Integrating m6A Regulators-Mediated Methylation Modification Models and Tumor Immune Microenvironment Characterization in Caucasian and Chinese Low-Grade Gliomas

Wangrui Liu; Chuanyu Li; Yuhao Wu; Wenhao Xu; Shuxian Chen; Hailiang Zhang; Haineng Huang; Shuai Zhao; Jian Wang

doi:10.3389/fcell.2021.725764

Integrating m6A Regulators-Mediated Methylation Modification Models and Tumor Immune Microenvironment Characterization in Caucasian and Chinese Low-Grade Gliomas

Front Cell Dev Biol. 2021 Nov 25:9:725764. doi: 10.3389/fcell.2021.725764. eCollection 2021.

Authors

Wangrui Liu^{1

2}, Chuanyu Li², Yuhao Wu³, Wenhao Xu⁴, Shuxian Chen⁵, Hailiang Zhang⁴, Haineng Huang², Shuai Zhao¹, Jian Wang¹

Affiliations

¹ Department of Transplantation, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.
² Department of Neurosurgery, Affiliated Hospital of Youjiang Medical University for Nationalities, Guangxi, China.
³ School of Medicine, Tongji University, Shanghai, China.
⁴ Department of Urology, Fudan University Shanghai Cancer Center, Shanghai Medical College, Fudan University, Shanghai, China.
⁵ Department of Oncology, Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.

Abstract

Background: As an important epigenetic modification, m6A methylation plays an essential role in post-transcriptional regulation and tumor development. It is urgently needed to comprehensively and rigorously explore the prognostic value of m6A regulators and its association with tumor microenvironment (TME) infiltration characterization of low-grade glioma (LGG). Methods: Based on the expression of 20 m6A regulatory factors, we comprehensively evaluated the m6A modification patterns of LGG after unsupervised clustering. Subsequent analysis of the differences between these groups was performed to obtain m6A-related genes, then consistent clustering was conducted to generate m6AgeneclusterA and m6AgeneclusterB. A Random Forest and machining learning algorithms were used to reduce dimensionality, identify TME characteristics and predict responses for LGG patients receiving immunotherapies. Results: Evident differential m6A regulators were found in mutation, CNV and TME characteristics of LGG. Based on TCGA and CGGA databases, we identified that m6A regulators clusterA could significantly predict better prognosis (p = 0.00016) which enriched in mTOR signaling pathway, basal transcription factors, accompanied by elevated immune cells infiltration, and decreased IDH and TP53 mutations. We also investigated the distribution of differential genes in m6A regulators clusters which was closely associated with tumor immune microenvironment through three independent cohort comparisons. Next, we established m6Ascore based on previous m6A model, which accurately predicts outcomes in 1089 LGG patients (p < 0.0001) from discovering cohort and 497 LGG patients from testing cohort. Significant TME characteristics, including genome heterogeneity, abidance of immune cells, and clinicopathologic parameters have been found between m6Ascore groups. Importantly, LGG patients with high m6Ascore are confronted with significantly decreased responses to chemotherapies, but benefit more from immunotherapies. Conclusion: In conclusion, this study first demonstrates that m6A modification is crucial participant in tumorigenesis and TME infiltration characterization of LGG based on large-scale cohorts. The m6Ascore provides useful and accurately predict of prognosis and clinical responses to chemotherapy, immunotherapy and therapeutic strategy development for LGG patients.

Keywords: RNA modificatio; immunotherapy; low-grade glioma; m6A; prognosis; tumor environment.