Thursday, April 03, 2008

Waiting for Sentinel: The Active Surveillance Revolution is Coming Soon

It sounds like the Food & Drug Administration is about ready to unveil its plans for a national active surveillance network, dubbed Sentinel. (Though hopefully that will be the only resemblence to the meanies in The Matrix.)

This long-discussed project was given a big push by the drug safety law enacted in 2007. The potential implications are truly revolutionary for biopharma companies—displacing the drug sponsor from its historic role as the center of the information flow about its own products.

The potential implications are one thing. The real implications will depend heavily on how the program is implemented—and by whom.

The new law directs FDA to set up a public-private partnership to help build an integrated national database of pharmacy claims, and develop the tools for analyzing signals to make regulatory decisions.

The law doesn’t say what role the pharmaceutical industry should or should not have in the system, how the agency should decide what signals to communicate publicly, or whether the claims data should be used for more than just safety assessments and made available to support other forms of research or even formulary decisions.

It is a safe bet that former Commissioner Mark McClellan will be involved; he championed the idea during the latter stages of the FDA legislative debate. In fact, the agency would really like to hand at least some of the project off to the newly formed Reagan-Udall Foundation chaired by McClellan but Congress has complicated that by prohibiting FDA from transferring funds to the Foundation.

The law does include some timelines and deliverables that mean the agency needs to get moving—something that regulators are eager to do anyway.

“Sentinel is a program you will be hearing about very soon,” FDA Commissioner Andrew von Eschenbach told the Food & Drug Law Institute annual meeting March 26.

How soon? Well, all we can say for sure is that it will be a month sooner than it was the last time we heard von Eschenbach discuss the idea, during an address to the National Press Club at the end of February.

But we do know a bit about where the agency is likely to be heading. Here are some “principles” that Associate Commissioner for Policy Jeff Shuren—who is the agency’s point person on Sentinel—outlined during Windhover’s FDA/CMS Summit in December. Shuren stressed that these are his personal thoughts.

(1) Scientific Credibility: Sentinel must win the “trust and confidence of patients, medical community, industry and other stakeholders.”

(2) Integrity: “The management structure and data analysis of this new system could be insulated from undue influence, whether by stakeholders or the political process.”

(3) Inclusiveness: “Stakeholders should have the opportunity to provide input into the standards and processes used by the system.”

(4) Transparency: “Protocols and study results should be made available to the public.”

(5) Privacy Protection and Data Security: This is “of paramount importance.”
Those principles seem hard to argue with.

Not everything is so simple, however. Shuren raised just a couple of the tough questions FDA and other stakeholders need to wrestle with.

For instance, there is the question of standards for interpreting safety signals in the Sentinel system. “When and with what level of confidence, can conclusions be drawn from the data?” Shuren asked.

“Even if we do not develop standards for such decision making—and it would be truly a challenging and controversial undertaking—we at least need to realize that there are likely different levels of confidence or certainty under which to make decisions regarding the treatment of a single patient versus formulary or coverage determinations versus decisions to change a product’s labeling," Shuren said. Which in turn raises the question: what access should FDA grant to the Sentinel data for other entities, like the Medicare agency or private payors, to support coverage decisions?

Those are the types of questions where most in industry would agree the right answer is worth waiting for.

No comments: