KRAS (Kirsten rat sarcoma 2 viral oncogene homolog) is a major predictive marker for anti-epidermal growth factor receptor treatment, and determination of KRAS mutational status is crucial for successful management of colorectal adenocarcinoma. More standardized and accurate methods for testing KRAS mutation, which is vital for therapeutic decision-making, are required. Digital droplet polymerase chain reaction (ddPCR) is an advanced digital PCR technology developed to provide absolute quantitation of target DNA. In this study, we validated the clinical performance of ddPCR in determination of KRAS mutational status, and compared ddPCR results with those obtained by Sanger sequencing and peptide nucleic acid-clamping. Of 81 colorectal adenocarcinoma tissue samples, three repeated sets of KRASG12/G13 mutation were measured by ddPCR, yielding high consistency (ICC = 0.956). Receiver operating characteristic (ROC) curves were constructed to determine KRASG12/G13 mutational status based on mutant allele frequency generated by ddPCR. Using the best threshold cutoff (mutant allele frequency of 7.9%), ddPCR had superior diagnostic sensitivity (100%) and specificity (100%) relative to the two other techniques. Thus, ddPCR is effective for detecting the KRASG12/G13 mutation in colorectal adenocarcinoma tissue samples. By allowing definition of the optimal cutoff, ddPCR represents a potentially useful diagnostic tool that could improve diagnostic sensitivity and specificity.
Keywords: PNA-clamping assay; ROC; Sanger sequencing; colorectal cancer; cutoff; ddPCR.