Abstract A28: DCIS disease-free survival in the population-based Wisconsin In Situ Cohort

Abstract

Abstract Ductal carcinoma in situ (DCIS) now makes up 20% of new breast cancer diagnoses. Women diagnosed with DCIS face an increased risk of developing invasive breast cancer. A number of randomized trials have demonstrated the effectiveness of radiation and tamoxifen in reducing the risk of invasive cancer after a DCIS diagnosis. However, few population-based studies have examined disease-free survival according to treatment in community practice. We examined disease-free survival in the Wisconsin In Situ Cohort (WISC) Study, a longitudinal study of health outcomes among a large population-based cohort including 1959 DCIS cases enrolled between 1995 and 2006. Data on the initial DCIS diagnosis, treatments, second breast cancer events, and prognostic factors were collected from the statewide cancer registry, patient interviews, and pathology reports. Multivariate Cox proportional hazards regression was used to evaluate disease-free survival in relation to treatment while adjusting for prognostic patient and tumor characteristics. After an average of 7.0 years of follow-up, 133 second breast cancer events occurred. The overall five-year disease-free survival rate was 94.6% (95% CI: 93.3–95.6%). Use of mastectomy declined over time, from 46% of DCIS cases in 1995–1997 to 30% of cases in 2004–2006. Tamoxifen use increased from 14% in 1995–1997 to 47% in 2004–2006. Overall five-year disease-free survival rates were 100% for women undergoing a bilateral mastectomy, 94.7% for women undergoing ipsilateral mastectomy, 95.4% for women treated with breast conserving surgery (BCS) and radiation, 89.9% for women treated with BCS only, and 86.5% for women receiving no further treatment beyond biopsy. These differences were driven mainly by variation in rates of ipsilateral events, whereas five-year disease survival for contralateral events was above 96% regardless of treatment type. No statistically significant differences were observed for any type of second event according to tamoxifen use. These results of this study provide population-based data that can be used to guide treatment choice for DCIS. Citation Information: Cancer Prev Res 2011;4(10 Suppl):A28.