1. Cataloging FDA’s Flexibility in Subpart H and Orphan Drug Approvals Frank Sasinowski, M.S., M.P.H., J.D Hyman, Phelps & McNamara, P.C. FJS@hpm.com 2014 MSG Scientific Annual Meeting September 22, 2014

2. Flexible approach to drug development for neuromuscular diseases –An analysis of FDA Subpart H approvals 21 non-cancer, non-HIV drugs approved via Subpart H pathway FDA flexibility evident in these precedents, but in different ways than seen in the orphan drug context –An analysis of drugs approved under Orphan Drug Act 135 orphan drugs approved from 1983 enactment of the Orphan Drug Act through July 1, 2010 2/3 of those approvals manifested FDA’s flexibility in applying the statutory requirements for establishing substantial evidence of effectiveness to gain marketing approval for orphan drugs Upcoming update –28 drugs approved between July 1, 2010 - June 30, 2014 FDA’s New Patient Focus Introduction September 22, 20142

3. Efficacy evidence required under the federal Food, Drug, and Cosmetic Act -“Substantial evidence” of effectiveness -“Adequate and well-controlled investigations” Orphan Drug Act of 1983 -Did not change the quantum or quality of either the safety or effectiveness evidence needed for drugs intended for rare disorders Quantum of Evidence Required September 22, 2014 3

4. FDA has regulations and policies providing greater flexibility for certain types of drugs (e.g., Subpart H), or under certain circumstances (e.g., May 1998 Evidence Guidance) FDA has publicly recognized its exercise of flexibility specifically for drugs treating Americans with rare diseases (Dr. Katz’s statement in October 2011 on Xenazine as approved under FDAMA 115) Formal FDA Policies and Statements September 22, 2014 4

5. Subpart H/Accelerated Approvals Part #1: Subpart H Approvals

6. Recent Milestones related to Subpart H An Analysis of Subpart H Precedents, according to Factors listed in FDA’s May 2014 Guidance on “Expedited Programs for Serious Conditions” Part #1: Subpart H Overview September 22, 20146

7. FDA Safety and Innovation Act (FDASIA) – July 2012 Four Recent Milestone Events, #1 7September 22, 2014

8. In July 2012, FDASIA refined the starting “Fast Track” codification of 21 C.F.R. Part 314, Subpart H, by defining a Subpart H therapy in this way: –“a product for a serious or life-threatening disease... that... has an effect on a surrogate... that is reasonably likely to predict clinical benefit, or on a clinical endpoint..., taking into account the severity, rarity, or prevalence of the condition and the availability or lack of alternative treatments.” Four Recent Milestone Events, #1 (con’t) September 22, 20148

9. President’s Council of Advisors on Science and Technology Report (PCAST) – Sept. 2012 “FDA’s interpretation of ‘reasonably likely... to predict’ can have a major impact on the pace of medical innovation....” “Historically, the use of [Subpart H] has been primarily used in a limited number of therapeutic areas— principally, HIV/AIDS, cancer, and inhalation anthrax (87 percent of cases)....” “We believe that the Nation would benefit if the FDA were to expand the use in practice of acceptable indicators to other serious or life-threatening diseases.” “The FDA should expand the use in practice of its existing authority for Accelerated Approval.” “FDA should direct its staff, across all divisions, to make full use of the Accelerated Approval track for all drugs meeting the statutory standard of addressing an unmet medical need for a serious... illness and demonstrating an effect on a clinical endpoint (other than survival or irreversible morbidity) or on a surrogate endpoint that is reasonably likely to predict clinical benefit.” Four Recent Milestone Events, #2 9September 22, 2014

10. FDA’s Guidance, “Expedited Programs for Serious Conditions– Drugs and Biologics” – May 2014 Lists and describes Factors that FDA views as critical to Accelerated Approval/ Subpart H Four Recent Milestone Events, #3 10September 22, 2014

11. Energy & Commerce Committee’s Subcommittee on Health 21 st Century Cures - May 2014 –Congress, for 1 st time in history, comprehensively addressing medical product development –Testimony from 1 st hearing focused on increasing visibility to Subpart H and exploring expanded use of intermediate clinical benefits (“ICE”) Four Recent Milestone Events, #4 11September 22, 2014

12. An Analysis of Subpart H Precedents – According to Factors in FDA’s Guidance By Frank Sasinowski and Alexander Varond Hyman, Phelps & McNamara, P.C.

