Identifying Predictors in the HPA Axis and Inflammatory Pathways for Suicidal Behavior in Youth

点击翻译按钮获取中文摘要

While suicidal behavior occurs in the context of many psychiatric disorders, relatively few subjects with a psychiatric disorder attempt suicide. One of the most challenging tasks for clinicians is to identify which patients will actually go on to attempt suicide. Studies find no association between clinicians' prediction for a patient and their actual suicidal behavior. Thus, it is critical to identify objective biological signatures for suicidal behavior, the 2nd leading cause of death among youth. In an R21 pilot study, we found that inpatients admitted for their first suicide attempt had lower hair cortisol concentrations (HCC) compared to those admitted for suicidal ideation and healthy controls. HCC provides a retrospective assessment of cortisol levels over the past few months and thus prior to attempt in our R21. Lower HCC were also associated with increased lethality of the attempt within attempters. Suicide attempters also differed by their lower glucocorticoid receptor (GR) mRNA and increased inflammation. In this R01, we propose to recruit a large sample of psychiatric inpatients (n=300), aged 18-30 years, with no prior history of suicidal behavior and enriched for suicidal ideation; and healthy controls (n=50); and follow them at 3, 6, and 12 months from intake. The risk for suicidal behavior is especially high during the first year after psychiatric hospitalization. We will collect biological data on HCC, gene expression in the HPA axis and inflammatory pathways, and systemic markers of inflammation (e.g., C- Reactive Protein, Tumor Necrosis Factor-α); and data on already-established clinical and behavioral predictors for suicidal behavior. We hypothesize that low HCC and downregulation of HPA axis genes together with upregulation of inflammatory genes and increased systemic inflammation at baseline will predict future suicidal behavior. Similarly, the trajectories of these biological alterations over time will be associated with worsening of clinical (impulsivity, aggression, sleep disturbances) and behavioral measures (decision-making, memory, and suicide-specific attentional biases and implicit cognitions). The models combining biological, clinical, and behavioral measures will show better performance in predicting attempts compared to models combining clinical and behavioral measures. We also propose to collect hair samples from subjects, aged 18-30 years, who died by suicide and compare them to those who died from accidental deaths on HCC, which reflects cortisol levels prior to death; we will also compare them on HCC to patients with no ideation, those with ideation, and those who go on to attempt suicide and thus examine the relation of HCC to risk across the full spectrum of suicidal behavior. This study is the first to examine the ability of biological markers in the HPA axis and inflammatory pathways to predict suicidal behavior and to examine them combined with clinical and behavioral predictors. This study will help better identify patients at highest risk who can then be targeted for closer monitoring and interventions; and will improve our understanding of the biological pathways for suicidal behavior, which will guide new therapeutic targets.