Background: The goal of breast conserving surgery (BCS) in women with invasive breast cancer (IBC) or ductal carcinoma in situ (DCIS) is to remove all malignant cells from the breast or to reduce the residual tumor burden to a level that is likely to be eradicated by adjuvant radiation therapy and/or systemic therapy. Theoretical concerns exist that radiography of lumpectomy specimens and subsequent pathologic processing may lead to artifactually positive lumpectomy margins (LM), and the need for subsequent re-excision. To improve the chances of obtaining negative final margins (FM) (generally defined as 2 mm or greater), a number of surgeons in the US and abroad have adopted a technique in which 4-6 additional margins surrounding the main lumpectomy specimen, so-called “cavity margins” (CM), are routinely obtained. The purpose of this study was to assess the clinical utility of the routine use of CM in reducing the need for re-excision.

Methods: The surgical pathology records at our institution were searched for all cases of BCS with additional complete routine CM sampling performed by a single surgeon between May 2008 and April 2010. Patients in whom additional CM were obtained based on intraoperative findings of grossly close margins were excluded from the analysis. The demographic characteristics of the patients and the histological features of the tumor were recorded. We then tabulated the number of patients who had positive LM, CM, and FM, defined as either DCIS or invasive carcinoma at or less than 2 mm from inked specimen margin. The number needed to treat (NNT) to prevent one re-excision was calculated.

Results: We identified 97 women (74 with IBC and 33 with DCIS-only) who had undergone BCS with routine CM sampling during the study period. Mean patient age was 62. Median specimen volume was 178 cc3. 90% of the IBCs were ductal subtype. Mean IBC size was 1.5 cm. 63% of tumors were ER+, 58% PR+, and 7% HER2+. Of the 97 patients in this study, 64 had +LM and 33 had -LM. Of the 64 with +LM, 38 did not have tumor in the CM and 26 had tumor in the CM. Of these 26, 18 had +FM (re-excision needed) and 8 had -FM (no re-excision needed). The proportion of patients with actual tumor at ink (not just close margins) in the lumpectomy specimen was significantly higher among the group with tumor present in the CM than it was in the group with no tumor present in the CM (14/26 versus 7/38, p = 0.006). The NNT with routine CM compared with standard BCS to prevent one re-excision was 2.0 (95% CI, 1.5 — 2.1).

Discussion: In this study the routine use of CM reduced the need for reexcision in women undergoing BCS for IBC or DCIS; one re-excision was avoided for every two patients treated with this surgical approach. The absence of tumor in the CM in the majority of patients with positive LM suggests that specimen radiography and/or pathologic processing techniques may result in artifactually positive LM. Our results indicate that improved specimen imaging and pathological processing techniques are needed to increase the fidelity of margin assessment. In the meantime, the routine use of cavity margins may reduce the need for re-excision in women undergoing BCS.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P5-14-03.