BACKGROUND: A recent phase III trial showed that the addition of bevacizumab, a monoclonal antibody to vascular endothelial growth factor-A, to first-line irinotecan, 5-fluorouracil, and leucovorin (IFL) prolonged median survival in patients with metastatic colorectal cancer. We carried out a retrospective analysis of patients in the trial to evaluate whether mutation status of k-ras, b-raf, or p53 or P53 expression could predict which patients were more likely to respond to bevacizumab. METHODS: Microdissected tumors from 295 patients (274 primary tumors, 21 metastases) were subject to DNA sequence analysis to identify mutations in k-ras, b-raf, and p53. Nuclear P53 expression was determined by immunohistochemistry. Hazard ratios and 95% confidence intervals (CI) for overall survival were estimated using Cox regression analysis. RESULTS: In all biomarker subgroups, estimated hazard ratios for risk of death were less than 1 for bevacizumab-treated patients as compared with those for placebo-treated patients. Mutations in k-ras and/or b-raf were observed in 88 of 213 patients (41%). Hazard ratios for death among patients with tumors with wild-type k-ras/b-raf status, as compared with those of patients with mutations in one or both genes, were 0.51 (95% CI = 0.28 to 0.95) among those treated with IFL plus bevacizumab and 0.66 (95% CI = 0.37 to 1.18) among those treated with IFL plus placebo. Mutations in p53 were found in 139 of 205 patients (68%), and P53 was overexpressed in 191 of 266 patients (72%); neither p53 mutation nor P53 overexpression was statistically significantly associated with survival. CONCLUSIONS: We did not find a statistically significant relationship between mutations of k-ras, b-raf, or p53 and the increase in median survival associated with the addition of bevacizumab to IFL in metastatic colorectal cancer.