`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`203567Orig1s000
`
`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:
`Dates:
`
`203567 / 000 (Resubmission)
`Jublia (efinaconazole) solution 10%
`Onychomycosis
`Dow
`Submitted: 12/20/2013
`PDUFA:
`6/20/2014
`
`Review Priority:
`
`Resubmission (Class 2)
`
`Division of Biometrics III
`Biometrics Division:
`Kathleen Fritsch, Ph.D.
`Statistics Reviewer:
`Concurring Reviewer: Mohamed Alosh, Ph.D.
`
`Medical Division:
`Clinical Team:
`Project Manager:
`
`Division of Dermatology and Dental Products
`Gary Chiang, M.D. / David Kettl /M.D.
`Strother Dixon
`
`Keywords: Labeling review
`
`Reference ID: 3500837
`
`

`

`1 Regulatory Background
`NDA 203567 for Jublia (efinaconazole) solution 10% for the treatment of onychomycosis
`was originally submitted on 7/26/2012. The NDA received a Complete Response due to
`Product Quality issues. These issues were:
`1.
`Inadequate manufacturing process and control information of the filling/capping/
` operation.
`Inadequate specification for the drug product.
`Inadequate integrity of the container closure system.
`Inadequate stability data to assure the expiration dating period.
`
`2.
`3.
`4.
`
`With this submission, the applicant has submitted information to address the Product
`Quality issues. A complete biostatistical review was conducted during the initial review
`cycle. There were no biostatistical issues raised in the initial review that would preclude
`the conclusion that efficacy had been established in the clinical trials. The team has
`determined that the changes in the manufacturing and control will not necessitate any
`new clinical trials. Thus the conclusions from the initial biostatistical review are still
`applicable. The remaining biostatistical issue that was not addressed in the initial review
`cycle was product labeling. This review will provide biostatistical recommendations on
`the product labeling.
`
`2 Biostatistical Conclusions from the Original Review Cycle
`The following is the Executive Summary from the biostatistical review for the original
`review cycle for Jublia. (Reviewer Kathleen Fritsch, dated 3/5/2013).
`
`Executive Summary
`Efinaconazole solution 10% was superior to vehicle in the treatment of onychomycosis in
`two studies. Studies P3-01 and P3-02 enrolled subjects age 18 to 65 with a clinical
`diagnosis of onychomycosis and positive mycology. Subjects applied treatment once
`daily for 48 weeks. The primary efficacy endpoint was complete cure at Week 52 (0%
`clinical involvement of target toenail plus negative KOH and negative culture). The
`secondary efficacy endpoints specified in the protocol were: (1) clinical efficacy rate at
`Week 52 (<10% affected target nail area), (2) mycological cure rate at Week 52 (negative
`KOH and culture), and (3) unaffected new nail growth at Week 52 (change from baseline
`in healthy target nail measurement). Secondary endpoints were analyzed in sequential
`order. The primary and secondary efficacy endpoints were all statistically significant and
`the results are presented in Table 1.
`
`Reference ID: 3500837
`
`2
`
`(b) (4)
`
`

