Systematic Identification and Pharmacological Targeting of Tumor Dependencies for Precision Cancer Medicine

项目摘要 癌症治疗的成功靶点分为三大类：引起肿瘤特异性 由于它们在肿瘤发生中的直接作用（癌基因依赖性）， 尽管在肿瘤发生中缺乏直接作用（非致癌基因依赖性），但特定突变具有致死性，以及 与肿瘤细胞与免疫系统相互作用相关的蛋白质（免疫检查点依赖性）。 然而，鉴于我们目前对癌症作为一个复杂和高度异质性系统的理解， 很难想象一个单独的蛋白质可能代表所有数十亿细胞的有效靶点， 组成一个典型的质量。事实上，虽然基于基因的靶向治疗和免疫肿瘤学有很大的希望， 大多数患者仍然没有反应或最终会复发耐药肿瘤，这表明 治疗靶点作为单一蛋白质的概念可能需要重新审视。 为了实现这一目标，我们将利用我们中心研究人员开发的非常成功的框架 用于识别和药理学靶向肿瘤依赖性， 在严格调节的肿瘤检查点模块内，少数主调节器（MR）蛋白的活性。 具体来说，我们将阐明和实验验证MR蛋白和相关的肿瘤检查点 罕见和不可治愈的恶性肿瘤的模块，在个别患者的基础上，通过执行基于网络的 使用从原发性肿瘤反向工程化的调节模型分析肿瘤样品特征 样品然后，我们将优先考虑一组FDA批准的药物和II期或III期后期研究药物， 肿瘤学（肿瘤药物）的III期研究，基于其靶向必需/合成致死 MR（OncoTarget）或逆转肿瘤的完整MR特征（OncoTreat）。 将使用可用的肿瘤药物干扰适当匹配的肿瘤模型的RNASeq谱， 获得并分析以评估由单个药物和药物组合物诱导的差异肿瘤检查点活性。 组合，随后进行低通量研究，以阐明其活性的调控基础。模型- 包括细胞系、来自肿瘤外植体的短期器官型培养物（EXPL）、类器官（ORG）和患者 衍生的异种移植物（PDX）-将基于MR蛋白保守性来选择。我们将验证这些发现 以及优先药物和组合的总体疗效、药效学特性、生物标志物 准确性和灵敏度，以及在合适的体外和体内模型中的抗性机制。 如果成功的话，这将代表精确癌症医学的第一种机械方法， 系统地推导出治疗靶点、相关抑制剂和人群分层生物标志物 从肿瘤状态调节及其药物诱导的调节的精确的、机械的理解。

PROJECT SUMMARY Successful targets for cancer therapy fall in three main categories: oncogenes that elicit tumor-specific essentiality because of their direct role in tumorigenesis (oncogene dependency), proteins that elicit synthetic lethality with specific mutations despite lack of a direct role in tumorigenesis (non-oncogene dependency), and proteins related to the interaction of tumor cells with the immune system (immune-checkpoint dependency). However, given our current understanding of cancer as a complex and highly heterogeneous system, it is difficult to imagine that an individual protein may represent an effective target for all the billions of cells that make up a typical mass. Indeed, while genetic-based targeted therapy and immunoncology hold great promise a majority of patients still does not respond or will eventually relapse with drug resistant tumors, suggesting that the concept of therapeutic targets as single proteins may need to be revisited. To accomplish this goal, we will leverage a highly successful framework developed by our center investigators for the identification and pharmacological targeting of tumor dependencies implemented by the concerted activity of a handful of Master Regulator (MR) proteins within tightly regulated tumor checkpoint modules. Specifically, we will elucidate and experimentally validate MR proteins and associated Tumor Checkpoint modules of rare and incurable malignancies, on an individual patient basis, by performing network-based analysis of tumor samples signatures using regulatory models reverse engineered from primary tumor samples. We will then prioritize a set of FDA approved drugs and late stage investigational drugs in phase II or phase III studies in oncology (oncology drugs) based on their ability to either target essential/synthetic-lethal MRs (OncoTarget) or to reverse the full MR signature of a tumor (OncoTreat). RNASeq profiles of appropriately matched tumor models perturbed with available oncology drugs will be obtained and analyzed to assess the differential tumor checkpoint activity induced by individual drugs and drug combinations, followed by low-throughput studies to elucidate the regulatory basis of their activity. Models – including cell lines, short-term organotypic culture from tumor explants (EXPL), organoids (ORG), and patient derived xenografts (PDX) – will be selected based on MR protein conservation. We will validate these findings as well as overall efficacy of prioritized drugs and combinations, pharmacodynamic properties, biomarker accuracy and sensitivity, and mechanisms of resistance in suitable in vitro and in vivo models. If successful, this would represent the first mechanistic approach for precision cancer medicine, where therapeutic targets, associated inhibitors, and population stratification biomarkers are systematically derived from precise, mechanistic understanding of tumor state regulation and of its drug-induced modulation.