细胞微环境介导肿瘤耐药性的多尺度系统建模、分析与控制

In the past decades, target therapy has become one of the standard therapies for the malignant tumors. But many types of tumors develop drug resistance during chemotherapies, resulting in tumor recurrence. The cellular and molecular mechanism of cancer drug resistance has been a controversial issue. The recent experimental evidence and our previous studies indicate that cell microenvironment might represent a potential mechanism underlying cancer drug resistance. This project aims to develop mathematical models for multiscale system of cancer drug resistance, from a systems biology perspective, to investigate the dynamic mechanisms and control strategies of microenvironment-mediated cancer drug resistance. Taking brain glioma as a realistic case, we will build cell population dynamics model to quantitatively study the influence of cell microenvironment on cancer drug resistance. We will next construct signaling pathway networks of cell-cell interaction between tumor cell and microenvironment based on single cell transcriptomic data, and further develop single cell-based multiscale spatial-temporal model to investigate the evolution mechanism of how microenvironment heterogeneity and stochasticity within molecular network mediate the emergence and development of cancer drug resistance. Furthermore, we will set up a multi-objective optimization model to study the dynamic control strategies of drug combinations. This project can not only help to advance our understanding of dynamic intra- and inter-cellular mechanisms of cancer drug resistance from the view of systems medicine, but can also provide guidance and theoretic basis for optimizing drug treatment regimens.

在过去几十年里，靶向治疗已成为治疗恶性肿瘤的标准疗法之一，但是多种肿瘤在化疗过程中因产生耐药性而导致肿瘤复发。肿瘤耐药性的细胞和分子机制一直是个争议的话题，但近期的实验证据和申请者的前期研究表明细胞微环境可能是介导肿瘤耐药性的一种潜在机制。本课题拟从系统生物学的观点建立多尺度肿瘤耐药性系统的数学模型，着重研究微环境介导肿瘤耐药性的动态机制和控制策略。将以脑胶质瘤为研究实例，通过建立细胞群体动力学模型研究细胞微环境对肿瘤耐药性的定量影响；基于单细胞转录组数据，构建肿瘤细胞与微环境细胞之间相互作用的信号通路网络，建立基于单细胞的多尺度时空模型，由此研究微环境的异质性和分子网络的随机性如何介导肿瘤耐药性的产生和发展的演化机制，并通过建立多目标优化模型研究药物组合的动态控制策略。本项目有助于从系统医学的观点深入理解肿瘤耐药性的细胞内外动态机制，并为优化药物治疗方案提供参考和理论依据。

在本项目的资助下，我们针对微环境介导的肿瘤耐药性的作用和机制，发展了一套单细胞转录组数据分析和多尺度建模的计算方法，为理解肿瘤耐药性和提高治疗效果提供了理论指导和技术支撑。主要取得以下研究成果：① 提出了一套基于单细胞RNA-seq数据构建细胞间和细胞内信号通路网络的方法（Briefings in Bioinformatics. 2021. 22(2):988-1005），并建立了基于单细胞的信号通路多尺度随机动力学模拟，采用ODEs和Gillespie算法耦合的方法模拟了EGFR信号网络模块决定细胞命运的机制（Briefings in Bioinformatics. 2020. 21(3):1080-1097）。② 基于自主发展的细胞通讯网络推断方法，分析胶质瘤单细胞转录组数据，发现了胶质瘤细胞和巨噬细胞互作的Gal-9/Tim-3细胞间信号及其下游调控网络，并与实验课题组合作通过动物实验验证表明Gal-9/Tim-3可作为PTEN缺失型脑胶质瘤的潜在免疫治疗新靶点（Science Advances. 2022, 8(27):eabl5165），为进一步的药物组合治疗提供了参考。③建立了新生血管介导肿瘤耐药的演化动力学模型（BMC Bioinformatics. 2019, 20(S7):203. 59-70），并考察了EGFR抑制剂与VEGFR抑制剂组合的时序控制。④研发了一种动力学模型和组学数据共同驱动的基因调控网络推断新方法（PLoS Computational Biology 2021；PLoS Computational Biology 2019；STAR Protocols 2022.），应用于乳腺癌、脑胶质瘤和感染免疫疾病等真实数据（Journal of Neuroinflammation 2018; Mathematical Biosciences and Engineering 2020; Frontiers in Immunology 2021）。已全部并超额完成预定研究目标。