By John Wilkerson, FDA’s Sentinel Initiative and other large data networks should be used to choose comparative effectiveness research projects, a cancer research advocacy groups says.
Many believe the government’s designs on comparative effectiveness are at odds with FDA’s emphasis on personalized medicine, so the Friends of Cancer Research is looking into how the two approaches could inform each other and not be seen as opposites.
The agency’s personalized medicine initiative is part of the Critical Path Initiative carried out by the Reagan-Udall Foundation, which was created by the FDA Amendments Act. FDAAA also created the Sentinel Initiative and authorized $25 million annually to create a data network, with the goal of collecting data on 100,000 patients by 2012.
The data from Sentinel is supposed to be on safety, and comparative effectiveness data would emphasize effectiveness. But Friends of Cancer Research Executive Director Jeff Allen said Sentinel queries could be run on efficacy, too. Allen said Sentinel is a good example of the type of data network that should be used to identify gaps in evidence on medical technologies and generate hypotheses for clinical studies based on subpopulation characteristics, which would help to further advance personalized medicine.
He did not know whether FDA is setting up Sentinel in such a way that it could be used to ask research questions about comparative efficacy. It might take revisions to how FDA manages data sources for Sentinel to be used for efficacy ends, he acknowledged.
Allen presented his recommendations on comparative effectiveness research March 20 at a public meeting held by the
Much of the comparative effectiveness research done so far has relied on literature reviews and meta-analyses of individual trials. These studies help synthesize information for clinical practice, but they cannot generate new knowledge beyond that included in the original studies and generally do not provide insights into the effectiveness of interventions in real-world settings, Allen said.
Without large data networks like Sentinel, researchers must create smaller networks each time they have a question.
The basic approach used by Sentinel should be used for comparative efficacy research, Allen said. Many private entities already keep large data sets on safety and effectiveness. The goal of Sentinel is to link those data sets so they catch signals of potential safety problems, which is a tall order. Congress gave FDA power to require observational trials if a safety signal is detected. Likewise, Allen said, Sentinel data could be mined to generate better questions on comparative efficacy that clinical trials would aim to answer. Once Sentinel and other data networks detect differences in response rates among patient populations, researchers could then design trials to study those biological differences.
Moreover, the use of harmonized data networks would help the public understand how comparative effective research projects are chosen and create an expansive collection of outcomes data for which comparisons of different treatment options could be performed. The information could also inform how research results are applied outside of clinical trials, providing better assessment of effectiveness in real-world populations.