The DDR-related gene signature with cell cycle checkpoint function predicts prognosis, immune activity, and chemoradiotherapy response in lung adenocarcinoma

Abstract

Background: As a DNA surveillance mechanism, cell cycle checkpoint has recently been discovered to be closely associated with lung adenocarcinoma (LUAD) prognosis. It is also an essential link in the process of DNA damage repair (DDR) that confers resistance to radiotherapy. Whether genes that have both functions play a more crucial role in LUAD prognosis remains unclear.

Methods: In this study, DDR-related genes with cell cycle checkpoint function (DCGs) were selected to investigate their effects on the prognosis of LUAD. The TCGA-LUAD cohort and two GEO external validation cohorts (GSE31210 and GSE42171) were performed to construct a prognosis model based on the least absolute shrinkage and selection operator (LASSO) regression. Patients were divided into high-risk and low-risk groups based on the model. Subsequently, the multivariate COX regression was used to construct a prognostic nomogram. The ssGSEA, CIBERSORT algorithm, TIMER tool, CMap database, and IC50 of chemotherapeutic agents were used to analyze immune activity and responsiveness to chemoradiotherapy.

Results: 4 DCGs were selected as prognostic signatures, and patients in the high-risk group had a lower overall survival (OS). The lower infiltration levels of immune cells and the higher expression levels of immune checkpoints appeared in the high-risk group. The damage repair pathways were upregulated, and chemotherapeutic agent sensitivity was poor in the high-risk group.

Conclusions: The 4-DCGs signature prognosis model we constructed could predict the survival rate, immune activity, and chemoradiotherapy responsiveness of LUAD patients.

Keywords: Cancer prognosis; Cell cycle checkpoint; DNA damage repair; Gene signature; Lung adenocarcinoma.

Fig. 1

The workflow chart of this study

Fig. 2

Identification of DDR-related genes with cell cycle checkpoint function (DCGs). A GO enrichment analysis of 296 DDR genes, BP: Biological process; MF: Molecular function; CC: Cellular components. B KEGG enrichment analysis of 296 DDR genes. C The Biological processes of cell cycle checkpoint involved by DDR genes

Fig. 3

Identification of differentially expressed DCGs in TCGA-LUAD and normal samples. A Differential expression volcano plot based on DCGs (red: P < 0.05 and |logFC|> = 1, blue: only P < 0.05). B The heatmap of 24 differentially expressed DCGs in LUAD and normal samples (blue: low expression level, orange: high expression level, **P < 0.01, ***P < 0.001). C Protein–protein interaction (PPI) network of 22 DCGs. D The hub genes calculated by the PPI network

Fig. 4

DNA methylation survival analysis of DCGs based on the MethSurv database. The Kaplan–Meier (K–M) survival curve of LUAD patients with different methylation levels in A cg17653972 from BRSK1, B cg18576335 from AURKB, C cg25653141 from BLM, D cg09161138 from CDT1, E cg22041712 from CENPF, F cg12148237 from PRKDC, and G cg07084161 from NBN

Fig. 5

LUAD classification based on the DCGs. A Consensus clustering matrix for k = 2. B Heatmap for the expression of DCGs based on the two clusters and clinical features. C Kaplan–Meier OS curve for two clusters in LUAD

Fig. 6

Prognosis model based on the DCGs signature. A A forest plot of univariate Cox regression analysis for 24 DCGs. B Cross-validation for optimal parameter selection in the LASSO regression. C Partial likelihood deviation under the number of different variables. D Distribution of patients based on the risk score. E, F Analysis of the survival rate and survival status in the two risk groups. G Principal component analysis (PCA) of the 4-DCGs signature. H The time-dependent receiver operating characteristic (ROC) of 4-DCGs signature

Fig. 7

The validation of the DCGs signature Prognosis model in two GEO cohorts. A, F Distribution of patients based on the risk score. B, C, G, H Analysis of the survival rate and survival status in the two risk groups. D, I Principal component analysis (PCA) of the 4-DCGs signature. E, J The time-dependent receiver operating characteristic (ROC) of 4-DCGs signature

Fig. 8

Establishment of a prognosis nomogram based on TCGA-LUAD. A Univariate COX regression analysis for the risk score and clinical characteristics. B Multivariate COX regression analysis for independent prognostic factors. C Heatmap for 4-DCGs and the correlation between clinical features and the risk groups (*P < 0.05). D Prognosis nomogram for predicting the survival of patients based on multivariate Cox regression analysis. E The calibration curves of the Nomogram for 1-year, 3-year, and 5-year survival in LUAD patients

Fig. 9

Functional enrichment of the 4-DCGs signature in TCGA-LUAD. A GO enrichment analysis of 4-DCGs, BP: Biological process; MF: Molecular function; CC: Cellular components. B KEGG enrichment analysis of 4-DCGs

Fig. 10

Immunoactivity analysis of two risk groups in TCGA-LUAD. A The composition of 22 types of tumor-infiltrating immune cells in the TCGA-LUAD samples. B The expression of 27 immune checkpoints in two risk groups. C–E Correlation analysis for the expression of immune checkpoints and risk scores

Fig. 11

Comparison of chemoradiotherapy response in TCGA-LUAD. A The ssGSEA scores of 16 DDR pathways in the two risk groups. B The ssGSEA scores of X-ray and UV response in the two risk groups. C–F The sensitivity analysis of LUAD common chemotherapy agents (Cisplatin, Crizotinib, Erlotinib, and Nilotinib) in two risk groups. G–J The sensitivity analysis of other cancers’ common chemotherapy agents (Axitinib, Camptothecin, Etoposide, and Gemcitabine)