`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`203214Orig1s000
`
`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 T A T I S T I C A L R E V I E W A N D E VA L U A T I O N
`CLINICAL STUDIES (ADDENDUM)
`
`NDA/BLA #:
`Drug Name:
`Indication(s):
`Applicant:
`Date(s):
`
`Review Priority:
`
`Biometrics Division:
`Statistical Reviewer:
`
`Concurring Reviewers:
`
`NDA 203,214
`Xeljanz (tofacitinib) 5 mg and 10 mg tablets
`Treatment of Rheumatoid Arthritis
`Pfizer Inc.
`Submitted: October 21, 2011; PDUFA: August 21, 2012
`Extension: November 21, 2012
`Standard
`
`Division of Biometrics II
`Yongman Kim, Ph.D.
`
`Joan Buenconsejo, Ph.D.
`Thomas Permutt, Ph.D.
`
`Division of Pulmonary, Allergy, and Rheumatology Products
`Nikolay Nikolov, M.D.
`Sarah Yim, M.D.
`Sally Seymour, M.D.
`Philantha Bowen
`Christine Chung
`
`
`Keywords: NDA review, clinical studies, safety
`
`
`
`Medical Division:
`Clinical Team:
`
`Project Manager:
`
`Reference ID: 3201228
`
`

`

`This is an addendum for NDA 203214 to reflect changes in Tables 12, 13, 20 and 22 from the
`safety statistical review dated October 1, 2012. I also would like to clarify that the comparison
`between tofacitinib and adalimumab was based on 12 month data, and not during the first 3
`months as stated in the Executive Summary. Specifically, we wrote in the original review
`
`“Based on comprehensive analyses of the major safety events of interest, there is a suggestion
`that tofacitinib might be associated with a higher risk of serious infections, and herpes infections
`compared to placebo, as well as to adalimumab, during the first 3 months of use.”
`
`The revised text is
`
`“Based on comprehensive analyses of the major safety events of interest, there is a suggestion
`that tofacitinib might be associated with a higher risk of serious infections, and herpes infections
`compared to placebo during the first 3 months of use, as well as to adalimumab, during the first
`12 months of use.”
`
`In section 3.3.1 of the review, zero death was reported in the combined tofacitinib 5 mg and 10
`mg group (Table 12). Our clinical team noted that there is one reported death in Study 1064 in
`the tofacitinib group. After checking the applicant’s report, we concluded that one patient
`receiving tofacitinib died in 29 days after the first 12 months of exposure and no deaths were
`reported in the adalimumab group. The revised tables with corrected results are presented in
`Tables 12 and 13.
`
`In Table 20 in the review, the reported number of patients treated with tofacitinib 10 mg in trials
`1045, 1046 and 1064 were incorrect. These counts have been updated to 297, 391 and 245
`respectively. Also in Table 20, number of patients treated with tofacitinib 10 mg who
`experienced a serious infection in trials 1046 and 1064 were incorrectly reported as 8 and 8
`respectively. The counts have been updated to reflect the correct counts: 7 and 9.
`
`
`Table 1. Events of interest among subjects randomized to tofacitinib 5 mg, 10 mg, and adalimumab
`in trials A3921035 and 1064
`
`Trial
`
`serious
`infections
`
`herpes
`zoster
`
`N
`
`Death MACE
`
`0
`0
`
`0
`1
`
`0
`3
`
`0
`2
`
`
`53
`A3921035¹
`A3921064² 204
`
`A3921035¹ 110
`A3921064² 405
`¹Months 0-3 only
`²Months 0-12
`
`
`
`
`Reference ID: 3201228
`
`Tofacitinib 5 mg, 10 mg and adalimumab
`lymphoma
`solid
`malignancy
`opportunistic
`tuberculosis
`organ
`excluding
`infections
`tumors
`NMSC
`Adalimumab
`0
`1
`1
`0
`1
`1
`tofacitinib 5 mg and 10 mg
`0
`0
`0
`3
`3
`0
`
`0
`0
`
`0
`0
`
`0
`0
`
`0
`2
`
`0
`3
`
`0
`15
`
`0
`5
`
`0
`19
`
`2
`
`

