Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jan 21:11:634634.
doi: 10.3389/fgene.2020.634634. eCollection 2020.

A Novel Prognostic Model of Early-Stage Lung Adenocarcinoma Integrating Methylation and Immune Biomarkers

Jin Ren  1 , Yun Yang  1 , Chuanyin Li  2 , Lu Xie  3 , Ronggui Hu  1   4 , Xiong Qin  5 , Menghuan Zhang  3   4
Affiliations

A Novel Prognostic Model of Early-Stage Lung Adenocarcinoma Integrating Methylation and Immune Biomarkers

Jin Ren et al. Front Genet. .

Abstract

Lung adenocarcinoma (LUAD) is caused by multiple biological factors. Therefore, it will be more meaningful to study the prognosis from the perspective of omics integration. Given the significance of epigenetic modification and immunity in tumorigenesis and development, we tried to combine aberrant methylation and tumor infiltration CD8 T cell-related genes to build a prognostic model, to explore the key biomarkers of early-stage LUAD. On the basis of RNA-seq and methylation microarray data downloaded from The Cancer Genome Atlas (TCGA), differentially expressed genes and aberrant methylated genes were calculated with "DEseq2" and "ChAMP" packages, respectively. A Chi-square test was performed to obtain methylation driver genes. Weighted correlation network analysis (WGCNA) was utilized to mine cancer biomarkers related to CD8 T cells. With the consequences of univariate Cox proportional hazards analysis and least absolute shrinkage and selection operator (LASSO) COX regression analysis, the prognostic index based on 17 methylation driver genes (ZNF677, FAM83A, TRIM58, CLDN6, NKD1, NFE2L3, FKBP5, ITGA5, ASCL2, SLC24A4, WNT3A, TMEM171, PTPRH, ITPKB, ITGA2, SLC6A17, and CCDC81) and four CD8 T cell-related genes (SPDL1, E2F7, TK1, and TYMS) was successfully established, which could make valuable predictions for the survival risk of patients with early-stage LUAD.

Keywords: CD8 T cell; lung adenocarcinoma; methylation; prognostic model; survival analysis.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
The workflow of the study.
FIGURE 2
FIGURE 2
The methylation driver genes. (A) Heatmap of the 277 methylation driver genes. The color from blue to red indicates the trend from hypomethylation to hypermethylation. (B) Scatter plot of the top six negatively correlated genes. The x-axis represents the RNA-seq expression value, and the y-axis represents the methylation beta value.
FIGURE 3
FIGURE 3
Functional enrichment analysis of methylation driver genes. (A) The top 10 enriched terms of the biological process of GO. The outer circle represents the expression values (log2FC) of methylation driver genes in each enriched GO term. Red dots imply upregulated genes and blue dots imply downregulated genes. The inner circle indicates the significance of GO terms. (B) The chord plot of top five KEGG enriched pathways. The left outer semicircle represents the log2FC value of methylation driver genes, and the right semicircle corresponds to five pathway entries.
FIGURE 4
FIGURE 4
Key modules and its correlation with CIBERSORT immune fractions. (A) Gene modules identified by hierarchical clustering. A total of 15 non-gray modules were generated. (B) The correlation between modules and CIBERSORT immune fractions. The turquoise module depicts the highest correlation (r = 0.4, p = 3e-15) with CD8 T cells.
FIGURE 5
FIGURE 5
Construction of risk assessment signature. (A) 241 promising candidates (P ≤ 0.05), including 32 protective and 209 risk markers. (B) The optimal λ value of 0.02774519 was selected to identify the most robust markers for prognosis in LASSO regression. (C) A combination of 21 genes remained with their individual non-zero LASSO coefficients. (D) The distribution of LASSO coefficients of the gene signature. The blue bars represent protective biomarkers and the red bars represent risk biomarkers.
FIGURE 6
FIGURE 6
Outcomes of the prediction model. (A) The distribution of risk scores for 345 early-stage LUAD patients. (B) The distribution of survival status, sorted by risk score. (C) Kaplan–Meier survival curve. It demonstrated the survival difference between the high-risk group and low-risk group. (D) ROC curve. It showed the performance of the model. (E) Kaplan–Meier survival curve using the validation cohort. Patients with higher risk score exhibited worse overall survival.
FIGURE 7
FIGURE 7
Construction of prognostic nomogram. (A) Nomogram for predicting 3- and 5-year survival probability, by integrating 21 biomarkers, age, risk score, sex, and tumor stage. (B) ROC curves of the nomogram. In the time-dependent ROC curve, the nomogram also displayed a robust performance in predicting the 3- and 5-year survival rates. The AUC was 0.78 and 0.76, respectively. (C,D) The calibration curves of 3- and 5-year survival rates. They showed an optimal agreement between the nomogram predictions and actual observations.
FIGURE 8
FIGURE 8
Survival analysis of four immune-related biomarkers. (A) Interaction network of TYMS, E2F7, SPDL1, and TK1. TYMS is the target of the FDA-approved drug Pemetrexed. (B) Heatmap of CD8 T cell-related genes. (C) Kaplan–Meier survival curve. It showed the significant survival difference between patients with low expression of TYMS, E2F7, SPDL1, TK1 (group 1), and other patients (group 2). (D) Kaplan–Meier survival curve. Group 2 was further classified into group 3 and group 4 according to our defined risk score index. It displayed the significant survival difference among the three groups, including the low proliferation group (group 1), the high proliferation and low risk score group (group 3), and the high proliferation and high risk score group (group 4).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Azimi F., Scolyer R. A., Rumcheva P., Moncrieff M., Murali R., McCarthy S. W., et al. (2012). Tumor-infiltrating lymphocyte grade is an independent predictor of sentinel lymph node status and survival in patients with cutaneous melanoma. J. Clin. Oncol. 30 2678–2683. 10.1200/jco.2011.37.8539 - DOI - PubMed
    1. Bernstein B. E., Meissner A., Lander E. S. (2007). The mammalian epigenome. Cell 128 669–681. 10.1016/j.cell.2007.01.033 - DOI - PubMed
    1. Blanche P., Dartigues J. F., Jacqmin-Gadda H. (2013). Estimating and comparing time-dependent areas under receiver operating characteristic curves for censored event times with competing risks. Stat. Med. 32 5381–5397. - PubMed
    1. Bos R., Sherman L. A. (2010). CD4+ T-cell help in the tumor milieu is required for recruitment and cytolytic function of CD8+ T lymphocytes. Cancer Res. 70 8368–8377. 10.1158/0008-5472.can-10-1322 - DOI - PMC - PubMed
    1. Cotto K. C., Wagner A. H., Feng Y.-Y., Kiwala S., Coffman A. C., Spies G., et al. (2018). DGIdb 3.0: a redesign and expansion of the drug-gene interaction database. Nucleic Acids Res. 46 D1068–D1073. - PMC - PubMed