Sacrificing standards

The FDA may be moving too fast

Today’s Food and Drug Administration moves much faster than it used to. That may not always be a good thing. A review of drug approvals by the agency from researchers at Harvard Medical School released last week found that the FDA is approving drugs more rapidly with weaker evidence than it did in the past. That can be beneficial when it leads to needed medicines getting to market quickly, and I believe that’s the agency’s intent. As the study’s authors highlight, however, this emphasis on speed and flexibility could be eroding standards. It may be time for a gut check.

The gold standard for demonstrating efficacy — and the surest way of winning drug approval — is to demonstrate success in large, wellcontrolled studies that result in a hard outcome. But there are faster ways to get to market.

In 1992, Congress created the accelerated approval program, which can greenlight medicines based on “surrogate” endpoints that predict rather than confirm benefit for patients, or those that have shown a shorter-term benefit. It’s one of several initiatives that have changed how the agency works. According to the study, 80.6% of approvals between 1995 and 1997 were supported by at least two pivotal trials. That number dropped to 52.8% between 2005 and 2017.

Companies that get accelerated approval have to prove their drug works with a confirmatory trial in order to gain full approval, but there’s no hard timetable for when that must be done. Thus, drugmakers often don’t hurry to conduct those tests. This is problematic at best, dangerous at worst.

Here’s just one case: In 2016, Sarepta Therapeutics Inc. sought approval of a medicine to treat a rare muscle-wasting disease in young boys based on weak evidence from a tiny trial. In the face of significant public pressure, the FDA approved Exondys 51 even though one of its scientists called the treatment “an elegant placebo” in a report. Sarepta is selling the drug for over $300,000 a year but has continually delayed a confirmatory trial. It’s now years away from completion, and there have been no real consequences for the delay.

When companies do complete post-approval trials, it sometimes reveals a mistake. Eli Lilly & Co.’s cancer drug Lartruvo got accelerated approval in 2016. Lilly then pulled the medicine from the market last year after a larger trial found no benefit. That’s a rare outcome, but there are many expensive drugs on the market that have never moved beyond surrogate endpoints. A study of 93 accelerated cancer drug approvals between 1992 and 2017 found that only 19 had proved to help patients live longer in a follow-up trial.

There are some good reasons for faster approvals, as former FDA Commissioner Scott Gottlieb outlined in a Twitter response this week to a critical New York Times editorial penned on Jan. 11. Scientists are better at evaluating the safety of medicines and trial design has improved, for example. And advances have made it easier to create drugs that target small populations and have dramatic effects, Mr. Gottlieb wrote.

He makes good points. But the agency arguably hasn’t found the right balance between embracing advances and maintaining a high bar. It certainly has a ways to go on post-approval follow up. America is entirely unable to control the price of new medicines; the approval of marginal drugs has financial consequences.

The FDA will soon face one of its most important and controversial decisions yet. Biogen Inc. is seeking approval for the first purportedly disease-modifying Alzheimer’s drug — a medicine that could be used by millions of people and cost billions — without good evidence that it works. The agency often uses unmet need as a justification for shifting standards, and there’s no bigger unmet need than Alzheimer’s. That doesn’t justify an approval based on one failed trial and another that is a questionable success at best.

The agency will have to decide whether to review or approve the medicine in the next year or so. This choice is an opportunity to resist public pressure and move back toward demanding firmer proof of efficacy before drugs hit the market.