13. Purpose To promote a better understanding of the circumstances under which Subpart H may be employed in order to: –Facilitate the development and expedited review of new drugs with the potential to address unmet medical needs for serious and life-threatening illnesses –Mobilize expanded FDA use of Subpart H To examine and describe the evidentiary foundation for FDA’s determinations that an unvalidated surrogate or clinical endpoint was “reasonably likely to predict” patient benefit sufficient to meet statutory standard of substantial evidence of effectiveness 13September 22, 2014

14. Methods Identified FDA’s 21 Subpart H approvals for conditions other than HIV/AIDS or cancer Provided analysis with narrative text describing most relevant information pertinent to each Factor listed in FDA Guidance –Including clinical evidence on the surrogate and, where available, on ultimate clinical benefit FDA said it could not address this most critical of factors in its guidance because the clinical evidence is not “generalizable.” In this analysis, we then chose not to generalize, but to describe specifics of each precedent. 14September 22, 2014

15. Methods Scoring (20-point scale) Part 1: Rarity of Condition (2 points) Part 2: Understanding of the Disease Process (4 points) –Understanding of the pathophysiology of the disease Part 3: Understanding of Relationship Between the Drug’s Effect on Surrogate and the Disease (4 points) –This understanding may come from epidemiological evidence, animal models, other drugs in similar pharmacologic class or other sources Part 4: Strength of Clinical Evidence (10 points) –Clinical evidence for surrogate (7 points) –Clinical evidence of benefit (3 points) 15September 22, 2014

16. Results – Summary Score Card Part 1: Rarity of Condition –Range: 1 - 2 points (out of maximum of 2) –Findings: Scores were consistently high Part 2: Understanding of the Disease Process –Range: 1 - 4 points (out of maximum of 4) –Findings: Although a clear understanding of the pathophysiology of the disease process will facilitate reliance upon a surrogate, the absence of a complete understanding of the disease process is not incompatible with Subpart H (e.g., Remicade) –“Crohn’s disease most likely represents a heterogeneous group of disorders. After much effort that has focused on the identification of a specific pathogenic cause, it is being recognized that disease manifestations could result from a combination of any, or all of, a number of factors.” (Medical Review at p. 2). 16September 22, 2014

17. Results – Summary Score Card Part 3: Understanding of the Relationship Between the Drug’s Effect on Surrogate and the Disease –Range: 1 - 4 points (out of maximum of 4) –Findings: Although a strong understanding of the relationship between the drug’s effect on surrogate and the disease is helpful, a weaker showing is not a bar to Subpart H (e.g., Fabrazyme) Part 4: Strength of Clinical Evidence –Range: 1 - 9 points (out of maximum of 10) –Findings: Clinical evidence on the surrogate: This is the most important approval factor to FDA reviewers and received the most weight; however, a low score here was still not a bar to approval, see e.g., Sulfamylon and Synercid Clinical evidence of benefit: 10 of 21 precedents had essentially no substantial positive evidence of clinical benefit (and one, Sirturo, was even negative) 17September 22, 2014

18. Betaseron (multiple sclerosis) (FDA’s first Subpart H approval) Part 1: Regulatory Factors Weighing into FDA Determination –Rarity of Condition – 2 of 2 pts Designated as orphan drug on November 17, 1988 Part 2: Understanding of the Disease Process –Understanding of the disease – 2 of 4 pts Pathophysiology of MS was known to a fair degree at the time of the conduct of the pivotal trial which permitted sponsor and FDA to have general agreement on co-primary endpoint of clinical utility related to exacerbations, but the understanding of the role of the somatic measure of the putatively key causal biologic marker, MRI lesion volume, was not definitively established. Results – Example September 22, 201418

