Image‐Based Modeling of the Invasive Vascular Front in Breast Cancer

Abstract

The vascular microenvironment at the tumor-host interface is known to play a crucial role in cancer progression and metastasis. This angiogenic interface involves the development of an invasive vascular front that undergoes dynamic remodeling as the tumor progresses. While immunohistochemistry for vascular markers and corrosion casting studies have improved our understanding of the properties of these interfacial blood vessels, they do not recapitulate the functional changes in this vascular bed. As a result, our understanding of this angiogenic microenvironment remains incomplete. Therefore, in this study we employed our existing library of eight whole-tumor microvascular networks from a preclinical breast cancer model to develop a computational image-based workflow to simulate the hemodynamics in the invasive vascular front. This was accomplished by computing tumor-wide 3D distributions of angiogenic characteristics such as vascular length density (mm/mm2) and surface area density (mm2/mm3), as well as simulating the perfusion density (ml/min/mm3) distributions using high-fidelity, 3D micro-CT images of the vasculature. Tumor boundary-to-center profiles were then employed to characterize how these parameters varied between the invasive vascular front and tumor center. We found that vascular and perfusion densities were highest within 0.5-1 mm from the boundary in all tumor samples. Interestingly, the vascular front in all tumor samples also exhibited patches of poorly vascularized regions. This structural and functional heterogeneity along the invasive vascular front was suggestive of a preferential direction of angiogenic growth in these tumors. In summary, this 3D image-based modeling workflow could be readily applied to any preclinical cancer model to characterize other features of the angiogenic microenvironment. Moreover, boundary-to-center profiles generated from this workflow could be incorporated in computational oncology drug delivery models or in silico investigations of the angiogenic microenvironment in cancer and other diseases.