FDA to Consider Real-World Evidence

Randomized clinical trials have long been the gold standard of the FDA drug-approval process, but the agency recently outlined how it plans to incorporate another type of data into regulatory decisions: real-world evidence (RWE; www.fda.gov/downloads/scienceresearch/specialtopics/realworldevidence/ucm627769.pdf).

According to FDA Commissioner Scott Gottlieb, MD, RWE is generated by analyzing data collected from sources outside of formal clinical trials, including electronic health records, medical claims, product and disease registries, laboratory test results, and technology used with mobile devices.

“The collective evaluation of these data sources has the potential to inform clinical decision making by patients and providers, develop new hypotheses for further testing of new products to drive continued innovation, and inform us about the performance of medical products,” Gottlieb says.

The framework for the FDA's RWE program was developed to satisfy a provision of the federal 21st Century Cures Act. Per the framework, the FDA will consider RWE to support new indications for approved drugs and to support or satisfy post-approval study requirements. When evaluating RWE, the agency will explore whether real-world data are fit for use, whether the trial or study design used to generate RWE provides sufficient scientific evidence to answer a regulatory question, and whether trial/study conduct meets FDA regulatory requirements such as those for study monitoring and data collection.

“This framework is intended to advance the collection of data that are appropriate, consistent, and provide information and knowledge that can better inform regulatory decision making,” says Gottlieb.

David Hyman, MD, of Memorial Sloan Kettering Cancer Center in New York, NY, is enthusiastic about incorporating nontraditional data sources into the drug approval process. “The whole idea here is to create frameworkswhere we feel that the data are as reliable as we might collect for a clinical trial, and I think that's possible,” he says. He thinks that RWE will be especially important in precision medicine, where traditional clinical trials often aren't feasible because researchers are studying rare cancers and need to divide a limited number of patients into even smaller groups based on mutations or tumor types.

Hyman is involved in the American Association for Cancer Research's Project GENIE, a large-scale research effort linking genomic data with clinical outcomes for patients with a range of cancers. “Where GENIE really fits into this framework is in using real-world data to develop real-world evidence in genomically defined patient populations,” Hyman says. He hopes that the FDA will eventually consider data from GENIE when making regulatory decisions.

Charles Fuchs, MD, MPH, of Yale University in New Haven, CT, notes that RWE studies can often quickly and more efficiently study large numbers of patients, thus providing greater statistical power and capturing a more representative population than traditional clinical trials.

Fuchs says that RWE may be particularly useful for determining an appropriate drug dose: The dose initially approved by the FDA typically depends upon a randomized clinical trial, “but if subsequent real-world evidence suggests that patients don't tolerate that dose, and most patients get a lower dose, then that could inform the FDA approval.” He adds that RWE could also be used to support new indications for approved drugs and to compare multiple drugs approved for the same indication.

However, Fuchs cautions that RWE can have inherent methodologic weaknesses such as biases and confounding variables. “I think the FDA's effort to include real-world evidence is going to be beneficial to the process, but you have to be careful with what you're doing with it. I thinkthe keys to the realm continue to be randomized clinical trials when you're developing a new drug,” he says. –Catherine Caruso

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.