- Speech by
Robert M. Califf, M.D., MACC
(Remarks as prepared for delivery)
I want to express my deep appreciation for the work of the Reagan Udall foundation. The reports over the past year on the Human Foods Program, the Center for Tobacco Products and Improving Understanding of FDA and FDA Related Products have given us excellent advice and guidance for improving our work at the FDA. This new report we are discussing today, “Enhancing Post-Market Evidence Generation for Medical Products,” is every bit as important. The Reagan Udall Foundation is providing tremendous support for the mission of the FDA.
Prior to enactment of the Kefauver Harris amendment to the Food, Drug and Cosmetic Act in 1962, clinical trials were developing as a tool to advance medical knowledge, but they were not seen as essential to the development of medical products. The evidence generation system was transformed with that law, which required “evidence consisting of adequate and well-controlled investigations, including clinical studies, conducted by experts qualified by scientific training and experience to evaluate the effectiveness of the drug involved, on the basis of which it could fairly and responsibly be concluded by such experts that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended or suggested in the labeling or proposed labeling thereof.” This remarkable change in the law stimulated a revolution in clinical trial methods and quality due in no small part to the hard work of the FDA.
While the system of premarket clinical trials has plenty of room for improvement, it is fundamentally working well and successfully separates medical products with benefits that exceed risks for specific indications from medical products that have risks that outweigh the benefits. I want to be clear, that from my perspective, we’ll continue to work on improving the pre-market trials system, especially in areas where precision medicine is moving quickly, such as cancer and rare diseases and through the FDA’s accelerated approval pathways. But this report covers a different domain that we need to define and improve now. The time is right.
The premarket evaluation of drugs, biologics and devices does its job, but leaves many questions on the table after initial approval as these medical products move into clinical use. How does the new treatment compare with the other treatment options? Can treatments be combined, and if so, how? How long should the treatment be used? Is there a better dose? What is the value of the treatment? Are there characteristics of patients that identify those with excess or risk or special benefit? What about additional indications? The point of this report is that although the fundamental principles of clinical trials pertain to all trials, the best practices for implementation in the post-marketing phase are different. Unfortunately, the clinical trials enterprise has not differentiated these different purposes for trials in a manner that enables a robust, fit for purpose post-marketing evidence generation system.
These post-market questions stretch beyond the direct purview of the FDA and take us into territory that intersects multiple federal agencies, provider organizations, clinicians and payers, and patients have much at stake in getting this right. The net result of the fragmentation of responsibility in the post-market phase is that we have no national system to generate the evidence we need to optimize the screening, diagnosis and treatment among the multitude of medical products on the market. Yet we continue to spend over $4.3 trillion for health outcomes that are inferior to other high income countries1 , exemplified by our significantly shorter life expectancy compared with peers.2
While this post-market phase of medical products is not FDA’s primary responsibility, we do have clear responsibility for safety and evidence to support new indications for medical products. Our mission statement includes the following statement: “FDA is responsible for advancing the public health by helping to speed innovations that make medical products more effective, safer, and more affordable and by helping the public get the accurate, science-based information they need to use medical products and foods to maintain and improve their health.”3 Study after study shows that less than 20% of major recommendations in specialty society clinical practice guidelines are supported by high quality evidence.4 We cannot provide accurate, science-based information when the evidence doesn’t exist. Additionally, the recommendations in this report, if implemented effectively, could open up evidence generation for nutrition and dietary supplements.
The report is a compilation of practical tactics that have been promoted in previous reports, but the comprehensive nature of the 30 recommendations gives us an excellent list from which to reform the system. Let me offer a few brief and general comments on them.
The first two overarching recommendations call for a concerted, direct approach to improving the degree to which post-market trials are efficient and affordable by reducing unnecessary complexity. Above all else, we need to work together to make the conduct of clinical trials less burdensome for participants and clinicians. By applying principles of quality by design we should be able to enable many more trials to answer critical questions. While the mechanism will need to be worked out, we definitely plan to work across the different agencies of the Department of Health and Human Services (HHS) on these issues. The timing for this is excellent with the confirmation of Dr. Monica Bertognolli as NIH Director.
We take to heart the plea to create a more favorable regulatory environment for simplified, less labor-intensive post-market trials. As simple as it might seem, developing a lexicon that can be used by all concerned to describe different types of clinical research is a fundamental issue that we intend to implement. A joint FDA-NIH committee is already working on this problem and we plan to bring in the broader community during the course of this upcoming year. Simply put, quality by design requires that we develop approaches that are generalizable for different purposes—a Phase I trial of a new product demands different criteria for quality than a trial comparing products that are already marketed with clear profiles. As long as the same terms, like pragmatic clinical trial or real world evidence, have different meanings among different experts, we will continue to see fragmented and slow progress.
The implementation recommendations include a checklist of approaches that would improve site efficiency, reduce the burden on clinicians participating in clinical research, and reduce the immensely frustrating issue of variability in contracting practices among institutions. Investigators would likely welcome a central mechanism for credentialing that would reduce the mind-numbing redundancy of duplicating paperwork for each trial, sponsor, or NIH institute. Multiple previous efforts have failed to bring contracting into a rational systematic approach, but perhaps with the momentum of multiple government and private elements working together, this vexing variation can be dramatically reduced or eliminated.
The remarkable progress in digitization of routine work should enable significant simplification of study processes and data collection. The ubiquity of electronic health records, claims data, and wearable sensors should enable collection of more complex, more relevant data without the current massive human labor with its huge cost. This is a place where artificial intelligence and machine learning should lead to the development of much more ability to “clean up” data and develop computable phenotypes to describe clinical status and endpoints. Across HHS, including the Office of the National Coordinator for Health Information Technology, NIH, FDA, and other agencies, we need to organize the library function to make sure these advances accrue for public good.
The recommendations for improving patient or participant recruitment and enrollment are sensible. The issue of consent continues to be frustrating, as there is general agreement that people would prefer simpler, less complex consent documents and, like many aspects in our society, a digital approach offers a more interactive capability that can be adapted to the individual. However, progress has been limited. I urge more attention to the option of posting consent forms on clinicaltrials.gov; sunshine and research on these documents would likely stimulate needed reform.
The recommendations on payment fall outside the primary remit of the FDA. However, as we advocate and contribute what we can to the effort, there is wisdom in the concept that everyone can pitch in and that there is a compelling societal rationale for answering critical questions that currently go unanswered. While the medical products industry should do its part, there is also wisdom in uncoupling the current product-sponsor approach, since medical products companies have direct financial disincentives from addressing questions that may reduce their revenue or profit.
In summary, while we will continue to improve our premarket clinical trials system for medical products, now is the time to make major renovations to our post-market system.
Technology is no longer our limitation. All of us would benefit from more knowledge of which medical products should be used for which patients and populations at a given point in their clinical journey.