Innovative Clinical Trial Designs for Assessing Treatment Effects on Kidney Function
Glomerular filtration rate (GFR) decline is a critical surrogate endpoint for evaluating kidney failure. However, interventions aimed at slowing chronic kidney disease (CKD) progression can cause acute GFR reductions, complicating the assessment of their long-term benefits. To address this, researchers have explored the utility of two alternative trial designs: the wash-out design and the active run-in randomized withdrawal design. These approaches aim to exclude the impact of acute effects, providing a clearer picture of long-term treatment benefits.
Post-hoc analyses of two clinical trials were conducted to evaluate these designs. The first trial, EMPA-REG Outcome, tested empagliflozin against placebo and included a wash-out period. The second trial, SONAR, tested atrasentan against placebo and featured an active run-in period with a randomized withdrawal. The studies compared the drug effect on GFR decline from the first on-treatment visit to the end of treatment (chronic slope in a standard randomized trial design) with GFR changes calculated from randomization to the end of the wash-out period or from treatment-specific baseline GFR values until the end of treatment.
Findings revealed that the effect of empagliflozin versus placebo on chronic GFR slope was 1.72 mL/min/1.73 m2/year, closely matching the total GFR decline from baseline to the end of the wash-out period. Similarly, the effect of atrasentan versus placebo on chronic GFR slope was 0.72 mL/min/1.73 m2/year, akin to the total slope estimated from treatment-specific baseline GFR values. Both designs demonstrated superior statistical power compared to the standard randomized design, affirming their appropriateness for computing treatment effects on GFR decline.
This research underscores the importance of innovative trial designs in accurately assessing the efficacy of treatments for CKD, offering valuable insights for future clinical trials and therapeutic strategies.