Mathematical models of tumor growth are often parameterized and driven by quantities that can only be obtained invasively or in isolated, idealized in-vitro or ex-vivo systems; very few models are conceived in terms of measurements that can be made noninvasively and repeatedly. This limits their applicability and the ability to test them experimentally. Our concept is to develop mathematical models of tumor growth and response that incorporate and are constrained by imaging data obtained for individual tumors. We explored in simulations how quantitative imaging data obtained from multiple imaging modalities (DW-MRI, SPECT and PET) may serve as initial conditions for three simple mathematical models of tumor growth and treatment response. We incorporate imaging measurements of cellularity, cell proliferation, apoptosis and hypoxia into these mathematical models. Overall, the models explored here are subject specific and different initial condition maps can lead to different tumor shapes and volumes. We conclude that integrating quantitative imaging data into mathematical models of tumor growth is a promising combination that can capture the salient features of tumor growth and treatment response, and this indicates the direction for additional research.

The full article is available here (free to read).