Multiscale computational modeling of breast cancer invasion: Towards a predictive patient-based tool

Background: Understanding the mechanisms of the growth and spread of cancer cells in the human breast and providing effective treatment continues to remain a significant challenge. Cancer cell invasion of breast tissue plays a pivotal role in the metastatic cascade through complex, coupled dynamics of the constituent processes occurring over a range of spatial and temporal scales, ranging from genetic events through intracellular signaling pathways through cell-matrix interactions to tissue scale changes.

Method: While most previous computational modeling has been carried out at the level of a single spatial or temporal scale, the work presented here has developed a novel multiscale modeling and computational simulation framework. Formulated in both two and three spatial dimensions, this approach is based on a new type of multiscale moving boundary computational model that explores the crucial role that matrix degrading enzymes, secreted by individual cancer cells (micro-scale) along the invasive edge of the tumour, play in the overall dynamics of the tissue-scale invasion (macro-scale).

Results: The computational simulation results we have obtained reveal a rich variety of cancer invasion patterns that emerge in the vicinity of a primary breast cancer. In the presence of heterogeneous tissue, our computational simulation results show well-defined “lobular” and “fingering” invasion patterns, consistent with the spread of a breast carcinoma, and which remain robust with respect to any changes made in the model parameters. These patterns are found to be compatible with respect to both the dynamic heterogeneity of cancer cell macro-scale distributions and to the evolution of the heterogeneity of the surrounding breast tissue.

Conclusions: By exploring the dynamics of the molecular processes of matrix degrading enzymes and of tissue-scale spatio-temporal cancer tissue evolution, this novel multiscale computational modeling framework faithfully replicates the cancer invasion process. Development of this computational tool should allow pre-operative breast cancer patient data to be used to predict the extent of breast cancer invasion and hence the extent of surgical resection.