Pols Piss Off Peto

In a public letter dated September 16, Oxford statistician Richard Peto wrote heated responses to questions from US Representatives John Dingell (D-MI) and Bart Stupak (D-MI) about the clinical investigation of Vytorin (ezetimibe/simvastatin; Merck/Schering-Plough). The congressmen's questions, many of which imply professional collusion between Peto and Merck/Schering-Plough, were originally submitted to the drugmakers in letters dated August 21 and September 2.

In his response, Peto seemed to be particularly vexed by the request that the companies "make Dr. Peto available for a staff interview as soon as possible." In reply, Peto advised that "the companies have no control over him." However, Peto did offer his assistance as an epidemiologist and statistician to help the congressmen and their subcommittee (of Oversight and Investigations) understand "statistically appropriate ways of interpreting the hypothesis-generating and hypothesis-testing data sets from trials."

Peto is referring specifically to his and others' recent analysis of cancer data from the ongoing Vytorin studies SHARP and IMPROVE-IT. The analysis was conducted in response to an observed association between the drug and cancer rates in the completed SEAS trial (background here). The conclusion of the analysis of Peto et al, which was recently published in the NEJM, is that there is "no credible evidence" that Vytorin (and in particular, ezetimibe) affects cancer rates.

The debate generated from the NEJM analysis rests on whether data from the 3 studies should have been combined to assess the potential cancer risk with Vytorin, as advocated by US cardiologist Steven Nissen, or whether it is inappropriate to lump data from the hypothesis-generating trial (SEAS) with data from the hypothesis-testing trials (SHARP and IMPROVE-IT), as urged by Peto and "any competent trial statistician."

In response to several questions regarding the NEJM analysis, Peto advised the congressmen, more or less, to read the article. Peto further declared that the analysis was undertaken independently of the drug companies and was coordinated by his Oxford colleague Rory Collins, in agreement with the various chairs of the Data Monitoring and Steering Committees of SHARP and IMPROVE-IT.

Other requests from the congressmen were answered with sharp rebukes, including a request for all correspondence between Peto's department, the Clinical Trial Service Unit (CTSU) at Oxford, and Merck or Schering-Plough. Peto replied that this request seems "to constitute inappropriate harassment," given that CTSU has conducted 4 major independent trials of Merck's cholesterol-modifying drugs for more than 10 years. Peto also denied the existence of a "secret" FDA report of his cancer analysis; nor did he send any advance drafts of the study results to Merck or Schering-Plough.

In a curious postscript, Peto requested that the subcommittee declare its own potential conflicts of interest, suggesting that Dingell, in particular, may hold a personal grudge against Peto and the CTSU. In 1997, Peto and others published an article in Controlled Clinical Trials, in which they described the procedures of the subcommittee in its investigation of the National Surgical Adjuvant Breast and Bowel Project (NSABP) as "inappropriate" and Dingell's statements as prosecutorial.

IMPROVE-IT = Improved Reduction of Outcomes: Vytorin Efficacy International Trial; SEAS = Simvastatin and Ezetimibe in Aortic Stenosis; SHARP = Study of Heart and Renal Protection.

Photograph of Richard Peto from CTSU website.