Kidney Week

Abstract: SA-OR087

Association between GFR Slope and Subsequent ESKD and the Impact of Adjustment for Measurement Error: Meta-Analysis of Observational Studies

Session Information

• New Considerations for Renoprotection Clinical Trials
  October 27, 2018 | Location: 1B, San Diego Convention Center
  Abstract Time: 05:54 PM - 06:06 PM

Category: CKD (Non-Dialysis)

• 1902 CKD (Non-Dialysis): Clinical, Outcomes, and Trials

Author

• Grams, Morgan, CKD Prognosis Consortium (CKD-PC), Baltimore, Maryland, United States

Morgan Grams, MD, PhD

Group or Team Name

• CKD Prognosis Consortium

Background

A recent NKF-FDA-EMA Workshop evaluated candidate surrogate endpoints for clinical trials to slow kidney disease progression, particularly among participants with relatively preserved baseline glomerular filtration rate (GFR). Here, we investigate difference in GFR slope as a candidate surrogate endpoint using observational studies assembled by the CKD-PC.

Methods

We evaluated the strength of association between GFR slope and the clinical endpoint of end-stage kidney disease (ESKD), including the impact of error in slope measurement and the length of the observation period in which slope is observed, using random-effects meta-analysis of 14 cohorts including 3,373,368 participants followed for a mean of 4.4 years.

Results

In covariate-adjusted analyses, a reduction in eGFR decline of 0.75 ml/min/1.73 m²/year was protective for ESKD in participants with baseline GFR ≥60 ml/min/1.73 m² (adjusted hazard ratio using 1-year slope, 0.93, 95% CI: 0.92-0.94). Associations were stronger among participants with GFR <60 ml/min/1.73 m² (0.88, 95% CI: 0.86-0.91), when slopes were observed over longer observation periods (2-year slope, 0.80, 95% CI: 0.78-0.82; 3-year slope, 0.71, 95% CI: 0.69-0.74), and when adjusted for measurement error (1-year slope 0.80, 95% CI 0.77-0.83; 2-year slope, 0.71, 95% CI: 0.69-0.74; 3-year slope, 0.64, 95% CI: 0.60-0.67). Compared to a control group with baseline GFR of 60 ml/min/1.73 m² and an expected 2-year GFR decline of 3 ml/min/1.73 m²/year, an intervention that slows GFR decline by 0.75 ml/min/1.73 m²/year would be expected to result in an absolute 10-year risk reduction of ESKD of 1.3% (0.9% to 3.6% risk reduction for a baseline eGFR of 75 to 45 ml/min/1.73 m², respectively).

Conclusion

Slope of GFR is associated with subsequent ESKD, with stronger associations after adjustment for measurement error as well as when observed over longer periods. These results, combined with analyses of clinical trials and simulation studies, may support the use of GFR slope as a surrogate endpoint in clinical trials.

Funding

• NIDDK Support