Meeting Banner
Abstract #1723

Predicting Pathological Response to Neoadjuvant Chemotherapy Using 3D Texture Feature Radiomics on Baseline Dynamic Contrast-Enhanced (DCE) MRI

Yifan Wu1, Daniel S. Hippe1, Ginger L. Lash1, Lanell M. Peterson1, Jennifer M. Specht2, and Savannah C. Partridge1

1Radiology, University of Washington, Seattle, WA, United States, 2Medicine, University of Washington, Seattle, WA, United States

There is emerging data supporting the value of texture and other radiomics features extracted from dynamic contrast-enhanced (DCE) MRI to characterize breast cancer subtypes and recurrence risk. DCE texture features may also provide unique value in predicting response to neoadjuvant chemotherapy (NAC). Our study investigated the predictive value of pretreatment DCE tumor texture features in 30 women with triple negative and luminal-B cancers undergoing NAC. We found higher-order texture features significantly predicted pathologic response, while other standard quantitative metrics did not. Our findings suggest texture features on DCE MRI may provide valuable information prior to treatment to help tailor therapies.

This abstract and the presentation materials are available to members only; a login is required.

Join Here