Drug Industry Defies Eroom's Law
In technology there’s Moore’s Law, which says that semiconductors will keep getting cheaper and better. In drug development there’s Eroom’s Law (Moore spelled backward, get it?), which posits that:
the number of new U.S. Food and Drug Administration (FDA)-approved drugs per billion U.S. dollars of R&D spending in the drug industry has halved approximately every nine years since 1950, in inflation-adjusted terms.
That’s from an article by three securities analysts and a venture capitalist that was published in the journal Nature Reviews Drug Discovery in March 2012 and got a lot of attention in the industry.
Just as it was published, though, the tide seemed to turn. The number of FDA new-drug approvals rose a lot in 2012 and stayed high in 2013. Last year it exploded. By the count of the Boston Consulting Group, 2014 was the biggest year ever for new therapeutic drug approvals:
There’s a happy way to look at this: Eroom’s Law has been repealed, the pharmaceutical industry is back, and the so-called biotech bubble is entirely rational exuberance.
Then there’s the other way. BCG estimates that peak sales from the drugs approved last year will actually add up to less than those approved in 2013, and much less than those approved in 1996, the previous high point for drug approvals. That is, more drugs are being approved, but most are aimed at smaller targets -- rare diseases, very specific cancers.
In fact, the authors of the Eroom’s Law article -- Jack W. Scannell, Alex Blanckley and Helen Boldon, then all of Sanford C. Bernstein in London, and Brian Warrington of Phoenix IP Ventures -- predicted back in 2012 that this might happen, in part because researchers were getting better at developing narrowly targeted drugs.
That’s better than not developing any drugs at all. But it’s an indication that diminishing investment returns on drug R&D may persist, even if Eroom’s Law doesn’t hold up in coming years. In their article, Scannell & Co. gave four main reasons for the poor return on investment:
- The better than the Beatles problem. The drug blockbusters of yesteryear are now low-cost generics, so even if you develop a new drug that treats a disease or symptom more effectively, it will be hard to get doctors to prescribe it and insurers to reimburse it. (To explain the name, it’s as if any new recording had to be much better than the Beatles for anyone to buy it.)
- The cautious regulator problem. It’s a lot harder to test drugs and get them approved than it used to be.
- The throw money at it tendency. Drug companies aren’t great at allocating money to R&D, and they tend to waste a lot.
- The basic research-brute force bias. In recent decades drug companies have put a lot of stock in molecular-biology research and high-volume screening techniques that may be less effective than the more experimental methods that preceded them.
No. 1 is as big a problem as ever. So is No. 2, although the FDA has made efforts to encourage the development of “orphan drugs” aimed at small populations. On No. 3, a lot of big drug companies in recent years have cut research spending, which may mean they will waste less but also may mean fewer new drugs. And although drug startups -- aka biotechs -- approach R&D in a different way than the established companies, there’s not much evidence yet that they’re any better at it. As for No. 4, the basic-research/brute-force approach has begun to yield more drugs, but they don’t serve big populations.
I caught up with Scannell, who left Sanford C. Bernstein in 2012 to become an executive at a biotech firm and is now an independent analyst and a fellow at a couple of drug-research think thanks, to ask what if anything had changed since the Eroom’s Law article. He cautioned that one shouldn’t read too much into a single year’s numbers, but said he was “fairly convinced there has been a real uptick in drug approvals over the last two or three years.”
Beyond that, though, it’s really hard to say. For a 2013 journal article, two Harvard Medical School professors surveyed physicians to identify the “most transformative drugs of the past 25 years.” Scannell says that when he read the article, what struck him was that “for a lot of them, no one would have known they were going to be transformative.” Despite recent attempts to systematize medical research such as the Human Genome Project, much about the workings of pharmaceuticals remains only partially understood. “I think there is a tendency for the commercial and policy debates on drug R&D to overestimate the degree to which drugs are designed and underestimate the degree to which they are discovered,” Scannell said.
And investing in discovery, it turns out, is a tricky business.
BCG's number "includes new molecular entities, as well as therapeutic and bioengineered vaccines, therapeutic blood products, new combination treatments and new single-agent formulations that were previously approved as part of a combination."
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Justin Fox at email@example.com
To contact the editor on this story:
James Greiff at firstname.lastname@example.org