KRAS mutations are proved as a predictor of response to EGFR-targeted therapies for patients with metastatic colorectal cancer. For identifying the wild-type KRAS (wt-KRAS) responder subset of patients who will benefit from novel agents our laboratory has introduced the TheraSreen K-RAS Mutation Kit(R) an allele-specific RT-PCR based assay. Our aim is to describe the validation procedure of this method in our laboratory, determine the portion of colorectal cancer patients with wt-KRAS status, and assess the prognostic power of mutational status for the anti-EGFR therapy outcome in colorectal cancer patients. In this study 302 samples from 273 patients with metastatic colorectal cancer were tested for 7 most common mutations on codon 12 and 13 of the KRAS gene. We used HT-29 and CCL-247 cell lines to determine the sensitivity of the method for different proportions of tumor cells in the sample. We determined that 2% of cells carrying a KRAS mutation must be present in the sample for an undisputable detection of mutated signal using the LightCycler Adapt Software. Among the tested patients 54.5% had a wt-KRAS genotype and 45.5% had a mutated KRAS genotype. The p.Gly12Asp was the most common detected mutation (38.5%). Among the cetuximab therapy responders, 85.7% had a wt-KRAS genotype. We have shown that the RT-PCR method introduced to discriminate between anti-EGFR therapy responders and non-responders is efficient, reliable and quickly applicable. The ratio of mutated versus wt-KRAS patients in our study is similar to ratios reported by other authors, as is the high correlation between wt-KRAS genotype and response to cetuximab therapy. Nevertheless the selection of patients for treatment solely on the basis of KRAS status is not perfect due to the fact that some responders are among the patients with mutated KRAS and some non-responders among the wt-KRAS patients.