Factors Associated with Presence/Absence of Residual Disease after Initial Breast-Conserving Surgery for Ductal Carcinoma In Situ (DCIS).
Shi Wei, Christopher Kragel, Omar Hameed. University of Alabama at Birmingham
Background: Breast-conserving surgery (BCS) is a standard treatment option in women with localized DCIS. Margin status is considered a major risk factor for residual disease and local recurrence. Reexcision is common in BCS, partly due to lack of consensus on what might constitute a "positive" or "close" margin. In this study, we aimed to identify potential predictive factors for the presence/absence of residual disease after initial BCS.
Design: A search of our surgical pathology files was performed to identify cases with a diagnosis of DCIS at initial needle core biopsies (NCB) between January 2005 and December 2009. Those with DCIS only and with subsequent BCS were reviewed. The histologic features, the extent of tumor and margin status were recorded to identify factors significant for residual tumor.
Results: There were 759 cases with a diagnosis of DCIS present at initial NCB. Of 232 with DCIS only, 94 underwent mastectomy, and 108 underwent BCS. Among the BCS cases, 20 (43%) out of 47 cases with close margins (<2mm) had residual disease on reexcision. None of the histologic features analyzed (nuclear grade, necrosis, type, and calcification) was associated with residual tumor, whereas % of blocks with DCIS (55±17% vs. 29±15%; p<0.0001), the total number of close margins (2.6±1.3 vs. 1.5±0.8; p=0.002) and the number of involved ducts/TDLUs at the margin (30±19 vs. 13±11; p=0.0003) were significant predictors. The parameters significantly associated with the absence of DCIS were <33% of blocks with DCIS (p<0.0001; odds ratio (OR)=0.01, 95% CI 0-0.25), only one close margin (p=0.002; OR=0.1, 95% CI 0.02-0.44), and <10 ducts/TDLUs with DCIS at the margin (p=0.004; OR=0.1, 95% CI 0.02-0.52). Separate "cavity" margins were obtained from 68 (63%) patients. Although lack of DCIS in 18 (26%) of these additional margins prevented unnecessary subsequent reexcision, the presence of DCIS in 12 (18%) cases led to additional surgery.
Conclusions: In this study, 57% (27/47) of BCS for DCIS with "close" margins were not associated with residual disease on reexcision. While none of the histologic features was associated with residual tumor, the extent of DCIS in the whole specimen, the total number of close margins and the extent of DCIS at the margin were significant predictors. In addition, additional margin sampling does not necessarily reduce the reexcision rate, probably related to the multifocal and patchy nature of DCIS.
Monday, February 28, 2011 1:00 PM
Poster Session II # 41, Monday Afternoon