Identification and Validation of a Novel Prognostic Signature of Gastric Cancer Based on Seven Complement System-Related Genes: An Integrated Analysis.

Abstract

The complement system (CS) is linked to the progression of gastric cancer (GC), which has a high mortality rate, though its mechanisms in GC remain unclear. This study aims to identify CS-related prognostic genes with causal links to GC, and to investigate their mechanisms. The intersection between differentially expressed genes (DEGs) obtained from the TCGA-STAD dataset and CS-related genes (CRGs) was defined as differentially expressed CRGs (DCRGs). Prognostic genes with a causal association with GC (pCDCRGs) were sequentially identified via Mendelian randomization (MR) analysis and Cox and least absolute shrinkage and selection operator (LASSO) regression analyses, followed by expression analysis. A gene signature and a nomogram were then established based on pCDCRGs and independent prognostic factors. Subsequent analyses focused on functional enrichment, immune relevance, drug sensitivity, gene interactions, and molecular regulatory networks. Eventually, reverse transcription-quantitative PCR (RT-qPCR) was employed to validate expression of pCDCRGs. DCRGs were obtained from the intersection of 8,418 DEGs and 241 CRGs. Among 12 DCRGs with causal association (CDCRGs) with GC, 7 genes were identified as pCDCRGs, including FANCG, FANCF, F2R, C4BPA, SERPINF2, PROC, and CD59. Notably, CD59 was markedly highly expressed in the normal group, whereas the other genes were markedly highly expressed in the GC group. Afterward, an accurate pCDCRG signature was developed. Risk score, age, and stage were recognized as independent risk factors, and the constructed nomogram demonstrated strong predictive accuracy. Additionally, analyses indicated that these 7 pCDCRGs may influence GC by affecting pathways such as complement and coagulation cascades, immune cell infiltration, immune characteristics, immunotherapy responses, and drug sensitivity. These effects may be linked to gene interactions and the regulatory roles of lncRNAs like RMRP and miRNAs such as hsa-mir-613. RT-qPCR showed C4BPA, PROC, F2R, and SERPINF2 were markedly up-regulated, whereas CD59 was markedly down-regulated in GC tissues. This study identified seven complement system-related prognostic genes with causal links to GC, based on which we developed a highly predictive 7-pCDCRG signature, providing valuable insights for clinical prognostic prediction and immunotherapy in GC patients.