CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`
`203214Orig1s000
`
`OFFICE DIRECTOR MEMO
`
`
`
`
`

`

`
`November 6, 2012
`
`Curtis J Rosebraugh, MD, MPH
`Director, Office of Dru Evaluation 11
`
`Subject
`Summary Review
`NDA/BLA #
`203214
`
`Sun t #
`
`Applicant Name
`Pfizer
`Proprietary /
`Xeljanz (tofacitinib)
`Established
`
`(USAN) Names
`Dosage Forms /
`Stren
`
`5 and 10 mg tablets given twice a day
`
`IMARDs
`
`Treatment of Adult Patients with Moderately to Severely Active
`Rheumatoid Arthritis (RA) and Inadequate Response to One or more
`‘
`'
`'u . -Anti-Rheumatic Dru- s
`
`1.
`
`Introduction and Discussion
`
`This review will be a brief summary of the basis for the regulatory action regarding tofacitinib.
`Please refer to the action package for other reviews containing more detailed discussion.
`Tofacitinib is a new molecular entity (NMLE) that inhibits Janus Associated Kinases (JAK) that
`is proposed for the treatment of adult patients with RA with inadequate response to one or
`more DMARDs. Tofacitinib is a small molecular entity and allows for oral dosing as opposed
`to that of biologic agents, which require injection, and is the first JAK inhibitor seeking
`approval for the treatment of RA.
`
`Tofacitinib’s inhibition action is promiscuous affecting multiple JAK family members (JAKl,
`JAKZ, JAK3, TYK2).
`JAK signaling is involved in multiple pathways, such as colony—
`stimulating factor and grth hormone, and many cytokines. As such, JAK inhibitors are
`under development for a wide range of immune and hematopoietic disorders. Knockout mice
`and various recessive disorders have demonstrated the importance of JAKs in normal immune
`function. Due to their wide ranging effects on immune and hematopoietic systems,
`manipulation of JAK family members through inhibition has the potential to lead to adverse
`events associated with decreased fimction of the immune and hematopoietic systems.
`
`Presently, there are many drugs approved for the treatment of RA as illustrated by the
`following tables from Dr. Chowdhury’s review.
`
`Table l. Non—biologic small molecule DMARDs approved for marketing in the United States
`
`Product Name (Trade Name)
`S u nsor
`
`Mechanism of Action
`in RA
`
`Year of First Approval
`
`for RA Sulfasalazine (AZULFIDINE)
`
`Pfizer
`
`Anti-inflammatory
`and antimicrobial
`
`1950
`
`Reference ID: 321 3367
`
`

