Integrated Analyses Reveal Potential Functional N6-Methyladenosine-Related Long Noncoding RNAs in Adrenocortical Adenocarcinoma

Background: Adrenocortical adenocarcinoma (ACC) is known to be a relatively uncommon malignant tumor of the adrenal gland with patients having a poor prognosis. At present, few reports are available concerning the m6A modifications of lncRNAs as well as their clinical and immunological significance in the occurrence and progression of ACC. Materials and Methods: In the present research, 21 m6A-related genes were analyzed. Both multivariate and univariate Cox regression analyses were conducted to examine the prognostic m6A-related lncRNAs. A sum of 165 m6A-related lncRNAs was obtained from The Cancer Genome Atlas (TCGA) dataset. Based on the expressions of m6A-related lncRNAs, all ACC patients were classified into distinct subgroups using the consistent clustering method. Finally, m6A-related lncRNAs that were shown to have prognostic value were utilized to develop an m6A-related lncRNA risk model, which may be employed in the prediction of prognosis and survival. Results: Using TCGA data set, 26 m6A-associated lncRNAs having putative prognostic values were identified according to their expression levels, TCGA-AAC patients were classified into two clusters with the aid of consistent clustering analysis. The correlation between the two clusters was low, in which cluster1 consisted of 42% of all ACC patients. The survival analysis showed that cluster1 was associated with an unfavorable prognosis relative to cluster2. A risk model was constructed incorporating 26 m6A-associated lncRNAs that were correlated with patient prognosis. The model was subsequently validated by univariate and multivariate Cox, receiver operating characteristic (ROC) curve, and survival analyses. We also observed that the m6A-related risk model performed well in anticipating prognoses and survival status of patients with AAC. The overall survival (OS) of the high-risk cohort, as predicted by the model, was lower as opposed to that of the low-risk cohort. Conclusion: In the present research, we developed a risk model consisting of 4 m6A-related long-noncoding RNAs (lncRNAs), which can exert independent predictive values in patients with ACC. Our findings demonstrated that these 4 m6A-related lncRNAs perform integral functions in the tumor immune microenvironment, and also revealed the possibility of using these lncRNAs to guide the development of prognostic classifications and therapy approaches for ACC patients.