Optimization of Immunotherapy Treatment Regimen for Glioblastoma

The most common and aggressive primary brain cancer, Glioblastoma multiforme (GBM), exhibits a highly immune-suppressed tumor microenvironment. Monotherapy with anti-PD-1—a common immunotherapy which targets the PD-L1/PD-1 axis—has proved to be unsuccessful likely due to added layers of immune suppression. Murine experiments show that CCR2⁺ myeloid cells are chemo-kinetically recruited by gliomas to the brain, where they add to the suppression of T cells. Further, combination treatment with anti-PD-1 and a CCR2 antagonist reduced tumor growth by targeting these two forms of immune suppression. Using a multidimensional ODE model, we evaluate the immune dynamics of the glioma microenvironment. We then optimize administration of the combination immunotherapy with our overall objective being minimization of the tumor burden and the effects of drug toxicity. Methods used in this work include parameter estimation via Approximate Bayesian Computation (ABC) rejection method, sensitivity analysis using extended Fourier Amplitude Sensitivity Test (eFAST), global stability analysis of the unique tumor-free equilibrium, and the use of optimal control methods set forth by Pontryagin’s minimum principle.