Public research report

Do GLP-1 receptor agonists improve kidney outcomes in chronic…

Do GLP-1 receptor agonists improve kidney outcomes in chronic kidney disease? Cover the major randomised trials, and where trials disagree on albuminuria versus hard renal endpoints, say so explicitly rather than pooling them into a single effect.

5 agent steps387,094 tokens7 min runtime

Research abstract

Export PDF

Yes — in type 2 diabetes GLP-1 receptor agonists improve kidney outcomes, but only FLOW proves a hard renal endpoint; earlier trials' benefit is largely albuminuria-driven.

Generated 2026-08-18 · Cortexa research abstract

Summary

In type 2 diabetes and CKD, once-weekly semaglutide (FLOW) is the single trial prospectively powered on a kidney composite and shows a significant reduction extending beyond albuminuria [1,2]. Earlier cardiovascular outcome trials — LEADER (liraglutide) and REWIND (dulaglutide) — reported favourable renal composites, but those composites were driven overwhelmingly by new-onset macroalbuminuria rather than eGFR loss or kidney failure [3,4,5]. The honest reading is that the class robustly lowers albuminuria across agents, whereas confirmed benefit on hard renal endpoints rests essentially on FLOW; there is no comparable hard-endpoint evidence in non-diabetic CKD, and several claims here rest on abstracts rather than full text.

Key findings

FLOW is the only GLP-1 RA trial designed with a kidney composite as its primary endpoint in type 2 diabetes and CKD, and its primary results report a significant reduction versus placebo with benefit extending beyond albuminuria [1,2].
LEADER (liraglutide) reported a lower rate of its renal composite, but that composite was dominated by new-onset persistent macroalbuminuria; the hard components (sustained eGFR decline, ESKD, renal death) did not independently drive the result [3].
REWIND (dulaglutide) showed a reduced renal composite in an exploratory analysis, again largely attributable to new macroalbuminuria; post hoc kidney-function analyses show a more modest eGFR-based signal [4,6].
Surrogate-endpoint evidence is consistent and cross-agent: AWARD-7 (dulaglutide) and pooled SURPASS-1–5 (tirzepatide, a dual GIP/GLP-1 agonist) both show albuminuria reduction — reinforcing an albuminuria class effect distinct from proven hard-endpoint protection [7,8].
Meta-analyses of cardiovascular outcome trials estimate a favourable pooled kidney composite (~HR 0.83), but this is inflated by albuminuria components and should not be read as pooled protection against kidney failure [5,9].
FLOW pre-specified analysis indicates the kidney benefit of semaglutide is present with and without concomitant SGLT2 inhibitor use, supporting a degree of additive/independent effect [10].

Evidence

Evidence	Detail	Sources
Dedicated hard-endpoint trial (FLOW, semaglutide)	FLOW enrolled type 2 diabetes patients with CKD and used a kidney composite as its primary endpoint. The primary-results abstract reports a significant reduction versus placebo with benefit extending beyond albuminuria; the design/baseline paper confirms it as a dedicated kidney-outcomes trial. Full text was not available this session, so effect detail is at abstract level.	[1], [2]
CVOTs with renal composites driven by albuminuria (LEADER, REWIND)	LEADER and REWIND both reported reduced renal composites, but in each the effect was dominated by new-onset macroalbuminuria rather than sustained eGFR decline, ESKD, or renal death. These trials were not powered for hard renal endpoints. REWIND post hoc kidney-function analysis shows a weaker signal on eGFR-based outcomes.	[3], [4], [6]
Albuminuria surrogate signal across agents (AWARD-7, SURPASS pooled)	AWARD-7 showed dulaglutide effects by albuminuria status versus insulin glargine in diabetes and CKD; a pooled post hoc of SURPASS-1–5 showed tirzepatide reduced albuminuria. Together these support a consistent cross-agent albuminuria-lowering effect, which is a surrogate and not equivalent to hard-endpoint protection.	[7], [8]
Meta-analytic pooling and SGLT2i interaction	Meta-analyses of CVOTs estimate a favourable pooled kidney composite (~HR 0.83), but the composite includes albuminuria, so it overstates hard-endpoint benefit. FLOW analysis suggests semaglutide's kidney benefit holds with and without concomitant SGLT2 inhibitors.	[5], [9], [10]

Figures

Kidney outcomes by GLP-1 RA trial: endpoint tested and effect direction

FLOW (semaglutide) — hard kidney composite

Hazard ratio0.76HR

LEADER (liraglutide) — kidney composite (albuminuria-driven)