19. Betaseron (multiple sclerosis) (cont’d) Part 3: Understanding of Relationship Between MRI Lesion Volume and Multiple Sclerosis – 3 of 4 pts –At the FDA Advisory Committee, there was a strong feeling among “many experts” that “the number and area of lesions detected on MRI are tantamount to a ‘surrogate’ endpoint that predicts disease progression in MS.” The strength of this opinion actually “surprised” the key FDA reviewers. Part 4: Clinical Evidence on MRI Lesion Volume and on Reduction in Exacerbations of MS –Clinical evidence for MRI Lesion Volume – 7 of 7 pts P-value for comparison between Betaseron and placebo arms ranged from 0.03 to 0.001, depending upon the analysis used. –Evidence of reduction in exacerbations – 2 of 3 pts Primary efficacy evaluations were based on reduction of exacerbations per subject and proportion of exacerbation-free subjects. –Met the first co-primary endpoint, reduction of exacerbations per subject, with a p-value of 0.0001. –Failed on the second co-primary endpoint, proportion of exacerbation-free subjects, with a p- value of 0.094. Results – Example September 22, 201419

20. Part 1: Regulatory Factors (0-2) Part 2: Understanding of Disease Process (0-4) Part 3: Understanding Relationship Between Effect on Surrogate & Disease (0-4) Part 4: Strength of Clinical Evidence (0-10) 20

21. Epidemiological, Pharmacologic and “Other Evidence” of the Surrogate or Disease (0-4) Part 3: Rarity (0-2) Part 1: Part 2: Understanding of the Disease (0-4) 21 Total Scores (0-20) Strength of Surrogate Endpoint Evidence (0-7) Strength of Clinical Evidence (0-3) Part 4:

22. An “intermediate clinical endpoint” (“ICE”) is “a measurement of a therapeutic effect that can be measured earlier than an effect on [irreversible morbidity or mortality (“IMM”)] and is considered reasonably likely to predict the drug’s effect on IMM or other clinical benefit.” Expedited Programs Guidance at p. 18. Effects on Quantum of Evidence: May Be Reasonable Because These are Less Important with an ICE-Based Approval –Part 2: Understanding of the Disease Process 3 out of 4 lowest scores (i.e., 1 out of 4 points) were for drugs approved using ICE. Northera (NOH), Makena (preterm birth), Remicade (Crohn’s). –Part 3: Understanding the Relationship Between the Drug’s Effect on Surrogate and the Disease 3 out of 4 lowest scores (i.e., 1 out of 4 points) were for drugs approved using ICE. Northera, Remodulin (PAH), and Remicade. Intermediate Clinical Endpoints (“ICE”) September 22, 201422

23. Northera (neurogenic orthostatic hypotension) (latest Subpart H approval) Northera was approved for the treatment of “orthostatic dizziness, lightheadedness, or the ‘feeling that you are about to black out’ in adult patients.” –Primary endpoint for pivotal studies: short term dizziness (i.e., the ICE) –Primary endpoint for confirmatory Phase 4 study: longer term dizziness The primary endpoint of short term dizziness (i.e., the ICE) is the primary endpoint that will be tested in the confirmatory Phase 4 study but will need to be shown that this benefit is sustained in a chronic setting (i.e., the ultimate clinical benefit) Northera scorecard: –Part 1: Rarity of Condition (2 of 2 pts) –Part 2: Understanding of Disease (1 of 4 pts) –Part 3: Understanding Relationship between Endpoint & Disease (1 of 4 pts) –Part 4: Strength of Clinical Evidence (7 of 10 pts) ICE – Example September 22, 201423

24. Degree of regulatory uncertainty with an ICE is reduced relative to an approval based on an unvalidated surrogate Intermediate Clinical Endpoints (“ICE”) September 22, 201424

25. Part 1: Regulatory Factors (0-2) Part 2: Understanding of Disease Process (0-4) Part 3: Understanding Relationship Between Effect on Surrogate & Disease (0-4) 25 Part 4: Strength of Clinical Evidence (0-10)

26. Epidemiological, Pharmacologic and “Other Evidence” of the Surrogate or Disease (0-4) Part 3: Rarity (0-2) Part 1: Part 2: Understanding of the Disease (0-4) 26 Total Scores (0-20) Strength of Surrogate Endpoint Evidence (0-7) Strength of Clinical Evidence (0-3) Part 4: Approved via ICE

