578 Background: Active surveillance is increasingly discussed for low-risk ductal carcinoma in situ (DCIS), yet randomized trials report only short-term outcomes. Long-term survival data stratified by nuclear grade remain limited. We evaluated the association between cancer-directed surgery and long-term breast cancer–specific survival (BCSS) and overall survival (OS) in DCIS. Methods: We conducted a retrospective population-based cohort study using SEER 17 registries, including patients diagnosed with DCIS between 2000 and 2019 with follow-up exceeding 36 months. Exposure was cancer-directed surgery (yes vs no). Outcomes were BCSS, analyzed using cause-specific Cox models, and OS. Prespecified covariates included age, year of diagnosis, race, tumor size, nuclear grade, estrogen and progesterone receptor status, median household income, and residence type. Missing data were handled using multiple imputation. Confounding was addressed using inverse probability of treatment weighting with stabilized weights. Analyses were stratified by nuclear grade (Grades 1–3). Results: Among 51,637 patients, 50,346 (97.5%) underwent surgery and 1,291 (2.5%) did not. Median follow-up was 120 months. In the overall cohort, surgery was associated with improved BCSS (hazard ratio HR 0.13; 95% CI, 0.09–0.19) and OS (HR 0.65; 95% CI, 0.56–0.75). Outcomes differed substantially by nuclear grade. In Grade 1 DCIS, surgery was not associated with improved BCSS (HR 0.57; 95% CI, 0.05–6.16), with 20-year BCSS exceeding 99% regardless of treatment, although OS favored surgery. In Grade 2 DCIS, surgery reduced breast cancer–specific mortality (HR 0.22; 95% CI, 0.10–0.45) without an OS benefit. In Grade 3 DCIS, surgery conferred marked improvements in both BCSS (HR 0.10; 95% CI, 0.06–0.15) and OS (HR 0.49; 95% CI, 0.40–0.60). Conclusions: In this large population-based study with up to 20 years of follow-up, omission of surgery in low-grade DCIS was not associated with worse breast cancer–specific survival, whereas clear survival benefits were observed in intermediate- and high-grade disease. These findings support risk-adapted management strategies and provide long-term survival context for active surveillance approaches in DCIS.
Aydogan et al. (Wed,) studied this question.