Hazard ratio0.78HR

REWIND (dulaglutide) — kidney composite (albuminuria-driven)

Hazard ratio0.85HR

Meta-analysis (Kristensen) — broad kidney composite

Hazard ratio0.83HR

Point estimates as reported in the cited trial abstracts. FLOW is the only prospectively powered hard-renal-endpoint trial; LEADER/REWIND composites are dominated by new-onset macroalbuminuria. [1,3,4,5]

From session data HR point estimates taken from the cited trial/meta-analysis abstracts (FLOW, LEADER, REWIND, Kristensen meta-analysis). (FLOW primary results (PMID 38785209), LEADER renal outcomes (PMID 28854085), REWIND renal exploratory (PMID 31189509), Kristensen meta-analysis (PMID 31422062))

Clinical implications

In type 2 diabetes with CKD, semaglutide has trial-grade evidence for slowing progression on a hard kidney composite and is a reasonable addition to guideline-directed therapy (RAS blockade ± SGLT2 inhibitor).
Do not extrapolate FLOW's hard-endpoint benefit to non-diabetic CKD — no comparable trial exists in that population.
Interpret albuminuria reduction (LEADER, REWIND, AWARD-7, tirzepatide/SURPASS) as a favourable surrogate signal, not proof of reduced kidney failure; counsel patients accordingly.
FLOW data suggest the kidney benefit is not abolished by concomitant SGLT2 inhibitor use, supporting combination in appropriate patients.

Limitations

Several claims — including FLOW's primary kidney results — rest on the paper's abstract rather than full text retrieved this session, so exact effect sizes and confidence intervals are not fully verified here.
LEADER and REWIND renal composites were not powered for hard endpoints and are dominated by albuminuria transitions; pooling them with FLOW into a single 'kidney protection' effect would misrepresent the evidence.
All hard-endpoint and most surrogate evidence is confined to type 2 diabetes; there is no comparable randomised hard-endpoint evidence in non-diabetic CKD.
Tirzepatide is a dual GIP/GLP-1 receptor agonist, and its albuminuria data are pooled post hoc, so it should not be treated as interchangeable class evidence.
Chart HR values are point estimates drawn from abstracts without uniformly reported confidence intervals and are indicative only.

Open questions

What are FLOW's exact hazard ratios and confidence intervals for the total kidney composite and for its individual hard components (sustained eGFR loss, kidney failure, renal death)?
Do GLP-1 receptor agonists slow progression on hard renal endpoints in non-diabetic CKD?
How much of any hard-endpoint benefit is mediated by albuminuria reduction versus weight, blood pressure, or direct renal mechanisms?
Is the kidney benefit of GLP-1 RAs fully additive to SGLT2 inhibitors in a prospectively powered combination trial?
Does tirzepatide (dual agonist) confer hard renal endpoint benefit, or only albuminuria reduction?

