The Neural Underpinnings of Depression and Cannabis Use in Young PLWH

项目总结/摘要 响应RFA-DA-21-012，“阐明SUD和其他疾病复杂发病机制 艾滋病病毒/艾滋病患者的精神疾病”（PLWH），我们建议调查奖励和疼痛回路， 大麻使用和抑郁症合并症，这是PLWH中两种非常普遍的情况。我们将专注于年轻人 成人（18-34岁），以尽量减少艾滋病毒神经元慢性效应，并鉴于高比率的物质使用 并减少了该年龄组对艾滋病毒治疗的坚持。我们提出的模型是：1）奖励功能障碍 （奖励学习，期望，成就，积极预测错误的缺陷）和疼痛超敏反应（疼痛 敏感性，厌恶，负预测误差）有助于大麻使用和抑郁症的年轻人 PLWH。2)缰核（Hb）是一个小的边缘中枢，在这些过程中起着关键的调节作用， 腹侧被盖区（VTA）在疼痛和损失后向丘脑核（NAc）发出奖励信号。3)THC， 大麻的主要成分，通过与大麻素1受体结合发挥其精神镇痛作用 在奖赏和疼痛系统中，包括前扣带回（ACC），导水管周围灰质（PAG），丘脑， 杏仁核，VTA，NAc和Hb，暂时缓解情绪和疼痛症状，但导致长期 奖励回路的改变会加剧抑郁和物质使用。4)利用改进 在fMRI分辨率方面，我们的新成像方法克服了先前的技术限制，研究了Hb和其他 对奖赏和疼痛处理至关重要的小结构。支持数据从PLWH年龄18-34在我们的健康 系统显示，43%人患有抑郁症，21%的人患有大麻使用障碍，只有68%的人患有无法检测到的艾滋病毒 病毒载量（VL）。使用奖励侧翼（RFT）和奖励预测错误（RPET）fMRI任务，我们记录了 不同的大脑活动在奖励预期，实现和预测错误，预测未来 抑郁症的严重程度此外，我们在RPET和疼痛任务期间检测Hb激活，并绘制Hb内在 功能连接（iFC）与奖励关键区域（VTA）、疼痛关键区域（PAG）或两个回路（NAc、ACC）。 记录了与抑郁症、快感缺乏和大麻使用相关的不同Hb iFC。我们假设 年轻的PLWH中使用大麻和抑郁具有叠加效应，导致奖励缺陷和疼痛 超敏反应，这种模式将预测1年随访时的不良结局。我们将使用2×2阶乘 设计：1）70名抑郁的大麻使用者; 2）70名抑郁的大麻非使用者; 3）70名非抑郁的大麻使用者 使用者;以及4）70名非抑郁的大麻非使用者。参与者将在以下地点进行全面评估： 基线、6个月和12个月，包括抑郁、物质、奖励、疼痛、焦虑、创伤、HIV治疗、CD 4 + 计数和VL。将进行基线认知测试和fMRI（静息状态、RFT、RPET、疼痛）。分析 方法将包括机器学习分类。

PROJECT SUMMARY / ABSTRACT In response to RFA-DA-21-012, “Elucidation of mechanisms underlying complex morbidities of SUD and other mental Illnesses in people living with HIV/AIDS” (PLWH), we propose to investigate reward and pain circuitry in cannabis use and depression comorbidity, two highly prevalent conditions in PLWH. We will focus on young adults (ages 18-34) to minimize HIV neuronal chronicity effects and in light of the high rates of substance use and reduced adherence to HIV treatment in this age group. Our proposed model is: 1) Both reward dysfunction (deficits in reward learning, expectancy, attainment, positive prediction errors) and pain hypersensitivity (pain sensitivity, aversion, negative prediction errors) contribute to cannabis use and depression comorbidity in young PLWH. 2) The habenula (Hb), a small limbic hub, plays a pivotal regulatory role in these processes by inhibiting ventral tegmental area (VTA) reward signals to the nucleus accumbens (NAc) following pain and loss. 3) THC, a major component of cannabis, exerts its psychoactive analgesic effects by binding to cannabinoid 1 receptors in the reward and pain systems, including the anterior cingulate (ACC), periaqueductal gray (PAG), thalamus, amygdala, VTA, NAc, and Hb, creating temporary relief of mood and pain symptoms but resulting in long-term alterations in reward circuitry that exacerbate depression and substance use. 4) Capitalizing on improvements in fMRI resolution, our novel imaging methods overcome prior technical constraints to study the Hb and other small structures critical to reward and pain processing. Supporting data from PLWH ages 18-34 in our health system show that 43% have depression, 21% have cannabis use disorders, and only 68% had undetectable HIV viral load (VL). Using the reward flanker (RFT) and reward prediction error (RPET) fMRI tasks, we documented distinct brain activity during reward anticipation, attainment and prediction error, which predicted future depression severity. Further, we detected Hb activation during RPET and a pain task, and mapped Hb intrinsic functional connectivity (iFC) with regions critical to reward (VTA), pain (insula, PAG), or both circuits (NAc, ACC). Distinct Hb iFC were documented in relation to depression, anhedonia and cannabis use. We hypothesize that cannabis use and depression in young PLWH have an additive effect, inducing both reward deficits and pain hypersensitivity and that this pattern will predict worse outcomes at 1 year follow-up. We will utilize a 2×2 factorial design: 1) 70 depressed cannabis users; 2) 70 depressed cannabis non-users; 3) 70 non-depressed cannabis users; and 4) 70 non-depressed cannabis non-users. Participants will have comprehensive evaluations at baseline, 6- and 12-months including depression, substance, reward, pain, anxiety, trauma, HIV treatment, CD4+ count, and VL. Baseline cognitive testing and fMRI (resting-state, RFT, RPET, pain) will be performed. Analytical approaches will include machine learning classifications.