By MICHAEL MCCAUGHAN

FDA officials are eager to experiment with “master trial” protocols to allow simultaneous studies of multiple targeted oncology therapies as an alternative to traditional, single-agent registrational studies.

An issue brief proposing a design for a multi-drug registration trial in metastatic non-small cell lung cancer was discussed during a Friends of Cancer Research/Brookings conference on cancer clinical research November 14. Based on the enthusiastic comments by FDA officials, notably Center for Drug Evaluation & Research Director Janet Woodcock, the agency is clearly ready and willing to work with sponsors to get such a trial up and running.

There are still important issues to be ironed out, including the key question of who exactly will be the sponsor of the study. However, FDA sounds eager to work with potential sponsors. For manufacturers, the evolving model offers some important potential advantages – but also the significant risk associated with losing control over development projects at the pivotal trial stage.

The Brookings/FOCR session has evolved into an important forum for FDA, industry, patient advocates and researchers to work together on shaping ideas that evolve into new policy. The conference also provides a forum for FDA officials to discuss concepts with each other as they wrestle through the implementation process for FDASIA.

Other panels in the meeting focused on concepts for implementing the new “Breakthrough Therapies” designation (““Breakthrough Therapy” Development Speed May Be Tempered By Manufacturing Hurdles” — “The Pink Sheet,” November 26 2012 12:00 AM) and for fleshing out an “enhanced” Accelerated Approval process (“Redefining Accelerated Approval Criteria Would Expand Pathway’s Use, Panel Says” — “The Pink Sheet,” November 26 2012 12:00 AM)

Enthusiastic Response from FDA

Unlike “Breakthrough Therapy” and Accelerated Approval changes, the master protocol concept is not something FDA is under legislative obligation to address.

However, it is clear that the enthusiasm for the concept within FDA runs deep.

FDA has been a vocal supporter of the I-SPY breast cancer trial (a Phase II study to screen multiple therapies sponsored by the Foundation of the National Institutes of Health). (“Biomarkers Consortium Launches Adaptive Trial To Test Targeted Breast Cancer Drugs” — “The Pink Sheet” DAILY, March 19 2010 12:00 AM)

In addition, Woodcock in particular has been an advocate for dramatically enhancing the clinical trial infrastructure in the US. (“The Future of Drug Regulation: An Interview With Center for Drug Evaluation & Research Director Janet Woodcock” — The RPM Report, July 1 2011 12:00 AM)

Her enthusiasm for using a multi-agent registrational trial model to advance that goal was impossible to miss. While not a panelist for this portion of the workshop, Woodcock took the microphone during Q&A and turned to address the large audience rather than the panelists.

“We need to start enrolling cancer patients in this country into clinical trials,” she said. There has to be a way to “get them to a treatment that is more likely to benefit them” rather than simply giving them standard of care—or the opportunity to enroll in a single agent study where their likelihood of expressing the right molecular target is low.

In addition, a stronger US trial infrastructure would help assure that new drug applications are less likely to be based overwhelmingly on overseas studies, where geographic variations could complicate analyses of potential benefits in the US.

Key Question: Who is the Sponsor

Woodcock made clear her view: “a trial should be set up.”

However, she added, the key question is: who will be the sponsor? All parties agree it must be run by an independent third-party, but there were different perspectives on the right type of independent entity to use.

Woodcock argued that the host of the trial not be an academic medical center. Instead, it needs to be a “free standing” trial (like I-SPY) that can reach patients in the community setting. “Because that is where they are.”

The discussion of the appropriate sponsor clearly struck a nerve among some audience members. Industry representatives though a CRO might be a good choice. Others wanted the National Cancer Institute’s Cooperative Groups involved. FNIH representatives noted their experience with I-SPY.

Sponsors Face Trade-Off

For drug developers, the master protocol model offers intriguing advantages and one clear disadvantage.

On the plus side, a master study protocol should significantly reduce Phase III costs, since the trial infrastructure is shared across many agents. In particular, the major inefficiency in recruitment would be addressed: a targeted therapy that applies to a small portion of the NSCLC population would typically require many patients to be screened to find the 5% or 10% with the right marker for the study.

In the master protocol model, the up-front screening would be used to direct patients into many different arms depending on which biomarkers they express.

The immediate efficiency gains from a shared trial are important, but there is another, long term implication that is worth considering as well: If many late stage patients enroll in broad studies covering a range of oncology indications, that could start to change the dynamics of the conversation around the cost of cancer care.

Right now, many late stage patients are treated with a “standard of care” that, as one clinician at the conference put it, delivers horrible outcomes at a high cost. Broader clinical trial participation would shift much of the “standard of care” cost out of the traditional payor setting, perhaps helping sponsors support premium prices for new targeted agents.(“Ending “Buy-And-Bill” in Oncology: The Question is How, Not If” — The RPM Report, October 24 2012 12:00 AM)

However, the tradeoff is inherently difficult: a loss of control of the development of a new agent.

By design, the master trial protocol relies on independent third-parties to evaluate candidates for inclusion in the protocol and to administer the study. As FDA’s Office of Hematology & Oncology Products Director Richard Pazdur put it, industry is “reluctant to give NMEs to a third party” because “you lose control of it.”

One element of the proposed trial in lung cancer may help with that dilemma: the working group members who drafted the proposal agreed that it probably would not be the type of trial used for a highly effective novel agent that might qualify for “Breakthrough” status. Instead, it would more likely be an option for drugs that showed promise but not dramatic early efficacy, or for agents that failed in a study but seemed to show promise in a biomarker identified subpopulation.

Head-to-Head Comparisons Inevitable

Still, the logic of the multi-agent trial design means that manufacturers may find their products included in an arm with several similar agents—making head-to-head comparisons nearly inevitable.

In fact, the FDA representative on the panel—Office of Biostatistics Director Lisa Lavange—focused her comments on the idea that a truly independent, third-party run protocol would allow sponsors to leverage control patients by having multiple agents targeting the same pathway studied against the same control group.

Lavange stressed that the analysis would not be comparative; each agent would be measured versus standard of care. But the number of patients enrolled could be sharply reduced and trials completed much faster by sharing a control group.

CDER Deputy Director for Clinical Science Robert Temple asked Lavange specifically about that concept, and especially her remark that the trial would not produce head-to-head comparisons. “I understand why you said that,” Temple said, since head-to-head comparisons “make everyone nervous.” But, “if you leverage the control group, you are comparing.”

Lavange responded that she wanted to “tread with caution.” Temple again acknowledged the need to reassure sponsors, but added “you really could” compare head to head in the proposed design.