He Says, She Says, NSAIDS

The much-in-the-news Cleveland Clinic announced yesterday it will conduct a four-year trial comparing painrelievers to determine if Pfizer's blockbuster Celebrex causes heart attacks. Vioxx and Bextra, similar drugs in the Cox-2 class, have already been pulled from the market. Pfizer will fund the $100 million study.

The study will be run by Dr. Steve Nissen, who chairs the FDA's Cardiovascular Drugs Advisory Committee. No stranger to industry-funded trials, Nissen says he turns all his fees over to charity.

The design of the trial seems straight forward enough: 20,000 people at risk of heart disease (they have diabetes, high blood pressure, etc.) will be divided into three groups: one will get Celebrex, the second will get ibuprofen and the third will get naproxen. With about 700 deaths normally expected in this at-risk group, any increase should be noticeable.

A couple of comments: the newspaper accounts of this trial suggest they plan to treat patients with Prilosec for gastrointestinal side effects. It was unclear if this was prophylactically or on an as-needed basis. Hopefully it will be the latter with good data collection so the public will finally learn if Celebrex is really better at preventing GI bleeds. The first time Pfizer paid for a study of this question, their hired researcher only reported six months of data. It later came out that full-year data showed no improvement over over-the-counter medicines.

Also, I hope Nissen adheres to the Journal of the American Medical Association's new guidelines for publication of industry-funded studies by hiring an independent biostatistician to analyze the data.