Emergence of suicidal symptoms has been reported as a potential antidepressant adverse drug reaction. Identifying risk factors associated could increase our understanding of this phenomenon and stratify individuals at higher risk. Logistic regressions were used to identify risk factors of self-reported treatment-attributed suicidal ideation (TASI). We then employed classifiers to test the predictive ability of the variables identified. A TASI GWAS, as well as SNP-based heritability estimation, were performed. GWAS replication was sought from an independent study. Significant associations were found for age and comorbid conditions, including bipolar and personality disorders. Participants reporting TASI from one antidepressant were more likely to report TASI from other antidepressants. No genetic loci associated with TAS I (p < 5e-8) were identified. Of 32 independent variants with suggestive association (p < 1e-5), 27 lead SNPs were available in a replication dataset from the GENDEP study. Only one variant showed a consistent effect and nominal association in the independent replication sample. Classifiers were able to stratify non-TASI from TASI participants (AUC = 0.77) and those reporting treatment-attributed suicide attempts (AUC = 0.85). The pattern of TASI co-occurrence across participants suggest nonspecific factors underlying its etiology. These findings provide insights into the underpinnings of TASI and serve as a proof-of-concept of the use of classifiers for risk stratification.

中文翻译：

---

超过 10,000 名服用抗抑郁药的澳大利亚成年人因治疗导致自杀的临床、人口统计和遗传风险因素

据报道，出现自杀症状是一种潜在的抗抑郁药物不良反应。识别相关的风险因素可以增加我们对这种现象的理解，并对风险较高的个体进行分层。逻辑回归用于识别自我报告的治疗归因的自杀意念（TASI）的风险因素。然后，我们使用分类器来测试所识别变量的预测能力。进行了 TASI GWAS 以及基于 SNP 的遗传力估计。GWAS 复制是从一项独立研究中寻求的。发现年龄和合并症（包括双相情感障碍和人格障碍）存在显着关联。从一种抗抑郁药报告 TASI 的参与者更有可能从其他抗抑郁药报告 TASI。没有与 TAS I 相关的基因位点（p  < 5e-8) 已确定。在具有暗示关联的 32 个独立变体中 ( p  < 1e-5)，GENDEP 研究的复制数据集中有 27 个先导 SNP。只有一个变体在独立复制样本中显示出一致的效果和名义关联。分类器能够对来自 TASI 参与者 (AUC = 0.77) 和那些报告治疗归因的自杀企图 (AUC = 0.85) 的非 TASI 进行分层。参与者之间的 TASI 共现模式表明其病因的非特异性因素。这些发现提供了对 TASI 基础的见解，并作为使用分类器进行风险分层的概念验证。