CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`
`204063Orig1s000
`
`SUMMARY REVIEW
`
`
`
`Page 1 of 29
`
`Biogen Exhibit 2374
`Coalition v. Biogen
`IPR2015-01993
`
`

`
`Cross Discipline Team Leader Review
`
`Cr oss-Discipline Team L eader Review
`
`Date
`From
`Subject
`NDAIBLAH
`Su pplement#
`Applicant
`Date of Submission
`PDUF A Goal Date
`
`P roprietary N arne f
`Established (USAN) n ames
`Dosage forms I Strength
`Proposed lndication(s)
`
`Recommended:
`
`1. Introduction
`
`2111113
`Billy Dunn, MD
`Cross-Discipline Team Leader Review
`204063
`
`Biogenldec
`2/27/ 12
`3/27/13
`
`Tecfidera/dimethyl fumarate
`
`Oral delayed release capsules/ 120 mg, 240 mg
`Treatment otJ>atients with relaE_sing_ fonns of mul~le
`sclerosis I
`(til<]
`I
`Approval
`
`I
`
`The sponsor (Biogen Idee) has submitted a new dmg application (NDA) to support the
`marketing of dimethyl fumarate (Tecfidera), a new oral chug with a pro osed indication for the
`S
`treatment of atients with relapsing fonns of multi le sclerosis
`(till"
`
`Dimethyl fumarate (D1v1F) has not been previously approved and is categorized as a new
`molecular entity. A related dmg product, a combination of DMF with other fumarate esters
`including the primaty metabolite of DMF, monomethyl fumarate (MMF), was approved in
`Germany in 1994 for the treatment of psoriasis and is marketed as Fumadenn. The proposed
`mechanism of action ofDMF it) MS is activation of the nuclear factor ( ety throid-derived 2)(cid:173)
`like 2 (Nr£2) b·anscriptional pathway that is involved in the cellular response to oxidative
`stress, ostensibly reducing inflanunatoty responses in both peripheral and central cells and
`promoting cytoprotection of central nervous system cells against toxic oxidative insults.
`
`The review team for this NDA included the following primruy reviewers:
`
`Chemistty - David Claffey, PhD
`Chemisby (Methods Validation Inspection) - Michael Trehy
`Chemisby (Biophannaceutics) - Elsbeth Chikhale, PhD
`Office of Manufacturing and Product Quality (Inspections) - Derek Smith, PhD
`Nonclinical- Melissa Banks-Muckenfuss, PhD
`Nonclinical (Carcinogenicity) - Steven Thomson, PhD
`Clinical Phru1nacology - Jagru1 Pru·epally, PhD
`
`Reference ID: 3282328
`
`Page 2 of 29
`
`I
`
`

`
`Cross Discipline Team Leader Review
`
`Clinical Pharmacology (IRT-TQT) – Qianyu Dang, PhD
`Division of Bioequivalence and GLP Compliance (Inspection) – Michael Skelly, PhD
`Statistics – Xiang Ling, PhD
`Clinical (Efficacy) – Heather Fitter, MD
`Clinical (Safety) – Gerard Boehm, MD
`Division of Medication Error Prevention and Analysis – Julie Neshiewat, PharmD
`Division of Risk Management – Kendra Worthy, PharmD
`Division of Medical Policy Programs – Shawna Hutchins, MPH, RN
`Pediatric and Maternal Health Staff (Maternal) – Carrie Ceresa, PharmD
`Pediatric and Maternal Health Staff (Pediatric) – Nadia Hejazi, MD
`Controlled Substance Staff – Alicja Lerner, MD, PhD
`Division of Pharmacovigilance – Andrew Fine, PharmD
`Division of Professional Drug Promotion – Quynh-Van Tran, PharmD
`Division of Consumer Drug Promotion – Meeta Patel, PharmD
`Study Endpoints and Labeling Development – Elizabeth Donohoe, MD
`Office of Scientific Investigations – Antoine El-Hage, PhD
`
` I
`
` discuss below the key conclusions of each reviewer and provide my recommendations
`regarding this submission.
