Abstract: Objective To investigate the molecular subtyping of Kashin-Beck disease (KBD) based on metabolomic profiling and to elucidate the impact of gene-environment interactions on disease pathogenesis and its distinct subtypes. Methods A total of 106 patients with KBD and 106 healthy controls (recruited from both endemic and non-endemic areas) were enrolled in this study. Untargeted metabolomic profiling was conducted utilizing liquid chromatography-mass spectrometry (LC-MS). Molecular subtyping was executed via K-means clustering, while the specific characteristics of each subtype were delineated through multivariate statistical modeling and pathway enrichment analyses. Furthermore, multivariate logistic regression was employed to evaluate gene-environment interactions. Results The KBD cohort was stratified into three distinct metabolic subtypes. Subtype 1 is primarily characterized by lipid metabolism dysregulation; Subtype 2 exhibits aberrant oxidative stress and extracellular matrix degradation; and Subtype 3 is associated with the reprogramming of amino acid and energy metabolism. Interaction analyses demonstrated a significant synergistic effect between low environmental selenium exposure and the rs1050450(ProLeu) allele, an association that was particularly pronounced within Subtype 2 (P_interaction=0.006). Conclusion This study reveals the existence of heterogeneous KBD subtypes from the perspective of metabolic phenotypes for the first time. By establishing a comprehensive "environment-genetics-metabolic phenotype" association framework, this research provides a novel scientific basis for the precision subtyping and early risk prognostication of KBD.

Key words: Kashin-Beck disease, Metabolomics, Molecular subtyping, Gene-environment interaction, Lipid metabolism, Oxidative stress