`

`Table 1 – Primary and Secondary Efficacy Endpoints at Week 52 (SAP 1)
`Study P3-01
`Study P3-02
`Vehicle
`Vehicle
`N = 214
`N = 201
`7 (3.3%)
`11 (5.5%)
`25 (12%)
`24 (12%)
`36 (17%)
`34 (17%)
`1.6 (0.4)
`0.9 (0.4)
`
`Efinacon.
`N = 656
`117 (17.8%)
`234 (36%)
`362 (55%)
`5.0 (0.2)
`
`Complete Cure
`Clinical Efficacy
`Mycologic Cure
`Unaffected new
`growth (mm)
`
`p-value
`
`<0.001
`<0.001
`<0.001
`<0.001
`
`Efinacon.
`N = 580
`88 (15.2%)
`180 (31%)
`310 (53%)
`3.8 (0.2)
`
`p-value
`
`<0.001
`<0.001
`<0.001
`<0.001
`
`The applicant created two versions of the Statistical Analysis Plan (SAP). The first
`version of the SAP was signed off about a week after the last subject completed Study
`P3-01 and the proposed analyses were consistent with the endpoints and analyses
`specified in the protocol. However, the applicant then revised the SAP about 5 weeks
`later. The second version of the SAP redefined the sets of secondary and supportive
`endpoints, and the order in which they were to be analyzed. The primary endpoint
`remained the same in both versions of the SAP, and thus the primary conclusions of the
`study are not affected by the changes to the SAP. This review will focus on the endpoints
`pre-specified in the protocol (and the first version of the SAP), rather than those specified
`only in the second version of the SAP. The secondary endpoints specified in the second
`version of the SAP were only proposed after the studies were completed. Although the
`applicant maintains that the studies were still blinded at that time the second SAP was
`written, changing endpoints after the studies are completed raises the concern that the
`Type I error rate could be inflated. Note that because all of the proposed secondary
`endpoints from either version of the SAP had p-values <0.001, the analyses from the
`second version of the SAP would lead to the same conclusions of efficacy as those from
`the original protocol/first version of the SAP.
`
`3 Applicant’s Proposed Labeling
`The following is the applicant’s proposed labeling for the Clinical Studies Section
`(submission dated 1/16/2014).
`14 CLINICAL STUDIES
`The safety and efficacy of once daily use of JUBLIA for the treatment of onychomycosis
`of the toenail were assessed in two
` 52-week prospective, multi-center,
`randomized,
` studies in patients 18 years and older (18 to 70 years of age) with
`20% to 50% clinical involvement of the area of the target toenail, without
`
`dermatophytomas or lunula (matrix) involvement.
` The
` compared 48-weeks of treatment with
`JUBLIA to the vehicle solution.
`
`
`
`
`
`
`
`
`
`3
`
`Reference ID: 3500837
`
`(b) (4)
`
`(b) (4)
`
`(b) (4)
`
`(b) (4)
`
`(b) (4)
`
`1 Page of Draft Labeling has been Withheld in Full
`as b4 (CCI/TS) immediately following this page
`
`

`

`(b) (4)
`
`4 Recommendations Regarding the Applicant’s Proposed
`LabeHng
`
`The key biostatistical recommendations regarding the applicant’s proposed labeling are:
`1. Present data from the two studies
`(m4)
`
`because efficacy is established in the individual studies.
`2. Do not include the proposed
`
`M“)
`
`mmtogether in one table, selecting the
`3. Present
`clinically relevant and statistically supp01ted secondary endpoints.
`
`The following is this reviewer’s recommended wording for the Clinical Studies section of
`labeling. Note that the final wording is not final, and may change.
`
`14 CLINICAL STUDIES
`
`The safety and efficacy of once daily use of JUBLIA for the treatment of onychomycosis
`
`of the toenail were assessed in two 52-week prospective. multi—center. randomized.
`
`double-blind clinical tn'als in patients 18 years and older (18 to 70 years of age) with 20%
`
`to 50% clinical involvement of the area of the target toenail The trials compared 48-
`
`weeks of treatment with JUBLLA to the vehicle solution. The Complete Cure rate was
`
`assessed at Week 52 (4-weeks after completion of therapy). Complete cure is defined as
`
`0% involvement of the target nail (no clinical evidence of onychomycosis of the target
`
`toenail) in addition to Mycologic Clu'e. defined as both negative fungal whine and
`
`negative KOH. Table 2 lists the efficacy results for trials 1 and 2.
`
`Reference ID: 3500837
`
`’Jt
`
`

