LAUSR

6516 Background: Due to the elevated risk of ipsilateral invasive breast cancer (iIBC) after diagnosis with primary ductal carcinoma in situ (DCIS), professional guidelines recommend surveillance screening within 6-12 months (mo) after completion of initial local treatment and annually thereafter. To characterize adherence to these guidelines, we explored longitudinal patterns of utilization and factors associated with the use of surveillance imaging (mammography, MRI, ultrasound) for women with primary DCIS treated with breast conserving surgery (BCS) ± radiotherapy (RT) within 6 mo of diagnosis. Methods: A treatment-stratified random sample of patients diagnosed with screen-detected and biopsy-confirmed DCIS in 2008-15 was selected from 1,330 Commission on Cancer-accredited facilities (up to 20/site) in the US. All imaging exams coded as asymptomatic were collected from 6 mo up to 10 years (yr) post-diagnosis. Time was defined according to 12-mo long surveillance periods. To be included in a given surveillance period, women had to be alive and free of a new breast cancer diagnosis through the end of the period. Women were classified as “consistent” screeners if they had at least one surveillance screen during each period, for the first 5 yr post-treatment or until censoring, whichever occurred first. Repeated measures multivariable logistic regression with generalized estimating equations was used to model receipt of surveillance breast imaging over time. The model included clinical and socioeconomic features. Results: The final analytic cohort contained 12,559 women; 8,989 (71.6%) received RT after BCS. Median age was 60 yr (interquartile range: 52-69) and median follow-up was 5.6 yr (95% confidence interval [CI] 5.6-5.7). Among women who received BCS (instead of BCS+RT), 62.5% (79.7%) underwent surveillance imaging within 6-18 mo after diagnosis. 38.7% (54.0%) were categorized as “consistent” screeners. Compared to white women, Black women were less likely to receive surveillance screening after treatment for primary DCIS (odds ratio [OR] 0.85, 95% CI 0.77-0.94). Hispanic ethnicity had a similar association (OR 0.86, 95% CI 0.74-0.99) compared to non-Hispanic ethnicity. Women with private insurance, compared to government insurance, were more likely to receive screening (OR 1.20, 95% CI 1.11-1.30). Prognostic tumor features indicative of a higher risk of subsequent iIBC, including higher grade, presence of comedonecrosis, and hormone receptor-negative DCIS, were not associated with screening uptake. Conclusions: Despite guidelines recommending annual surveillance imaging, many women with primary DCIS do not undergo regular imaging after BCS. The findings from this US-based study suggest that disparities in screening uptake are associated with race/ethnicity and insurance status rather than prognostic tumor features.