`
`
`2. Background
`
`
`DMF is not an approved drug product anywhere in the world. It has been under
`investigational development (IND 73061) in the United States for the treatment of multiple
`sclerosis since 2006. As noted above, Fumaderm is approved in Germany for the treatment of
`psoriasis.
`
`As primary support for the proposed indication, the sponsor presents the results from two
`controlled Phase 3 efficacy study (studies 109MS301 and 109MS302). Both studies were of
`similar design and evaluated the effect of 240 mg bid and 240 mg tid of DMF in patients with
`MS on a variety of outcomes. In addition, as further support, the sponsor presents the results
`of a controlled Phase 2 dose-finding study (study C-1900) and interim results of an ongoing
`open-label, dose and rater-blinded extension study (109MS303).
`
`One meeting with the sponsor focused on this submission took place, a pre-NDA meeting on
`1/25/12. There are no significant outstanding issues from this meeting.
`
`
`3. CMC/Device
`
`
`Dr. Claffey reviewed this submission and found it acceptable.
`
`Dr. Chikhale reviewed this submission and found it acceptable.
`
`Mr. Trehy reviewed this submission and found it acceptable.
`
`2
`
`Reference ID: 3282328
`
`Page 3 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`Dr. Smith completed the manufacturing inspection and found it acceptable.
`
`There are no outstanding CMC issues. There are no CMC post-approval recommendations.
`
`
`4. Nonclinical Pharmacology/Toxicology
`
`
`Dr. Thomson reviewed this submission and found the statistical considerations of the
`carcinogenicity studies acceptable.
`
`Dr. Banks-Muckenfuss reviewed this submission and found it unacceptable. She does not
`recommend approval. She bases her recommendation on nonclinical findings of renal toxicity,
`including tumors in rodents, at clinically relevant doses in all species assessed.
`
`As described by Dr. Banks-Muckenfuss, animal data have demonstrated that DMF causes
`multiple toxicities across organ systems, including “kidney, testes, stomach (nonglandular),
`pancreas, liver, thymus, lymphatic system, and eye (retina).”
`
`It is the renal toxicity that is most concerning. The renal tubular and interstitial toxicity seen
`in animals was widespread and somewhat insidious. It appears to occur at lesser doses with
`increasing duration of exposure, and damage may not clearly be seen in studies of lesser
`duration. Predictors of toxicity in the animals were not seen consistently in different species
`(urinary protein only in rats) and the utility of such assessments in humans as predictors of
`toxicity is uncertain. The renal findings in rodents included renal tumors. These tumors may
`or may not be species specific. In addition to tumors, the renal findings may be irreversible, as
`seen in the chronic monkey study.
`
`Dr. Banks-Muckenfuss is concerned that the toxicities, particularly the carcinogenicity, may be
`compatible with the known actions of DMF. (Dr. Boehm discusses this to some degree, as
`well). She is perhaps most troubled by the notion that the enhanced clinical monitoring in
`humans may have been inadequate and that the toxicity may not yet be seen in trials of
`possibly insufficient duration. Taken together, she is left to conclude that the safety database
`from the clinical trials was potentially inadequate to detect possible “irreversible tissue damage
`and loss of function” along with renal tumors associated with human doses of DMF that are
`linked to relevant toxic doses in animals. She does acknowledge that the relevance of the
`animal findings to human risk is unclear.
`
`Dr. Banks-Muckenfuss’s supervisor, Dr. Lois Freed, performed an independent secondary
`review with specific attention to renal factors. She, too, observed evidence of widespread
`multi-organ toxicity across multiple species (rodent, dog, monkey), with clear evidence of
`renal toxicity.
`
`Upon detailed review of the data, she is somewhat more hopeful, though still cautious, that
`predictive human monitoring (urinary albumin) may be useful in the avoidance of potential
`renal toxicity. That said, the chronic toxicity study in monkey and dog resulted in the
`development of irreversible interstitial fibrosis consistent with low level chronic renal toxicity
`
`3
`
`Reference ID: 3282328
`
`Page 4 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`and, while BUN and creatinine were decreased (consistent with the findings in rat) there were
`no urinary findings consistent with renal toxicity.
