The Underrepresentation of Women in CNS Clinical Trials

 Introduction 

Between 1989 and 1993, the inclusion of women in National institute of Health (NIH) supported clinical research was policy but not statutory law. In 1993, congress wrote the inclusion requirementinto federal law through the NIH revitalisation act of 1993 (public law 103-43), specifically the section titled women and minorities as subjects in clinical research (National Institutes of Health, 2025).This act was passed because of several scientific, political, and social developments in the 1980s-early 1990s pushing the U.S congress to act.  

Women are not defined as minorities in the central nervous system (CNS) ) clinical trial space but are a part of required or suggested inclusion categories for recruitment of participants in research.These come from the U.S federal research inclusion policies, primarily developed by the NIH and reinforced by the NIH Revitalisation Act of 1993.  

The NIH introduced a still standing policy in 2016 requiring “sex as a biological variable” (SABV) in any preclinical and clinical research it funds. This expects researchers to consider sex differences in their design, analysis and reporting, requiring scientific justification. Private preclinical and clinical research has no equivalent federal mandate.  

For industry-sponsored research there is still no federal “inclusion law” like the NIH revitalisation Act. However, FDA approval requirements indirectly push companies to include women. 

The U.S. Food and Drug Administration (FDA) does not require demographic quotas either. However, it does: 

Despite policy revisions in the early 1990s, the inclusion of women in clinical research remains an obvious underrepresentation of actual patient populations. For instance, a 2020 analysis of data from ClinicalTrials.gov found that women represented 42% of trial participants in psychiatric disorders while they represent 60% of patients.  

Trials led by female principal investigators (PI) are more likely to recruit a higher percentage of women (mean of 54.1%) compared to trials led by men (46.9%) (Gender Diversity of research teams and clinical trial enrolment (JAMA network open, 2025). 

This demonstrates the feasibility of including women is not barriered by any real limiting inclusion/exclusion criteria but by the directive of PIs, sponsors, and investors to uphold scientific representation and inclusion of their target populations. 

Within the general clinical trial space, there are positive signs of improvement on the inclusion of women in research. Analysis of 4,616 clinical trial submissions showed that 90% of trials intend to recruit both sexes, while 3.7% are female-only and 6.1% are male only (Medicines and Healthcare Products Regulatory Agency, MHRA, 2025). Although the intended recruitment of both sexes, does not necessarily imply a 50/50 distribution within the trial population, it also does not account for the fact that recruitment targets (e.g., reaching 90% of a quota) are not directly comparable to the number of published reports or the final representation of females within a clinical trial population. 

Justifications 

All NIH-supported clinical research must include women unless a clear scientific justification explains why inclusion is inappropriate. The justifications given for inclusion of women being inappropriate is the main growing issue in clinical trials today.  

Inclusion and exclusion criteria for females in early-phase (Phase I and early Phase II) clinical trials are heavily influenced by reproductive safety, hormonal variability considerations, drug-drug interactions and historically, the lack of preclinical reproductive toxicity data.  

While FDA guidelines have evolved to encourage inclusion, many early-phase trials still disproportionately exclude women of childbearing potential to manage risk and legal liabilities. 

Inclusion criteria: Including women in early-phase clinical trials often requires meeting specific safety-related criteria, particularly related to reproductive risk. These may include: 

Exclusion criteria: Common exclusions applied to women in early-phase trials often include: 

Participant: prevalence ratio (PPR) 

PPR is a metric used in clinical trial analysis to evaluate whether the sex distribution of trial participants reflects the sex distribution of the disease in the real world. 

Formula: 

Key: 0.8–1.2 is ideal; below 0.8 suggests under-representation, above 1.2 suggests over-representation.  

Without this adjustment there lies a risk of: 

Example underrepresented CNS diseases:  

Epilepsy: A recent analysis of epilepsy trials registered in the clinicaltrails.gov database (2006-2022) calculated PPRs for sex representation. Among 86 epilepsy trials reporting sex data, females were under-represented (PPR<0.8) being included for 12% of trials (Bulter M, 2025).  

Neurodegenerative diseases: The representation of sexes varies significantly based on the specific neurodegenerative condition and its underlying epidemiology. However recent papers have demonstratedthat in a large analysis of Parkinson’s Disease (PD) the aggregated female PPR was 0.83. Women comprised 38.8% of PD trial participants versus their prevalence in the wider PD population (Tsai CC, 2025). 

Phase I significance  

The underrepresentation of women in Phase I clinical trials is more significant than other stages of clinical trials because of its focus on dose determination, pharmacokinetics (PK) and pharmacodynamics (PD), safety and efficacy. When women are underrepresented in Phase I CNS trials, the entire development pathway is built on incomplete biology. Sexual androcentrism at this stage of clinical trials is costly to later stage interpretation and patient population outcomes.  

Improper early dosing due to sex underrepresentation can: 

Takeaways  

The field of CNS research still underrepresents woman in early phase clinical trial populations, importantly at the most significant phase, Phase I.  By the frequent underrepresentation of women in Phase I populations women are the population at greatest risk of adverse reactions, side effects, and medications not working as intended. This could be overcome by designing trials for female biology. Moving away from sexual androcentricity, to models that include not only the female biology but the evolving variables of the female body over a woman’s lifetime. This could be addressed in several ways, such as increasing the Phase I study population to allow averaging of results across menstrual cycle phases between participants, or including women using contraception, as this reflects the intended target population. These sex differences are not limiting to research outcomes and should not be considered as such.  With small design changes, CNS drug trials can become more representative and scientifically robust.  

References: