Margin status in ductal carcinoma in situ of the breast: radiological-pathological correlation in slice radiographs and histological large-format slides of breast conserving surgery specimens

Abstract

AIM: Positive margins in breast conserving surgery (BCS) specimens are a risk factor for local recurrence in ductal carcinoma in situ (DCIS). As DCIS often bears microcalcifications, intraoperative specimen or slice radiographs are widely used to guide BCS. As in some DCIS parts microcalcifications may be absent, margin widths based on imaging of microcalcifications may differ from final DCIS margin status. Our aim was to analyse the correlation of radiological and final pathological DCIS margin widths in slice radiographs of BCS specimens. METHODS: A total of 646 whole slices of 46 BCS specimens were analysed. 168 slices contained both DCIS and microcalcifications. Margin width of microcalcifications in the slice radiograph was measured by two radiologists and compared with final histopathological DCIS margins in whole slice large-format slides. RESULTS: The overall correlation between radiological and pathological margin widths was weak (r = 0.390; p < 0.001). Radiological overestimation of pathological margin widths occurred in 499 (74%) and underestimation in 77 (11 %) of 672 margins. The rate of histologically compromised margins depended on radiological margin width and was 36.8% for <2 mm, 29.3% for 2 to 5 mm, 12.2% for 6 to 9 mm, 7.1% for 10 to 19 mm, and 4% for ≥20 mm, respectively. CONCLUSION: Radiological margin widths tend to be considerably wider than those based on histological evaluation. A radiologically free rim of ≥10 mm in the BCS specimen reduces the rate of histologically compromised margins to <9%.