Results of a recent meta-analysis, combined with the report from the 2012 workshop, GFR Decline as an Endpoint for Clinical Trials in CKD, and three reviews from the workshop analytical group, provide support for the use of lesser declines in estimated glomerular filtration rate (eGFR) as a surrogate end point in trials of progression of chronic kidney disease (CKD).
That was the primary finding of the meta-analysis conducted by Lesley A. Inker, MD, MS, and colleagues. Analysis results were reported in the American Journal of Kidney Diseases [2014;64(6): 848-859].
In late 2012, the National Kidney Foundation and the US Food and Drug Administration sponsored the scientific workshop to assess data that might support new definitions of GFR decline in clinical trials of therapies related to CKD. The clinical trial meta-analysis reported here was designed to assess the consistency of treatment effects on the established and alternative end points.
For trials of progression kidney disease, alternative end points for trials of lesser decline in GFR may increase the number of end point events during the trials, perhaps leading to more precise estimates of the treatment effects, providing greater statistical power, and/or enabling trials with shorter time periods or smaller sample sizes. The current FDA accepted end point for clinical trials of progression of kidney disease is kidney failure and halving of GFR, defined as doubling of serum creatinine level (equivalent to a 57% decline in eGFR).
The current diagnostic test study included 9488 participants from 37 randomized controlled trials of progression of CKD across five intervention types. The index test was alternative end points including 20%, 30%, 40%, and 57% change in eGFR from baseline throughout duration of the study and during follow-up of 12, 18, and 24 months. The reference test was the historically established end point of time to composite of treated kidney failure (end-stage renal disease [ESRD]), untreated kidney failure (GFR <15 mL/min/1.73 m2), or doubling of serum creatinine level throughout the duration of the study.
Among the studies included in the analysis, median follow-up was 3.62 years. During that time period, there were 3070 established events, including 2029 cases of ESRD, 1151 cases of eGFR <15 mL/min/1.73m2, and 2086 doubling of serum creatinine. There were differences in results among the interventions based on participant and study characteristics; eg, eGFRs were lower at baseline in studies of renin-angiotensin system (RAS) blockade and studies of RAS blockade had shorter duration than studies of immunosuppressive therapy.
The median time from initial visit to the second visit for the confirmed end point was 3.16 months. Generally, compared with the established end point, the number of confirmed alternative end points was greater for smaller versus larger declines in eGFR and longer versus shorter follow-up intervals. For the overall study duration, compared with the established end point, there were approximately 30% more end points for the 40% decline and 60% more for the 30% decline, whereas at 24 months, the 30% decline continued to result in more end points and the 40% decline resulted in fewer end points compared to the established end point during the overall study duration.
With end points defined by a lesser decline in eGFR, there was a general trend toward attenuation of the treatment effect. The attenuation was greater with nonconfirmed end points, with the exception of the intervention with the low-protein diet, for which there was an association with a stronger treatment effect.
The range of the ratio of the hazard ratio for the alternative to established end point for the five types of interventions was 0.91 to 1.12 for 40% decline and from 0.88 to 1.15 for 30% decline for the overall study duration. This result indicated a consistency of treatment effects, the researchers said.
Limitations to the study cited by the authors included limiting the analyses to the specific diseases and interventions included in the published and unpublished literature available in 2007, the limitation of statistical power created by the sample size and variation in treatment effects among CKD trials, and the confirmed results was not a planned visit as part of the study design, but was the next available visit and may have occurred long after the original visit.
“in summary, the results presented here, together with those of the workshop report and the other three articles from the workshop analytical group, suggest that a confirmed eGFR decline of 40% and possibly 30% may be used as an end point for a trial of kidney disease progression in certain circumstances. Acute effects of interventions of GFR, nonproportional effects of interventions on GFR decline, and effects of interventions on the endogenous filtration markers used to estimate GFR could affect the utility of one or both of these end points, depending on the magnitude of the effect. Exploration as to the existence of these factors should be undertaken early in the design of drug development and potentially could be evaluated during phase 3 trials that use eGFR decline as an end point,” the researchers said.
- For trials of progression kidney disease, alternative end points for trials of lesser decline in GFR may increase the number of end point events during the trials, perhaps leading to more precise estimates of the treatment effects.
- Generally, compared with the established end point, the number of confirmed alternative end points was greater for smaller versus larger declines in eGFR and longer versus shorter follow-up intervals.
- Results of a recent meta-analysis, combined with the report from the 2012 workshop, GFR Decline as an Endpoint for Clinical Trials in CHD, and three reviews from the workshop analytical group, provide support for the use of lesser declines in eGFR as a surrogate end point in trials of progression of CKD.