Exploiting Ecology and Evolution to Prevent Therapy Resistance in EGFR-Driven Lung Cancer

摘要 肺癌是美国癌症死亡的主要原因，2016年估计有15.8万人死亡。这个 这些死亡中的大多数直接原因是最终出现了对最初有效的治疗方法的抗药性。 这种耐药性的演变代表了肿瘤学中最大的未得到满足的需求之一。虽然大多数研究 集中在赋予这种抗性的单个分子改变上，所以我们建议把重点放在 产生这些变化的生态进化过程。研究达尔文进化论和生态学 发生在异质肿瘤内的相互作用，我们将紧密集成定制的数学模型和 设计用来告知它们的实验技术。关注EGFR突变的非小细胞肺癌 癌症类型，我们在实验室有经验的高效靶向治疗，我们将接近 这个问题有三个正交的、综合的数学实验目标。首先，要了解 在异质肿瘤中细胞间发生的生态相互作用，我们将结合我们的经验 用进化博弈论和我们一流的进化博弈分析，我们设计了 特别是为了这个目的。在这里，我们希望学习如何瞄准驱动耐药性的相互作用--一种新的 这一战略可能会为药物设计开辟全新的途径。第二，我们将允许进化论向我们展示 它在面对特定选择压力时通过长期定向创造的趋同表型 进化论。在这一长期的进化过程中，我们将以小组药物敏感性的形式来测量表型 定期进行化疗和靶向治疗，创造了第一个暂时性侧支敏感性 在任何实体肿瘤中进行映射。通过汇集在整个进化生命历史中观察到的共同表型，我们 然后将使用交互和基于种子的协议来生成这些状态的分子签名 灵敏度，我们将在公开可用的数据中和在PDX线路库中进行验证。最后，我们将深入研究 深入到我们在前两个目标中研究的生态和进化过程的相关时间尺度。 我们计划将进化博弈论的复制者-变异者框架应用于我们 是癌症领域的先驱。为了验证和参数化这些模型，我们还将测试 我们用博弈分析通过进化时间执行分析来测量生态动态 在一项长期的进化实验中。我们的三个正交目标中的每一个都得到了最近的高影响力的支持 出版物，每一种都代表紧密耦合的实验和计算协议。我们干净的，好的- 精心设计的整合，加上对进化生物学直接研究的创新关注，有望 阐明了癌症研究的这一困难领域，并提供了提供概括性见解的可能性。

ABSTRACT Lung cancer is the leading cause of cancer deaths in the USA, with an estimated 158,000 deaths in 2016. The direct cause of the majority of these deaths is the eventual emergence of resistance to initially effective therapies. This evolution of drug resistance represents one of the greatest unmet needs in oncology. While most research is focused on the individual molecular alterations that confer this resistance, we instead propose to focus on the eco-evolutionary processes that generate these alterations. To study the Darwinian evolution and ecological interactions occurring within heterogeneous tumors, we will tightly integrate bespoke mathematical models and experimental techniques designed to inform them. Focusing on EGFR mutated non-small cell lung cancer, a cancer type with a highly efficacious targeted therapy with which we have experience in our lab, we will approach this problem with three, orthogonal, integrated mathematical-experimental Aims. First, to understand the ecological interactions occurring at the inter-cellular level in heterogeneous tumors, we will couple our experience with evolutionary game theory with our first-in-class evolutionary game assay, which we have designed to specifically for this purpose. Here, we hope to learn to target the interactions that drive resistance – a novel strategy which could open up entirely new avenues of drug design. Second, we will allow evolution to show us the convergent phenotypes it creates in the face of specific selective pressures through long-term directed evolution. During this long-term evolution, we will measure phenotype, in the form of drug sensitivity to a panel of chemotherapeutics and targeted therapies at regular intervals, creating the first temporal collateral sensitivity map in any solid tumor. By pooling common phenotypes observed throughout the evolutionary life history, we will then use interactomic and seed-based protocols to generate molecular signatures of these states of sensitivity, which we will validate in publicly available data and in an in library of PDX lines. Finally, we will delve deeply into the relevant time scales of the ecological and evolutionary processes we study in the first two aims. We plan to apply the replicator-mutator framework of evolutionary game theory to a spatial transform that we pioneered in cancer. To validate and parameterize these models, we will also test the evolutionary stability of the ecological dynamics we measure with our game assay by performing the assay through evolutionary time during a long-term evolution experimental. Each of our three orthogonal aims is supported by recent high impact publications, and each represent tightly coupled experimental and computational protocols. Our clean, well- designed integration, together with innovative focus on the direct study of the evolutionary biology, promises to shed light on this difficult area of cancer research, and offers the possibility of providing generalizable insights.