Polygenic Risk, Trait Variables, and External Stressors in Fatal and Nonfatal Suicidal Behavior.

Abstract

Suicidal behavior is heritable and influenced by various clinical traits and external stressors. Clarifying these genetic and environmental influences may guide targeted intervention. To examine the interplay among polygenic scores for suicide attempt (suicide-PGSs), clinical characteristics, and external stressors in suicidal behavior. This case-control study included genome-wide genotyping of individuals with nonfatal suicide behavior, individuals who died by suicide, and alive and deceased control individuals with no suicidal behavior or death by suicide, respectively. Data were collected in New York, New York; Montreal, Canada; and Munich, Germany, between 1991 and 2011, and analyses were performed from July 1 to October 30, 2024. Suicide-PGSs, internal traits (hostility, impulsivity, aggression), clinical variables (depression severity, suicidal ideation, number and lethality of suicide attempts, number of depressive episodes), and external stressors (childhood abuse, recent life events). Associations between suicide-PGS and suicidal behavior, traits, clinical features, and environmental stressors were assessed via logistic regression, linear regression, and Poisson models. The sample included 1699 individuals across 2 cohorts. The live cohort included 1275 participants, of whom 239 had attempted suicide (mean [SD] age, 41.8 years; 147 female [61.5%]) and 1036 were control participants who had not (mean [SD] age, 37.3 [17.3] years; 574 female [55.4%]), and the postmortem cohort included 424 individuals, of whom 294 died by suicide (mean [SD] age, 45.1 [17.0] years; 219 male [74.5%]) and 130 were control individuals who died of other causes (mean [SD] age, 49.6 [18.3] years; 102 male [78.5%]). Suicide-PGS was associated with suicide attempts in the live (odds ratio [OR], 1.35; 95% CI, 1.17-1.56) and postmortem (OR, 1.34; 95% CI, 1.07-1.70) cohorts. Among live participants who attempted suicide, higher suicide-PGS was associated with lifetime aggression severity (b = 0.67; 95% CI, 0.41-0.94), depression severity (b = 0.20; 95% CI, 0.12-0.28), and less hostility (b = -0.51; 95% CI, -0.82 to -0.19) but not with impulsivity or lethality. Suicide-PGS was also associated with more depressive episodes (b [SE], 0.11 [0.04]) but not the number of lifetime suicide attempts in the live cohort. Suicide-PGS was associated with reported childhood abuse (OR, 1.16; 95% CI, 1.02-1.33) and recent life stress (b [SE], 0.17 [0.05]) in the live cohort, though these stressors did not moderate genetic associations with suicide. This case-control study found that genetic liability for suicide attempt was associated with clinical characteristics, internal traits, and external stressors, highlighting the complexity of genetic and environmental interactions. Larger studies with detailed phenotyping are needed to clarify genetic-environment contributions to suicide risk.