Luan Mingyuan, Song Fucheng, Qu Shuyuan, Meng Xi, Ji Junjie, Duan Yunbo, Sun Changgang, Si Hongzong, Zhai Honglin
School of Basic Medicine, Qingdao University Medical College, Qingdao, Shandong 266071, P.R. China.
Department of Public Health, Qingdao University Medical College, Qingdao, Shandong 266071, P.R. China.
Oncol Lett. 2020 Oct;20(4):58. doi: 10.3892/ol.2020.11919. Epub 2020 Jul 29.
Lung cancer is a major cause of cancer-associated mortality worldwide. However, the association between multi-omics data and survival in lung cancer is not fully understood. The present study investigated the performance of the methylation survival risk model in multi-platform integrative molecular subtypes and aimed to identify copy number (CN) variations and mutations that are associated with survival risk. The present study analyzed 439 lung adenocarcinoma cases based on DNA methylation, RNA, microRNA (miRNA), DNA copy number and mutations from The Cancer Genome Atlas datasets. First, six cancer subtypes were identified using integrating DNA methylation, RNA, miRNA and DNA copy number data. The least absolute shrinkage and selection operator (LASSO) regression algorithm was used to extract methylation sites of survival model and calculate the methylation-based survival risk indices for all patients. Survival for patients in the high-risk group was significantly lower compared with that for patients in the low-risk group (P<0.05). The present study also assessed methylation-based survival risks of the six subtypes and analyzed the association between survival risk and non-silent mutation rate, number of segments, fraction of segments altered, aneuploidy score, number of segments with loss of heterozygosity (LOH), fraction of segments with LOH and homologous repair deficiency. Finally, the specific copy number regions and mutant genes associated with the different subtypes were identified (P<0.01). Chromosome regions 17q24.3 and 11p15.5 were identified as those with the most survival risk-associated copy number variation regions, while a total of 29 mutant genes were significantly associated with survival (P<0.01).
肺癌是全球癌症相关死亡的主要原因。然而，多组学数据与肺癌生存之间的关联尚未完全明确。本研究调查了甲基化生存风险模型在多平台整合分子亚型中的表现，旨在识别与生存风险相关的拷贝数（CN）变异和突变。本研究基于来自癌症基因组图谱数据集的DNA甲基化、RNA、微小RNA（miRNA）、DNA拷贝数和突变，分析了439例肺腺癌病例。首先，通过整合DNA甲基化、RNA、miRNA和DNA拷贝数数据，识别出六种癌症亚型。使用最小绝对收缩和选择算子（LASSO）回归算法提取生存模型的甲基化位点，并计算所有患者基于甲基化的生存风险指数。高危组患者的生存率显著低于低危组患者（P<0.05）。本研究还评估了六种亚型基于甲基化的生存风险，并分析了生存风险与非沉默突变率、片段数、改变的片段比例、非整倍体评分、杂合性缺失（LOH）片段数、LOH片段比例和同源修复缺陷之间的关联。最后，确定了与不同亚型相关的特定拷贝数区域和突变基因（P<0.01）。染色体区域17q24.3和11p15.5被确定为与生存风险相关拷贝数变异区域最多的区域，同时共有29个突变基因与生存显著相关（P<0.01）。