27. Part 1: Subpart H Conclusions Our analysis of Subpart H precedents demonstrates FDA’s flexibility: –Robust compliance with Factors 2, 3, and 4 cited in FDA’s May 2014 Guidance is not required –Part 1 is essentially a prerequisite Our analysis also: –Clarifies the basis for existing FDA’s Subpart H approvals –May open this pathway to more therapies for patients with serious illnesses in need of therapy –Highlights that the demands on Sponsors and FDA may be reduced if more ICE are used 27 September 22, 2014

28. Part #2: Orphan Drug Flexibility Orphan Drug Approvals

29. Purpose –To examine whether FDA exercises flexibility when reviewing applications for orphan diseases, and –If so, to illustrate the nature and scope of that flexibility Cataloguing FDA’s Orphan Drug Actions September 22, 2014 29

30. Scope of analysis: all 135 orphan drug new chemical entities approved from 1983 to June 30, 2010 (excluding those for rare cancers) For each of the 135 drugs: –Reviewed FDA’s publicly-available documents (totaling 27 boxes, primarily medical and statistical reviews) –Classified the level of “efficacy evidence” determined by FDA to be adequate for drug approval Review & classification done by author experienced in orphan drugs –Author’s analysis of 1983 law while at FDA led to 1984 and 1985 amendments, and since leaving FDA in 1987 author has been working with FDA, drug sponsors, researchers and patient advocates on orphan therapies, including key contributions on 30 of the 135 drugs in this analysis. Methods September 22, 2014 30

31. 1.“Conventional” –Evidence would satisfy the two adequate and well-controlled studies standard 2.“Administrative Flexibility” – formal FDA policy –May 1998 Guidance: “Clinical Evidence of Effectiveness” –FDAMA 115 –Subpart H of 21 C.F.R. Part 314 (accelerated approval, Fast Track) –Subpart E of 21 C.F.R. Part 312 3.“Case-by-Case Flexibility” –For each of the 58 drugs in this class, an explanation is provided since each has some unique feature Classes of Efficacy Evidence September 22, 2014 31

32. Results September 22, 2014 32

33. 90 of the 135 orphan drug approvals or 67% resulted from some exercise of FDA flexibility in applying the statutory standard for evidence of effectiveness. Results September 22, 2014 33

34. Stability of Flexibility September 22, 2014 34

35. The principal evidence of efficacy was a single Phase 3 placebo-controlled study in 371 subjects. Study included two co-primary endpoints (frequency of exacerbations per patient and the proportion of exacerbation free patients), and several secondary endpoints, including the percent change in MRI lesion area. Study met the first co-primary endpoint (p = 0.0001). Failed on the second co-primary endpoint (p = 0.094). Results on the MRI secondary endpoint were statistically significant (p = 0.0001), and were viewed by FDA as an unvalidated surrogate that was reasonably likely to predict ultimate clinical benefit. Betaseron – Interferon Beta-1b September 22, 2014 35

36. Despite having failed to meet one of the specified primary endpoints, FDA concluded that the "clinical relevance of Betaseron in reducing the frequency of exacerbations in MS is a significant benefit by itself in considering the approval of this product. However, the review of the patient MRI data also played a significant supportive role in the overall evaluation [of the product] and may be an important surrogate in the pathology of MS." FDA Summary Basis of Approval at p. 13 attached to July 23, 1993 Approval Letter from Dr. Janet Woodcock. Betaseron – Interferon Beta-1b September 22, 2014 36

37. May 1996 approval for treating patients with relapsing forms of multiple sclerosis. Avonex was approved on the basis of a single study (and without reliance on the efficacy for the previously approved interferon beta drug for MS) in which the primary endpoint results are not “very persuasive” (that is, not less than a p of 0.01). Primary endpoint, time to progression, p-value of 0.02. Secondary clinical endpoints were generally significant: –Change in Expanded Disability Status Scale (p = 0.006), –Number of exacerbations (p = 0.03), –Percentage exacerbation free (p = 0.10, not significant) and –Annual exacerbation rate (p = 0.04). –The secondary MRI endpoints were number of lesions at end of year 1 (p = 0.02), at end of year 2 (p = 0.05), T2 lesion volume at end of year 1 (p = 0.02) and at end of year 2 (p = 0.36). Avonex – Interferon Beta-1a September 22, 2014 37

