Healthy Skepticism Library item: 5268

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: news

Herper M.
Vioxx's Descent : Eighteen Inflated Months
Forbes.com 2006 Jun 26
http://www.forbes.com/business/2006/06/26/vioxx-merck-drug-trials-cz_mh_0626vioxx.html?boxes=author

Keywords:
Vioxx

Notes:

Ralph Faggotter’s Comments:

“From the day Vioxx was withdrawn from the market, the principal assertion from Merck has been that patients exposed for less than 18 months were not at risk,”

The ‘safe for the first 18 months’ defence is looking shaky now the evidence has been put under the microscope.

Full text:

Pharmaceuticals
Eighteen Inflated Months

Once again, Merck’s legal battle over its withdrawn pain pill, Vioxx, is magnifying a scientific quibble into a major problem for the embattled drug giant.

On Sept. 30, 2004, Merck’s (nyse: MRK – news – people ) then chief executive, Raymond Gilmartin, stood up at a press conference and announced a stunning decision: The company was pulling its $2.5-billion-per-year painkiller, Vioxx, off the market because the results of a new study showed the drug doubled patients’ risk of heart attack and stroke.

But Gilmartin and his chief scientist, Peter Kim, repeatedly pointed to a silver lining in the data. The study had to run 18 months before it became clear that patients on Vioxx were more at risk than those taking a placebo. That made it easier for Merck to stand behind its contention that earlier studies, which had been shorter, showed no increase in risk. It also opened the possibility that when patients sued, saying their heart attacks were caused by Vioxx, Merck could ask to have the vast majority of cases tossed out because the plaintiffs didn’t take Vioxx for very long.

For almost two years, Merck has repeated the 18-month conclusion. But now it is under attack from The New England Journal of Medicine, the most influential medical journal in the world. The Journal published a paper on the study, called APPROVe, in February 2005, in which the authors again asserted that it takes 18 months for Vioxx’s risk to emerge. But today, the Journal posted a correction removing the 18-month difference from the abstract and discussion of that article.

It is the second time the NEJM has taken issue with a Vioxx trial—in December, it issued an expression of concern about the way heart risk was described in another big Merck study.

“From the day Vioxx was withdrawn from the market, the principal assertion from Merck has been that patients exposed for less than 18 months were not at risk,” says Steven Nissen, who heads cardiology at the Cleveland Clinic and has been one of Merck’s biggest critics on Vioxx—including in a letter published alongside the NEJM correction. “Here we now see a retraction of the claim.”

Curt Furberg, a professor at Wake Forest University and another longtime critic, says that based on new data that has been made available, “there is no evidence whatsoever” for an 18-month delay.

Trial lawyers may salivate at the correction because Merck hammered so hard on the importance of the 18-month difference. Trent Miracle, an attorney at SimmonsCooper, says that the correction could double the number of Vioxx lawsuits his firm files by September to more than 2,000.

So far, Merck has lost one of two long-term-use cases so far tried, compared to two of four short-term-use cases. Ten new cases are set to begin by the end of the year. It is possible that by persevering in battling each case, Merck might outlast the trial lawyers.

Merck disagrees that the correction damages the 18-month claim at all. In a press release and a five-page letter to the scientific community, Peter Kim asserts that although Merck had to correct a statistical method, the 18-month conclusion is still valid based upon the data Merck had at the time. “It is important to understand that the correction centered on the description of a single statistical method,” Kim said in a statement. “Merck used a number of analytical and graphical methods to evaluate whether the relative risk changed over time.”

Scientists usually demand that results be statistically significant—that mathematically determined odds of a result happening by random chance are less than 5%. There is one measure that gives this result for the APPROVe data, but it is not the one prespecified in the study.

In an interview, Jeffrey Drazen, the NEJM editor in chief, said that picking a method of analysis after seeing the data is bad scientific form. “You specified a method in the paper,” he says. “You have to stick with the method that you specified.”

From Drazen’s point of view, the change is scientifically unimportant, because the main conclusion of the paper was that Vioxx poses a risk to the heart, not that such a risk takes 18 months to emerge. “To me, it’s not a major thing,” he says. “If it weren’t for all the external events going on, it would not cause a blip on the radar screen.”

John Baron, a professor of medicine at Dartmouth University who co-authored the APPROVe study, makes a similar point: “In general, from what I understand, scientifically the change is quite small. All this is kind of a tempest in a teapot, but also because with the extended followup data, the discussion isn’t even about the current data.”

But the long legal battle seems likely to continue to tarnish what was once the most sterling scientific reputation in the drug industry. Wake Forest’s Furberg says further analyses of the data that are now ongoing are only likely to further weaken the claim that there is a delay in Vioxx’s affect on the heart. “I think it’s unwise of them to come down so hard now, when they must know what’s coming down the road.”