`
`Given the availability of Fumaderm clinical data, Dr. Freed briefly reviewed its toxicology
`studies and found a similar, though perhaps somewhat less severe, toxicological profile.
`
` re-evaluation of mouse and rat carcinogenicity data by the sponsor’s expert consultant
`resulted in no substantial change in the findings of the mouse study but, in the rat study, a
`reconsideration of the renal tumors resulted in a change in renal tumor incidence such that
`their incidence was only slightly increased, only in females, and was no longer considered
`drug-related.
`
`Reproductive and developmental toxicity findings remained significant.
`
`Taken together, Dr. Freed feels the sponsor has conducted an adequate battery of nonclinical
`studies to support marketing of DMF for treatment of patients with relapsing forms of multiple
`sclerosis.
`
`She finds that rodent forestomach, rodent and dog testes, and pan-species (mouse, rat, dog,
`monkey) kidney were the primary target organs. She describes that forestomach is of
`questionable relevance to humans. She feels that testicular findings can and should be
`described in clinical labeling. Finally, she agrees with Dr. Banks-Muckenfuss that the data
`demonstrate a potential for human renal toxicity, suggesting the possibility of irreversible
`injury due to low level chronic injury and repair.
`
`Recognizing that the review team is in agreement that clinical trial monitoring may not have
`been able to detect renal injury consistent with that seen in animals, Dr. Freed feels that the
`efficacy findings in clinical trials along with the available safety data from those clinical trials,
`limited though it may be, combined with the Fumaderm postmarketing experience (namely, no
`indication of renal toxicity with longer-term exposure) are sufficient to support approval. She
`agrees with the plans for the large 5 year observational post-approval study discussed below.
`
`Thus, with appropriate labeling, she recommends approval, along with a nonclinical
`postmarketing requirement to conduct a juvenile animal toxicology study to support pediatric
`clinical development.
`
`
` A
`
`5. Clinical Pharmacology/Biopharmaceutics
`
`
`Dr. Parepally, Dr. Dang, and Dr. Skelly reviewed this submission and found it acceptable.
`
`Detailed labeling recommendations are found in the clinical pharmacology review.
`
`The clinical pharmacology review notes that MMF is the active metabolite of DMF and that
`DMF is not detectable in systemic circulation due to rapid and complete hydrolysis. The
`conclusions below were based on evaluation of plasma concentrations of MMF.
`
`4
`
`Reference ID: 3282328
`
`Page 5 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`
`There are no outstanding clinical pharmacology issues. There are no clinical pharmacology
`post-approval recommendations.
`
`Pharmacokinetics
`Although MMF exposure in individuals was highly variable, overall exposure increased
`approximately proportionally in response to single and multiple doses of 120-360 mg of DMF.
`Tmax was achieved in 2 to 2.5 hours in the fasted state. Protein binding was 27-45%. MMF
`has a volume of distribution of 53-73 liters. DMF was extensively and rapidly metabolized to
`MMF, the only active metabolite, which is further metabolized through the citric acid cycle.
`The elimination half-life was 0.5 to 1.4 hours, leading to no accumulation with multiple doses.
`The major elimination route of MMF was exhalation as carbon dioxide, accounting for 60% of
`the dose. Urinary and fecal elimination were minor routes at 15% and 1%, respectively.
`
`Food effect
`Food roughly doubled the Tmax. It also led to a modest reduction in flushing from 94% to
`68%.
`
`Pharmacodynamics
`The mechanism of action may be activation of the nuclear factor (erythroid-derived 2)-like 2
`(Nrf2) transcriptional pathway, reducing inflammation and promoting cytoprotection.
`
`Intrinsic factors
`Age and race – no dose adjustments are recommended, although no meaningful conclusions
`can be drawn regarding race or the elderly due to a lack of patient variability.
`
`Gender – no dose adjustments are recommended.
`
`Renal impairment – not studied as this was a minor route of elimination.
`
`Hepatic impairment – not studied as this was a minor route of elimination.
