Leveraging genetic and electronic health record data to identify novel targets and drugs for treating alcohol use disorder

项目总结/摘要 虽然四种治疗酒精使用障碍（AUD）的药物已获得食品和药物管理局的批准， (FDA)许多患者并没有从中受益。此外，虽然全基因组关联研究（GWAS） 酒精消费和有问题的酒精使用（PAU;即，一种结合了AUD诊断和 有害饮酒的测量）已经产生了许多重要的单核苷酸多态性（SNP）， 风险，这些已经产生了治疗AUD的药物靶点。因此，确定药物靶点的需求尚未得到满足 用于开发治疗AUD的新型和/或再用途药物。最近的研究表明， 遗传支持的疾病机制可以提高药物开发的成功率， (i.e.,围绕疾病相关基因的生物网络）被富集用于批准药物的靶点。 因此，影响酒精消费和PAU风险的基因及其相关模块可能会产生新的目标 和用于治疗的药物。此外，大型电子健康记录（EHR）的可用性 数据集使我们有可能探索暴露于FDA批准的药物是否可以改善 除批准的医疗条件外的其他医疗条件，如AUD，并可能被重新利用。 这项提案将建立在研究小组先前的工作基础上，并利用基因组学的进步和获得 退伍军人事务部（VA）EHR通过VA信息学和计算基础设施（芬奇）：1）阐明 与酒精消费和PAU相关的模块（目标1和3）;以及2）确定有前途的药物用于再利用 治疗AUD（目标2）。目的1-2的一般假设是：1）与所鉴定的模块有关的基因 将成为众多获批药物的靶向药物; 2）在VA EHR中具有足够患者数据的药物中， 在倾向评分分析中，将有证据表明他们减少了酒精消费。假设为 目标3是分析将识别针对生物过程富集的排名靠前的模块， 与酒精消费和PAU相关。总之，这项建议结合了精神病遗传学和 药物流行病学方法，以确定新的目标和评估有前途的药物，以重新用于 治疗AUD。一种非理论的，遗传数据驱动的方法来选择有前途的FDA批准的药物， 然后在EHR中使用倾向评分方法对其进行测试，这在精神病学中以前没有做过， 包括澳元。这个项目是由于最近在酒精消费和PAU的GWAS方面的进展， 药物靶点连接，以及用于遗传和其他分析的EHR的培养。这种药物优先排序的方法 可以发现在后续临床试验中测试的独特药物和在临床前评估的新靶点， 问题研究

PROJECT SUMMARY/ABSTRACT While four medications for treating alcohol use disorder (AUD) are approved by the Food and Drug Administration (FDA), many patients do not benefit from them. Moreover, while genome-wide association studies (GWASs) of alcohol consumption and problematic alcohol use (PAU; i.e., a phenotype that combines AUD diagnoses and a measure of harmful drinking) have yielded many significant single nucleotide polymorphisms (SNPs) that affect risk, these have yielded few drug targets for treating AUD. Hence, there is an unmet need to identify drug targets for the development of novel and/or repurposed drugs to treat AUD. Recent research indicates that targeting disease mechanisms with genetic support can increase the success rate in drug development and that modules (i.e., biological networks surrounding disease-associated genes) are enriched for targets of approved drugs. Thus, genes affecting alcohol consumption and risk of PAU and their associated modules could yield new targets and drugs for therapeutic repurposing. Furthermore, the availability of large electronic health records (EHR) datasets makes it possible to explore whether exposure to FDA-approved drugs can lead to improvements in medical conditions other than the ones for which they are approved, such as AUD, and potentially be repurposed. This proposal will build upon prior work by the study team and leverage advances in genomics and access to the Veterans Affairs (VA) EHR through the VA Informatics and Computing Infrastructure (VINCI) to: 1) elucidate modules linked to alcohol consumption and PAU (Aims 1 and 3); and 2) identify promising drugs for repurposing to treat AUD (Aim 2). The general hypotheses for Aims 1-2 are: 1) the genes implicated in the identified modules will be targeted by numerous approved drugs; and 2) of the drugs with sufficient patient data in the VA EHR, there will be evidence that they reduce alcohol consumption in propensity score analyses. The hypothesis for Aim 3 is that the analysis will identify top ranked modules that are enriched for biological processes with relevance to alcohol consumption and PAU. In sum, this proposal combines psychiatric genetic and pharmacoepidemiologic methods to identify novel targets and evaluate promising drugs to be repurposed for treating AUD. An atheoretical, genetic data-driven approach to selecting promising FDA-approved drugs and then testing them in the EHR using propensity score methods has not previously been done in psychiatry, including for AUD. This project is made possible by recent advances in GWAS of alcohol consumption and PAU, drug target linking, and the cultivation of EHRs for genetic and other analyses. This approach to drug prioritization could uncover unique drugs to be tested in follow-up clinical trials and novel targets to be evaluated in preclinical studies.