38. December 1995 approval for treating amyotrophic lateral sclerosis (ALS or Lou Gehrig’s disease) Two studies, both of which failed to hit their primary endpoints of time to tracheostomy or death according to the prespecified analysis in these placebo-controlled, randomized trials. The primary endpoint results by the prespecified analysis in these two trials were p-values of 0.076 and 0.12. In both cases, FDA salvaged each trial by permitting a post- hoc analysis that in each case yielded a p-value of exactly 0.05 in each trial, not less than 0.05. Rilutex – Riluzole September 22, 2014 38

39. In addition, there had been one interim analysis in study 301 with an alpha “cost” of 0.001 so that the hypothesis was being treated to determine not if it were less than a p-value of 0.05 but less than 0.049. It is also noteworthy that both trials had numerous secondary endpoints of muscle strength and neurological indices and not only did these not show any statistically significant separation between placebo and drug arms, there was hardly any numerical difference between the groups on these indices. Finally, in both studies, “there was no statistical significance in mortality at the end of the study.” The FDA medical reviewer noted that the apparent improvement in survival occurs early in each study period and the Kaplan- Meier curves came nearly together at the end of the study period, so that the FDA medical reviewer further observed: “Of course, the unanswered questions are whether the Kaplan-Meier curves eventually meet and follow a common path therefore [or] potentially the curves could cross with cumulative survival being worse on drug after 2-3 years.” Rilutex – Riluzole September 22, 2014 39

40. November 2008 approval for treating Lennox-Gastaut Syndrome. Single placebo-controlled randomized study (n=138) which was robustly statistically positive on all three co-primary endpoints of seizure activity (p-values of 0.0015, <0.0001, 0.0041). However, Institutional Review Boards may not have found it unethical for a second study to be conducted. Therefore, the “statistically very persuasive finding” in this one trial may not have satisfied the strict application of FDA “single study” policy in its Evidence Guidance (See Section II.C.3). However, this sponsor also conducted 2 large studies of this drug in a prevalent disorder, “partial seizures,” and while FDA did not find the efficacy evidence in these 2 “partial seizure” trials adequate to warrant the drug’s approval for that prevalent indication, FDA found that the efficacy evidence in these 2 studies provided “additional support” for the orphan indication as noted in the final sentence of the conclusionary paragraph by the medical reviewer on the efficacy evidence for the Lennox-Gastaut use: “The agent is additionally supported by the evidence from the partial seizure trials which indicate anticonvulsant activity.” Banzel – Rufinamide September 22, 2014 40

41. August 2008 approval for treating chorea associated with Huntington’s disease. In the larger of two efficacy trials (n=30), the primary endpoint of change from baseline in the Chorea Score (a subset of the Motor Assessment Scale of the Unified Huntington’s Disease Rating Scale) for the average of weeks 9 and 12 was statistically significant (p = 0.0001). In addition, the Clinical Global Impression of Change (key secondary endpoint) was also statistically significant. However, the primary endpoint of the smaller, staggered withdrawal study (n=18) had only a trend suggestive of efficacy, but was not statistically significant for its primary endpoint, although the magnitude of change was similar to that seen in the larger trial. Xenazine – Tetrabenazine September 22, 2014 41

42. August 2009 approval for infantile spasms FDA showed considerable flexibility: –Undertook its own post hoc analysis of data using different endpoints –Accepted data from open-label studies –Ignored the requirement for predefined hypotheses –Showed willingness to rely on post-marketing studies Vigabatrin – Sabril September 22, 2014 42

43. 3 controlled studies formed approval basis, despite limitations, including: –Lack of predefined protocol –Interim statistical plans –Amendments to increase n size during trial –Questions of completeness of blinding In the single key pivotal study, the primary endpoint was change in average spasm frequency as measured over a 2-hour period (p=0.562). –A secondary endpoint included a 24-hour observation window that showed statistical significance (p=0.03). –FDA accepted the 24-hour secondary endpoint finding as “substantial evidence of effectiveness” despite being a secondary endpoint in a single trial which robustly failed on its primary endpoint. Vigabatrin – Sabril September 22, 2014 43

