CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`203214Orig1s000
`
`MEDICAL REVIEW(S)
`
`
`
`
`
`
`
`
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service
`
`
`
`
`Food and Drug Administration
`Center for Drug Evaluation and Research
`ODE II / DPARP / HFD-570
`10903 New Hampshire Ave.
`Silver Spring, MD 20993
`
`
`Memo to File
`
`Addendum to Primary Clinical Review
`
`
`NDA#:
`203,214
`SD#s:
`30, 31, 33, 36, 37, 38, 39, 40, 41
`Reviewer: Nikolay P. Nikolov, M.D., CDER/OND/DPARP
`Submission: October 21, 2011 (Original NDA)
`
`
`August 01, 2012 (Major Amendment Part 1, SD#36)
`
`
`August 10, 2012 (Major Amendment Part 2, SD#38)
`Reviewed: September 26, 2012
`Product:
`Tofacitinib (CP-690,550), an inhibitor of Janus kinase (JAK) family of kinases
`Proposed use: Treatment of rheumatoid arthritis (RA)
`Sponsor:
`Pfizer
`Summary: This is an addendum to the primary clinical review of NDA 203,214, tofacitinib for
`treatment of patients with RA, dated June 26, 2012. This document updates the primary review
`of the original NDA with:
`• A review of the revised safety analyses included in the major NDA amendment,
`• An updated risk-benefit assessment of tofacitinib for the intended indication, and
`• Updated labeling recommendations based on the new data.
`This document does not discuss issues covered in the primary review such as regulatory
`background, CMC, Pharmacology-Toxicology, Clinical Pharmacology, or clinical development.
`Efficacy is discussed in the context of the updated risk-benefit assessment.
`Based on the information from this amendment, the PDUFA goal date was extended by three
`months from August 21, 2012 to November 21, 2012 and the Sponsor was formally notified of
`the action on August 20, 2012.
`The results of the amended safety analyses are consistent with the results of the original NDA
`analyses. Further, the Agency’s analyses are in general agreement with the sponsor’s
`analyses of the safety data. The amended submission provided the data to refine the
`quantitative assessment of the relative risk of tofacitinib to placebo, adalimumab and between
`the two tofacitinib doses. No new or different safety issues were identified to significantly
`change the overall risk benefit assessment from the primary review.
`
`
`
`
`Reference ID: 3195340
`
`1
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`1. Recommendations/Risk Benefit Assessment
`
`1.1 Recommendation on Regulatory Action
`Recommend approval of NDA 203,214 for tofacitinib 5 mg BID with revisions to the proposed
`labeling, as outlined in Section 3. Labeling Recommendations.
`
`1.2 Risk Benefit Assessment
`
`Brief Overview of the Clinical Program
`The original 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
`patients with moderately-to-severely active 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 pivotal Phase 2 dose-ranging studies (1025 and 1035) and the five
`confirmatory Phase 3 studies are summarized in Table 1. For an overview of these studies, the
`reader is referred to the primary clinical review.
`
`Summary of Efficacy Review and Conclusions
`The amendment does not contain efficacy information. For review and conclusions regarding
`efficacy in this application, the reader is referred to the primary clinical review.
`
`Summary of Safety Review and Conclusions
`The review of tofacitinib clinical safety database, updated with the information in this
`amendment, identified several areas of major safety concerns potentially associated with
`tofacitinib administration:
`
`
`1. An increased risk of solid and hematologic malignancies, including lymphoproliferative
`disorder (LPD) indicating a potential safety signal.
`a. Risk of malignancy, excluding NMSC, increased numerically in a dose and time-
`dependent fashion:
`i. Malignancy was numerically higher in the tofacitinib 10 mg BID compared
`to 5 mg BID group during the 0-12 month controlled period of the pivotal
`randomized clinical studies, assessed both by:
`•
`Incidence rate ratio of 1.38 with 95% CI (0.44, 4.35) and
`• Mantel-Haenszel risk ratio of 1.41 with 95% CI (0.45, 4.42)
`accounting for differences across trials.
`ii. Malignancy was also numerically higher in the tofacitinib compared to
`adalimumab group during the 0-12 month controlled period of study 1064,
`as assessed both by increased:
`•
`Incidence rate ratio of 1.77 with 95% CI (0.18, 16.99) and
`• Mantel-Haenszel risk ratio of 1.51 with 95% CI (0.16, 14.43).