`
`Drug-drug interactions
`No effects were seen. The potential for interactions is low.
`
`Thorough QT study
`DMF did not show any potential for prolonging the QTcF interval compared with placebo.
`
`Pharmacometrics
`N/A
`
`Pharmacogenomics
`N/A
`
`
`5
`
`Reference ID: 3282328
`
`Page 6 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`6. Clinical Microbiology
`
`
`N/A
`
`
`7. Clinical/Statistical- Efficacy
`
`
`Dr. Ling and Dr. Fitter reviewed this submission. Both recommend approval.
`
`As discussed by Dr. Ling and Dr. Fitter, as primary support for the application, the sponsor
`submitted two adequate and well-controlled efficacy studies, 109MS301 and 109MS302
`(study 301 and study 302).
`
`Studies 301 and 302 were multi-national, multi-center, randomized, double-blind, placebo-
`controlled, parallel-group studies to evaluate the efficacy and safety of two doses of DMF (240
`mg bid and 240 mg tid) in patients with relapsing-remitting multiple sclerosis (RRMS).
`Dosing was initiated in each group at half the target dose for one week before escalating to the
`full dose (120 mg bid and 120 mg tid for 7 days, followed by 240 mg bid and 240 mg tid long-
`term).
`
`Study 302 also included an additional open-label active comparator arm using glatiramer
`acetate. As Dr. Fitter discusses on page 85 of her review, although the sponsor argues that
`DMF was superior to the active comparator, these data are essentially uninterpretable by
`design and I will not discuss them further.
`
`Enrollment criteria for studies 301 and 302 were typical of MS trials including the following
`notable key criteria: diagnosis (per 2005 revised McDonald criteria) of RRMS with EDSS of
`0 to 5, at least 1 relapse over the preceding year with MRI findings consistent with MS, and no
`relapses in the 50 days prior to randomization with a stable clinical course at the time of
`randomization.
`
`The primary efficacy endpoint for study 301 was the proportion of subjects relapsed (PR) at 2
`years.
`
`The primary efficacy endpoint for study 302 was the annualized relapse rate (ARR) at 2 years.
`
`The “key” secondary efficacy endpoints for study 301 were, in rank order, number of new or
`newly enlarging T2 hyperintense lesions, number of Gd-enhancing lesions, ARR, and
`disability progression.
`
`The “key” secondary efficacy endpoints for study 302 were, in rank order, number of new or
`newly enlarging T2 hyperintense lesions, number of new T1 hypointense lesions, proportion of
`subjects relapsed, and disability progression.
`
`
`6
`
`Reference ID: 3282328
`
`Page 7 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`Disability progression was defined as at least a 1.0 point increase on the EDSS from a baseline
`EDSS of 1.0 or greater or a 1.5 point increase on the EDSS from a baseline EDSS of 0 that
`was sustained for 12 weeks.
`
`The above primary and secondary endpoints were analyzed in a hierarchical fashion.
`
`Dr. Ling and Dr. Fitter have provided a discussion of these various measurements and the
`statistical approach used in their analyses. Their use in these trials is acceptable.
`
`Study 301
`
` total of 1237 patients were randomized while 1234 were randomized and treated as follows:
`
`
`408 subjects to placebo
`410 subjects to 240 mg bid
`416 subjects to 240 mg tid
`
`Patients were enrolled from 198 centers in 28 countries. Approximately 15% were from the
`United States. Overall, patients were distributed widely throughout the world.
`
`The MRI cohort comprised 540 treated subjects, well balanced across all three treatment
`groups.
`
`282 patients, well balanced across all three treatment groups, did not complete the study. The
`discontinuation rate was considered by both Dr. Ling and Dr. Fitter. Neither concluded that
`they called the results of the study into question.
`
`Demographic and baseline characteristics of the patients were well-matched. As is typical for
`MS trials, most patients were relatively young white women.
