EMA dossier review puts sex representation in central nervous system drug development under the microscope
A retrospective cross-sectional study in central nervous system medicine development examined how sex was represented in the regulatory evidence base submitted to the European Medicines Agency between 2012 and 2022. The work focused on marketing authorisation dossiers, a source that reaches beyond published trial reports and into the material regulators actually review when judging safety, efficacy and product labeling.
That distinction matters because much of the public conversation about trial diversity stops at journal articles. Regulatory dossiers can contain trial datasets, sponsor responses to agency questions, and product information that never appears in the published paper. For a field such as neurology and psychiatry, where disease burden, pharmacokinetics, adverse-event profiles and therapeutic response may differ by sex, those documents can reveal whether sex-specific evidence is being generated, analyzed and presented with enough precision to support confident prescribing.
The study looked across phase 2 and phase 3 studies, a design choice that allows comparison at two critical points in development. Phase 2 work often shapes dose selection and signal detection, while phase 3 programs determine the evidence package used for authorization. When sex imbalance appears early, it can cascade into later development, leaving pivotal trials underpowered for meaningful subgroup analysis.
Regulatory material offers a clearer view of what developers actually submitted
Marketing authorisation dossiers provide a rare window into the full submission ecosystem. Investigators can examine the trial data themselves, the questions raised by regulators, and the final product information that clinicians later rely on in practice. In a domain as heterogeneous as neurology and other central nervous system disorders, that access matters because published reports may compress sex-specific details into a sentence or omit them altogether.
The study also reflects a broader historical problem in biomedical research. Sex differences in health and disease have been recognized for decades, yet preclinical and clinical research has often used a default male model, both biologically and statistically. In practice, that can distort dose selection, blur adverse-event patterns, and weaken confidence in how a drug will perform across the full treated population. The regulatory dossier is one of the few places where those omissions can be traced back to the source material.
A drug developed on an incomplete biological template is a risk multiplier, not a neutral shortcut.
Why sex representation matters in CNS medicine
Central nervous system disorders create a particularly difficult testing environment. Symptoms can be subjective, endpoints may depend on scales that embed clinician judgment, and treatment response often varies with age, comorbidity, hormonal milieu, metabolism and concomitant medication use. Sex-related differences can therefore show up not only in enrollment patterns, but also in dose tolerability, discontinuation rates, exposure levels and the interpretation of benefit-risk balance.
The abstract makes clear that the investigators were not just counting men and women. They were also trying to assess sex-related differences in the development pathway itself. In regulatory science, that can include how trials define sex categories, whether analyses are stratified, whether reproductive subgroups are separated, and whether product information reflects clinically meaningful differences. When those elements are handled inconsistently, the result is not just a statistical gap. It can affect the instructions physicians receive at the bedside.
The study’s framing is especially relevant because sex and gender are not interchangeable variables. Aggregate data may classify participants in broad categories while leaving reproductive status, gender identity and other clinically relevant distinctions unresolved. That limits the precision of any downstream analysis and makes it harder to know whether a signal is truly biological, social, pharmacologic or a mixture of all three.
What the study design can and cannot answer
Several features strengthen the analysis. Access to regulatory dossiers permits direct examination of submitted evidence rather than relying solely on the published record. Inclusion of both phase 2 and phase 3 studies also gives a more complete look at the transition from exploratory development to registrational testing. For regulators, clinicians and sponsors, that stepwise view can identify whether sex representation improves, stalls or worsens as programs mature.
The same design also carries important constraints. Restricting the review to centrally authorised products means the findings may not apply to every clinical development program or every authorisation route in Europe. Medicines approved through different pathways may have different sponsor strategies, different disease areas, and different evidence expectations. Even within centrally authorized products, the epidemiologic sources used to judge whether a trial mirrors the affected population can vary substantially, making direct comparison difficult.
Heterogeneity in those epidemiologic sources complicates any attempt to determine whether trial enrollment matches real-world disease burden. A condition may appear to enroll too few women or too few men, but the answer depends on which population estimate is used, how the condition is defined, and which country or region is considered. In cross-border regulatory science, even small definitional shifts can change the apparent degree of underrepresentation.
Reproductive subgroups and gender identity remain difficult to measure
The abstract also points to a persistent blind spot in clinical research: aggregate data and inconsistent sex categories limit the assessment of reproductive subgroups, gender identity and the precision of sex-specific clinical analyses. Those limitations are not administrative details. They determine whether a development program can detect pregnancy-related safety concerns, menopause-related pharmacology, differences tied to hormonal status, or adverse effects that emerge only in narrowly defined subgroups.
Without consistent definitions, subgroup analysis becomes unstable. Without enough participants, even a well-defined subgroup may be too small for inference. And without prespecified sex-specific analyses, investigators may never ask the most clinically relevant question: whether the same dose, endpoint and safety profile apply equally across sexes.
A regulatory dossier can only answer the questions sponsors chose to ask before submission.
Why this matters for routine prescribing
Clinicians depend on labels, summaries of product characteristics and trial data to guide treatment decisions. If sex-related differences are not adequately characterized during development, the uncertainty is transferred to everyday prescribing. That uncertainty can influence dose selection, monitoring intensity and the threshold for suspecting a drug-related adverse event. In central nervous system medicine, where therapeutic windows may be narrow and treatment response is frequently individualized, those downstream effects can be substantial.
The study therefore sits at the intersection of regulatory review and bedside care. Its value lies not in a single outcome measure, but in exposing how sex representation is handled before a medicine reaches the market. Even without the numerical results in the abstract excerpt, the framework itself is important: it tests whether current development practices are producing evidence robust enough to support sex-aware regulation.
The unresolved issue is not whether sex differences exist. The harder question is whether development programs are designed tightly enough to measure them with confidence, across the full path from phase 2 signal detection to phase 3 authorization dossiers. The abstract makes clear that the answer depends on better definitions, more consistent subgroup reporting and a regulatory culture that treats sex as a core analytic variable rather than a background demographic field.