Gut-Liver-Pancreas Axis-Derived Composite Index for Glycemic Stratification in Type 2 Diabetes - A Cross-Sectional Study

Background: Dysregulated gut-derived metabolites such as bile acids (BA) and short-chain fatty acids (SCFAs) have been implicated in altered glycemic control and insulin dynamics in type 2 diabetes. Understanding their combined diagnostic utility may improve metabolic causes and therapeutic strategies. Objective: To evaluate the discriminative power of BA, SCFAs, and a composite gut-liver-pancreas axis-based index in predicting poor glycemic control, and to explore their clustering behaviour using Principal Component Analysis (PCA) in individuals stratified by HbA1c status. Methods: Serum samples were analysed from individuals with varying glycemic control (HbA1c >8% vs ≤8%) for the quantification of BA, SCFAs, and insulin levels were. A composite metabolic index was computed as (Bile Acids × Insulin) / SCFA. PCA was performed to visualize clustering of metabolic signatures. Receiver Operating Characteristic (ROC) curve analysis evaluated the predictive accuracy of individual and composite markers. Regression analysis identified significant predictors of postprandial insulin levels. Results: The composite index demonstrated the highest discriminative performance (AUC = 0.87), outperforming bile acids (AUC = 0.73) and SCFAs (AUC = 0.61) in identifying individuals with poor glycemic control. PCA revealed distinct separation of glycemic subgroups, with insulin and the composite index vectors aligned along PC1, while SCFA contributed orthogonally, indicating inverse metabolic association. Multivariable regression identified HbA1c (β = +3.19, p = 0.02), bile acids (β = +1.62, p = 0.01), and the composite index (β = +0.92, p = 0.005) as positive predictors of insulin, whereas SCFAs were negatively associated (β = -2.24, p = 0.03). Conclusion: The gut-liver-pancreas axis-derived composite index offers superior predictive value for glycemic status compared to individual metabolites. Its integration into metabolic profiling may enhance early identification of dysglycemia and guide targeted interventions.