Exploring The Functional Role Of FAM81A In Pancreatic Cancer Based On Machine Learning Identification Of Tumor Stemness

Background and Purpose:Pancreatic cancer,one of the deadliest malignancies with a steadily increasing global incidence and mortality rate,often remains undiagnosed until advanced stages due to limited understanding of its genetic evolution and the pancreas’s concealed anatomical location,resulting in poor treatment outcomes.Tumor stem cells are critical factors in the progression,relapse,metastasis,and chemoresistance of pancreatic cancer;however,extant research indicates that various cells within the tumor microenvironment possess potential stem cell-like characteristics,challenging the conventional unidirectional hierarchical model of cancer stem cells and contributing to the suboptimal efficacy of diverse therapeutic strategies.Consequently,there is an urgent need for biomarkers capable of identifying tumor stemness.As one of the earliest studied epigenetic modifications,DNA methylation is believed to play a pivotal role in inducing and maintaining specific cancer cell stemness properties.Thus,investigating the interplay and regulatory mechanisms between DNA methylation and tumor stemness features in pancreatic cancer may provide novel insights and approaches for its treatment.Materials and Methods:In this study,based on the mRNA expression profiles and DNA methylation data from the Cancer Genome Atlas(TCGA),we employed a particular logistic regression algorithm to quantify the stemness indices(mRNAsi and mDNAsi)of pancreatic cancer samples.Subsequently,we utilized weighted gene co-expression network analysis to identify modules most correlated with stemness(mRNAsi)and conducted Gene Ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway analyses to examine the enrichment functions and pathways of the characteristic modules.We then visualized the interaction networks of the characteristic modules using Cytoscape and further refined central methylation sites employing the Maximal Clique Centrality(MCC)scoring method.Ultimately,by performing survival and correlation analyses on central methylation sites and their target genes across multiple databases,we identified key genes.We employed correlation tests to investigate how DNA methylation regulatory factors modulate the expression of key genes and further explored their potential involvement in pancreatic cancer through enrichment analyses.Additionally,we analyzed the correlations between the key genes,potential regulatory factors involved in the pathways,and stemness markers.Subsequently,we employed CIBERSORT and Timer analyses to assess the impact of stemness features and key genes on the tumor immune microenvironment.Lastly,we conducted differential expression and survival analyses of the key genes in pan-cancer contexts.Results:This study quantified the mRNAsi and mDNAsi of 151 pancreatic cancer patients in the TCGA PAAD cohort using a particular logistic regression algorithm,revealing significant differences in stemness indices among distinct clinical-pathological and molecular features and a strong correlation with patient prognosis.Employing Weighted Gene Co-expression Network Analysis(WGCNA),we identified the most correlated module with mRNAsi,and through GO and KEGG analyses,discovered that these modules were primarily enriched in biological processes such as cell growth and differentiation,as well as pathways involving tumor cell proliferation,adhesion,and apoptosis.By applying the Maximal Clique Centrality(MCC)scoring method,we screened 20 relevant central DNA methylation sites and found FAM81A to be a key prognostic gene in pancreatic cancer through multi-database survival analyses.The expression of FAM81A was significantly correlated with DNA methyltransferases,demethylases,CDK1,PDK1,and stemness markers(NANOG,SOX2,KLF4).FAM81A was significantly enriched in pathways involving tumor cell cycle,cell adhesion molecules,and glycolysis/gluconeogenesis.Additionally,the tumor immune microenvironment in pancreatic cancer with high FAM81A expression exhibited a higher degree of immune cell infiltration.Furthermore,FAM81A demonstrated differential expression and prognostic value in various types of cancer.Conclusion:1.Multi-omics approaches revealed that the 3-UTR methylation site(cg2262574)may potentially mediate FAM81A mRNA expression regulation,influencing the stemness of pancreatic cancer cells.2.The key gene FAM81A holds significant prognostic value in pancreatic cancer,exhibits a strong correlation with mRNAsi,and participates in various cellular processes and pathways,potentially playing a crucial role in the onset,progression,and prognosis of pancreatic cancer.3.FAM81A expression in pancreatic cancer is significantly associated with CKD1 and stemness markers,potentially participating in the induction and maintenance of tumor stemness in pancreatic cancer through the regulation of the cell cycle pathway.4.FAM81A expression exhibits a significant correlation with TMB and tumor proliferative capacity,implying its potential involvement in the regulation of genomic stability and growth in tumors.5.FAM81A expression demonstrates a notable positive correlation with the abundance of multiple immune cell infiltrations,suggesting its potential influence on the tumor microenvironment and immune regulation.6.In pan-cancer analyses,FAM81A displays significant differential expression and prognostic value in eight types of cancer,indicating its potential importance across various malignancies.