`
`The results for the primary outcome in the standard ITT population (assigned treatment at
`randomization with at least 1 day of study medication exposure), presented by the sponsor and
`confirmed by the review team, are below:
`
` A
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`p-value
`< 0.0001
`< 0.0001
`
`risk reduction compared to placebo
`
`
`49%
`
`
`
`
`50%
`
`
`
`PR
`
`240 mg bid 0.27
`240 mg tid
`0.26
`Placebo
`0.46
`
`Dr. Ling (pages 17-18 of her review) conducted or confirmed multiple sensitivity analyses,
`including worst case scenario analyses designed to account for treatment discontinuation and
`potential biased relapse assessment due to unblinding, and found that all analyses were highly
`consistent with the primary analysis. She also examined the effect of flushing on treatment
`effect to account for unblinding and found the treatment effect preserved. Dr. Fitter also
`presents and discusses these analyses on pages 53-54 of her review. Taken together, these
`various analyses are strongly supportive of the primary analysis.
`
`7
`
`Reference ID: 3282328
`
`Page 8 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`
`The results for the secondary outcomes described above, assessed in hierarchical fashion,
`presented by the sponsor and confirmed by the review team, are below:
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`T2 lesions
`2.6
`
`4.4
`
`17.0
`
`risk reduction compared to placebo
`
`
`85%
`
`
`
`
`74%
`
`
`
`Gd T1 lesions risk reduction compared to placebo
`0.1
`
`
`
`90%
`
`
`0.5
`
`
`
`73%
`
`
`1.8
`
`
`
`
`
`ARR
`0.172
`0.189
`0.364
`
`risk reduction compared to placebo
`
`
`53%
`
`
`
`
`48%
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`p-value
`<0.0001
`<0.0001
`
`p-value
`<0.0001
`<0.0001
`
`p-value
`<0.0001
`<0.0001
`
`disability progression
`0.164
`
`
`
`0.177
`
`
`
`0.271
`
`risk reduction compared to placebo p-value
`
`38%
`
`
`
`0.0050
`
`34%
`
`
`
`0.0128
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Ling describes various sensitivity analyses of these secondary outcomes on pages 18-21 of
`her review, and all are consistent with the findings described above except for an analysis of
`disability progression sustained for 24 weeks (rather than 12 weeks as was pre-specified).
`Although it numerically favored DMF, it did not reach significance for either group, as can be
`seen below:
`
`
`
`disability progression
`0.128
`
`
`
`0.119
`
`
`
`0.169
`
`risk reduction compared to placebo p-value
`
`23%
`
`
`
`0.1893
`
`31%
`
`
`
`0.0760
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Fitter examined this issue (page 58 of her review) and postulates that the larger number of
`placebo patients who began alternative MS medications after progression (there were no such
`patients in the DMF groups) could have disproportionately reduced the number of patients
`with 24 week sustained progression in the placebo group.
`
`Overall, Dr. Ling and Dr. Fitter agree that study 301 provides convincing evidence of
`effectiveness on the primary and all secondary endpoints.
`
`Dr. Fitter also provides an exploration of various exploratory endpoints on pages 58-64 and
`102-106. None of her findings argue against the findings described above.
`
`
`8
`
`Reference ID: 3282328
`
`Page 9 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`
`Study 302
`
` A
`
` total of 1430 patients were randomized while 1417 were randomized and treated as follows:
`
`
`363 subjects to placebo
`359 subjects to 240 mg bid
`345 subjects to 240 mg tid
`350 subjects to glatiramer acetate (again, these subjects will not be discussed below; they are
`included for the sake of completeness with regard to the total number of patients in the trial; it
`is also worth noting that patients randomized to this group constitute 10 of the 13 patients who
`were randomized but not dosed, and 8 of those 10 withdrew consent upon learning of their
`open-label treatment assignment)
`
`Patients were enrolled from 200 centers in 28 countries. Approximately 20% were from the
`United States. Overall, patients were distributed widely throughout the world.
`
`The MRI cohort comprised 681 treated subjects, well balanced across all four treatment
`groups.
`
`290 patients, well balanced across all four treatment groups, did not complete the study. The
`discontinuation rate was considered by both Dr. Ling and Dr. Fitter. Neither concluded that
`they called the results of the study into question.
