Abstract
PURPOSE: Treatment advances have improved outcomes in clinical trials of patients with metastatic colorectal cancer (mCRC). Less is known about these effects for patients in real-world settings. This study evaluated treatment patterns and survival in older, demographically diverse patients with mCRC. METHODS: A retrospective cohort analysis was performed for 4,250 patients from January 1, 2000 to December 31, 2007 using linked Surveillance, Epidemiology, and End Results-Medicare database. Patients were ≥ 66 years, enrolled in Medicare parts A and B, and received first-line treatment with fluorouracil and leucovorin (5-FU/LV), capecitabine (CAP), 5-FU/LV plus oxaliplatin (FOLFOX), or CAP and oxaliplatin (CAPOX). Cox regression with backward elimination and propensity score-weighted Cox regression estimated relative risk of death. Date of last follow-up was December 2009. Statistical comparisons were made between 5-FU/LV vs. CAP and FOLFOX vs. CAPOX. RESULTS: Compared to 5-FU/LV, patients treated with CAP were older (mean age 78 vs. 76; P<0.0001) and more likely female (61 vs. 54 %; P=0.0017), while patients receiving CAPOX and FOLFOX were similar in age (mean age 74 vs. 73; P=0.0924). Complications requiring medical resource utilization following initiation of therapy were significantly higher among patients administered with 5-FU/LV (54 %) vs. CAP (17 %; P<0.0001) and FOLFOX (75 %) vs. CAPOX (57 %; P<0.0001). The multivariate analysis revealed no significant differences in survival between 5-FU/LV and CAP and between FOLFOX and CAPOX. CONCLUSIONS: Overall survival was comparable between CAP and 5-FU/LV and between CAPOX and FOLFOX with fewer complications requiring medical resource utilization associated with CAP and CAPOX, thus confirming clinical trial results.