Multi-Scale Cancer Modeling: From Cell Phenotypes to Cancer Spread and Response

总体而言，该项目旨在（1）增加对肿瘤和微环境中分子和细胞尺度变化与肿瘤生长之间联系的生物物理机制的见解，（2）开发技术，使我们能够轻松地将尖端实验测量纳入多尺度模型，从而更快地确定其对整体肿瘤进展和治疗反应的影响，（3）研究和量化肿瘤对治疗反应的时空动力学，（4）改善体内-计算机模拟发展反馈回路，形成真正的综合建模的支柱，因此（4）推动多尺度综合癌症建模的前沿。为了量化复杂的癌症现象在不同尺度之间的关系，我们利用离散和连续建模方法的优点，采用混合建模。我们实施混合，多尺度算法作为癌症建模的下一阶段，特别是淋巴瘤和白血病建模。这涉及到动态耦合肿瘤规模的模型和分子/细胞规模的模型开发的Cristini，Macklin和同事与细胞信号和进化/遗传模型开发的研究项目2。这还需要与Gambhir及其同事的肿瘤生长、血管形成和对化疗反应的最先进的活体时间过程测量相结合。这些实验将（1）为我们提供第一手的生物学数据，这些数据将塑造模型的发展，（2）为我们提供关键模型参数的精确测量，这些参数唯一地约束建模框架，（3）为模型验证和测试提供额外的独立测试，以及（4）为真正的综合建模提供机会，我们的第一轮研究校准模型导致后续实验，以测试新的癌症生物学假设并改进多尺度模型。

Overall, the project is designed to (1) add insights as to the biophysical mechanisms that link molecular- and cell-scale variations in the tumor and the microenvironment to the tumor's growth, (2) develop techniques that allow us to readily incorporate cutting-edge experimental measurements into the multi-scale model and thus more quickly determine their impact on overall tumor progression and response to therapy, (3) investigate and quantify the spatiotemporal dynamics of tumor response to therapy, (4) improve the in vivo-in silico development feedback loop that forms the backbone of true integrative modeling, and so (4) push the frontier of multi-scale, integrative cancer modeling. In order to quantify the relationships between complex cancer phenomena at different scales, we harness the advantages of both discrete and continuum modeling approaches by employing hybrid modeling. We implement hybrid, multi-scale algorithms as the next stage of cancer modeling in general, and lymphoma and leukemia modeling in particular. This involves dynamically coupling tumor-scale models and molecular/cell-scale models developed by Cristini, Macklin and coworkers with cell signaling and evolutionary/hereditary models developed by Research Project 2. This also requires integration with state of- the-art intravital time-course measurements of tumor growth, vascularization, and response to chemotherapy by Gambhir and co-workers. These experiments will (1) provide us with first-hand biological data that will shape the model development, (2) provide us with precise measurements of key model parameters that uniquely constrain the modeling framework, (3) provide additional, independent tests for model validation and testing, and (4) provide an opportunity for true integrative modeling, where our first round of investigating the calibrated model leads to follow-up experiments to test new cancer biology hypotheses and improve the multiscale model.