Drug companies should have less leeway to withhold clinical trial data from public scrutiny on the grounds of commercial confidentiality, argue US researchers from Boston’s Brigham and Women’s Hospital and the Harvard School of Public Health.

In particular, safety data could be more routinely disseminated with little threat of competitive damage to the product sponsor, claim Aaron Kesselheim and Michele Mello in the March/April issue of US journal Health Affairs. One way of achieving this would be to replace the Summary Basis of Approval issued by the Food and Drug Information with a “more informative public document”.

Earlier disclosure of comprehensive safety data on new medicines – and particularly on new uses of existing medicines where there are implications for off-label use – “might lead to more rapid detection of risks and decisions to withdraw or restrict unsafe drugs”, Kesselheim and Mello contend. If the FDA is not prepared to grasp this nettle, they add, congressional action may the only alternative.

The review of US confidentiality laws and the FDA’s disclosure policy comes in the light of several recent cases in which approved drugs were later found to pose serious safety risks. These range from Scios’ congestive heart failure treatment Natrecor (nesiritide) to the COX-2 inhibitors Vioxx (rofecoxib, Merck & Co) and Bextra (valdecoxib, Pfizer).

In the latter instance, the authors point out, Pfizer successfully petitioned the FDA to restrict disclosure of trial data for an acute pain indication. Only after Public Citizen filed a lawsuit (in 2004) was the relevant information released, showing that the FDA had reviewed multidose safety data for Bextra and found “an excess of serious adverse events including death” in that setting. The drug was nonetheless approved for three other indications – and subsequently withdrawn in April 2005 when additional studies linked it to an increased incidence of postoperative heart attacks.

Legal challenges from consumer groups have in some cases forced the FDA to ease its policy of strict confidentiality for data submitted as part of the drug approval process. Despite the new disclosure obligations imposed by the 1966 Freedom of Information Act (FOIA), the agency continued to limit the release of trial data by invoking the Act’s commercial and confidential information exemption, Kesselheim and Mello note. Moreover, FOIA requests are a “suboptimal vehicle” for disseminating safety data, given the need for invariably protracted and expensive litigation if the FDA and/or manufacturer turn them down.

The authors do not ignore efforts by companies and industry associations to deflect pressure for full data disclosure by setting up their own clinical trial registries. But “evidence suggests that such registries remain incomplete”, they comment. Last year the World Health Organization launched the International Clinical Trials Registry Platform, a global network calling for the registration of all trials on humans. However, industry has voiced concerns about disclosure of early-phase studies slowing development or weakening the patentability of new compounds.

For Kesselheim and Mello, revealing safety data would rarely pose a competitive threat, whereas the case for full disclosure of efficacy data is “less compelling”. Safety data alone “cannot support product approval, and their public health significance is particularly high”, they comment. The FDA and the courts should therefore “closely scrutinise claims that clinical trial data are confidential commercial information – for example, by requiring drug companies to make more specific showings of competitive harm”.

The FDA should also expand the routine disclosure of safety data beyond responses to FOIA requests, the researchers argue. Reliance on the Summary Basis of Approval “can be problematic because FDA rules allow those who submit the data to serve as architects of the disclosures that are made”. The authors believe the agency has the authority under the Food, Drug, and Cosmetic Act to make these changes on its own initiative. Failing that, it will be down to congressional initiatives such as the Enhancing Drug Safety and Innovation Act (S 484), recently re-introduced by Senators Edward Kennedy and Mike Enzi.

Proposals such as this, which would publicise the results of clinical trials within 30 days of FDA approval, “merit consideration by the full Congress”, Kesselheim and Mello say.