`

`
`
`Mechanism of Action
`in RA
`Anti-metabolite
`Interference with
`antigen processing
`Cytostatic
`
`Unknown
`
`Unknown
`
`Year of First Approval
`for RA
`1953
`
`1955
`
`1968
`
`1970
`
`1985
`
`T-cell activation inhibitor
`
`1995, 1990
`
`Anti-metabolite
`
`1998
`
`Product Name (Trade Name)
`[Sponsor]
`Methotrexate sodium (METHOTREXATE SODIUM)
`[Multiple]
`Hydroxychloroquine (PLAQUENIL)
`[Sanofi-Aventis]
`Azathioprine (IMURAN)
`[Prometheus Labs]
`Penicillamine (CUPRIMINE)
`[Alton]
`Auranofin (RIDAURA)
`[Prometheus Labs]
`Cyclosporine (NEORAL) (SANDIMMUNE)
`[Novartis]
`Leflunomide (ARAVA)
`[Sanofi-Aventis]
`
`Table 2. Biologic large molecule DMARDs approved for marketing in the United States
`Claims for adult RA §
`Product Name (Trade Name)
`Presentation
`Description
`and MOA ‡
`and ROA †
`[Sponsor] {year} *
`Etanercept (ENBREL)
`Vial 25 mg
`Fusion protein consisting of
`[Immunex/Amgen] {1998}
`Prefilled syringe 25 or 50 mg/mL
`TNF-R and human IgG1 Fc
`TNF-α inhibitor
`SureClick Autoinjector 50 mg/mL
`SC injection
`
`Vial 10 mg/mL
`Chimeric IgG1 k mAb
`TNF-α inhibitor
`IV infusion
`
`Infliximab (REMICADE)
`[Centocor] {1999}
`
`Anakinra (KINERET)
`[Amgen] {2001}
`
`Adalimumab (HUMIRA)
`[Abbott] {2002}
`
`Abatacept (ORENCIA)
`[Bristol Myers Squibb] {2005}
`
`Rituximab (RITUXAN)
`[Genentech and Biogen]
`{2006}
`Golimumab (SIMPONI)
`[Centocor] {2009}
`
`Certolizumab Pegol (CIMZIA)
`[UCB Inc] {2009}
`
`Prefilled syringe 100 mg
`SC injection
`
`Prefiled syringe 40 mg/0.8 mL
`Prefilled syringe 20 mg/0.4 mL
`Humira Pen 40 mg/0.8 mL
`SC injection
`Lyophilized powder 250 mg/vial
`IV infusion
`
`Vial 10 mg/mL
`IV infusion
`
`Prefiled syringe 50 mg/0.5 mL
`SmartJect Autoinjector 50 mg/0.5
`mL
`SC injection
`Lyophilized powder 200 mg/vial
`Prefilled syringe 200 mg/mL
`SC injection
`
`Recombinant polypeptide
`IL-1 receptor antagonist
`
`Human IgG1 k mAb
`TNF-α inhibitor
`
`Fusion protein consisting of
`CTLA-4 and human IGg1 Fc
`T cell activation inhibitor
`through B7-1 and B7-2
`Chimeric murine/human IgG1
`k mAb
`Anti CD20, B cell depletor
`Humanized IgG1 k mAb
`TNF-α inhibitor
`
`Humanized Fab fragment
`TNF-α inhibitor
`
`Tocilizumab (ACTEMRA)
`[Genentech/Roche] {2010}
`
`Vial 20 mg/mL
`IV infusion
`
`Humanized IgG1 k mAb
`IL-6 receptor inhibitor
`
`Clinical response
`Major clinical response
`Physical function response
`Radiographic response
`Clinical response
`Major clinical response
`Physical function response
`Radiographic response
`Clinical response
`Physical function response
`Radiographic response
`Clinical response
`Major clinical response
`Physical function response
`Radiographic response
`Clinical response
`Major clinical response
`Physical function response
`Radiographic response
`Clinical response
`Physical function response
`Radiographic response
`Clinical response
`Physical function response
`
`
`Clinical response
`Major clinical response
`Radiographic response
`Physical function response
`Clinical response
`Major clinical response
`Radiographic response
`Physical function response
`
`* Year = Year of first approval for RA
`† ROA = Route of administration
`‡ MOA= Mechanism of action
`§ Claims: Clinical response (or reducing signs and symptoms) assessed by ACR 20, 50, and 70 response over at least 3-6 month;
`Major clinical response defined as achieving ACR 70 response continuously over 6 months period; Physical function response (or
`improving physical function) assessed by health assessment questionnaire (HAQ) over at least 6 month period; Radiographic response
`(or inhibiting progression of structural damage) assessed radiographically by Total Sharp Score (TSS) and sometimes its components
`of erosion score (ES) or joint space narrowing (JSN) score over 6 or 12 months
`
`
`
`
`Reference ID: 3213367
`
`2
`
`

