Institute of Applied Simulation, Zurich University of Applied Sciences, Einsiedlerstrasse 31a, 8820 Wädenswil, Switzerland; ETH Zurich, Universitätstrasse 16, 8092 Zürich, Switzerland; Institute for Social and Preventive Medicine, University of Bern, Finkenhubelweg 11, 3012 Bern, Switzerland; Department of Biology, Stanford University, 371 Serra Mall, Stanford CA 94305, USA.
ETH Zurich, Universitätstrasse 16, 8092 Zürich, Switzerland.
J Theor Biol. 2020 May 7;492:110185. doi: 10.1016/j.jtbi.2020.110185. Epub 2020 Feb 6.
Cancer immunotherapies rely on how interactions between cancer and immune system cells are constituted. The more essential to the emergence of the dynamical behavior of cancer growth these interactions are, the more effectively they may be used as mechanisms for interventions. Mathematical modeling can help unearth such connections, and help explain how they shape the dynamics of cancer growth. Here, we explored whether there exist simple, consistent properties of cancer-immune system interaction (CISI) models that might be harnessed to devise effective immunotherapy approaches. We did this for a family of three related models of increasing complexity. To this end, we developed a base model of CISI, which captures some essential features of the more complex models built on it. We find that the base model and its derivates can plausibly reproduce biological behavior that is consistent with the notion of an immunological barrier. This behavior is also in accord with situations in which the suppressive effects exerted by cancer cells on immune cells dominate their proliferative effects. Under these circumstances, the model family may display a pattern of bistability, where two distinct, stable states (a cancer-free, and a full-grown cancer state) are possible. Increasing the effectiveness of immune-caused cancer cell killing may remove the basis for bistability, and abruptly tip the dynamics of the system into a cancer-free state. Additionally, in combination with the administration of immune effector cells, modifications in cancer cell killing may be harnessed for immunotherapy without the need for resolving the bistability. We use these ideas to test immunotherapeutic interventions in silico in a stochastic version of the base model. This bistability-reliant approach to cancer interventions might offer advantages over those that comprise gradual declines in cancer cell numbers.
癌症免疫疗法依赖于癌症和免疫系统细胞之间相互作用的构成方式。这些相互作用对于癌症生长的动态行为的出现越重要，它们就越有可能被用作干预机制。数学建模可以帮助揭示这些联系，并帮助解释它们如何塑造癌症生长的动态。在这里，我们探讨了是否存在简单、一致的癌症-免疫系统相互作用（CISI）模型特性，可以利用这些特性来设计有效的免疫治疗方法。我们对一组越来越复杂的三种相关模型进行了研究。为此，我们开发了一个 CISI 的基础模型，它捕捉了建立在其基础上的更复杂模型的一些基本特征。我们发现，基础模型及其衍生物可以合理地再现与免疫屏障概念一致的生物学行为。这种行为也与癌细胞对免疫细胞的抑制作用超过其增殖作用的情况一致。在这些情况下，模型家族可能表现出双稳态模式，其中两种不同的、稳定的状态（无癌症和完全生长的癌症状态）是可能的。提高免疫引起的癌细胞杀伤的有效性可能会消除双稳态的基础，并突然使系统的动力学进入无癌症状态。此外，结合免疫效应细胞的给药，对癌细胞杀伤的修饰可以被用于免疫治疗，而无需解决双稳态问题。我们在基础模型的随机版本中，使用这些想法在计算机上测试免疫治疗干预。这种依赖于双稳态的癌症干预方法可能优于那些逐渐降低癌细胞数量的方法。
