Distinct Metabolic Signatures Linked to High-Resolution Computed Tomography Radiographic Phenotypes in Stable and Progressive Fibrotic Lung Disease

Abstract Rationale and Objectives This study aimed to identify distinct metabolic signatures associated with disease progression by integrating high-resolution computed tomography (HRCT) visual scoring with comprehensive metabolomic profiling. Materials and Methods This single-center, cross-sectional study enrolled 60 idiopathic pulmonary fibrosis/interstitial lung disease (IPF/ILD) patients from December 2021 to October 2022. Participants underwent standardized pulmonary function testing, HRCT imaging, and peripheral blood collection for metabolomic analysis using one-dimensional hydrogen nuclear magnetic resonance ( 1 H NMR) spectroscopy and ultra-high-performance liquid chromatography coupled to tandem mass spectrometry (UPLC-MS/MS). A comprehensive HRCT scoring system quantified total HRCT composite scores. Linear regression analysis integrated radiographic scores with metabolomic profiles, adjusted for multiple covariates Results Stable IPF/ILD exhibited moderate negative correlations between the six most significant metabolites and HRCT scores (r = −0.27 to −0.51), along with a high abundance of specific phospholipids (triacylglycerol, monoacylglycerol, phosphatidylglycerol, phosphatidylethanolamine, diacylglycerol), sphingomyelin, ceramide, and acylcarnitine. In contrast, progressive disease showed weak positive correlations between the six most significant metabolites and HRCT scores (r = 0.19–0.26), and moderate negative correlation between specific triacylglycerol species and HRCT scores (r = −0.37-0.4). Furthermore, metabolomic analysis in individuals with progressive disease revealed both high and low abundances of specific phospholipid species (including high and low triacylglycerol species, as well as low levels of phosphatidylglycerol, phosphatidylethanolamine, phosphatidylcholine, phosphatidylserine, and phosphatidylinositol), along with high levels of certain sphingomyelin, ceramide, taurine, and purine bases, and low levels of xanthine and lactic acid observed. Conclusion Integration of systematic HRCT visual scoring with metabolomic profiling successfully differentiated stable from progressive IPF/ILD through distinct molecular-radiographic signatures. Funding Supporting Effective Educator Development (SEED) grant