Most Drug Approvals by FDA are Based on Clinical Trials in White People
​[Posted on: Thursday, February 16, 2017]

There is significant lack of diversity in clinical trial populations. FDA approved 67 new drugs in 2015 and 2016, and of the total trial populations, 5-7% were African-Americans, 11-12% were Asians, while 76-79% were White. Women were more equally represented with about 40-48% study populations being female. The situation was worse in certain indications like oncology and cardiovascular indications with 1-2% and less than 5% African-Americans, respectively. Indications of the central nervous system did better with about 40% of the subjects being African-American. These trial populations do not represent the real population for any of the diseases for which these drugs were approved and it is reflected in the real life experience with these drugs. These data also indicate some interesting facts about clinical trials. The very high caucacian participation indicates that most of the pivotal clinical trials used for market approval are conducted in the US and Western Europe. Clinical trials in Asia and South America have not yet become significant contributors to FDA approvals. It also indicates that FDA’s efforts to encourage minority participation are still lagging. From a regulatory point of view, this confirms that one does not require diversity of population to seek market approval giving companies no incentives to push for diversity in their trials. At the same time, it point to a significant opportunity of untapped pools of potential clinical trials participants that could boost recruitment and quality of clinical trial data. It is a regulatory conundrum for FDA as legally it cannot require a sponsor to increase diversity in its clinical trial population while being fully aware of the limitations of the data provided. Recently, there has been a push towards using real world evidence to support post-market commitments and even pre-market data. The lack of diversity in clinical trial populations creates a stronger case for robust collection of real world evidence that would likely be more representative of the real population for a disease.