Purpose: The association between a positive resection margin and the risk of ipsilateral breast tumor recurrence (IBTR) after conservative surgery and radiation is controversial. The width of the resection margin that minimizes the risk of IBTR is unknown. While adjuvant systemic therapy mag decrease the risk of an IBTR in all patients, its impact on patients with positive or close margins is largely unknown. This study examines the interaction between margin status, margin width, and adjuvant systemic therapy on the 5- and 10-year risk of IBTR after conservative surgery and radiation. Methods and Materials: A series of 1,262 patients with clinical Stage I or II breast cancer were treated by breast-conserving surgery, axillary node dissection, and radiation between March 1979 and December 1992. The median follow-up was 6.3 years (range 0.1-15.6). The median age was 55 years (range 24-89). Clinical size was T1 in 66% and T2 in 34%. Seventy-three percent of patients were node-negative. Only 5% of patients had tumors that were EIC-positive. Forty-one percent had a single excision, and 59% had a reexcision. The final margins were negative in 77%, positive in 12%, and close (less than or equal to 2 mm) in 11%. The median total dose to the tumor bed was 60 Gy with negative margins, 64 Gy with close margins, and 66 Gy with positive margins. Chemotherapy +/- tamoxifen was used in 28%, tamoxifen alone in 20%, and no adjuvant systemic therapy in 52%. Results: The 5-year cumulative incidence (CI) of IBTR was not significantly different between patients with negative (4%), positive (5%), or close (7%) margins. However, by 10 years, a significant difference in IBTR became apparent (negative 7%, positive 12%, close 14%,p = 0.04). There was no significant difference in IBTR when a close or positive margin was involved by invasive tumor or DCIS. Reexcision diminished the IBTR rate to 7% at 10 years if the final margin was negative; however, the highest risk was observed in patients with persistently positive (13%) or close (21%) (p = 0.02) margins. The median interval to failure was 3.7 years after no adjuvant systemic therapy, 5.0 years after chemotherapy +/- tamoxifen, and 6.7 gears after tamoxifen alone. This delay to IBTR was observed in patients with close or positive margins, with little impact on the time to failure in patients with negative margins. The 5-year CI of IBTR in patients with close or positive margins was 1% with adjuvant systemic therapy and 13% with no adjuvant therapy. However, by 10 years, the CI of IBTR was similar (18% vs. 14%) due to more late failures in the patients who received adjuvant systemic therapy. Conclusion: A negative margin (> 2 mm) identifies patients with a very low risk of IBTR (7% at 10 Sears) after conservative surgery and radiation. Patients with a close margin (less than or equal to 2 mm) are at an equal or greater risk of IBTR as with a positive margin, especially following a reexcision. A margin involved by DCIS or invasive tumor has the same increased risk of IBTR. A reexcision of an initially close or positive margin that results in a negative final margin reduces the risk of IBTR to that of an initially negative margin. A close or positive margin is associated with an increased risk of IBTR even in patients who are EIC-negative or receiving higher boost doses of radiation. The median time to IBTR is delayed; however, the CI is not significantly decreased by adjuvant systemic therapy in patients with close or positive margins-the 5 year results in these patients underestimate their ultimate risk of recurrence. (C) 1999 Elsevier Science Inc.
