conclusions and Recommendations
`
`The study reviewed provides adequate evidence of the analgesic effect of Subsys. Cancer
`patients receiving Subsys for breakthrough pain experienced a greater reduction in pain intensity
`compared to patients receiving placebo.
`
` 5.2.1 Labeling
`
`The applicant submitted the following wording for the draft label:
`
`
`
`
`
`
`Reference ID: 3051809
`
`12
`
`(b) (4)
`
`2 Page(s) of Draft Labeling has been Withheld in Full as B4 (CCI/TS) immediately following
`this page
`
`

`

`
`
`
`
`November 18, 2011
`
`Signature/Distribution List
`
`Primary Statistical Reviewer: Yan Zhou, Ph.D.
`
`
`
`
`Mathematical Statistician
`
`Date:
`
`Concurring Reviewer:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Dionne Price, Ph.D.
`Team Leader
`
`
`
`
`
`Reference ID: 3051809
`
`15
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`YAN ZHOU
`11/30/2011
`
`DIONNE L PRICE
`11/30/2011
`Concur
`
`Reference ID: 3051809
`
`

`

`
`
`Statistics Filing Checklist of New NDA
`Division of Biometrics II
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Date: 4/26/11
`
`Priority Classification: S
`
`Applicant: Insys Therapeutics, Inc.
`
`
`
`
`NDA #: 202-788
`
`Proposed Trade Name: SUBSYS
`
`Generic Name: Fentanyl Sublingual Spray Date of Submission: 3/4/11
`
`Indication: for the management of breakthrough cancer pain in patients with
`malignancies who are already receiving and who are tolerant to around-the-clock opioid
`therapy for their underlying persistent cancer pain.
`
`No. of Controlled Studies: 1
`
`
`
`
`User Fee Goal Date: 1/4/12
`
`Date of 45-Day Meeting: 4/13/11
`
`Medical Officer: Yip, Luke, M.D. (DAAP)
`
`Project Manager: Davies, Kathleen (DAAP)
`
`Statistical Reviewer: Zhou, Yan, Ph.D.
`
`Statistical sections: Sections 2.5, 2.7, and 5.3.5
`
`Anticipated Review Completion Date: 11/30/11
`
`
`
`Comments:
`
`
`1. It is fileable.
`
`
`NDA 202-788 Stat Filing Checklist.doc
`Zhou, Yan
`
`Page 1
`
`5/3/2011
`
`Reference ID: 2941325
`
`

`

`
`
`CHECKLIST
`Item
`
`
`Index sufficient to locate necessary reports, tables, etc.
`
`
`Original protocols & subsequent amendments available in the
`NDA
`
`
`Designs utilized appropriate for the indications requested
`
`
`Endpoints and methods of analysis spelled out in the
`protocols
`
`
`Interim analyses (if present) planned in the protocol and
`appropriate adjustments in significance level made
`
`
`Appropriate references included for novel statistical
`methodology (if present)
`
`
`Sufficient data listings and intermediate analysis tables to
`permit statistical review
`
`
`Data from primary studies in electronic data room
`
`
`Intent-to-treat analysis
`
`
`Effects of dropouts on primary analyses investigated
`
`
`Safety and efficacy for gender, racial, and geriatric subgroups
`investigated
`
`
`
`
`
`
`Check
`(NA if not applicable)
`
`Yes
`
`
`Yes
`
`
`
`Yes
`
`
`Yes
`
`
`
`NA
`
`
`NA
`
`
`Yes
`
`
`
`Yes
`
`
`Yes
`
`
`Yes
`
`
`
`Yes
`
`NDA 202-788 Stat Filing Checklist.doc
`Zhou, Yan
`
`Page 2
`
`5/3/2011
`
`Reference ID: 2941325
`
`

`

`
`
`BRIEF SUMMARY OF CONTROLLED CLINICAL TRIALS
`
`Study
`Number
`(Dates
`Conducted)
`
`
`INS-05-001
`
`(10/07 – 2/10)
`
`
`
`Number of
`Centers
`(Locations)
`
`
`35 centers
`
`(All US)
`
`Sample
`Size
`
`Type of
`Control
`
`Design
`
`Duration of
`Treatment
`
`
`Placebo
`
`
`Titration:
` n = 130
`
`Randomization:
` Fentanyl
` n= 98
` Placebo
` n= 98
`
`
`Randomized,
`Double-blind,
`Cross-over,
`Placebo-
`controlled,
`Multicenter with
`an open-label
`titration phase
`
`
`
`Titration:
` 26 days
`
`Double-Blind Treatment:
` 26 days
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Concur: Price, Dionne, Ph.D.
`
` Team Leader
`
`
`
`_____________________________
`Zhou, Yan
`Mathematical Statistician
`
`
`NDA 202-788 Stat Filing Checklist.doc
`Zhou, Yan
`
`Page 3
`
`5/3/2011
`
`Reference ID: 2941325
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`YAN ZHOU
`05/03/2011
`
`DIONNE L PRICE
`05/03/2011
`concur
`
`Reference ID: 2941325
`
`