Protocol for Developing and Validating a Multimarker-Clinical Prediction Model of SGLT2 Inhibitor-Induced Acute eGFR...

Area of Science:

• Nephrology
• Proteomics
• Biomarker Discovery

Background:

• Sodium-glucose cotransporter 2 (SGLT2) inhibitors are effective in reducing renal and cardiovascular events.
• An acute decline in estimated glomerular filtration rate (eGFR) is a common early side effect of SGLT2 inhibitors.
• Predictors of this eGFR dip show significant heterogeneity across studies.

Purpose of the Study:

• To identify urinary protein biomarkers predicting the early eGFR dip after SGLT2 inhibitor initiation.
• To develop a clinically actionable prediction model integrating novel biomarkers with routine variables.
• To improve risk stratification for patients initiating SGLT2 inhibitors.

Main Methods:

• A three-stage study (retrospective discovery, prospective internal validation, external validation) with ~600-700 participants.
• Proteomic screening using DIA mass spectrometry to identify urinary proteins associated with ≥10% eGFR decline at 1 month post-SGLT2i initiation in CKD stages 3-4.
• A LASSO-logistic regression model integrating top proteins with clinical variables (age, BMI, diabetes, heart failure, SBP, baseline eGFR, diuretic use).

Main Results:

• The study aims to identify specific urinary proteins linked to the early eGFR dip.
• A prediction model will be developed and validated using established statistical metrics (C-statistic, NRI, IDI).
• The model's calibration and clinical utility will be rigorously assessed.

Conclusions:

• Identifying biomarkers for the SGLT2 inhibitor-associated eGFR dip is crucial for personalized medicine.
• An integrated prediction model may enhance clinical decision-making regarding SGLT2 inhibitor therapy.
• This research could lead to better management of chronic kidney disease patients on SGLT2 inhibitors.