Women, black and Hispanic/Latinx participants were underrepresented in pivotal clinical trials for drugs approved from 2007 to 2017, according to a new report by the Tufts Center for the Study of Drug Development.
In the pivotal clinical trials, 44.9% of patients were women, short of the 48.6% that would be expected based on sex distribution by disease prevalence and population census. Participants who identified as black or of African descent were the most underrepresented participant group, representing 5.4% of participants in clinical trials rather than the 15.6% expected based on disease prevalence and population census.
Hispanic/Latinx participants and participants that fell in the "other" category were also underrepresented.
White participants were overrepresented in clinical trials, making up 75.9% of participants, compared to an expected 67.1%. Asian patients were also overrepresented on an overall basis "due, in part, to global demographic requirements and classification practices in which an individual's race and/or ethnicity may be categorized differently in different parts of the world," the report notes.
Looking at the individual drugs, though, 59% of approved drugs had a 20% or greater underrepresentation of Asian participants. Hispanic/Latinx and "other" participants were underrepresented by 20% or more in studies for 62% and 76% of approved drugs, respectively. Like the combined results, black participants had the greatest underrepresentation on an individual drug basis, with 80% of approved drugs having a 20% or greater underrepresentation of black participants.
Some therapeutic areas were especially troublesome. For example, all the drugs approved for pulmonary/respiratory and rheumatology diseases had a 20% or greater underrepresentation of black participants, and 88% of drugs approved for neurology disease were underrepresented for black participants. Hispanic/Latinx participants, on the other hand, were underrepresented in oncology, with 93% of approved cancer treatments having a 20% or greater underrepresentation of Hispanic/Latinx participants.
The data didn't come from all the clinical trials of approved drugs because demographics weren't always reported. While 89.7% of 757 pivotal clinical trials had data on the sex of the participants and 72.8% reported the race of participants, only 36.7% reported the ethnicity of participants. Earlier studies in the clinical programs of the approved drugs were less likely to report the breakdown of participants, with just 45.5%, 27.4% and 13% of the 4,723 studies reporting data on sex, race and ethnicity, respectively.
Looking over time, the diversity of women in clinical trials improved, going from a -5.1% disparity for studies of drugs approved from 2007-2010 to a 0.1% disparity for studies of drugs approved from 2014-2017. Hispanic/Latinx participants also improved from -44.8% disparity for studies of drugs approved from 2007-2010 to -8.8% disparity for studies of drugs approved from 2014-2017. "No improvement was observed in the underrepresentation of black participants during the entire 2007-17 period," the report notes, with disparity hovering around -50%.
FDA encouraging diversity
In June 2019, the FDA issued draft guidance for sponsors with recommendations on how to increase enrollment of underrepresented populations in their clinical trials.
While some populations have to be excluded for safety reasons and medically complex patients may have adverse clinical events that make it hard to determine if the events are drug related, "certain populations are often excluded from trials without strong clinical or scientific justification (e.g., the elderly, those at the extremes of the weight range, individuals with organ dysfunction, those with malignancies or certain infections such as HIV, and children)," the guidance notes.
The guidance encourages sponsors to improve the eligibility criteria to help increase the diversity of clinical trials. For example, it may be possible to eliminate or modify restrictive phase II criteria when moving to phase III studies, especially if additional data from other studies conclude the restrictions are unnecessary.
Drug metabolism and clearance can vary across populations – the elderly or patients with liver or kidney dysfunction, for instance – so the agency recommends sponsors characterize the pharmacokinetics in specific populations early in development, eliminating needless exclusion in larger studies.
The FDA notes that adaptive clinical trials can help with study diversity if sponsors start with a narrow population that's expanded to a broader population based on interim data. Alternatively, a company might start broadly with an adaptive design set up to make enrollment more restrictive if the interim data meet prespecified criteria.
Even when populations for primary endpoints need to be narrow, studies can enroll broader populations, from all disease stages or syndrome presentations, for example, and only look at secondary endpoints for the broader population. "This approach allows the study to utilize enrichment to help demonstrate effectiveness while also providing information on effectiveness and safety in a broader population and not decreasing the chances of achieving success on the primary clinical endpoint," the guidance notes.
Beyond inclusion criteria, the FDA suggests sponsors improve logistical issues for participants and other factors that may limit participation in clinical trials. For example, reducing the frequency of study visits, or substituting electronic communication for the visits, can increase the diversity of participation by making it easier for patients who may have travel issues.
Sponsors can also make sure potential participants know about the availability of financial reimbursements for expenses, which may allow participation by lower-income patients. There's also potential to pay participants in exchange for their participation in clinical trials, although the agency notes that sponsors should discuss payments with institutional review boards about the appropriate level of payments.
Since participants may not want to travel far for study visits, locating clinical trial sites in areas with a higher concentration of racial and ethnic minorities can improve diversification in studies. The FDA also notes that recruitment events in the community – at churches or barbershops, for instance – as well as at times outside of normal working hours, can help improve diversity of participants.
Finally, participants from an underrepresented group already in the study may be a good resource for sponsors to understand why they enrolled in the study and the challenges they face. The information can then be used to improve the recruitment of additional patients from the underrepresented group.