Plasma Proteomic Profiling Identifies Biomarkers and Therapeutic Targets for Predicting Osteoporosis Risk in Patients with Inflammatory Bowel Disease.

BACKGROUND/AIMS: Inflammatory bowel disease (IBD) is a systemic disorder that affects not only the gastrointestinal tract but also extraintestinal organs and is associated with a significantly increased risk of osteoporosis. This study aims to investigate the association between plasma proteomic profiles and the subsequent risk of osteoporosis in patients with IBD. MATERIALS AND METHODS: Using data from 52 953 participants in the UK Biobank Pharma Proteomics Project, overlapping plasma proteins associated with both IBD and osteoporosis using Cox regression analysis were identified. Among these proteins, those with high predictive importance for IBD-osteoporosis multimorbidity were selected using machine learning and multistate modeling. Single-cell- type expression analysis was conducted to evaluate the cell type-specific enrichment of the candidate predictive protein-coding genes in the intestine. RESULTS: Over a median follow-up of 13.6 years, 339 proteins were significantly associated with IBD and 183 with osteoporosis, with 104 proteins overlapping between the 2 conditions. Machine learning and multistate modeling narrowed these down to 8 candidate predictive proteins (PLAUR, TNFRSF10A, SPINK1, HLA_E, CHGA, SPP1, TNFRSF10B, and IGFBP2) for IBD-osteoporosis multimorbidity. Exploratory single-cell expression analysis of 5 available candidate genes characterized their intestinal expression patterns, providing supportive biological insights. CONCLUSION: The study delineates the proteomic signatures associated with IBD-osteoporosis multimorbidity, identifies specific proteins associated with both conditions, highlights key molecular players linking the diseases, and identifies both established and potentially novel candidate biomarkers.