Background: The GERCOR index based on performance status and serum LDH was reported to be useful to predict survival for patients with previously untreated mCRC. However, the validity of the GERCOR index has not been reported in patients treated with bevacizumab (Bev)-based first line chemotherapy. Methods: 115 patients with mCRC treated with Bev contained first line chemotherapy were registered from 15 centers in Japan. Univariate and multivariate analysis for overall survival (OS) were performed using patient characteristics. Survival analyses were performed with the Kaplan-Meier method, log-rank test and the Cox proportional hazards model. The analysis was also designed to determine whether the GERCOR index could be extended to progression-free survival (PFS). Results: All data were available for prognostic categorization in 108 patients. Patients with the GERCOR index of low, intermediate and high risk were 45, 57, and 6, respectively. The pts characteristics between low risk (L) and intermediate/high risk (I/H) were generally balanced except for prior colorectomy (75.6% in L, 54.0% in I/H; p = 0.027), based cytotoxic agent (oxaliplatin) (80.0% in L, 93.7% in I/H; p = 0.039), liver metastasis (53.3% in L, 79.4% in I/H; p = 0.006) and median number of metastatic organ (1 in L, 2 in I/H; p = 0.024). The distribution and median OS / PFS for the GERCOR index were as follows: L (n = 45; 29.9/10.0 months), I/H (n = 63; 17.0/8.5 months). For OS, there was significant difference between L and I/H (p = 0.003). For PFS, there was not significant difference between L and I/H (p = 0.522). In the Cox multivariate analysis, GI did not show an independent prognostic impact (L vs I/H ; HR 1.499, p = 0.120) and predictive impact (L vs I/H ; HR 0.922, p = 0.733). Conclusions: In this analysis, the GERCOR index might be neither the predictive nor prognostic factor in the bevacizumab combined first line chemotherapy for patients with mCRC.