Research Paper Volume 11, Issue 16 pp 6153—6174

Prognostic value of genome-wide DNA methylation patterns in noncoding miRNAs and lncRNAs in uveal melanomas

Zheng Zheng1, *, , Dan Xu1, *, , Keqing Shi2, , Minfeng Chen1, , Fan Lu1, ,

  • 1 School of Ophthalmology and Optometry, Eye Hospital, Wenzhou Medical University, State Key Laboratory and Key Laboratory of Vision Science, Ministry of Health and Zhejiang Provincial Key Laboratory of Ophthalmology and Optometry, Wenzhou, Zhejiang 325000, China
  • 2 The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang 325000, China
* Equal contribution

Received: January 15, 2019       Accepted: August 9, 2019       Published: August 20, 2019
How to Cite

Copyright © 2019 Zheng et al. This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 3.0) License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Background: Uveal melanomas are the most common primary intraocular malignant tumors in adults, associated with a high metastasis rate and a low 5-year survival rate. It is a clinic urgency and importance to identify prognostic factors for UVMs.

Results: 55 aberrantly methylated sites of miRNAs and 47 aberrantly methylated sites of lncRNAs were observed between Alive < 2 years group and Alive > 2 years group of UVMs. Two prognostic classifiers were generated. For 13- miRNAs-CpG-classifier, the AUC were 0.958, 0.848 and 0.824 at 1 year, 2 years and 3 years, respectively. For 9- lncRNAs-CpG-classifier, the AUC were 0.943, 0.869 and 0.866 at 1 year, 2 years and 3 years, respectively.

Conclusion: The correlation between genome-wide DNA methylation patterns of miRNAs and lncRNAs and the overall survival in UVMs were identified in this study. This novel finding shed new light on developing biomarkers of prognosis for UVMs.

Methods: DNA methylation profiles of noncoding miRNAs and lncRNAs for UVMs were accessed from The Cancer Genome Atlas. Then the prognostic value was analyzed by least absolute shrinkage and selection operator method Cox regression and tested by Time-dependent Receiver Operating Characteristic curve.


UVMs: Uveal melanomas; miRNAs: microRNAs; lncRNAs: long ncRNAs; TCGA: The Cancer Genome Atlas; LASSO: least absolute shrinkage and selection operator method; tdROC: Time-dependent Receiver Operating Characteristic curve; MAPK: mitogen-activated protein kinase; TSS: transcriptional start site; HC: Hierarchical clustering.