Epidemiologic studies have evaluated the association between KRAS mutations and resistance to the treatment of epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitors (TKIs) in patients with non-small cell lung cancer (NSCLC). However, results were inconclusive. To derive a more precise estimation of the relationship, we performed this meta-analysis. Systematic computerized searches of the PubMed and Medline databases (up to Jun 30, 2009) were performed. A total of 22 studies were included in the final meta-analysis, consisting of 1470 NSCLC patients, of whom 231 had KRAS mutations (16%). Current or former smokers had a higher frequency of KRAS mutations than never smokers (25% versus 6%; OR = 4.36; P< 0.01). Mutations were more common among adenocarcinoma than other histologies (26% versus 16%; OR = 1.98; P< 0.01). The objective response rate (ORR) of NSCLC patients with mutant KRAS was 3% (6/210), whereas the ORR of NSCLC patients with wild-type KRAS was 26% (287/1125). The overall pooled RR for ORR was 0.29 (95% CI: 0.18-0.47; P< 0.01). Subgroup analyses were conducted on the basis of ethnicity and study treatment, all the results were not materially altered and did not draw different conclusions, indicating that our results were robust. In summary, this meta-analysis suggests that KRAS mutations may represent negative predictive biomarkers for tumor response in NSCLC patients treated with EGFR-TKIs. However, due to a mutually exclusive relationship between KRAS and EGFR mutation and no difference in survival between KRAS mutant/. EGFR wild-type and KRAS wild-type/. EGFR wild-type NSCLC, the clinical usefulness of KRAS mutation as a selection marker for EGFR-TKIs sensitivity in NSCLC is limited. © 2009 Elsevier Ireland Ltd.