Background: No consensus exists for clear margins for breast-conserving surgery for pure ductal carcinoma in situ (DCIS). We examined the implications of applying a "no ink on tumor" standard for pure DCIS by correlating clear margin width with rates of residual disease.
Methods: Lumpectomies with complete shaved cavity margins (SCMs) for pure DCIS at our institution from 2004 to 2007 were reviewed and patients with microinvasive cancer or multifocal disease requiring multiple wires excluded. Rates of residual disease in shaved margins were determined based on margin status of the main lumpectomy specimen using margin widths of "ink on tumor," ≤1, >1 to <2, and ≥2 mm.
Results: Overall, 182 women undergoing lumpectomy for pure DCIS met eligibility criteria. In patients with "ink on tumor" in the main lumpectomy specimen, 88 % had residual disease in the SCMs. Rates of residual disease in SCMs for lumpectomies with margins of <2 mm (but not on ink) were 52 % compared with 13 % for lumpectomies with margins ≥2 mm (p < 0.0005). Multivariate analyses confirmed the association of lumpectomy margin width and residual tumor in shaved cavity margins. Odds of residual disease in the SCM for postmenopausal patients were 74 % less than for pre/perimenopausal women (odds ratio 0.26; confidence interval 0.08-0.82).
Conclusions: Application of a "no ink on tumor" lumpectomy margin standard to patients with DCIS results in a significant increase in the rates of residual disease in cavity margins compared with use of a ≥2-mm margin standard. Use of narrower margins may have important implications for use of adjuvant therapy.