Premarket approval (PMA) studies for coronary stents were found to not fully reflect the intended-use population (IUP) due to significant underrepresentation of minority participants and modest underrepresentation of women, according to research on the enrollment implications of site selection published April 28 in JACC.

Using pooled data from nine coronary stent PMA studies submitted to the U.S. Food and Drug Administration from 2003-2018, Wayne B. Batchelor, MD, MHS, MBA, FACC; Robert M. Califf, MD, MACC; Roxana Mehran, MD, FACC; and colleagues identified 8,859 U.S. participants across 196 sites classified by U.S. region, surrounding county demographics, teaching status, Veterans Administration affiliation, trial volume, female principal investigator (PI) involvement and number of acute hospital beds.

Among the total patient population, median age was 64 years, 30% were women and 88% self-identified as non-Hispanic White, compared with 12% as racial or ethnic minorities, including 6.0% who identified as Black and 4.1% as Hispanic.

Results showed that minority participants and women were underrepresented, with participant-to-prevalence ratios (PPRs) of 0.48 and 0.77, respectively (with 0.80 considered the threshold for adequate representation). This did not change significantly across the 15-year study period.

Racial and ethnic minority enrollment varied between sites and was predicted by West and South regions, county minority population, population density and per-capita income (R2=0.50; p<0.001). Female enrollment, on the other hand, showed less variation and was poorly predicted by any research site characteristics, including presence of female PI involvement (non-VA status only, R2=0.095; p<0.001), although this was so rare (<6%) that researchers noted the study may be underpowered to report on a possible relationship.

Additionally, researchers found that using modeling estimates that reallocated enrollment from low to high minority enrolling sites could move PPRs within the 0.80 adequate representation threshold. "A site selection strategy that accounts for regional and county-level demographics may normalize minority representation without compromising that of non-Hispanic Whites," they write. "However, the effect on enrollment of women would be minimal, underscoring the need for additional targeted interventions to enhance representation of women in cardiovascular device trials."

In an accompanying editorial comment, James C. Blankenship, MD, MACC, and Carolina Ponce-Orellana, MD, FACC, write that "the responsibility to include adequate representation of all patient groups in research extends to the entire scientific community."

They suggest that study sponsors could limit enrollment to minority populations, prioritize site selection on past enrollment performance, and focus on sites with female or minority PIs as additional ways to improve PPR – all answers to "a question that has not only been a thorny problem for decades, but that takes on new relevance in the current social climate of the [U.S.].