Introduction: Recently, contrast-enhanced mammography (CEM) has emerged as a reliable alternative to breast magnetic resonance imaging (MRI) for the assessment of pathological response in breast cancer patients. Our study sought to determine the diagnostic accuracy of CEM to predict pathological complete response (pCR) in patients who received neoadjuvant chemotherapy (NACT).

Methods: We retrieved the medical records of patients who underwent NACT at our institution. Using post-surgery pCR, morphological evidence and CEM enhancement tumours were classified as follows: 1) radiologic complete response (rCR); 2) functional radiological complete response (frCR); and 3) non-complete response. Initially, we used multivariate analyses adjusted by clinical variables and frCR or rCR to determine which variables affected pathological response. Then, CEM diagnostic accuracy to discriminate pCR was assessed using receiver operating characteristic curves in univariate and multivariate models including either frCR or rCR.

Results: A total of 48 patients were included in our study. Most patients (68.7%) were hormone receptor (HR)+ and 41.6% (20) of the patients achieved pCR. Using univariate logistic regression analyses we found that HR status, HER2 status, rCR and frCR had a significant impact on CEM diagnostic accuracy. Exploratory analyses found that CEM sensitivity was higher for HR− tumours. Multivariate logistic regression analyses found 60% sensitivity, 92.9% specificity and 79.2% accuracy in a model that included clinical variables and rCR.

Conclusion: CEM is a reliable alternative to high-cost, time-consuming breast MRI that predicts pCR in patients undergoing NACT; CEM diagnostic accuracy was higher among patients who harboured HR− tumours.