SUICIDE RISK OF THIRD-GENERATION ANTIPSYCHOTICS IN PERSONS WITH SCHIZOPHRENIA AND SCHIZOAFFECTIVE DISORDERS: A SYSTEMATIC REVIEW AND META-ANALYSIS

Background

Clozapine is the only antipsychotic with protective effects against suicide in schizophrenia and schizoaffective disorders (SCZ). However, the high side effect burden and monitoring protocol pose additional barriers to adherence. Third-generation antipsychotics (TGA) are characterized by unique partial agonism of dopamine D2 and D3 receptors and primarily include aripiprazole, oxaripiprazole, brexpiprazole, cariprazine, and lumateperone. TGAs have high tolerability and low rates of discontinuation. Additionally, several TGAs modulate serotonin, dopamine, and N-methyl-d-aspartate neurotransmission that have been implicated in suicide. Given the clear need for a more benign antipsychotic with protective effects against suicide, we aim to investigate the effects of TGAs of suicide in SCZ.

Methods

We searched PubMed, MEDLINE, PsycINFO, Scopus, Cochrane Central Register of Controlled Trials, Europe PMC, and ClinicalTrials.gov databases from inception to December 2023 for TGA studies in SCZ that reported suicide data. Aripiprazole studies were excluded to focus on newer antipsychotics. The primary outcome was suicide deaths and attempts; suicidal ideation was included as a secondary outcome. The revised Cochrane risk of bias tool (RoB 2) was used to assess study quality. Stata SE18 was used to conduct random effects and single proportion meta-analyses for randomized controlled trials (RCT) and single-arm open label extension studies (OLE), respectively. Sensitivity analyses were conducted for RCTs. Subgroup analyses explored the impact of dose, drug type, and comparator arm.

Results

Of 4418 identified articles, 15 RCTs (N = 5700) met inclusion criteria – five brexpiprazole (N = 2260), seven cariprazine (N = 2703), two lumateperone (N = 701), and one oxaripiprazole (N = 36). Five OLEs (N = 2200) were included – three brexpiprazole (N = 1521) and two cariprazine (N = 679). Sample sizes ranged from 36 to 1031, and trial durations ranged from 4 to 72 weeks. The overall incidence of suicide events was low, and 13 studies excluded participants with higher suicide risk. Two RCTs and all OLEs had moderate bias risk; the remaining studies had low bias risk. Compared to placebo control, TGAs neither increased nor decreased the risk of suicide deaths and attempts [RR = 0.65, 95% CI = 0.24–1.72, p = 0.38] or suicide deaths, attempts, and ideation [RR = 0.63, 95% CI = 0.34–1.19, p = 0.15]. Subgroup and sensitivity analyses did not show significant impacts of TGAs on suicide outcomes. Single proportion meta-analysis of OLEs showed a significant increase in the incidence of suicide attempts and deaths [pooled incidence = 0.004, 95% CI = 0.000–0.013, p = 0.048] and suicide deaths, attempts, and ideation [pooled incidence = 0.024, 95% CI = 0.005–0.054, p = 0.0013], but there was marked heterogeneity across OLEs for primary [I2 = 71.68%; Q(4) = 12.75, p = 0.013] and secondary [I2 = 91.32%; Q(4) = 55.50, p < 0.0001] suicide outcomes.

Discussion

Clozapine remains unique in its protective effect against suicide. There is no current evidence to show that TGAs positively or negatively affect suicide outcomes. The increased incidence of primary and secondary suicide outcomes in OLEs should be interpreted with caution due to high heterogeneity and requires replication in other data. There remains a need for an alternative to clozapine that is equally effective in treating SCZ and preventing suicide while avoiding unwanted side effects. Additionally, alternative ways to include higher risk populations in suicide prevention research should be considered. Future studies into rare suicide outcomes should consider using alternative high-quality study designs, larger sample sizes, and longerterm follow up.