This letter, by Health Research Group Researchers Elizabeth Barbehenn, Ph.D., Larry Sasich, Pharm.D., M.P.H., Deputy Director Peter Lurie, M.D., M.P.H. and Director Sidney Wolfe, M.D., appeared in the Lancet on July 15, 2000.
Sir – Kristina Schoonjans and Johan Auwerx stated in their review (March 18, p 1008) of the thiazolidinediones (a family of diabetes drugs) that “much of the information discussed here has had to come from company websites”. We sympathise with the authors’ difficulty in finding peer-reviewed data on these drugs. This stems from the failure of drug companies to publish their findings – it appears that only one of 11 efficacy studies submitted to the US Food and Drug Administration (FDA) as part of the approval packages for rosiglitazone and pioglitazone has been published.
However, drug company websites are not the only, and certainly not the best, place to find unpublished data on drugs. Consulting readily available US governmental sources on the internet can lead to different conclusions regarding drug safety and efficacy. Using FDA data sources, we have just completed our own assessment of the thiazolinediones. This research resulted in a petition to the FDA to relabel all three drugs to state that they are not only less efficacious than already-approved drugs, but appear to have class toxicities, particularly cardiac failure (www.citizen.org/hrg1514; accessed June 23, 2000), a serious finding not previously adequately acknowledged in the product labelling or on the company websites.
These conclusions could never have been reached had we depended upon the company websites alone. Unlike the company websites, which provide only the product labelling and superficial information about the drugs, the FDA website includes detailed summaries and assessments of the actual data submitted in support of the New Drug Application. Reviews of pre-clinical studies and clinical trials submitted to the FDA for troglitazone, rosiglitazone, and pioglitazone were available to Schoonjans and Auwerx through a Freedom of Information Act (FOIA) request (http://www.fda.gov/foi/foia2.htm; accessed June 23, 2000). For some drugs approved after 1997, reviews can be downloaded directly from the agency’s web site (http://www.fda.gov/cder/approval/index.htm; accessed June 28, 2000). Use of the FOIA is not limited to U.S. citizens.
In the past, FDA clinical, statistical, and pharmacology reviews were only made publicly available after a drug was approved, severely limiting scientific debate over the appropriateness of drug approval. As a result of a lawsuit we brought against the agency, the materials distributed to FDA Advisory Committee members before approval, including summaries of the FDA reviews, are now posted on the internet the day before the Advisory Committee meeting (http://www.fda.gov/ohrms/dockets/ac/00mtbc.htm; accessed June 23, 2000). The transcripts of these FDA Advisory Committee meetings are another valuable source of information (http://www.fda.gov/foi/electrr.htm; accessed June 23, 2000).
The failure of Schoonjans and Auwerx to consult these free sources, and their reliance instead on self-serving data from the company, seriously undermines the credibility of their review. Public Citizen’s Health Research Group is a non-profit, consumer research and advocacy organisation.