Expedited Approval Pathway Concept Draws Strong Support But Different Proposals

Debate at a recent cancer research meeting reflects the challenges that lay ahead, and the diverging viewpoints that will have to be balanced, in creating any new approval mechanism.

Growing momentum for finding new ways to speed development and approval of drugs for serious diseases has led to draft legislation in the Senate and efforts under the Innovation Initiative at FDA. Yet, debate at a recent cancer research meeting reflected the challenges that lay ahead, and the diverging viewpoints that will have to be balanced, in creating any new approval mechanism.

At the Nov. 10 Conference on Clinical Cancer Research in Washington, D.C., a panel of agency representatives and stakeholders collectively proposed an expedited development pathway model for drugs with early evidence of large treatment effects. Under this approach, a randomized Phase IIb screening trial could serve as the basis for full approval.

One proposal would allow randomized Phase IIb screening trials to serve as the basis for full approval.

The group’s consensus approach was reached after a good deal of disagreement and plenty of compromise among the panel members.

However, Center for Drug Evaluation and Research Director Janet Woodcock, who was part of the panel, also presented separate proposals for approval of “breakthrough” cancer therapies. In her view, full or accelerated approval could be based upon single-arm studies with adaptive designs in which the investigational agent demonstrates high efficacy, compared to historical controls, in either durable complete responses or overall response rate.

Yet another panelist, Wyndham Wilson of the National Cancer Institute, weighed in with his own proposal premised upon loosening of requirements under the current accelerated approval mechanism.

The multiple proposals reflect strong support for creating a new system that speeds development and approval time for drugs that demonstrate strong treatment effects early on. However, the specifics for how this should be accomplished, and the threshold level of efficacy and safety data needed, seems likely to be the subject of extensive debate both inside and outside the agency.

Wanted: Faster Drug Development

One of the key elements of FDA’s innovation report, released in October, is development of an expedited approval pathway for drugs that show “exceptional promise” in early studies. The agency intends to hold a series of scientific meetings to discuss issues related to creation of a new pathway and will publish a draft guidance based on the outcome of these meetings (Also see "FDA Biomedical Innovation Initiative Draws On Familiar Themes" - Pink Sheet, 10 Oct, 2011.).

With a nod to the length of time this outreach process is likely to take, the agency said it will take two more immediate and related steps toward expedited drug development. One of these is issuance of a draft guidance on enrichment strategies in clinical trials – a document that has been in the works for a number of years.

CDER also will publish draft guidance on use of pathologic complete response as a surrogate endpoint for accelerated approval in primary high-risk breast cancer. FDA said the document will describe a “relatively seamless pathway” to accelerated approval that could be followed from a multi-drug screening trial, such as the ongoing I-SPY 2 trial in women with high-risk breast cancers (Also see "Biomarkers Consortium Launches Adaptive Trial To Test Targeted Breast Cancer Drugs" - Pink Sheet, 19 Mar, 2010.).

However, these FDA-initiated efforts have not quelled external pressure to speed up approvals for promising drugs.

Legislation drafted by Sen. Kay Hagan, D-N.C., would establish “progressive” and “exceptional” approval pathways aimed at facilitating the development and expediting approval of drugs that provide a meaningful advancement in the treatment of serious or life-threatening diseases (Also see "FDA “Progressive Approval” And “Exceptional Approval” Pathways Possible Under Senate Bill" - Pink Sheet, 14 Nov, 2011.).

The draft legislation, which could be attached to the Prescription Drug User Fee Act reauthorization legislation that must pass Congress next year, embodies a number of the Biotechnology Industry Organization’s proposals for improving FDA’s regulatory processes.

BIO’s “big ideas” called for drug approvals based on Phase II data if a product fills an unmet medical need, significantly advances the standard of care, is targeted, or is already approved by the European Medicines Agency or another mature regulatory agency (Also see "BIO’s "Big Ideas" For FDA Reform Are Too Big For PDUFA" - Pink Sheet, 4 Jul, 2011.).

The Hagan draft is short on specifics about how the progressive and exceptional approval processes would actually work and the threshold level of evidence needed to support approval under these routes. The absence of such details could give FDA considerable flexibility in how it decides to implement any new approval pathway.

FDA, however, may wish to avoid the legislative route altogether.

