Abstract
Background: Breast cancer is the most common cancer in women. The differences in clinical outcomes are often associated with racial/ethnic background. Although the prognostic factors are associated with cancer outcome, it may not be obvious whether a factor truly makes the same contribution to outcome in racial/ethnic groups. There is currently no study for integrating and comparing combinations of tumor size, lymph node status, histological grade, ER status and age in Black and White women. Objective: To conduct comparative survival analyses for breast cancer according to similar combinations of prognostic factors in black and white women.
Methods: Breast cancer data were obtained from SEER Cancer Registry (1990–2000). An algorithm was created with the statistical programming language “R”. Disease specific survival rates were calculated by the Kaplan-Meier method. Patients who were lost to follow-up or died from causes other than breast cancer were censored by the algorithm. Survival rates were compared by the log-rank test, and the level of significance set at p < 0.05.
Results: Without including histologic grade and ER status, the effect of T and N alone on survival was evident, the survival rates decrease as increase in T and N among both black and white women. After 10 years, the proportion surviving among white women with grade 1 was 95%, whereas for grade 2 this decreased to 87%, and for grade 3 was 79%. For black women, the proportion surviving with grade 1 was 92%, for grade 2 was 74%, and for grade3 was 66%. When comparing increased grade with ER+ status, there was a decrease in survival among both black and white; however, the decrease in survival were more pronounced when considering increased grade with ER- status, with lowest survival overall evidenced in G3, ER- status. The survival rates of white women were always more favorable than blacks for any combinations of the five prognostic factors. The case fatality rate was always less favorable for black women than for whites when similar combinations of prognostic factors were compared.
Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):A87.