Urinary Proteomics to Guide Early Intervention to Prevent Complications in Type 2 Diabetes - a Feasibility Study

Title: Body fluid proteome SIGnatures for persoNALised intervention to prevent cardiovascular and renal complications in diabetes. Aim: To explore the feasibility of using urinary proteomic risk scores in clinical practice to identify patients at risk of developing end organ damage and identify which patients should receive additional renocardiovascular protective treatment.

Background: Diabetes and its associated complications impose a significant burden on both patients and societies. Despite advancements in lowering blood glucose, the elevated risk of developing cardiovascular disease (CVD) and chronic kidney disease (CKD) remains a pressing concern, underscoring the need for optimized prevention strategies and improved therapeutic options. Recent developments in glucose-lowering drugs, such as sodium-glucosecotransporter- 2-inhibitors (SGLT2-i) and glucagon-like-peptide-1 receptor agonists (GLP1-RA), as well as the use of the non-steroidal mineralocorticoid receptor antagonist (nsMRA) finerenone, have shown promising cardiovascular and renal protection. Currently, there is no reliable method for predicting personalized treatment responses in diabetic complications. Consequently, benefits of treatment are under dispute, due to a large number of patients not responding. The use of SGLT2-i, nsMRA and GLP1-RA in CKD has happened largely in parallel, all agents have demonstrated benefit, but it is not yet clear how to prioritize between the drugs or if all should be combined. This study builds upon previous scientific work that have investigated the urine proteome and identified several biomarkers able to predict early diabetes associated complications. CKD273 urine proteomic risk score is a well-established tool used to predict the risk of chronic kidney disease (CKD) progression. CAD160 is urine proteomic risk score to predict the risk of coronary artery disease (CAD). HF2 urine proteomic classifier is used to predict the risk of heart failure (HF). Urine sample analysis is based on capillary electrophoresis coupled with mass spectrometry (CE-MS) to determine these risk scores. Urine proteomic scores are continous numerical values. Higher score means that the urinary peptide pattern is more similar to that of patients with progressive disease. A lower score indicates a peptide profile more typical of healthy individuals. In addition a Support Vector Machine (SVM), a supervised machine learning algorithm will perform in silico treatment simulations and calculate the change in classification scores for 3 different potential interventions: GLP1-RA semaglutide, SGT2-i dapagliflozin and GLP1-RA finerenone. Based on these changes (with the largest beneficial change indicating the most effective treatment), the most suitable intervention can be selected and the participent will be allocated. Design: Single-centre, open-label, parallel group (intervention group) with 6 months intervention. Population: Type 2 diabetes without history of heart failure NYHA Class IV or advanced diabetic kidney disease with an estimated Glomerular Filtration Rate (eGFR) \< 30 ml/min/1.73m2 or urinary albumin creatinine ratio (UACR) \> 200 mg/g. Objectives: To assess the feasibility of using proteomic classifiers in clinical practice for response prediction in a prospective study. We will use urinary proteomic classifiers: CKD273, CAD160 and HF2 to identify patients suited for additional medical treatment with sodium-glucose-cotransporter-2 (SGLT2)- inhibitors, glucagon-like-peptide-1 GLP-1 receptor agonists or non-steroidal mineralocorticoid receptor antagonist. Interventions: The SGLT2 inhibitor dapagliflozin 10 mg daily, the nsMRA finerenone 10-20 mg daily, and the GLP-1 receptor agonist semaglutide 0.25-1.0 mg once weekly. The medication will be given stepwise according to a prespecified algorithm and guided by the response on UACR. Endpoints: Primary endpoint is feasibility of using urinary proteomic classifiers in clinical practice, while secondary endpoints are changes in UACR and urinary proteomic signatures after 6 months of treatment. Time schedule: The study is expected to start inclusion June 1st 2025. The recruitment period is 6 months, the intervention period is 6 months and hence the study is expected to be terminated May 31st 2026.