Identification Of Candidate Prognostic Markers For Gallbladder Cancer Using Network-based Module Analysis

Gallbladder cancer is a rare and aggressive malignancy of the biliary tract with a dismal survival rate.Effective biomarkers and therapeutic targets are urgently needed.We analyzed gene expression profiles of GBC to identify differentially expressed genes(DEGs)and then used these DEGs to identify functional module biomarkers based on protein functional interaction(FI)networks.We further evaluated the module-gene protein expression and clinical significance with immunohistochemistry staining(IHC)from 80 GBC samples.Five functional modules were identified.Module 0 included classical cancer signaling pathways,such as Ras and PI3K-Akt;and modules 1-4 included genes associated with muscle cells,fibrinogen,extracellular matrix,and integrins,respectively.We validated the expression of LIFR,PIK3R1,and MMP12,which were hubs or functional nodes in modules.Compared with paired peritumoural tissues,we found that the expression of LIFR(P=0.002)and PIK3R1(P=0.046)proteins were significantly downregulated,and MMP12(P=0.006)was significantly upregulated.Further prognostic analysis showed that patients with low expression of LIFR had shorter overall survival than those with high expression(log-rank test,P=0.028),the same trend as for PIK3R1(P=0.053)and MMP12(P=0.006).Multivariate analysis indicated that expression of MMP12 protein(hazard ratio[HR]=0.429;95%confidence interval=[CI]0.198-0.930;P=0.032)was one of the significant independent prognostic factors for overall survival.We found a highly reliable FI network,which revealed LIFR,PIK3R1 and MMP12 as novel prognostic biomarker candidates for GBC.These findings could accelerate biomarker discovery and therapeutic development in this cancer.