On Monday, I wrote about a new bill introduced by Sens. Ted Cruz (R-TX) and Mike Lee (R-UT) to overhaul the Food and Drug Administration. Among other things, the bill would try to speed up FDA approvals in order to boost innovation and address the current dearth of new drugs in the pipeline.
I was skeptical. In recent years, Congress has passed a number of measures to greatly shorten the amount of time that drug applications languish in the FDA bureaucracy, and that hasn't led to a lasting boom in new drugs that are significant improvements over existing treatments. So far, cutting through red tape hasn't ushered in a new age of innovation.
Economist Alex Tabarrok took issue with my piece and wrote a response. He noted that the creation of the Prescription Drug User Fee Act (PDUFA) in 1992, which allowed the FDA to collect fees on pharmaceutical companies in order to increase funding for the drug approval process, dramatically shortened approval times for drug applications and led to more drugs coming to market. The average review time for a drug dropped from three years to about one year.
That part is true. The FDA under Ronald Reagan in the 1980s was underfunded, sluggish in approving new drugs, and arguably holding back innovation. PDUFA really did speed things up. But that's not the whole story, either. Let's take a look at the longer view:
What you see is that the annual approval rate for new drugs has mostly held steady since the 1950s, with a (short-lived) boost in the 1990s — the period after PDUFA gave the FDA more funding to clear out a backlog of applications.
But this boost was only temporary. During the late 1990s and 2000s, as drug review times kept falling to all-time lows, there was no increase in new drug approvals. Today, the FDA approves drugs much more rapidly than it did in the 1980s — it's the fastest regulatory agency in the world. But the number of new drugs coming to market each year has fallen back to historical averages.
This suggests the FDA is no longer the main barrier to innovation. And it suggests that speeding up approval times even further won't always lead to new and better drugs.
Faster approvals don't automatically mean better drugs
Importantly, we don't just want to look at the raw number of drugs being approved. What we care about is how many innovative drugs are being approved that offer significant improvements over existing treatments. And on that score, there's little evidence that speeding up approval times has helped here, either.
In the 1990s, Congress created four programs to expedite the development and approval process for new pharmaceuticals: the breakthrough drug designation, priority review, fast track, and accelerated approval. These programs were all intended to push innovative new drugs — medicines to treat rare, serious, or life-threatening diseases — through the FDA more quickly, often on the basis of more limited and preliminary clinical trial data.
Researchers who have looked into this question found that speeding up these pathways did surprisingly little to increase the number of innovative drugs. The majority of drugs that have come onto the market offer little or no improvement over the status quo. More specifically, several studies have found that since the mid-1990s, about 85 to 90 percent of new drugs don't offer any clinical advantages for patients.
Recent research in the BMJ looked at a database of all new medicines approved by the FDA between 1987 and 2013 to find out whether the increase in the number of products coming through on expedited pathways led to an increase in innovative drugs getting to market. Unfortunately, there was no such correlation. Over the past two decades, the number of drugs qualifying for the FDA's expedited programs has increased 2.6 percent per year, but "this trend is being driven by drugs that are not first in class and thus potentially less innovative," the study authors wrote.
There's good reason many of these new drugs haven't been effective. As the physician and epidemiologist Ben Goldacre points out, accelerated approvals allow the use of less robust data, including "surrogate endpoints" in studies. This means companies can submit things like the results of blood tests on cholesterol levels as evidence for the efficacy of drugs — rather than waiting to find out how the drugs affect disease and mortality rates.
That's why, as this analysis from MedPage Today and the Milwaukee Journal shows, the use of surrogate endpoints for approvals of new cancer drugs over the past decade hasn't actually helped patients very much. Three-quarters of the new cancer drugs got to the market "without proof that they extended life," the reporters wrote. "Seldom was there proof of improved quality of life, either." (You can read more about the perils of relying on surrogate endpoints here.)
Faster approvals could mean more harmful drugs
If speeding up the FDA process just meant more drugs of questionable efficacy, that wouldn't be a huge problem. But there have also been questions about whether riskier drugs with safety issues were coming onto market. A 2008 study (that Tabarrok cites) suggests there's no evidence of a link between speeding up FDA approvals and the number of adverse events. But that study isn't the only one on this question. A more recent 2014 study in Health Affairs found that after PDUFA, there was a 25 percent increase in medicines that needed to be pulled from the market out of safety concerns or that received a new black box warning (which signals that the drug carries very serious side effects).
It's difficult to tease out causation here. Have speedier review times involving more preliminary data caused regulators to put more drugs on the market that they're unsure about? But at the very least, more black box warnings mean that there are more drugs reaching patients with what regulators consider to be sketchy safety profiles.
Are we trying to fix the wrong problem?
To be fair, it's possible that the past is not a perfect guide, and that a policy that streamlines the FDA further will finally lead to a surge of real innovation without corresponding side effects. But what evidence we have suggests we shouldn't assume that automatically. And there are potential risks and trade-offs with slimming down the drug approval process. Where we draw that line exactly is, in some sense, a judgment call. It's possible to look at the numbers above — about 10 to 15 percent of new drugs offer real clinical value, an uptick in black box warnings — and conclude that the benefits are worth the risks, that we should keep speeding up drug regulation.
Or you could read those numbers and decide that perhaps the dangers of poorly regulated medicines are becoming worrisome and we should focus on other ways to boost innovation. Again, the FDA is already the fastest drug regulatory agency in the world. Yet the number of new and innovative drugs coming to market has stagnated.
The causes of this innovation crisis are complex. Some say we need more funding for very basic research because we've maxed out medicines that can be made based on our current understanding of the body. Others argue that incentives in the pharmaceutical industry are skewed toward marketing and "me too" drugs (which are minor variations on existing therapies).
I don't know what the solution is, but I'm skeptical of further speeding up or whittling down the regulatory process. We should focus on other measures to boost R&D instead.
By Julia Belluz - Vox/Dec. 16, 2015