Multimodal Machine Learning-Based Marker Enables Early Detection and Prognosis Prediction for Hyperuricemia.

Area of Science:

• Metabolic disorders
• Bioinformatics
• Machine Learning

Background:

• Hyperuricemia (HUA) is a prevalent metabolic disorder with a long asymptomatic phase, often leading to gout and other metabolic complications.
• Early detection and risk prediction for HUA and gout are critical for timely intervention and management.
• Existing diagnostic methods may not fully capture the complex interplay of genetic and clinical factors contributing to HUA and its sequelae.

Purpose of the Study:

• To develop and validate a novel in-silico quantitative marker for hyperuricemia (ISHUA) using a stacked multimodal machine learning model.
• To assess the performance of ISHUA in detecting HUA and predicting the risk of gout and related metabolic outcomes.
• To investigate the impact of lifestyle factors on gout and metabolic outcomes within different risk strata.

Main Methods:

• Integration of genetic and clinical data from large cohorts (UK Biobank and Nanfang Hospital).
• Development and validation of a stacked multimodal machine learning model to generate the ISHUA.
• Statistical analysis to evaluate ISHUA's association with gout and metabolic outcomes, and to assess the influence of lifestyle.

Main Results:

• The ISHUA model demonstrated robust performance in HUA detection across train, internal, and external test sets (AUCs ranging from 0.779 to 0.859).
• ISHUA effectively stratified individuals into low- and high-risk groups for gout (AUCs ~0.81) and identified increased susceptibility to metabolic outcomes in the high-risk group.
• Favorable lifestyle profiles were associated with significantly reduced hazard ratios for gout and other metabolic outcomes, particularly in the high-risk ISHUA group.

Conclusions:

• The multimodal machine learning-based ISHUA marker offers a powerful tool for personalized risk stratification of hyperuricemia, gout, and associated metabolic disorders.
• ISHUA facilitates early identification of individuals at high risk, enabling targeted preventive strategies.
• Lifestyle interventions are effective in mitigating gout and metabolic risks, even within high-risk populations identified by ISHUA, underscoring the importance of personalized, lifestyle-based management.