Construction and validation of a prognostic model of angiogenesis-related genes in multiple myeloma

• Published in: BMC cancer Vol. 24; no. 1; pp. 1269 - 15
• Main Authors: Hu, Rui, Chen, Fengyu, Yu, Xueting, Li, Zengzheng, Li, Yujin, Feng, Shuai, Liu, Jianqiong, Li, Huiyuan, Shen, Chengmin, Gu, Xuezhong, Lu, Zhixiang
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 11.10.2024; BioMed Central; BMC
• Subjects: Angiogenesis; Angiogenesis-related genes; Biomarkers, Tumor - genetics; Databases, Genetic; Evaluation; Gene Expression Profiling; Gene Expression Regulation, Neoplastic; Gene Ontology; Gene Regulatory Networks; Genetic aspects; GEO; Health aspects; Humans; Immune microenvironment; Mathematical models; Medical research; Medicine, Experimental; Multiple myeloma; Multiple Myeloma - genetics; Multiple Myeloma - pathology; Neovascularization; Neovascularization, Pathologic - genetics; Prognosis; China; GEO; Immune microenvironment; Prognosis; Angiogenesis-related genes; Multiple myeloma
• ISSN: 1471-2407; 1471-2407
• DOI: 10.1186/s12885-024-13024-9

Angiogenesis is associated with tumour growth, infiltration, and metastasis. This study aimed to detect the mechanisms of angiogenesis-related genes (ARGs) in multiple myeloma (MM) and to construct a new prognostic model. MM research foundation (MMRF)-CoMMpass cohort, GSE47552, GSE57317, and ARGs were sourced from public databases. Differentially expressed genes (DEGs) in the tumour and control cohorts in GSE47552 were determined through differential expression analysis and were enriched with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. Weighted gene coexpression network analysis (WGCNA) was applied to derive modules linked to the ARG scores and obtain module genes in GSE47552. Differentially expressed ARGs (DE-ARGs) were selected for subsequent analyses by overlapping DEGs and module genes. Furthermore, prognostic genes were selected using univariate Cox and least absolute shrinkage and selection operator (LASSO) regression analyses. Depending on the prognostic genes, a risk model was constructed, and risk scores were determined. Moreover, MM samples from MMRF-CoMMpass were sorted into high- and low-risk teams on account of the median risk score. Additionally, correlations among clinical characteristics, gene set variation analysis (GSVA), gene set enrichment analysis (GSEA), immune analysis, immunotherapy predictions and the mRNA‒miRNA‒lncRNA network were carried out. A total of 898 DEGs, 211 module genes, 24 DE-ARGs and three prognostic genes (AKAP12, C11orf80 and EMP1) were selected for this study. Enrichment analysis revealed that the DEGs were related to 86 GO terms, such as 'cytoplasmic translation', and 41 KEGG pathways, such as 'small cell lung cancer'. A prognostic gene-based risk model was created in MMRF-CoMMpass and confirmed with the GSE57317 dataset. Moreover, a nomogram was established on the basis of independent prognostic factors that have proven to be good predictors. In addition, the immune cell infiltration results suggested that memory B cells were enriched in the high-risk group and that immature B cells were enriched in the low-risk group. Finally, the mRNA‒miRNA‒lncRNA network demonstrated that hsa-miR-508-5p was tightly associated with EMP1 and AKAP12. RT‒qPCR was used to validate the expression of the genes associated with prognosis. A new prognostic model of MM associated with ARGs was created and validated, providing a new perspective for exploring the connection between ARGs and MM.