`iii. Increased with prolonged tofacitinib exposure (from 0.8 events per 100 pt-
`yrs for <6 months exposure to 1.4 events per 100 pt-yrs for >24 months
`
`
`
`Reference ID: 3195340
`
`2
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`
`exposure). The dose-relatedness in the long-term extension studies could
`not be adequately assessed due to study design limitations.
`b. There appears to be an increased risk of lymphoma in particular.
`i. Seven cases of LPD occurred in the overall RA development program as
`of May 2012, all in tofacitinib-treated patients. Two of the cases occurred
`in highly atypical locations (CNS and breast).
`ii. Five lymphoma cases in 218 (2.3%) renal transplant patients who had
`received 15 mg BID. Four of 5 patients were on 15 mg BID for 6 months
`and one of 5 patients was on 15 mg BID for 3 months prior to
`maintenance treatment with 10 mg BID.
`iii. Lymphomas occurred in the highest dose group (3 of 8, 37%) of the
`chronic toxicology study in monkeys.
`
`
`
`2. An increased risk of serious infections, including opportunistic infections, identifying a
`profile of tofacitinib as a major immunosuppressant.
`a. Serious infections associated with tofacitinib use were common in the RA
`program with pneumonia being the most common (occurring only in tofacitinib-
`treated patients).
`b. Tuberculosis occurred only in tofacitinib treated patients in a clearly dose-
`dependent fashion.
`c. Opportunistic infections were not uncommon and included cases of cryptococcal
`infections, Pneumocystis jiroveci pneumonia, and BK virus encephalitis, which
`are seen exclusively in severely immunocompromized patients.
`
`
`3. Laboratory abnormalities, to include abnormal hematologic parameters, lipid parameter
`changes, liver enzymes, and serum creatinine elevation. Tofacitinib administration was
`associated with:
`a. Dose-dependent sustained neutropenia and progressive lymphopenia. Severe
`lymphopenia was also associated with increased risk of infections.
`b. Dose-dependent sustained elevations in total, LDL, and HDL cholesterol. Total
`and LDL cholesterol levels reversed with lipid-lowering therapy. These lipid
`abnormalities did not appear to translate into increases in cardiovascular events.
`c. Liver enzymes elevations which were observed mostly in patients on background
`DMARDs and significant liver abnormalities were uncommon. However, in one
`case, drug-induced liver injury could not be excluded and the case is considered
`to meet Hy’s law criteria.
`d. Dose-dependent small but significant elevations of mean serum creatinine.
`These increases were associated with an increasing incidence of patients
`meeting the protocol criteria for discontinuation due to creatinine increases
`(confirmed creatinine increases of more than 50% of baseline). There did not
`otherwise appear to be an increase in the proportion of patients experiencing
`serious adverse events of renal failure.
`
`
`In conclusion, the safety data from tofacitinib RA development program is consisted with the
`profile of a potent immunosuppressant, with associated inherent risks, such as serious
`infections, including opportunistic infections and tuberculosis. Tofacitinib administration was
`also associated with malignancy (excluding non-melanoma skin cancer, NMSC) in a manner
`that may be consistent with a dose- and duration of exposure- dependent manner.
`
`3
`
`Reference ID: 3195340
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`Gastrointestinal perforations and interstitial lung disease were observed in the clinical trials,
`however the relative risk and role of tofacitinib treatment in the development of these adverse
`events is not well defined. Treatment with tofacitinib resulted in dose-dependent changes in
`laboratory parameters, such as sustained neutropenia and progressive
`lymphopenia,
`sustained elevations in total, LDL, and HDL cholesterol, small but significant elevations of
`mean serum creatinine, and liver enzymes elevations. While most of these were not
`associated with clinical adverse events in the controlled setting of the clinical trials, severe
`lymphopenia was associated with increased risk of infections. One case of Hy’s law occurred
`with tofacitinib treatment. Using the estimate of severe drug-induced liver injury as occurring
`at 1/10th the rate of Hy’s Law cases, 1 case of severe liver injury might be expected in 50,000
`patients treated with tofacitinib.
