A prespecified pooled analysis of 30,787 participants from the SELECT, FLOW, and SOUL trials shows semaglutide reduces major kidney outcomes by 16% and a kidney-specific composite by 20%, with benefits seen in people with and without diabetes. Safety was consistent with the known GLP-1 receptor…
On August 7, 2026, The Lancet Diabetes & Endocrinology published a prespecified pooled analysis of the SELECT, FLOW, and SOUL trials showing that semaglutide reduced the risk of a composite of major kidney outcomes by 16% compared with placebo. The analysis, conducted by Novo Nordisk, pooled participant-level data from 30,787 randomized adults across the three placebo-controlled phase 3 trials. The population spans the cardio-kidney-metabolic disease spectrum: FLOW enrolled people with chronic kidney disease, while SELECT and SOUL enrolled people with atherosclerotic cardiovascular disease, with and without type 2 diabetes.
The primary kidney composite occurred in 973 semaglutide-treated participants versus 1134 placebo participants, a hazard ratio of 0.84 95% CI 0.77 to 0.91 . A secondary composite that excluded cardiovascular-related death showed a 20% relative reduction: 347 events with semaglutide versus 416 with placebo, a hazard ratio of 0.80 95% CI 0.69 to 0.92 . Kidney benefits appeared in people with diabetes and in people without it, which matters because the earlier evidence base for kidney protection with GLP-1 receptor agonists was built largely around glucose-lowering trials in type 2 diabetes.
Safety outcomes were similar between groups and consistent with the known adverse-event profile of GLP-1 receptor agonists, and serious adverse events were numerically lower with semaglutide. The analysis is the first prespecified pooled analysis of participant-level data from these three programs to focus specifically on a composite kidney outcome across diverse cardio-kidney-metabolic populations, including people without diabetes.
The primary endpoint was time to first occurrence of a persistent ≥50% reduction in estimated glomerular filtration rate eGFR , kidney failure , kidney-related death, or cardiovascular-related death. Kidney failure was defined as persistent eGFR <15 mL/min per 1.73 m² or initiation of kidney replacement therapy. The secondary endpoint was the same kidney composite with cardiovascular-related death removed.
The 50% eGFR threshold is a common surrogate in kidney outcome trials: a sustained reduction of that size closely tracks progression to kidney failure. The eGFR threshold of <15 mL/min per 1.73 m² is the conventional definition of kidney failure, the stage at which replacement therapy is typically considered. Because the composite counts only the first event per participant, the event totals of 973 and 1134 are conservative measures of the total burden of kidney and cardiovascular events.
The three trials used different configurations of the drug. SELECT used 2.4 mg of once-weekly subcutaneous semaglutide in people with established atherosclerotic cardiovascular disease; FLOW used 1.0 mg once-weekly subcutaneous in people with chronic kidney disease; SOUL used 14 mg of once-daily oral semaglutide. Mean follow-up ranged from 39.5 to 47.5 months across the trials. The two hazard ratios, 0.84 for the composite including cardiovascular death and 0.80 for the kidney-specific composite, point in the same direction. Their confidence intervals 0.77 to 0.91 and 0.69 to 0.92 overlap, so the difference between the two estimates should not be over-read, but the direction is consistent with a kidney effect that does not depend on cardiovascular death.
Pooling participant-level data is not the same as meta-analyzing published summary results. With individual records, one endpoint definition can be applied uniformly: every eGFR measurement, kidney replacement therapy initiation, and death is classified against identical rules, and every participant enters the same time-to-event model with trial as a stratification factor. That uniformity increases precision and makes event counts directly comparable across trials. The prespecified design, with the analysis plan fixed before the pooled data were examined, limits the scope for selective reporting.
The design also has boundaries. Participants were randomized to semaglutide or placebo within each trial, not across doses or routes of administration. The analysis can therefore estimate the average effect of semaglutide as tested in these three programs, but it cannot establish equivalence between the oral and subcutaneous formulations or between the 1.0 mg, 2.4 mg, and 14 mg doses. Baseline characteristics differed among the three study populations, and the follow-up durations were not identical. A pooled analysis of three trials remains a secondary analysis; it does not carry the same evidentiary weight as a single prospective trial designed around a kidney primary endpoint.
Within those limits, the analysis does establish an average 16% reduction on the broad composite and a 20% reduction on the kidney-specific composite across three independent programs. The randomized comparison inside each trial remains intact, so confounding of the kind that plagues observational cohorts is not the concern here. The limitation is about averaging across heterogeneous trials, not about the internal validity of each treatment contrast.
Semaglutide is a synthetic analog of human glucagon-like peptide-1, a 31-amino-acid peptide with structural changes that block degradation by dipeptidyl peptidase-4 and a fatty-acid side chain that binds albumin to extend its half-life. GLP-1 receptors are expressed beyond the pancreatic beta cell, including in vascular endothelium, vascular smooth muscle, immune cells, and renal tissue. That broad expression gives the drug a pharmacological reach that glucose control alone would not predict.
