Evaluation of molecular mechanisms of treatment response in late-life depression

描述：在过去的几十年里，抗抑郁药和心理治疗一直是治疗抑郁症的一线药物。 治疗LLD。尽管安全且耐受性良好，但大量患者不能实现完全和完全的治疗。 初始治疗后持续缓解。大约50%的LLD患者在两次治疗后没有反应。 抗抑郁药试验，符合治疗抵抗的共识定义（TR-LLD）。持续存在 老年人的慢性和升高的抑郁症状具有重要的临床和公共卫生 影响这与总体健康状况不佳、生活质量下降以及更高的风险有关。 死亡率与治疗后持续缓解的患者相比。尽管与公共卫生相关， 关于TR-LLD的生物学机制的信息很少，也没有可靠的临床预测模型来 在抗抑郁治疗开始时评估谁会或不会对治疗产生反应。 利用NIMH资助的临床试验，晚年抑郁症的不完全反应： 缓解”（IRL-GREY），在本研究中，我们建议评估生物学机制 与晚年抑郁症的治疗反应有关，并开发一种基于机器学习的算法， 预测这些受试者的治疗反应。我们将进行一个全面的，多重的蛋白质组学研究， 分析了来自完成临床试验1期和2期患者的542份样本。我们 假设衰老相关的生物学途径（即炎症反应控制，蛋白质稳定控制， 细胞损伤反应、内皮功能）将与LLD中较差的治疗反应相关。 此外，我们假设机器学习衍生的生物标志物组将具有灵敏度和特异性。 大于80%预测LLD的治疗反应。最后，我们将评估生物学机制 与治疗后不同的抑郁症状轨迹有关。 这项工作将为生物驱动的治疗反应模型奠定基础，该模型将有助于指导， 在抗抑郁治疗开始时，那些将从特定治疗中获益更多的人。如果成功，我们的工作 可以加速针对抑郁症的治疗努力和创新，并减少大量患者的痛苦。 老人及其家人。

DESCRIPTION: Over the past decades, antidepressants and psychotherapy have been the first-line treatments for LLD. Despite being safe and well-tolerated, a large number of patients do not achieve full and persistent remission after initial treatment. About 50% of patients with LLD do not respond after two antidepressant trials, meeting the consensus definition of treatment resistance (TR-LLD). The persistence of chronic and elevated depressive symptoms in older adults has significant clinical and public health implications. This has been correlated to poor general health, reduced quality of life, and a higher risk of mortality when compared to those with sustained remission after treatment. Despite the relevance to public health of TR-LLD, there is little information about the biological mechanisms and no robust clinical prediction model to evaluate at the outset of antidepressant therapy who will or will not respond to treatment. Leveraging an NIMH funded clinical trial, the Incomplete Response in Late-Life Depression: Getting to Remission” (IRL-GREY), across 3 sites, in this study, we propose to evaluate the biological mechanisms related to treatment response in late-life depression and to develop a machine learning based algorithm for prediction of treatment response in these subjects. We will carry out a comprehensive, multiplexed proteomic analysis on 542 samples from patients who completed phase 1 and phase 2 of the clinical trial. We hypothesise that ageing-related biological pathways (i.e. inflammatory response control, proteostasis control, cell damage response, endothelial function) will be associated with poorer treatment response in LLD. Moreover, we hypothesize that a machine learning derived biomarker panel will have sensitivity and specificity greater than 80% to predict treatment response in LLD. Finally, we will evaluate the biological mechanisms related to different depressive symptoms trajectories after treatment. This work will set the stage for a biologically-driven model of treatment response that will be useful to guide, at the outset of antidepressant treatment, those who will benefit more from a specific treatment. If successful, our work can accelerate therapeutic efforts and innovation targeting depression and reduce suffering for large numbers of elderly and their families.