NAMPT shows tissue-dependent immune-metabolic associations in osteoarthritic synovium and aging-related skeletal muscle.

Introduction: Nicotinamide phosphoribosyltransferase (NAMPT) is a key immune-metabolic regulator linking nicotinamide adenine dinucleotide (NAD) salvage, inflammatory signaling, and tissue stress responses, but whether it occupies comparable transcriptomic contexts in osteoarthritic synovium and skeletal muscle remains unclear. Methods: We performed a retrospective multi-cohort transcriptomic analysis using public GEO datasets with separate discovery and external supporting analysis stages. Discovery analyses included knee osteoarthritis synovium cohorts GSE55235 and GSE55457 and aging-related skeletal muscle proxy cohorts GSE38718 and GSE25941. Muscle-side findings were interpreted with explicit separation between aging-related skeletal muscle proxy cohorts and the clinically phenotyped sarcopenia external-support cohort. NAMPT abundance was evaluated by differential expression analysis and effect-size synthesis, immune and metabolic pathway activity was quantified by single-sample gene set enrichment analysis, and disease-adjusted partial correlations were used to define discovery-stage association structure. External supporting analyses were then performed in GSE12021, GSE254682, and the clinically phenotyped sarcopenia cohort GSE111016 to assess whether the direction of NAMPT abundance differences was reproduced in independent disease-relevant cohorts. Results: Across the discovery stage, NAMPT showed a downward abundance signal in both tissue contexts, although meta-analytic pooling in knee osteoarthritis synovium was limited by between-study heterogeneity. External datasets provided partial support for lower NAMPT expression in disease-relevant synovium in GSE12021 and in clinically phenotyped sarcopenia muscle in GSE111016, whereas GSE254682 showed the same direction without statistical significance. Discovery-stage disease-adjusted coupling analyses indicated that osteoarthritic synovium retained stronger NAMPT associations with inflammatory and glycolytic programs, whereas aging-related skeletal muscle proxy cohorts showed weaker immune coupling and a pattern more consistent with metabolic deficit. Cohort-stratified stability analyses suggested that these associations were directionally consistent in osteoarthritic synovium but weaker and more heterogeneous in the muscle proxy cohorts. A reduced external association analysis showed partial module-level consistency with this contrast, including preserved inflammatory linkage in GSE254682 and a more oxidative-phosphorylation-oriented pattern in GSE111016. Exploratory cohort-level separation analyses and therapeutic inference were retained as secondary analyses and were not interpreted as evidence of clinical utility or efficacy. Discussion: These findings suggest reduced NAMPT abundance with partial external support together with observational discovery-stage evidence of tissue-dependent immune-metabolic association patterns, and they favor context-specific interpretation and targeted experimental follow-up rather than uniform extrapolation across tissues.