`

`Table 2:
`
`Efficacy Endpoints
`
`Trial 1
`
`Trial 2
`
`Complete
`Curea
`
`Complete or
`Almost
`Complete Cureb
`
`Mycologic Curec
`
`JUBLIA
`
`N = 656
`
`117
`
`17.8%
`
`173
`
`26.4%
`
`362
`
`55.2%
`
`Vehicle
`
`N = 214
`
`7
`
`3.3%
`
`17
`
`7.9%
`
`36
`
`16.8%
`
`JUBLIA
`
`N = 580
`
`88
`
`15.2%
`
`136
`
`23.4%
`
`310
`
`53.4%
`
`Vehicle
`
`N = 201
`
`11
`
`5.5%
`
`15
`
`7.4%
`
`34
`
`16.9%
`
`a. Complete cure is defined as 0% clinical involvement of the target toenail plus negative
`KOH and negative culture.
`
`b. Complete or almost complete is defined as ≤5% affected target toenail area involved
`and negative KOH and culture.
`
`c. Mycologic cure is defined as negative KOH and negative culture.
`
`Signatures/Distribution List
`
`Primary Statistical Reviewer: Kathleen Fritsch, Ph.D.
`Date: 3/5/2014
`
`Statistical Team Leader: Mohamed Alosh, Ph.D.
`
`cc:
`DDDP/Kettl
`DDDP/Chiang
`DDDP/Dixon
`OBIO/Patrician
`DBIII/Wilson
`DBIII/Alosh
`DBIII/Fritsch
`
`Reference ID: 3500837
`
`6
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`KATHLEEN S FRITSCH
`05/05/2014
`
`MOHAMED A ALOSH
`05/05/2014
`
`Reference ID: 3500837
`
`

`

`
`
`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:
`Dates:
`
`Review Priority:
`
`
`
`203567 / 000
`TRADENAME (efinaconazole ) solution 10%
`Onychomycosis
`Dow
`Submitted: 7/26/2012
`PDUFA:
`5/26/2013
`
`Standard review
`
`Biometrics Division:
`Division of Biometrics III
`Statistics Reviewer:
`Kathleen Fritsch, Ph.D.
`Concurring Reviewer: Mohamed Alosh, Ph.D.
`
`
`Medical Division:
`Division of Dermatology and Dental Products
`Clinical Team:
`Gary Chiang, M.D. / David Kettl, M.D.
`Project Manager:
`Strother Dixon
`
`
`Keywords: Onychomycosis, secondary endpoints, SAP
`
`
`Reference ID: 3271427
`
`

`

`
`
`Table of Contents
`
`1
`
`2
`
`EXECUTIVE SUMMARY .............................................................................3
`
`INTRODUCTION..........................................................................................4
`
`2.1 Overview ............................................................................................................ 4
`2.1.1
`Clinical Studies ........................................................................................... 4
`2.1.2
`Regulatory History...................................................................................... 5
`
`2.2 Data Sources...................................................................................................... 6
`
`3
`
`STATISTICAL EVALUATION......................................................................6
`
`3.1 Data and Analysis Quality ............................................................................... 6
`
`3.2 Evaluation of Efficacy....................................................................................... 6
`3.2.1
`Study Design and Statistical Analysis ........................................................ 6
`3.2.2
`Subject Disposition..................................................................................... 9
`3.2.3
`Baseline Characteristics............................................................................ 10
`3.2.4
`Primary Efficacy Endpoint ....................................................................... 12
`3.2.5 Missing Data Handling ............................................................................. 13
`3.2.6
`Secondary Efficacy Endpoints.................................................................. 15
`3.2.7
`Efficacy over Time ................................................................................... 16
`3.2.8
`Efficacy by Center .................................................................................... 20
`3.2.9
`Phase 2 Dose-Ranging Study.................................................................... 21
`
`3.3 Evaluation of Safety........................................................................................ 22
`3.3.1
`Extent of Exposure.................................................................................... 22
`3.3.2 Adverse Events ......................................................................................... 23
`
`4
`
`FINDINGS IN SPECIAL/SUBGROUP POPULATIONS.............................25
`
`4.1 Gender, Race, Age, and Geographic Region ................................................ 25
`
`4.2 Other Special/Subgroup Populations............................................................ 27
`
`5
`
`SUMMARY AND CONCLUSIONS.............................................................27
`
`5.1
`
`Statistical Issues and Collective Evidence..................................................... 27
`
`5.2 Conclusions and Recommendations.............................................................. 28
`
`SIGNATURES/DISTRIBUTION LIST .................................................................28
`
`
`
`
`Reference ID: 3271427
`
`
`
`2
`
`

