Abstract Introduction: Trastuzumab is a targeted antibody to the human epidermal growth factor receptor 2 (HER2) that induces cell cycle arrest and is used in the treatment of HER2+ positive breast cancer. We have recently presented in vivo evidence in a murine model that trastuzumab also improves vascular delivery of subsequent cytotoxic therapies. The mechanism by which trastuzumab and the immune system interact to regulate tumor-associated angiogenesis is not well characterized. Therefore, we offer a preliminary report on a mathematical framework to systematically investigate the potential interactions among the immune response, tumor cells, vasculature, and other environmental factors based on experimental data for the BT474 murine model of HER2+ breast cancer. Experimental: BT474 breast cancer cells were implanted subcutaneously into athymic nude mice. After tumors reached 250 mm3, mice were treated with trastuzumab or saline, tumor volumes were recorded, and tumors were extracted at various times over seven days. Immunohistochemistry for treated and control tumors were evaluated for 0, 1, 3, 4, and 7 days post trastuzumab treatment. Histology data includes: percent hypoxia (pimonidazole), percent necrosis, and vascular maturation index (VMI, ratio of alpha-smooth muscle actin to total vessel counts as stained with CD31). Ongoing studies are quantifying immune cell infiltration through immunofluorescent imaging of F4/80 and CD11c expression. Modeling: We developed a system of five coupled, ordinary differential equations that accounts for the temporal variation in tumor growth, vasculature, hypoxia, necrosis, and immune response. The general immune response component corresponds to the mouse's pro-inflammatory responses, as T cell driven responses are absent in this murine model. Uncertainty analysis was performed to verify plausible overlap between the model's predictions and the experimental data—where local and global samplings of all parameters were used to generate potential model results to be compared to the data for both control and treated mouse sets. Sensitivity analysis was performed using Sobol' Indices to determine the driving parameters of the system to identify target parameters for experimental estimation. The model parameters were calibrated using mean and standard deviations for the available data (tumor volume, VMI, percentage of hypoxia, and percentage of necrosis) for each experimental time point to predict the differences of the immune component between the treated and control tumors. Results and Discussion: The model is well behaved and can be calibrated with the available data. The model predicts distinct differences for the immune response between the control and treated groups—showing increasing versus decreasing immune component values over time for treated versus control results, respectively. Preliminary results from the immunofluorescent imaging data support the immunological predictions of the model; in particular, the amount of immune infiltration (i.e., more immune cells) in necrotic areas is greater in the treated than the untreated tumors (p<0.025). We acknowledge the support of CPRIT RR160005 and NCI R01CA186193. Citation Format: Jarrett AM, Yankeelov TE, Ehrlich LI, Godfrey W, Sorace AG. A mathematical model for predicting immune response in a trastuzumab treated HER2+ breast cancer animal model: Preliminary efforts [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P3-05-16.
Read full abstract