Most clinical trials fail to reflect America’s racial and ethnic diversity

A new study finds just 6% of clinical trials used to approve new drugs in the U.S. reflect the country's racial and ethnic makeup, with an increasing trend of trials underrepresenting Black and Hispanic individuals.

The findings arrive amidst a push for personalized medicine, which creates treatments designed specifically for an individual's genetic makeup. 

Researchers at UC Riverside and UC Irvine examined data from 341 pivotal trials-the large, final-stage studies used to gain FDA approval for new drugs-between 2017 and 2023. They observed a decline in Black and Hispanic enrollment beginning in 2021, even as calls for greater equity in science and medicine intensified. Asian representation increased over this period, while white participation remained largely stable.

Precision medicine relies on understanding how genetic differences influence treatment outcomes. If clinical trials under-sample large segments of human genetic variation, critical signals for safety and efficacy may be missed."

Sophie Zaaijer, geneticist with both UCR and UC Irvine, and co-lead author of the study

Zaaijer and co-author Simon "Niels" Groen, a UCR geneticist, argue that while ancestry alone shouldn't guide clinical treatment decisions, it plays a critical role in the early stages of drug development. People from different backgrounds often carry different versions of genes, called alleles, that affect how the body responds to medications.

"When a trial includes only a narrow slice of humanity, we can't be confident a drug will work - or be safe - for everyone it's meant to help," Groen said.

Clinical trials used to approve drugs in the United States are conducted both in the U.S. and in other countries that follow International Council for Harmonisation, or ICH, standards. While this ensures consistency between trials and speeds up approvals, it also concentrates evidence in a few regions such as the U.S., Europe, China, and Japan. 

Sub-Saharan Africa and much of Latin America, which host less than 3% of pivotal trials, are often left out of the data that shapes medicines used by millions of Americans.

This could be changing for Hispanics. Brazil joined ICH in 2016, followed by Mexico in 2021 and Argentina in 2024. Expanding trial networks to these and other underrepresented regions may help future studies better capture the genetic variation of patients worldwide.

Zaaijer began this line of research as a postdoctoral fellow at Cornell Tech, studying how little human genetic diversity is taken on board in preclinical drug development when patient-derived cells are used to model disease and test potential therapies. 

"I kept wondering," Zaaijer said, "If our preclinical models are this skewed, what happens once those drugs move into clinical trials?" Bias in preclinical models is an early warning sign, but bias in clinical trials becomes medical practice, she noted. 

Her collaboration with Groen's lab developed naturally. His lab at UCR studies how tiny worms metabolize plant toxins, and the parallels with human biology are striking. "Many of the same genes used to break down chemicals in worms are also involved in drug metabolism in humans," Groen said.

"The genes worms use to detoxify chemicals are ancient," Groen continued. "We carry many of the same ones. But small natural variations in forms of these genes can have a big effect."

Published in Communications Medicine, the study offers several recommendations: set diversity goals at the beginning of the drug development pipeline at the preclinical stage, choose testing locations that reflect the health needs and genetic backgrounds of local populations, and collect biological samples, such as blood or saliva, that can help researchers understand how people's bodies react to a drug. 

Even as DNA testing becomes more common in doctor's offices, the researchers stress that realizing the full promise of personalized medicine depends on stronger, more ancestry-aware data from the start.

"Precision medicine becomes possible only when clinical trials map the biology of all patients, not just a subset," Groen said. "Our analysis could offer a roadmap for how to get there."