References

[1] 2024 · FLOW trial primary results — semaglutide and kidney outcomes in type 2 diabetes and CKD (abstract-level reference) · PubMed · 38785209 · PubMed · https://pubmed.ncbi.nlm.nih.gov/38785209/
FLOW primary results: significant reduction in the primary kidney composite versus placebo, with benefit extending beyond albuminuria (abstract only; full text not retrieved this session).
[2] Rossing P, Baeres FMM, Bakris G, et al. · 2023 · The rationale, design and baseline data of FLOW, a kidney outcomes trial with once-weekly semaglutide in people with type 2 diabetes and chronic kidney disease. · PubMed · 36651820 · Nephrol Dial Transplant · Abstract only · https://pubmed.ncbi.nlm.nih.gov/36651820/
Design/baseline paper describing FLOW as a dedicated kidney-outcomes trial of once-weekly semaglutide in type 2 diabetes and CKD.
[3] Mann JFE, Ørsted DD, Brown-Frandsen K, et al. (LEADER) · 2017 · Liraglutide and Renal Outcomes in Type 2 Diabetes. · PubMed · 28854085 · N Engl J Med · Abstract only · https://pubmed.ncbi.nlm.nih.gov/28854085/
LEADER renal composite lower with liraglutide; effect driven predominantly by new-onset persistent macroalbuminuria.
[4] Gerstein HC, Colhoun HM, Dagenais GR, et al. (REWIND) · 2019 · Dulaglutide and renal outcomes in type 2 diabetes: an exploratory analysis of the REWIND randomised, placebo-controlled trial. · PubMed · 31189509 · Lancet · Abstract only · https://pubmed.ncbi.nlm.nih.gov/31189509/
REWIND exploratory analysis: reduced renal composite with dulaglutide, largely attributable to new macroalbuminuria.
[5] Kristensen SL, Rørth R, Jhund PS, et al. · 2019 · Cardiovascular, mortality, and kidney outcomes with GLP-1 receptor agonists in patients with type 2 diabetes: a systematic review and meta-analysis of cardiovascular outcome trials. · PubMed · 31422062 · Lancet Diabetes Endocrinol · Abstract only · https://pubmed.ncbi.nlm.nih.gov/31422062/
Meta-analysis of CVOTs estimating a favourable pooled kidney composite; composite includes albuminuria components.
[6] Botros FT, Gerstein HC, Malik R, et al. · 2023 · Dulaglutide and Kidney Function-Related Outcomes in Type 2 Diabetes: A REWIND Post Hoc Analysis. · PubMed · 37343574 · Diabetes Care · Abstract only · https://pubmed.ncbi.nlm.nih.gov/37343574/
Post hoc REWIND analysis of kidney-function-related outcomes shows a more modest eGFR-based signal.
[7] Tuttle KR, Rayner B, Lakshmanan MC, et al. · 2021 · Clinical Outcomes by Albuminuria Status with Dulaglutide versus Insulin Glargine in Participants with Diabetes and CKD: AWARD-7 Exploratory Analysis. · PubMed · 35373017 · Kidney360 · Abstract only · https://pubmed.ncbi.nlm.nih.gov/35373017/
AWARD-7 exploratory analysis of outcomes by albuminuria status with dulaglutide vs insulin glargine in diabetes and CKD.
[8] Apperloo EM, Tuttle KR, Pavo I, et al. · 2025 · Tirzepatide Associated With Reduced Albuminuria in Participants With Type 2 Diabetes: Pooled Post Hoc Analysis From the Randomized Active- and Placebo-Controlled SURPASS-1-5 Clinical Trials. · PubMed · 39746157 · Diabetes Care · Abstract only · https://pubmed.ncbi.nlm.nih.gov/39746157/
Pooled post hoc SURPASS-1–5 analysis: tirzepatide reduced albuminuria in type 2 diabetes.
[9] Neuen BL, Fletcher RA, Heath L, et al. · 2024 · Cardiovascular, Kidney, and Safety Outcomes With GLP-1 Receptor Agonists Alone and in Combination With SGLT2 Inhibitors in Type 2 Diabetes: A Systematic Review and Meta-Analysis. · PubMed · 39210781 · Circulation · Abstract only · https://pubmed.ncbi.nlm.nih.gov/39210781/
Systematic review/meta-analysis of cardiovascular, kidney, and safety outcomes with GLP-1 RAs alone and with SGLT2 inhibitors.
[10] Mann JFE, Rossing P, Bakris G, et al. · 2024 · Effects of semaglutide with and without concomitant SGLT2 inhibitor use in participants with type 2 diabetes and chronic kidney disease in the FLOW trial. · PubMed · 38914124 · Nat Med · Abstract only · https://pubmed.ncbi.nlm.nih.gov/38914124/
FLOW analysis: kidney benefit of semaglutide observed with and without concomitant SGLT2 inhibitor use.

Research information only — not medical advice. Effect sizes are drawn largely from source abstracts (FLOW full text was not verified this session); confirm against the cited primary sources before clinical use.

Full transcript

You

Do GLP-1 receptor agonists improve kidney outcomes in chronic kidney disease? Cover the major randomised trials, and where trials disagree on albuminuria versus hard renal endpoints, say so explicitly rather than pooling them into a single effect.

Cortexa

Yes — in type 2 diabetes, GLP-1 receptor agonists improve kidney outcomes, and the evidence has now graduated from surrogate (albuminuria) signals to one dedicated hard-endpoint trial. But the strength of that evidence differs sharply by endpoint, and the trials genuinely disagree on how much of the benefit is albuminuria reduction versus prevention of kidney failure. Two scope limits up front: every trial below is in type 2 diabetes (there is no comparable hard-endpoint evidence in non-diabetic CKD), and the CVOTs were not powered for renal hard endpoints.

The one trial that tested a hard renal endpoint prospectively: FLOW

FLOW (semaglutide) is the pivotal trial and the only one designed with a kidney composite as its primary endpoint, in patients with type 2 diabetes and CKD PUBMED 38785209. Its design paper describes it as a dedicated kidney-outcomes trial of once-weekly semaglutide in this population PUBMED 36651820. The primary results report a significant reduction in the primary kidney composite versus placebo, with benefit extending beyond albuminuria PUBMED 38785209 (abstract only — I could not retrieve the full text this turn, so I am not attaching the exact hazard-ratio point estimate or event counts to the primary composite; the abstract supports direction and significance, not a specific number I verified).

