Mutation Profile of Tibetan Lung Cancer Revealed by Whole Exome Sequencing
Xin Wang,Yutian Lai,Wei Dai,Jintao He,Guowei Che
DOI: https://doi.org/10.1016/j.jtho.2019.08.006
IF: 20.121
2020-01-01
Journal of Thoracic Oncology
Abstract:As reported by three major hospitals of Tibet autonomous region in last 2 decades, lung cancer ranked the sixth or fifth malignant tumor in a total of 4656 Tibetan cancer patients, which implies that maybe there are some unique features in Tibetan lung cancer.1Duan W.X. Ze Y.G. Ding J.W. Zhao Y.L. Bianba Z.X. Composition of patients with malignant tumors in a general hospital of Tibet [In Chinese].Bull Chinese Cancer. 2018; 27: 114-117Google Scholar, 2Li W.P. Chen C. Zang B.C. Min X.X. Hu F. Analysis on the syn-position of causing malignance of 2727 cases in Lhasa, TAR [In Chinese].J Tibet Univ. 2011; 26: 52-55Google Scholar Besides, epidemiologists found a negative correlation between the incidence and mortality rate of some specific cancer types and the altitude.3Van Pelt W.R. Epidemiological associations among lung cancer, radon exposure and elevation above sea level—a reassessment of Cohen's county level radon study.Health Physics. 2003; 85: 397-403Crossref PubMed Scopus (15) Google Scholar So, we conducted this research to investigate the mutational profile of lung cancer in high-altitude Tibetan population and enrolled 17 Tibetan lung cancer patients including 7 adenocarcinomas (Tibetan-LUAD) and 10 squamous cell carcinomas (Tibetan-LUSC). Paired tissues from tumors and normal lymph nodes were subjected to whole-exome sequencing by Illumina X TEN platform using paired-end 150X strategy. The Cancer Genome Atlas (TCGA)–LUAD/LUSC mutation data were downloaded by R package TCGA biolinks. Han-NSCLC mutation data were extracted from the article reported by Zhang et al.4Zhang X.C. Wang J. Shao G.G. et al.Comprehensive genomic and immunological characterization of Chinese non–small cell lung cancer patients.Nat Commun. 2019; 10: 1772Crossref PubMed Scopus (112) Google Scholar Six types of base substitution were shown in Figure 1A. The base substitution spectrum was dominated by C>T/G>A (38.76%) and followed by C>G/G>C (25.14%) in Tibet-LUSC. In Tibet-LUAD, the base substitution spectrum was dominated by C>T/G>A (38.97%) and followed by C>A/G>T (24.83%). We noticed higher proportion of C>T (p < 0.001) and lower proportion of C>A (p < 0.001) in Tibetan-NSCLC rather than TCGA-NSCLC or Han-NSCLC (Fig. 1B). A total of 4715 somatic mutations, including 3526 nonsynonymous mutations and 1189 synonymous mutations were detected. For nonsynonymous mutations, the median single nucleotide variations (SNVs) and the total SNVs of these patients were 128 (range, 5-1182 SNVs) and 3390, respectively, including 3029 missense mutations, 282 nonsense mutations, and 8 start/stop-loss mutations. And in 15 patients, a total of 136 indels were identified, including 8 splice site mutations, 73 frameshift mutations, 54 in-frameshift mutations, and 1 nonsense mutation. The median tumor mutation burden of Tibetan-LUSC and Tibetan-LUAD was 4.95 per MB (range, 0.11-26.42 MB) and 0.84 per MB (range, 0.58-4.36 MB), respectively. Mutation landscape was shown in Figure 2. Most frequently mutated genes in Tibetan-LUSC and Tibetan-LUAD were tumor protein p53 (TP53) and EGFR, respectively. Meanwhile, we compared the sample frequency of top mutated genes in Tibetan-NSCLC with TCGA-NSCLC and Han-NSCLC. Mutational signature 1 and 4, which were defined by the Catalog of Somatic Mutations in Cancer and associated with aging and smoking, contributed largely in most Tibetan NSCLC patients (Fig. 3).Figure 3Mutational Signature of Tibetan NSCLC and TCGA NSCLC. Weight of each signature was calculated individually. LUAD, lung adenocarcinoma; LUSC, lung squamous carcinoma; TCGA, The Cancer Genome Atlas.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Given the smoking history of the patients and the analysis of mutational signature, smoking accounted for the most important and prevalence exogenous carcinogen in our Tibetan NSCLC cohort. It is well known that signature 4 closely associated with C>A mutation. Although smoking made similar contribution in our Tibetan-NSCLC cohort as well as TCGA-NSCLC cohort and Han-NSCLC cohort, we found lowered proportion of C>A in our cohort. This might be explained by another important factor, the environment. For example, higher oxygen level caused DNA damage of G>8-oxodG and in the end resulted in higher frequency of C>A/G>T transversion.5Kino K. Sugiyama H. Possible cause of G-C-->C-G transversion mutation by guanine oxidation product, imidazolone.Chem Biol. 2001; 8: 369-378Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar In summary, this study was the first to describe mutational profile of Tibetan patients with NSCLC who lived in an average altitude of 3500 m. We provide some insights about the possible role of lowered oxygen level on NSCLC and uncovered a different mutation landscape of NSCLC which need further investigation. This research was supported by Cadre Health Care Research Project of Sichuan Province(HZ2019-103), and Key Project of Science & Technology Department of Sichuan Province(2018SZ0153). The authors thank Genetron Health who provided the next-generation sequencing platform and we appreciate the technique support of bioinformatic analysis provided by Feilong Zhao.