FDA can and has shown flexibility in encouraging sponsors to adapt their clinical programs in the face of strong early efficacy results. In the case of Roche’s melanoma treatment Zelboraf (vemurafenib), for example, FDA urged the sponsor to modify the ongoing Phase III trial and conduct earlier interim analyses. Zelboraf received full approval after a review lasting less than four months (Also see "Zelboraf Approval Hastened By FDA Officials Impressed With Early Efficacy" - Pink Sheet, 22 Aug, 2011.).

The proposals aired at the cancer research conference – concepts that were either developed with agency input or came directly from the agency itself – offer insight into FDA’s thinking on how to speed drug approvals but do not appear to require new legislative authority.

The Consensus Proposal

Sponsored by the Engelberg Center for Health Care Reform at the Brookings Institution and the Friends of Cancer Research, the conference brought together representatives from FDA, NCI, academia, industry and the patient community to identify consensus approaches for new, expedited development pathways for drugs that show substantial efficacy in early development.

Conference on Clinical Cancer Research Panel On Development Paths for New Drugs With Large Treatment Effects Seen Early
Panel Member	Title/Institution
Thomas Fleming	Professor, Biostatistics, University of Washington
Edward Korn	Mathematical Statistician, Biometric Research Branch, NCI
Grazyna Lieberman	Director, Biostatistics, Genentech
Jane Perlmutter	President and Founder, Gemini Group
Mikkael Sekeres	Director, Leukemia Program, Associate Professor of Medicine, Cleveland Clinic*
Rajeshwari Sridhara	Director, Division of Biostatistics V, FDA/CDER
Wyndham Wilson	Senior Investigator, Chief, Lymphoma Therapeutics Section, NCI*
Janet Woodcock	Director, CDER/FDA

*Current member, FDA Oncologic Drugs Advisory Committee

The panel’s work included multiple conference calls leading to the development of an issue brief presented at the meeting.

The finished proposal seeks to limit reliance upon single-arm studies for purposes of gaining accelerated approval. This approach should be used only in exceptional circumstances, ideally while confirmatory trials already have begun accruing, the brief says.

The panel’s preference for moving away from single-arm studies mirrors that of FDA’s Oncologic Drugs Advisory Committee. At a February 2011 meeting, the committee said sponsors seeking accelerated approval should conduct randomized trials except for indications with a small patient population or where drug activity is high (Also see "Accelerated And Accompanied: Oncology Drugs Should Use Randomized Trials, FDA Panel Says" - Pink Sheet, 8 Feb, 2011.).

However, the CCCR panel favored a randomized trial approach even when early evidence suggests treatment effects are high. The group suggested a small but well-conducted randomized Phase IIb study capable of confirming early results that suggest large treatment effects could support full approval.

Under this scenario, sponsors who see extraordinary effects on overall response rate and duration of response in a single-arm Phase I trial would move into a small, randomized Phase IIb study. “The purpose of the trial is to demonstrate a large treatment effect in a small number of patients, while maintaining the same statistical significance currently used in trials that seek small benefits,” the panel’s brief states. “This trial design maximizes efficiency in bringing highly active drugs to patients quickly, and in resources used to explore efficacy and safety.”

A Phase IIb trial encompassing 120-150 patients would be of sufficient size to support full approval if large treatment effects are seen, but small enough to serve as a screening trial before Phase III if the treatment shows only moderate efficacy.

“This can be part of the agreed-upon study design upon study initiation,” the brief states. “If results are poor, fewer patients will have been exposed, and the most ineffective treatments will be screened out.”

The panel’s proposal is aimed at screening out ineffective therapies and screening in effective ones.

Panel member Thomas Fleming, a University of Washington biostatistician who frequently serves on FDA advisory committees, said the goal of the small randomized Phase IIb study is to screen out ineffective therapies and screen in effective ones with a high probability of statistical confidence.

Although the proposal calls for small Phase IIb trials, “if they are properly done with high quality and they’re showing very strong effects on registrational endpoints, shouldn’t this be accepted or considered as a potential basis for registration?” Fleming asked.

A trial that shows only moderate effects does not necessarily mean it was a failed study, Fleming hastened to add. Rather, such results could suggest the agent is worthy of further exploration in a larger trial adequately powered to detect more moderate treatment effects.

The brief describes numerous issues under this model that would warrant further clarification between a sponsor and the agency. These include acceptable endpoints for the Phase IIb trial, what results (including hazard ratio and p-value) would be considered sufficient for registration purposes, and whether a Special Protocol Assessment would be required.