`

`1 Executive Summary
`Efinaconazole solution 10% was superior to vehicle in the treatment of onychomycosis in
`two studies. Studies P3-01 and P3-02 enrolled subjects age 18 to 65 with a clinical
`diagnosis of onychomycosis and positive mycology. Subjects applied treatment once
`daily for 48 weeks. The primary efficacy endpoint was complete cure at Week 52 (0%
`clinical involvement of target toenail plus negative KOH and negative culture). The
`secondary efficacy endpoints specified in the protocol were: (1) clinical efficacy rate at
`Week 52 (<10% affected target nail area), (2) mycological cure rate at Week 52 (negative
`KOH and culture), and (3) unaffected new nail growth at Week 52 (change from baseline
`in healthy target nail measurement). Secondary endpoints were analyzed in sequential
`order. The primary and secondary efficacy endpoints were all statistically significant and
`the results are presented in Table 1.
`Table 1 – Primary and Secondary Efficacy Endpoints at Week 52 (SAP 1)
`
`Study P3-01
`Study P3-02
`
`p-value Efinacon.
`Efinacon.
`Vehicle
`Vehicle
`N = 580
`N = 656
`N = 214
`N = 201
`117 (17.8%) 7 (3.3%) <0.001 88 (15.2%) 11 (5.5%) <0.001
`234 (36%)
`25 (12%) <0.001
`180 (31%)
`24 (12%)
`<0.001
`362 (55%)
`36 (17%) <0.001
`310 (53%)
`34 (17%)
`<0.001
`5.0 (0.2)
`1.6 (0.4)
`<0.001
`3.8 (0.2)
`0.9 (0.4)
`<0.001
`
`Complete Cure
`Clinical Efficacy
`Mycologic Cure
`Unaffected new
`growth (mm)
`
`The applicant created two versions of the Statistical Analysis Plan (SAP). The first
`version of the SAP was signed off about a week after the last subject completed Study
`P3-01 and the proposed analyses were consistent with the endpoints and analyses
`specified in the protocol. However, the applicant then revised the SAP about 5 weeks
`later. The second version of the SAP redefined the sets of secondary and supportive
`endpoints, and the order in which they were to be analyzed. The primary endpoint
`remained the same in both versions of the SAP, and thus the primary conclusions of the
`study are not affected by the changes to the SAP. This review will focus on the endpoints
`pre-specified in the protocol (and the first version of the SAP), rather than those specified
`only in the second version of the SAP. The secondary endpoints specified in the second
`version of the SAP were only proposed after the studies were completed. Although the
`applicant maintains that the studies were still blinded at that time the second SAP was
`written, changing endpoints after the studies are completed raises the concern that the
`Type I error rate could be inflated. Note that because all of the proposed secondary
`endpoints from either version of the SAP had p-values <0.001, the analyses from the
`second version of the SAP would lead to the same conclusions of efficacy as those from
`the original protocol/first version of the SAP.
`
`p-value
`
`
`
`Reference ID: 3271427
`
`
`
`3
`
`

