Gout is a common metabolic disease caused by chronic urate deposition. Although hyperuricemia is a well-established risk factor, only a subset of individuals with elevated serum urate eventually develop gout. Moreover, among patients with gout, recurrence is frequent and difficult to predict. Current clinical markers (e.g., serum urate, C-reactive protein) lack sufficient accuracy for early prediction and recurrence monitoring.
This project aims to leverage the unique breadth of the UK Biobank to build data-driven risk prediction models for gout. Specifically, we will:
Early risk prediction: Use genetic variants (e.g., SLC2A9, ABCG2), proteomic and metabolomic markers, and routine biochemical/clinical variables to evaluate individual risk of progression from hyperuricemia to gout. Polygenic risk scores (PRS) and multi-omics signatures will be tested for incremental predictive value beyond serum urate.
Recurrence risk assessment: Among participants with documented gout, evaluate the role of urate fluctuation, renal function, inflammatory markers, medication adherence (e.g., allopurinol, febuxostat use), and multi-omics profiles in predicting recurrence. Time-to-event and dynamic models (e.g., Cox regression, machine learning approaches) will be employed.
Comorbidity burden: Explore associations between gout and its major comorbidities (cardiovascular disease, chronic kidney disease, diabetes) using longitudinal hospital admission records and biomarkers, to identify shared molecular pathways.
Scientific rationale: UKB provides large-scale and harmonized genomic, proteomic, metabolomic, lifestyle, and hospital record data, uniquely enabling integrative multimodal analyses. By focusing solely on UKB resources, we can establish robust, generalizable models of early gout risk and recurrence. The findings are expected to improve understanding of the molecular determinants of gout progression, enhance patient stratification, and guide preventive interventions.