`
`Most of these potential safety issues are seen with other traditional and biologic DMARDs and
`have historically been handled via appropriate labeling and risk evaluation and mitigation
`strategies, which should also be adequate in this case. The clinical trial experience has been
`extensive, but may not be sufficient to capture the full extent of safety concerns that may arise
`with long-term JAK inhibition with tofacitinib, which is a new molecular entity.
`
`The risk of malignancy and serious infections increased numerically with prolonged tofacitinib
`exposure; however due to limitations in the design of the long-term safety database, the
`incremental increase in risk associated with dose and duration of exposure could not be
`adequately characterized. However, as noted in the main clinical review, the long-term data
`submitted raise a concern that the risk could be increased in the higher dose with increasing
`duration of exposure. Therefore, to further address residual uncertainties regarding the long-
`term safety of tofacitinib, a long-term study of both tofacitinib doses 5 mg and 10 mg BID along
`with an active comparator is warranted, which will be consistent with the recommendations
`form the Arthritis Advisory Committee panel.
`
`Risk Benefit Overview
`Tofacitinib is a new molecular entity and if approved, would be the first in class of drugs, Janus
`associated kinase (JAK) inhibitors, for the treatment of patients with RA. Importantly, tofacitinib
`is an oral agent providing potential advantages over exiting injectable DMARDs.
`
`At both tested doses, 5 mg and 10 mg BID, tofacitinib demonstrated clinical benefit on signs
`and symptoms and physical function in patients with established moderately-to-severely active
`RA, both as a monotherapy and in combination with MTX or other traditional DMARDs, with
`results in the range of approved biologic DMARDs such as TNF inhibitors. Tofacitinib, however
`did not show a clear evidence of radiographic benefit in this patient population, based on the
`results of a single study.
`
`Tofacitinib has a safety profile of a potent immunosuppressant associated with increased risk
`of serious infections, including opportunistic infections and tuberculosis, solid and hematologic
`malignancy, possible
`idiosyncratic hepatotoxicity, and dose-dependent
`laboratory
`abnormalities, among which elevated LDL, HDL and total cholesterol.
`
`
`
`
`Reference ID: 3195340
`
`4
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`The amended safety analyses of the major events of interest from the seven pivotal studies
`using Poisson regression analyses and accounting for differential study design were consistent
`with the observations in the original application.
`
`During the 12 month controlled period of the seven pivotal studies, malignancy, excluding
`NMSC, serious infections, including tuberculosis, and opportunistic infections were identified
`as safety signals associated with the use of tofacitinib. Dose-dependent numerical increases in
`the relative risk of malignancies (excluding NMSC) and tuberculosis were observed, reaching
`statistical significance (p-value=0.03 before adjusting for multiplicity) only for tuberculosis.
`
`In conclusion, the overall risk to benefit profile of tofacitinib in RA appears to be favorable, with
`many more patients potentially benefiting from treatment compared to those at potential risk.
`Most of the potential safety issues such as malignancy, tuberculosis, serious and opportunistic
`infections, hepatotoxicity and laboratory abnormalities are seen with other traditional and
`biologic DMARDs and have historically been handled via appropriate labeling and risk
`evaluation and mitigation strategies, which should also be adequate in this case.
`
`Dosing Recommendations
`
`The clinical benefit of tofacitinib 5 mg BID and 10 mg was generally comparable. Numerical
`differences were observed in ACR and DA828 response rates and change in HAQ-Dl between
`the two doses; however these were not consistent across the randomized controlled studies
`and any differences were relatively small and of questionable clinical significance.
`“"0
`
`(I!) (4)
`
`herefore, based on the available data, the overall risk to benefit profile is more
`favorable for the 5 mg BID dose and until additional data are available to better characterize
`the long-term safety profile of the two doses, only the 5 mg BID dose should be approved.
`
`Recommendations for Postmarketing Risk Evaluation and Mitigation Strategies
`
`A Risk Evaluation and Mitigation Strategy (REMS)
`communication to healthcare providers about the risks of:
`
`is warranted and should assure
`
`0 Serious infections, including opportunistic infections, and tuberculosis
`
`o Malignancy
`0 Cardiovascular major adverse events associated with tofacitinib-induced elevation in
`lipid parameters
`. Changes in laboratory parameters, such as decreases in neutrophil counts, lymphocyte
`counts, hemoglobin levels,
`increases in low density lipoprotein cholesterol (LDL-c),
`serum creatinine, transaminase elevations and potential hepatotoxicity associated with
`tofacitinib.
