Gene Signature and Identification of Clinical Trait-Related m 6 A Regulators in Pancreatic Cancer

Jie Hou 1, Zhan Wang 1, Hong Li 1, Hongzhi Zhang 1, Lan Luo 1

Affiliations

PMID: 32754191
PMCID: PMC7367043
DOI: 10.3389/fgene.2020.00522

Free PMC article

Abstract

Pancreatic cancer (PC) has a very poor prognosis and is usually diagnosed only at an advanced stage. The discovery of new biomarkers for PC will help in early diagnosis and a better prognosis for patients. Recently, N6-methyladenosine (m6A) RNA modifications and their regulators have been implicated in the development of many cancers. To investigate the functions and mechanisms of m6A modifications in the development of PC, 19 m6A regulators, including m6A-methyltransferases (ZC3H13, RBM15/15B, WTAP, KIAA1429, and METTL3/14), demethylases (FTO and ALKBH5), and binding proteins (YTHDF1/2/3, YTHDC1/2, IGF2BP1/2/3, HNRNPC, and HNRNPA2B1) were analyzed in 178 PC tissues from the cancer genome atlas (TCGA) database. The results were verified in PC cell lines Mia-PaCa-2, BXPC-3, and the control cell line HDE-CT. The m6A regulators-based sample clusters were significantly related to overall survival (OS). Further, lasso regression identified a six-m6A-regulator-signature prognostic model (KIAA1429, HNRNPC, METTL3, YTHDF1, IGF2BP2, and IGF2BP3). Model-based high-risk and low-risk groups were significantly correlated with OS and clinical traits (pathologic M, N, and clinical stages and vital status). The risk signature was verified as an independent prognostic marker for patients with PC. Finally, gene set enrichment analysis revealed m6A regulators (KIAA1429, HNRNPC, and IGF2BP2) were related to multiple biological behaviors in PC, including adipocytokine signaling, the well vs. poorly differentiated tumor pathway, tumor metastasis pathway, epithelial mesenchymal transition pathway, gemcitabine resistance pathway, and stemness pathway. In summary, the m6A regulatory factors which related to clinical characteristics can be involved in the malignant progression of PC, and the constructed risk markers may be a promising prognostic biomarker that can guide the individualized treatment of PC patients.

Keywords: biomarker; clinical traits; m6A regulators; pancreatic cancer; prognostic model.