`

`Table 2. Incidence ratio of events of interest, comparing tofacitinib (5 mg and 10 mg) vs.
`adalimumab during months 0 – 12 + 28 days based on randomized treatment
`assignment in trial A3921064
`Incidence ratio
`(95% CI)
`
`
`
`Mantel-Haenszel
`RR (95% CI)
`
`Adalimumab¹
`events/100 py
`
`Tofacitinib²
`events/100 py
`
`Risk Difference³
`(95% CI)
`per 100 py
`0.25 (-0.24, 0.73)
`-1.01 (-2.89, 0.87)
`-
`0.28 (-1.07, 1.63)
`
`ND
`0.39 (0.07, 2.35)
`ND
`1.77 (0.18,16.99)
`
`ND
`0.34 (0.06, 1.99)
`ND
`1.51 (0.16, 14.43)
`
`1.77 (0.18,16.99)
`
`1.51 (0.16, 14.43)
`
`0.28 (-1.07, 1.63)
`
`ND
`ND
`2.59 (0.75, 8.94)
`
`ND
`ND
`2.52 (0.74, 8.60)
`
`-
`0.53 (-0.20, 1.26)
`2.43 (-0.19, 5.06)
`
`2.47 (-0.67, 5.60)
`
`death
`MACE
`lymphoma
`Solid organ
`tumors
`malignancy
`excluding
`NMSC
`opportunistic
`infections
`tuberculosis
`serious
`infections
`herpes
`1.91 (0.73, 5.05)
`1.98 (0.74, 5.31)
`zoster
`ND = not defined. Reference level is adalimumab
`¹A total of 194.5 years of exposure were observed among patients on adalimumab
`²A total of 377.1 years of exposure were observed among patients on tofacitinib
`³Events in Tofacitinib per 100 py – Events in Adalimumab per 100py
`
`
`Table 3. Events of interest during months 0-12 + 28 days among subjects randomized to tofacitinib
`5mg or 10mg + patients who transitioned from placebo to tofacitinib at either month3 or month 6
`(by design or by response).
`
`0
`1.54
`0
`0.51
`
`0.51
`
`0
`0
`1.54
`
`2.57
`
`0.26
`0.53
`0
`0.80
`
`0.80
`
`0
`0.53
`3.98
`
`5.04
`
`
`
`Tofacitinib 5mg and 10mg, Months 0-12 + all patients who transitioned from placebo to tofacitinib
`Trial
`N
`death MACE
`lymphoma
`solid
`malignancy
`opportunistic
`tuberculosis
`organ
`excluding
`infections
`tumors
`NMSC
`
`serious
`infections
`
`herpes
`zoster
`
`
`89
`A3921025
`191
`A3921032
`74
`A3921035
`394
`A3921044
`298
`A3921045
`388
`A3921046
`255
`A3921064
`** Total ** 1689
`
`A3921025
`A3921032
`A3921035
`A3921044
`A3921045
`
`74
`187
`61
`384
`297
`
`
`
`0
`0
`0
`3
`0
`1
`1
`5
`
`0
`1
`0
`1
`1
`
`0
`3
`0
`0
`0
`2
`1
`6
`
`0
`0
`0
`2
`2
`
`0
`0
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`1
`0
`
`tofacitinib 5mg
`1
`0
`0
`2
`0
`0
`2
`5
`tofacitinib 10mg
`0
`0
`0
`3
`1
`
`0
`0
`0
`3
`0
`1
`0
`4
`
`0
`0
`0
`3
`0
`
`1
`0
`0
`2
`0
`0
`2
`5
`
`0
`0
`0
`2
`1
`
`0
`0
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`1
`1
`
`1
`4
`0
`16
`2
`3
`8
`34
`
`1
`3
`0
`9
`4
`
`1
`1
`1
`20
`2
`15
`7
`47
`
`2
`2
`0
`21
`5
`
`3
`
`Reference ID: 3201228
`
`

