Clinical significance of multi-sample next-generation sequencing for detection of gene mutations in non-small cell lung cancer

陈苍宋,陈振,郭雪晶,周建娅,孙文佳,方梁杰
DOI: https://doi.org/10.3760/cma.j.cn115355-20191229-00597
2020-01-01
Abstract:Objective:To explore the clinical significance of multi-sample next-generation sequencing (NGS) in detecting gene mutations in non-small cell lung cancer (NSCLC), so as to provide a basis for individualized targeted treatment.Methods:The data of 51 patients with NSCLC in the First Affiliated Hospital of Zhejiang University Medical College from June 2016 to February 2019 was retrospectively analyzed. The patients' age, gender, smoking status, pathological type, tumor staging, pleural invasion status and meningeal invasion status were collected, and the tissues, cerebrospinal fluid, pleural effusion and plasma samples were detected by NGS. The correlations between the driver gene mutations [epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), and ROS proto-oncogene 1, receptor tyrosine kinase (ROS1)] and the clinicopathological features of patients were analyzed by χ 2 test or Fisher exact probability method. Cohen Kappa coefficient analysis was used to compare the consistency of NGS detection results in different samples. Results:Among the 51 patients, 3 different types of specimens were examined in 7 patients. Of which, the driver gene mutations were all positive in 3 cerebrospinal fluid samples and double gene mutations were found in 2 hydrothorax samples. Driver gene mutations were more common in female [90.48% (19/21) vs. 50.00% (15/30), χ 2 = 9.107, P = 0.003], non-smokers [80.00% (24/30) vs. 47.62 (10/21), χ 2 = 5.829, P = 0.016], adenocarcinoma patients [72.34% (34/47) vs. 0, P = 0.017] and patients with malignant pleural effusion [92.86% (13/14) vs. 56.76% (21/37), χ 2 = 4.443, P = 0.035], and the differences were statistically significant. However, there was no statistically significant difference in mutations of driver gene among patients with different age or tumor stages (both P > 0.05). The consistency rate of genetic test results of the driver gene mutations in 37 plasma samples and matched tissue samples was 70.27% (26/37), κ value was 0.430, which suggested a good consistency between them. Conclusions:In patients with advanced NSCLC, gene detection of cerebrospinal fluid is of high application value for patients with meningeal metastasis; for patients with pleural invasion, gene detection of pleural effusion is an effective means to screen targeted therapy drugs. NGS detection of multiple specimens can better reflect the status of gene mutations and guide the individualized targeted therapy of lung cancer patients.
What problem does this paper attempt to address?