`
`The risks of tofacitinib treatment, including malignancy, serious infections, hepatotoxicity, and
`necessity of laboratory monitoring should also be communicated to patients.
`
`Recommendations for Postmarketing Requirements and Commitments
`
`o A long-term prospective safety study:
`
`Reference ID: 31 95340
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`A study of both tofacitinib doses 5 mg and 10 mg BID along with an active comparator
`to assess the safety for major AEs of interest that may manifest after prolonged
`tofacitinib exposure, such as cardiovascular adverse events, malignancies, serious
`infections,
`and hepatotoxicity. Such a study would help addressing residual
`uncertainties regarding the long-term safety of tofacitinib and would be consistent with
`the recommendations by the Arthritis Advisory Committee members for collecting long-
`term safety data.
`
`0 Studies to achieve compliance with PREA:
`Polyarticular juvenile idiopathic arthritis (leA) is considered to be the pediatric
`equivalent of adult RA. Therefore, in accordance with the Pediatric Research Equity Act
`(PREA) of 2003, studies in leA are mandated. With this submission, the applicant has
`requested a deferral for patients age 2-17 with leA, and a waiver for children 0-2,
`since leA is extremely rare in this age group. These requests have been granted for
`other therapeutic biologics, as, for ethical reasons, it is desirable to have an adequate
`experience with the safety profile of a treatment in adults before proceeding with
`extensive studies in children. The applicant has already discussed details of their
`proposed Phase 3 program in leA and systemic juvenile idiopathic arthritis (leA) with
`the Agency.
`"m
`
`Reference ID: 31 95340
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`2. Review of the Amendment
`
`2.1 Background
`
`The original 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
`patients with moderately-to-severely active 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 efficacy statistical review was on the five Phase 3
`studies 1032, 1044, 1045, 1046, and 1064.
`
`The pivotal Phase 2 dose-ranging studies (1025 and 1035) and the five Phase 3 confirmatory
`studies are summarized in Table 1.
`
`
`
`
`
`Reference ID: 3195340
`
`7
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, MD.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`Table 1 . Key Design Features of NDA 203,214 Pivotal Randomized Controlled Studies for Efficacy and Safety
`
`Patient Population
`
`Patients with incom lete res nse to rior TNF inhibitor
`Moderate—to—severe RA
`MTX-IR,
`Stable background MTX
`
`Enrolled
`Randomization
`
`507
`1:1:1:1:1:1.'1
`
`Moderatetesevere RA
`1NF-IR,
`Stable background MTX
`
`Treatment Arms (transition and escape for NR)
`
`Primary
`Endpoints
`
`Timepomt
`
`CP1mgBlD(—.CP5mgBID@M03ifNR)+MI'X
`CP3mgBID(—.CP5mgBID@Mo3ifNR)+MTX
`CP5mgBID+M'D(
`CP10mgBID+MTX
`CP15mgBID+MTX
`CP20mgOD(—»CP5mgBID@Mo3ifNR)+M'D(
`Placebo (—.CP5mg BID@ Mo3ifNR)+MTX
`CP 5 mg BID + MTX
`CP 10 mg BID + MTX
`Placebo (—.CP 5 mg BID @ Mo3)+ MTX
`Placebo —»CP 10
`
`ACR20
`HAQ-DI
`DA828<2.6
`
`Patients with incomplete response to MTX or other DMARD:
`Moderatetosevere RA
`R, DB, AC
`DMARD-IR,
`Phase 2,
`No background therapy
`Doseranging
`6 months
`
`384
`1:1:1:1:1:1:1
`
`DA828<2.6
`
`CP1 mg BID (—.CP5mg BID@ Mo3ifNR)
`CP 3 mg BID (—.CP 5 mg BID @ M03 if NR)
`CP 5 mg BID
`CP 10 mg BID
`CP 15 mg BID
`Adalimumab (—.CP 5 mg BID @ M03)
`Placebo (—.CP 5 mg BID @ M03 if NR)
`CP 5 mg BID + MTX
`CP 10 mg BID + MTX
`PBO («CP5mgBlD@M060rMo3ifNR)+ MTX
`PBO (—pCP 10 mg BID @ M0 6 or M03 if NR)+ MIX
`CP5mg BID
`CP 10 mg BID
`PBO—vCPSmgBID@M03
`PBO —vCP 10
`o BID @
`CP 5 mg BID + DMARD
`CP 10 mg BID + DMARD