Several candidate mechanisms could explain the kidney signal. GLP-1 receptor activation increases natriuresis and modestly lowers blood pressure, which reduces intraglomerular pressure. The drugs also dampen inflammatory signaling and oxidative stress, and they lower body weight and blood glucose, both of which reduce long-term metabolic strain on the kidney. Because the benefit in this pooled analysis appeared in people without diabetes, and because the kidney-specific composite showed a numerically larger relative reduction than the composite that included cardiovascular death, the analysis supports the interpretation that the benefit may not be explained solely by glycaemic or weight effects.
What the analysis cannot do is distinguish among those mechanisms. Clinical event data do not reveal whether the protection comes from hemodynamic changes at the glomerulus, from anti-inflammatory effects on renal immune cells, or from reduced tubular metabolic demand. Mechanistic substudies measuring albuminuria, intraglomerular hemodynamics, and inflammatory biomarkers in the pooled trial populations would be needed to separate these pathways.
The Peptide Atlas registry file for semaglutide lists 668 registered clinical trials, with 10 currently recruiting. Of the records on file with a phase classification, four are Phase 2, four are Phase 4, and one is Phase 3. Peptide Atlas indexes 197 PubMed papers for semaglutide. The active trial roster shows how far the peptide has moved beyond diabetes and obesity:
Recent indexed literature tracks the same expansion. A systematic review and meta-analysis in Clin Ter July 1, 2026; PMID 42340790 examined long-term safety and renal outcomes of semaglutide in non-diabetic obesity with chronic kidney disease or hypertension. The STRIDE trial, reported in European Heart Journal July 1, 2026; PMID 41780559 , assessed semaglutide in peripheral artery disease and diabetes by baseline disease severity and age. A randomized clinical trial in JAMA Psychiatry July 1, 2026; PMID 42054055 tested semaglutide and effort-based decision-making in major depressive disorder. Also indexed are a systematic review of weight-lowering drugs and natural female fertility in Clinical Obesity July 1, 2026; PMID 42307450 and a scoping review of retinal vascular events in Ophthalmologica June 25, 2026; PMID 42348481 .
The eight third-party lab purity tests on file for semaglutide, with a highest observed purity of 99.979%, indicate the quality standard that contract laboratories are achieving for this peptide. For a molecule injected at milligram doses, purity bears directly on consistent pharmacology and on the interpretation of preclinical and translational experiments.
For clinicians, the pooled analysis gives a single, readable estimate to use when discussing kidney risk with patients on semaglutide, including patients without diabetes. A 16% reduction in first events of a composite that includes kidney failure, kidney-related death, and cardiovascular-related death is a meaningful effect size, and the 20% reduction on the kidney-specific composite points to an effect on the kidney itself. The numerically lower rate of serious adverse events with semaglutide, against a safety profile otherwise consistent with the GLP-1 receptor agonist class, is relevant to prescribing decisions in a chronic disease population that may take the drug for years.
For researchers, the analysis demonstrates that kidney outcomes can be extracted from cardiovascular outcome trials through prespecified participant-level pooling, and that kidney endpoints are now tractable in GLP-1 receptor agonist development. The registry data reinforce that semaglutide research is no longer confined to metabolic medicine; recruiting trials in type 1 diabetes, severe mental illness, endometrial cancer, atrial fibrillation, and childhood obesity mean the evidence base will accumulate across several specialties at once. The oral program, which delivered the 14 mg dose tested in SOUL, depends on co-formulation with an absorption enhancer to make the peptide bioavailable across the gut, a distinct production challenge from the subcutaneous versions.
For the peptide supply chain, the analysis strengthens the demand case for semaglutide as a therapeutic beyond glycemic control, with consequences for manufacturing capacity for both injectable and oral forms. Regulators evaluating semaglutide's kidney effects now have a single prespecified analysis spanning three phase 3 programs to weigh, but the pooled design, differing doses, and differing routes mean the evidence is likely to be treated as supportive rather than definitive on its own.
The analysis carries explicit caveats. The three trials differed in baseline characteristics, in semaglutide dose and route of administration subcutaneous 1.0 mg in FLOW, subcutaneous 2.4 mg in SELECT, oral 14 mg in SOUL , and in mean follow-up, which ranged from 39.5 to 47.5 months. The pooled design cannot eliminate those differences; it can only average over them. The follow-up period is also short relative to the natural history of chronic…
Peptides referenced: Semaglutide, Glucagon, GLP-1.
Related reading: AI Peptide That Locks STING Dimer Speeds Diabetic Wound Healing, Bank of America spends $250 million a year on weight loss drugs for staff: ‘We see a great impact, Peptide Research Enters New Era as High-Purity Demand Surges, GLP-1 Receptor Agonists Tied to Lower Pseudarthrosis Odds After ACDF.