Meeting Banner
Abstract #1593

Predicting response to neoadjuvant therapy using tumor morphologic features on breast MRI

Wen Li1, Rohan Nadkarni1, David C Newitt1, Natsuko Onishi1, Jessica Gibbs1, Lisa J Wilmes1, Bonnie N Joe1, Efstathios D Gennates2, John Kornak2, Mark Magbanua3, Laura van't Veer3, Barbara LeStage4, I-SPY 2 Consortium5, Laura J Esserman6, and Nola M Hylton1
1Radiology & Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, 2Epidemiology & Biostatistics, University of California, San Francisco, San Francisco, CA, United States, 3Laboratory Medicine, University of California, San Francisco, San Francisco, CA, United States, 4I-SPY 2 Advocacy Group, San Francisco, CA, United States, 5Quantum Leap Healthcare Collaborative, San Francisco, CA, United States, 6Surgery, University of California, San Francisco, San Francisco, CA, United States

Synopsis

Functional tumor volume (FTV) has been used to longitudinally assess tumor response to neoadjuvant therapy longitudinally in I-SPY trials. Previously, we found that a single 3D shape feature, sphericity, is associated with pathologic complete response (pCR). In this study, we expanded the analysis by including all radiomic shape features in an elastic net model and evaluated their role in prediction of pCR in the multi-center I-SPY 2 trial. Our results showed that the model with shape features outperformed the models with FTV or sphericity alone.

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

Join Here