CRO

Data key to boosting trial diversity—and its time for CROs to act, says Phesi

Regulators' efforts to boost trial diversity are not enough, according to Phesi, which says data from unrepresentative studies should not be considered good quality.

The artificial intelligence technology firm made the comments this week to coincide with a new analysis of clinical studies conducted over the past 15 years. The analysis looked at 589,295 patients participating in 6,372 U.S.-recruiting cancer trials and revealed that 48% of those studies had no Hispanic or latin American representation, while 42% did not include a single Black patient.

The challenge is partly historical and partly to do with industry, says Phesi President Gen Li.

“African American, Hispanic and Latin Americans have not been well served by the industry to date, which impacts sponsors and CROs’ ability to recruit from these groups," Li said. "Secondly, many sponsors and CROs are not considering how to reach these communities.”

When a trial does not include protocols to recruit a diverse population, it "will simply not be diverse or inclusive," Li continued. 

The FDA called on industry to do more to improve diversity in clinical trials in draft guidance published in April. The agency encouraged clinical trial sponsors to define enrollment goals for underrepresented groups as early as practicable in clinical development. While this is a positive move, the FDA needs to do more, according to Li.

“The guidance is a welcome addition to helping to tackle diversity failings, but we know that the biopharmaceutical industry can take some time to respond to guidance due to the complexities of the sector," Li said. 

The guidance also lacks incentives and consequences, and therefore sponsors and CROs need to make changes to their processes to ensure that diversity is "built in," Li said. This could include use of real-world data for site selection and defining inclusive enrollment criteria. 

Phesi’s survey did uncover some positive findings.

“Despite the poor representation of minority groups at trial level overall, there is still a small percentage of trials that did very well in including various minority groups," Li said.