Reporting signals of selected extrapyramidal adverse events with four newer second-generation antipsychotics: a...

Newer second-generation antipsychotics (SGAs) differ in pharmacology and approved indications, and extrapyramidal symptoms (EPS) remain clinically important. We investigated whether lurasidone, brexpiprazole, cariprazine, and lumateperone showed disproportionate reporting of selected EPS in the FDA Adverse Event Reporting System (FAERS).

We conducted a retrospective pharmacovigilance signal-detection study of spontaneous reports using OpenVigil 2.1-MedDRA-v24 (FAERS Q1 2004-Q3 2025). Analyses were restricted to reports in which the target drug was the primary suspect. Narrow MedDRA sub-SMQs for Akathisia, Dyskinesia, and Dystonia and their Preferred Terms (PTs) were evaluated with the reporting odds ratio (ROR), proportional reporting ratio (PRR), and Bayesian information component (IC). Sensitivity analyses used a common calendar window (Q1 2020-Q3 2025) and the first 24 post-approval quarters for each drug.

All four drugs had positive reporting signals for the three selected outcomes, and Akathisia had the largest within-drug IC025. Drug-specific PT patterns included akathisia/restlessness, tardive dyskinesia, and focal dystonia terms. The four separately calculated drug-versus-database estimates were not tested against one another and are not comparative risk estimates. All 24 estimates in the two time-window sensitivity analyses retained a lower ROR confidence bound above one and IC025 above 0, although signal magnitudes varied across windows.

The findings generate pharmacovigilance hypotheses for selected EPS only; they do not estimate incidence, comparative risk, or causality and should not guide preferential prescribing. Although positive signals persisted in both time-window sensitivity analyses, variation in signal magnitude, reporting behavior, differences in indication and concomitant treatment, the unexplained denominator discrepancy, and omission of Parkinson-like events require cautious interpretation and independent validation.