AUTHOR=Liu Jingchao , Wang Jianlong , Wu Meng , Zhang Wei , Meng Lingfeng , Wang Jiawen , Lv Zhengtong , Xia Haoran , Zhang Yaoguang , Wang Jianye 

TITLE=Comprehensive Analysis of N6-methyladenosine Modification Patterns Associated With Multiomic Characteristics of Bladder Cancer

JOURNAL=Frontiers in Medicine

VOLUME=Volume 8 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2021.757432

DOI=10.3389/fmed.2021.757432

ISSN=2296-858X

ABSTRACT=Purpose: 
To comprehensively analyze N6-methyladenosine modification patterns in bladder tumors and to further systematically explore the inherent relationships between these modification patterns and multiomic tumor characteristics.
Materials and Methods:
A total of 901 bladder tumor samples, including 405 bladder tumors from TCGA database, 188 bladder tumors from GEO database GSE13507 and 308 bladder tumors from GEO database GSE32894, were included in this systematic analysis. The N6-methyladenosine modification patterns were identified utilizing unsupervised clustering analysis. To quantify N6-methyladenosine modification patterns, the m6Ascore of individual samples was developed using principal component analysis algorithms. The relationships among immune infiltration, tumor mutation burden, various clinical characteristics, molecular subtypes and the m6Ascore were systematically analyzed. The guiding value of the m6Ascore in bladder tumor immunotherapy was further validated in an external trial cohort.
Results:
We determined three different N6-methyladenosine modification patterns for 901 bladder tumors. The quantitative m6Ascore of individual samples derived from N6-methyladenosine modification patterns could play a significant role in predicting overall survival (p < 0.001), immune cell infiltration (p < 0.05), and classic oncogene mutations (p < 0.05). A low m6Ascore combined with high tumor mutation burden indicated better survival outcomes. (p < 0.001) A higher m6Ascore also tended to indicate a higher grade (p < 0.001), higher T stage (p < 0.001), older age (p < 0.001), and higher death rate (p < 0.001). The Basal type tended to exhibit significantly higher m6Ascores than the Luminal and Neuronal subtypes. (p < 0.001)  Furthermore, a higher m6Ascore was also positively correlated with higher PD-L1, PD-1, and CTLA4 expression. (p < 0.01) External immunotherapy cohorts demonstrated that no difference in therapeutic effects was noted between the high and low m6Ascore groups when anti-PD1 immunotherapy was exclusively administered. When anti-PD1 and anti-CTLA4 immunotherapy were simultaneously administered, the high m6Ascore group had a significantly better prognosis than the low m6Ascore group (p < 0.001).
Conclusions: This study systematically revealed the important roles of m6A methylation modification patterns in bladder tumors. Detailed quantification of m6A modification patterns could improve our understanding of the bladder tumor microenvironment and could provide guidance for future immunotherapy strategies.