`

`
`
`
`
`This submission supports the approval of tofacitinib 5 mg twice a day for moderately to
`severely active RA as monotherapy for patients not tolerating methotrexate or in combination
`with methotrexate or other nonbiologic DMARDs. My summary of efficacy and safety and
`conclusions follows below.
`
`Efficacy
`
`Efficacy for RA drugs traditionally have been evaluated on the basis of American College of
`Rheumatology (ACR) response1, Health Assessment Questionnaire-Disability Index (HAQ-
`DI)2 assessment of physical functioning effect, Disease Activity Score (DAS)-283 and
`standardized radiographic scoring assessing effect on structural damage progression. Dose-
`ranging studies utilizing a range from 1 mg BID to 15 mg BID demonstrated a relatively flat
`dose-response for doses of 3 mg BID and above in subjects with inadequate response to
`background MTX therapy, with suggestions of dose-response for 5 mg BID and above as a
`monotherapy in patients with inadequate response to DMARDs . Based on this information,
`the sponsor chose to bring forth the 5 mg and 10 mg twice a day dosing for further
`development.
`
`Five trials form the basis of efficacy and safety for tofacitinib as summarized in the table
`below from page 11 of Dr. Yim’s review.
`
`
`1 A compilation of 1) tender joint count (68 joints), 2) swollen joint count (66 joints) and 3 of 5 ACR core set
`measures (patient global VAS, physician global VAS, patient assessment of pain VAS, patient assessment of
`physical function, acute phase reactant ESR or CRP). This compilation is measured as 20% or 50% or 70% or
`greater improvement.
`2 20 questions in 8 categories measuring fine movement of upper extremity, locomotor activities of lower
`extremities and activities of both. Activities include dressing, rising, eating, walking, etc. and are averaged based
`on a four-level difficulty scale from zero (no difficulty) to three (unable to do). Minimal clinical important
`difference for improvement is considered to be 0.22 units.
`3 Composite incorporating number of tender and swollen joints (out of 28), a patient global assessment VAS and
`ESR. Assess level of disease at a given timepoint. High activity DAS28 score >5.1, low activity <3.2.
`
`
`
`Reference ID: 3213367
`
`3
`
`

`

`
`
`Table 3: Summary of the Phase 3 Studies in RA Submitted for the NDA
`
`
`
`
`Dr. Yim has summarized the primary endpoints for each study (From page 15, italics below).
`
`All five studies are designed to establish superiority of the two doses of tofacitinib (5 mg and
`10 mg BID) to placebo for all the primary endpoints.
`
`Studies 1032, 1045, 1046, and 1064 had three primary endpoints, in sequence, as follows:
`1. Proportion of patients with ACR20 improvement
`a. At Month 3 for Studies 1032 and 1045
`b. At Month 6 for Studies 1046 and 1064
`2. Mean change from baseline to Month 3 in HAQ-DI
`3. Proportion of patients with DAS28<2.6
`a. At Month 3 for Studies 1032 and 1045
`b. At Month 6 for Studies 1046 and 1064
`
`
`Study 1044 had four primary efficacy endpoints, in sequence, as follows:
`1. Proportion of patients with ACR20 improvement at Month 6
`2. Mean change from baseline to Month 6 in modified Total Sharp Score (mTSS)
`3. Mean change from baseline to Month 3 in HAQ-DI
`4. Proportion of patients with DAS28<2.6 at Month 6.
`
`
`
`
`
`
`
`
`Reference ID: 3213367
`
`4
`
`

`

`
`
`ACR
`
`The primary endpoint results are summarized in the table below from Dr. Yim’s review (page
`17).
`Table 3: Summary of ACR20 Response Rates (Primary Endpoint) in Phase 3 RA Studies
`
`
`
`
`This shows that tofacitinib demonstrated efficacy for both the 5 mg and 10 mg BID dosage on
`ACR20 compared to placebo with a trend of numerical improvement for the 10 mg BID
`dosage compared to the 5 mg BID dosage. ACR50 and ACR70 results demonstrate that the
`tofacitinib groups had greater improvement compared to placebo control groups with slightly
`higher numerical response rates in the 10 mg BID compared to the 5 mg BID dosed groups.
`
`HAQ-DI
`
`HAQ-DI results are presented in the table below from page 18 of Dr. Yim’s review.
`
`
`
`
`Reference ID: 3213367
`
`5
`
`

`

`
`
`Table 4: HAQ-DI Endpoint Results
`
`
`HAQ-DI results are similar to those of the ACR, again with slightly higher response rates in
`higher dosage groups.
`
`Radiographic outcomes
`
`The primary radiographic endpoint of Van der Heijde modified Sharp Score4 for Study 1044 is
`presented below from page 20 of Dr. Yim’s review.
`
`
`
`
`
`4 Measures joint space narrowing (JSN) and erosions at 16 locations in each hand and wrist and 12 locations in
`each foot using a 6-point scale from 0 to 5 for a maximum total erosion score of 280. JSN based on 15 locations
`in each hand and wrist and 6 locations in each foot using a 5-point scale from 0 to 4 for a maximal score of 168.
`Theoretical maximum score of erosions + JSN is 448.
`
`
`
`Reference ID: 3213367
`
`6
`
`

