Abstract B70: Race, age, and advanced stage colorectal cancer survival

Abstract

Abstract African Americans (AA) have worse survival from colorectal cancer (CRC) compared to European Americans (EA), but the reasons for the racial differences in stage-specific survival are poorly understood. To address this gap, we analyzed population-based data from the South Carolina Central Cancer Registry (SCCCR) to examine racial differences in advanced stage survival before and after the introduction of the new chemotherapeutic regimens in 2004 and to integrate information on race, age, and pathologic prognostic tumor characteristics and survival. The study focused on metastatic disease because of similar recommended standard therapy, the large racial difference in the relative survival, and a documented change in chemotherapy usage in 2004. The study population was comprised of data from patients with newly diagnosed advanced stage CRC (SEER staging distant) from 1/1/96 through 12/31/06 with follow-up data until 12/31/07. Analyses were designed to compare AA to EA by computing median survival and hazard ratios (HR) and 95% CIs to model the hazard of death as a function of race, controlling for age, sex, year of diagnosis, and first-line chemotherapy treatment. We assessed interactions between race and age, gender, colorectal location, histologic type, and grade on overall survival. Data from 3463 (1057 AA and 2406 EA) patients with advanced stage colorectal adenocarcinoma were evaluated. 732 AA and 1707 EA were diagnosed before 2004. Compared to EA, AA were more likely to be younger, female, have proximal cancers, have lower tumor grades, and were less likely to report first line chemotherapy usage. Histologic type or year of diagnosis did not significantly differ by race. Overall survival among AA and EA differed by age, with the racial difference in survival most pronounced among those <50 years of age. Median survival under 50 years of age for EA and AA was 17 months (95% CI 16–19 months) and 12 months (95% CI 10–14 months), respectively; among those 50 years and older, median survival for EA and AA was 12 months (95% CI 11–13 months) and 11 months (95% CI 10–12 months), respectively. In Cox proportional hazard analyses, a significant interaction was observed between race and age on survival. Overall, younger AA compared to younger EA had a 1.32 (95% CI 1.09–1.60) times higher risk of death; post 2003, the racial disparity in the risk of death among the young was even higher (HR 1.73 (95% CI 1.14–2.62)). Older AA compared to older EA had similar risks of death overall (HR 1.05 95% CI 0.97–1.14) and post 2003 (HR 1.14 95% CI 0.97–1.34). Among the younger, proximal tumor location in AA compared to EA was associated with a HR of 1.42 (95% CI 1.01–2.01) and HR 1.24 (95% CI 0.96–1.61) for non-proximal tumors. In this same age comparison, post 2003, the associations for proximal cancers were stronger ((HR 2.31 (95% CI 1.01–5.28) compared to non-proximal, HR 1.48 (95% CI 0.82–2.68)). Higher tumor grade was also associated with poorer survival among younger AA compared to EA (HR 1.49 (95% CI 1.00–2.23)). In contrast to the younger patients, clinicopathologic features were not strongly associated with survival among older patients. Younger AA had poorer overall survival than younger EA, a disparity not present in the older age group. The survival disparity worsened since the advent of the new chemotherapies introduced in 2004. The results also indicated that proximal location and higher tumor grade were associated with poorer survival among the young AA compared to younger EA. These findings uncovered some clear patterns underlying the racial disparity in advanced stage CRC, providing direction for future and more focused studies. Citation Information: Cancer Epidemiol Biomarkers Prev 2011;20(10 Suppl):B70.