There is a substantial unmet need for effective and patient-acceptable drugs to treat severe mental illnesses such as schizophrenia (SZ). Computational analysis of genomic, transcriptomic, and pharmacologic data generated in the past 2 decades enables repurposing of drugs or compounds with acceptable safety profiles, namely those that are U.S. Food and Drug Administration approved or have reached late stages in clinical trials. We developed a rational approach to achieve this computationally for SZ by studying drugs that target the proteins in its protein interaction network (interactome). This involved contrasting the transcriptomic modulations observed in the disorder and the drug; our analyses resulted in 12 candidate drugs, 9 of which had additional supportive evidence whereby their target networks were enriched for pathways relevant to SZ etiology or for genes that had an association with diseases pathogenically similar to SZ. To translate these computational results to the clinic, these shortlisted drugs must be tested empirically through randomized controlled trials, in which their previous safety approvals obviate the need for time-consuming phase 1 and 2 studies. We selected 2 among the shortlisted candidates based on likely adherence and side-effect profiles. We are testing them through adjunctive randomized controlled trials for patients with SZ or schizoaffective disorder who experienced incomplete resolution of psychotic features with conventional treatment. The integrated computational analysis for identifying and ranking drugs for clinical trials can be iterated as additional data are obtained. Our approach could be expanded to enable disease subtype–specific drug discovery in the future and should also be exploited for other psychiatric disorders.

中文翻译：

---

精神分裂症相互作用组衍生的可再利用药物和两名候选人的随机对照试验


对于治疗精神分裂症（SZ）等严重精神疾病的有效且患者可接受的药物的需求存在大量未满足的需求。对过去 20 年生成的基因组、转录组和药理学数据进行计算分析，能够重新利用具有可接受安全性的药物或化合物，即那些已获得美国食品和药物管理局批准或已进入临床试验后期的药物或化合物。我们开发了一种合理的方法，通过研究靶向其蛋白质相互作用网络（相互作用组）中蛋白质的药物，在计算上为 SZ 实现这一目标。这涉及对比在疾病和药物中观察到的转录组调节；我们的分析产生了 12 种候选药物，其中 9 种有额外的支持性证据，表明它们的靶标网络丰富了与 SZ 病因相关的途径或与 SZ 致病相似疾病相关的基因。为了将这些计算结果转化为临床，这些入围药物必须通过随机对照试验进行实证测试，其中它们之前的安全性批准消除了耗时的 1 期和 2 期研究的需要。我们根据可能的依从性和副作用情况，从入围候选人中选择了 2 名。我们正在通过辅助随机对照试验对精神分裂症或分裂情感障碍患者进行测试，这些患者在常规治疗中精神病特征未完全缓解。当获得更多数据时，可以迭代用于识别和排序临床试验药物的综合计算分析。我们的方法可以扩展以在未来实现疾病亚型特异性药物的发现，并且也应该用于其他精神疾病。