Kidney Compass: GFR Slope and Hierarchical Composite Endpoints in Glomerular Disease

Welcome back to Kidney Compass!

On this episode, recorded at the inaugural International Society of Glomerular Diseases (ISGD) Clinical Trials Summit in Galveston, Texas, host Brendon Neuen, MBBS, PhD, is joined by Lesley Inker, MD, MS, a professor of medicine at Tufts University School of Medicine and director of the Kidney and Blood Pressure Center at Tufts Medical Center. Following her summit talk on GFR slope and hierarchical composite endpoints (HCEs), she explains how slope earned acceptance as a surrogate and what work remains before HCEs see broad use in glomerular disease.

How Did GFR Slope Become a Surrogate Endpoint for CKD Progression?

Inker traced the field’s move away from doubling of serum creatinine, equivalent to a 57% GFR decline, toward 40% and 30% declines. Those thresholds made more trials viable, but she noted they still capture events late in the disease course. Testing treatments in earlier CKD required an endpoint measurable sooner, which led her group to GFR slope.

At the 2018 National Kidney Foundation workshop with the FDA and EMA, Inker’s trial-level analysis compared treatment effects on slope with treatment effects on the clinical endpoint within each trial. The published meta-analysis of 47 randomized trials and 60,620 participants found an R² of 0.97 for 3-year total slope and 0.96 for chronic slope. A treatment effect of 0.75 mL/min/1.73 m²/year or more predicted clinical benefit with at least 96% probability.¹

An updated analysis in Nature Medicine expanded the dataset to 66 studies and 186,312 participants. Total slope remained strongly associated with the clinical endpoint (R², 0.97), with no evidence of heterogeneity across diabetes, glomerular disease, and other CKD groups.² Inker explained her team used a partial pooling model to borrow strength from the full dataset while producing subgroup estimates. In glomerular disease, the association was as strong or stronger than overall, though credible intervals were wider given fewer trials.

Neuen asked whether trials should assess change in GFR over a fixed follow-up or model slope across variable follow-up. Inker urged investigators to keep following early enrollees, since longer observation yields more safety data and more clinical events. The primary analysis is a separate question. In populations with few deaths or dialysis starts, such as IgA nephropathy, she said change from baseline using mixed models for repeated measures is reasonable, as recent trials have done.

She also flagged sources of nonlinearity specific to glomerular disease. Intercurrent illness, reduced oral intake, and courses of corticosteroids can produce systematic deviations in GFR trajectories. Historical trials collected limited data on treatment interruptions, and defining a discontinuation analytically remains unresolved. “Those kind of questions actually aren’t known right now,” Inker said. “I think it’s an important area for us to investigate.”

What Role Could Hierarchical Composite Endpoints Play in Glomerular Disease Trials?

The kidney HCE developed by Heerspink and colleagues ranks all-cause mortality, kidney failure, sustained 57%, 50%, and 40% GFR declines, and GFR slope, analyzed as win odds.³ Inker cautioned no single endpoint fits every trial. A population with high GFR and slow progression will generate few events and is a poor fit for an HCE. The ideal population accrues both kidney failure events and GFR decline, with slope capturing everyone else.

On how to compare slopes between individuals within an HCE, Inker said a minimum threshold difference lends face validity, even though empirical work suggests it has little effect on the win ratio. She separated this individual-level threshold from the between-arm difference regulators weigh. Her group estimates a between-arm difference of about 0.75 mL/min/1.73 m²/year is clinically meaningful. “It seems very small, but truly it translates into a hazard ratio of 0.7, which we all would be very happy with in our trials,” Inker said.

Neuen pointed to FIND-CKD as real-world confirmation. Finerenone slowed annual eGFR decline by 0.7 mL/min/1.73 m² versus placebo in adults with nondiabetic CKD, alongside a reduction in the composite kidney and cardiovascular outcome.⁴

Inker sees room to tailor HCE components by disease. “I think for all glomerular diseases, proteinuria or albuminuria is a really important component of the disease process,” she said. She cited work by Heerspink incorporating proteinuria in focal segmental glomerulosclerosis and suggested remission could serve as a component in lupus nephritis and membranous nephropathy, pending further validation.

Her group is now comparing treatment effects on HCEs with effects on the clinical endpoint and on slope across a broad set of trials, stratified by disease. Remaining technical questions include how fixed versus extended follow-up affects HCE use and how to handle therapies with acute hemodynamic effects on GFR.

Editors’ note: Disclosures for Neuen include AstraZeneca, Bayer, Boehringer Ingelheim, Janssen, and others. Disclosures for Inker include Chinook, Alexion, Novartis, AstraZeneca, and others.

References:

1. Inker LA, Heerspink HJL, Tighiouart H, et al. GFR slope as a surrogate end point for kidney disease progression in clinical trials: a meta-analysis of treatment effects of randomized controlled trials. J Am Soc Nephrol. 2019;30(9):1735-1745. doi:10.1681/ASN.2019010007

2. Inker LA, Collier W, Greene T, et al. A meta-analysis of GFR slope as a surrogate endpoint for kidney failure. Nat Med. 2023;29(7):1867-1876. doi:10.1038/s41591-023-02418-0

3. Heerspink HJL, Jongs N, Schloemer P, et al. Development and validation of a new hierarchical composite end point for clinical trials of kidney disease progression. J Am Soc Nephrol. 2023. doi:10.1681/ASN.0000000000000243

4. FIND-CKD expands finerenone evidence beyond diabetic CKD, with Rajiv Agarwal, MD. HCPLive. Published July 14, 2026. Accessed November 14, 2026. https://www.hcplive.com/view/find-ckd-expands-finerenone-evidence-beyond-diabetic-ckd-with-rajiv-agarwal-md