`
`Similar to study 301, demographic and baseline characteristics of the patients were well-
`matched, and most patients were relatively young white women.
`
`Analyses and analytic strategy for study 302 are similar to those for study 301 and will be
`presented below in a similar fashion.
`
`The results for the primary outcome in the standard ITT population (assigned treatment at
`randomization with at least 1 day of study medication exposure), presented by the sponsor and
`confirmed by the review team, are below:
`
`
`
`
`
`
`
`ARR
`0.224
`0.198
`0.401
`
`
`
`
`
`
`
`
`
`p-value
`<0.0001
`<0.0001
`
`risk reduction compared to placebo
`
`
`44%
`
`
`
`
`51%
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Ling (pages 21-22 of her review) conducted or confirmed multiple sensitivity analyses,
`including worst case scenario analyses designed to account for treatment discontinuation and
`potential biased relapse assessment due to unblinding, and found that all analyses were highly
`consistent with the primary analysis. She also examined the effect of flushing on treatment
`effect to account for unblinding and found the treatment effect preserved. Dr. Fitter also
`presents and discusses these analyses on page 76 of her review. Taken together, these various
`analyses are strongly supportive of the primary analysis.
`
`9
`
`Reference ID: 3282328
`
`Page 10 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`
`The results for the secondary outcomes described above, assessed in hierarchical fashion,
`presented by the sponsor and confirmed by the review team, are below:
`
`T2 lesions
`5.1
`
`4.7
`
`17.4
`
`T1 lesions
`3.0
`
`2.4
`
`7.0
`
`risk reduction compared to placebo
`
`
`71%
`
`
`
`
`73%
`
`
`
`risk reduction compared to placebo
`
`
`57%
`
`
`
`
`65%
`
`
`
`
`
`
`
`risk reduction compared to placebo
`
`
`34%
`
`
`
`
`45%
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`p-value
`<0.0001
`<0.0001
`
`p-value
`<0.0001
`<0.0001
`
`p-value
` 0.0020
`< 0.0001
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`
`
`PR
`
`240 mg bid 0.29
`240 mg tid
`0.24
`Placebo
`0.41
`
`
`
`disability progression
`0.128
`
`
`
`0.130
`
`
`
`0.169
`
`risk reduction compared to placebo p-value
`
`21%
`
`
`
`0.25
`
`24%
`
`
`
`0.20
`
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Ling describes various sensitivity analyses of these secondary outcomes on pages 22-24 of
`her review, and all are consistent with the findings described above except for analyses of
`disability progression. In this case, as the disability findings were not statistically significant,
`the sensitivity analyses could not confirm the finding regardless of the outcome. That said, the
`pre-specified sensitivity analysis of disability progression sustained for 24 weeks (rather than
`12 weeks as was primarily pre-specified) neared statistical significance in the bid dose group,
`as seen below:
`
`
`disability progression
`0.078
`
`
`
`0.086
`
`
`
`0.125
`
`risk reduction compared to placebo p-value
`
`38%
`
`
`
`0.0630
`
`33%
`
`
`
`0.1172
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Ling again conducted a sensitivity analysis of disability sustained through the end of the
`study, with a significant result in the bid dose group, as seen below:
`
`
`disability progression
`0.064
`
`
`
`0.091
`
`
`
`0.125
`
`240 mg bid
`240 mg tid
`Placebo
`
`
`risk reduction compared to placebo p-value
`
`48%
`
`
`
`0.0200
`
`27%
`
`
`
`0.2123
`
`10
`
`Reference ID: 3282328
`
`Page 11 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`Dr. Fitter points out the sponsor’s comment that the placebo disability progression rate was
`low in study 302, approximating the treated progression rate in the study 301. She argues that
`the primary findings of the trial trump this post-hoc cross-study comparison, a reasonable and
`valid position. She also notes that a baseline imbalance in the placebo group in study 302
`might support the notion of an unusually low disability progression rate, but she points out that
`such an imbalance did not exist. Given the similarity in the trials, Dr. Fitter performed a
`pooled analysis for the 12 week disability secondary outcome assessed in both trials:
`
`
`disability progression
`0.146
`
`
`
`0.155
`
`
`
`0.222
`
`risk reduction compared to placebo p-value
`
`32%
`
`
`
`0.0034
`
`30%
`
`
`
`0.0059
`
`240 mg bid
`240 mg tid
`Placebo
`
`Dr. Ling and Dr. Fitter agree that study 302 provides convincing evidence of effectiveness on
`the primary and all secondary endpoints, except for disability progression.
