Abstract A86: Investigation into racial/gender disparities in postsurgery survival of colorectal cancer: A retrospective study from a community hospital in New York City

Abstract

Background: Racial disparities and gender differences in the prevalence and mortality of colorectal cancer have been reported. African Americans (AA) have the highest incidence and worst prognosis of colorectal cancer amongst ethnic groups. Men also have a younger age at diagnosis than women along with a poorer prognosis. It is uncertain whether differences persist when patients are of similar biological status at diagnosis, and they have equal opportunity to receive surgery. The aim of this study is to investigate post-surgery survival between AAs and Whites as well as between men and women in the setting of a community hospital in which a majority of the population is AA. Methods: This is a retrospective analysis of data from The Brooklyn Hospital Center's cancer registry from 1997 to 2010. Of all of 1068 registered patients, 806 (75.46%) are AA, 243 (22.75%) are White and 19 (1.8%) are Asian. 445 are male and 623 are female. Among all of the subjects, 817 (76.5%) underwent surgery. Among the 817 surgical patients, 619 are African American, 174 are White, and 34 are Asian (not included in the data analysis). 330 are men, and 403 are women (exclude Asian). Statistical analyses were done by SPSS and Epi info software. Chi square test, Fisher's exact test, and one-way ANOVA analysis are used for the baseline characteristics. Kaplan–Meier survival probabilities are calculated and multivariate Cox proportional hazards models were applied to estimate hazard ratios (HR) with 95% confidence intervals (95% CI). Results: The average age at diagnosis is 67.62±13.23 in AAs and 69.54±14.25 in Whites with no significant difference. Average age at diagnosis in men is 65.81±11.78 and in women is 68.09±13.5 (P=0.014). There is significant difference in location of the presenting tumor; AAs right sided colon cancer in AAs is 40.87% as compared with Whites in whom right sided colon cancer occurs 30.46% of the time. 50.57% of Whites have colorectal cancer located in rectosigmoid region vs. only 38.93% of AAs (p=0.002). There was no significant difference in tumor location between men and women. No significant difference existed in the distribution of histological grade (p=0.19), pathologic stages (p=0.82), people who received surgery, and percentage of patients who received chemotherapy between ethnic groups as well as between gender groups. Hemicolectomy is the most common procedure performed. Significant difference for the distribution of the surgery types was present with 51.4% of AAs receiving hemicolectomy vs. 39.1% of Whites (p=0.022). Also, a significant difference in surgery type was measured between genders with 47% of men receiving hemicolectomy versus 49.9% of women (p=0.044). Kaplan Meier survival analysis: no significant differences presented between these two ethnic groups (Log-rank p=0.2958) with regards to the 5 years survival probability; 5 year survival in AAs is 61.07% and in Whites is 57.36 %. There was a significant difference between men and women in 5 year survival with Log rank p=0.035 and a 5-year survival probability for men of 54.38% vs. 64.24% for women with Cox hazard ratio 1.248. Conclusions: Much attention has been paid to racial disparities and gender differences for the past two decades. Our study shows that AAs are younger at diagnosis and are more likely to present with proximal tumors than Whites, but there is no significant difference in survival between these two ethnic groups post-surgery. Based on similar biological background and with equal opportunity to receive surgery, the racial disparity is diminished perhaps even ameliorated. However, a gender difference is present in our data analysis even when taking into account no difference in the biological characteristics. Men have a younger age at diagnosis and a poorer prognosis. As some studies state, that gender difference may be attributed to hormone levels. Further study is needed to uncover the underlying mechanisms. Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):A86.