Data-driven Neurotrophic Phenotypes in Schizophrenia: A Principal Component and Cluster Analysis with Mediation of...

To investigate whether disrupted neurotrophic signalling contributes to schizophrenia pathophysiology and antipsychotic response, focusing on brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF). This study used principal component analysis and cluster-based phenotyping to identify neurotrophic and clinical response patterns across antipsychotic treatments.

In a 12-week prospective three-arm cohort, 90 adults with schizophrenia received risperidone (n = 30), aripiprazole (n = 30), or clozapine for treatment-resistant schizophrenia (n = 30). Plasma BDNF and NGF were measured at baseline and week 12 by ELISA, and symptoms were rated with the Positive and Negative Syndrome Scale. Principal component analysis was applied to eight standardized demographic, biomarker, and symptom-change variables, followed by k-means clustering. Structural equation modelling evaluated BDNF and NGF as mediators of treatment effects on clinical response.

All antipsychotics significantly increased BDNF and NGF, with significant drug-by-time interactions. Three phenotypic clusters were identified: excellent responders with moderate neurotrophic changes, treatment-resistant patients with elevated NGF but poor symptom improvement, and intermediate responders with minimal BDNF increases. Mediation analysis indicated that BDNF changes accounted for most of the treatment effect on clinical response, whereas NGF did not show significant mediation.

BDNF appears to be a key mechanistic mediator of antipsychotic efficacy, while NGF reflects pharmacological exposure without directly translating into symptom improvement. These findings support BDNF-informed phenotyping to refine treatment strategies in schizophrenia.