GFR decline as an alternative end point to kidney failure in clinical trials: a meta-analysis of treatment effects from 37 randomized trials.

BACKGROUND There is increased interest in using alternative end points for trials of kidney disease progression. The currently established end points of end-stage renal disease and doubling of serum creatinine level, equivalent to a 57% decline in estimated glomerular filtration rate (eGFR), are late events in chronic kidney disease (CKD), requiring large clinical trials with long follow-up. As part of a comprehensive evaluation of lesser declines in eGFR as alternative end points, we describe the consistency of treatment effects of intervention on the alternative and established end points in past trials. STUDY DESIGN Diagnostic test study. SETTING & POPULATION 9,488 participants from 37 randomized controlled trials of CKD progression across 5 intervention types. INDEX TEST Alternative end points including percentage change in eGFR from baseline (20%, 30%, 40%, and 57%) throughout study duration and to 12, 18, and 24 months. eGFR change confirmed versus nonconfirmed at the next visit. REFERENCE TEST The historically established end point of time to composite of treated kidney failure (end-stage renal disease), untreated kidney failure (GFR<15mL/min/1.73m(2)), or doubling of serum creatinine level throughout study duration. RESULTS Over a median of 3.62 years' follow-up, there were 3,070 established end points. Compared to the established end point, the number of alternative end points was greater for smaller versus larger declines in eGFR and longer versus shorter follow-up intervals. There was a general trend toward attenuation of the treatment effect with end points defined by a lesser eGFR decline, with greater attenuation with nonconfirmed end points, except for the low-protein-diet intervention, for which a stronger treatment effect was observed. The ratio (95% credible interval) of the HR for the alternative to established end point for the 5 intervention types ranged from 0.91 (0.64-1.43) to 1.12 (0.89-1.40) for 40% decline and from 0.88 (0.63-1.39) to 1.15 (0.88-1.54) for 30% decline for the overall study duration, indicating consistency of treatment effects. LIMITATIONS Limited variety of interventions tested and low statistical power for many CKD clinical trials. CONCLUSIONS These results provide some support for the use of lesser eGFR declines as a surrogate end point, with stronger support for the 40% than 30% decline.