Tuesday, March 27, 2007

"Drug Crazy"

Every time the issue of conflicts of interest affecting the US Food and Drug Administration (FDA) comes up, one is sure to see at least one opinion piece decrying how excess concern with such conflicts will stifle innovation and keep wonder drugs off the market. There seems to be a whole raft of economists, lawyers, and policy-wonks out there who believe that any slowdown in drug approvals or any excess concern about drug side-effects will lead to unmitigated disaster.

Recently, the FDA announced it would restrict physicians and scientists with major conflicts of interest from participating on its advisory boards. To summarize quickly (with quotes from a Washington Post article),

The biggest change bars anyone from serving on a committee who has a financial interest of $50,000 or more that could have an impact on the drug, device or other issue the committee is considering. Overriding this rule would require a special waiver. If such a waiver is granted, that person could attend meetings, consult and express his views, but could not vote.

In addition, anyone with a conflicting financial interest of less than $50,000 could sit on an advisory committee, but could not vote. There will be a 60-day comment period before the proposals take effect.

"The limit would cover financial interests over the proceeding 12 months and cover, for example, stock and consulting fees," [said an Agency spokesperson.]
See discussion of these changes on GoozNews and the links from PharmaGossip. I could easily argue that allowing people who get up to $50,000 a year from a drug company to sit on an advisory board would given undue influence to people who are paid substantial money by those with vested interests in having the board advise in a certain way.

Nonetheless, soon after these new rules were announced a commentary appeared in the Wall Street Journal aptly entitled "Drug Crazy," by Richard A Epstein, a law professor at the University of Chicago, which practically warned the sky would fall if people with large conflicts of interest are kept off FDA advisory boards. Here are some choice quotes.

The main justification for this new regime is the widespread perception that drug and device makers have "hijacked" the FDA process, allowing dangerous drugs to slip into the market. The most conspicuous illustration involves the various COX-2 inhibitors such as Vioxx and Bextra....

The common perception is that the FDA acted too late in removing these drugs from the marketplace, and indeed should never have approved them at all. That perception rests on a crucial tradeoff: how much of an increased risk of heart attack should people be allowed to run in order to mitigate the ravages of arthritis and other degenerative diseases.

'Suffering boomers want to fill Vioxx void' is a headline in last Friday's Chicago Tribune. The gist of the story: Baby boomers and others have returned to Pfizer's Celebrex, which reported an 18% increase in sales notwithstanding the FDA's stringent black box warning. Similarly, many individuals stashed away their Vioxx for a rainy day after it was recalled by Merck. These individuals understand the risk they're running, but trust their physicians to minimize its impact.

This evidence should put the FDA's zeal in a new light. Just how do these advisory committees 'protect' informed product users by limiting their choices for dealing with crippling chronic conditions?

The FDA's major problem is not laxity, but zealotry. Its current get-tough view on conflict of interest only aggravates the fundamental flaw in its institutional design. Transfixed on the harms drugs can cause, the FDA remains largely oblivious to the harms they can prevent. Any delay in the use of a successful drug is costly: The delay matters little to the FDA, but a great deal to the thousands who plea for compassionate exemptions to try a drug that has not met with FDA approval.

The current get-tough policy on conflicts of interest only magnifies these costly FDA biases. The agency rules cut out the persons who know most about the drugs, and who could well counter unsound objections by critical committee members. But keeping these experts off the committees also skews the deliberative process in a more subtle and powerful way.

Anyone who consults for drug companies is likely to come with the presumption that new drugs may well fill important therapeutic voids. Once these individuals, as a class, are kept off advisory committees, the remaining drug experts are more likely to have the populist views held by the likes of a Marcia Angell or Sidney Wolfe, who suspiciously view most new drugs as insignificant advances over prior treatments. I tremble at the thought that these "untainted" experts will markedly slow down the FDA approval process. This would be bad news for the tens of millions who suffer from arthritis, and the countless individuals who suffer from other conditions.

Right now, we all have a simple expedient to protect ourselves against dangerous drugs that make it to the market: Don't take them. But we have no protection at all when the FDA denies us that choice in the first place. Right now the pace of drug approval is too slow. We don't need the FDA to slow it up still further.

My response is, to be polite, oh, balderdash. Prof Epstein mainly succeeded in revealing his lack of knowledge about drugs and clinical epidemiology. Let me address his main points.

Did Taking Rofecoxib Off the Market Deprive Patients of a Uniquely Effective Drug?

His first major point is that removing Cox-2 inhibitors from the market deprives arthritis sufferers of some magically effective treatment.

