Objective: To investigate the value of changes in apparent diffusion coefficient (ADC) of diffusion-weighted MRI imaging for predicting the efficacy of neoadjuvant chemotherapy in breast cancer.  
Methods: The clinical data of 80 patients with locally advanced breast cancer treated from January 2017 to December 2018 were retrospectively analyzed. The breast cancer in all patients was confirmed by hollow needle aspiration. All patients underwent neoadjuvant chemotherapy, and routinely underwent MRI examination before neoadjuvant chemotherapy and before surgery. The relationship between the changes in tumor ADC values (ΔADC) and pathological response (MP grade) were analyzed. The effectiveness of ΔADC in estimating the efficacy of neoadjuvant chemotherapy was determined by the ROC curve.
Results: In the entire group of patients, the ADC value of tumors after chemotherapy was significantly higher than that before chemotherapy (1.27×10–3 mm2/s vs. 0.98 ×10–3 mm2/s, P=0.000). The cancer type-stratified analysis showed that except of no statistical difference between ADC values before and after neoadjuvant chemotherapy in triple negative breast cancer (P>0.05), the ADC values after chemotherapy was significantly higher than those before chemotherapy in all the remaining types of breast cancer (all P<0.05). The ΔADC value in patients with major histological response was significantly higher than that in those without significant histological response (0.448×10–3 mm2/s vs. 0.209×10–3 mm2/s, P=0.004). The area under ROC curve of ΔADC in estimating the efficacy of breast cancer neoadjuvant chemotherapy was 0.72, with the sensitivity of 73.1% and specificity of 66.7%. When the cutoff value ΔADC value was 0.239×10–3 mm2/s, the positive predictive value was 51.4%, negative predictive value was 83.7% and the accuracy was 68.8%.
Conclusion: In most types of breast cancer, the changes in ADC has certain value for early predicting and evaluating the treatment response of neoadjuvant chemotherapy.