Race as an independent factor for survival in breast cancer patients according to analysis of the National Cancer Database (NCDB).

Abstract

e18155 Background: Black (B) women after early stage of diagnosis have been shown to have double the risk of white (W) women for failing to receive adjuvant chemotherapy. Despite lower incidence of breast cancer in B patients, they are more likely to die of the disease. Worse outcomes in B populations have been related to clinical and socioeconomic factors. However, the determination of improved access and its effects on survival in black patients remains uncertain. Methods: 1042844 patients diagnosed between 2004 and 2014 with stage I-III breast cancer were identified in the NCDB. Only W and B races were analyzed with established risk factors of age, stage, comorbidity score, and insurance status. Data was analyzed using univariable and multivariable logistic and Cox proportional hazards regression models. Odds Ratio (OR) for binary outcome, Hazard Ratio (HR) for time-to-event (survival) outcome along with 95% confidence interval (95% CI) are reported. Results: Among the total population 85.5% were W, 10.6% B, and 3.9% other. B were more likely to be uninsured (OR: 1.66; 95% CI: 1.59 - 1.72; p < 0.0001), or have Medicaid (OR: 2.01; 95% CI: 1.96 - 2.07; p < 0.0001). B were also diagnosed at later stage (stage 3 OR: 1.59; 95% CI: 1.57 - 1.63; p < 0.0001) with higher co-morbidities (OR: 2.49; 95% CI: 2.34 - 2.67; p < 0.0001) consistent with prior studies. B were more likely to experience delayed treatment (OR: 2.15; 95% CI: 2.10 - 2.20; p < 0.0001). B race remained an independent factor associated with higher likelihood of death compared to W patients (HR: 1.32; 95% CI: 1.3 - 1.34; p < 0.0001) in multivariable analysis. Conclusions: This large database study demonstrates that even when controlling for established risk factors such lack of insurance or Medicaid, higher comorbidities, and later stage at diagnosis, B patients were more likely to experience delays in treatment initiation and worse overall survival. This suggests race remains an independent risk factor for poor outcome even when clinical factors are matched. Further analysis including tumor biology should be examined to better understand this persistent disparity.