High sensitivity of both sequencing and real-time PCR analysis of KRAS mutations in colorectal cancer tissue.

Abstract

The KRAS mutation status predicts the outcome of treatment with epidermal growth factor receptor targeted agents, and therefore the testing for KRAS mutations has become an important diagnostic procedure. To optimize the quality of this test, we compared the results of the two most commonly used KRAS mutation tests, cycle sequencing and a real-time PCR-based assay, in DNA extracted from formalin-fixed paraffin-embedded (FFPE) colorectal cancer samples of 511 patients. The results were interpreted in the context of the tumour cell percentage and the assay parameters. In 510 samples KRAS mutation status assessment was successful. A KRAS mutation was detected in 201 tumours (39.4%). Sequencing and the real-time PCR-based assay generated the same result in 486 samples (95.3%). The sequencing result was considered false positive in one (0.2%) and false negative in nine samples (1.8%). The assay result was considered false positive in six (1.2%) and false negative in seven samples (1.4%). Explanations for discrepant test results were a higher sensitivity of the assay in samples with a low tumour cell percentage, occurrence of mutations that are not covered by the assay and delta Ct values approximating the cut-off value of the assay. In conclusion, both sequencing and the real-time PCR-based assay are reliable tests for KRAS mutation analysis in FFPE colorectal cancer samples, with a sensitivity of 95.5% (95% confidence interval [CI] 91.7-97.9%) and 96.5% (95% CI 93.0-98.6%), respectively. The real-time PCR based assay is the method of choice in samples with a tumour cell percentage below 30%.