Is FDA Getting Impatient with Conventional Clinical Trials?
​​[Posted on: Thursday, September 28, 2017]

Last week at a conference on real world evidence (RWE) organized by the National Academy of Sciences in Washington DC, the FDA Commissioner, Dr. Gottlieb, and The CDER chief, Dr. Woodcock, emphasized the limitation of conventional clinical trials and the importance of RWE, promising new guidance documents and new regulatory paradigms to increase the use of RWE in regulatory decisions. Dr. Woodcock made the strongest statement by an FDA official that “I personally believe the clinical trial system is broken”. There is no doubt that with the wealth of health related information available from our smartphones, wearable devices, and electronic health records, we now have access to astronomical amounts of data that gives the most holistic information about one’s health than ever before. And we have the technology to analyze this data quickly, completely and, in many cases, automatically. It would be unfortunate to not take advantage of this information, particularly if it means better medical products and improved disease management. RWE has the potential to determine unmet medical needs and the value of intervention, improve health policy and resource allocation, guide development of new molecules, and evaluate the real effectiveness of new drugs. Particularly for new drugs for rare diseases, where the drugs are approved with “imperfect data”, RWE can be used to collect “perfect” post-market data to support or reject a new drug based “real clinical information” and make educated reimbursement decisions. However, RWE has major limitations due the practical “imperfections” in the data. First, the data could be highly biased based on the preferences of the physicians and manufacturers. There is little incentive for the prescribers and manufacturers to report data negative to their interests. This can be addressed to some extent by independent reporting of the RWD to FDA but there could be challenges in implementing such reporting. Second, RWE is usually able to capture only large effects due to the noise in the data. So, several clinically beneficial treatments which provide incremental improvement may not be helped by RWE. Third, RWE is practical for analysis of post-market data to find new uses, improve treatment regimens, and find practical hurdles to patient compliance. On the other hand, conventional controlled and randomized clinical trials are still the most reliable technique for pre-market investigations. There is some talk of using hybrid data, i.e., a mix of clinical trial and RWE datasets but FDA’s guidance for that is not expected soon. The current estimate is by the year 2021. And even then, best estimates based on the medical device RWE guidance show that FDA would want very strict criteria for using RWE for pre-market decisions making it hard, if not impossible to use RWE except for a few cases. So, for all intents and purposes, our estimate is that for new product developers, conventional clinical trials are here to stay. Stay tuned.