Crowdsourcing optimal cancer treatment strategies that maximize efficacy and minimize toxicity

 描述（由申请人提供）：了解肿瘤启动、演变和对治疗作出反应的复杂时空过程是肿瘤学界的一个主要焦点，也是需要多学科整合的一个焦点。现代已经开发了一套多样化的疗法，导致许多癌症的生存率显着提高。然而，许多治疗方法都有一个短期成功然后复发的周期，通常是更具侵袭性的肿瘤。在过去的十年中，癌症研究界已经开始认识到跨基因型、表型和环境尺度的异质性的重要性，这是耐药性和治疗失败的关键驱动因素。肿瘤细胞和环境之间的复杂对话选择了表型上最适合生存的克隆，而不管可能促进肿瘤进展的特定突变。这些发生在异质性肿瘤和异质性环境（癌症生态系统）之间的动力学几乎不可能通过实验进行剖析。此外，向混合物中添加多种处理通常会导致非线性和不直观的动态。因此，了解肿瘤进化和生态如何随着治疗而变化是控制治疗后侵袭性和耐药性克隆出现的关键。我们的中心假设是，在治疗癌症时，我们应该利用异质性，而不是忽视它，通过开发众包序贯和联合疗法，引导肿瘤演变和生态产生更有效，毒性更小和更持久的反应。我们计划通过开发一个基于治疗异质性进化癌症的研究游戏来测试这一假设。游戏的核心引擎将是实体肿瘤生长的校准数学模型，通过不同的相关肿瘤表型，环境和治疗方案为特定器官部位量身定制。基于在与我们的研究游戏互动时观察到的模式，成功的玩家将基于对癌症对先前治疗的适应性反应以及癌症如何在真实的时间内对当前治疗作出反应的理解来选择后续治疗。由于众包计算和人类智慧的力量，我们将在各种癌症生态系统中得出一套最佳治疗策略。

 DESCRIPTION (provided by applicant): Understanding the complex spatial and temporal process by which tumors initiate, evolve and respond to therapy is a major focus of the oncology community and one that requires the integration of multiple disciplines. A diverse suite of therapies have been developed in the modern era, leading to significantly improved survival rates across many cancers. However, many treatments share a cycle of short-term success followed by recurrence, often of a more aggressive tumor. In the past decade, the cancer research community has begun to acknowledge the importance of heterogeneity across genotypic, phenotypic, and environmental scales as a key driver in drug resistance and treatment failure. The intricate dialogue between tumor cells and environment selects for clones that are best adapted phenotypically to survive, regardless of specific mutations that may facilitate tumor progression. These dynamics, occurring between a heterogeneous tumor and a heterogeneous environment (the cancer ecosystem) are almost impossible to dissect experimentally. Further, adding multiple treatments to the mix often leads to nonlinear and unintuitive dynamics. Therefore, understanding how tumor evolution and ecology changes with treatment is key to controlling the emergence of aggressive and resistant clones following therapy. Our central hypothesis here is that when treating cancer we should exploit heterogeneity, rather than ignore it, by developing crowdsourced sequential and combination therapies that steer tumor evolution and ecology producing more effective, less toxic and longer lasting responses. We plan to test this hypothesis through the development of a research game based on treating a heterogeneous evolving cancer. The core engine of the game will be a calibrated mathematical model of solid tumor growth, tailored to specific organ sites through different associated tumor phenotypes, environment and treatment options. Based on patterns observed while interacting with our research game, successful players will choose the follow-up treatments based on an understanding of the cancer's adaptive response to previous treatments as well as how the cancer is responding to the current therapy in real time. As a result of the power of crowdsourced computation and human intelligence we will derive a suite of optimal treatment strategies across a diverse set of cancer ecosystems.