Gastrointestinal adenocarcinoma analysis identifies promoter methylation-based cancer subtypes and signatures

Abstract

Gastric adenocarcinoma (GAC) and colon adenocarcinoma (CAC) are the most common gastrointestinal cancer subtypes, with a high incidence and mortality. Numerous studies have shown that its occurrence and progression are significantly related to abnormal DNA methylation, especially CpG island methylation. However, little is known about the application of DNA methylation in GAC and CAC. The methylation profiles were accessed from the Cancer Genome Atlas database to identify promoter methylation-based cancer subtypes and signatures for GAC and CAC. Six hypo-methylated clusters for GAC and six hyper-methylated clusters for CAC were separately generated with different OS profiles, tumor progression became worse as the methylation level decreased in GAC or increased in CAC, and hypomethylation in GAC and hypermethylation in CAC were negatively correlated with microsatellite instability. Additionally, the hypo- and hyper-methylated site-based signatures with high accuracy, high efficiency and strong independence can separately predict the OS of GAC and CAC patients. By integrating the methylation-based signatures with prognosis-related clinicopathologic characteristics, two clinicopathologic-epigenetic nomograms were cautiously established with strong predictive performance and high accuracy. Our research indicates that methylation mechanisms differ between GAC and CAC, and provides novel clinical biomarkers for the diagnosis and treatment of GAC and CAC.

Machine learning in the identification of prognostic DNA methylation biomarkers among patients with cancer: a systematic review of epigenome-wide studies

“…Fewer studies used other metrics, including C-index ( N =16) and receiver operating characteristic curves ( N = 39) to evaluate the discriminative power of their prognostic models. Only 18 studies (6, 31, 36, 39-42, 51, 54, 55, 57, 59, 66, 67, 78, 80, 84, 88) evaluated the calibration accuracy of prognostic models.…”

“…Among the 39 studies that adopted clustering-based workflows (5, 18-23, 29, 32-35, 37-39, 43, 45-47, 50, 53-57, 61, 62, 64, 69, 70, 75, 77-80, 83, 84, 87, 88), most started with a preliminary screening of CpG sites (Figure 4.1). The most frequently used pre-selection approach was selecting CpG sites that were significantly associated with prognosis by first performing univariable Cox regression analysis, and then performing multivariable Cox analysis with adjustment for clinical variables.…”

“…In contrast, four studies (23,33,34) (47) directly trained classification models with clusters as labels using Bayesian networks or random forest, while 17 studies constructed prognostic models (37-39, 45, 46, 53-57, 64, 69, 70, 75, 77, 78, 80). Lastly, models developed in the previous stage were mostly internally validated by split-sample validation, albeit four studies validated the model externally in an independent dataset (33,50,57,78). As to studies using more than one supervised ML model for feature selection, four studies intersected CpGs selected by different ML techniques (9,31,48,84), whereas two studies only selected the biomarkers retained in one model with the best performance (72,85).…”

Identification of stem cell-related subtypes and risk scoring for gastric cancer based on stem genomic profiling

“…Recent studies have focused on the establishment of prognostic RS models based on protein-coding genes [ 35 ], non-coding genes (e.g., lncRNA, miRA, and circRNA) [ 58 ], and CpG island methylation sites [ 59 ] to evaluate survival outcomes in GC patients. As the generation of the model is in the preliminary research stage, there is still a lack of a widely accepted model for clinical applicability.…”

Transcriptional and Epigenetic Bioinformatic Analysis of Claudin-9 Regulation in Gastric Cancer

“…In relation to gastric cancer, the hypermethylation of the Claudin-11 promoter has been associated with increased invasive potential [95], and the hypermethylation of the Claudin-3 promoter is considered a predictor of poor prognosis in advanced gastric adenocarcinoma [96]. In fact, promoter hypermethylation and claudin expression have been associated with gastric cancer TNM stage [97,98].…”