The spectre of Sanofi-Aventis' Ketek (telithromycin) continued to haunt the Food and Drug Administration as a cross-party alliance of US politicians asked Congress' Government Accountability Office to investigate the FDA's reliance on non-inferiority studies to approve antibiotics and other drugs.

Unlike the gold standard of the double-blind, placebo-controlled trial, a non-inferiority study compares the drug candidate with a known effective medicine. The aim is to demonstrate the equivalence of the two therapies (or non-inferiority of the new drug) and, by implication, the effectiveness of the product under review. While non-inferiority trials can have ethical and other advantages, they may also generate more equivocal results than a placebo-controlled study.

The call for GAO action came from four Democrats - Representatives Edward Markey, John Dingell, Henry Waxman and Bart Stupak of the House Energy and Commerce Committee - and from Charles Grassley, Republican chairman of the Senate Finance Committee. Grassley, Markey and Waxman have previously sought a Congressional investigation into the approval process for Ketek, amid allegations that the FDA glossed over evidence of fraud and misconduct in a key clinical study for the antibiotic. Ketek has been linked to a number of cases of serious liver injury and liver failure.

The Ketek affair has raised broader questions about the acceptance of non-inferiority studies as proof of effectiveness by the FDA's Office of Antimicrobial Products, noted the politicians in a letter to the GAO's comptroller general, David Walker. They pointed to concerns voiced by authorities such as the International Conference on Harmonization about the use, design and limitations of non-inferiority trials, particularly in relation to assay sensitivity and the “implicit lack of any measure of internal validity."

Representative Waxman put it more bluntly. “Permitting drug companies to take shortcuts in their clinical trials poses real risks to Americans,” he commented. “An antibiotic that is no better than a placebo can't fight off an infection. Worse, it can subject you to serious, even life-threatening side-effects without any compensating benefit.”

The politicians want the GAO to address the issue of non-inferiority trials in two phases. The first would provide details of all drugs approved by the Office of Antimicrobial Products over the last 10 years on the basis of non-inferiority studies, as well as the relative proportions of drugs for which efficacy was established from non-inferiority studies or placebo-controlled/other superiority trials.

In the second phase, the GAO would prepare a report analysing the FDA's oversight of, and reliance on, non-inferiority studies. Among the specific questions raised by Markey et al in this respect is whether and how the FDA has tackled the phenomenon of “bio-creep” mentioned in a 'Points to Consider' document published by the Division of Anti-Infective Drug Products in October 1992.

This refers to the selection of successively less effective comparator drugs in non-inferiority studies, with the end result that equivalence may be presumed for products that are in fact statistically and clinically inequivalent. The problem may be exacerbated by delays in relabelling products whose recognised effectiveness has been compromised by changes in resistance patterns and advances in scientific knowledge.

The politicians also asked the GAO about an internal regulatory briefing held by the FDA in July 2005 to discuss the use of non-inferiority studies as a measure of effectiveness in indications such as acute exacerbation of chronic bronchitis. By Peter Mansell