GLP-1 Receptor Agonists and Their Link to Suicide Behavior: A Review of Controlled Trials

Suicide represents a significant global public health concern. According to global statistics, more than 700 000 individuals die by suicide each year, with many more attempting it. Suicide can occur at any stage of life and has become the third leading cause of death among individuals aged 15–29 [1]. The factors influencing suicide are multifaceted, involving physiological conditions, social environment, and other determinants. Notably, obesity and diabetes are recognized as important risk factors for suicidal behavior.

It is well established that there is a strong correlation between obesity and mental health issues such as depression and anxiety [2]. Individuals with obesity often experience feelings of shame related to their body weight, which can lead to significant psychological distress. Since the onset of the obesity epidemic in the United States, the incidence of suicidal behavior has been found to be higher among obese adolescents compared to their non‐obese peers [3]. Similarly, patients with type 2 diabetes mellitus (T2DM) are also at an increased risk for suicidal tendencies. Many individuals with T2DM suffer from complications such as amputations and cardiovascular diseases, which contribute to a diminished quality of life. A retrospective cohort study found that patients with T2DM had a 1.8‐fold higher risk of developing depressive symptoms compared to individuals without T2DM over a 6.5‐year follow‐up period [4]. Given that depressive symptoms are a known risk factor for suicidal behavior [5], patients with T2DM may be at an elevated risk for suicide.

Glucagon‐like peptide‐1 receptor agonists (GLP‐1 RAs) are widely used therapeutics for managing T2DM and obesity. Theoretically, GLP‐1 RAs could reduce the risk of suicide by improving the management of these conditions. However, reports from the Icelandic Medicines Agency have raised concerns, indicating that some individuals using liraglutide and semaglutide have experienced suicidal thoughts and self‐harm. In response, the European Medicines Agency (EMA) has requested that certain license holders of GLP‐1 RAs provide further evidence regarding suicidal ideation and self‐harm [6].

A previous meta‐analysis examining the relationship between GLP‐1 RAs and mental health, based on randomized controlled trials (RCTs), concluded that no significant association was found between GLP‐1 RAs and mental health outcomes. However, this meta‐analysis had several limitations, including a small sample size of included RCTs and the lack of more detailed subgroup analyses [7]. Therefore, this meta‐analysis aims to comprehensively evaluate the published RCTs to better assess the relationship between GLP‐1 RAs exposure and suicidal behavior in patients with T2DM or obesity.

This meta‐analysis synthesized evidence from 25 RCTs to evaluate the potential association between GLP‐1 RAs and suicide‐related behaviors. Our findings demonstrate no significant correlation between the use of GLP‐1 RAs and suicidal behavior in patients with T2DM or obesity.

Obesity is an independent risk factor for suicidal behavior. Individuals with obesity often face societal stigma, discrimination, and social exclusion due to poor body image, which can lead to low self‐esteem and social dysfunction. These societal pressures may exacerbate psychological distress, subsequently elevating the risk of suicide. Furthermore, excessive circulating adipokines in obese individuals promote the increased expression of inflammatory mediators [33], potentially contributing to a chronic low‐grade inflammatory state. These inflammatory mediators may cross the blood–brain barrier (BBB), disrupting the metabolism and function of neurotransmitters such as serotonin and dopamine, which are essential for emotional and behavioral regulation. Additionally, chronic inflammation activates the hypothalamic–pituitary–adrenal (HPA) axis. Prolonged exposure to glucocorticoids can induce glucocorticoid resistance, stimulate immune cell activation, and enhance the production of inflammatory mediators, thus exacerbating neurotoxic effects. Concurrently, chronic glucocorticoid exposure has been shown to cause neuronal damage and loss in structures of the limbic system (e.g., hippocampus and amygdala), potentially increasing the susceptibility to suicidal behavior [34].

Patients with T2DM exhibit higher rates of depression and suicide compared to the general population [35]. This elevated suicide risk stems from a combination of physiological, psychological, social, and nutritional factors [36]. Similar to individuals with obesity, T2DM patients often experience chronic inflammatory states, which may contribute to depression and suicidal tendencies [37]. Animal studies have shown that T2DM mouse models exhibit hippocampal neuronal loss and atrophy [38]. Given that the hippocampus plays a critical role in emotional processing and memory formation, structural deficits in this region correlate with depressive symptoms, such as food avoidance and behavioral immobilization in murine models. These findings suggest that T2DM may predispose individuals to psychiatric disorders, including depression and suicidality, through neurological dysfunction [39]. Psychosocially, T2DM patients face multiple stressors, including diabetes‐related anxieties (e.g., fear of complications), the burdens of disease management (e.g., daily glucose monitoring, insulin injections, and dietary restrictions), as well as financial challenges and insufficient social support—each of which is a potential suicide risk factor [4].

A study has proposed a potential mechanism linking GLP‐1 RAs to suicide risk through their hypothalamic effects. During obesity treatment, GLP‐1 RAs bind to GLP‐1R in the hypothalamus, brainstem, and septal nuclei, particularly regulating feeding behavior in the brainstem and hypothalamic regions [40]. Activation of GLP‐1R may increase the expression of interleukin‐6 (IL‐6) in the brainstem and elevate IL‐6/IL‐1β levels in the hypothalamus [41]. Although these inflammatory mediators suppress appetite and promote weight loss, their overexpression could theoretically increase the risk of suicide. Furthermore, hypothalamic overactivation induced by GLP‐1 RAs may disrupt stress regulation through hyperactivation of the HPA axis, potentially leading to hippocampal dysfunction, cognitive deficits, and suicidal behavior, especially in patients with coexisting psychosocial risk factors [42].

