Association Between Race/Ethnicity and Survival in Women With Advanced Ovarian Cancer.

Introduction Ovarian cancer is the fifth-leading cause of cancer-related mortality in US women. There are survival disparities between non-Hispanic black (NHB) and non-Hispanic white (NHW) women. We assessed if insurance status or extent of disease modified the effect of race/ethnicity on survival for ovarian cancer. Methods A historical cohort was assembled using the 2007-2015 National Cancer Institute's Surveillance, Epidemiology, and End Result (SEER) dataset. Adult NHB and NHW (>18 years) diagnosed with regional and distant ovarian cancer were included. The outcome was five-year cause-specific mortality. Multivariable Cox regression models were fitted, including race by the extent of disease and race by insurance status interaction terms. Results For each significant interaction, separate Cox models were fitted. In total 8,043 women were included. The insurance status/race interaction was not statistically significant, but the extent of disease modified the effect of race on survival. NHB survival was lower in regional disease (adjusted hazard ratio (HR) =1.6; 95% confidence interval (CI) 1.1-2.4), while there was no difference in survival between women with distant disease (adjusted HR =1.0; 95%CI 0.9-1.2). Conclusions Ovarian cancer mortality is similar between NHB and NHW women with the distant disease but higher in NHB women with regional disease. Further research should clarify whether this difference is due to access to quality cancer treatment or other factors affecting treatment response.