Triggered by concerns over an increased risk of cardiovascular side effects, Merck announced on 30 September the withdrawal of rofecoxib (Vioxx), its leading drug for the management of rheumatoid arthritis. This decision prompts questions about the current drug regulatory system in the US and the profound influence of pharmaceutical companies on decision-makers, doctors and patients.

The US Food and Drug Administration (FDA) first approved rofecoxib, a nonsteroidal anti-inflammatory drug (NSAID) with inhibitory action against cyclooxygenase-2 (COX-2), for the treatment of osteoarthritis-related pain and inflammation, and subsequently to treat rheumatoid arthritis. Rofecoxib seemed advantageous over other NSAIDs such as naproxen and ibuprofen, which are associated with an increased risk of gastrointestinal ulcers and bleeding. A study—the so-called VIGOR trial—submitted by Merck to the FDA in June 2000 showed that patients taking Vioxx had fewer gastrointestinal side effects than patients taking naproxen. But crucially, the same study disclosed a four-fold greater risk of myocardial infarction in people taking Vioxx.

After discussing the data with its Arthritis Advisory Committee, the FDA ruled that new safety information from the VIGOR trial be added to the Vioxx label in April 2002. Nevertheless, these findings could have prompted the FDA to mandate a postmarketing clinical trial designed to specifically assess the cardiovascular risks associated with rofecoxib use.

A prospective phase 4 trial would have established whether the difference between the two drugs was due to a protective effect of naproxen on the cardiovascular system (as Merck originally argued) or to a true adverse effect of rofecoxib. Moreover, Vioxx was a fast-track product, approved by the FDA on an accelerated timeline. Fast-track status is granted to drugs that “treat serious and life-threatening illnesses and that provide meaningful therapeutic benefit over existing therapies.” Although the FDA Modernization Act of 1997 specifically stipulates that these fast-tracked drugs be subjected to postmarketing surveillance, Vioxx was not.

Subsequent epidemiological studies, which included an independent analysis of the VIGOR data, also hinted at possible cardiovascular complications in patients taking Vioxx. But the FDA still failed to impose on Merck the requirement of a postmarketing trial. It was only after Merck set out to study the usefulness of Vioxx in preventing the recurrence of colon polyps that alarms went off. The results from this trial showed that Vioxx was associated with a higher incidence of myocardial infarction and stroke when compared to placebo.

The reasons behind the paucity of FDA actions in this case are not clear, but it already faces accusations of silencing one of its own experts who had raised safety concerns about Vioxx (see p. 1149). It is also uncertain whether other COX-2 inhibitors on the market (such as celecoxib and valdecoxib) have similar cardiovascular effects. Pfizer, which markets celecoxib (Celebrex), has stood firm by its product, saying that three studies involving 6,000 people found no link between the drug and increased risk of heart disease, but the data will only become available next year.

Merck also showed strong support for its drug from the start of the controversy, repeatedly asserting the cardiovascular safety of Vioxx and disputing the accumulating epidemiological evidence. During the period the drug was in the market, Merck's direct-to-consumer advertising of Vioxx was intense despite the growing concerns with its safety profile, likely increasing consumption of the drug. In 2003, Vioxx accounted for 11% of Merck's revenue.

Direct-to-consumer marketing, an activity that is also regulated by the FDA, has been the subject of intense discussion since 1997, when new guidelines led to its current proliferation. Advocates of this form of advertising argue that it helps patients stay informed of the treatment options available to them. Opponents claim that it undermines the physician's authority to prescribe an alternative drug for fear that patients will take their business elsewhere.

The Vioxx case highlights another consequence of direct-to-consumer marketing—it increases consumption of drugs prior to assessment of their long-term safety. In the context of the concerns about the cardiovascular effects of rofecoxib, it is remarkable that the FDA did not act to regulate the marketing of Vioxx. And taking into account the availability of other NSAIDs to manage arthritis, it will be important to consider all the factors that led to the decision to grant Vioxx fast-track status in the first place.

Direct-to-consumer advertising is not the only way in which pharmaceutical companies exert influence on healthcare issues; they spend more money lobbying the US Congress than any other healthcare organization. A study published last April in the American Journal of Medicine shows that drug companies spend on lobbying more than twice the amount that physicians and other health professionals do and six times more than disease advocacy and public-health organizations.

In a climate in which the influence of pharmaceutical companies on healthcare decisions is so profound, the FDA must exercise its leadership and authority to keep the playing field level in the best interest of consumers. But cases such as the Vioxx withdrawal, the flu vaccination fiasco and the suicide risk in children taking antidepressants have undermined confidence in the agency. Regaining public trust should be at the top of the FDA agenda.