Quantitative in vivo imaging to enable tumor forecasting and treatment optimization

Abstract

Current clinical decision-making in oncology relies on averages of large patient populations to both assess tumor status and treatment outcomes. However, cancers exhibit an inherent evolving heterogeneity that requires an individual approach based on rigorous and precise predictions of cancer growth and treatment response. To this end, we advocate the use of quantitative in vivo imaging data to calibrate mathematical models for the personalized forecasting of tumor development. In this chapter, we summarize the main data types available from both common and emerging in vivo medical imaging technologies, and how these data can be used to obtain patient-specific parameters for common mathematical models of cancer. We then outline computational methods designed to solve these models, thereby enabling their use for producing personalized tumor forecasts in silico, which, ultimately, can be used to not only predict response, but also optimize treatment. Finally, we discuss the main barriers to making the above paradigm a clinical reality.

Integrating mechanism-based modeling with biomedical imaging to build practical digital twins for clinical oncology

“…The behavior and interaction of tumor cells with, for example, healthy cells, immune cells, and vasculature can be described through the language of ordinary differential equations (ODEs), which describe the change in the quantity of species over time, or partial differential equations (PDEs), which describe the change in the quantity of species over both time and space [35,36,99,[132][133][134] (Figure 2b).…”

“…This limitation is overcome by PDE models, which often extend ODE formulations to incorporate the movement of the modeled species and their interaction(s) with spatially-varying tissue properties [35,36,99,[132][133][134]138]. Tumor cell movement is often modeled via a diffusion term which can be randomly defined or informed by tissue type [52], mechanical properties [154], or tissue anisotropy [155].…”

“…Additionally, by integrating certain PDE model variables over a region of interest it is possible to estimate the scalar quantities of interest usually employed in ODE models (e.g., tumor volume). However, to solve and parameterize PDE models it is necessary to employ more advanced numerical methods, which are usually more computationally intensive [132,138].…”

Learn-Morph-Infer: A new way of solving the inverse problem for brain tumor modeling

“…Personalization of the clinical (irradiation) target volume, could spare more healthy tissue and increase progression-free survival by potentially avoiding recurrence (Stupp et al, 2014, Harpold et al, 2007, Jackson et al, 2015, Lipkova et al, 2019. Current computational approaches for personalizing radiotherapy planning often rely on solving an inverse problem for GBM growth models (Hogea et al, 2008, Konukoglu et al, 2010b, Geremia et al, 2012, Menze et al, 2011, Le et al, 2017, Lipkova et al, 2019, Scheufele et al, 2020, Subramanian et al, 2020a, Lorenzo and et al, 2021. In this context, the growth (forward) models are based on partial differential equations (PDEs) that describe the evolution of tumor cell density in the brain anatomy.…”

Learn-Morph-Infer: a new way of solving the inverse problem for brain tumor modeling

“…Personalization of the clinical (irradiation) target volume, could spare more healthy tissue and increase progression-free survival by potentially avoiding recurrence [1][2][3][4] . Current computational approaches for personalizing radiotherapy planning often rely on solving an inverse problem for GBM growth models [4][5][6][7][8][9][10][11][12][13] . In this context, the growth (forward) models are based on partial differential equations (PDEs) that describe the evolution of tumor cell density in the brain anatomy.…”