# Comparison of In Vitro Metrics With Real-World Risk of Drug-Induced Parkinsonism Due to Antipsychotic Drugs: Retrospective Cohort Study

## TL;DR

This study compares lab-based drug metrics with real-world risk of antipsychotic-induced parkinsonism, finding that combining receptor binding and brain penetration data improves prediction accuracy.

## Contribution

The study introduces a novel composite metric combining receptor kinetics and BBB penetration to better predict real-world DIP risk from antipsychotics.

## Key findings

- Haloperidol had the highest DIP risk (HR=4.56), while aripiprazole had the lowest (HR=2.11).
- Metric 4 (pKr × BBB penetration rate) showed the strongest correlation with real-world DIP risk (R²=0.95).
- Including aripiprazole reduced the correlation strength (R²=0.58) due to its partial agonist mechanism.

## Abstract

Drug-induced parkinsonism (DIP) predominantly occurs due to antipsychotic drugs (APDs) blocking dopamine D2 receptors (D2Rs). However, in vitro assays often fail to fully reflect real-world variability in clinical outcomes.

This study aimed to evaluate whether in vitro pharmacological metrics correspond to real-world risk of DIP associated with APD use.

For 8 commonly used APDs, key in vitro parameters—including inhibition constants (Ki) of D2Rs and the serotonin 2A receptor, reversal rate (Kr) of D2Rs, and blood-brain barrier (BBB) penetration rate—were compiled to construct 6 composite DIP risk metrics. The real-world DIP risk was assessed using the Seoul National University Hospital common data model (2002‐2021). APD users were matched 1:1 to selective serotonin reuptake inhibitor users using propensity score matching, and Cox proportional hazard regression was performed to estimate the hazard ratios (HRs) for DIP risk. Correlation between each in vitro metric and real-world DIP risk was evaluated using logarithmic regression models.

Among 44,664 patients from 8 matched cohorts, haloperidol showed the highest DIP risk (HR=4.56, 95% CI 2.29‐9.07), whereas aripiprazole exhibited the lowest risk (HR=2.11, 95% CI 1.56‐2.86). Metric 4 (pKr × BBB penetration rate) exhibited the strongest correlation with real-world DIP risk (R2=0.95). The correlation decreased when aripiprazole, a partial D2R agonist, was included in the analysis (R2=0.58).

Integrating receptor-binding kinetics with BBB penetration may provide an in vitro framework that reflects real-world variation in DIP risk among D2R-antagonizing APDs. These findings support the relevance of combining kinetic and central nervous system exposure parameters for early safety evaluation.