FDA Seeks Dedicated Meta-Analysis Team As Part Of PDUFA V User Fee Funding

FDA wants to create a meta-analysis team to conduct its own studies with funding from the next round of application user fees.

The new team would give the agency the ability to verify and respond to those meta-analyses published by academics and others, a move that could avoid some of the public relations problems like those experienced with Avandia.

FDA took some well-documented hits after a meta-analysis was published by researchers documenting cardiovascular safety issues with the diabetes drug. The new team would create more agency capacity to analyze the studies other researchers conduct and look at the proprietary data it holds to determine if safety signals have been identified.

The FDA meta-analysis group also would help develop standards for the conduct and use of meta-analyses, and eventually guidance on best practices.

Agency officials plan to develop a proposal that will be considered during future Prescription Drug User Fee Act reauthorization meetings, according to recently released minutes of a Nov. 22 negotiation with industry.

The agency has stated there was no consensus as to best practices for meta-analyses related to drug safety questions.

FDA's Meta-Analysis Proposal

The agency is developing a plan to spend user fee money on:

• Public meeting to talk about science and methods for such studies

• Guidance on best practices

• Dedicated review team to handle such studies

With more clinical data about drugs becoming publicly available, the agency likely will need to respond to more published meta-analyses raising safety questions about products. The agency said the work requires extensive evaluation and sometimes re-analysis of data, according to minutes of a Sept. 27 PDUFA negotiation with industry.

The agency said in the Nov. 22 minutes the proposal for a meta-analysis team also includes plans for a public meeting to talk about science and methods, as well as areas for standardization. Industry and the agency agreed in the minutes standards are needed to measure the quality of meta-analyses that are released.

FDA plans to use the comments to write a draft guidance on best practices that would be available for comment during PDUFA V, which is scheduled to begin in 2012.

Industry officials appeared supportive of meta-analysis standards, but unsure of how the studies should be applied. Robert Clark, Pfizer vice president of worldwide regulatory strategy, said Dec. 9 during a panel discussion at the Elsevier Business Intelligence FDA-CMS Summit that meta-analyses are a single data point and the agency has to decide how it will treat the information it gleans from them.

“I think if there’s scientific standards and the analyses are transparently done and they’re done to the right standards, that’s a good thing,” Clark said. “I think where more work and consideration needs to be done is, is there regulatory decision-making made off of those meta-analyses?”

Agency Wants Better Response Times

The new meta-analysis team would give the agency the ability to respond faster to those published by academics and potentially allow the agency to address new safety concerns sooner.

That became an issue when the Cleveland Clinic’s Steve Nissen published a meta-analysis of clinical trial data on GlaxoSmithKline’s Avandia (rosiglitazone) in 2007 showing the drug was causing increased cardiovascular risks.

The analysis led to an advisory committee meeting and arguments among FDA officials about whether the drug should be withdrawn from the market. After several more years of debate and data gathering, it was eventually put on a risk management plan that is expected to allow "minimal" sales (Also see "FDA, EMA Decisions On Avandia Reflect The Power Of REMS" - Pink Sheet, 27 Sep, 2010.).

The agency has clearly been wary of relying on the studies for regulatory decisions. Still, the agency used its own meta-analysis of Avandia trials to help it make decisions about whether the drug should be pulled (Also see "Avandia Shows FDA Discomfort With Meta-Analyses, Observational Studies In Decision-Making" - Pink Sheet, 11 Oct, 2010.).

Now it appears the agency wants to determine on its own the best applications for a meta-analysis as well as how best to respond to others when they are published.

Office of New Drugs Director John Jenkins said during the panel discussion at the FDA-CMS Summit that when looking back on the case, the agency wished it would have communicated earlier that it already was looking at the signal Nissen had identified.

Jenkins said the agency now is more focused on early communication of safety signals it finds, so the public better understands what it is working on (Also see "FDA Clears Drugs With Post-Market Safety Questions Faster Than It Re-labels" - Pink Sheet, 9 Nov, 2009.).

The meta-analysis team theoretically would allow the agency to better deal with a similar situation. FDA can be in the best position to respond to situations like Avandia because it has access to all the patient-level data, Jenkins said.

“We have to have the capacity to analyze these data when they become available,” he said. “We need to go back and take it apart and sometimes put it back together again before we can say how do we view this from a regulatory action?”

Increased public disclosure of clinical trial data is a positive development, Jenkins said during the panel discussion. He said he is expecting more meta-analysis-related signals to reach the agency.

“I don’t see it as necessarily a bad thing, but it’s a new source of signaling that we have to be ready to respond to,” Jenkins said.

Guidance Development Ongoing

Development of the meta-analysis guidance already is under way and expected to explain the requirements for presenting and challenging drug safety information with them. This summer, an FDA official predicted the public meeting could occur in January, although the agency's new proposal suggests the meeting might be pushed into the PDUFA V era.

The Institute of Medicine also is working on meta-analysis standards (Also see "Meta-Analyses Will Get Separate Guidance Documents From FDA And IoM" - Pink Sheet, 9 Aug, 2010.). An IoM spokesperson said the reports are undergoing peer review and could be released in late February or early March.

Jenkins said FDA and others also are struggling with how much weight meta-analysis should carry in the decision-making process. Statistically, they are not as well thought of as randomized clinical trials and observational studies.

Nissen, also a member of the Dec. 9 panel, said the meta-analysis becomes a useful tool once it is clear all data is available. “By getting it all out there, you give us the opportunity as scientists, but we have to act responsibly,” he said.

By Derrick Gingery