Predicting breast cancer response to therapy with MRI and modeling

Background/problem

Neoadjuvant therapy (NAT) for cancer treatment is currently based on generalized population-based data such as receptor status, tumor grade, body surface area, and genetic markers. This approach is inherently limited, as it does not account for the unique spatial and physiological characteristics of individual tumors. Consequently, approach to treatment plans and predicting tumor response to therapy are also suboptimal.

Current clinical trial systems can’t test all possible combinations, timings, and doses for specific cancer subtypes or individual patients. Mechanism-based modeling aims to incorporate various biological mechanisms into predictive models, but often requires extensive data and training on large populations. Thus, there is a critical need for technologies that can predict individual tumor responses and optimize therapeutic regimens based on each patient's unique tumor characteristics.

Tech overview/solution

This patented technology developed at The University of Texas at Austin is a groundbreaking approach to cancer treatment, leveraging quantitative MRI data to predict individual tumor responses to therapy. It uses advanced imaging techniques, such as dynamic contrast-enhanced MRI (DCE-MRI) and diffusion-weighted MRI (DW-MRI), to gather detailed information about a tumor’s size, cellularity, and vascular properties. These data inform a biophysical, reaction-diffusion model to simulate tumor growth, cellularity, and vascular characteristics, accounting for spatial drug distribution and effects on tumor cells. This approach enables data-based predictions of individual tumor responses to NAT, enabling personalized treatment plans that optimize effectiveness for individual patients.

Benefits/competitive advantage

This technology surpasses traditional methods that rely on generalized population data, offering a personalized approach that accounts for each patient’s unique tumor characteristics.

Key advantages include:

Patient-specific predictions use patient-specific MRI data to predict accurate therapy outcomes.
Integration of multiple MRI modalities combines data from DCE-MRI and DW-MRI for a comprehensive tumor analysis.
Mechanism-based modeling employs a biophysical, reaction-diffusion model for simulating tumor growth and therapy response at a cellular level.
Spatially resolved predictions provide spatially resolved predictions of tumor response, informing precise surgical planning and targeted therapies.
Optimization of treatment regimens suggests optimized treatment regimens based on predicted responses, potentially improving patient outcomes.

This technology offers a non-invasive method to monitor and predict tumor response, potentially reducing trial and error in treatment planning. Its implementation in community-based radiology centers also increases accessibility, making it a game-changer in the field of cancer therapy.

Opportunity

This technology offers a significant opportunity in the following fast-growing sectors: MRI market, personalized cancer therapy planning, predictive oncology and diagnostics, advanced medical imaging and clinical trial design (therapy optimization, patient-specific drug efficacy models).