The Glasgow Prognostic Score (GPS), based on C-reactive protein and albumin levels, has shown its prognostic value in metastatic colorectal carcinoma (mCRC) patients receiving conventional cytotoxic therapy. Bevacizumab, a monoclonal antibody to vascular epidermal growth factor, improves the overall survival in mCRC. The aim of the present study was to assess the prognostic value of GPS in mCRC patients receiving antivascular epidermal growth factor therapy. From August 2005 to August 2012, consecutive patients with mCRC who received chemotherapy plus bevacizumab were eligible for the present analysis. The clinical stage, C-reactive protein, albumin and the Eastern Cooperative Oncology Group performance status were recorded at the time of initiation of bevacizumab. Patients received 5-fluorouracil-based chemotherapy plus bevacizumab in accordance with the digestive oncology multidisciplinary staff proposal and in line with the French recommendations for the treatment of mCRC. Eighty patients were eligible (colon n = 59, rectum n = 21), with a median follow-up of 14 months (range 1-58 months). Chemotherapy given with bevacizumab and 5-fluorouracil was oxaliplatin (n = 41, 51%) or irinotecan (n = 27, 34%). At baseline, 56, 31 and 13% of patients had a GPS of 0 (n = 45), 1 (n = 25) and 2 (n = 10), respectively. The median progression-free survival in these groups was 10.1, 6.5 and 5.6 months (P = 0.16), respectively. The median overall survival was 20.1, 11.4 and 6.5 months, respectively (P = 0.004). Our study confirmed the prognostic value of GPS in mCRC patients receiving chemotherapy plus bevacizumab. Given the poor survival observed in patients with an GPS of 2, studies dedicated to these patients could identify optimal treatment modalities.