Secondary Analysis and Integration of Existing Data Related to Chronic Orofacial Pain and Placebo Effects - Administrative Supplement

本项目响应RFA-DE-22-011，旨在分析代表参与者子集的现有数据 患有颞下颌关节紊乱病(TMD)，他们接受了深入的临床、行为和心理研究 根据R01 DE025946(PI：Colloca，截至2022年9月30日)进行表型鉴定。建议的目标是 与最初的R01有很大不同-探索与预期诱导相关的遗传变异的工作 慢性口腔面部疼痛的止痛作用、心理因素对安慰剂反应的预测作用及相关基因 与预期镇痛相关的前额叶和边缘区域的神经元变化。vbl.使用 院长的倡议基金分配给科洛卡博士，我们采集了血液并从一个子集提取了RNA-SEQ数据 在74名TMD参与者中。有了这个子集的行为、心理、临床和现在的转录数据 在TMD参与者中，中心假设是不同的差异表达基因(DEG)和途径 与内源性疼痛调制(EPM)相关的特征是表现出最高TMD的参与者 安慰剂效应。我们将通过安慰剂比较与高EPM和低EPM相关的转录图谱 在TMD参与者中测试的效果(AIM1)，我们将预测高EPM整合转录， 使用机器学习的社会人口学、临床和心理数据(探索性特定目标2) 模特们。为了确定高安慰剂反应性的转录图谱，TMD参与者将被 根据平均报告分为高安慰剂应答者(HLR)和低安慰剂应答者(LPR) 在视觉模拟评分(VAS)上，疼痛评分从0=无痛到100=最高，以30为界值 想象得到的痛苦。基于我们之前发布的结果、信息性的初步结果和DEG能力 计算，我们希望有足够的能力来识别HPR中的关键DEG关联，而不是那些TMD 无反应和/或安慰剂反应较低，同时控制性别、年龄和疼痛严重程度。重要的是 将进行无偏见的浓缩分析，以确定与EPM相关的转录过程。 机器学习方法(例如，广义增强模型)将允许我们整合社会人口统计学， 临床和心理的转录标记物，以进一步表征TMD参与者的HPR。我们的团队 具有强大的互补性专业知识，确保这项研究将提供综合模型，以实现步骤- 循序渐进地发现那些表现出EPM最大活性的分子机制 安慰剂效应。这是第一个使用转录特征分析和机器学习模型来预测 未被充分研究的TMD人群中的EPM。研究结果将具有很高的临床相关性，并将提供更多信息 广泛的研究产生知识，这些知识将对指导未来迈向一体化和 翻译精准医学。

This project in response to RFA-DE-22-011 aims to analyze existing data representing a subset of participants with Temporomandibular disorders (TMD) who underwent in-depth clinical, behavioral, and psychological phenotyping under R01 DE025946 (PI: Colloca, ending September 30, 2022). The proposed aims are substantially different from the original R01-work exploring genetic variants associated with expectancy-induced analgesia in chronic orofacial pain, psychological factors predicting placebo responders, and genes-related neuronal changes in the prefrontal and limbic areas associated with expectancy-induced analgesia. Using Dean’s Initiative Funds allocated to Dr. Colloca, we collected blood and extracted RNA-seq data from a subset of 74 TMD participants. With the behavioral, psychological, clinical and now, transcriptomic data from this subset of TMD participants, the central hypothesis is distinct Differently Expressed Genes (DEG) and pathways associated with Endogenous Pain Modulation (EPM) characterize those TMD participants who show the highest placebo effects. We will compare transcriptomic profiles associated with high versus low EPM via placebo effects tested in TMD participants (AIM1) and we will predict high EPM integrating transcriptomic, sociodemographic, clinical and psychological data (Exploratory Specific Aim 2) using machine learning models. In order to identify transcriptomic profiles of high placebo responsiveness, TMD participants will be divided into High Placebo Responders (HLR) and Low Placebo Responders (LPR) based on an average reported pain score cut-off of 30 on a visual analogue scale (VAS) anchored from zero=no pain to 100=maximum imaginable pain. Based on our prior published results, informative preliminary results, and DEG power calculation, we expect enough power to identify key DEG associations in HPR compared to those TMD who do not respond and/or have lower placebo responses while controlling for sex, age and pain severity. Importantly, unbiased enrichment analyses will be conducted to identify transcriptomic processes associated with EPM. Machine learning approaches (e.g., generalized boosted models) will allow us to integrate sociodemographic, clinical and psychological with transcriptomic markers to further characterize HPR in TMD participants. Our team is strong with complementary expertise, ensuring that this research will provide integrative models towards step- by-step discoveries of molecular mechanisms characterizing those who show the largest activation of EPM via placebo effects. This is the first project to use transcriptomic profiling and machine learning models to predict EPM in an understudied TMD population. Findings will have high clinical relevance and will inform more extensive studies generating knowledge that will be critical to guide future steps towards integrative and translational precision medicine.