In fact, even the clinical trials supported by Merck failed to show that Vioxx (rofecoxib) was any better at relieving pain or other arthritis symptoms than older, cheaper non-steroidal anti-inflammatory drugs (NSAIDs) like ibuprofen and naproxen. See:
Day R, Morrison B, Luza A et al. A randomized trial of the efficacy and tolerability of the COX-2 inhibitor rofecoxib vs ibuprofen in patients with osteoarthritis. Arch Intern Med 2000; 160: 1781-1787. (link here)
Lisse JR, Perlman M, Johannsson G et al. Gastrointestinal tolerability and effectiveness of rofecoxib versus naproxen in the treatment of osteoarthritis: a randomized controlled trial. Ann Intern Med 2003; 139: 539-546. (link here)

To quote the latter study, "in terms of osteoarthritis efficacy, no statistically significant difference in PGADS scores was observed between the rofecoxib and naproxen groups over 12 weeks.... Improvement in AUSCAN scores was not statistically signifcantly different between the rofecoxib and naproxen groups." So even the authors of a Merck supported study were not claiming that Vioxx offered any more pain or symptom relief than did naproxen.

It is possible that for a few patients a Cox-2 inhibitor offers better symptom relief than a NSAID. But there is no evidence of any wide-spread or major advantages. The notion that Cox-2 inhibitors are better pain or symptom relievers seems to be an urban myth. And Counselor Epstein seems to have bought into this myth.

In fact, the only clear advantage of rofecoxib seemed to be that it lead to a slightly smaller risk of upper gastrointestinal (GI) problems than did traditional NSAIDs. For example, one study showed that the risk of upper GI bleeding was 1.3%/year for rofecoxib vs 1.8%/year for various NSAIDs. But even that is a small advantage. One would have to treat 200 patients with rofecoxib to avoid one such a bleed. See:
Langman MJ, Jensen DM, Watson DJ et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282: 1929-1933. (link here)

Are the Risks of Drugs Well-Known, and Can be Easily Avoided by Not Taking Them?

Prof Epstein asserted that it is easy to avoid drug adverse effects. This might be true when these effects are well-known. But the whole point of the Vioxx case was that the adverse effects of rofecoxib were not well-known, and there were allegations that the adverse effects were concealed while the drug was vigorously promoted to a wide-spectrum of patients who were not at particular risk of GI side-effects, and could have been satisfactorily treated with other drugs. (See some relevant posts here, here, here and here.)

Do Conflicted Board Members Provide Needed Opposition to "Populists"?

The notion that without board members who are investors in or paid by pharmaceutical companies, boards will be lead by "populists" who "view most new drugs as insignificant advances over prior treatments." Underlying this is the unstated assumption that many drugs offer significant advances over prior treatments. This is why Prof Epstein "trembled" at the thought of any slow-down in drug approvals.

As a physician, I wish that most new drugs were major advances over existing treatments. But my knowledge of evidence-based medicine suggests that they aren't. Unfortunately, most new drugs (and most new treatments) offer at best only small incremental improvements over other relevant options. And some are simply "me-too" efforts that really offer the most inconsequential advantages.

For example, see the data presented on Vioxx (rofecoxib) above. Clinical research never showed that this drug was a much better pain or symptom reliever than old-fashioned NSAIDs. At best, it lead to slightly fewer gastrointestinal side effects. But we now know that these were traded for increased risk of adverse cardiovascular effects.

In summary, I would suggest that lawyers and economists who write about drugs ought to look at the relevant clinical research before composing their first sentences. Maybe that would give them a more realistic idea of what new drugs and other health care products really are innovations, and what their downsides might be.

By the way, the author of the article, Richard Epstein, did disclose that he has "consulted frequently with pharmaceutical companies." Maybe employment by these companies influenced him to be less than skeptical about the benefits and harms of their products.


. said...

Bravo. Great posting. I agree whole-heartedly with you. The best practices committees should also be examined in the same way as the FDA committees.

Philip R. Alper, M.D. said...

I have no ties to the drug industry, but I do see advantages to creation of drug diversity even without major differences from existing drugs (the new ones often derided as "me-too" drugs). Pharmaceutical leaps aren't always transformative; most aren't. Seemingly small incremental benefits may relate to better response in sub-groups of patients that may not be readily apparent in studies not specifically designed to check for them.

When some patients have responded better to Vioxx than to other NSAIDs in my own practice, I have wondered why. This is true with all drugs. Better than average response may be seen as placebo effect, but it may also be real and serve as the basis for further research and help in developing superior drugs.

Marcia Angell made similar criticisms of Epstein to those in this blog in her Wall Street Journal letter-to-the-editor. The problem, it seems to me, is more the cost of the new drugs and the incentive to keep pumping the profits than the nuances of efficacy.

Philip R. Alper, M.D.