PATHOLOGICAL MARGIN INVOLVEMENT AND THE RISK OF RECURRENCE IN PATIENTS TREATED WITH BREAST-CONSERVING THERAPY

BACKGROUND. The relationship between the microscopic margins of resect ion and ipsilateral breast recurrence (IBR) after breast-conserving th erapy for carcinomas with or without an extensive intraductal componen t (EIC) has not been adequately defined. METHODS. Of 1,790 women with unilateral clinical Stage I or II breast carcinoma treated with radiat ion therapy as part of breast-conserving therapy, 343 had invasive duc tal histology evaluable for an extensive intraductal component (EIC), had inked margins that were evaluable on review of their pathology sli des, and received greater than or equal to 60 Gray to the tumor bed; t hese 343 women constitute the study population. The median follow-up w as 109 months. All available slides were reviewed by one of the study pathologists. Final inked margins of excision were classified as negat ive > 1 mm (no invasive or in situ ductal carcinoma within 1 mm of the inked margin); negative less than or equal to 1 mm, or close (carcino ma less than or equal to 1 mm from the inked margin but not at the mar gin); or positive (carcinoma at the inked margin). A focally positive margin was defined as invasive or in situ ductal carcinoma at the marg in in three or fewer low-power fields. The first site of recurrent dis ease was classified as either ipsilateral breast recurrence (IBR) or d istant metastasis/regional lymph node failure. RESULTS. Crude rates fo r the first site of recurrence were calculated first for all 340 patie nts evaluable at 5 years, then separately for the 272 patients with EI C-negative cancers and the 68 patients with EIC-positive cancers. The 5-year rate of IBR for all patients with negative margins was 2%; and for all patients with positive margins, the rate was 16%. Among patien ts with negative margins, the 5-year rate of IBR was 2% for all patien ts with close margins (negative less than or equal to 1 mm) and 3% for those with negative > 1 mm margins. For patients with dose margins, t he rates were 2% and 0% for EIC-negative and EIC-positive tumors, resp ectively; the corresponding rates for patients with negative margins > 1 mm were 1% and 14%. The 5-year rate of IBR for patients with focall y positive margins was 9% (9% for EIC-negative and 7% for EIC-positive patients). The 5-year crude rate of IBR for patients with greater tha n focally positive margins was 28% (19% for EIC-negative and 42% for E IC-positive patients). CONCLUSIONS. Patients with negative margins of excision have a low rate of recurrence in the treated breast, whether the margin is > 1 mm or less than or equal to 1 mm and whether the car cinoma is EIC-negative or EIC-positive. Among patients with positive m argins, those with focally positive margins have a considerably lower risk of local recurrence than those with more than focally positive ma rgins, and could be considered for breast-conserving therapy. (C) 1996 American Cancer Society.