Epigenetically regulated gene expression profiles decipher four molecular subtypes with prognostic and therapeutic implications in gastric cancer

Abstract

Background Gastric cancer (GC) is one of the most common malignant tumors of the digestive tract which seriously endangers the health of human beings worldwide. Transcriptomic deregulation by epigenetic mechanisms plays a crucial role in the heterogeneous progression of GC. This study aimed to investigate the impact of epigenetically regulated genes on the prognosis, immune microenvironment, and potential treatment of GC. Results Under the premise of verifying significant co-regulation of the aberrant frequencies of microRNA (miRNA) correlated (MIRcor) genes and DNA methylation-correlated (METcor) genes. Four GC molecular subtypes were identified and validated by comprehensive clustering of MIRcor and METcor GEPs in 1521 samples from five independent multicenter GC cohorts: cluster 1 was characterized by up-regulated cell proliferation and transformation pathways, with good prognosis outcomes, driven by mutations, and was sensitive to 5-fluorouracil and paclitaxel; cluster 2 performed moderate prognosis and benefited more from apatinib and cisplatin; cluster 3 was featured by an up-regulated ligand–receptor formation-related pathways, poor prognosis, an immunosuppression phenotype with low tumor purity, resistant to chemotherapy (e.g., 5-fluorouracil, paclitaxel, and cisplatin), and targeted therapy drug (apatinib) and sensitive to dasatinib; cluster 4 was characterized as an immune-activating phenotype, with advanced tumor stages, benefit more from immunotherapy and displayed worst prognosis. Conclusions According to the epigenetically regulated GEPs, we developed four robust GC molecular subtypes, which facilitated the understanding of the epigenetic mechanisms underlying GC heterogeneity, offering an optimized decision-making and surveillance platform for GC patients.

Bioinformatics Analysis and Validation of Potential Markers Associated with Prediction and Prognosis of Gastric Cancer

“…However, rational application of these classifications in clinical settings is still not widespread, mainly due to their intricacy. Beyond these studies, many researchers have sought other surrogate molecular classifications of GC to detect biomarkers and validate therapeutic prediction based on various analyses, including epigenome, proteome, metabolome, immunity, and tumor mutational burden [ 56 , 72 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 ]. However, no studies can be said to have succeeded in identifying alternative markers that allow more efficient categories than those of TCGA and ACRG.…”

Molecular Classifications in Gastric Cancer: A Call for Interdisciplinary Collaboration

“… Weng’s classification. Clinicopathological and molecular features of GC types [ 172 ]. MSI: microsatellite instability; CIMP: CpG island methylator phenotype.…”

“…The C3 type was the least sensitive to gefitinib, gemcitabine, and sorafenib. [172] In their investigation, Weng et al analyzed data from 1521 GC samples across five independent datasets from GEO and TCGA databases [172]. Utilizing an integrated clustering algorithm that combined miRNA expression and DNA methylation profiles, they identified four distinct molecular types of GC, subsequently validated into four independent multicenter cohorts as follows: cluster 1 (C1, 30.4%), C2 (22.7%), C3 (15%), and C4 (32%).…”

Gastric Cancer in the Era of Epigenetics

“…In contrast, the C4 cluster involves immune-related processes, microsatellite stability, and TP53 mutations and relates to high-grade carcinomas and poor prognosis. Clusters 2 and 3 seem to correlate to a moderate prognosis, and advanced stages and are related to histone and DNA alterations, signaling pathways and immune invasion [ 11 ].…”

“…Beyond TCGA and ACRG classifications, several other classifications have been proposed. Recently, Weng et al (2023) suggested an epigenetic-based classification. According to their proposal, the analysis of 1521 GC cases from GEO and TCGA databases in combination with miRNA-expression and DNA-methylation profiles led to four molecular GC clusters.…”