Abstract B012: Sociodemographic and clinical determinants associated with early-onset breast cancer

Abstract Purpose Early-onset breast cancer (EOBC) diagnosed before age 50 is increasing globally and comprises approximately 15% of all breast cancer (BC) cases in the United States. In Botswana, almost 50% of patients present with EOBC, but risk factors are poorly understood. We aimed to identify sociodemographic and clinical factors associated with EOBC in Botswana. Methods We analyzed a prospective cohort of newly diagnosed BC patients at Princess Marina Hospital between 2015 and 2023. Logistic regression examined the association of sociodemographic and clinical factors, including BC molecular subtypes (i.e, ER+/Her2-, ER+/Her2+, ER-/Her2+, triple negative BC), with EOBC. Factors that had a crude association of p<0.20 were retained in the multivariable model. EOBC was compared to BC patients diagnosed at ≥50 years. A secondary exploratory analysis of BC diagnosed before age 40 was also conducted. Analysis done using STATA 19.5. Results There were 641 BC patients included in the analyses (median age 51.0 (IQR 42.0, 62.9)) with 299 ( 46.7%) EOBC, and 117 (18.3%) diagnosed before 40 years. Of the whole cohort, 31.8% had HIV, 55.9% were from socioeconomically disadvantaged districts (SDD), 44.5% were ever married, 72.9% diagnosed at stage III-IV, and 22.8% were triple negative BC. In the univariate analysis, people with HIV had higher odds of EOBC (OR: 2.32; 95% CI: 1.65-3.26), while living in a SDD (OR: 0.59; 95% CI: 0.43-0.80) and ever being married (OR: 0.33; 95% CI: 0.23-0.46) were associated with lower odds of EOBC. There were no significant associations between EOBC and smoking history, alcohol use, family history of breast cancer, stage at diagnosis, or BC subtypes. In the multivariable model, adjusting for HIV, SDD, marital status, and alcohol, HIV remained significantly associated with higher odds of EOBC (aOR:1.86; 95% CI:1.30-2.66), while living in an SDD (aOR: 0.64; 95% CI: 0.46-0.89) and ever being married (aOR: 0.38; 95% CI: 0.27-0.54) showed reduced odds of EOBC. In the exploratory analysis of BC diagnosed before 40, alcohol use was associated with a greater odds of EOBC (aOR: 2.52, 95%CI: 1.26-5.05), and ever being married was associated with lower odds (aOR: 0.16; 95% CI: 0.09-0.28). HIV status was not associated with diagnosis before age 40 (aOR: 0.78; 95%CI: 0.48-1.26). Conclusion Our analysis found that women with HIV had a higher odds of EOBC, aligning with reports of younger age at diagnosis in this group. Lower odds were observed among married patients, possibly reflecting social support, and among those in SDDs, which may relate to lifestyle or environmental exposures. Alcohol use showed increased odds with a diagnosis before age 40, but not before age 50. This finding should be explored further. Future studies incorporating detailed behavioral, lifestyle, environmental, and clinical data in high-burden populations may identify targetable risk factors and guide precision BC control strategies relevant to efforts to reduce early detection disparities among racial/ethnic minorities, rural populations, and low-income women. Citation Format: Tara M. Friebel-Klingner, Tlotlo Ralefala, Lebogang Laletsang-Mokokwe, Babe Gaoleabale, Nkhabe Chinyepi, Peter Vuylsteke, Dipho I. Setlhako, Scott Dryden-Peterson, Mosepele Mosepele, Robert Gross, Yehoda M. Martei. Sociodemographic and clinical determinants associated with early-onset breast cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: The Rise in Early-Onset Cancers—Knowledge Gaps and Research Opportunities; 2025 Dec 10-13; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(23_Suppl):Abstract nr B012.