`

`A3921046
`
`A3921 064
`
`391
`
`245
`
`"" Total "‘
`
`1639
`
`2
`
`0
`
`5
`
`1
`
`1
`
`6
`
`0
`
`0
`
`1
`
`2
`
`1
`
`6
`
`2
`
`1
`
`7
`
`1
`
`0
`
`4
`
`2
`
`2
`
`6
`
`7
`
`9
`
`33
`
`7
`
`1 3
`
`50
`
`In section 4 of the review, results from the analyses of selected safety endpoints were presented
`in Table 22. The analyses were based on the as treated population defined as all randomized
`patients plus patients who transitioned from placebo to tofacitinib by study design. In
`consultation with the clinical review team, we decided to include, in addition to the definition
`above, all placebo patients who advanced to tofacitinib by response in the analyses. The change
`is on the footnote. The numbers are the same except in malignancy where the upper bound is
`now 0.7% instead of 0.8%.
`
`Table 4: Inte ated Anal ses of Efficac and Safe
`
`ted
`
`Tofacitinib 5
`
`Tofacitinib 10
`
`Tofacitinib
`
`Tofacitinib 10
`
`mg versus
`placebo
`
`mg versus
`placebo
`
`versus
`placebo
`
`mg versus 5
`mg
`
`Efficacy Endpoints '1‘
`
`ACR20 at Week 12
`
`28%
`
`34%
`
`31%
`
`6%
`
`(difference in proportions)
`
`(24%, 32%)
`
`(30%, 38%)
`
`(27%, 35%)
`
`(3%, 10%)
`
`HAQ-DI at Week 12
`(mean difference)
`
`-0.23
`(-0.28, -0.18)
`
`-0.32
`(-0.36, -0.27)
`
`-0.27
`(-0.32_, -0.23)
`
`-0.08
`(-0.13, -0.04)
`
`Selected Safety Endpoints :1:
`(difference in rates)
`
`Serious Infection (0 — 3
`months)
`
`Herpes Infection (0 — 3
`months)
`
`Tuberculosis Infection (0 —
`12 months)
`
`Malignancy (0 — 12 months)
`
`1.3%
`(-0.2%, 2.8%)
`
`1.9%
`(-0.3% 4.2%)
`
`0.5%
`(0.1%, 0.9%)
`
`0.2%
`(-0.4%, 0.7%)
`
`1‘ includes all randomized patients. Nonresponder imputation was applied to missing binary outcome data and baseline
`observation was carried forward to missing HAQ—DI score at week 12.
`1 includes all randomized patients plus patients who transitioned from placebo to tofacitinib by study design and by response
`
`Reference ID: 3201228
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`YONGMAN KIM
`10/09/2012
`
`JOAN K BUENCONSEJO
`10/09/2012
`I concur with the review amendment.
`
`Reference ID: 3201228
`
`

`

`
`
`
`
`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 T A T I S T I C A L R E V I E W A N D E VA L U A T I O N
`CLINICAL STUDIES
`
`NDA/BLA #:
`Drug Name:
`Indication(s):
`Applicant:
`Date(s):
`
`Review Priority:
`
`Biometrics Division:
`Statistical Reviewer:
`
`NDA 203,214
`Xeljanz (tofacitinib) 5 mg and 10 mg tablets
`Treatment of Rheumatoid Arthritis
`Pfizer Inc.
`Submitted: October 21, 2011; PDUFA: August 21, 2012
`Extension: November 21, 2012
`Standard
`
`Division of Biometrics II
`Joan Buenconsejo, Ph.D.
`Yongman Kim, Ph.D.
`Eugenio Andraca-Carrera, Ph.D.
`Concurring Reviewers: Mat Soukup, Ph.D.
`Thomas Permutt, Ph.D.
`
`Division of Pulmonary, Allergy, and Rheumatology Products
`Nikolay Nikolov, M.D.
`Sarah Yim, M.D.
`Sally Seymour, M.D.
`Philantha Bowen
`Christine Chung
`
`
`Keywords: NDA review, clinical studies, safety
`
`
`Medical Division:
`Clinical Team:
`
`Project Manager:
`
`Reference ID: 3197461
`
`

`

`
`Table of Contents
`1 EXECUTIVE SUMMARY.................................................................................................... 4
`
`2
`
`INTRODUCTION.................................................................................................................. 5
`
`3 STATISTICAL EVALUATION........................................................................................... 8
`3.1 REVIEW OF THE APPLICANT’S ORIGINAL SAFETY EVALUATION......................................... 8
`3.2 REVIEW OF THE APPLICANT’S NEW SAFETY EVALUATION AND LABEL............................ 11
`3.3 REVIEW OF THE SAFETY DATA ......................................................................................... 14
`3.3.1 Tofacitinib vs. placebo.............................................................................................. 14
`3.3.2 Tofacitinib vs. adalimumab....................................................................................... 18
`3.3.3 Tofacitinib 5 mg vs. tofacitinib 10 mg ...................................................................... 19
`3.3.3.1 Tofacitinib 5 mg vs. 10 mg during months 0-12 based on randomized treatment
`assignment.............................................................................................................................. 20
`3.3.3.2 Tofacitinib 5 mg vs. 10 mg during months 0 – 12 based on randomized treatment
`assignment + subjects who switched treatment from placebo to tofacitinib ......................... 21
`3.3.4
`Summary of results.................................................................................................... 25
`4 SUMMARY AND CONCLUSIONS................................................................................... 26
`
`
`
`
`
`Reference ID: 3197461
`
`2
`
`

