The percentage of unnecessary mastectomy due to false size prediction by preoperative imaging studies in patients with breast cancer who underwent neoadjuvant chemotherapy.

Abstract

573 Background: It was shown that, with neoadjuvant chemotherapy (NCT), about 40-50% of breast conserving surgery (BCS) ineligible breast cancer converted to BCS-eligible. Because the estimated tumor extent with imaging after NCT tend to be discordant with pathologic extent, we assumed that proportion of unnecessary total mastectomy (TM) might be greater in NCT patients compared with non-NCT patients. We aimed to determine the proportion of TM for potential BCS candidate due to false size prediction after NCT and factors affecting that. Methods: We prospectively enrolled patients with invasive breast cancer (Stage II or III) who were scheduled for TM from 2018 to 2021 at Seoul National University Hospital (SNUH). Patients who were determined to undergo TM by only patient’s preference, who had BRCA1/2 mutation, multicentric cancer, diffuse malignant calcification over more than a quadrant, and inflammatory cancer were excluded. The decision of TM was done by operating surgeon considering the tumor extent, location, and breast size. Breast imaging including mammography, ultrasonography (USG), and MRI were done in patients. Before surgery, each surgeon recorded a hypothetic maximum tumor size (HMS) that the surgeon would have been able to try BCS if the patient had actually less than that size of tumor at that location in the patient. After operation, the HMS was compared with final pathologic total extent of the tumor including invasive and in situ cancer. In SNUH, decisions and tries of BCS led to eventual BCS with negative margin in 97.4% of patients during same period with this study. Results: Among 360 patients enrolled, 130 were patients who underwent NCT, and 230 were chemotherapy naïve as a control group. 62 of 130 (52.3%) in the NCT group and 49 of 230 (21.3%) in the non-NCT group were found to have smaller tumor extent than the pre-recorded HMS, respectively (NCT vs non-NCT p-value <0.001). Further analyses were done only for NCT group. In an analysis according to subtype, the proportion of TM with false size prediction were more in HER2-positve (63.3%) and triple-negative breast cancer (TNBC) (57.6%) than in ER-positive/HER2-negative breast cancer (25.0%) (p-value <0.001). In an imaging analysis, MRI size were larger than HMS in 73.8% of all patients. Both MRI-pathology size and USG-pathology size discrepancies were significantly associated with false decision of TM (both p<0.001), but not with mammography. Without MRI, the false decision would have been reduced by 26.1%. Conclusions: We found that 52.3% of patients who received TM after NCT were actually BCS eligible, which was significantly higher than non-NCT patients. The proportion of false prediction were more in HER2-positive and TNBC. MRI imaging contributed to false size prediction and subsequent TM for BCS eligible patients. Clinical trial information: NCT04689529.