Sex as a biological variable in nonclinical studies: Bridging scientific rigor, animal welfare, and regulatory expectations

Nonclinical studies in animal models have often treated male animals as the default, based on the assumption that estrous-cycle hormonal variation in females increases variability and complicates interpretation. This convention is a structural limitation in study design, particularly in therapeutic areas where sex differences in pharmacokinetics (PK), pharmacodynamics (PD), and toxicity are substantial and clinically relevant. Clinical experience shows higher adverse drug reaction rates in women and recurring examples of sex-biased toxicities, underscoring the translational cost of limited sex consideration early in development. Sex differences in nonclinical PK and toxicity arise from defined mechanisms, including sex-dependent growth hormone secretion patterns that shape hepatic and renal enzyme/transporter expression, as well as sex-specific determinants of gastrointestinal absorption and cardiac electrophysiology. Omics analyses further show that sex effects are embedded in system-wide molecular networks across immune, metabolic, cardiovascular, renal, and neural systems, supporting a shift from descriptive to predictive and stratified animal study designs. Sex bias is also an animal welfare concern: male-only strategies can generate surpluses of unused females in breeding colonies, challenging the 3Rs-particularly Reduction-when evaluated at the project or facility level. Regulatory expectations are converging through International Council for Harmonization (ICH) principles, the National Institutes of Health (NIH) sex-as-a-biological-variable (SABV) policy, and reporting frameworks such as Sex and Gender Equity in Research (SAGER), with increasing emphasis on transparent sex reporting and analysis across nonclinical evidence streams. This review integrates scientific, welfare, and regulatory perspectives to provide a practical framework for sex-aware nonclinical study design, enhancing translational validity and ethical rigor.