Medical Device Daily Washington Editor

WASHINGTON – The idea of parallel review of devices by FDA and the Centers for Medicare & Medicaid Services seems to have died on the vine, but a session held here Tuesday by the Food and Drug Law Institute (FDLI; Washington) suggests that observers of both agencies are divided on whether these two entities can realistically collaborate meaningfully on even less-contentious issues.

The split also extended to how observers assume FDA and CMS would make use of comparative effectiveness data, given that their data needs differ substantially.

Michael Gaba, a partner in the DC law firm of Holland & Knight, said he sees "a much greater push" at the Department of Health and Human Services to bring leaders of the department's agencies together. "This is part of a growing effort at HHS to get the agencies to collaborate," he said.

Gaba said that much of the collaboration between CMS and FDA would be on risk assessment functions. One of the stumbling blocks at present, he said, is that "FDA depends on data reliability" from randomized, controlled clinical trials (RCTs), which do not "necessarily reflect what goes on in the real world," a point of great interest to CMS, which relies more on data applicability than data reliability.

All the same, such requirements are going up. Since the passage of the Food and Drug Administration Amendments Act of 2007 (FDAAA), the agency has issued 21 letters to drug and device makers mandating post-market studies and also has approved 13 risk evaluation and mitigation strategies (REMS) for drugs and devices.

Another feature of FDAAA is the Sentinel system, launched this year, which is a response to episodes such as the problems associated with Vioxx, and which will enable FDA to be much more proactive "at heading these things off" earlier, Gaba said.

"I see a comparative effectiveness (CE) institute as an entity that can further inform FDA on the risk side," he said. The idea is for a CE agency to invoke "simple real-world studies of new technologies," he added, although the meaning of the word "simple" in this context may prove nettlesome.

On the other hand, Gaba did not advocate the use of the UK's National Institute for Clinical Effectiveness (NICE) as a model, inasmuch as NICE is "known as a choke point for getting new technologies to market."

He posed the question of the day in a CE context. "Should there be interagency collaboration on risk assessment? Reasonable minds will differ," but CMS has expanded its data collection requirements, he said, and "risk is part of that analysis."

Gaba added that the "framework [for collaboration] is being built now" via a data-sharing protocol overseen by HHS's assistant secretary for planning and evaluation.

One of the details to be worked out is the precise mission of a CE institute – i.e., whether it will collect data on relative risk only or also extract and analyze data on the absolute risks that are FDA's principal interest. If the latter, FDA could put some of the cost of its risk assessment work on the CE institute, but this dilemma is not yet resolved.

Scott Gottlieb, MD, of the American Enterprise Institute (AEI; Washington), was not so optimistic about collaboration. He said the current approach "reduces transparency," citing the adverse events related to erythropoietin stimulating agents (ESAs). Gottlieb said CMS's coverage decision was based on "the expertise and nuanced judgment of FDA," but post-market events turned that coverage decision sour.

Gottlieb seemed to see a fork in the road. "I'm of the opinion that the agencies are growing farther apart, a trend that's going to continue," he stated. Part of this is engendered by "an expertise gap" which is in turn due in part to a change in mission at CMS, which "never thought of itself until recently as a public health organization."

FDA and CMS also employ "very different evidentiary standards," Gottlieb noted. As an example, he said the coverage with evidence development (CED) mechanism, "a process set up by Dr. [Mark] McClellan to allow broader reimbursement," ran afoul of FDA because of perceptions that "it would obviate the FDA approval process" for expansions of indications.

"FDA raised a lot of concerns that CED" would allow use of prospective registry data to support reimbursement for expanded indications without the support of FDA-style clinical trial data, he said.

"FDA is pushing toward increasing information on drug labels," Gottlieb said, but CMS wants to make greater use of post-market data to guide its coverage decisions. Comparative effectiveness, he observed, does not resolve the dilemma because "the kinds of clinical constructs they're talking about developing ... do not reach the level of evidence needed" to change a drug label.

FDA is institutionally inclined to prefer reimbursement policies that will stem off-label use, Gottlieb indicated, which means "the agencies are increasingly at odds with each other," a source of tension that will likely persist. "To the extent that CMS continues to make determinations based on lower evidence standards" than FDA requires, "it is going to be important to provide transparency, but also clinical discipline." He said clinical expertise is tough to come by at CMS, which currently employs "probably fewer than 30 clinicians" and no oncologists.

Still, Gottlieb said that any attempt to bring more expertise in house would be "a fool's errand," remarking that a smarter approach would be to "infuse CMS with some kind of process ... where you have external advisory committees." This would help with clinical discipline and transparency in one stroke.

"I'm not as optimistic that you'll see broad collaboration" between FDA and CMS, Gottlieb said.

During the question-and-answer period, one attendee asked how CMS will grapple with the least-costly-alternative (LCA) standard, given the outcome of Hayes v. Leavitt, a case in which a U.S. District Court ruled that CMS is not statutorily empowered to use cost as a consideration.

Gaba seemed to suggest that a decision to appeal might not be taken up until after the Obama administration takes over at HHS. If CMS loses that appeal, the problem falls into Congress's lap, he said. "I don't think anyone on Capitol Hill wants to be on record as voting that cost" be a factor.

Panel moderator Kathleen McGuan, a partner at the DC law firm of Reed Smith, said the decision does not necessarily apply across the nation, but another such outcome in court probably would force a showdown. Until then, "I don't think the agency will feel as boxed in" as is often reported, she said.