`
`Dr. Fitter again provides a similar exploration of various exploratory endpoints on pages 82-87
`and 102-106.
`
`Dr. Ling (pages 25-28 of her review) conducted subgroup analyses by demographic and
`baseline characteristics of study 301 and 302 and found that the results were generally
`consistent across all subgroups, with no findings suggesting a particularly different pattern of
`effect than that seen in the main group analyses.
`
`Overall, based upon both study 301 and 302, both Dr. Ling and Dr. Fitter feel that DMF had a
`convincing and significant effect on all primary and secondary endpoints except for disability
`progression in study 302. Despite this, both are supportive of DMF's nominal effect on
`disability, variously citing the strong results in study 301, the possibility of a somewhat
`underpowered sample size in study 302 relative to study 301, and the suggestion of benefit in
`various additional analyses of study 302 as well as the pooled data from both studies.
`
`Dr. Ling feels that both doses of DMF used in studies 301 and 302 are efficacious.
`
`Dr. Fitter agrees that both doses of DMF are efficacious and supports the approval of the 240
`mg bid regimen. She argues that as the 240 mg tid regimen confers no apparent benefit over
`the 240 mg bid regimen, the latter is the correct choice for approval. I note that while the 240
`mg bid regimen was numerically superior on nearly all endpoints in study 301, the 240 mg tid
`regimen was numerically superior on all endpoints in study 302. Further, the differences
`favoring each regimen were not particularly dramatic in either study, and they did not appear
`on face to represent clinically meaningful effects.
`
`Dr. Fitter also discusses study C-1900 on pages 28-30 of her review. This was a dose-finding
`study that assessed imaging outcomes of 256 MS patients in four dose groups: placebo, 120
`mg daily, 120 mg tid, and 240 mg tid. According to the sponsor’s pre-specified analysis
`strategy, only the 240 mg tid dose group demonstrated a significant effect on imaging
`
`11
`
`Reference ID: 3282328
`
`Page 12 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`outcomes (Gd T1 lesions on MRI). Additional analyses supported the superiority of the 240
`mg tid dose, leading to its use in the study 301 and 302.
`
`Finally, Dr. Fitter discusses study 109MS303 (study 303) on pages 88-90 of her review. This
`is an ongoing long-term extension of studies 301 and 302 primarily intended to assess safety,
`although efficacy measures are being assessed. Subjects who completed study 301 or 302
`either continued on their current dose of DMF or, if originally randomized to placebo or
`glatiramer acetate, were re-randomized to one of the two dose groups of DMF used in the
`original study. In this manner, all patients in study 303 receive DMF in a blinded fashion.
`Interim efficacy data as of August 3, 2011, were submitted by the sponsor and are descriptive
`in nature. They are unable to contribute to a determination of effectiveness.
`
`
`8. Safety
`
`
`Dr. Boehm reviewed this submission and found no obstacles to approval related to safety.
`
`As noted above, though DMF has not been previously approved, Fumaderm, a combination of
`DMF with other fumarate esters including the primary metabolite of DMF, monomethyl
`fumarate (MMF), was approved in Germany in 1994 for the treatment of psoriasis.
`
`EXPOSURE
`
`As Dr. Boehm discusses, the safety database for DMF exceeds standard ICH guidelines for the
`characterization of common adverse events at the intended recommended dose. Safety data on
`3424 subjects in clinical trials of healthy volunteers, MS, psoriasis, and rheumatoid arthritis
`were submitted by the sponsor.