`

`
`LIST OF TABLES
`
`Table 1: Applicant’s timelines for submission of FDA safety request .........................................................6
`Table 2: List of all studies included in analysis............................................................................................7
`Table 3: Safety Analysis Population in Phase 3 Rheumatoid Arthritis Studies, as randomized and treated
`(Applicant’s) .................................................................................................................................................9
`Table 4: Selected Common Adverse Events...............................................................................................13
`Table 5: Applicant’s Analyses of Major Events of Interest........................................................................13
`Table 6. Events of interest during months 0 – 3 among subjects randomized to placebo ..........................15
`Table 7. Events of interest during months 0 – 3 among subjects randomized to tofacitinib 5 mg .............15
`Table 8. Events of interest during months 0 – 3 among subjects randomized to tofacitinib 10 mg ...........16
`Table 9. Incidence ratio of events of interest, comparing tofacitinib vs. placebo during ...........................16
`Table 10. Events of interest during months 3 – 6 among subjects who switched from placebo to tofacitinib
`by design at month3 in trials A3921032 and 1045 .....................................................................................17
`Table 11. Incidence ratio of events of interest, comparing tofacitinib vs. placebo during .........................17
`Table 12. Events of interest among subjects randomized to tofacitinib 5 mg, 10 mg, and adalimumab in
`trials A3921035 and 1064...........................................................................................................................18
`Table 13. Incidence ratio of events of interest, comparing tofacitinib (5 mg and 10 mg) vs. adalimumab
`during months 0 – 12 + 28 days based on randomized treatment...............................................................19
`Table 14. Events of interest during months 0 – 12 + 28 days among subjects randomized to ...................20
`Table 15. Incidence ratio of events of interest, comparing tofacitinib 10 mg vs. tofacitinib 5 mg during
`months 0 – 12 + 28 days based on randomized treatment assignment .......................................................21
`Table 16. Events of interest during months 3 – 12 + 28 days among subjects who switched from placebo
`to tofacitinib by at month 3.........................................................................................................................21
`Table 17. Events of interest during months 6 – 12 + 28 days among subjects who switched from placebo
`to tofacitinib by at month 6.........................................................................................................................22
`Table 18. Events of interest during months 0 – 12 + 28 days among subjects randomized to tofacitinib 5
`mg or 10 mg + patients who transitioned from placebo to tofacitinib by study design (month 3 for studies
`1032 and 1045, and Month 6 for studies 1044, 1046, and 1064)................................................................23
`Table 19. Incidence ratio of events of interest, comparing tofacitinib 10 mg vs. tofacitinib 5 mg during
`months 0 – 12 + 28 days based on randomized treatment assignment + patients who transitioned from
`placebo to tofacitinib by study design (month 3 for studies 1032 and 1045, and Month 6 for studies 1044,
`1046, and 1064). .........................................................................................................................................23
`Table 20. Events of interest during months 0 – 12 + 28 days among subjects randomized to tofacitinib 5
`mg or 10 mg + patients who transitioned from placebo to tofacitinib at either month3 or month 6 (by
`design or by response). ...............................................................................................................................24
`Table 21. Incidence ratio of events of interest, comparing tofacitinib 10 mg vs. tofacitinib 5 mg during
`months 0 – 12 + 28 days based on randomized treatment assignment + patients who transitioned from
`placebo to tofacitinib at either month3 or month 6 (by design or by response)..........................................25
`Table 22: Integrated Analyses of Efficacy and Safety................................................................................27
`
`
`
`
`
`
`Reference ID: 3197461
`
`3
`
`

