Session Item

To evaluate 15-year outcomes for patients with ductal carcinoma in situ (DCIS) of the breast, receiving mastectomy, breast conserving surgery (BCS) alone or with whole breast radiotherapy (BCS+RT). Previously described patient prognostic score and tumour resection margins were also assessed.

The study cohort consisted of all eligible patients with DCIS treated between years 1994 and 2011. Data on type of surgery, tumour resection margins and RT were retrieved retrospectively using medical records. For each patient previously described prognostic score was calculated, considering patient age (>60 years = 0, 40–60 years = 1, <40 years = 2), tumour size (<16 mm = 1, 16–40 mm = 1, >40 mm = 2) and tumour grade (G1 = 0, G2 = 1, G3 = 2). The Kaplan-Meier method was used to calculate estimated survival curves for locoregional recurrence rates (LRR) and invasive relapse-free survival (iRFS) and compared by the log-rank test. Cox’s proportional hazards model was used to assess the effects of covariates on survival.

Of the 883 patients included in the study, 311 (35.2%) received mastectomy, 265 (30%) received BCS+RT and 307 (34.8%) received BCS. The median age was 56 years old (range, 18–87). Prognostic score calculation was available for 859 patients as follows: 289 (33.6%) scored <2 points and 570 (64.6%) scored ≥2 points. Resection margin status was as follows: positive (R1) in 36 (4.1%), close (<2 mm) in 144 (16.3%), ≥2 mm in 657 (74.1%), and unknown margin status in 46 (5.2%) patients. At the median follow-up of 14.8 years (range of 0.1–27), 37 (4.2%), 45 (5.1%), 5 (0.5%), and 18 (2%) patients experienced non-invasive ipsilateral breast tumor recurrence (IBTR), invasive IBTR, regional recurrence and distant relapse, respectively, and 151 (17.1%) patients died. The 15-year LRR were 3.6%, 10.6% and 15% for the mastectomy, BCS+RT and BCS groups, respectively. The corresponding 15-year iRFS rates were 94%, 94% and 77% (p<0.001), respectively (Figure 1).

Compared to mastectomy and BCS+RT groups, BCS group had more invasive IBTR (2.3% and 2.6% versus 10.1%; p<0.001). There were less non-invasive IBTR in the mastectomy group compared to BCS+RT or BCS groups (0.6% versus 7.5% and 4.9%; p>0.001). Compared to positive resection margins (R1), negative resection margins (either <2 mm or ≥2 mm), were associated with improved iRFS (Hazard Ratio (HR) 0.75, 95% confidence intervals (CI) 0.61–0.92; p=0.010). Among patients treated with BCS (n=572), not-receiving postoperative RT (HR 4.06, 95 % CI 1.83–9.04; p<0.001) and a higher patient prognostic score (≥2 points; HR 2.33, 95% CI 1.15–4.72; p=0.018) were both associated with worse iRFS (Figure 2).

Our study demonstrated equivalent invasive IBTR and iFRS rates in patients treated with either mastectomy or BCS+RT.  Not-receiving RT following BCS was associated with worse outcomes regardless of patient prognostic score.