Background: Studies suggest women living with HIV (WHIV) may have lower breast cancer incidence compared to the general population. To minimize the effect of potential differences in the prevalence of risk factors for breast cancer incidence and detection between WHIV and women in the general population, we used data from Medicaid, the predominant provider for WHIV, to compare breast cancer incidence by HIV status. Methods: We used administrative claims from women aged 18-64 years who were enrolled in Medicaid (2001-2009) in 14 states (AL, CA, CO, GA, FL, IL, MA, MD, NC, NY, OH, PA, TX, WA), and had ≥7 months of continuous enrollment in their first enrollment period. We excluded women who were dual Medicaid/Medicare enrollees or had evidence of a current or past breast cancer diagnosis during the first 6 months of enrollment. We identified all incident breast cancers that occurred during or after the 7th month of enrollment using the CMS Chronic Conditions Warehouse (CMS-CCW) definition. HIV status was defined using a modified CMS-CCW definition of two HIV claims within one year rather than two.

Women meeting the definition for HIV before breast cancer diagnosis were considered HIV positive from enrollment. We estimated adjusted cumulative incidence curves for breast cancer over age and by HIV status to allow non- proportional hazards. To account for differences in women by HIV status, we weighted curves by the inverse probability of HIV status, predicting HIV status with race, Charlson comorbidity index, state, age at and year of enrollment, census tract level median income and proportion rural. From the adjusted survival curves, we calculated the difference in the restricted mean survival time (RMST) for breast cancer incidence by HIV status and bootstrapped 95% confidence intervals. RMST estimates the area under the curve and is interpreted as the average time to breast cancer over a given time interval. Analyses were stratified by age, presuming women ≥55 as post-menopausal, and evaluated separately by race/ethnicity (Black, Latina, White). Results: There were 68,179 women, 341 breast cancers, and 280,420 person- years (PY) among WHIV and 11,920,902 women, 24,791 breast cancers, and 25,458,327 PY among women without HIV. Compared to women without HIV, WHIV were more likely to be Black, older, enrolled longer, and have more comorbidities.

For women <55, there was no significant difference in time to breast cancer by HIV status (RMST difference for WHIV compared to women without HIV -0.60 months; 95% CI: -1.68, 0.49). Results were similar by race. For women ≥55, WHIV had on average 1.08 (95% CI: 0.79, 1.37) greater months to breast cancer compared to women without HIV. Results were similar by race except among Black women where the RMST difference was attenuated and non-significant. Conclusion: In a large sample of Medicaid enrollees (2001 to 2009), time to incident breast cancer did not differ by HIV status among women <55. Among women ≥55, WHIV had a slightly longer expected time to incident breast cancer than women without HIV.

Citation Format: Maneet Kaur, Keri Calkins, Bryan Lau, Corinne E. Joshu. Breast cancer incidence by HIV status and race among women enrolled in Medicaid, 2001 to 2009 [abstract]. In: Proceedings of the AACR Virtual Conference: Thirteenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2020 Oct 2-4. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(12 Suppl):Abstract nr PO-179.