`

`1 EXECUTIVE SUMNIARY
`
`Pfizer proposes tofacitinib (CP-690,550) 5 mg and 10 mg orally administered twice a day (BID)
`for the treatment of rheumatoid arthritis (RA). To support this marketing application, the
`applicant submitted data from 21 Phase 1 studies, six Phase 2 studies, and five Phase 3 studies.
`The focus of the statistical review was on the five Phase 3 studies 1032, 1044, 1045, 1046, and
`
`1064. Please refer to the statistical review by Yongman Kim dated on June 23, 2012 for a
`detailed review of the efficacy data. That review concluded that there is substantial evidence of
`efficacy of tofacitinib 5 mg or 10 mg for the treatment of rheumatoid arthritis. This is based on
`consistent findings in the domains of reducing signs and symptoms of RA (as measured by
`ACR20) and improving physical function (as measured by HAQ-DI).
`
`The focus of the safety statistical review was on the controlled portion of the two Phase 2 studies
`1025 and 1035, and the five Phase 3 studies 1032, 1044, 1045, 1046, and 1064. Based on
`
`comprehensive analyses of the major safety events of interest, there is a suggestion that
`tofacitinib might be associated with a higher risk of serious infections, and herpes infections
`compared to placebo, as well as to adalimumab, during the first 3 months of use
`”(0
`
`Therefore, the direction of the difference in risk. if
`any. is not known with much confidence. The analyses conducted in this review indicate no
`evidence of a difference in the risk of death, MACE, lymphoma, solid organ tumors,
`opportunistic infections excluding tuberculosis, serious infections and herpes zoster infections
`between patients treated with tofacitinib 5 mg or 10 mg.
`
`Overall, the clinical benefit of tofacitinib 5 mg and 10 mg twice a day was generally comparable.
`Numerical differences were observed in ACR20 response rates and change in HAQ-DI between
`the two doses; however these were not consistent across the randomized controlled studies and
`any differences were small.
`(hm)
`
`Clinical assessment of the balance of the risk and benefit of the 10 mg dose is needed as
`the differences in the efficacy between tofacitinib 10 mg and 5 mg for symptomatic treatment of
`RA may not justify the potential risk of serious diseases like malignancy and serious infections.
`The risk-benefit profile as it currently stands may favor the tofacitinib 5 mg group.
`
`Reference ID: 31 97461
`
`

`

`2
`
`INTRODUCTION
`
`This document is a joint statistical safety review representing the work of Joan Buenconsejo,
`Yongrnan Kim, and Eugenio Andraca-Carrera. The application under consideration is NDA
`203,214 from Pfizer for the new molecular entity tofacitinib (also known as CP—690,550) for the
`treatment of patients with moderately to severely active rheumatoid arthritis RA)
`
`This review will focus only on the results from the analyses of major safety events of interest
`including death, lymphoma, solid organ tumor, opportunistic infection, tuberculosis, serious
`infection, herpes zoster, and cardiovascular (MACE) events. While tofacitinib therapy was
`shown to be associated with changes in some laboratory values, as well as increase in other
`adverse events by the applicant, we will not be presenting these results. The objectives of this
`review are to highlight some of the issues identified during the review of safety data including
`the steps we have taken, to discuss the outcome from the re-analyses of the safety data, and to
`provide cements on how adverse events should be reported in the label. For additional review
`of the safety data including laboratory values, and other AEs, please refer to the clinical review
`of Dr. Nikolay Nikolov.
`
`The NDA was submitted on October 21, 2011 by Pfizer. Pfizer proposes tofacitinib (CP—
`690,550) 5 mg and 10 mg orally administered twice a day (BID) for the treatment of rheumatoid
`arthritis (RA). To support this marketing application, the applicant submitted data from 21 Phase
`1 studies, six Phase 2 studies, and five Phase 3 studies. The focus of the statistical review was on
`
`the five Phase 3 studies A3921032, A3921044, A3921045, A3921046, and A3921064. Please
`refer to the statistical review by Yongman Kim dated on June 23, 2012 for a detailed review of
`the efficacy data. That review concluded that there is substantial evidence of efficacy of
`tofacitinib 5 mg or 10 mg for the treatment of rherunatoid arthritis. This is based on consistent
`findings in the domains of reducing signs and symptoms of RA (as measured by ACR20) and
`improving physical function (as measured by HAQ-Dl).
`(m4)
`
`The applicant’s safety evaluation was based on the integrated safety data from five key double-
`blind phase 3 studies in RA (A3921032, A3921044, A3921045, A3921046, and A3921064).
`These data were supported by integrated data from five completed phase 2 studies (A3921025,
`A3921039, A3921019, A3921035, and A3921040) and integrated data from two ongoing long—
`term extension O..TE) studies (A3921024 and A3921041), see Table 2.
`
`Dr. Kim, in his review, expressed our concern that the safety summaries the applicant provided
`may not be adequate to make regulatory decision regarding the safety of tofacitinib 5 mg or 10
`mg doses. Specifically, we were concerned with the number of studies included in the main
`integrated safety analyses, on how the results were summarized based on pooled data without
`controlling for differential study designs and patient populations, and the inability to properly
`account for those placebo patients who advanced to tofacitinib. These issues made it difficult for
`us to interpret the findings. The applicant reported the results primarily based on crude rates.
`Given the complexity of the study design, we believed that a model-based approach is more
`
`Reference ID: 31 97461
`
`

