Making Sense of Sex in Clinical Trials
By Deborah Borfitz
November 12, 2025 | Roughly two-thirds of pivotal clinical trials leading to a drug approval by the U.S. Food and Drug Administration (FDA) now include women at rates either at or above the disease burden on the female population. Yet progress on their inclusion has “plateaued” since the FDA began releasing its Drug Trials Snapshots in 2015 because of “fundamental issues in not considering female biology,” according to Sophie Zaaijer, Ph.D., clinical and translational scientist at the University California Irvine and Riverside.
“We are stuck in diagnosis; we are stuck in treating women as if they are mini men ... we are stuck in our enrollment efforts for specific diseases and ... in disseminating sex-specific findings” linked to the labeling of marketed drugs, she says. Over the past three decades, after a 1993 edict by the National Institutes of Health (NIH) to include women in clinical trials, “we have picked the low-hanging fruit” by simply taking a headcount of women in studies.
That isn’t enough to bridge the gap between the setup of preclinical studies and clinical trial design, Zaaijer says, which will require considering the “alternate disease states” of women in clinical trial design. It’s a topic that she will explore in depth at the inaugural Health Executive & Research Summit (HERS) in San Diego next March, with four panelists illuminating different aspects of the problem.
For every research question, be it about molecular pathways or clinical applications, it is worth considering reasons women might be different than men by looking at those groups separately, says Zaaijer. Hormonal fluctuations across a woman's lifespan can dramatically influence the onset and progression of conditions that include cardiovascular disease, type 2 diabetes, and autoimmune disorders.
Stalled Progress
Zaaijer says her fascination with female health began as a preclinical researcher working on human cell models used in the discovery of new drugs. This was when she encountered the reality that both the sex and ancestral backgrounds of cell models are rarely considered (Cell, DOI: 10.1016/j.cell.2021.03.041).
And that got her thinking about what was happening on the clinical trial side of things to understand current enrollment patterns and the biomarkers being measured in women to identify the shortcomings, she continues. It helps to remember that a clinical trial is “the most rigorous tool to answer the question does the intervention work and is it safe ... for the people who will use it.”
The human population is highly diverse and perhaps the biggest subgroups are men and women, Zaaijer says. For decades, women were excluded from clinical trials based on fears about the childbearing potential of women. The NIH Revitalization Act of 1993 was designed to relieve that tension by coming up with safeguards that included the need for reproductive toxicology studies, pregnancy testing, and contraception protocols to make the inclusion of women safer.
As the FDA’s Drug Trials Snapshots has evidenced, the enrollment of women into studies relative to their disease burden has improved considerably. But, by Zaaijer’s estimation, women are still adequately represented in only about 67% of trials that led to FDA approved therapies over the past 10 years.
Progress has stalled for many reasons, but the essence of the entire problem is that current diagnostics capture only snapshots of disease rather than disease trajectories, as will be shared by HERS panelist Michael Liebman, Ph.D., managing director at digital healthcare modeling and analytics company IPQ Analytics. His take, Zaaijer says, is that “we are currently searching for our keys under a lamppost ... but not beyond that beam and therefore we miss a lot.”
Liebman claims that nearly 90% of drug failures can be traced back to how a disease is defined in terms of both diagnosis and the disease state of the people being tested. “This ... is much better understood for men compared to women.”
His position is that when diseases are diagnosed incorrectly, everything downstream—trial enrollment, drug dosing, and patient outcomes—is affected. Liebman’s work also shows that the application of artificial intelligence in drug development has been thwarted by the continued use of “old-fashioned, static diagnostics ... with male-dominated outputs,” Zaaijer shares.
Risk and Resilience
Meanwhile, over at Bayer AG, Kirsten Leineweber, Ph.D., is looking at linking patients’ symptoms to their genomic signatures to map “how small variants shape the risk and resilience to disease,” says Zaaijer. Leineweber, another HERS panelist, is a senior expert in human disease biology whose research focuses on pulmonary hypertension and early development within the therapeutic areas of cardiovascular, renal and immunology.
