B. J. Schmidt,1 D. W. Bartlett,2 S. Agrawal,1 M. Reed,2 M. Jure-Kunkel,1 A. A. Gutierrez,1 R. A. Clynes,1 B. S. Fischer,1 A. Kadambi,2 C. Friedrich,2 K. Kudrycki,2 A. Roy,1 T. A. Leil1; 1Bristol-Myers Squibb, Princeton, NJ, 2Rosa & Co., San Carlos, CA

BACKGROUND: Mechanistic models capable of integrating datasets from the molecular, cellular, and tissue level to provide research predictions of tumor response are well-positioned to play a central role in translational research and clinical development for the emerging immuno-oncology therapeutic paradigm. The availability of calibration and validation data from clinical trials from the first successful immuno-oncology therapies such as ipilimumab and nivolumab (including CA184004, MDX1106-03, CA209004, CA209009) facilitates comparison of the simulated outcomes with clinical data.
METHODS: A multidisciplinary team developed the biological scope of a mechanistic, ODE-based simulation platform. The initial platform focuses on the interactions of multiple immune cell types, cancer cells, soluble mediators, cell-cell contact effects, as well as ipilimumab and nivolumab therapies within the microenvironment of a prototypical simulated lesion and effect on tumor shrinkage.
RESULTS: The platform was calibrated, taking into account nivolumab and ipilimumab plasma concentrations, circulating absolute lymphocyte counts, trends in tumor cytokines, an IFNγ gene expression signal, changes in tumor infiltrating lymphocytes, and lesion size data. In agreement with clinical observations, an enhancement in lesion response was observed with the combination therapy.
CONCLUSION: The platform recapitulates essential immune response pathways in a simulated lesion and exhibits qualitative agreement with patient response phenotypes to immuno-oncology agents. Having demonstrated proof-of-principle with a preliminary calibration, the platform will serve as a framework to facilitate biomarker identification, integrate additional therapeutic mechanisms, propose new combination strategies, and serve as a sub-model within a broader simulation framework for the cancer-immunity cycle.