`PBO (—pCP 5 mg BID @ Mo 60r M03 ifNR) + DMARD
`PBO (—pCP 10 mg BID @ M0 6 or M03 if NR) + DMARD
`CP 5 mg BID + PBO SC+ MTX
`CP10mgBID+ PBOSC+MTX
`PBO («CP5mg BID@ Mo6orMo3ifNR)+ PBO SC + MTX
`PBO (—.CP 10 mg BID @ M0 6 or M03 ifNR) + PBO SC + MTX
`P80 + Adalimumab + MTX
`Source: Sunmzy of Clinical Efficacy, Ciniml Study Reports for studies A3921032. A3921044, A3921045, A3921046, A3921064
`'-One year eflimcy data submitted tor Study A3921044; '-Background DMARD herapy 'n Study A3921046: 84% of subjects on MIX, -1I2 on corrbination DMARDs; Ac-active control (adalimumab, study
`A3921064); BID-two ines daily, DMARDs—disease—rmdilying anti-rheumatic dmgs; IR-inoomplete response; MTX—methotrexate; mTSS-rmdifred total Sharp Score; NRnon-rasponderdefned as patients who failed
`
`Moderate-tdsevere RA
`MTX—IR,
`Stable background MTX
`
`Moderate-to-severe RA
`DMARD-IR,
`No background to Month 3
`
`Moderate-to—severe RA
`DMARD-IR,
`Stable background DMARDs'
`
`Moderate-tdsevere RA
`MTX—IR,
`Stable background MTX
`
`ACR20
`mTSS
`HAQ-DI
`DA328<2_6
`ACR20
`HAQ—Dl
`DAS28<2.6
`
`ACR20
`HAQ-DI
`DA328<2_6
`
`ACR20
`HAQ-Dl
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`The review of the efficacy data concluded that there is substantial evidence of efficacy of
`tofacitinib 5 mg or 10 mg for the treatment of rheumatoid arthritis 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.
`m"
`
`The applicant’s safety evaluation was based on the pooled safety data from five confirmatory
`double-blind phase 3 studies in RA (1032, 1044, 1045, 1046, and 1064). These data were
`supported by pooled data from five completed phase 2 studies (1025, 1039, 1019, 1035, and
`1040) and pooled data from two ongoing long-term extension (LTE) studies (1024 and 1041).
`For an overview of these studies, the reader is referred to the primary clinical review.
`
`The methods used in the safety analyses of the original NDA were consistent with those from
`other RA development programs. These safety analyses provided a qualitative but inadequate
`quantitative assessment of the long-term safety and the relative safety of the two dosing
`regimens (5 mg vs. 10 mg BID) of tofacitinib to allow for regulatory decision and accurate
`labeling.
`
`Due to the complexity of the trial design, such as early escape options and unequal
`randomization, the following issues were identified as potential limitations with regard to the
`safety data presentation and analyses in the original application:
`
`1. Data were analyzed for patients as randomized and not as treated. For example,
`patients who were on placebo and transition to active treatment at Months 3 or 6 were
`counted under the respective active treatment arm from that point on, instead of time
`“zero". Presenting the safety data for patients as treated may allow for a more precise
`assessment of crude proportions of patients with AEs for any given time period, even
`though this approach may not affect the incidence rates of AEs, adjusted for actual
`exposure to the drug, as discussed in this document. One important consideration with
`using this approach however,
`is that patients who escape from placebo to active
`treatment due to active disease represents a group of patients with more active disease
`who may have a different baseline risk of developing adverse events and therefore may
`represent a somewhat different population than the one originally randomized to active
`treatment.
`
`2.
`
`It was not clear from the submission whether the rules for capturing and reporting of
`AEs were applied consistently within the RA development program which may introduce
`a bias in AE reporting. For example, only deaths were reported if they occurred on
`treatment or within 30 days of last dose.
