Inflammation-related Genes as Potential Shared Biomarkers for Metabolic Dysfunction-associated Fatty Liver Disease and Acute Pancreatitis.

Background

People with Acute Pancreatitis (AP) who also have Metabolic Dysfunctionassociated Fatty Liver Disease (MAFLD) are more likely to experience worse consequences. However, the mechanisms that link MAFLD and AP are not fully understood. Our study identified shared biomarkers utilizing bioinformatics and machine learning techniques.

Methods

We selected datasets for AP and MAFLD from the GEO database. Differentially Expressed Genes (DEGs) were identified from the AP dataset. Weighted Gene Co-expression Network Analysis (WGCNA) was conducted on the MAFLD dataset to identify the module highly correlated with the disease. Subsequently, we obtained shared genes by taking the intersection of the DEGs and the module genes. The shared genes were analyzed for GO and KEGG enrichment. A Protein-Protein Interaction (PPI) network and machine learning techniques were used to identify diagnostic genes, which were evaluated for diagnostic efficacy using ROC curves.

Results

The AP dataset yielded 454 upregulated and 380 downregulated DEGs. WGCNA identified 715 module genes in MAFLD, producing 42 shared genes. Enrichment analyses implicated inflammatory responses, calcium signaling, and lipopolysaccharide-related immune pathways. FPR1 and S100A9 emerged as the final diagnostic candidates, with AUC values exceeding 0.7 across all datasets.

Discussion

FPR1 and S100A9 are involved in immune activation, inflammatory signaling, and oxidative stress, suggesting roles in the pathogenesis of both AP and MAFLD. MAFLD may indirectly worsen AP severity through inflammatory and lipid pathways.

Conclusion

FPR1 and S100A9 are promising common biomarkers for AP and MAFLD, and may provide important insights into common mechanisms and therapeutic opportunities.