Moreover, acute central administration of GLP‐1 RAs activates GLP‐1R in the raphe nucleus, directly inducing anxiety‐like behaviors. This pharmacological intervention also increases the expression of serotonin (5‐HT2A and 5‐HT2C) receptors and serotonin turnover in the amygdala [43]. Animal studies show that overexpression of the 5‐HT2C receptor in the limbic system increases anxiety‐like behaviors in mice [44]. These pathways may contribute to the induction of negative emotions and increase suicide risk.

Paradoxically, chronic GLP‐1 RA administration demonstrates antidepressant effects that may reduce suicidal behavior. Animal models indicate that exenatide and liraglutide exert antidepressant effects in T2DM mice and antipsychotic‐treated rats, respectively [45, 46]. Notably, the blockade of 5HT2A/5HT2C receptors does not affect the antidepressant efficacy of GLP‐1 RAs [47], suggesting that distinct emotional pathways are activated during acute versus chronic GLP‐1R activation. Therefore, patients may be more likely to benefit from the antidepressant effects of GLP‐1 RAs in long‐term anti‐obesity or anti‐T2DM treatments. Emerging evidence suggests that GLP‐1 RAs may mitigate suicide risk through anti‐inflammatory mechanisms, including direct and indirect immunomodulation via suppression of myeloid cell‐driven inflammation (monocytes/macrophages). Additionally, GLP‐1R activation reduces the induction of plasma tumor necrosis factor‐alpha (TNF‐α) by various Toll‐like receptor agonists, such as lipopolysaccharide (LPS) [48]. Furthermore, following GLP‐1R activation, anti‐inflammatory effects are mediated via GPCR signaling: Gαs‐mediated cAMP elevation activates the Epac/PKA pathways, while PI3K/Akt‐CREB signaling enhances Bcl‐2/Bcl‐xL expression and inhibits pro‐apoptotic proteins (BAD/p53) [49]. Collectively, these mechanisms reduce neuroinflammation, protect neural cells from oxidative stress, and restore neurotransmitter balance.

The aforementioned studies have demonstrated that GLP‐1R is expressed in brain regions involved in emotion regulation, such as the hypothalamus and limbic system. Upon activation of GLP‐1R, GLP‐1 RAs can influence affective states through various mechanisms. However, our meta‐analysis suggests that these pathways do not translate into clinically significant behavioral risks in humans.

Although obese participants may theoretically experience greater psychosocial stress than those with T2DM, our subgroup analyses revealed no significant differences in the incidence of suicide‐related behaviors between GLP‐1 RA exposure groups and control groups in either population. Adolescents, in particular, are more susceptible to social exclusion, discrimination, and stigmatization compared to adults. Moreover, trauma and abuse during sensitive developmental periods can impact cortical and subcortical development, directly influencing suicidal thoughts, behaviors, and self‐harm tendencies in adolescents. Previous studies have shown that suicidal ideation is prevalent among adolescents, with a prevalence rate ranging from 15% to 25% [50]. Despite these factors, our study conducted a subgroup analysis based on age and found no significant differences between the exposed group and the control group, regardless of whether participants were adolescents or adults.

A study has suggested that liraglutide exhibits relatively poor permeability across the BBB, whereas lixisenatide has been shown to cross the BBB and exert neuroprotective effects [49]. Consequently, it is conceivable that different types of GLP‐1 RAs may have varying impacts on the central nervous system. However, our subgroup analyses revealed no significant differences between the exposure and control groups for any of the GLP‐1 RA types. Additionally, no intergroup variations were observed for specific subtypes of suicidal behavior. No significant differences were observed between the exposure group and control group when comparing placebo or others, potentially due to interstudy participant heterogeneity and uncertain definitions of suicide behavior events. The underlying mechanisms require further investigation.

Our meta‐analysis is strengthened by large sample sizes, high‐quality RCTs, and low heterogeneity. Furthermore, the comprehensive subgroup analyses systematically demonstrate that there is no significant association between GLP‐1 RAs and suicidal behavior in populations with T2DM or obesity. However, certain limitations should be considered when interpreting these results. Despite the substantial sample sizes, statistical power remains limited for rare outcomes, such as suicidal behaviors. Low‐incidence events are inherently susceptible to multifactorial confounding, which may impact the precision of our findings.

In conclusion, no significant correlation was observed between GLP‐1 RAs exposure and suicidal behaviors, regardless of whether the participants had T2DM or obesity. Subgroup analyses, stratified by indication, age, type of GLP‐1 RAs, and specific subtypes of suicidal behaviors, further confirmed the absence of such a correlation.

Jingqi Chen: conceptualization, methodology, writing – original draft. Qiufeng Zhang: conceptualization, writing – original draft. Qingping Wu: data curation, methodology. Xiaoming Zhang: data curation. Zhiyi Xiang: software, validation. Sidong Zhu: software; visualization. Tianfu Dai: writing – review and editing. Yuexiu Si: writing – review and editing.

The authors have nothing to report.

The authors have nothing to report.

The authors declare no conflicts of interest.

The authors have nothing to report.

Chen J., Zhang Q., Wu Q., et al., “Impact of GLP‐1 Receptor Agonists on Suicide Behavior: A Meta‐Analysis Based on Randomized Controlled Trials,” Journal of Diabetes 17, no. 9 (2025): e70151, 10.1111/1753-0407.70151.

The datasets supporting the conclusions of this article are included within the article itself. For further details, please contact the corresponding author.