HDNA methylation data-based molecular subtype classification related to the prognosis of patients with hepatocellular carcinoma

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

BACKGROUND

DNA methylation biomarkers have great potential in improving prognostic classification systems for patients with cancer. Machine learning (ML)-based analytic techniques might help overcome the challenges of analyzing high-dimensional data in relatively small sample sizes. This systematic review summarizes the current use of ML-based methods in epigenome-wide studies for the identification of DNA methylation signatures associated with cancer prognosis.

METHODS

We searched three electronic databases including PubMed, EMBASE, and Web of Science for articles published until 2 January 2023. ML-based methods and workflows used to identify DNA methylation signatures associated with cancer prognosis were extracted and summarized. Two authors independently assessed the methodological quality of included studies by a seven-item checklist adapted from 'A Tool to Assess Risk of Bias and Applicability of Prediction Model Studies (PROBAST)' and from the 'Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK). Different ML methods and workflows used in included studies were summarized and visualized by a sunburst chart, a bubble chart, and Sankey diagrams, respectively.

RESULTS

Eighty-three studies were included in this review. Three major types of ML-based workflows were identified. 1) unsupervised clustering, 2) supervised feature selection, and 3) deep learning-based feature transformation. For the three workflows, the most frequently used ML techniques were consensus clustering, least absolute shrinkage and selection operator (LASSO), and autoencoder, respectively. The systematic review revealed that the performance of these approaches has not been adequately evaluated yet and that methodological and reporting flaws were common in the identified studies using ML techniques.

CONCLUSIONS

There is great heterogeneity in ML-based methodological strategies used by epigenome-wide studies to identify DNA methylation markers associated with cancer prognosis. In theory, most existing workflows could not handle the high multi-collinearity and potentially non-linearity interactions in epigenome-wide DNA methylation data. Benchmarking studies are needed to compare the relative performance of various approaches for specific cancer types. Adherence to relevant methodological and reporting guidelines are urgently needed.

MJDs family members: Potential prognostic targets and immune-associated biomarkers in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most common gastrointestinal malignancies. It is not easy to be diagnosed in the early stage and is prone to relapse, with a very poor prognosis. And immune cell infiltration and tumor microenvironment play important roles in predicting therapeutic response and prognosis of HCC. Machado-Joseph domain-containing proteases (MJDs), as a gene family extensively involved in tumor progression, has pro-cancer and anti-cancer effects. However, the relationship between MJDs family members and immune cell infiltration and tumor microenvironment in HCC remains unclear. Therefore, cBio Cancer Genomics Portal (cBioPortal), The Cancer Genome Atlas (TCGA), UALCAN, Human Protein Atlas (HPA), MethSurv, and Tumor Immune Estimation Resource (TIMER) databases were performed to investigate the mRNA expression, DNA methylation, clinicopathologic features, immune cell infiltration and other related functions of MJDs family members in HCC. The results indicated that the expression of ATXN3, JOSD1, and JOSD2 was dramatically increased in HCC tissues and cell lines, and was correlated with histological grade, specimen type, TP53 mutation, lymph node metastatic, gender, and age of patients with HCC. Meanwhile, these genes also showed clinical value in improving the overall survival (OS), disease-specific survival (DSS), progression free survival (PFS), and relapse-free survival (RFS) in patients with HCC. The prognostic model indicated that the worse survival was associated with overall high expression of MJDs members. Next, the results suggested that promotor methylation levels of the MJDs family were closely related to these family mRNA expression levels, clinicopathologic features, and prognostic values in HCC. Moreover, the MJDs family were significantly correlated with CD4⁺ T cells, CD8⁺ T cells, B cells, neutrophils, macrophages, and DCs. And MJDs family members’ expression were substantially associated with the levels of several lymphocytes, immunomoinhibitors, immunomostimulators, chemokine ligands, and chemokine receptors. In addition, the expression levels of MJDs family were significantly correlated with cancer-related signaling pathways. Taken together, our results indicated that the aberrant expression of MJDs family in HCC played a critical role in clinical feature, prognosis, tumor microenvironment, immune-related molecules, mutation, gene copy number, and promoter methylation level. And MJDs family may be effective immunotherapeutic targets for patients with HCC and have the potential to be prognostic biomarkers.

Epigenetic modifications precede molecular alterations and drive human hepatocarcinogenesis

Development of primary liver cancer is a multi-stage process. Detailed understanding of sequential epigenetic alterations is largely missing. Here, we performed Infinium Human Methylation 450k BeadChips and RNA sequencing analyses for genome-wide methylome and transcriptome profiling of cirrhotic liver (n=7), low- (n=4) and high-grade (n=9) dysplastic lesions, early (n=5) and progressed (n=3) hepatocellular carcinomas (HCC) synchronously detected in eight HCC patients with chronic hepatitis B infection. Integrative analyses of epigenetically driven molecular changes were identified and validated in two independent cohorts comprising 887 HCC. Mitochondrial DNA sequencing was further employed for clonality analyses and indicates multi-clonal origin in the majority of investigated HCC. Alterations in DNA methylation progressively increased from CL to dysplastic lesions and reached a maximum in early HCC. Associated early alterations identified by IPA pathway analyses involved apoptosis, immune regulation and stemness pathways, while late changes centered on cell survival, proliferation and invasion. We further validated putative 23 epi-drivers with concomitant expression changes and associated with overall survival. Functionally, Striatin 4 (STRN4) was demonstrated to be epigenetically regulated and inhibition of STRN4 significantly suppressed tumorigenicity of HCC cell lines.Overall, application of integrative genomic analyses defines epigenetic driver alterations and provides promising targets for novel therapeutic approaches.