Early Prediction of the Breast Cancer Response to Neoadjuvant Chemotherapy Based on Intratumoral Interstitial Fluid Pressure using 3-D Subharmonic-Aided Pressure Estimation

作者：Sung In Choi, Basak Dogan, Corinne E. Wessner, Jessica H. Porembka, Priscilla Machado, B Bersu Ozcan, Nisha Unni, Abu-Khalaf, MD, MBA, Maysa, Kenneth Hoyt, Flemming Forsberg, Kibo Nam · 发表于：Ultrasound in Medicine & Biology · 年份：2026 · DOI：10.1016/j.ultrasmedbio.2026.06.023 · 研究领域：Ultrasound and Hyperthermia Applications、Ultrasound Imaging and Elastography、Breast Cancer Treatment Studies

Elevated intratumoral interstitial fluid pressure (IFP) may negatively impact the efficacy of neoadjuvant chemotherapy (NAC). This study aimed to predict the breast cancer response to NAC based on intratumoral IFP using 3D contrast-enhanced ultrasound (CEUS) with subharmonic-aided pressure estimation (SHAPE). This prospective study enrolled breast cancer patients undergoing NAC from two clinical sites. 3D CEUS was performed at 0%, 10%, and 30% of NAC completion using a GE Logiq E9 with Definity. The pressure gradients between the tumor and adjacent normal tissue were estimated by SHAPE, utilizing the inverse linear relationship between the ultrasound contrast agent's subharmonic signal amplitude and hydrostatic pressure. The obtained SHAPE gradients were evaluated for the prediction of pathological complete response (pCR) and radiological response (RR, tumor volume reduction ≥90%), using t-test and receiver operating characteristic curve analysis. Additionally, the effect of incorporating tumor volume into the SHAPE gradient was evaluated. This study included thirty-seven patients. SHAPE gradients were significantly different between pCR and non-pCR at 0% (p = 0.03), 10% (p = 0.03), and 30% (p = 0.001) completion of NAC, and between RR and non-RR at 30% (p = 0.007) completion of NAC. The optimal predictive performance for SHAPE was achieved at 30% NAC completion, yielding area under the curve (AUC) values of 0.85 for pCR and 0.79 for RR. By combining the tumor volume-based pa...