Factors associated with local recurrence and cause-specific survival in patients with ductal carcinoma in situ of the breast treated with breast-conserving therapy or mastectomy.

PURPOSE: We reviewed our institution's experience treating patients with ductal carcinoma in situ (DCIS) of the breast to determine risk factors for ipsilateral breast tumor recurrence (IBTR) and cause-specific survival (CSS) after breast-conserving therapy (BCT) or mastectomy.
MATERIALS AND METHODS: Between 1981 and 1999, 410 cases of DCIS (405 patients) were treated at our institution; 367 were managed with breast-conserving surgery (54 with lumpectomy alone and 313 with adjuvant radiation therapy (RT) [median dose, 45 Gy]). Of these 313 patients, 298 received also a supplemental boost of RT to the lumpectomy cavity (median dose, 16 Gy). Forty-three patients underwent mastectomy; 2 (5%) received adjuvant RT to the chest wall. A true recurrence/marginal miss (TR/MM) IBTR was defined as failure within or adjacent to the tumor bed in patients undergoing BCT. Median follow-up for all patients was 7 years (mean: 6.1 years).
RESULTS: Thirty patients (8.2%) experienced an IBTR after BCT (25 [8%] after RT, 5 [9.3%] after no RT), and 2 patients (4.7%) developed a chest wall recurrence after mastectomy. Of the 32 local failures, 20 (63%) were invasive (18/30 [60%] after BCT and 2/2 [100%] after mastectomy), and 37% were DCIS alone. Twenty-four (80%) of the IBTRs were classified as TR/MM. The 10-year freedom from local failure, CSS, and overall survival after BCT or mastectomy were 89% vs. 90% (p = 0.4), 98% vs. 100% (p = 0.7), and 89% vs. 100% (p = 0.3), respectively. Factors associated with IBTR on Cox multivariate analysis were younger age (p = 0.02, hazard ratio [HR] 1.06 per year), electron boost energy < or = 9 MeV (p = 0.03, HR 1.41), final margins < or = 2 mm (p = 0.007; HR, 3.65), and no breast radiation (p = 0.002, HR 5.56). On Cox univariate analysis for BCT patients, IBTR, TR/MM failures, and predominant nuclear Grade 3 were associated with an increased risk of distant metastases and a reduced CSS.
CONCLUSIONS: After treatment for DCIS, 10-year rates of local control, CSS, and overall survival were similar after mastectomy and BCT. Young age (<45 years), close/positive margins (< or = 2 mm), no breast radiation, and lower electron boost energies (< or = 9 MeV) were associated with IBTR. Local failure and predominant nuclear Grade 3 were found to have a small (4%-12%) but statistically significantly negative impact on the rates of distant metastasis and CSS. These results suggest that optimizing local therapy (surgery and radiation) is crucial to improve local control and CSS in patients treated with DCIS.