Non-linear tumor-immune interactions arising from spatial metabolic heterogeneity

A hybrid multiscale mathematical model of tumor growth is used to investigate how tumoral and microenvironmental heterogeneity affect the response of the immune system. The model includes vascular dynamics and evolution of metabolic tumor phenotypes. Cytotoxic T cells are simulated, and their effect on tumor growth is shown to be dependent on the structure of the microenvironment and the distribution of tumor phenotypes. Importantly, no single immune strategy is best at all stages of tumor growth. I. TUMOR-IMMUNE INTERACTIONS Both the innate and adaptive arms of the immune system have been shown to induce tumor cell death [1, 2]. Immune surveillance is widely thought to eradicate many microscopic cancers before they become clinically-apparent. Conversely, clinical cancers are assumed to have developed adaptive strategies to evade immune attack. We examine a simplified tumor-immune model that pits T cells against a small growing tumor. In some cases, the immune system is able to remove a small tumor before it becomes detectable, but here we focus on the case where a tumor does eventually escape, and look at the tumor-immune interactions. There are multiple mechanisms of immune evasion that tumors use, including immunosuppressive surface markers such as PDL1 [3], down-regulation of antigen presentation machinery [4], recruitment of immunosuppressive immune cells [5], and secretion of immunosuppressive factors such as TGF-ß [6] and extracellular acid [7]. We focus on the latter here, a consequence of the altered glycolytic metabolism seen in many tumors. II. TUMOR-ENVIRONMENT HETEROGENEITY Heterogeneity is increasingly being recognized as an important aspect of tumor biology. Recently, Swanton and colleagues [8] showed that multiple biopsies from the same tumor display distinct genetic profiles and yet are phenotypically similar. This phenotypic convergence despite genotypic divergence has previously been examined theoretically [9] and may be a predictable evolutionary consequence of the tumor ecosystem [10, 11]. Quantifying this heterogeneity and the dynamics of its evolution remain a challenge, as does the understanding of how it relates to overall outcome [12-14]. Here we examine an environment that is temporally and spatially heterogeneous, largely due to variations in blood flow, resulting in local fluctuations of Research supported by the Moffitt Cancer Center PSOC, NIH/NCI U54CA143970. M. Robertson-Tessi (813-745-6818; e-mail: mark.robertsontessi@ moffitt.org), R. J. Gillies (email: [email protected]), R. A. Gatenby (email: [email protected]) and A. R. A. Anderson ([email protected]) are with the Moffitt Cancer Center, Tampa, FL 33612 USA. nutrients as well as variable infiltration of immune cells. The tumor cells are allowed to evolve, and selection pressures in each niche lead to the development of different tumor cell phenotypes. This heterogeneity then has implications for the immune response, both with respect to the tumor cells’ properties and the environments in which they reside.