Evolutionary dynamics of non-genetic mechanisms of drug resistance in cancer

尽管传统上认为癌症是一种遗传性疾病，但人们越来越认识到，肿瘤细胞群异质性的非遗传来源在疾病进展和耐药性导致治疗失败中起着重要作用。例如，常见的非遗传异质性来源，如表观遗传现象和环境噪声，已被证明可诱导肿瘤细胞的短暂或可逆耐药状态。由于这些表型开关通常在比遗传进化快得多的时间尺度上运作，它们可以对肿瘤的进化过程和对治疗的反应产生很大的影响。这些观察结果对新药的临床潜力产生了巨大的兴趣，如表观遗传疗法，其目标是控制这些表型开关的细胞机制；然而，到目前为止，这些疗法的成功是有限的。这些努力的一个障碍是对个体细胞水平上的随机表型转变如何产生和影响肿瘤群体水平对治疗的反应的机制理解不足。该项目旨在通过数学理论的发展来阐明这些问题，这些理论将亚细胞水平的表型转变机制与群体水平的肿瘤进化动力学联系起来，在不断变化的肿瘤微环境背景下。这一框架将应用于了解白血病和结直肠癌治疗耐药的发展，并探索新的治疗策略，以防止治疗失败。由于在使用抗生素治疗的细菌群体中也会发生类似的现象，因此也可以利用这些模型来提供对抗菌素治疗耐药性的见解。在该项目的第一部分，PI将开发和分析由肿瘤微环境中基质细胞分泌因子响应的可逆表型转变驱动的耐药性的连续时间马尔可夫过程模型。在该项目的第二部分，PI将开发一种新的多尺度建模框架，使用分支随机游走将随机亚细胞表观遗传动力学与种群水平的进化动力学联系起来，并使用该模型探索表观遗传过程对癌症耐药性进化的影响。将对这些模型进行分析，以阐明驱动人群非遗传异质性的潜在过程如何影响总体进化转向耐药性和肿瘤复发。这项工作将涉及非标准几何上的状态依赖分支随机漫步的分析以及连续时间马尔可夫过程的分支过程逼近。特别是对这些模型的极限行为、随机命中时间（如复发时间、基因沉默/激活时间）和灭绝概率进行分析。PI将与实验合作者密切合作，应用这些模型和结果来了解由非遗传机制驱动的耐药性的特定癌症系统。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Although cancer is traditionally viewed as a genetic disease, it is becoming increasingly recognized that non-genetic sources of heterogeneity amongst tumor cell populations play an important role in disease progression and drug resistance leading to treatment failure. For example, common sources of non-genetic heterogeneity such as epigenetic phenomena and environmental noise have been shown to induce transient or reversible drug-resistant states in tumor cells. Since these phenotypic switches often operate at significantly faster time scales than genetic evolution, they can exert large influences on the course of tumor evolution and response to therapy. These observations have generated tremendous interest in the clinical potential of novel drugs, such as epigenetic therapies, that target the cellular machinery controlling these phenotypic switches; however, the success of these therapies has been limited so far. One obstacle to these efforts has been an inadequate mechanistic understanding of how stochastic phenotypic transitions at the level of individual cells arise and influence tumor population-level responses to therapy. This project aims to elucidate these questions through the development of mathematical theories that link the mechanisms of phenotypic transitions at the subcellular level to population-level tumor evolutionary dynamics, within the context of a changing tumor microenvironment. This framework will be applied to understand the development of therapeutic resistance in leukemias and colorectal cancers, and to explore novel treatment strategies for preventing treatment failure. Since analogous phenomena occur in the context of bacterial populations treated with antibiotics, such models can also be leveraged to provide insights into antimicrobial therapy resistance.In the first part of the project, the PI will develop and analyze a continuous-time Markov process model of drug resistance driven by reversible phenotypic transitions in response to stromal cell-secreted factors in the tumor microenvironment. In the second part of the project, the PI will develop a novel multiscale modeling framework using branching random walks to link stochastic subcellular epigenetic dynamics to population-level evolutionary dynamics, and use this model to explore the impact of epigenetic processes on the evolution of drug resistance in cancer. These models will be analyzed to elucidate how the underlying processes driving non-genetic heterogeneity in a population influence overall evolutionary shifts towards drug resistance and tumor recurrence. This work will involve the analysis of state-dependent branching random walks on nonstandard geometries as well as branching process approximations of continuous time Markov processes. In particular, analyses of the limiting behaviors of these models, stochastic hitting times (e.g. recurrence time, gene silencing/activation times), and extinction probabilities will be conducted. The PI will work closely with experimental collaborators to apply these models and results to understand specific cancer systems in which drug resistance is driven by non-genetic mechanisms.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.