Systems biology modeling of radiation resistance and chemotherapy-radiation combination therapies in head and neck squamous cell carcinoma

肿瘤对放射治疗的抵抗仍然是癌症患者长期存活的重要障碍， 尤其是头颈部鳞状细胞癌（HNSCC），这是一种长期预后不良的癌症类型 (less 5年生存率50%以上）。为了克服抗辐射的问题， 治疗与放射增敏化学疗法相结合。预测一个人对 放疗和特定的化疗-放疗联合治疗可以改善 为癌症患者制定个性化治疗计划。正在努力建立系统， 用于生物标志物发现和治疗反应预测的癌细胞生物学模型;然而，由于 方法上的缺陷，未能在基因组规模上整合多组学数据，以及缺乏特异性， 对于单个患者肿瘤，这些预测模型尚未在临床上实施。为了满足这些需求， 本项目的目标是开发一个个性化的系统生物学建模平台， 预测HNSCC患者肿瘤对放射和化疗-放射组合疗法的反应。 这些模型将首先通过整合来自研究中心的个体患者的全面生物学数据来创建。 癌症基因组图谱（TCGA）。这种方法将允许比较代谢、信号传导和 放射敏感性和放射抗性患者肿瘤之间的表型特征。然后通过整合 放射治疗和放射增敏化疗的作用机制 框架，在个别放射治疗中对特定化疗-放射联合治疗的反应- 可以预测耐药患者肿瘤。机器学习分类器将根据TCGA患者数据开发 模型预测，以确定哪些生物和临床因素最能预测辐射敏感性 和化疗-放疗联合治疗的成功。据推测，不同的反应， 在放射抗性HNSCC肿瘤中的化学疗法-放射组合疗法已经完成 通过氧化还原代谢和信号传导，以及建模中的氧化还原生物学成分， 该框架将显著丰富联合治疗成功的预测性生物标志物列表。 虽然这个项目的重点将在HNSCC，这种系统生物学建模方法将是适用的 任何癌症类型。该项目的结果将是一组减少的临床可测量的生物标志物， 准确预测HNSCC患者对放疗和特异性化疗-放疗的反应 联合疗法，以及一个精确的医学平台，以测试临床相关的治疗策略。 这个项目是创新的，因为它结合了多组学癌症患者数据与最先进的系统 生物建模技术，以研究辐射抗性的生物机制，以及 预测个体放射抵抗患者的化疗-放射联合治疗反应。

Tumor resistance to radiation therapy remains a significant obstacle to long-term cancer patient survival, especially for head and neck squamous cell carcinoma (HNSCC), a cancer type with poor long-term outcomes (less than 50% advanced stage five-year survival). To overcome the problem of radiation resistance, radiation therapy is being combined with radiation-sensitizing chemotherapies. Prediction of an individual’s sensitivity to radiation and specific chemotherapy-radiation combination therapies prior to treatment would improve the development of personalized treatment plans for cancer patients. Efforts are being made to create systems biology models of cancer cells for biomarker discovery and prediction of treatment response; however, due to methodological shortcomings, failure to integrate multi-omic data on a genome-scale, and lack of specificity to individual patient tumors, these predictive models have yet to be implemented clinically. To address these needs, the objective of this project is to develop a personalized systems biology modeling platform for individualized prediction of HNSCC patient tumor response to radiation and chemotherapy-radiation combination therapies. These models will be created by first integrating comprehensive biological data on individual patients from The Cancer Genome Atlas (TCGA). This approach will allow for the comparison of metabolic, signaling, and phenotypic signatures between radiation-sensitive and radiation-resistant patient tumors. By then integrating the mechanisms of action of radiation therapy and radiation-sensitizing chemotherapies into the modeling framework, the response to particular chemotherapy-radiation combination therapies in individual radiation- resistant patient tumors can be predicted. Machine learning classifiers will be developed from TCGA patient data and model predictions to determine which biological and clinical factors are most predictive of radiation sensitivity and chemotherapy-radiation combination therapy success. It is hypothesized that differential response to chemotherapy-radiation combination therapies in radiation-resistant HNSCC tumors is accomplished through redox metabolism and signaling, and components of redox biology within the modeling framework will significantly enrich the list of predictive biomarkers for combination therapy success. Although the focus of this project will be on HNSCC, this systems biology modeling approach will be applicable to any cancer type. The outcomes of this project will be a reduced set of clinically-measurable biomarkers for accurate prediction of HNSCC patient response to radiation therapy and specific chemotherapy-radiation combination therapies, as well as a precision medicine platform to test clinically relevant therapeutic strategies. This project is innovative because it combines multi-omic cancer patient data with state-of-the-art systems biology modeling techniques to investigate the biological mechanisms of radiation resistance, as well as to predict chemotherapy-radiation combination therapy response in individual radiation-resistant patients.