Abstract
Background: Gastric cancer (GC) is one of the main mortality cause worldwide. Previously, we found Forkhead box protein (FOXM1) or Urokinase-type plasminogen activator (PLAU) are independent prognostic markers of GC. This study aims to explore the combining prognostic efficacy and the potential insights underlying additive effect of FOXM1 to PLAU in GC progression through in-silico analyses. Method: The expression of FOXM1 and PLAU were profiled in 33 cancer types using public data. A merged GC expression dataset containing 598 samples was used for evaluating prognostic significance of FOXM1/PLAU. Gene Set Enrichment Analysis (GSEA) was performed to elucidate the mechanisms underlying FOXM1/PLAU promoted GC progression. The Cancer Genome Atlas (TCGA) was used for analyzing the association between FOXM1/PLAU and tumor immune infiltration. Genomic and proteomic differences between FOXM1+PLAU+ and FOXM1-PLAU- groups were also computed using TCGA GC data. Drugs targeting FOXM1/PLAU associated gene expression pattern was analyzed using LINCs database. Results: FOXM1 and PLAU are overexpressed in 17/33 cancer types including GC. Kaplan-Meier analyses indicate that the FOXM1+PLAU+ subgroup have the worst prognosis, while FOXM1-PLAU- subgroup have the best survival. Bioinformatics analysis indicated that FOXM1+PLAU+ associated genes are enriched in TGF-beta, DNA repair and drug resistance signaling pathways; FOXM1 and PLAU expression are negatively correlated with tumor immune infiltration. Genomic and proteomic differences between FOXM1+PLAU+ and FOXM1-PLAU- groups were presented. Data mining from LINCs suggested several chemicals or drugs that could target the gene expression pattern of FOXM1+PLAU+ patients. Conclusion: FOXM1+PLAU+ can serve as effective prognostic biomarkers and potential therapeutic targets for GC. Due to the additive effect of these two genes, screening for drugs or chemicals that targeting the expression patterns PLAU+FOXM1+ subgroup may exert important clinical impact on GC management.