New Haven, Conn.-based Yale School of Medicine and Atlanta-based Emory University researchers found a weak association between FDA drug proxies and disease outcomes — putting into question the effectiveness of using proxies in the approval process.
A recent movement to speed up clinical trial conclusions and get new medications to patients has relied heavily on surrogate markers, or proxies, which are possible predictors of disease outcomes months and years in advance. Proxies allow studies to finish before it's known if the disease actually improves or if patients are likely to live longer, according to a May 28 Yale news release.
The study, published April 22 in the Journal of the American Medical Association, looked at 37 surrogate markers for 32 chronic diseases, such as hypertension and diabetes. The proxies were taken from the FDA list of surrogate markers that create the basis for prior drug approvals. The FDA does not approve proxies separately and once a proxy is used in a successful drug application, it is fair game for others targeting the same condition.
The analysis found that published meta-analyses did not exist for 60% of markers, and for the other 40%, the association between markers and disease outcomes was weak.
Joseph Ross, MD, professor of medicine and of public health at Yale and senior author of the study, said it is unclear why so many of the markers on the published list lack meta-analysis-based evidence, but the findings add a level of uncertainty to treatment choices and suggest that drugs approved based on proxies may not work.
He and his team said they plan to conduct meta-analyses of surrogate markers with insufficient documentation.