Integrating Methylation Signature Modifications and Subtype Classification into a Prognostic Risk Model for Survival Prognosis and Predicting Immunotherapy Response in Colorectal Cancer.
TL;DR
This study shows that the integrated analysis of multiple RNA methylation modifications provides a more comprehensive view of epitranscriptomic heterogeneity in CRC than single-modification approaches and suggests that multi-modification-based models may improve prognostic accuracy and help identify CRC patients who may benefit from immunotherapy.
Abstract
Background
The prognosis for colorectal cancer (CRC) is poor, and the disease is marked by high rates of morbidity and death, highlighting the need for reliable biomarkers. RNA methylation modifications play important roles in cancer biology. However, the integrated role of m6A/m5C/m1A/m7G modifications in CRC has not been fully characterized and requires further investigation.
Methods
Data from The Cancer Genome Atlas colon adenocarcinoma cohort (TCGA-- COAD) were used as the training set, and GSE17536 served as the validation cohort. We further investigated the probable biological processes and prognostic significance of methylation modifications-related genes in CRC using immune infiltration analysis, enrichment analysis, consistent clustering, Kaplan-Meier (KM) survival analysis, TIMER database exploration, and differential expression analysis. To evaluate the association between risk grade and survival prognosis, medication sensitivity, and immune infiltration, we created the Methylation Modifications Risk Model (MMRM) and confirmed its validity using immunohistochemical (IHC) and immunofluorescence (IF) staining.
Results
Using information from 45 methylation modifications-related genes, consensus clustering to identify 3 categories of patients. There were variations across the three groups in terms of immune cell infiltration, survival, and immunological scores. Then, using LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis, we constructed MMRM that could distinguish between high- and low-risk populations utilizing BCL10, LEPROTL1, DPP7, P4HA1, SLC39A8, and AC008735.2. Using the validation set, KM survival analysis, TIMER database exploration, and IHC and IF staining, its accuracy was further supported. These findings suggest that m6A/m5C/m1A/m7G are associated with the immunological milieu of CRC tumors, and that MMRM is an excellent predictor of CRC patient survival.
Discussion
This study shows that the integrated analysis of multiple RNA methylation modifications provides a more comprehensive view of epitranscriptomic heterogeneity in CRC than single-modification approaches. The association between the MMRM, immune landscape, and immune checkpoint expression suggests the potential mechanistic link between RNA methylation regulation and tumor immune evasion. These findings suggest that multi-modification-based models may improve prognostic accuracy and help identify CRC patients who may benefit from immunotherapy.
Conclusion
In this work, we emphasized the correlation between alterations in the CRC immunotumor microenvironment and the m6A/m5C/m1A/m7G subtypes. We developed and validated MMRM, which is useful for predicting treatment sensitivity, immune infiltration, and survival in patients with colorectal cancer. This contributes to the understanding of m6A/m5C/m1A/m7G methylation and may offer potential approaches to the treatment of colorectal cancer.