`

`appropriate to summarize the safety data. The clinical team was also concerned about the entry
`criteria, the inclusion of the Japanese studies, and the dose adjustments provisions in the long-
`term extension studies (see Section 3.1). Because of the potential limitations of the safety data
`presentation and analyses in the original application, the Division asked the applicant to provide
`alternative approaches to analyze the safety data, particularly those major events of interest. The
`Division sent information requests on May 22, June 4, and June 22 requesting safety datasets
`with selected variables from existing database and additional analyses accounting for differences
`in length of exposure and the cross-over nature of the design. Teleconferences were held
`between the Division and the applicant to clarify some issues regarding the requests. A Type A
`meeting between the Division and the applicant was held on July 10, 2012 to discuss the
`potential for alternative safety analyses to address the Agency’s questions, and the potential
`impact of the submission timing on the review of the NDA, particularly in light of the August 21,
`2012 action date. Since multiple information requests were sent to the applicant, the Division
`clarified at the meeting that the analyses requested from the June 20, 2012 letter supersedes the
`June 4, 2012 request, and that the applicant should submit all requested analyses from the June
`20 letter prior to the PDUFA action date. Furthermore, the Division acknowledged the receipt of
`the requested datasets (from the May 22 and June 4 letters) and informed them that we will be
`conducting additional analyses to the safety data. The applicant provided two timelines for
`submission presented in Table 1.
`
`Table 1: Applicant’s timelines for submission of FDA safety request
`
`August 19, 2012*
`
`* Date corrected during the July 20, 2012 meeting
`
`In response to the Information Request dated June 20, 2012, the applicant submitted the
`following on August 1, 2012 (serial number 34).
`
`
`1. Exposure Estimates and Incidence Rates Tables (As Treated)
`2. Kaplan-Meier Plots and Survival Function Tables (As Randomized)
`3. Treatment-emergent AE Tables (As Randomized)
`4. Laboratory Data As Randomized Tables
`
`
`On August 10, 2012 (serial number 37), they submitted the following:
`
`
`
`
`
`6
`
`Reference ID: 3197461
`
`

