Determinants and disparities of survival in triple-negative breast cancer patients: A population-based retrospective longitudinal cohort design utilizing the Cox proportional hazard analytical model
A significant racial disparity in breast cancer mortality exists among women in the United States. Triple-negative breast cancer (TNBC) is a breast cancer phenotype that may explain, in part, this disparity between white and African American women. The objective of this study was to determine the predictors of survival in TNBC and non-triple-negative breast cancer (NTNBC) patients. Data on 168,756 female patients with a diagnosis of invasive breast cancer in the Surveillance Epidemiology and End Results (SEER) program were stratified based on breast cancer receptor phenotypes in this retrospective longitudinal cohort study design. Multiple logistic regressions were used for exploring predictors of treatment which showed that not receiving surgery as standard treatment was associated (odds ratio: 95% CI) with TNBC (OR 1.151: 1.042, 1.177), uninsured (OR 3.552: 3.206, 3.937) and African American (OR 1.804: 1.702, 1.912) while not receiving radiation was associated with TNBC (OR 1.151: 1.113, 1.190), uninsured (OR 1.318; 1.217, 1.429). Cox’s hazard models were used, regressing age, race, ethnicity, marital status, health insurance status, histological tumor grade, and treatment status on survival time, the outcome measure. Analysis revealed that the mean survival time is lower for TNBC [15.60 (± 10.29)] months compared with NTNBC [16.01 (± 10.18)] (p < 0.0001), a difference though small is statistically significant. The independent determinants of survival in TNBC were: young age at diagnosis [(β = 0.033, HR 1.033 (1.026, 1.041)]; being African American [(β = 0.182, HR 1.200 (1.117, 1.289)], being married [(β = - 0.362, HR 0.697 (0.658, 0.737)]; higher tumor histological grades [β = 1.034, HR 2.812 (2.159,3.661)]; uninsured [(β = 0.541, HR 1.717 (1.481, 1.992)]; no surgery [(β = 2.156, HR 8.633 (8.152, 9.143)], or no radiation treatment [(β = 0.489, HR 1.630 (1.535,1.73)]. African American race, uninsured status, higher grade at diagnosis, inadequate treatment are independent predictors of poor survival among breast cancer patients; importantly, TNBC had a lower survival than that of NTNBC patients. A higher proportion of TNBC patients had a diagnosis at younger age, with higher tumor grade and was of the African American race. The survival disparity in African American patients may be partially explained by disproportionately higher TNBC cases among them, as well as, rates of not receiving standard treatments.
Belcon, Michael C, "Determinants and disparities of survival in triple-negative breast cancer patients: A population-based retrospective longitudinal cohort design utilizing the Cox proportional hazard analytical model" (2015). ProQuest ETD Collection for FIU. AAI10165725.