The significance of KRAS in advanced colorectal cancer (CRC) treated with bevacizumab (B) is not well understood. We conducted a systematic review and pooled analysis of published trials with the aim to assess the predictive and prognostic role of KRAS status in patients treated with B. We performed a systematic search of PubMed, EMBASE, Web of Science, and Cochrane Register of Controlled Trials. The primary endpoints included objective response rate (RR), progression-free survival (PFS), and overall survival (OS). The odds ratio (OR) for RR and hazard ratios (HRs) were calculated or extracted by published data either using a fixed effect model or a random effect model. A total of 12 studies were included. A total of 2,266 patients were analysed (54 % were KRAS wt). The pooled RRs for KRAS wild-type (wt) versus mutated (mut) patients were 54.8 and 48.3 %, respectively (OR 1.42, P = 0.02). Median PFS was significantly longer in KRAS wt patients compared with that in KRAS mut patients (HR = 0.85; 95 % confidence interval (CI) 0.74-0.98; P = 0.02). Similarly, median OS was significantly better in wt KRAS patients compared with that in mut KRAS patients (HR = 0.65; 95 % CI 0.46-0.92; P = 0.01). This pooled analysis of 12 published studies shows that KRAS wt status is a good prognostic factor for B-based chemotherapy. Also, KRAS wt CRC is associated with a better RR with B plus chemotherapy than mut counterpart.