[Value of fused 18F-FDG PET/CT images in predicting efficacy of neoadjuvant chemotherapy on breast cancer].

BACKGROUND OBJECTIVE: Previous studies confirmed that fluorine-18 fluorodeoxyglucose positron emission tomography (18F-FDG PET) imaging relates to clinical or pathologic responses of tumors to neoadjuvant therapy. This study was to evaluate the correlation of fused 18F-FDG PET/CT images to cell apoptosis of breast cancer after neoadjuvant chemotherapy, and explore its value in predicting the efficacy of neoadjuvant chemotherapy on breast cancer. METHODS: Forty-five patients with primary breast cancer, proven by core needle biopsy, underwent 3 cycles of neoadjuvant chemotherapy. PET/CT was performed before and after treatment, and the ratio of tumor area activity to non-tumor area activity (T/N) was calculated. The apoptosis index (AI) was determined using TUNEL technique. RESULTS: Of the 45 patients, 4(8.9%) achieved complete remission (CR), 29 (64.4%) achieved partial remission (PR), 10 (22.2%) presented stable disease (SD), and 2 (4.4%) presented progressive disease (PD) after neoadjuvant chemotherapy. The mean T/N ratio was decreased from 3.23±0.63 before chemotherapy to 2.31±0.49 after chemotherapy (P=0.006) by 6.4%-50.8%. The mean AI was increased from (2.81±0.76)% before chemotherapy to (17.31±6.85)% after chemotherapy (P0.001) by 1.9%-41.3%. The T/N ratio reduction rate was positively correlated to AI change (rs=0.850, P0.001). At a threshold of 20% decrease from baseline in T/N ratio, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of PET/CT in predicting clinical response were 90.9%, 83.3%, 93.8%, 76.9%, and 92.1%, respectively. CONCLUSIONS: Neoadjuvant chemotherapy might effectively induce cell apoptosis in breast cancer and inhibit the glucose uptake. Fused PET/CT imaging is closely related to cell apoptosis status of breast cancer after neoadjuvant chemotherapy, and may be applied to predict the response of breast cancer to neoadjuvant chemotherapy.