Allostatic Load and Racial and Rural Disparities in Breast Cancer Survival.

Importance: Racial and geographic disparities in breast cancer survival persist. Biological consequences of chronic stress, measured by allostatic load (AL), may contribute to these disparities, but their role in breast cancer prognosis is not well understood. Objective: To evaluate the association between AL and overall survival among patients with breast cancer and assess the contribution of AL to racial and rural disparities in survival. Design, Setting, and Participants: This cohort study used data from an institutional electronic health record and cancer registry at the University of Virginia Comprehensive Cancer Center. Participants were women diagnosed with stage I to III breast cancer between 2014 and 2024. Data were analyzed from February 2025 through April 2025. Exposures: Allostatic load was derived from 14 clinical biomarkers and medication history and categorized as low (score ≤3) or high (score >3). Main Outcomes and Measures: The primary outcome was overall survival. Cox proportional hazards models were used to assess the association with overall survival. Blinder-Oaxaca decomposition was performed to assess the extent to which AL explained disparities in survival by race and rurality. Results: Among 3069 patients with stage I to III breast cancer followed up for a median (IQR) of 55.4 (30.1-83.3) months (353 Black [11.5%], 2530 White [82.4%], and 178 other race [5.8%]; 1565 aged ≤66 years [51.0%] and 1504 aged >66 years [49.0%]), most individuals were postmenopausal (1523 patients [49.6%]) and married (1766 patients [57.5%]). There was a higher mean (SD) AL score in patients who were older (4.39 [1.68] for ages >66 years vs 3.84 [1.68] for ages ≤66 years), Black (4.54 [1.81] vs 4.07 [1.68] for White patients), rural residents (4.25 [1.72] vs 4.08 [1.70] for urban residents), uninsured (4.35 [1.55]) or on public insurance (eg, 4.33 [1.93] for Medicaid vs 3.70 [1.56] for private insurance), and unemployed (4.14 [1.77]) or retired (4.39 [1.65] vs 3.63 [1.60] for employed) and who reported no alcohol use (4.33 [1.74] vs 3.94 [1.66] for reported use) or a history of tobacco use (4.20 [1.69] vs 4.05 [1.72] for no reported use). High AL score (>3) was associated with an increased risk of mortality (hazard ratios [HRs] ranged from 1.26; 95% CI, 1.04 to 1.54 for model 5 [adjustment for demographics, social economic status, lifestyle factors, and clinical factors] to 1.53; 95% CI, 1.26 to 1.86 for model 1 [crude model]) compared with low AL score (≤3). Stratified analyses indicated a larger HR in the association between AL score and mortality among rural Black patients (HR per 1-unit increase in AL, 3.33; 95% CI, 1.27-8.77). In decomposition analyses, the total explained portion accounted for -0.0708 of the racial disparity, with AL score accounting for -0.0100 (95% CI, -0.0219 to 0.0018), or 14.2% of the explained portion, while the total explained portion of the geographic disparity was -0.0127, with AL score accounting for -0.0043 (95% CI, -0.0107 to 0.0021), or 34.3% of the explained rural disparity, although estimates were not statistically significant. Conclusions and Relevance: In this study, high AL was independently associated with worse overall survival. The contribution of AL to the observed racial and rural disparities did not reach statistical significance.