Improving the FDA, Ctd

by Elizabeth Nolan Brown

Should the U.S. Food and Drug Administration inch closer to Europe’s drug approval model, in which certified, independent bodies can (and compete to) review new products? A Dish reader says, hey, it worked for the Federal Communications Commission:

The European system you describe has been in place for more than 15 years for much of the equipment that’s approved by the FCC.  Those of us who worked in the field at the time wondered if it really would be as reliable as the system the FCC used before, which required applications that were reviewed by the FCC and had to be approved before the equipment would be sold.  In practice, though, this system works much better than the old way – products come to market faster, the process is quicker and cheaper and, most important, there’s no evidence that there’s any more cheating than their used to be.  The FCC also has, as suggested, agreed that testing that meets European requirements also is good in the U.S., saving companies the trouble of re-testing equipment they know is compliant.  It’s a model the FDA definitely should consider.

Another Dish reader is more skeptical—but offers his own FDA improvement plan:

I’m the Director of Engineering at a small medical device company… While the European method of getting device/drug approval does seem attractive, I just wanted to clarify that the FDA does have an “independent reviewer” option that allows third party organizations that are certified by the FDA to perform the review for market clearance for many types of devices. The Catch-22 here is that the third party reviewers, while generally being more efficient than direct FDA review, are very expensive, certainly more expensive than a standard FDA 510(k) fee ($5,100 per device). So do we move to more third party review to expedite device approvals, and make medical device development more expensive, or do we bolster the FDA staff so review can be faster there?

One way the FDA could become more efficient is if they took a more libertarian approach to approval. Right now, a device/drug has to be proven “safe and effective” to get pre-market clearance. Why not have the FDA only concern itself with the “safe” part and let the market take care of the “effective” part? If a device or drug is ineffective, no one will buy it. That radically simplifies clinical trials as well, lowering the requirements for FDA approval while maintaining the safety of patients.

But another reader takes issue with the idea that Europe’s drug approval process is something to emulate:

I actually work in drug discovery and am currently working on a filing. But I don’t know where this animus is really coming from in terms of approval times. Consider this paper from the New England Journal of Medicine. In it they analyze the drug approval times for the FDA, the (European Medicines Agency) and Health Canada. There are a lot of statistics and different ways to look at the data, but these are the main conclusions:

For novel therapeutic agents approved between 2001 and 2010, the FDA reviewed applications involving novel therapeutics more quickly, on average, than did the EMA or Health Canada, and the vast majority of these new therapeutic agents were first approved for use in the United States.

In that study, a total of 289 unique novel therapeutic agents were approved, including 190 that earned approval in both the U.S. and Europe. Of this group, about 64 percent were first approved in the United States. Some 154 agents were approved in both Canada and the U.S., with us first 86 percent of the time.

The same reader points to another study, this one comparing the EMA, FDA and Japan’s Pharmaceutical and Medical Devices Agency (PMDA):

If you look at the first figure you’ll see the median values for the FDA are the best over the entire period. There is some variability, but I think that’s because of a couple outliers. Drugs are also generally submitted to the FDA first. So even thought submissions are completed by the FDA faster, the other agencies have more information from the drugs approved/being processed in the US.

I must say, all of the data I’ve seen says the FDA is doing a decent job in terms of approval times.

The researchers behind the New England Journal of Medicine paper, published in 2012, may provide some insight into mixed perceptions of the FDA’s approval process. Writing in Forbes, Joseph Ross and Nicholas Downing tease out some nuances in their findings:

… there was much more variation in time to approval among applications to the FDA. More than half of approvals were complete within one year, but there were many examples of the FDA requiring 800, 1000, even 1200 total days before approval. For instance, the well-known anti-cancer drugs Sanofi‘s Eloxatin and Novartis‘ Gleevec were both approved in less than 80 days, however it took more than 10 years from initial submission to approval for Sabril, and (sic) anti-seizure medication, and Asclera, a sclerosing agent to treat varicose veins.

A lot of the variation in FDA time to approval can be attributed to whether one or more cycles of review were required. Among the 62% of applications the FDA approved after a single review, the median time to approval was 278 days. In contrast, the median time to approval was 765 days among the 38% of applications that required multiple cycles of review.

Interestingly, applications within the hematology, oncology, and immune-modulating and anti-infective therapeutic classes were most likely to receive FDA approval after a single review. Applications within the musculoskeletal and pain and psychiatry and central nervous system therapeutic classes were most likely to require multiple cycles of review.

Okay, but can we all agree the FDA needs to hurry it up on the sunscreen already?