`

`2
`
`Introduction
`
`2.1 Overview
`
`Inclusion criteria
`
`2.1.1 Clinical Studies
`Efinaconazole solution 10% is a new molecular entity antifungal intended for the
`treatment of onychomycosis. This product was submitted as a 505(b)(1) application.
`Efinaconazole solution was evaluated in one Phase 2 and two Phase 3 studies. The Phase
`2 study evaluated three treatment regimens (10% solution, 10% solution with occlusion,
`and 5% solution) and vehicle over a 36-week treatment period. The 10% solution
`(without occlusion) regimen was selected for Phase 3 development. The first Phase 3
`study (P3-01) enrolled 870 subjects (656 efinaconazole/214 vehicle) and the second
`Phase 3 study (P3-02) enrolled 785 subjects (781 in the ITT: 580 efinaconazole/201
`vehicle). Four subjects in Study P3-02 were randomized in error (3 efinaconazole and 1
`vehicle), did not receive study medication, and were not included in the ITT population.
`Both studies enrolled subjects age 18 and older with 20-50% involvement of the target
`toenail. Treatment was applied once daily at bedtime to all affected toenails for 48
`weeks. An overview of the studies is presented in Table 2 and Table 3. This review will
`focus primarily on the two Phase 3 studies.
`Table 2 – Clinical Studies Overview – Phase 3 Studies
`Study Numbers
`DPSI-IDP-108-P3-01 and DPSI-IDP-108-P3-02
`Study Design
`Randomized, double-blind, vehicle-controlled
`Age 18 - 70, clinical diagnosis of onychomycosis, 20-50%
`involvement of target nail without dermatophytomas or lunula
`involvement, uninfected length ≥ 3mm, thickness ≤ 3mm, positive
`KOH, and positive culture (dermatophyte or mixed
`dermatophyte/Candida)
`Once daily at bedtime to all affected nails for 48 weeks. Solution
`Treatment
`applied to nail folds, nail bed, hyponychium, and undersurface of the
`regimen
`nail plate.
`Primary endpoint Complete cure at Week 52 (0% clinical involvement of the target
`nail, negative KOH, and negative culture)
` P3-01 P3-02
`Efinaconazole, 10% 656 583*
`Vehicle 214 202*
`P3-01: US – 510 subjects (34 centers), Canada – 117 subjects (7
`centers), Japan – 243 subjects (33 centers)
`P3-02: US – 649 subjects (36 centers), Canada – 132 subjects (8
`centers)
`*Four subjects in Study P3-02 were randomized in error (3 efinaconazole and 1 vehicle) and did not receive
`medication and were not included in the ITT population.
`
`
`
`
`
`Treatment arms
`and sample size
`
`Study location
`
`
`
`Reference ID: 3271427
`
`
`
`4
`
`

`

`Inclusion criteria
`
`Treatment
`regimen
`
`Table 3 – Clinical Studies Overview – Phase 2 Study
`Study Number
`DPSI-IDP-108-P2-01
`Study Design
`Randomized, dose-ranging, vehicle-controlled
`Age 18 - 65, clinical diagnosis of onychomycosis, 20-50%
`involvement of target nail without dermatophytomas or lunula
`involvement, uninfected length ≥ 3mm, thickness ≤ 3mm, positive
`KOH, and positive culture (dermatophyte or mixed
`dermatophyte/Candida)
`Once daily at bedtime to all affected nails for 36 weeks. The 10%
`solution was applied with or without overnight semi-occlusion
`Solution applied to nail folds, nail bed, hyponychium, and
`undersurface of the nail plate.
`Primary endpoint Various visual and mycological assessments
`Efinaconazole, 10% 39
`Efinaconazole, 10% (with semi-occlusion) 36
`Efinaconazole, 5% 38
`Vehicle 22
`Mexico – 135 subjects (11 centers)
`
`Treatment arms
`and Sample Size
`
`Study location
`
`
`2.1.2 Regulatory History
`The IND for efinaconazole was opened in 2007 with a cumulative irritation safety study.
`The Phase 3 clinical studies plan was discussed at an End-of-Phase 2 meeting on
`8/4/2009. The protocols were amended three times. The protocols were not submitted as
`Special Protocol Assessments. The dates that the versions of the protocols were signed
`and submitted to the Agency are listed below.
`• Original – protocol date 9/14/2009; submitted 11/6/2009
`• Amendment 1 – amendment date 11/23/2009; submitted 12/28/2009
`• Amendment 2 – amendment date 2/22/2010; submitted 3/3/2010
`• Amendment 3 – amendment date 5/3/2010; submitted 5/19/2010
`
`
`Subjects were first enrolled under Amendment 1 of the Phase 3 protocols. The first
`subject was screened on 12/3/2009 and the last subject visit was 10/14/2011. The Phase
`3 protocols (Amendment 1) were reviewed by the Agency and an Advice Letter was sent
`to the sponsor on 4/14/2010. The Advice Letter contained two comments on the efficacy
`assessment: advising the sponsor to clarify how the investigator calculates percent nail
`involvement, and noting that the proposed supportive efficacy endpoints would have
`limited regulatory utility. The primary and secondary endpoints and proposed statistical
`analysis plan were the same across all four versions of the protocol. None of the
`amendments modified the efficacy evaluations or analyses (with the exception that
`Amendment 1 added a quality of life questionnaire). The amendments made changes to
`exclusion criteria, clarified clinical procedures, and added ECG assessments.
`
`The sponsor also submitted two versions of the Statistical Analysis Plan (SAP). Version
`2 of the SAP was actually submitted to the Agency first (SAP date 11/9/2011 with
`
`
`
`Reference ID: 3271427
`
`
`
`5
`
`