44. January 2010 approval as potassium channel blocker indicated to improve walking in patients with multiple sclerosis (MS). FDA relied on conventional efficacy evidence: –2 adequate and well-controlled studies, each of which hit its prespecified primary endpoint (walking speed) by its prespecified analysis. This proved that the drug arm’s improved walking speed over the placebo arm was not due to chance. –However, the key question here was: is this statistically significant difference in walking speed also clinically meaningful to patients? Ampyra – Dalfampridine September 22, 2014 44

45. Ampyra – Dalfampridine 45 Figure 1: Average walking speed change (%) from baseline during the double-blind phase of Trial 1 P=0.001p=0.001 p<0.001 p=0.005 p=0.258 p>0.99p>0.999

46. Ampyra – Dalfampridine 46 Figure 2: Average walking speed change (%) from baseline during the double-blind phase of Trial 2 p=0.008 p=0.002 p<0.001 p=0.333 p>0.99p>0.999

47. Historically, FDA has regularly exercised regulatory flexibility in approving therapies for Americans with rare diseases This analysis aides FDA’s regulation of therapies for persons with rare diseases Conclusions on Orphan Drug Act Analysis September 22, 2014 47

48. Analysis will summarize and score the approval of the 28 orphan drugs approved between July 1, 2010 - June 30, 2014 Upcoming Update to Orphan Drug Act Analysis September 22, 2014 48

49. Part #3: Rise of Patient Power Rise of Patient Power

50. No reference to patients in Pure Food and Drug Act of 1906 or subsequent legislation FDASIA added patient perspective (July 2012) –FDA must “implement strategies to solicit the views of patients... and consider the perspectives of patients during regulatory discussions.” FDASIA § 1137 –This could include FDA hiring a patient representative as a Special Government Employee (SGE) to be a part of the internal FDA review team. Patients as Part of the Internal FDA Review Team September 22, 2014 50

51. Ms. House is President of the Acid Maltase Deficiency Association. Speaking at FDA’s Rare Disease Patient Advocacy Day, March 1, 2012. After the Myozyme review, FDA medical reviewers told me that they learned from Ms. House that being stable for a person with a uniformly progressive disease is a HUGE benefit. Patient perspective is a key factor for evaluating both safety and efficacy. September 22, 2014 51 Tiffany House – A Patient Representative

52. FDA Patient-Focused Drug Development Initiative September 22, 2014 52 Diseases Selected for FY2013 – 2015:

53. In February 2013, FDA asked for comments on “Structured Approach to Benefit-Risk Assessment in Drug Regulatory Decision-Making, Draft PDUFA V Implementation Plan”. This is FDA’s plan for involving patients more in FDA’s pivotal benefit-risk decisions. September 22, 2014 53 Structured Approach to Benefit-Risk Assessment

54. Registries may lead to natural histories, of sufficient quality, to have regulatory consequence –Natural history as a historical control arm (Lumizyme) –Using patient experience in period prior to baseline as a control arm for a clinical trial (iPlex) –Establish relationship of genotype and phenotype to disease progression (Kalydeco) –Helps define what is “clinically meaningful ” Giving Voice to Patients Beyond FDASIA: Creating Natural Histories and Registries September 22, 2014 54

55. Charge: Develop consensus on characterization of diseases –If doctors agree on a uniform way to capture disease data, useful natural history data can be developed In a meeting with Dr. Temple earlier this month, he expressed frustration/dismay that the medical community and patient advocacy groups have not generally developed systematic methods to uniformly measure progression of diseases. –This was in the context of a muscular disease. –He said: “Why don’t those caring for such patients see that it is critical to measure the same clinical and laboratory endpoints in the same order (e.g., to minimize the role of fatigue) using the same tests?” Giving Voice to Patients Beyond FDASIA: Creating Natural Histories and Registries (Part 2) September 22, 2014 55

56. Conclusions Three lessons: 1.FDA can be flexible, especially with respect to therapies for serious diseases when the therapies truly address unmet medical need. 2.Engage patients in trial design and all drug development activities. 3.Shoulder your responsibilities to – not only provide best quality of care for each patient with rare or serious disease – but to gather key health information in systematic, uniform way to contribute to “natural history” understanding of the progression of the condition. Thank you FJS@hpm.com