Woodcock’s Advice: Stop And Think

The process of arriving at the panel’s consensus proposal was highly contentious, several of the group’s members said. Evidence of this disagreement could be seen in the existence of alternative proposals proffered by some panel members.

CDER’s Woodcock said the consensus proposal “produces a very efficient development program that could be used for a wide variety of endpoints and actually lead to full approval.” Yet she presented two alternatives that could allow for full or accelerated approval based upon single-arm studies demonstrating large improvements over currently available treatments.

Though many metastatic solid tumors do not have good options for remission, cure or durable stabilization, industry and FDA should be prepared for groundbreaking therapeutic cures, Woodcock said. The current system is not adequately prepared with a development path for something that comes along and is spectacular.

“What are you going to do if you get something that obviously appears to be so much better than anything going before it,” she asked rhetorically. In metastatic solid tumors, if a treatment shows a large amount of complete responses in Phase I, “we should all stop and think about what to do next. We shouldn’t just say well we’re going to do Phase I, and we’re going to do Phase II, then we’re going to do Phase III.”

Rather than proceeding down this traditional development route, the sponsor should consider either adding a significant number of patients at the same dose in Phase I or rapidly opening a multicenter single-arm trial, Woodcock said.

“The purpose of this trial would be to refine the estimate of percentage complete response by adding sufficient numbers of patients, and also following these individuals to assess durability in a fairly large number of people,” Woodcock’s proposal states. “This would address the criticism that the initial estimate of the treatment effect in Phase I trials is usually an overestimate, and also begin to get a handle on the crucial issue of relapse.”

If the durable complete response rate is actually much lower than was initially seen, then the sponsor would proceed to a randomized trial. “If the durable CR rate remains well superior to any historical control, then the development program should also focus on evaluating all major toxicities and determining if they outweigh the probable benefit, as well as establishing the length of CR and if possible the reasons for relapse,” the proposal states.

“I would think that if you had a long enough, durable enough tail of complete responses over a substantial amount of time, that we, FDA, could grant a regular approval based on that finding,” Woodcock told the conference. “That’s what I’m proposing for debate.”

The CDER director’s second proposal is similar in concept but premised upon robust effects on overall response rates.

“What if you had an investigational drug that resulted in a very high response, well over that obtained with current therapy, but not necessarily this high rate of complete responses where the tumor went away,” Woodcock said. “Then you do the same thing, and you expand the cohort significantly to get away from the complaints that you’re only generalizing from a very small subset of pretty healthy people with the disease. You enter a lot more people, you follow them as well. And you again look at durability of this response and how long did it last.”

This pathway could lead to accelerated approval, with the need to follow a large cohort of patients to eventual relapse.

Woodcock acknowledged problems inherent in determining the appropriate historical control, particularly for a biomarker-identified population.

Woodcock acknowledged one problem inherent in her proposal – the need to use an appropriate historical control against which an investigational drug’s efficacy is gauged. “Who is the historical control? If it is going to be a targeted therapy, it would have to be the same people who would be entered into the trial and people with the same type of prognosis,” she said.

Finding a historical control to use for a targeted therapy could be particularly problematic given that such controls generally predate the identification of a biomarker, said audience member Richard Schilsky, University of Chicago and chair of the American Society of Clinical Oncology’s government relations committee.

Woodcock’s proposal suggests a control group could be established either by previously identified and followed cohorts of patients with the same biomarker results, or by assembling a concurrent, though non-randomized, control group at sites not involved in the study at hand.

ODAC’s Chairman Weighs In

Yet another proposal was offered up by Wilson, senior investigator and chief of the lymphoma therapeutics section at NCI and current ODAC chairman.

Wilson said he disagreed with expedited development measures that involve looking at earlier or more limited endpoints for approval. “I actually don’t see this as the solution to the problem to actually getting the drugs out there,” he said, adding that such an approach results in limited efficacy and safety data and discourages further study.

Rather, Wilson chose to focus his remarks on how the accelerated approval process could be modified to expedite drug development.

Accelerated approval is available for drugs intended to treat serious diseases and fill an unmet medical need. This latter criteria, Wilson argued, warrants re-examination.

“As more and more targeted drugs are coming up, it is becoming more and more difficult to do a clinical trial where you actually show that it is fulfilling an unmet need,” he said. This requirement is “very artificial,” he added, because although early trials may show drug activity, the agent may be best suited in other settings, something that will not be known until post-marketing studies are conducted.