`

`submission date 11/14/2011). Although the definition of the primary endpoint was
`identical to that in the protocol, the set of secondary endpoints in Version 2 of the SAP
`differed from those defined in the protocol (one secondary endpoint was reclassified as
`supportive, one new secondary endpoint was added, and the analysis order was changed).
`The Agency sent an Advice Letter on 2/27/2012 noting that changing the planned
`analysis when the studies are nearly completed raises concerns about unblinding, and
`could affect the Type I error and the interpretation of the results. Subsequently, on
`3/23/2012, the sponsor submitted the original version of the SAP (dated 9/29/2011),
`along with a proposal to include analyses from both versions of the SAP in the final
`clinical study reports.
`
`The sponsor requested a Pre-NDA meeting, but canceled the meeting after receiving the
`pre-meeting communication and determining that no face-to-face discussion was required
`(final minutes dated 5/14/2012).
`
`Data Sources
`2.2
`This reviewer evaluated the applicant’s clinical study reports, datasets, clinical
`summaries, and proposed labeling. This submission was submitted in eCTD format and
`was entirely electronic. Both SDTM and analysis datasets were submitted. The analysis
`datasets used in this review are archived at \\cdsesub1\EVSPROD\NDA203567\0000\m5\
`datasets.
`3 Statistical Evaluation
`
`Data and Analysis Quality
`3.1
`The databases for the studies required minimal data management prior to performing
`analyses and no requests for additional datasets were made to the applicant.
`
`
`3.2
`
`Evaluation of Efficacy
`
`3.2.1 Study Design and Statistical Analysis
`Studies P3-01 and P3-02 were identically-designed, randomized, double-blind, vehicle-
`controlled studies of the efficacy and safety of efinaconazole solution 10% in the
`treatment of onychomycosis. The studies enrolled subjects aged 18 – 70 with a clinical
`diagnosis of onychomycosis, including 20-50% involvement of target nail without
`dermatophytomas or lunula involvement, uninfected length ≥ 3mm, thickness ≤ 3mm,
`positive KOH, and a positive culture (dermatophyte or mixed dermatophyte/Candida).
`Subjects were randomized in a 3:1 ratio to efinaconazole or vehicle. Treatment was
`applied once daily at bedtime to all affected toenails for 48 weeks. Subjects were
`evaluated at screening, baseline, and then every four weeks through Week 52.
`
`Efficacy assessments included percent involvement of the target toenail, length of the
`unaffected part of the target toenail, KOH examination, fungal culturing, and assessment
`of presence/absence of onychomycosis in non-target toenails. These assessments were
`conducted every 12 weeks and end of study (screening, baseline, and Weeks 12, 24, 36,
`
`
`
`Reference ID: 3271427
`
`
`
`6
`
`

