Abstract 4263: Characterization of patient-specific drug delivery for breast cancer using image-guided computational fluid dynamics

Abstract

Abstract Introduction: A tumor's blood supply and interstitial flow play an essential role in tumor growth, invasion, and treatment response. We have developed a methodology that employs quantitative MRI data to constrain a patient-specific, computational fluid dynamics (CFD) model of blood flow and interstitial transport within breast tumors. The dynamics of solute transport are characterized based on steady state flow fields, so that delivery of oxygen, nutrients, or therapies through the circulation can be estimated. It is a fundamentally new way to characterize breast tumor hemodynamics using MRI. Method: Eleven malignant and five benign lesions from 12 patients were included in this study. Vessel segmentation and tracking were performed to reconstruct the whole breast vasculature and to identify vessels feeding or draining tumors. Ultrafast dynamic contrast-enhanced MRI and diffusion-weighted imaging data were analyzed to estimate the bolus arrival time, Ktrans (volume transfer coefficient), and ADC (apparent diffusion coefficient); these data were used to spatially assign flow direction, local vascular permeability, and tissue density, respectively. The model of steady state flow was described by three components: 1) blood flow following Poiseuille's law, 2) interstitial flow following Darcy's law, and 3) flux transmitted across the vascular walls following Starling's law. The behavior of drug delivery was described by an advection-diffusion equation in the interstitial tissue, with the profile of the vascular bolus as a source term. The whole system was solved with a 1D-3D coupled implementation. At the end of analysis, the tracer's propagation through the tissue-of-interest was visualized and hemodynamic characteristics are derived to compare the malignant and benign lesions. Result: Visualization of the time-resolved distribution and propagation of solute demonstrated the intratumoral heterogeneity of accessibility to drugs. Furthermore, we are currently calculating the spatially-resolved accumulation, wash-in rate, and infiltration duration of different drugs for each lesion, so that a quantitative comparison can be performed between malignant and benign lesions. Conclusion: We have developed a computational model, informed by patient-specific MRI data, to simulate the blood supply, interstitial fluid environment, and intratumorally heterogeneous access to therapies for breast tumors. It represents the first methodology of integrating CFD with patient specific MRI data for quantifying the spatiotemporally resolved drug propagation as well as the entire pressure and flow fields within the breast. NCI U01CA142565, U01CA CA174706, and R01 CA218700. CPRIT RR160005. Citation Format: Chengyue Wu, David A. Hormuth, Federico Pineda, Gregory S. Karczmar, Robert D. Moser, Thomas E. Yankeelov. Characterization of patient-specific drug delivery for breast cancer using image-guided computational fluid dynamics [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4263.