The shared biomarkers and molecular mechanisms of systemic lupus erythematosus and type 2 diabetes

Abstract

Systemic lupus erythematosus (SLE) and type 2 diabetes mellitus (T2DM) share inflammatory and metabolic disturbances, yet the molecular mechanism of their overlap remains unclear. This study used integrated bioinformatics to identify transcriptomic signatures and potential biomarkers common to both conditions. Gene expression profiles from publicly available datasets (SLE: 38 patients/32 controls; T2DM: 6 patients/6 controls; validation cohorts: 79/30 and 41/15, respectively) were analyzed to detect shared differentially expressed genes and co-expression modules. Functional enrichment, protein-protein interaction networks, and immune-cell composition analyses were performed. Diagnostic gene panels were constructed using random forest feature selection and logistic regression and evaluated through receiver operating characteristic analysis with external validation. A total of 551 shared differentially expressed genes were identified, enriched predominantly in type I interferon signaling, Toll-like/NOD receptor pathways, TNF signaling, necroptosis, and neutrophil extracellular trap formation. Across analytical methods, a 10-gene interferon-related hub (STAT1, IRF7, OAS1, OAS2, ISG15, MX2, IFI35, RSAD2, SAMD9, SAMD9L) genes demonstrated strong discriminative performance in both SLE and T2DM. A three-gene model further showed potential clinical utility (AUC 0.872–1.00 in discovery; 0.665–0.928 in validation).These signatures align with therapeutic axes in SLE (IFN/JAK-STAT) and intersect inflammatory-metabolic pathways in T2DM, supporting assayable biomarkers and compact diagnostic models that warrant validation in larger, medication-annotated cohorts.