`

`48, and 52). In addition, an onychomycosis quality of life questionnaire was
`administered to native-English speaking subjects at baseline, Week 24, and Week 52.
`Localized skin reactions were recorded at each visit. Burning, itching, and vesiculation
`were recorded as present or absent. Redness and swelling were recorded on a 4-point
`scale (none, mild, moderate, severe).
`
`The primary efficacy endpoint was complete cure (0% clinical involvement of target
`toenail plus negative KOH and negative culture) at Week 52 (4 weeks post-treatment).
`Complete cure was analyzed using a Cochran—Mantel-Haenszel test stratified on analysis
`center. The applicant wrote two statistical analysis plans (SAPs) for the studies. The
`primary efficacy endpoint and analysis were identical in both versions and consistent
`with the protocol. The two SAPs differed in the definition and ordering of the secondary
`and supportive endpoints. According to the applicant, both SAPs were approved prior to
`database lock. The two versions of the SAP are dated 9/29/2011 (Version 1) and
`11/9/2011 (Version 2). The database locks occurred on 11/21/2011 for Study P3-01 and
`12/6/2011 for Study P3-02.
`
`The first version of the SAP listed the secondary and supportive endpoints as they were
`presented in the protocol. This list included three secondary endpoints as follows:
`0 Clinical efficacy rate at Week 52 (<10% affected target nail area)
`
`0 Mycological cure rate at Week 52 (negative KOH and culture)
`0 Unaffected new nail growth at Week 52 (change from baseline in healthy target
`nail measurement)
`
`The second version of the SAP and removed one secondary endpoint, added one
`secondary endpoint, and re-ordered the list as follows:
`
`0 Complete or almost complete cure rate at Week 52 (55% affected target nail area
`and negative KOH and culture)
`
`0 Unaffected new nail growth at Week 52 (change from baseline in healthy target
`nail measurement)
`
`0 Mycological cure rate at Week 52 (negative KOH and culture)
`
`Each SAP proposed to test the hypotheses for the secondary endpoints in sequential order
`(in the order listed). The applicant states that the motivation for revising the SAP was
`primarily to provide a “more statistically robust and clinically relevant evaluation for the
`secondary endpoint.” (page 55 of the clinical study report for Study P3-01 and page 52 of
`the clinical study report for Study P3-02) The Agency advised the sponsor in an Advice
`Letter dated 2/27/2012 that changing the secondary endpoints and their ordering afier the
`studies were complete or nearly complete could impact the Type I error and would make
`the results of the study difficult to interpret. Subsequently, the sponsor proposed
`including analyses from both versions of the SAP in the clinical study report. M“)
`
`Reference ID: 3271427
`
`

`

`Response rate endpoints were analyzed using a Cochran-Mantel-Haenszel test stratified
`on analysis center. Unaffected new toenail growth was analyzed using ANOVA with
`factors for treatment and analysis center.
`
`The two SAPs also differed in the list of supportive efficacy analyses. All supportive
`efficacy endpoints were to be summarized by descriptive statistics. The first SAP
`included the following list of supportive endpoints:
`• Change from baseline in number of affected toenails
`• Target nail growth from baseline
`• Change from baseline to Week 24 and Week 52 in onychomycosis quality of life
`(OnyCOE-t)
`
`
`The second SAP added three new supportive endpoints to those listed in the first SAP so
`that the list included:
`• Clear nail (0% affected nail)
`• Almost clear nail (≤5% affected nail)
`• Clinical efficacy (≤10% affected nail)
`• Change from baseline in number of affected toenails
`• Target nail growth from baseline
`• Change from baseline to Week 24 and Week 52 in onychomycosis quality of life
`(OnyCOE-t)
`
`
`Note that the definition of clinical efficacy (≤10% affected nail) in the second SAP is
`slightly different from the definition of clinical efficacy (<10% affected nail) in the first
`SAP. Because many investigators reported affected nail area to the nearest 5%, whether
`or not subjects are included who have 10% affected nail affects the response rates.
`
`Small centers were combined into analysis centers for the CMH and ANOVA analyses.
`Centers with fewer than 9 efinaconazole and 3 vehicle subjects were pooled into analysis
`centers. Among the centers with fewer than 9 efinaconazole or 3 vehicle subjects, the
`smallest center was pooled with the largest, etc. until all analysis centers met the
`minimum size. Consistency of treatment response across analysis centers for the primary
`endpoint was assessed with the Breslow- Day test. If the Breslow-Day test was
`significant at 0.10, sensitivity analyses were conducted to assess the impact of extreme
`centers.
`
`The ITT population was defined as all subjects randomized and dispensed study drug.
`The per protocol population included subjects who
`• met all inclusion/exclusion criteria unless a waiver was granted prior to
`randomization
`• did not take any interfering concomitant medications
`• completed the Week 52 visit
`• missed no more than 20% of the total number of expected doses during the
`treatment period
`• did not miss more than 14 cumulative doses in the 28 days leading up to the date
`of the last dose
`
`
`
`Reference ID: 3271427
`
`
`
`8
`
`

