THE RELATIONSHIP BETWEEN MICROSCOPIC MARGINS OF RESECTION AND THE RISK OF LOCAL RECURRENCE IN PATIENTS WITH BREAST-CANCER TREATED WITH BREAST-CONSERVING SURGERY AND RADIATION-THERAPY

Background. The relationships among the involvement of tumor at the fi nal margins of resection, the presence of an extensive intraductal com ponent (EIC), and the risk of local recurrence are important considera tions in patients treated with conservative surgery and radiation ther apy for early stage breast cancer but have not been defined adequately . Methods. Between 1982 and 1985, 885 patients were treated for clinic al Stage I or II invasive breast cancer. The study population was limi ted to 181 patients with an infiltrating ductal carcinoma who received a radiation dose to the surgical site of 60 Gy or greater, whose fina l microscopic margins of resection were evaluable, and who had at leas t 5 years of follow-up. A positive margin was defined as tumor present at the inked margin of resection, a close margin as tumor within 1 mm of the inked margin, and a negative margin as no tumor within 1 mm of the inked margin. A focally positive margin was defined as tumor at t he margin in three or fewer low-power fields. In 157 patients (87%), t he tumor was evaluable for the presence or absence of an EIC. The medi an follow-up was 86 months. Results. In 12 of 181 patients (7%), a rec urrence developed at or near the primary site (true recurrence/margina l miss [TR/MM]) within 5 years. The 5-year rate of TR/MM (with 95% con fidence intervals) among patients with negative, close, focally positi ve, and more than focally positive margins was 0% (0-4%), 4% (0-20%), 6% (1-17%) and 21% (10-37%), respectively. Patients with positive marg ins also were more likely to have a distant failure within 5 years (14 %, 8%, 25%, and 32% in the four groups, respectively). However, patien ts with positive margins more often had positive axillary lymph nodes than patients with negative or close margins (59% vs. 38%, P < 0.02). The 5-year rate of TR/MM was 20% for patients with an EIC-positive tum or and 7% for patients with an EIC-negative tumor. However, among the 127 patients with an EIC-negative tumor, the 5-year rate of TR/MM was less than 10% in all margin groups. Among the 30 patients with an EIC- positive tumor, the 5-year rate of TR/MM was 0% when margins were nega tive or close but 50% when margins were more than focally positive. Co nclusions. These results provide support for the use of breast-conserv ing surgery and breast irradiation in all patients with uninvolved mar gins, whether the tumor is EIC-positive or EIC-negative. This study su ggests that breast-conserving therapy (including a radiation boost to the primary site) also may be a reasonable option for some patients wi th an EIC-negative tumor and margin involvement.