Assessment of the Risk of Residual Disease in Patients Operated for Invasive Breast Cancer with Positive Margins or Cavity Margin Shavings.

Introduction: The status of the surgical margin is an important factor associated with local recurrence after lumpectomy in breast cancer patients. Standard surgical practice is to obtain clear margins even if this requires a second surgical procedure. The aim of the present study was to identify factors predicting a positive histological re-excision specimen and to propose a score to evaluate the risk of residual disease (RRD). Material and Methods: A retrospective search of our Database from 2000 to 2005 was conducted to identify patients with the following inclusion criteria: invasive carcinoma, pT1-2 with positive margins or cavity margin shavings, pN0 to pN2, second surgery for re-excision. The exclusion criteria were multi-centric or multifocal carcinomas; pT3-4; extensive intraductal carcinoma with micro-invasive tumor, pre-operative treatments. A negative margin was defined by the absence of tumor cells at the inked margin. Positive margins were defined as tumor cells directly at the cut edge of the specimen, and close margins as tumor cells ≤ 2 mm from the cut edge. For cavity margin shavings, positivity was defined by the presence of tumor cells anywhere in the cavity shaving. A multivariate logistic regression model was used on the basis of the prior univariate analysis results to assess the independent relative risk on the presence of residual tumor and to establish a score to predict the RRD. Results: 1611 patients benefited from breast conservative surgery, and 186 met our selection criteria. Median age was 56 years, and post-menopausal status was 67%. Tumor size was pT1 in 75% (pT1a 11%, pT1b 27%, pT1c 37%) and pT2 in 24%. Node status was pN0 in 70% patients. Histotype at first surgery was invasive ductal in 77%, invasive lobular in 19% and both in 4%. An extensive intraductal component was associated in 25%. Cavity margin shavings were performed for 96% of the patients, and all had at least one positive shaving. Lumpectomy margins were positive in 51%, close in 25% and negative in 24%. Margin status was unknown in 11%. Residual disease was found in 66% of the re-excision. The second surgery consisted in conservative surgery for 56% of patients. The median time interval between primary surgery and re-excision was 32 days (range: 6-86). Histotype at re-excision could be mixed and was intraductal in 72%, invasive ductal in 27%, invasive lobular in 15%. In univariate analysis, the following factors were associated with a higher rate of residual disease: age Conclusion: In this highly selected population, the RRD is high, even when the score =0. Further study will compare the score groups with local relapse and survival. Meanwhile, re-excision should be considered for all patients with positive margins or cavity margin shavings involved. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 3116.