OBJECTIVETo detect K-ras gene mutations in plasma free DNA by peptide nucleic acid clamp PCR assay (PNA-PCR) and nested primer PCR, and to analyze the correlation between K-ras mutations and prognosis in patients with metastatic colorectal cancer (mCRC).

METHODSPeripheral blood was collected and free DNA was extracted from plasma in 106 patients with mCRC. Nested primer PCR and PNA-PCR were used to detect K-ras gene mutation in the plasma free DNA. The patients were divided into three groups by K-ras status: wild-type group (wild-type determined by both methods), low mutation group (mutation by PNA-PCR method, wild-type by nested primer PCR method) and high mutation group (mutation by two methods). The correlation between K-ras mutations and prognosis was analyzed.

RESULTSThe mutation rate of K-ras in tumor tissues of the 106 patients was 40.6%. The Mutation rate of K-ras in plasma free DNA detected by PNA-PCR was 31.1%, significantly higher than that of 15.1% detected by nested primer PCR (P = 0.006). The consistent rate of the K-ras status in plasma free DNA detected by PNA-PCR and that in tumor tissue detected by traditional method was up to 83.0%. The median overall survival (OS) of patients of the wild type, low mutation and high mutation groups was 23.5 months, 17.3 months and 13.9 months, respectively (P = 0.002). The median progression-free survival (PFS) of the K-ras wild-type, low mutation and high mutation groups with first-line chemotherapy was 6.8 months, 6.1 months and 3.2 months, respectively (P = 0.002), and the median OS of them were 23.0 months, 15.5 months and 13.9 months, respectively (P = 0.036). The overall response rate (ORR) was improved in the K-ras wide-type patients who received cetuximab combined with chemotherapy as first-line therapy (75.0% vs. 23.4%, P = 0.058). Cetuximab combined with in second-line therapy chemotherapy led to a significant improvement in disease control rate (DCR) ( 100% vs. 35.7%, P < 0.001) as compared with those of chemotherapy alone. COX regression model showed that K-ras status detected by PNA-PCR, ECOG PS, number of surgery and initially metastatic site were independent factors for prognosis.

CONCLUSIONSPNA-PCR for the detection of K-ras mutation in plasma free DNA can be used to substitute the traditional method for detection of K-ras mutation in tumor tissues. The abundance of K-ras mutation in plasma free DNA is an independent prognostic factor for patients with metastatic colorectal cancer.