August 01, 2007

There are dueling perspectives on the FDA today in the Boston Globe and the Wall Street Journal. Susan Woods and David Michaels argue in the Globe that the FDA's ability to assess the risks of new drugs and devices has been undermined by user fees and conflicts of interest on advisory committees:

The FDA's ability to identify the adverse risks of approved drugs is further limited by the Prescription Drug User-Fee Act, which became law in 1992. The law directs the agency to collect user fees from drug manufacturers for the purpose of speeding up the review process, and it limits how the agency can spend the funds. As it has turned out in practice, this system has also drawn funding away from other critical FDA functions, including inspections, oversight of advertising, and the scientific infrastructure of FDA. A similar law created user fees for medical devices. This also focuses on quick approvals rather than inspections, long-term safety studies or other safeguards.

User fees appear to save the taxpayer money, but the cost to the public is unacceptable. More medical products are approved more quickly, but with less information about their dangers. But the efficient review of applications for new drugs, vaccines, and devices does not have to conflict with improved safety analysis and monitoring. With adequate support, we can develop a system that maximizes both.

On the other hand, Richard Miller, writing in the Journal, argues that the FDA is rolling back the clock on regulations and slowing access to new cancer therapies:

Current FDA policies are discouraging the development of groundbreaking treatments for cancer and other killer diseases, turning the clock back on hard-won regulations put in place in response to the AIDS crisis that allow patients faster access to new drugs. Case in point: This week, facing rejection by the Agency, GPC Biotech withdrew its New Drug Application (NDA) for Satraplatin, a drug to treat prostate cancer -- despite data from a large controlled clinical trial showing the drug delayed tumor growth in patients where the disease is widespread.

Sadly, far from being an aberration, Satraplatin is the fifth promising cancer treatment set back by the FDA this year.

I tend to agree more with Miller's critique than with Woods and Michaels. Just because we are detecting more high profile drug risks doesn't necessarily mean that the background rate has changed.

The new 24 hour media environment, concerns over health care costs, more treatments for chronic disease, massive electronic databases that are open to search (and conflicting interpretations) all combine t0 generate controversy and a feeling of crisis at the FDA when a safety signal is detected, even though the agency might actually be doing as good - or better - a job as it has ever done. (This analysis from the Milken Institute is very much worth reading on the relative trade-offs of speed and safety. Note: registration required.)

There is no similar hue and cry if the FDA declines to approve a drug that was, in fact safe and effective.

Also, blaming PDUFA restrictions for the FDA's budget woes is unfair. The NIH's budget has seen a vast increase in recent years, while the FDA's has been basically flat. If Congress wanted to give the FDA substantial new funds to improve drug safety tools it could have. I also think that Woods and Michaels are alarmist about conflict of interest rules - as John Calfee argued on MPT last year.

Having said that, Woods and Michaels do raise very important points. They are correct that the agency needs new resources and tools to help improve postmarket surveillance. And it should look for new tools to streamline and improve pre-market testing, through substantial new funding for the Critical Path.

My vote would be to use technological tools to speed more new drugs to market at less cost in return for enhanced post-market surveillance. This would maximize innovation and improve safety - and reconcile the goals of (nearly) all the FDA's critics.

Posted by Paul Howard at August 1, 2007 12:17 PM

Comments

Post a comment