`

`1. Treatment-emergent AE Tables (As Treated)
`2. Laboratory Data As Treated Tables
`
`Protocol
`
`
`These two submissions fulfilled the requested analyses from the June 20, 2012 information
`request.
`
`
`Table 2: List of all studies included in analysis
`
`
`Patient Population
`
`Enrolled
`Design
`Randomization
`Duration
`Patients with incomplete response to prior TNF inhibitor
`R,DB, PC
`507
`A3921025
`Moderate-to-severe RA
`MTX-IR,
`Dose-
`1:1:1:1:1:1:1
`Stable background MTX
`ranging
`6 months
`
`A3921032
`
`Moderate-to-severe RA
`TNF-IR,
`Stable background MTX
`
`R, DB, PC
`6 months
`
`399
`2:2:1:1
`
`Patients with incomplete response to MTX or other DMARDs
`A3921035
`Moderate-to-severe RA
`R,DB, AC
`797
`DMARD-IR,
`Dose-
`1:1:1:1:1:1:1
`No background MTX
`ranging
`6 months
`
`A3921044
`
`Moderate-to-severe RA
`MTX-IR,
`Stable background MTX
`
`R, DB, PC
`Two years*
`
`797
`4:4:1:1
`
`A3921046
`
`Moderate-to-severe RA
`DMARD-IR,
`Stable background
`DMARDs#
`
`R, DB, PC
`One-year
`
`792
`4:4:1:1
`
`A3921064
`
`Moderate-to-severe RA
`MTX-IR,
`Stable background MTX
`
`R, DB, AC
`One year
`
`717
`4:4:1:1:4
`
`Moderate-to-severe RA
`DMARD-IR,
`No background to Month
`3
`
`R, DB, PC
`6 months
`
`610
`4:4:1:1
`
`A3921045
`
`
`
`
`
`Reference ID: 3197461
`
`Treatment Arms
`
`CP 1 mg BID (→CP 5 mg BID @ Mo3 if NR)+ MTX
`CP 3 mg BID (→CP 5 mg BID @ Mo3 if NR )+
`MTX
`CP 5 mg BID + MTX
`CP 10 mg BID + MTX
`CP 15 mg BID + MTX
`CP 20 mg QD (→CP 5 mg BID @ Mo3 if NR )+
`MTX
`Placebo (→CP 5 mg BID @ Mo3 if NR)+ MTX
`CP 5 mg BID + MTX
`CP 10 mg BID + MTX
`Placebo (→CP 5 mg BID @ Mo3)+ MTX
`Placebo (→CP 10 mg BID @ Mo3)+ MTX
`
`CP 1 mg BID (→CP 5 mg BID @ Mo3 if NR)
`CP 3 mg BID (→CP 5 mg BID @ Mo3 if NR )
`CP 5 mg BID
`CP 10 mg BID
`CP 15 mg BID
`Adalimumab (→CP 5 mg BID @ Mo3)
`Placebo (→CP 5 mg BID @ Mo3 if NR)
`CP 5 mg BID + MTX
`CP 10 mg BID + MTX
`PBO (→CP 5 mg BID @ Mo 6 or Mo3 if NR)+ MTX
`PBO (→CP 10 mg BID @ Mo 6 or Mo3 if NR)+
`MTX
`CP 5 mg BID + DMARD
`CP 10 mg BID + DMARD
`PBO (→CP 5 mg BID @ Mo 6 or Mo3 if NR) +
`DMARD
`PBO (→CP 10 mg BID @ Mo 6 or Mo3 if NR) +
`DMARD
`CP 5 mg BID + PBO SC+ MTX
`CP 10 mg BID + PBO SC + MTX
`PBO (→CP 5 mg BID @ Mo 6 or Mo3 if NR) +
`PBO SC + MTX
`PBO (→CP 10 mg BID @ Mo 6 or Mo3 if NR) +
`PBO SC + MTX
`PBO + adalimumab + MTX
`CP-690,550 5 mg BID
`CP-690,550 10 mg BID
`PBO → CP 5 mg BID @ Mo 3
`PBO → CP 10 mg BID @ Mo 3
`
`7
`
`

`

` STATISTICAL EVALUATION
`
`
`
`
` 3
`
`
`3.1 Review of the Applicant’s Original Safety Evaluation
`
`The applicant focused on two main safety populations in their report. This includes patients in
`the Phase 3 RA studies (pooled) and patients in the two LTE studies (pooled).
`
`For simplicity, the last four digits will be used to denote the study number. For example, study
`1025 refers to study A3921025.
`
`
`Phase 2 and Phase 3 RA Studies
`
`The Phase 3 studies group includes data from the five Phase 3 RA studies which were pooled
`(1032, 1045, 1046, 1064, and 1044). Four of these studies were designed to administer
`tofacitinib along with a background DMARD (typically MTX), and one study administered
`tofacitinib as monotherapy.
`
`Studies 1032, 1044, 1045, 1046, and 1064 were of similar design, except for the population
`studied, the study duration, and treatment arms. They were all phase 3, randomized, double-
`blind, placebo-controlled studies and are presented to allow for comparison of tofacitinib to
`placebo over 3 – 6 months and, in one study, to the active control adalimumab over 12 months.
`In four of these studies, patients were randomized to CP5, CP10, placebo (→CP5 at month 3 or
`6), placebo (→CP10 at month 3 or 6) in a 4:4:1:1 ratio, and in study 1032 in a 2:2:1:1 ratio.
`Patients originally randomized to placebo were advanced to either tofacitinib 5 mg or 10 mg at
`3 or 6 months, per protocol design, resulting in 3 treatment periods: 0 to 3 months (placebo-
`controlled portion of the studies), 3 to 6 months (some patients remained on placebo; some
`advanced to tofacitinib treatment), and >6 months (no placebo group; all patients advanced to
`tofacitinib treatment). In other words, there are two ways for placebo patients to advance to
`tofacitinib:
`
`
`1. cross-over by response: placebo patients who did not meet response criteria at a specified
`timepoint will advance to tofacitinib
`
`
`2. cross-over by design: (all remaining) placebo patients at specified timepoint will advance
`to tofacitinib.
`
`
`In general, the applicant presented the Phase 3 safety data by the overall 0 to 12 months duration,
`except for adverse events, serious adverse events, and discontinuations due to adverse events
`where data were also presented by treatment period. Incidence rates adjusted for patient-years of
`exposure and the corresponding 95% confidence intervals were calculated based on the number
`of patients with events. The applicant reported the rates by treatment group in two ways: (1)
`based on their original randomized treatment arms (i.e. as randomized), and (2) based on
`combining CP5 and CP10 groups and adding those placebo patients who advanced to tofacitinib
`
`8
`
`Reference ID: 3197461
`
`

