Baseline Risk and Longitudinal Changes in kidneyintelX.dkd and Its Association With Kidney Outcomes in the CANVAS and CREDENCE Trials

OBJECTIVE: We evaluated the prognostic and clinical utility of kidneyintelX.dkd, a biomarker-based risk score, in patients with type 2 diabetes and a broad range of chronic kidney disease (CKD) by assessing its association with kidney outcomes at baseline and longitudinally, comparing it with the established Kidney Disease Improving Global Outcomes (KDIGO) risk classification, and examining its responsiveness to canagliflozin. RESEARCH DESIGN AND METHODS: We measured tumor necrosis factor receptor-1 (TNFR-1), TNFR-2, and kidney injury molecule-1 (KIM-1) in banked plasma samples at baseline and year 1 and calculated kidneyintelX.dkd scores of participants with CKD G1-G3b from two large randomized controlled trials (Canagliflozin Cardiovascular Assessment Study CANVAS and Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation CREDENCE). We assessed concordance between KDIGO and kidneyintelX.dkd risk levels, evaluated associations of baseline and 1-year changes in kidneyintelX.dkd with kidney outcomes, and examined treatment effects of canagliflozin versus placebo. RESULTS: Mean kidneyintelX.dkd scores increased across higher KDIGO risk categories, but individual-level differences revealed improved risk reclassification. The kidneyintelX.dkd score was independently associated with kidney outcomes and more strongly predictive than KDIGO classification. At 1 year, canagliflozin significantly lowered kidneyintelX.dkd score versus placebo, and longitudinal reductions by 1 year were associated with lower subsequent risk of kidney outcomes, independent of changes in estimated glomerular filtration rate or urinary albumin-to-creatinine ratio. Absolute risk reductions with canagliflozin were largest among those at high kidneyintelX.dkd risk. CONCLUSIONS: The kidneyintelX.dkd score adds prognostic value beyond clinical classification, reflects canagliflozin treatment response, and helps identify individuals most likely to benefit from therapy. These findings support a role for the kidneyintelX.dkd score in personalized risk assessment and monitoring in type 2 diabetes and CKD in prospective studies and clinical practice.