Ultrasonic RF Time Series for Early Assessment of Tumor Response to Chemotherapy: First in Vivo Studies on Mice Breast Cancer Model

To evaluate whether ultrasonic radio-frequency (RF) time series allows assessing early changes in tumor microstructures in early prediction of tumor response to chemotherapy. Experiments were approved by the regional animal care committee. Fifty MCF-7 breast cancer bearing nude mice were treated either with cisplatin and paclitaxel or sterile saline. Sequential ultrasound imaging was performed on days 0, 3, 6 and 8 after the treatment to simultaneously collect B-mode images and RF data. The ultrasonic RF data was analysized with MATLAB-based software to extract 6 RF time series features including slope, intercept, S1, S2, S3, S4. The changes in RF time series features were compared to microstructures changes in histopathology. Treatment with cisplatin and paclitaxel significantly reduced tumor growth in comparison with the control tumors on day 6 and 8 (P＜0.05) while there was no significantly change between the control and treated tumors on day 0 and 3 (P =0.442, P =0.429, respectively). Intercept, S1 and S2 were significantly increased and slope was significantly decreased in the treated tumors on day 3, 6 and 8 as compared with day 0 (P＜0.01), but not in the control tumors (P＞0.05). On day 3, 6 and 8, intercept, S1, S2, S3 and S4 were significantly increased and slope was significantly decreased in treated tumors as compared with control tumors (P＜0.01). Tumor cell density was significantly decreased 3 days after treatment as compared with control tumors (P＜0.01). Ultrasonic RF time series analysis provided a simple way to noninvasively assess early tumor response to chemotherapy without injection of any contrast agents.