Statistics Smackdown: Experts Duel Over Vytorin Cancer Data
Posted by Jacob Goldstein
September 18, 2008, 12:36 pm
The Oxford stats guru who said Vytorin doesn’t increase the risk of cancer in the face of some data raising that possibility has sent a sharply worded response to questions from a couple of congressman looking into the issue.
Sir Richard Peto’s letter (online here)
ctsu.ox.ac.uk
says that “any competent trial statistician would endorse” his conclusion that there’s “no credible evidence” that the cholesterol drug is tied to increased cancer risk.
The cancer twist in the Vytorin saga started in July, when researchers said patients who received the drug in one study were more likely to die of cancer than those who received a placebo. Based on that finding, Peto and his colleagues analyzed data from other ongoing trials of Vytorin and found that, overall, the data showed the drug didn’t increase the risk of cancer.
Reps. John Dingell and Bart Stupak, a pair of congressional Dems who’ve been looking into Vytorin for a while now, sent letters asking for more info to Merck and Schering-Plough, which co-market the drug, as well as to Peto.
Since then, the New England Journal of Medicine has published not only Peto’s analysis but also a piece by Thomas Fleming, a University of Washington statistician, said it’s impossible to rule out a connection between Vytorin and cancer-related deaths.
In his response to the congressmen, Peto, who is the statistician on one of the ongoing Vytorin trials, says that Merck and Schering -Plough “did not know the results or receive any draft of my report before it was finalised and sent out by email to FDA and others.” He adds that it’s not in the public interest to “label potentially useful drugs as unsafe if there is no credible evidence that they are.”
The letter says Merck and Schering-Plough have given grants to the Oxford Clinical Trial Services Unit worth more than 100 million pounds since 1997 for work on several trials. Peto also says that complying with Dingell and Stupak’s request for all correspondence with the companies would require an “unreasonably vast amount of effort,” given that his unit at Oxford has worked with the companies on trials involving about 60,000 patients and “extensive correspondence.”
Permalink | Trackback URL: blogs.wsj.com |
