Evaluation of preclinical antipsychotic models used to support first‐in‐human clinical trials

Schizophrenia is regarded as a complex and heterogeneous psychiatric disorder, characterised by diverse symptoms and comorbidities, which complicate both clinical management and drug development. Current pharmacological treatment, primarily based on dopamine D₂ receptor antagonism or partial agonism, which has not markedly progressed since the emergence of chlorpromazine in the 1950s, remains inadequate in addressing the full spectrum of clinical symptoms. Despite decades of preclinical research, many novel compounds with different mechanisms that show efficacy in animal models subsequently fail in Phase II or III clinical trials. This translational gap may reflect limitations in model selection, reliance on behavioural endpoints with poor clinical correspondence and the inherent inability of rodent paradigms to capture the full heterogeneity of human schizophrenia. This review provides a systematic overview of how rodent models and behavioural assays have been applied by the pharmaceutical industry over the past 30 years to evaluate antipsychotic efficacy, both for marketed drugs and investigational compounds entering first-in-human (FIH) clinical trials. We examine the extent to which these models have informed regulatory submissions and clinical development, whilst also analysing the translational challenges that arise from their limited ability to capture the complexity and heterogeneity of schizophrenia, as well as the impact of inclusion criteria on the testing of antipsychotic drug efficacy. By highlighting these limitations, we propose key considerations for refining model selection, behavioural endpoints, and biomarker integration to strengthen the predictive value of preclinical research and improve the likelihood of success for novel antipsychotics in clinical trials.