Abstract
In patients with lung adenocarcinoma (LUAD), the prognostic role of adjacent nontumor tissues is still unknown. Alterations in the activity of immunologic and hallmark gene sets in adjacent nontumor tissues may have a potential influence on cell proliferation of normal lung cell after pulmonary lobectomy. We sought to discover LUAD subgroups and prognostic gene sets based on changes in gene set activity in tumor and adjacent nontumor tissues. Firstly, we used gene set variation analysis (GSVA) to characterize the activity changes of 4922 gene sets in LUAD and nontumor samples. Luckily, we identified three novel LUAD subtypes using the nonnegative matrix factorization (NMF) algorithm. In detailed, patients with subtype-3 had a favorable prognosis, but subtypes 1 and 2 had a bad prognosis. In addition, patients with subtype-3 in the validation cohort also lived longer. Meanwhile, using the LASSO-Cox algorithm, we discovered 15 prognostic gene sets in tumors (T gene sets) and two prognostic gene sets in adjacent nontumors (N gene sets). Interestingly, genes from N gene sets were related with immune response in nontumor tissues, but genes from T gene sets were correlated with DNA damaging and repairing in tumor tissues. These findings highlighted the possibility of a stronger immune response in nearby nontumor tissues. In conclusion, our study established a theoretical foundation for selecting therapy strategy for LUAD patients that should be guided by changes in activity in tumor and adjacent nontumor tissues, particularly after pulmonary lobectomy.
Highlights
According to the World Health Organization (WHO) reports, lung cancer will continue to be one of the most lethal types of cancer globally by 2020 [1], with a 5-year survival rate of just 19% [2]
Previous bioinformatic studies have focused on lung adenocarcinoma (LUAD) tissues, differentiating subtypes or risk stratifying based on the integration of whole gene expression profiles in LUAD tissues and according to specific gene expression, but have grossly underestimated the role of adjacent nontumor tissues in HCC subtypes
GO and KEGG enrichment analyses revealed that in nontumor tissues, genes were related with immune response, but genes in tumor tissues were correlated with DNA damaging and repairing
Summary
According to the World Health Organization (WHO) reports, lung cancer will continue to be one of the most lethal types of cancer globally by 2020 [1], with a 5-year survival rate of just 19% [2]. Non-small-cell lung cancer (NSCLC) accounts for 85 percent of all lung cancers patients, with lung adenocarcinoma (LUAD) being a subset of NSCLC (NSCLC) [3]. The majority of LUAD patients diagnosed are already at an advanced stage [4]. Investigations of pathway activity alterations in LUAD have tended to concentrate on particular pathways rather than systematically evaluating plenty of pathways in tumor and nontumor samples. In some LUAD patients after pulmonary lobectomy, the remaining cells will proliferate or repair and may sometimes become cancerous again [10].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
