Algorithmic asymptotic analysis: extending the arsenal of cancer immunology modeling

TL;DR

This paper applies algorithmic asymptotic analysis to a complex cancer immunology model, simplifying it to reveal key mechanisms and improve predictive accuracy during tumor progression, aiding treatment development.

Contribution

It introduces an algorithmic approach using Computational Singular Perturbation to derive a reduced, mechanistic model from a complex tumor-immune interaction system.

Findings

The long-term tumor dynamics depend on an initial explosive growth stage.

The reduced model accurately predicts explosive stage dynamics with fewer parameters.

Insights are consistent across different tumor-immune and patient scenarios.

Abstract

The recent advances in cancer immunotherapy boosted the development of tumor-immune system models aiming to provide mechanistic understanding and indicate more efficient treatment regimes. However, the complexity of such models, their multi-scale dynamics and their overparameterized character renders them inaccessible for wide utilization. In this work, the dynamics of a fundamental model formulating the interactions of tumor cells with natural killer cells, CD8$^+$ T cells and circulating lymphocytes is examined. It is first shown that the long-term evolution of the system towards high-tumor or tumor-free equilibria is determined by the dynamics of an initial \emph{explosive stage} of tumor progression. Focusing on this stage, the algorithmic Computational Singular Perturbation methodology is employed to identify the underlying mechanisms confining the system's evolution towards the equilibrium and the governing slow dynamics along them. It is shown that these insights are preserved along different tumor-immune system and patient-dependent realizations. Utilizing the obtained mechanistic understanding, a novel reduced model is constructed in an algorithmic fashion, which accurately predicts the dynamics of the system during the explosive stage and includes half of the parameters of the detailed model. This present analysis demonstrates the potential of algorithmic asymptotic analysis to simplify the complex, overeparameterized and multi-scale nature of cancer immunology models and to indicate the interactions and cell types to target for more effective treatment development.