`

`• did not miss 28 or more consecutive doses during the treatment period
`• were not out of the visit window (± 5 days) for the Week 52 visit
`
`
`The primary method of handling missing data for the primary efficacy analysis was last
`observation carried forward (LOCF). As a sensitivity analysis, subjects with missing
`Week 52 complete cure assessments were imputed as failures. A second sensitivity
`analysis imputes subjects with missing values as successes.
`
`3.2.2 Subject Disposition
`Study P3-01 randomized 656 subjects to efinaconazole and 214 to vehicle, and all
`subjects were included in the ITT population. Study P3-02 randomized 583 subjects to
`efinaconazole and 202 to vehicle, however, 3 efinaconazole subjects and 1 vehicle
`subject were not included in the ITT population, so the ITT population includes 580
`efinaconazole and 201 vehicle subjects. All four subjects were noted as having been
`randomized in error and none of the four were dispensed medication. The four subjects
`were enrolled at three different centers and all of them failed at least one of the
`inclusion/exclusion criteria which defined the extent or clinical characteristics of the
`onychomycosis.
`
`Similar proportions of efinaconazole and vehicle subjects discontinued the study early in
`Study P3-01 (around 12% per arm), while a slightly higher proportion of vehicle subjects
`discontinued early in Study P3-02 (15% for efinaconazole and 21% for vehicle). The
`disposition and reasons for discontinuation are presented in Table 4 and Table 5. The
`most common reasons for discontinuation were subject request and loss-to-follow-up,
`and the rates for these categories were slightly higher on the vehicle arm than the
`efinaconazole arm in each study. However, discontinuation due to adverse events was
`higher on the efinaconazole arm than the vehicle arm in each study. One center in Study
`P3-01 (Site 121) closed down before the study was completed and 8 subjects were
`discontinued due to the site closing.
`
`Table 4 – Disposition of Subjects (Study P3-01)
`Efinaconazole
`
`Subjects Randomized
`656
`Discontinued study
`81 (12.3%)
`Reasons for discontinuation
`Adverse event
` Application site
` Other
`Subject request
` Moved or couldn’t make visits
` Withdrew consent
` Lack of Efficacy
` Adverse event
`--Table continues on next page.--
`
`Vehicle
`214
`27 (12.6%)
`
` 1 (0.5%)
`0 (0.0%)
`1 (0.5%)
`12 (5.6%)
`5 (2.3%)
`4 (1.9%)
`3 (1.4%)
`0 (0.0%)
`
`
`
`21 (3.2%)
`19 (2.9%)
` 2 (0.3%)
`31 (4.7%)
`16 (2.4%)
`10 (1.5%)
` 4 (0.6%)
` 1 (0.2%)
`
`
`
`Reference ID: 3271427
`
`
`
`9
`
`

`

`Table 4 continued - Disposition of Subjects (Study P3-01)
`
`Efinaconazole
`Vehicle
`Subjects Randomized
`656
`214
`Lost to follow-up
`20 (3.0%)
`11 (5.1%)
`Protocol violation
` 0 (0.0%)
` 1 (0.5%)
`Other
` 9 (1.4%)
` 2 (0.9%)
` Clinic Closing
`6 (0.9%)
`2 (0.9%)
` Lack of Efficacy
`1 (0.2%)
`0 (0.0%)
` Moved
`1 (0.2%)
`0 (0.0%)
` Can’t asses nail growth
`1 (0.2%)
`0 (0.0%)
`Note: The bolded terms are the categories from the CRFs. The Adverse event, Subject request, and Other
`classifications required the investigator to specify additional details.