Margin index is not a reliable tool for predicting residual disease after breast-conserving surgery for DCIS

Objective: We previously introduced the concept of margin index as a method for prediction of residual disease after attempted breast-conserving therapy (BCT). We sought to apply the margin index to patients with ductal carcinoma in situ (DCIS) to determine its reliability in predicting residual disease. Methods: We identified all patients with DCIS who were treated with BCT from 2004 to 2010. Margin index was calculated as follows: margin index = closest margin (mm)/tumor size (mm) ×100. A receiver operating curve was created using the derived margin index and the presence or absence of residual disease in the re-excision specimen. Sensitivity and specificity were calculated at various margin indices to identify the optimum margin index. Results: Of 380 patients undergoing attempted BCT, 109 (29%) underwent re-excision. Of 109 patients undergoing re-excision, 46 (42%) had positive margins and were excluded from the study, 15 (14%) were excluded due to inability to determine the size of DCIS on pathology reports, and 48 (44%) met study criteria and were included in the analysis. Of 48 patients undergoing re-excision, 19 (40%) had residual disease. The receiver operating curve c index was 0.65. However, there was no optimum margin index that reliably predicted the presence or absence of residual disease. Conclusions: Margin index is not a reliable method for prediction of residual disease after attempted BCT with close margins in patients with DCIS only. This may be a reflection of the complexities in accurately determining DCIS size and margin status in pathologic specimens.