`
`3. Only safety data from the five Phase 3 randomized controlled trials were included in the
`pooled safety analyses (A3921032, A3921044, A3921045, A3921046, and A3921064).
`However, the two pivotal Phase 2 dose-ranging studies (A3921025, and A3921035)
`were of similar design and patient population and could be included in the pooled
`analyses to increase the overall sample size for assessment of controlled data.
`
`4. Long-term extension study A3921041 included patients who completed Japanese
`Phase 2 studies which were not included in the pooled analyses of the controlled
`studies. Therefore, excluding these patients in the open label extension analyses may
`
`Reference ID: 31 95340
`
`9
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`
`allow for more accurate comparison with the analyses from the controlled studies. One
`caveat of this approach however, is that important safety information may be excluded.
`
`
`To address these concerns, the Division had asked the sponsor to reconsider how best to
`analyze the safety data, particularly those major events of interest. The Division sent several
`information requests to the applicant 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
`sponsor to clarify some issues or roadblocks regarding the requests.
`
`At a Type A meeting on July 10, 2012, between the Division and the sponsor the following
`additional analyses were agreed upon as a path forward to addressing the information
`requests and the Agency’s concerns:
`• Seven trials to be included in integrated analysis:
`o Five phase 3 studies – 1032, 1046, 1044, 1045, 1064
`o Two phase 2 studies – 1025, 1035,
`• Timing of Events . Events within 30 days of stopping treatment or decreasing dose. No
`30 day window (hard stop) for placebo to tofacitinib cross over, increasing tofacitinib
`dose.
`• Time Intervals
`o 0-3 months
`o 0-6 months
`o 0-12 months
`• Events of interest
`o death
`o
`lymphoma
`o solid organ tumor
`o opportunistic infection
`o TB
`o SAE infection
`o herpes zoster
`o CV MACE events
`• Events of interest for labeling (These events can be for 0-3 month time interval as long
`as laboratory changes have reached a plateau).
`o Hemoglobin
`o Lipids
`o Neutrophils
`o LFTs
`o Common AEs
`For transparency, the Agency provided the sponsor with its plan for additional safety analyses
`to include.
`1. Assessment of the safety profile of tofacitinib 5 mg BID relative to 10 mg BID for the 0-6
`and 0-12 month periods of the 7 studies, with respect to events of interest, stratified by
`study, using Poisson regression with an offset term to account for differing exposure
`times, and study to account for differing patient populations and/or study designs.
`a. Only patients originally randomized to tofacitinib 5 and 10 mg BID
`
`
`
`Reference ID: 3195340
`
`10
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`
`b. 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).
`c. 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) + patients who
`escaped to tofacitinib 5 and 10 mg BID due to active disease.
`2. Comparison of the safety profile of tofacitinib, relative to adalimumab, with respect to
`events of interest using data from studies 1035 and 1064, separately.
`
`
`The Sponsor has submitted the requested analyses as a major NDA amendment on August
`10, 2012 which was within three months of the user fee goal date. Therefore, the Agency
`extended the goal date by three months to provide time for a full review of the submission. The
`extended user fee goal date is November 21, 2012.
`
`Long-term extension studies considerations
`Analyses of the long-term extension data from studies 1024 and 1041 proved to be challenging
`for quantitative assessment of the relative risk of the two tofacitinib doses for several reasons,
`including:
`• The study design allowed for inconsistent dosing between the index and extension
`studies. For example:
`o
`In study 1024, following implementation of Protocol Amendment 3
`(January 21, 2009), enrolling patients initiated open-label tofacitinib 10 mg
`BID at the baseline visit (except patients in China, who initiated open-label
`tofacitinib 5 mg BID at the baseline visit). Prior to implementation of
`Amendment 3, all patients initiated open-label tofacitinib 5 mg BID at the
`baseline visit. Patients who are taking tofacitinib 10 mg BID were allowed
`to decrease dosing to 5 mg BID for mild to moderate safety concerns and
`those who are taking 5 mg BID were allowed to increase their dose to 10
`mg BID at the instruction of the investigator.
`In study 1041, all patients were to initiate at a dose of tofacitinib 5 mg BID.
`The dosage may be increased from 5 mg BID to 10 mg BID, reduced from
`10 mg BID to 5 mg BID, or temporarily discontinued (up to 28 consecutive
`days) based on consideration of the risks and benefits to the patient.
