Vytorin and Cancer

The news that the cholesterol-lowering drug Vytorin (a combination of Merck's Zocor, a statin, and Schering-Plough's Zetia) didn't reduce heart valve disease but raised the risk of cancer sent the stocks of both companies plunging yesterday. I don't have much to add to the basic story. The New York Times got it right. Headline: "Trial Intensifies Concerns About Safety of Vytorin."

The trial's sponsors brought in Richard Peto of Oxford to explain away the cancer results, which, while small, reached statistical significance. Of the 1,900 people in the four-year trial, "102 patients taking Vytorin developed cancer, compared with 67 taking the placebo. Of those, 39 people taking Vytorin died from their cancer, compared with 23 taking placebo," according to the paper.

Said Peto, who compared those results to a still uncompleted set of trials of the combination drug: “I think we should not be diverted by fears of cancer.” His argument? It triggered many different types of cancers, not just one type. So it couldn't have been the drug.

What kind of science is that? According to Times reporter Alex Berenson, the cancer signal reached the 95 percent level of certainty, the benchmark used by medical researchers as statistical significance. You don't get to move the goal posts after the ball is kicked, in this case, away from where the data landed.

I guess I shouldn't be surprised. Though Peto has been a leading voice in the crusade against second-hand smoke, the last time we heard from him was when he explained away the late Richard Doll's undisclosed conflicts of interest with Monsanto. The two men issued a famous report in the early 1980s that downplayed the environmental causes of cancer.

He's always claimed to be a great stickler for the sanctity of the data. This appears to be a case where he should have let the data speak. If those larger trials come back without a cancer signal, then the famed epidemiologist (or others) can do a meta-analysis (a pooling of all the data) and say the risk is not statistically significant. But as of now, it would appear that this combination pill, still taken by millions of people, raises the risk of cancer.