Conflicted Medicine

Last year, arthritis patients were stunned to learn that two popular painkillers are dangerous. The meds were backed by studies conducted at medical centers with pharmaceutical-company funding, a situation that has left many wondering whether they can trust drug trials anymore.

It seemed like the ideal wedding of wealth and brains. Pharmaceutical and biotech companies have poured cash into universities and medical centers since the 1980s, funding research that has helped spur innovations including new cancer medicines and the hepatitis B vaccine. But today, critics are questioning whether money has corrupted academia's biomedical research mission--particularly in assessing the safety and effectiveness of drugs--and debating what, possibly, should be done to safeguard scientific integrity and public health.

Consider the stunning downfall of the painkillers called COX-2 inhibitors, including Merck's Vioxx and Pfizer's Celebrex, best-selling treatments for arthritis that turned out to boost the risk of heart attacks and strokes. Merck executives, who last year pulled Vioxx from the market, insist they acted with integrity in monitoring and disclosing the pill's cardiovascular risks. But critics say that Merck officials knew of the dangers as early as 2000 yet buried the information. Last August, a Texas jury found the firm liable in one Vioxx user's death, awarding a $253.5 million punitive verdict.

The catastrophe raises serious concerns not just about big pharma's practices and the U.S. Food and Drug Administration's (FDA's) ability to guarantee drug safety but also about the reliability of academic research conducted by investigators who receive funding from medicine manufacturers. The tale of the COX-2 inhibitors illustrates how "commercial interests can and do distort the scientific evidence that guides our medical decisions," write John Abramson, a family physician who teaches at Harvard Medical School in Boston, and Barbara Starfield, a health care researcher at Johns Hopkins University in Baltimore, Maryland, in the Journal of the American Board of Family Practice.

In November 2000, for instance, a clinical trial published in the New England Journal of Medicine showed that arthritis patients who took Vioxx experienced fewer stomach problems than did those who took naproxen. Heart attacks were more common in the Vioxx group, but the study's authors said that this elevated risk occurred only in those with previous cardiovascular trouble. The report helped clinch the drug's blockbuster status.

However, says Abramson, the article omitted crucial--and less favorable--information. The full data from the trial, made public at an FDA agency briefing in February 2001, showed that Vioxx users, regardless of their prior heart health, developed more total cardiovascular problems--heart attacks, strokes, and blood clots--and more serious illness overall than naproxen users did, Abramson says. All 13 study authors, he notes, either were Merck employees or had received grants or consulting fees from the company.

Disputing Abramson's analysis, Merck executives defend the journal article as a fair, accurate description of the study's results, saying that "any suggestion of bias is unfounded." Company officials say that the original publication focused specifically on heart attacks because those accounted for most of the difference in cardiovascular dangers between the two groups. The study's authors also argued--without any actual evidence, critics say--that naproxen protects the ticker: As a result, Vioxx appeared to boost cardiovascular risks by comparison. Furthermore, data from Merck's other trials had not found a heart risk.

Nonetheless, a cloud hangs over the company. And the questions about bias aren't unique to Merck. Industry affiliations are now widespread among academic experts who conduct drug studies and review experimental therapies. For instance, one-third of the members of the panel that advised FDA to keep COX-2 inhibitors on the market last February had financial connections with the medicines' manufacturers. If their votes had been excluded, the panel would have ruled the opposite way. Meanwhile, 35% of the researchers who've penned clinical practice guidelines for treating common ailments from headache to depression have conflicting ties to drug companies.

"You see this conflict of interest permeating all the sources of information"--journals, clinical guidelines, FDA--"that doctors have been taught to trust," says Abramson, a plaintiff's expert in several Vioxx lawsuits. As a result, he concludes, the medical knowledge that physicians and patients rely upon is often flawed.

Financial conflicts arising from the ivory tower's commercialization pose the most serious threat to research integrity today, says Frederick Grinnell, a cell biologist and founder of the Ethics in Science and Medicine Program at the University of Texas Southwestern Medical Center in Dallas. According to a survey of members of the Association of University Technology Managers, approximately two-thirds of educational institutions hold equity in start-up companies that underwrite research by their faculty. Studies show that about one in four academic biomedical investigators now receive industry grants, and at one typical top-notch institution, the University of California, San Francisco, a review of financial disclosure forms found that 8% of scientists own stock in, serve as paid advisers to, or hold other stakes in the companies sponsoring their work.

Evidence suggests that commercial funding and the potential for profit might sway investigators' findings, even if they're not consciously aware of it, Grinnell says. For instance, a survey of 70 articles and letters to journal editors about high blood pressure drugs called calcium channel blockers found that 96% of authors who favored the medicines had financial relationships with the pills' manufacturers--versus 37% of authors who took a critical view. More recently, a review of 370 clinical trials of drugs--for conditions as widely varied as warts and asthma--concluded that the odds were five times greater that studies would recommend a new medicine as the treatment of choice when their funding came from the drug's maker instead of a nonprofit sponsor. “It's very hard for people to dissociate themselves from those kinds of influences," says Grinnell.

Still, having a conflict of interest doesn't guarantee bad behavior, says bioethicist Thomas Murray of the Hastings Center in Garrison, New York. For instance, before Vioxx and Celebrex crashed, a few physicians who received research grants from Merck or Pfizer raised red flags about potential cardiovascular risks. And academic scientists often value their corporate partnerships not just for extra income but also for the chance to turn basic lab discoveries into medicines that benefit society, Murray says.

However, even the suspicion that study findings might be biased by monetary incentives can damage the public's faith in science, says Hastings Center ethicist Josephine Johnston: "You have a problem as soon as you get that perception." Although most academic investigators with industry ties are honest, says Eric Topol, a cardiologist at the Cleveland Clinic Lerner College of Medicine in Ohio and a critic of Vioxx, "we're at a nadir of public trust of medical research."

To restore confidence, he says, physician-researchers must severely limit industry's influence over how they execute studies and communicate results. Ideally, clinical scientists shouldn't consult or lecture on the corporate dime, adds Topol, who cut all his own ties with drug firms in 2004. He and others argue that FDA advisory committees should bar experts with conflicting industry affiliations. Congress is currently debating such legislation, but opponents say that stricter rules would make it difficult to find qualified reviewers because pharmaceutical firms tend to work with the leaders in a given medical specialty. Topol, however, counters that plenty of other knowledgeable investigators are available. Another agency, the National Institutes of Health, has already cracked down on corporate links this year: It has banned its employees from consulting for drug or biotech outfits, amid objections from critics who fear that excessively tough regulations will impede scientific discourse with industry researchers and trigger a brain drain from the agency.

Abramson, meanwhile, calls for major reform to weed commercial bias from our medical knowledge base. In his book Overdosed America: The Broken Promise of American Medicine, he proposes the creation of an independent board to oversee the registration and design of clinical trials, and to review and certify the full data from drug studies before publication in journals.

It's too late, Abramson notes, for medicine to divorce itself from industry. But like a team of stern marriage counselors, regulatory agencies and universities need to set some ground rules to ensure that academia and pharma can live in harmony and enjoy a fruitful, honest relationship.

Ingfei Chen, a freelance writer in Santa Cruz, California, now feels conflicted about taking any medicine.