`

`
`
`Table 5: Analyses of Change from Baseline in Modified Total Sharp Scores, Study A3921044
`
`
`The clinical and statistical reviewers question whether the results above are robust. The
`following graph from page 21 of Dr. Yim’s review illustrates their concern.
`
`Figure 1: Frequency Distribution of Change from Baseline in mTSS at Month 6
`
`
`
`Source: Figure 3 of the statistical review by Dr. Yongman Kim
`
`
`Dr. Yim notes that the results for the 10 mg dose are driven mainly by two outliers, one of
`which is from extrapolated data. The review team concludes that there is uncertainty in the
`radiographic outcome and lack of robustness of the primary endpoint to sensitivity analysis
`
`
`
`
`
`Reference ID: 3213367
`
`7
`
`

`

`
`
`because there was a low amount of progression in the placebo group and small overall
`magnitude to the treatment effect.
`
` I
`
`LS Mean
`
`-0.34
`-0.40
`
`
`-0.73, 0.04
`-0.79, -0.02
`
`
`p-value
`
`0.079
`0.038
`
`
`0.024
`0.198
`
`
` would note that radiographic trials in the current environment are very challenging. The total
`possible score for a mTSS is 448, yet the scale above demonstrates that trials are attempting to
`measure very small changes in score. The potential difference in scores between treatment
`groups is further reduced because of the ethical necessity to minimize the use of placebo, even
`as add-on treatment, and the need to limit the duration of inadequate control of disease
`activity. For this trial, the primary time point was at month 6, but many patients were advanced
`from one treatment group to another at Month 3 (49% of placebo patients and 26% of
`tofacitinib 5 mg patients) for non-response. This may account for the low amount of
`progression in the placebo group. This type of study design requires imputation methods to
`account for a large amount of missing data. The table below from Dr. Chowdhury’s review
`demonstrates the effect different imputation methods have upon the results.
`
`Table 7. mTSS score shown as change from baseline to month 6, study 1044, Study IV, or “Scan” *
`Treatment †
`N
`LS mean change
`Difference vs placebo
`
`
`from baseline
`95% CI
`Primary analysis, Parametric, Linear extrapolation method
`Tof 5 mg + methotrexate
`278
`0.12
`Tof 10 mg + methotrexate
`290
`0.06
`Placebo + methotrexate
`140
`0.47
`Alternate pre-specified analysis, Non-parametric ‡
`-77, -6
`-41
`334
`Tof 5 mg + methotrexate
`278
`-59, 12
`-23
`352
`Tof 10 mg + methotrexate
`290
`
`
`376
`Placebo + methotrexate
`140
`Alternate sensitivity analysis, excluding one patients from Tof 10 mg group with change of over 20 units
`0.056
`Tof 5 mg + methotrexate
`278
`0.11
`-0.34
`-0.69, 0.01
`0.061
`Tof 10 mg + methotrexate
`289
`0.12
`-0.33
`-0.68, 0.02
`
`Placebo + methotrexate
`140
`0.45
`
`
`* Study ID shown as Pfizer’s study number, and as referred in tofacitinib product label, and as Pfizer
`“identified” at the May 9, 2012, AAC meeting
`† Tof = Tofacitinib oral tablets
`‡ LS means from non-parametric analysis were based on the rank-transformed radiograph data
`
`
`Because of the earlier use of DMARDs in general, and the use of methotrexate as background
`therapy in this trial, it is not surprising that radiographic progression in this cohort of RA
`patients is not as pronounced as in historical RA patient cohorts. However, this illustrates the
`increasing challenge of demonstrating treatment effect on structural damage outcomes. The
`improvement we have attained in therapy over the years have made such studies difficult and
`will require further thought
` In any event, Study
`1044, while suggestive of a beneficial effect on radiographic progression,
`
`
`
`DAS-28 < 2.6
`
`The DAS-28 results for those achieving a score less than 2.6 are listed below in a table from
`page 22 of Dr. Yim’s review.
`
`
`
`
`
`Reference ID: 3213367
`
`8
`
`(b) (4)
`
`(b) (4)
`
`(b) (4)
`
`