`• The duration of exposure among patients in the extension studies was highly variable.
`For example, some patients may have had only 3 months of tofacitinib exposure
`(studies 1045 and 1032) versus 12 months for patients from studies 1044, 1046, and
`1064.
`• The overall duration of exposure to tofacitinib 5 mg BID dose was significantly longer
`compared with 10 mg BID (see Table 29 in the primary clinical review).
`
`
`Because of these shortcomings in the design of the open label extension studies, quantitative
`assessment of the relative risk of the two tofacitinib doses was limited to assessment of safety
`data from the controlled periods of the pivotal randomized clinical trials.
`
`
`
`o
`
`Reference ID: 3195340
`
`11
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`To further address residual uncertainties regarding the long-term safety of tofacitinib, a long-
`term study of both tofacitinib doses 5 mg and 10 mg BID along with an active comparator is
`warranted. This will help assessing the safety for major AEs of interest that may manifest after
`prolonged tofacitinib exposure, such as cardiovascular adverse events, malignancies, serious
`in section Recommendations
`for
`infections, and hepatotoxicity as
`recommended
`Postmarketing Requirements and Commitments.
`
`
`2.2 Methods
`
`As agreed with the sponsor, the amended safety data analyses consist of an integrated
`analysis of the controlled periods of the seven pivotal studies (five confirmatory studies – 1032,
`1046, 1044, 1045, 1064, and two dose-ranging studies – 1025, 1035) described in the primary
`clinical review and summarized in Table 1.
`
`
`1. Analyses of major events of interest (death, malignancy excluding NMSC, serious and
`opportunistic infections, tuberculosis, herpes zoster, and MACE) to compare the relative
`safety profile of the two tofacitinib dosing regimens (5 mg vs. 10 mg BID) during the
`controlled periods of the studies, i.e. 0-12 months.
`
`
`
`To account for the differences in study design, i.e. patients transitioning to active
`treatment by design or due to active disease at either 3 or 6 months in the different
`studies, the Division’s statistical review team identified and conducted three different
`analyses using Poisson regression modeling for:
`a) Only patients originally randomized to tofacitinib 5 and 10 mg BID (Analysis 1
`population)
`b) Patients originally randomized to tofacitinib 5 and 10 mg BID + patients who
`transitioned to tofacitinib 5 and 10 mg BID by study design, i.e. month 3 for
`studies 1032 and 1045, and Month 6 for studies 1044, 1046, and 1064 (Analysis
`2 population).
`c) 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) + patients who
`escaped to tofacitinib 5 and 10 mg BID due to active disease (Analysis 3
`population).
`
`
`The risk of events is assessed through a Poisson regression model stratified by study
`with an offset term given by the logarithm of time until first event or censoring. The
`parameter estimated by the Poisson model after anti-logarithmic transformation is the
`Incidence Rate Ratio (IRR). The Mantel-Haenszel Risk Ratio is also shown as a
`secondary analysis method. For rare events, both measures of risk are expected to
`produce similar results. Note that the MH Risk Ratio does not take into account the
`subject-specific time of exposure, only the number of subjects in each treatment arm.
`Also shown in tables is the observed pooled event rate per 100 patient years of
`exposure for each event of interest and treatment arm.
`
`
`
`
`Reference ID: 3195340
`
`12
`
`

`

`Addendum to Primary Clinical Review
`Reviewer: Nikolay P. Nikolov, M.D.
`NDA 203,214
`Tofacitinib for Rheumatoid Arthritis
`
`
`2. Analyses of common AEs and laboratory parameters to compare the relative safety
`profile of tofacitinib (5 mg and 10 mg BID) to placebo during the controlled periods of
`the studies, i.e. 0-3 months:
`a) Patients originally randomized to tofacitinib 5 and 10 mg BID or placebo
`b) Patients originally randomized to tofacitinib 5 and 10 mg BID or placebo +
`patients who transitioned from placebo to tofacitinib 5 and 10 mg BID by study
`design at Month 3 (studies 1032 and 1045 only). This analysis did not include
`patients who transitioned to active treatment due to active disease because they
`may have a different baseline risk of developing adverse events. Similarly, this
`analysis did not include patients who continued to receive placebo through Mon