The pharmaceutical industry has long operated on a one-size-fits-all model, developing drugs primarily tested on, and thus best suited for, people of European descent. This approach ignores — and potentially harms — the billions of people of color on the planet. Lack of diversity occurs at all levels of the pharmaceutical ecosystem, from the makeup of C-suite and research staffs to participation in clinical trials. It even extends to preclinical research.
In 2015, the FDA began reporting on the representation of individuals in clinical trials, with the hope that raising awareness would drive meaningful change. This effort, however, has not yet yielded enough progress. Data from 2023 paint a stark picture of the ongoing racial disparity in clinical trials: among 4,522 people enrolled in 14 cancer drug trials, 62% were white, 23% were Asian, while only 2% were Black and 4% were Hispanic/Latino.
Take, for example, capivasertib (Truqap), made by AstraZeneca to treat HR-positive/Her2-negative breast cancer. The FDA’s 2023 approval of the drug was based on a clinical trial with 708 participants, 57.5% of whom were white of European/Caucasian descent; just 1.1% of participants were Black. That is surprising given that Black women are just as likely as white women to develop HR-positive/Her2-negative breast cancer but are much more likely to die from it.
In an effort to right this wrong, the FDA released its Diversity Action Plans draft guidance in June, outlining how pharmaceutical companies should work to increase clinical trial enrollment of populations that have historically been underrepresented in clinical studies.
But a major flaw in drug development pipelines continues to be overlooked: diversity in the preclinical risk assessments that precede human trials. These early evaluations, often based on human cell lines, help determine a drug’s safety and effectiveness. Researchers increasingly use human cell lines and miniature replicas of human organs grown in labs (organoids, spheroids, or microphysiological systems) to predict how new drug compounds might behave in humans. However, as I reported in a recent commentary in the journal Cell, the majority of human cell lines hail from donors of European descent. One landmark study published in 2019 indicated that, among the 1,000 most commonly used human cell lines in preclinical drug discovery, 62% were of European descent, 29% were of East Asian descent, 6% were of sub-Saharan African descent, and 2% were of Hispanic/Latino descent.
The ancestry of cell models used in experiments is rarely reported in scientific literature or elsewhere. The cell-based preclinical safety data therefore offers a biased, non-inclusive foundation.
A panel of cell lines established by the U.S. National Cancer Institute, called NCI-60, is made up of 60 human cancer cell lines from nine different tissues, and has been used to screen more than 100,000 compounds. Of these cell lines, 95% came from individuals of European descent, and 5% from those of sub-Saharan African descent.
This disparity is ironic, since the first human cell line came from cancerous cervical cells removed from a Black woman named Henrietta Lacks in 1951. The HeLa cell line — which Lacks did not know about or give consent to — has been responsible for numerous advances in biomedical research. However, a single cell line derived from one individual of sub-Saharan African descent does not represent the diversity of the Black U.S. population.
This ongoing monochromatic approach in preclinical R&D fails to capture the rich tapestry of human genetic variation, painting a portrait of humanity with a limited palette.
Prospective participants in clinical trials are currently unaware of this disparity and don’t request racially specific preclinical safety data. Yet this information is crucial for truly informed consent, in particular for individuals of under-represented, non-white groups. While the HeLa story unfolded in an era of limited technical expertise and holistic oversight, the narrative of inclusive preclinical data collection is now emerging at a time when holistic oversight and expertise are available.
The lack of collecting inclusive preclinical data is apparent for large pharmaceutical companies: none report this variable in their environmental, social, and governance (ESG) reports, diversity reports, or scientific publications.
A glimmer of progress emerged at this year’s Microphysiological Systems World Summit, where I heard Kimberly Homan of Genentech break new ground by presenting statistics on the ancestral backgrounds of human cell models used in the company’s research. This signals a shift towards greater deliberate consideration of genetic diversity in early-stage drug development.
The lack of readily available cell models representing African, South American, and South Asian genetic backgrounds, highlighting a problematic deficiency in tools for inclusive studies, represents a critical challenge in biomedical research. Developing a more diverse range of cell lines and implementing effective ancestry annotation policies should be an essential undertaking for the scientific community. Though this would require substantial time and effort, the potential benefits to medical research and health equity make this investment well worth pursuing.
According to the FDA, 20% of drugs show different effects across racial groups. The time and cost of ensuring that preclinical safety data matches the target patient population is minimal compared to the time commitment and expenses wasted on patient recruitment without representative preclinical safety data and the risk of clinical trial failure.
As the FDA’s guidance on diversity shakes up the status quo, forward-thinking researchers and pharmaceutical leaders should take it a step further by creating — and using — cell lines derived from diverse populations in preclinical research.
Sophie Zaaijer, Ph.D., is a life sciences consultant and the founder of FIND Genomics.