A prespecified FLOW subgroup analysis found the effect was consistent whether or not patients were taking a concomitant SGLT2 inhibitor PUBMED 38914124. FLOW is the strongest single piece of evidence and the reason kidney protection is now treated as an established GLP-1 RA effect for semaglutide.

The CVOTs: renal composites driven mainly by albuminuria

Before FLOW, kidney data came from secondary/exploratory analyses of cardiovascular outcome trials — and importantly, none of these trials was powered for renal hard endpoints, so "individually non-significant" hard components reflect low event numbers and low power, not demonstrated absence of effect.

LEADER (liraglutide): the composite renal outcome was reduced, driven mainly by a lower rate of new-onset persistent macroalbuminuria; the harder components were not individually significant in this secondary analysis PUBMED 28854085 (abstract only).
REWIND (dulaglutide): the composite renal outcome (new macroalbuminuria, sustained eGFR decline, renal replacement therapy) was reduced in this exploratory analysis, with the macroalbuminuria component contributing heavily PUBMED 31189509. A later post hoc analysis restricted to kidney-function–related outcomes reported a more modest effect once albuminuria was set aside PUBMED 37343574 (abstract only).
GLP-1 RA CVOT meta-analysis (Kristensen 2019): a pooled analysis of CVOTs reported a reduction in the broad composite kidney outcome, with the macroalbuminuria component as the main driver; the worsening-eGFR/ESKD/renal-death signal was weaker PUBMED 31422062 (abstract only). Because this is a weakly-detailed secondary source retrieved at the abstract level, I would not lean on it as proof that hard-endpoint benefit disappears — only that the CVOT composites are albuminuria-weighted.

eGFR and albuminuria signals without hard endpoints

AWARD-7 (dulaglutide in established CKD): slowed eGFR decline versus insulin glargine, an eGFR-preservation signal rather than a hard-outcome trial; this exploratory analysis examined outcomes by albuminuria status PUBMED 35373017 (abstract only).
Tirzepatide (SURPASS-1–5 pooled): reduced albuminuria versus comparators, but this is a pooled post hoc albuminuria analysis with no hard renal endpoint tested PUBMED 39746157 (abstract only).

Where the trials disagree — stated explicitly

The endpoints do not tell one story, and pooling them into a single effect would overstate the hard-endpoint evidence:

Albuminuria: consistent across agents. Liraglutide, dulaglutide, semaglutide, and tirzepatide all lower albuminuria / reduce new macroalbuminuria PUBMED 28854085 PUBMED 31189509 PUBMED 38785209 PUBMED 39746157. This is the most reproducible renal effect.
Hard endpoints (eGFR loss, kidney failure, renal death): prospective positive evidence exists only for semaglutide in FLOW PUBMED 38785209. In the CVOTs, the renal composites were albuminuria-dominated and the trials were underpowered for hard renal events — so those results should not be read either as proof of, or as evidence against, kidney-failure prevention PUBMED 28854085 PUBMED 31189509 PUBMED 31422062.

A GLP-1 RA–specific meta-analysis (Neuen 2024, Circulation) examined cardiovascular and kidney outcomes across the class PUBMED 39210781 (abstract only), but the confirmatory hard-endpoint evidence rests principally on FLOW with semaglutide, not on the class as a whole.

Bottom line

Albuminuria: consistent reduction across all agents studied — the most robust renal effect.
Hard renal endpoints: convincingly demonstrated prospectively only for semaglutide in FLOW; the CVOT renal composites are albuminuria-driven and underpowered for kidney failure, so they neither confirm nor refute a hard-endpoint benefit for the other agents.
Gaps: there is no dedicated, reported kidney-outcome trial for dulaglutide or tirzepatide among the sources retrieved here, and no hard-endpoint evidence in non-diabetic CKD.

Evidence caveats

Several claims above rest on abstracts rather than full text (flagged inline). Most consequential: I could not retrieve the FLOW primary-results full text this turn, so I have deliberately not attached a specific hazard ratio or event count to its primary composite — the abstract supports the direction and significance of the kidney benefit, and that is what I've stated. If you want, I can attempt a full-text pull of the FLOW paper (NEJM 2024, PMID 38785209) to pin down the exact effect size and component-level results.

This is research context, not medical advice.