Clinical relevance of RNA editing profiles in lung adenocarcinoma

Si Shi; Shibin Chen; Menghang Wang; Bingchen Guo; Yaowu He; Hong Chen

doi:10.3389/fgene.2023.1084869

Clinical relevance of RNA editing profiles in lung adenocarcinoma

Front Genet. 2023 Mar 10:14:1084869. doi: 10.3389/fgene.2023.1084869. eCollection 2023.

Authors

Si Shi¹, Shibin Chen², Menghang Wang¹, Bingchen Guo³, Yaowu He¹, Hong Chen¹

Affiliations

¹ The Respiratory Department, The Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China.
² Medical Research Center, Beijing Chao-Yang Hospital, Capital Medical University, Beijing, China.
³ Department of Cardiology, The first Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China.

Abstract

Background: Lung adenocarcinoma (LUAD) is the most frequently occurring lung cancer worldwide, with increasing death rates. It belongs to the non-small cell lung cancer (NSCLC) type and has a strong association with previous smoking history. Growing evidence has demonstrated the significance of adenosine-to-inosine RNA editing (ATIRE) dysregulation in cancer. The aim of the present study was to evaluate ATIRE events that might be clinically useful or tumorigenic. Methods: To explore survival-related ATIRE events in LUAD, its ATIRE profiles, gene expression data, and corresponding patients' clinical information were downloaded from the Cancer Genome Atlas (TCGA) and the synapse database. We evaluated 10441 ATIRE in 440 LUAD patients from the TCGA database. ATIRE profiles were merged with TCGA survival data. We selected prognostic ATIRE sites, using a univariate Cox analysis (p < 0.001). Cox proportional hazards regression and lasso regression analysis were used to determine survival-related ATIRE sites, create risk ratings for those sites, and build a prognostic model and a nomogram for assessing overall survival (OS). Six ATIRE sites were used in the prognostic model construction and patients were randomly divided into a validation cohort (n = 176) and a training cohort (n = 264). The "Pheatmap" program was used to create risk curves that included risk score, survival time, and expression of ATIRE sites. We also determined the clinical prediction model's discrimination. The decision curve analysis and the 1-, 2-, and 3-year corrective curves were simultaneously used to evaluate the nomogram. We also evaluated the relationship between the amount of ATIRE sites and host gene expression and the impact of ATIRE expression on transcriptome expression. Results: The pyroglutamyl-peptidase I (PGPEP1) chr19:18476416A > I, ankyrin repeat domain 36B pseudogene 1 (ANKRD36BP1) (dist = 3,795), T-box transcription factor (TBX19) (dist = 29815) chr1:168220463A > I, Syntrophin Beta 2 (SNTB2) chr16:69338598A > I, hook microtubule-tethering protein 3 (HOOK3) chr8:42883441A > I, NADH dehydrogenase flavoprotein 3 (NDUFV3) chr21:44329452A > I, and FK506-binding protein 11 (FKBP11) chr12:49316769A > I were used in the prognostic model construction. High levels of risk score were significantly associated with worse OS and progression-free survival. Tumour stage and risk score were related to OS in LUAD patients. The predictors were among the prognostic nomogram model's risk score, age, gender, and tumor stage. The calibration plot and C-index (0.718) demonstrated the significant accuracy of nomogram's predictions. ATIRE level was markedly elevated in tumor tissues and was highly variable between patients. Conclusion: Events involving ATIRE in LUAD were highly functional and clinically relevant. The RNA editing-based model provides a solid framework for further investigation of the functions of RNA editing in non-coding areas and may be used as a unique method for predicting LUAD survival.

Keywords: RNA editing; lung adenocarcinoma; nomogram; overall survival; prognostic model.