Schizophrenia is a complex neuropsychiatric disorder manifesting with diverse positive and negative symptoms as well as cognitive impairments. Current antipsychotics primarily address positive symptoms and frequently cause substantial side effects, highlighting the need for novel therapeutics targeting alternative pathomechanisms. Animal models are widely used to investigate schizophrenia's neurobiology and guide drug discovery, yet their clinical relevance for proof-of-concept (PoC) studies remains controversial. This Current Opinion critically examines the key limitations of animal models in schizophrenia research from a clinical perspective. The disorder's multifaceted nature, involving genetic, environmental, and neurodevelopmental factors, makes accurate replication in animals challenging. Additionally, core human-specific symptoms, like hallucinations and thought disorders, cannot directly be modeled, although some underlying cross-species constructs can be operationalized. We argue that animal models are most informative when used to test specific, well-defined mechanistic hypotheses and when readouts are anchored to human-relevant biomarkers and neurophysiology, rather than interpreted as proxies for diagnostic categories. Accordingly, translational utility is not uniform: predictive performance depends on the induction paradigm, the construct validity of behavioral and neurophysiological readouts, and the clinical endpoint being modeled. We advocate for close collaboration between preclinical and clinical researchers to refine existing models, establish new and translationally relevant paradigms, and clearly define their interpretive scope. Integrating complementary advanced in vitro and in silico models may enhance mechanistic understanding and help prioritize hypotheses and candidates. Still, these approaches should be viewed as adjuncts, not superior replacements, because they cannot yet capture the circuit- and systems-level dynamics relevant to schizophrenia.
Dammann-Bawadkji et al. (Tue,) studied this question.