`

`
`
`
`Table 6: Proportion of Patients with DAS28<2.6 Response
`
`
`
`
`FDA did a slightly different analysis using ITT population and nonresponder imputation for
`missing data. With that analysis, Studies 1032 and 1064 fail to reach statistical significance.
`
`Overall efficacy
`
`Tofacitinib doses of 5 mg and 10 mg BID demonstrated efficacy for signs and symptoms of
`RA as measured by ACR responses, DAS28 and HAQ-DI. For most outcomes, 10 mg BID
`provided a greater numerical response compared to the 5 mg BID dose.
`
`Tofacitinib probably has not demonstrated an effect on structural outcomes, although that may
`be more due to limitations of the trial design than activity of the drug.
`
`Safety
`
`As Dr. Chowdhury discusses, additional safety analyses of clinical data were requested from
`the sponsor during the review. This was because the sponsor had originally performed various
`pooling which limited quantification of safety events. These changes and their ramifications
`are thoroughly discussed in Dr. Chowdhury’s review. As a result, events of interest including
`death, lymphoma, solid organ tumor, opportunistic infection, tuberculosis, SAE infections,
`herpes zoster, MACE events, hemoglobin level, lipid profile, neutrophil counts, liver function
`test and common adverse events were re-analyzed at time intervals of 0-3 months, 0-6 months,
`0-12 months with appropriate treatment group assignment for patients switching from one
`treatment group to another. The re-analyzed safety is based on the five definitive efficacy and
`safety studies and two dose ranging studies (1025 and 1035). Results were also compared to
`adalimumab in those studies containing it as an active control. The tables below from Dr.
`Chowdhury’s review demonstrate the results.
`
`
`
`
`
`Reference ID: 3213367
`
`9
`
`

`

`
`
`Table 7. Summary of adverse event of interest in 0-12 months of treatment from seven studies *
`
`Patients as randomized
`Patients as treated ‡
`Treatment group †
`Placebo
`Tof
`Placebo
`Tof
`5mg BID
`5mg BID
`Number of patients
`1336
`1689
`Exposure, patient-years
`1056
`1241
`Deaths
` Number of deaths, n
` Incidence rate, per 100 PY
`Serious infection events
` Patients with ≥ 1 SIE, n (%)
` Incidence rate, per 100 PY
`Opportunistic infections
` Patients with ≥ 1, n
` Incidence rate, per 100 PY
`Tuberculosis
` Patients with ≥ 1 TB, n (%)
` Incidence rate, per 100 PY
`Herpes zoster infections
` Patients with ≥ 1 , n (%)
` Incidence rate, per 100 PY
`Malignancy, solid organ §
` Patients with ≥ 1 malignancy, n
` Incidence rate, per 100 PY
`Malignancy, lymphoma
` Patients with ≥ 1 lymphoma, n
` Incidence rate, per 100 PY
`MACE
`6
`2
`4
`2
` Patients with ≥ 1 MACE, n
`0.48
`0.83
`0.38
`0.83
` Incidence rate, per 100 PY
`* Two dose ranging efficacy and safety studies 1025 and 1035, and five definitive efficacy and safety studies 1045,
`1046, 1064, 1044, and 1032 (ID as Pfizer’s study number)
`† Tof = Tofacitinib oral tablets
`‡ Patients as treated are those who were randomized to the group plus placebo patients who were switched from
`placebo to tofacitinib treatment by study design or because of lack of response. Number of patients in the tofacitinib as
`treated group is larger than as randomized group because some patients from the placebo group advanced to tofacitinib
`groups at month 3 or month 6 by study design, and these patients are counted under both placebo group and tofacitinib
`group for the as treated group.
`§ Excludes non-melanoma skin cancer NMSC)
`
`Table 8. Adverse event of interest from studies 1035 and 1064 for patients as treated *
`
`Study 1035, 0-3 months ‡
`Study 1064, 0-12 months
`Treatment group †
`Adalimumab
`Tof, pooled
`Adalimumab
`Tof, pooled
`40 mg
`5 mg and 10 mg
`40 mg
`5 mg and 10 mg
`Number of patients
`53
`110
`204
`500
`Deaths
` Number of death, n
`Serious infection events
` Patients with ≥ 1 SIE, n
` Incidence rate, per 100 PY
`Opportunistic infections
` Patients with ≥ 1, n
`Tuberculosis
` Patients with ≥ 1 TB, n
` Incidence rate, per 100 PY
`Herpes zoster infections
`
`809
`240
`
`1
`0.42
`
`3 (<1)
`1.3
`
`-
`0
`
`-
`0
`
`5 (<1)
`2.1
`
`0
`0
`
`-
`0
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`5
`0.40
`
`34 (2)
`2.74
`
`4
`0.32
`
`-
`0
`
`47 (3)
`3.79
`
`5
`0.40
`
`-
`0
`
` 1
`
`
`
`17 §
`3.84
`
`-
`
`2
`0.45
`
`
`
`
`
`
`
` 0
`
`
`
`3
`1.54
`
`-
`
`-
`0
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`809
`240
`
`1
`0.42
`
`3
`1.3
`
`-
`0
`
`-
`0
`
`5 (<1)
`2.1
`
`-
`0
`
`-
`0
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`5
`0.47
`
`31 (2)
`2.94
`
`3
`0.28
`
`-
`0
`
`42 (2)
`3.98
`
`5
`0.47
`
`-
`0
`
` 0
`
`
`
`-
`0
`
`-
`
`-
`0
`
`
`
`
`
`
`
` 0
`
`
`
`-
`0
`
`-
`
`-
`0
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3213367
`
`10
`
`(b) (4)
`
`(b) (4)
`
`