`

`Placebo
`681
`221
`
`
`CP5 mg
`1216
`1451
`1056
`
`CP10 mg
`1214
`1439
`1046
`
`All Doses*
`2430
`2890
`2102
`
`Adalimumab
`204
`204
`204
`
`(i.e. as treated). During the first 3 months, safety data are summarized based on as randomized
`population (Table 3). At month 3, 460 out of 681 patients treated with placebo were advanced to
`either tofacitinib 5 mg or 10 mg at month 3, by either protocol design or by response. Even when
`these patients started tofacitinib therapy, they were counted in the 3 to 6 month group. Similarly,
`those patients who advanced to either tofacitinib 5 mg or 10 mg at month 6 were counted in the >
`6 month group even though they had just started tofacitinib therapy at that point.
`
`Table 3: Safety Analysis Population in Phase 3 Rheumatoid Arthritis Studies, as randomized and
`treated (Applicant’s)
`
`Studies
` 0 – 3 months
` 3 – 6 months
` > 6 months
`
`As noted earlier, we were concerned that the safety summaries the applicant provided may not be
`adequate to support regulatory decision regarding the safety of tofacitinib 5 mg or 10 mg doses.
`Together with the clinical team, we agreed that the two dose-ranging studies (studies 1025 and
`1035) are sufficiently similar in design and patient population to the Phase 3 studies to be
`included in the key integrated analyses. We also agreed that data from these studies should not
`be simply pooled for analysis but instead should be adjusted by study given the differing study
`designs and patient populations. Furthermore, the analyses should take into account placebo
`patients who advanced to tofacitinib and to take advantage of this unique design by applying
`modeling approaches to analyze the data, instead of relying on crude rates described by time
`point of measurement. As an example, suppose a patient who was initially treated with placebo
`but who advanced to tofacitinib 5 mg at month 3 experienced an adverse event at month 5, the
`applicant reported the event in the tofacitinib 3 – 6 months when in actuality this patient had
`been receiving tofacitanib for only 2 months.
`
`In summary, we identified three sets of comparison for safety evaluation:
`
`
`1. tofacitinib versus placebo (at 0 to 3 months), discussed in Section 3.3.1.
`2. tofacitinib versus adalimumab (at 0 to 12 months), discussed in Section 3.3.2.
`3. tofacitinib 5 mg versus tofacitinib 10 mg (at 0 to 12 months), discussed in Section 3.3.3.
`
`The risk of events was assessed through a Poisson regression model stratified by study with an
`offset term given by the logarithm of time until first event or censoring to the following safety
`data:
`
`
`1. Only patients originally randomized to tofacitinib 5 and 10 mg BID.
`2. Patients originally randomized to tofacitinib 5 and 10 mg BID + patients who
`transitioned to tofacitinib 5 and 10 mg BID by study design (month 3 for studies
`1032 and 1045, and Month 6 for studies 1044, 1046, and 1064).
`3. Patients originally randomized to tofacitinib 5 and 10 mg BID + patients who
`transitioned to tofacitinib 5 and 10 mg BID by study design (month 3 for studies
`
`
`
`Reference ID: 3197461
`
`9
`
`

`

`1032 and 1045, and Month 6 for studies 1044, 1046, and 1064) + patients who
`escaped to tofacitinib 5 and 10 mg BID due to active disease.
`
`
`Poisson regression was chosen since it models count data. It can also model count data per unit
`time to account for different length of exposure through an offset term (a variable that is forced
`to have a regression coefficient of