Background:
Urinary bladder cancer is one of the most common malignancies of the urinary tract; however, the mechanism of its high recurrence and responses to immunotherapy remains unclear, resulting in difficulties in clinical outcome prediction. Epigenetic alterations, especially DNA methylation, play important roles in bladder cancer development and are increasingly investigated as biomarkers for diagnostic or prognostic predictions. However, little is known regarding hydroxymethylation since previous studies based on bisulfite-sequencing approaches could not differentiate between 5mC and 5hmC signals, resulting in entangled methylation results.
Methods:
Here, we used a multi-omics approach to provide the genome, transcriptome, methylome, and hydroxymethylome landscape of both primary and recurrent bladder cancers.
Results:
By whole exome sequencing, we identified driver mutations that are involved in the UBC development, such as FGFR3, KDMTA and KDMT2C. However, few of these driver mutations are associated with the recurrence or the PD-L1 down-regulation in UBC. By integrating RRBS and oxRRBS-seq data, we identified fatty acid oxidation-related genes significantly enriched in 5hmC-associated transcription alterations in recurrent bladder cancers. We also observed a serial of 5mC hypomethylation DMRs in the gene body of NFATC1, a gene that is highly involved in the T-cell immune response, in bladder cancer samples with high expression of PD-L1. And since 5mC and 5hmC alternations are globally anti-correlated, RRBS-seq based markers which combine the 5mC and 5hmC signals, attenuate the cancer-related signals and therefore are not optimal to be used as clinical biomarkers.
Conclusions:
By multi-omics profiling of UBC samples, we showed that epigenetic alternations are more involved than genetic mutations in the recurrence and PD-L1 regulation of UBC. In addition, as a proof of principle, we demonstrated that the combined measurement of 5mC and 5hmC level by bisulfite-based method will compromise the prediction accuracy of epigenetic biomarkers.