`

`Treatment group 1'
`
`Stud 1035, 0-3 months
`Adalimumab
`Tof. pooled
`40m-
`5m- and 10m
`
`Stud 1064, 0-12 months
`Adalimumab
`Tof. pooled
`40m-
`5m- and 10 n-
`
`Patients with Z l . n
`Incidence rate. . . 100 PY
`
`Malignancy. solid organ //
`Patients with Z l malignancy. n
`Incidence rate. - ‘ 100 PY
`
`Malignancy. lymphoma
`Patients with 2 1 lymphoma. n
`Incidence rate. -
`' 100 PY
`MACE
`Patients with Z l MACE. n
`Incidence rate. . . 100 PY
`
`5
`2.57
`
`20 1|
`4.52
`
`
`
`* Two studies that included adalimumab as a comparator (1035 was a dose ranging efficacy and safety study and 1064
`was a definitive efficacy and safety study). Patients as treated are those who were randomized to the group plus
`placebo patients who were switched from placebo to tofacitinib treatment by study design or because of lack of
`response.
`T Tof = Tofacitinib oral tablets
`I Comparison between adalimumab and tofacitinib were done for 0-3 months because at month 3 all adalimumab
`patients and non-responders from placebo and tofacitinib 1 mg and 3 mg dose groups were advanced to tofacitinib 5
`mg
`§ 8 in tofacitinib 5 mg and 9 in tofacitinib 10 mg group based on patients as treated
`1| 7 in tofacitinib 5 mg and 13 in tofacitinib 10 mg group based on patients as treated
`I/ Excludes non-melanoma skin cancer NMSC
`
`An examination of Table 10 reveals that tuberculosis and malignancy rates appear slightly
`increased when examining groups dosed with 5 mg BID compared to groups dosed with 10 mg
`BID. There were too few events in Table 11 to examine for dose-ordering except perhaps for
`serious infection rates (8 in the 5 mg group and 9 in the 10 mg group) and herpes zoster (7 in
`the 5 mg group and 13 in the 10 mg group) where there were no clear trends.
`
`Lymphoproliferative disorders including lymphoma are always a concern with any drug that
`exerts its effects through the immune system. Only one case was noted in the 12-
`month data and this was in the 10 mg twice a day group. However, as noted in Dr.
`Chowdhury’s review, overall lymphoma was reported in seven patients, all in the tofacitinib
`group (without dose-ordering). Tofacitinib is also being developed for allografl rejection, and
`at doses of 15 mg BID, 5 cases were noted in 218 treated patients (most EBV-positive). It was
`also noted in animal studies that in the high dose cynomolgus monkeys, 5 of 8 animals had
`lymphoma.
`(m4)
`
`Laboratory tests of interest included lymphocyte and neutrophil counts, lipid parameters and
`liver enzyme tests. Lymphocyte count decrease below 500 cells/cmm occurred in 0.04% of
`patients in both the 5 and 10 mg twice-daily tofacitinib treatment groups during the first 3
`months of treatment. Neutrophil count decrease below 1000 cells/cmm occurred in 0.07% of
`patients in both groups during the first 3 months of treatment.
`
`Total cholesterol, LDL cholesterol, HDL cholesterol and triglyceride elevations occurred in a
`dose-related manner in patients treated with tofacitinib. Dming the first 3 months of exposure,
`
`Reference ID: 321 3367
`
`1 1
`
`

