Clinical Study On Combined Detection Of Plasma KRAS, NRAS, BRAF And PIK3CA Genes In Patients With Colorectal Cancer

Objective:With the advent of anti-epidermal growth factor receptor-targeted therapies in recent years,the overall survival of patients with colorectal cancer(CRC)has been significantly extended.For most patients with advanced metastasis or recurrence,tumor tissue specimens cannot be obtained for genetic testing to guide targeted therapy.In this study,by detecting mutations in KRAS,NRAS,BRAF,and PIK3 CA genes in tumor tissues and corresponding plasma of patients with CRC,analyzing the consistency of detecting KRAS,NRAS,BRAF,PIK3 CA gene mutations in plasma and tumor tissues,and exploring the feasibility of plasma instead of tumor tissues to detect genetic mutations..Methods:A total of 175 patients with stage Ⅰ-Ⅳ CRC were enrolled.The patients’ plasma was collected before surgery,and at the same time,tumor tissues embedded in formaldehyde and fixed with paraffin were collected after surgical resection.DNA was extracted from plasma and tumor tissues respectively.The mutation status of KRAS,NRAS,BRAF and PIK3 CA genes in tumor tissues and corresponding plasma were detected by next-generation sequencing.Analyzing the correlation between KRAS,KRAS,NRAS,BRAF,PIK3 CA in colorectal cancer patients and its relationship with clinicopathological features;Using tumor tissue test results as a control,the consistency,sensitivity and specificity,positive predictive value and negative predictive value of KRAS,NRAS,BRAF and PIK3 CA gene mutations detected in plasma were analyzed.Results:1.Among the 175 patients,60 had KRAS mutations in tissue samples,115 had wild type,and the mutation rate was 34.3%.There were 39 patients with KRAS mutation and 136 patients with wild type in the plasma samples,and the mutation rate was 22,3%.Tissue samples included 5 patients with NRAS mutations,170 with wild type,and a mutation rate of 2.9%;4 patients with NRAS mutations in plasma,and 136 with wild type,had a mutation rate of 2.3%.Among the tissue samples,there were 5 patients with BRAF mutation,170 with wild type,and the mutation rate was 2.9%.There were 4 cases with BRAF mutation in plasma,136 cases of wild type,and the mutation rate was 2.3%;There were 10 patients with PIK3 CA mutations in tissue samples,156 wild-type patients,and a mutation rate of 5.7%;8 patients with mutations in plasma samples,and 167 wild-type patients,with a mutation rate of 4.6%.One patient had double mutations in the KRAS gene and PIK3 CA gene,and the tissue was wild-type.2.Using tissue test results as controls,the sensitivity and specificity of KRAS mutations detected in plasma were 55.0% and 94.8%,and the positive and negative predictive values were 84.6% and 79.4%;the sensitivity and specificity of NRAS were 80.0% and 100%,positive predictive value and negative predictive value were 100% and 99.4%;the sensitivity and specificity of BRAF were 80.0% and 100%,positive predictive value and negative predictive value are 100% and 99.4%;the sensitivity and specificity of PIK3 CA were 70.0% and 99.4%,with positive and negative predictive values of 87.5% and 98.2%.The consistency rate of KRAS gene detection in plasma and tissue was 81.1%(kappa = 0.543);the consistency rate of NRAS gene detection was 99.4%(kappa = 0.886);the consistency rate of BRAF gene test was 99.4%(kappa = 0.886);the consistency rate of PIK3 CA gene test It was 97.7%(kappa = 0.655).3.There is a correlation between total gene mutation rates in plasma samples and pathological stages,stage Ⅰ(8.3%)vs stage Ⅱ(5%)vs stage Ⅲ(26.0%)vs stage Ⅳ(48.9%)(p <0.05),the difference had statistical significant;but had no statistical significant with gender and age(p> 0.05).Plasma samples and tissue samples detected 50.0% consistency rate of genetic mutations in stage Ⅰ(Kappa = 0.122);consistency rate of stageⅡ was 61.1%(Kappa = 0.137);consistency rate of stageⅢwas 80.0%(Kappa = 0.542);consistency rate of stageⅣ was 88.9%(Kappa = 0.778),and the coincidence rate between plasma samples and tissue samples increased with the increase of staging.4.Cross-mutation of PIK3 CA and KRAS occurred in 4 of 175 patients with CRC,but no cross-mutation was found in KRAS,NRAS,and BRAF.There was no correlation between KRAS,NRAS,BRAF,and PIK3 CA gene mutations,indicating that KRAS,NRAS,BRAF,and PIK3 CA gene mutations were independent of each other in CRC patients.The KRAS gene is related to pathological stage(p <0.05),and the mutation rate is higher in patients with higher pathological stage,regardless of gender and age(p> 0.05);NRAS,BRAF,PIK3 CA genes have no relationwith gender,age,and pathological stage(p> 0.05).Conclusion:1.There is no correlation between KRAS,NRAS,BRAF and PIK3 CA gene mutations in CRC patients,and they are independent2.Detection of plasma KRAS,NRAS,BRAF,PIK3 CA gene mutations by second-generation sequencing has higher sensitivity and specificity,and has a higher consistency rate compared with tumor tissue detection.3.When patients with CRC are unable to obtain tumor tissue specimens,especially advanced patients,plasma can be used instead of the mutation status detected by tissues to guide the patient’s targeted individualized treatment.