`
` A
`
` total of 2665 MS subjects have received at least one dose of DMF. Of these 2665 MS
`patients, 1787 were exposed for at least 1 year, 1198 were exposed for at least 2 years and 712
`were exposed for at least 3 years.
`
`The intended recommended dose is 240 mg bid and 2537 MS patients have received this dose
`or greater. Of those 2537, 1136 received the intended recommended dose.
`
`The bulk of the safety analysis stems from two pools of MS subjects:
`
`Pool A – data from placebo-controlled trials in MS (C-1900 part 1, 109MS301, and
`109MS302) representing 1720 subjects (2323.5 patient years of exposure).
`
`Pool B – data from placebo-controlled and uncontrolled trials in MS (C-1900 parts 1 and 2,
`109MS301, 109MS302, and 109MS303) representing 2513 subjects (4306.7 patient years of
`exposure).
`
`
`
`12
`
`Reference ID: 3282328
`
`Page 13 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`A similar arrangement exists for two pools of psoriasis subjects:
`
`Pool C – data from placebo-controlled trials in psoriasis representing 213 subjects (52.5
`patient years of exposure).
`
`Pool D – data from placebo-controlled and uncontrolled trials in psoriasis representing 296
`subjects (317.7 patient years of exposure).
`
`Additional clinical development safety data comes from assorted small early phase trials.
`
`Finally, the sponsor submitted post-marketing safety data for Fumaderm.
`
`Overall, little difference was seen between the MS data and the psoriasis data, and the MS data
`will be emphasized.
`
`DEATHS
`
`There were 11 deaths in the DMF clinical development program (9 on drug, 1 on glatiramer
`acetate, and 1 on placebo). Of the 9 on drug, 7 occurred in MS trials, and 2 in psoriasis trials.
`These are briefly described below:
`
`
`1. A 54 year old woman with MS treated with 120 mg bid for 5 days died from a
`traumatic brain injury resulting from a bicycle accident.
`2. A 38 year old woman with MS treated with 240 mg tid for 61 days died from a motor
`vehicle accident.
`3. A 55 year old woman with MS treated with either 240 mg bid or tid for 196 days died
`from an acute malignant tumefactive MS relapse with eventual intraventricular
`hemorrhage.
`4. A 31 year old woman with severe MS treated with 240 mg bid for 848 days died from
`an MS relapse accompanied by infection leading to cardiopulmonary arrest.
`5. A 40 year old woman with MS treated with 240 mg tid for 760 days died from suicide.
`6. A 32 year old woman with MS treated with 240 mg bid for 346 days (interrupted) died
`from sepsis from decubitus ulcers.
`7. A 49 year old woman with MS treated with 240 mg bid for 406 days died from
`mesothelioma.
`8. A 44 year old man with psoriasis and multiple cardiovascular risk factors treated with
`240 mg tid for 75 days died from likely sudden cardiac arrest.
`9. A 48 year old man with psoriasis and multiple cardiovascular risk factors treated with
`240 mg tid for at least 249 days died from likely sudden cardiac arrest.
`
`
`As Dr. Boehm discussed in his review on page 21, the Pool A deaths do not seem to differ in
`frequency among treatment groups.
`
`
`
`13
`
`Reference ID: 3282328
`
`Page 14 of 29
`
`

`
`Cross Discipline Team Leader Review
`
`SERIOUS ADVERSE EVENTS (SAEs)
`
`In Pool A, there was no individual SAE reported by at least 1% of DMF patients and more
`commonly compared to placebo (Dr. Boehm’s review, page 24). In fact, the overall frequency
`of SAEs was slightly greater in the placebo group.
`
`In Pool B, Dr. Boehm notes that the only SAE that occurred in ≥1% of DMF exposed patients
`was MS relapse (9%, 227/2513). In Pool A, MS relapse was more common in placebo
`subjects.
`
`The psoriasis pools (C and D) were equally unconcerning.
`
`On pages 26-29 of his review, Dr. Boehm explores several sporadic SAEs of potential concern
`(including skin reactions and allergic reactions) and finds nothing to clearly implicate DMF
`and plausible alternative