`

`LDL cholesterol increased by 15% and 19% in tofacitinib 5 mg and 10 mg groups
`respectively. There were too few cardiovascular events to evaluate if the changes in lipid
`profile translated into adverse events.
`
`There was one case of probable drug-induced liver injury (DILI) reported in a patient treated
`with tofacitinib 10 mg twice daily. There did not appear to be transaminitis shifts associated
`with tofacitinib use, which some may feel is necessary to categorize this case specifically as a
`‘Hy’s Law’ case. It is an important distinction to make as any extrapolation of potential
`population effects may be limited to those cases of DILI that fulfill the strict definition of
`‘Hy’s Law’. This is not to say that tofacitinib may not have the potential to cause DILI, only
`that our usual estimate of DILI as occurring at 1/10 the rate of Hy’s Law cases based upon
`exposure (1/50,000 exposures in this case) may not be accurate.
`
`In summary, tofacitinib use resulted in some safety concerns that occur with
`immunosuppressant drugs. These would include various infections and malignancy. There
`may be dose-ordering for some of these signals (tuberculosis, herpes zoster infection and
`malignancy). Laboratory tests demonstrated decreased lymphocyte and neutrophil counts
`comparing tofacitinib to placebo. Adverse lipid parameter changes were also noted with
`tofacitinib. It is also important to note that the events above were evaluated on a limited
`duration of 12 months of exposure which may not fully capture time-dependent events such as
`malignancy, lymphoma and others. Finally a case of probable DILI has been identified, which
`considering the size of the database may indicate rates of 1/50,000 (if Hy’s Law), or less (if
`not).
`
`DPARP plans on making a post marketing required study a condition of approval. This study
`will explore potential cardiovascular risks as well as others including further elucidation of
`malignancy and infectious risks. A REMS will also be required including a medication guide
`and communication plan outline the potential risks of therapy. I agree with this course of
`action.
`
`Advisory Committee Meeting
`
`This NDA was discussed at the May 9, 2012 Arthritis Advisory Committee meeting.
`Discussion of the members are summarized nicely in Dr. Yim and Chowdhury’s reviews.
`Overall, regarding whether there the efficacy and safety data provide substantial evidence to
`support approval for the treatment of moderately to severely active rheumatoid arthritis in
`patients who have had inadequate response to one or more DMARDs, the panel voted yes=8
`and no=2.
`
`Conclusions and Recommendations
`
`Tofacitinib 5 and 10 mg twice a day doses have both demonstrated efficacy. The 10 mg dose
`provided greater numeric response than the 5 mg dose for the primary and many secondary
`efficacy parameters.
`)(4)
`
`Reference ID: 321 3367
`
`12
`
`

`

`DPARPs position is that the 5 mg twice a day dose is the optimum dose for patients. This is
`perhaps true for the majority ofpatients. However, fliere are probably some patients for which
`the 10 mg twice a day dose would
`ovide
`ter relief and who would be willin to tolerate
`
`
`
`ater risk to achieve this oal.
`
`
`
`As it stands at present, I agree that the 5 mg twice a day dose should be approved for treatment
`of adult patients with moderately to severely active RA as monotherapy for patients who
`cannot tolerate methotrexate, and as a combination with methotrexate or other non-biologic
`DMARDs and that a PMR study should be conducted as outlined above. A REMS
`communicating the risks of therapy and appropriate monitoring is also appropriate.
`
`Reference ID: 3213367
`
`13
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`CURTIS J ROSEBRAUGH
`11/06/2012
`
`Reference ID: 3213367
`
`