Women are genetically different than men in subtle but important ways that affect their disease risk and resilience, Zaaijer says. That is, they carry two X chromosomes, one of which is inactivated in each female cell, while men have only one to begin with that’s being expressed.
However, “X chromosome inactivation is not systematic through all cells in the body,” she points out. “We are sort of a mosaic.” But as women age, X chromosome silencing can erode, and previously quiet genes can start expressing. “This [awakening] is thought to also be associated with immune responses, cancer (Communications Biology, DOI: 10.1038/s42003-025-07691-y), and autoimmune disease (Science Advances, DOI: 10.1126/sciadv.adn6537),” per recent research that “should have begun years ago.”
Zaaijer’s point here is that genomic signatures are important to the linking of female-specific genetics and genomics with phenotypes—what is being called the “genomic rhythm” of female biology. But the biggest differences between men and women that needs to be considered when designing clinical trials is the hormonal state of individuals.
This is where Katie Baca-Motes, the CEO and co-founder of GSD (Generational Science & Discovery) Health Research, comes in. In her view, “hormonal health is the missing vital sign,” says Zaaijer, who couldn’t agree more. Women go through major hormonal changes throughout life, including perimenopause (the transition period) and menopause (the point after 12 consecutive months without a period) as well as monthly fluctuations earlier on with the start and stop of their monthly cycle.
These hormonal changes affect the inflammatory system, metabolism, and immunity, and are associated with many different symptoms in diseases that include migraine and arthritis, she adds. There is also “a whole list of drugs that are known to impact the female body differently when in different hormonal states.”
Not considering these biological realities in designing clinical trials can result in “noisy data of unknown origin” from women and impact study outcomes, says Zaaijer. Baca-Motes has done a lot of research examining the potential of wearables and digital tools (e.g., eDiaries) to provide “more fine-grained, longitudinal patterns” in trials that sponsors can include in their treatment response analyses.
‘Profitable Endeavor’
Rounding out the story arc on the topic of how clinical trials can better serve women’s health is Diana Torgersen, Ph.D., head of innovation ecosystem integration at Organon. She has thought a lot about the potentially vast number of clinical trials that have failed due to the hormonal variability of women and their changing chromosomal state as they age, Zaaijer says.
These notably include studies that have been shelved because no clear signal was detected, although women might have benefitted had the data been analyzed based on hormonal state, she continues. “[Some of] those treatments would have been helpful for women ... [and] pharma companies,” meaning lost opportunities to “do good but also make it a profitable endeavor” given that half the population represents a massive market.
The overarching message of Zaaijer and her colleagues is that women are not smaller men and inclusion is not the same thing as understanding sex-linked traits. “We need more fine-grained longitudinal tracking of human health in general, which is especially important in women but at the end of the day also important for men.”
Trial sponsors can now take advantage of FDA’s 2019 “enrichment guidance” offering opportunities based on guidance from the FDA allowing studies to enrich for “time-specific measurement points such as hormonal changes when assessing whether treatments work for women, notes Zaaijer. A major opportunity for female-aware trial design. Also; rather than retrospectively asking women if they had a migraine at the same time they were menstruating, for example, companies often do their measurements at a pre-determined point in women’s menstrual cycle and employ eDiaries to prospectively capture the timing of migraines and periods in a manner that isn’t subject to modification later—a practice that is ripe for adoption in other disease areas.
The “other side of the coin” is that clinical trial protocols for migraines are so well aligned with female physiology that they have an inclusion rate for women that is “very much higher than the disease burden,” Zaaijer says. “Now the men are a little bit underrepresented in those studies.”
But the days of simply counting women in trials need to end along with a move to the “next level” where longitudinal patterns of disease and hormonal states are instead counted, she says. Given the availability of wearables and eDiaries, and precision medicine companies like Tempus and Natera to do longitudinal tracking of genomic biomarkers, the time has at long last come.







Leave a comment