The percentage of unnecessary mastectomy due to false size prediction using preoperative ultrasonography and MRI in breast cancer patients who underwent neoadjuvant chemotherapy: a prospective cohort study

Background: Imaging-estimated tumour extent after neoadjuvant chemotherapy tends to be discordant with the pathological extent. The authors aimed to prospectively determine the proportion of decisions regarding total mastectomy for potential breast-conserving surgery candidates owing to false size prediction with imaging in neoadjuvant chemotherapy and non-neoadjuvant chemotherapy patients. Materials and methods: The authors prospectively enroled clinical stage II or III breast cancer patients who are scheduled for total mastectomy between 2018 and 2021. This study was conducted at Seoul National University Hospital at South Korea. Before surgery, each surgeon recorded the hypothetical maximum tumour size at which the surgeon would have been able to attempt breast-conserving surgery if the patient had actually less than the size of the tumour at that location in the breast. After surgery, the hypothetical maximum tumour size was compared with the final pathologic total extent of the tumour, including invasive and in situ cancers. Results: Among the 360 enroled patients, 130 underwent neoadjuvant chemotherapy, and 230 did not undergo neoadjuvant chemotherapy. Of the total of each group, 47.7% in the neoadjuvant chemotherapy group and 21.3% in the non-neoadjuvant chemotherapy group had a smaller pathologic tumour extent than the pre-recorded hypothetical maximum tumour size (P<0.001). Further analyses were conducted for the neoadjuvant chemotherapy group. The proportions of total mastectomy with false size prediction were higher in HER2-positive (63.3%) and triple-negative (57.6%) patients compared with ER-positive/HER2-negative (25.0%) patients (P<0.001). Both magnetic resonance imaging-pathology and ultrasonography-pathology size discrepancies were significantly associated with false decisions for total mastectomy (both P<0.001). Without magnetic resonance imaging, the false decision may be reduced by 21.5%. Conclusion: A total of 47.7% of patients who received total mastectomy after neoadjuvant chemotherapy were breast-conserving surgery eligible, which was significantly higher than that of non-neoadjuvant chemotherapy patients. Magnetic resonance imaging contributed the most to false size predictions.