Multi-omics analysis reveals critical metabolic regulators in bladder cancer

BACKGROUND

The crosstalk between genomic alterations and metabolic dysregulation in bladder cancer is largely unknown. A deep understanding of the interactions between cancer drivers and cancer metabolic changes will provide novel opportunities for targeted therapeutic strategies.

METHODS

Three primary bladder cancer specimens with paired normal tissues or blood samples were subjected to whole-exome sequencing, DNA methylation array and whole-transcriptome sequencing by next-generation sequencing technology. We applied the methods to multi-omics data combining the Cancer Genome Atlas (TCGA) bladder cancer samples, including somatic mutation, DNA copy number, DNA methylation and gene expression profile for validation.

RESULTS

We identified 34 mutated cancer driver genes in bladder cancer. KDM6A was the most significantly mutated cancer driver gene. Metabolic pathways were enriched in both differentially methylated regions (DMRs) and differentially expressed genes. Twenty-nine DMRs in the TSS200 region were highly correlated with the upregulation of gene expression, and 24 DMRs in the genome were highly correlated with the downregulation of gene expression. A total of 201 genes had highly correlated DNA methylation and expression. Thirty-four genes, including the known metabolic genes CXXC5, PRR5, ABCB8 and BAHD1, were further validated in the TCGA cohort. Multi-omics alterations identified two new candidate driver genes, WIPI2 and GFM2, that warrant future studies.

CONCLUSIONS

This study provides a comprehensive and systematic analysis, focusing on identifying key regulatory factors that may lead to cancer metabolic heterogeneity. Further understanding and verification of the cancer genes driving metabolic reprogramming and their role in the progression of bladder cancer will help to identify new therapeutic targets.

Gene Expression Patterns Unveil New Insights in Papillary Thyroid Cancer

Background and objectives: Papillary thyroid carcinoma is the most frequent variety of all malignant endocrine tumors. It represents a heterogeneous malignancy with various clinical outcomes, emphasizing the need to identify powerful biomarkers with clinical relevance. Materials and Methods: Available gene expression data (level 3) for thyroid cancers were downloaded from the Cancer Genome Atlas (TCGA), followed by bioinformatic analyses performed on the data set. Results: Based on gene expression analysis, we were able to identify common and specific gene signatures for the three main types of papillary thyroid carcinoma (classical, follicular variant, and tall-cell). The survival rate was not significantly different among the main subtypes, but we were able to identify a biological adhesion signature with impact in patient prognostic. Conclusions: Taken together, the gene expression signature and particular adhesion signature, along with ITGA10 and MSLN in particular, could be used as a prognostic tool with important clinical relevance.