BACKGROUND: Histologically positive margins are generally considered unacceptable with breast conserving therapy (BCT) given the increased risk of local recurrence (LR). What constitutes an adequate negative margin remains controversial. Margin status was explored as a predictor of LR post-BCT. METHODS: Manitoba women with loco-regional progression and/or mastectomy >6 months following BCT for Stage I/II invasive cancer (1995-2004) were identified from the Manitoba Cancer Registry; LR cases were confirmed by chart review. Three controls per case were matched by age, grade, stage, and adjuvant chemotherapy use. Margin status was categorized as histologically positive, < or =1 mm, < or =2 mm or >2 mm. Conditional logistic regression determined the odds ratio of LR by margin category. RESULTS: There were 50 LR cases in 3,017 patients who underwent BCT, with a median follow-up of 60 months. Wider margins were associated with a non-significant reduction in LR: >1 mm versus < or =1 mm (OR 0.69; 95% CI 0.28-1.69) and >2 mm versus < or =2 mm (OR 0.90; 95% CI 0.44-1.84). CONCLUSIONS: No clear benefit to wider histologically negative margins is demonstrated. (c) 2008 Wiley-Liss, Inc.
BACKGROUND: Histologically positive margins are generally considered unacceptable with breast conserving therapy (BCT) given the increased risk of local recurrence (LR). What constitutes an adequate negative margin remains controversial. Margin status was explored as a predictor of LR post-BCT. METHODS: Manitoba women with loco-regional progression and/or mastectomy >6 months following BCT for Stage I/II invasive cancer (1995-2004) were identified from the Manitoba Cancer Registry; LR cases were confirmed by chart review. Three controls per case were matched by age, grade, stage, and adjuvant chemotherapy use. Margin status was categorized as histologically positive, < or =1 mm, < or =2 mm or >2 mm. Conditional logistic regression determined the odds ratio of LR by margin category. RESULTS: There were 50 LR cases in 3,017 patients who underwent BCT, with a median follow-up of 60 months. Wider margins were associated with a non-significant reduction in LR: >1 mm versus < or =1 mm (OR 0.69; 95% CI 0.28-1.69) and >2 mm versus < or =2 mm (OR 0.90; 95% CI 0.44-1.84). CONCLUSIONS: No clear benefit to wider histologically negative margins is demonstrated. (c) 2008 Wiley-Liss, Inc.