Exploration of the potential association between GLP-1 receptor agonists and suicidal or self-injurious behaviors: a pharmacovigilance study based on the FDA Adverse Event Reporting System database

Abstract

Background: Establishing whether there is a potential relationship between glucagon-like peptide 1 receptor agonists (GLP-1RAs) and suicidal or self-injurious behaviors (SSIBs) is crucial for public safety. This study investigated the potential association between GLP-1RAs and SSIBs by exploring the FDA Adverse Event Reporting System (FAERS) database.

Methods: A disproportionality analysis was conducted using post-marketing data from the FAERS repository (2018 Q1 to 2022 Q4). SSIB cases associated with GLP-1RAs were identified and analyzed through disproportionality analysis using the information component. The parametric distribution with a goodness-of-fit test was employed to analyze the time-to-onset, and the Ω shrinkage was used to evaluate the potential effect of co-medication on the occurrence of SSIBs.

Results: In total, 204 cases of SSIBs associated with GLP-1RAs, including semaglutide, liraglutide, dulaglutide, exenatide, and albiglutide, were identified in the FAERS database. Time-of-onset analysis revealed no consistent mechanism for the latency of SSIBs in patients receiving GLP-1RAs. The disproportionality analysis did not indicate an association between GLP-1RAs and SSIBs. Co-medication analysis revealed 81 cases with antidepressants, antipsychotics, and benzodiazepines, which may be proxies of mental health comorbidities.

Conclusions: We found no signal of disproportionate reporting of an association between GLP-1RA use and SSIBs. Clinicians need to maintain heightened vigilance on patients premedicated with neuropsychotropic drugs. This contributes to the greater acceptance of GLP-1RAs in patients with type 2 diabetes mellitus or obesity.

Keywords: FAERS database; GLP-1RAs; Obesity; Pharmacovigilance; Suicidal or self-injurious behaviors; Type 2 diabetes.

Fig. 1

TTO analysis of SSIBs reported for each GLP-1 RA. Reported TTO analysis and duration of treatment of SSIB associated with a liraglutide, b semaglutide, c dulaglutide, and d exenatide. Black diamonds represent the TTO of SSIBs following the administration of GLP-1RAs, while gray bars indicate the duration of GLP-1 treatment for each case. Due to limitations in data availability, the duration of therapy could only be plotted for a subset of the cases. Goodness-of-fit test of SSIB associated with e liraglutide, f semaglutide, g dulaglutide, and h exenatide. Positioned above the plot is the quantile boxplot. The green bars correspond to the case number of occurrences within the distribution, while the green line represents the fitted curve for the model exhibiting the most optimal outcomes. we adopted log-normal distribution to describe the latency of liraglutide and exenatide, gamma distribution for semaglutide, and Weibull distribution for dulaglutide. The raw data of TTO analysis in the 52 valid cases can be found in Additional file 1: Table S7. TTO, time-to-onset; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist

Fig. 2

IC₀₂₅ of SSIBs associated with GLP-1 RA in the primary analysis and false negative analysis. When IC₀₂₅ > 0, a disproportionate reporting signal was detected. FAERS, US Food and Drug Administration Adverse Event Reporting System; IC, information component; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist

Fig. 3

IC₀₂₅ of SSIBs associated with GLP-1 RA in the sensitivity analysis. When IC₀₂₅ >0, a disproportionate reporting signal was detected. FAERS, US Food and Drug Administration Adverse Event Reporting System; IC, information component; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist