FDA Pushes the Envelope on Clinical Trial Design and Research Requirements

FDA is encouraging biopharma companies to adopt innovative research methods and drug development tools to reduce the cost and time of bringing new therapies to patients. Similarly, the agency is exhibiting more flexibility in approving cutting-edge therapies that follow less conventional and streamlined R&D programs. 

In July, for example, FDA staff supported advisory committee approval of the first CAR-T (chimeric antigen receptor T-cell) therapy, Novartis’ tisagenlecleucel, which was found highly effective in treating a rare leukemia affecting children, although safety and long-term effectiveness remain uncertain. More than 100

experimental CAR-T therapies are in development, generating optimism about greater future access to regenerative medicines to treat or cure a range of cancers and other serious conditions. 

FDA also indicated it may approve innovative drugs with less clinical data, as with Amicus Therapeutics’ new treatment for Fabry disease. A year ago, the company reported that FDA requested it to conduct another Phase III study for migalastat, which would have added two years to development. But after reviewing additional Amicus data, the sponsor said that the agency agreed to accept an application based on the current development program.

Earlier this year, FDA demonstrated a more flexible regulatory approach by expanding approval for the Vertex cystic fibrosis drug Kalydeco based on in vitro data, and not additional clinical trials. Another regulatory first was to authorize Merck & Co.’s cancer drug Keytruda for the treatment of tumors associated with a specific genetic abnormality, as opposed to cancers affecting parts of the body, such as lung or breast. 

FDA Commissioner Scott Gottlieb further proposes to spur development of treatments for rare diseases by improving the process for evaluating orphan drug designation requests, which have more than doubled in recent years. He told the Senate Appropriations subcommittee in June that a new Orphan Drug Modernization Plan will eliminate a backlog of 200 pending designation requests by mid-September and establish procedures for vetting such requests within 90 days in the future.    

Modeling and master protocols

Another strategy for accelerating clinical research is to make greater use of master protocols to evaluate multiple drugs or multiple diseases at the same time. Instead of designing and conducting individual studies for each investigational therapy, Janet Woodcock, director of the Center for Drug Evaluation and Research (CDER), encourages sponsors to support joint research collaborations able to test therapies more quickly and efficiently. Woodcock and Lisa LaVange, director of CDER’s Office of Biostatistics, recently wrote in The New England Journal of Medicine that launching such projects requires additional time and resources but can answer multiple questions more quickly and efficiently. Such master protocols have been developed for lung and breast cancer, but could also apply to infectious and neurological diseases and to therapies for rare conditions that affect small patient populations.

Computer modeling and simulation can significantly inform the design and outcomes of clinical trials, as highlighted by Commissioner Gottlieb in outlining FDA’s plan for implementing the 21st Century Cures Act in July. Greater use of in silico tools to evaluate new drugs and medical devices, explained the commissioner, may help predict clinical outcomes and safety issues, support evidence of effectiveness, optimize dosing, and evaluate potential adverse event mechanisms. 

The “Cures” bill provides $500 million over nine years for FDA to carry out its many additional tasks and initiatives, including the development and publication of new guidance on the use of in silico tools for advancing drug development, similar to draft guidance already issued on submitting physiologically-based pharmacokinetic model data in drug applications. The program also will support FDA’s Center for Devices and Radiological Health in building in silico regulatory models for product design and evaluation, including the development of a digital library of models and a family of “virtual patients” for device testing. Gottlieb noted that modeling and simulation may assist in creating natural history databases to support more efficient drug development, as seen in current collaborations on natural history models in Parkinson’s disease, Huntington disease, Alzheimer’s disease, and muscular dystrophy.

Such approaches also may support development of targeted drugs by modeling how individual physiology and genetics can identify patient subgroups that need dose adjustments. In silico methods, moreover, can help assess the effects of drug interactions, renal impairment and hepatic insufficiency in patients.

These efforts to advance state-of-the-art technologies involve collaboration with external parties able to provide expertise and infrastructure. And they require further FDA investment in high performance computing that would enable FDA review staff to manipulate large data sets. FDA’s Scientific Computing Board is poised to help expand the agency’s computing capabilities and modeling approaches to support these initiatives.