A leading cardiologist at the Cleveland Clinic, Steve Nissen, is charging that the NIH had a panic attack and should have allowed the ADAPT trial to continue.

The ADAPT trial, for the newcomers, was the Alzheimer’s Disease Anti-Inflammatory Prevention Trial. It was halted in December 2004 after similar trials of Vioxx and Celebrex had been stopped and garnered national attention. By that time, Vioxx was off the market.

In a late December 2006 letter to the peer-reviewed journal PLoS Clinical Trials, Nissen says that it’s now clear that the NIH jumped the gun in halting a clinical trial of Celebrex, a COX-2 inhibitor. It took two years for that to come to light. But that’s because several medical journals rejected the incomplete, partial findings of the ADAPT team before PLoS finally agreed to publish them.

SOPs Abandoned

Writes Zissen: “Management of clinical trials is a major public responsibility. It takes self-discipline and a precise understanding of statistical methods. For ADAPT, accepted scientific procedures were not followed, resulting in an inappropriate public warning. Accordingly, the trial results cannot be reliably interpreted.” Ouch. We can’t recall such an open condemnation of the NIH.

According to Nissen, the ADAPT trial was stopped not by an Institutional Review Board (IRB) supervising the trial, nor by a Data Safety Monitoring Board (DSMB). Rather, he says, the trial was stopped by a fearful and unscientific NIH. (Nissen cites a Washington Post article to buttress his point, but that article appears to be an editorial written by Dartmouth professor-consultants, not a news story.)

Nissen: ‘They abruptly stopped the trial’

Anyway, the Dartmouth professors and Nissen believe that the NIH rushed to a premature, statistically baseless conclusion. They think the NIH was swept up in the larger hysteria around painkillers and drug safety in late 2004.

As Nissen writes in PLoS:  “It appears that NIH officials, concerned about revelations regarding the safety of coxibs in the colon polyps studies, simply decided to look at the cardiovascular event results in ADAPT. Seeing a marginally significant difference between naproxen and placebo, they abruptly stopped the trial. Indeed, NIH officials went even further and issued the public warning. Both the termination of this trial and the issuance of a public warning about naproxen were inappropriate and reflected faulty logic. These actions unnecessarily created public consternation and denied the participants in the trial the opportunity for their commitment to result in any benefit to themselves or society.”

IRB v. DSMB

One interesting issue relates to clinical trial governance by DSMB vs. the IRB. This is an evolving issue. The exchange of letters in PLoS only begins to illuminate the degree to which an IRB or a DSMB may render judgment on the interim data in any particular situation. Neither article, alas, mentions a less well known body: the PERB, or Poultry-Emulating Review Board. PERBs can occasionally be found in organizations that are more concerned with public relations than with scientific progress.

The official response to Nissen’s letter is just as interesting. A group of ADAPT physician-scientists, lead by John Breitner at the University of Washington, says no—the NIH did not stop the trial on its own.

Rather, Breitner and colleagues contend the study was stopped by the ADAPT steering committee, plus a little bit of FDA input, plus a bit of NIH participation, and (mostly) a lot of hunches and guesses and warm thoughts about their mothers. We’ll get to those in a minute.

Shifting The Blame

As Breitner et al. outline the reasons for the halt of the ADAPT trial, they become fuzzier and less intelligible. The DSMB for the ADAPT trial appears to have looked at their own preliminary safety data but not been alarmed. As the ADAPT investigators note in reply to Nissen, the early comparator and Celebrex safety data “would not in themselves have led to a decision to suspend either treatment.” A miscellaneous bag full of other factors intruded. One was Pfizer’s suspension of another trial of Celebrex, which did show a real safety signal.

Once the media blitz on the COX-2 inhibitors had begun, the ADAPT scientists apparently became fearful. They seem to have believed they would have to come up with new informed consent paperwork, among other issues. And they invoked the vaunted “mother” test. To wit, the physicians making the decision had to decide if they’d want their own moms to continue participating in the trial, given the physicians’ own limited knowledge about the COX-2 inhibitors at the time. None of the doctors could vote “yes” so the trial was stopped. Simple as that.

Or is it? We’re probably supposed to think, “Hey, these doctors love their mothers. Isn’t that great?” No. Actually, it’s a weak argument. The mommy test is lame and ineffectively presented. It’s a transparent attempt to seem reasonable and kind-hearted but shirk science’s long-term responsibility to society at large and to the 2,400 patients in the ADAPT trial.

Invoking Mothers

More than anything, the mommy test injects emotion and guesswork into a process that should remain as rational as possible. Some people might want their mothers to have the best possible pain relief. Other people might want to give their mothers flower-scented, homeopathic quackery. Still other people might want their mothers’ doctors to prescribe medication based on scientific facts, not hunches and voodoo.

We’re sure the ADAPT scientists understand that clinical trials always entail a judicious balancing of risk and opportunity. This was true for the ADAPT trial, and it is true for trials of toenail fungus. The best way to take care of the mothers of the world is not to dodge the scientific, procedural or publicity-oriented challenges that come with finding out if a drug is safe and effective. Some hard-headed mothers we know would think the people running the ADAPT study lost their heads just when society needed them to remain sober and unaffected by emotion.

When All Else Fails…

Both sides looking at the halted ADAPT trial agree that (in that study) there seems to be no increased cardiovascular risk to Celebrex. We’ll never know the full safety picture, of course, because the trial was stopped. We’re not defending Celebrex. We are defending the existing clinical trial processes and procedures, which (had they been followed) would not have forced anyone to choose between scientific knowledge and the collective safety of the mothers of the world.

Thankfully, the two sides of this controversy also agree on one more thing. They think the news media got just about everything wrong and created a bad atmosphere for a scientific decision. Here’s how the ADAPT authors put it: “We concur with Dr. Nissen, nonetheless, about the lamentable and somewhat misleading publicity that surrounded our decision.”

Well, go ahead—shoot the messengers. Why not? It’s fun! But the writing skills of the NIH and the physicians in the ADAPT trial should be carefully sharpened before they try to communicate with people outside the pharmaceutical industry and the medical community. If you don’t believe us, show the PLoS articles to your mother and ask her if they make any sense.