FDA-Catalyst Alignment with the CMS Linkage to the PCORI RELIANCE Trial

In collaboration with the Office of Medical Policy (OMP) at U.S. Food and Drug Administration (FDA), the Sentinel Operations Center (SOC) will work with the Department of Population Health Sciences (DPHS) at the Duke University School of Medicine to link patients enrolled in the (Patient-Centered Outcomes Research Institute) PCORI RELIANCE trial to traditional fee-for-service Medicare data. The RELIANCE trial is a parallel, pragmatic, non-inferiority trial designed to evaluate chronic roflumilast versus azithromycin therapy for the prevention of chronic obstructive pulmonary disease (COPD) exacerbations among up to 3,200 adults across 20-40 academic medical centers, led by the PCORnet COPD Patient-Powered Research Network (PPRN).

Linkage of the RELIANCE trial to Medicare data will provide additional information on the primary outcome, all-cause rehospitalization or death, select secondary outcomes such as all-cause healthcare utilization and medication adherence; and will also provide an opportunity to test distributed regression methods with vertically partitioned data. This activity will support FDA’s assessment of the use of real world evidence to help support the approval of new indications for approved drugs which is mandated by the 21st Century Cures Act, as well as build and test infrastructure and methods to conduct pragmatic clinical trials.

This linkage project will occur in four phases: preparatory, planning, data linkage, and analysis. RELIANCE trial enrollees will be linked twice to Centers for Medicaid and Medicare Services (CMS): first with an interim Medicare data linkage to pilot the linkage process, and second with an end-of-trial Medicare data linkage. The pilot distributed regression analysis will use the end-of-trial data; the analysis will compare results obtained from distributed regression with the results generated from pooled patient-level data.