P-283 Analysis of KRAS mutation allele in circulating cell-free DNA derived from urine in colorectal cancer patients

Abstract

Liquid biopsy is widely introduced in various malignant diseases and it provides crucial clinical information on genetic profiles. Cell-free DNA (cfDNA) derived from peripheral blood has been studied mainly. However, other body fluids also have potential to provide molecular information. Urine could be a novel and completely noninvasive source to provide genetic information. The aim of this study was to evaluate the quantity of cfDNA derived from urine in colorectal cancer patients. The accuracy of KRAS mutation allele detection was also analyzed using droplet digital PCR (ddPCR). Urine, peripheral blood and tissue samples were collected from consecutively resected colorectal cancer patients. DNA was extracted from each sample and the quantity was measured. From each DNA sample, ddPCR was performed to detect common point mutations in KRAS oncogene. A total of 143 patients were enrolled. In all patients, plasma and urine cfDNA were successfully extracted and the quality and quantity were appropriate for genetic analysis. The concentration of urine cfDNA was significantly higher than that of plasma (2448 ng/ml vs. 267 ng/ml, p< 0.05). Neither plasma nor urine cfDNA concentration correlated with cancer staging. In tumor tissue DNA, KRAS mutation allele was detected in 95 patients (66.4%), KRAS mutation was detected in 21 patients from plasma and 13 patients from urine. The concordant rate of KRAS mutation allele detection between tumor tissue DNA and plasma cfDNA was 22.1%, and between tumor tissue DNA and urine cfDNA was 13%. In the early stage patients (stage I and II), the detection rate of KRAS mutation allele was similar in plasma and urine cfDNA (15.6% vs. 15.6%), however, in advanced-stage patients (stage III and IV), the detection rate in urine cfDNA was lower than that in plasma DNA (15.9% vs. 25.4%). Furthermore, urine cfDNA had a lower detection rate of KRASmutation than plasma cfDNA in stage I and IV patients (stage I; plasma cfDNA 14%, urine cfDNA 0%, stage IV; plasma cfDNA 52%, urine cfDNA 12%), however, the detection rates were equivalent in stage II and III patients (stage II; plasma cfDNA 16%, urine cfDNA 20%, stage III; plasma cfDNA 16%, urine cfDNA 16%). Our data indicated that urine cfDNA is a useful source in complementing genetic analysis in colorectal cancer patients. The sensitivity seems to be inferior especially in advanced stage, however, it will be considered useful for evaluation of minimal residual disease in stage II and III colorectal cancer patients.