On this page
- EGFR and urine albumin mean in
- FLOW tested
- The primary result
- Kidney function declined more slowly on
- Heart events and deaths fell as well
- FLOW sits next to the other kidney
- The combination question FLOW could not
- US and Canadian approvals actually cover
- FLOW did not test
- Each semaglutide kidney outcomes claim is
- Short version
- Questions
- Sources
This article works through what FLOW actually measured: how kidney function is scored, who the trial enrolled, the primary result in both relative and absolute terms, the heart and survival results, how the trial sits next to the other kidney-protective drug classes, what the US and Canadian approvals actually say, and what the trial did not test.
What eGFR and urine albumin mean in semaglutide kidney outcomes research
Chronic kidney disease means the kidneys filter blood more slowly, leak protein into the urine, or both, over months to years. Two laboratory numbers describe it. The estimated glomerular filtration rate (eGFR) estimates how much blood the kidneys filter each minute, adjusted for body size; higher is better, and a healthy young adult usually sits above 90. The urinary albumin-to-creatinine ratio (UACR) measures albumin, a blood protein, in the urine; healthy kidneys leak almost none, so a high ratio is an early sign of kidney damage.
Those two numbers defined who could join FLOW. Participants needed an eGFR of 50 to 75 with a UACR above 300 and below 5,000, or an eGFR of 25 to below 50 with a UACR above 100 and below 5,000. In plain terms, everyone enrolled had measurably reduced filtration plus substantial protein in the urine, a combination that carries a high risk of progressing to kidney failure. Participants had a mean age of 66.6 years, and the group's mean eGFR at entry was 47.0, well inside the moderate-to-severe range.
The endpoint language matters the same way. Kidney failure in FLOW meant starting chronic dialysis, receiving a transplant, or a sustained fall to an eGFR below 15. A sustained fall of at least 50% in eGFR from baseline meant losing half of the filtration a participant started with, confirmed over time rather than on a single test. These are events patients feel and health systems pay for, not laboratory markers alone, which is part of what made FLOW different from the kidney signals reported in earlier trials of this drug class.
What FLOW tested: one trial built for kidney outcomes
FLOW (Perkovic et al., N Engl J Med 2024; ClinicalTrials.gov NCT03819153) was a double-blind, randomized, placebo-controlled trial run at 387 sites in 28 countries. Researchers randomly assigned 3,533 adults with type 2 diabetes and chronic kidney disease to semaglutide 1.0 mg once weekly, as administered in the study, or to placebo. Both groups continued standard care, which included a renin-angiotensin system inhibitor, a standard kidney-protective blood pressure medicine, at the highest tolerated dose unless it was not tolerated or not advised.
The primary outcome was a composite of major kidney disease events: the onset of kidney failure, at least a 50% reduction in eGFR from baseline, or death from kidney-related or cardiovascular causes. Prespecified confirmatory secondary outcomes were the yearly rate of change in eGFR, major adverse cardiovascular events, and death from any cause, tested in a fixed order.
Two design facts belong next to every result below. First, the trial was stopped early: at a prespecified interim analysis, the independent monitoring committee recommended ending it because the efficacy criteria for early cessation had been met, after a median follow-up of 3.4 years. Early stopping for clear benefit is an ethical safeguard, and it can also make an effect estimate slightly optimistic, which is one reason the grades below stop where they do. Second, Novo Nordisk, which sells semaglutide, funded the trial. That is standard for approval trials, and it is a reason to want independent replication, not a reason to discard randomized data.
Our semaglutide research profile tracks the full trial record for this compound, from the diabetes and weight programs to the cardiovascular outcome trials that pointed toward FLOW.
The primary result: major kidney events fell 24%
A primary-outcome event occurred in 331 of 1,767 participants on semaglutide and 410 of 1,766 on placebo. The hazard ratio was 0.76 (95% CI, 0.66 to 0.88; P=0.0003), a 24% lower risk of a major kidney event on semaglutide over the follow-up period.
Read the same result in absolute terms. The event counts work out to about 18.7% of the semaglutide group and 23.2% of the placebo group experiencing a first primary event during the trial, a gap of roughly 4 to 5 people in every 100 treated over about 3.4 years. Relative percentages make headlines; the absolute gap is what a person weighing treatment is actually trading against side effects, cost, and effort, and in a population this high-risk it is a substantial gap.
The kidney-specific parts of the composite pointed the same way. A composite limited to the kidney components of the primary outcome (kidney failure, a fall of at least 50% in eGFR, or death from kidney causes) gave a hazard ratio of 0.79 (95% CI, 0.66 to 0.94), a 21% lower risk. That matters because the primary composite also counts cardiovascular deaths; the kidney-only analysis shows the result was not carried by the heart component alone.
Kidney function declined more slowly on semaglutide
Beyond the event counts, FLOW measured how fast filtration itself was falling. The mean annual eGFR slope was less steep on semaglutide by 1.16 units per year (P<0.001): filtration fell by about 2.19 units per year on semaglutide versus 3.36 units per year on placebo.
A difference of about one filtration unit per year sounds small. Compounded over a decade of living with chronic kidney disease, it is the difference between reaching kidney failure in a given year and reaching it several years later, or not at all. Slope results like this are also the trial's internal consistency check: a drug that truly slows kidney decline should show both fewer kidney failure events and a slower measured decline, and FLOW showed both, in the same direction, in the same participants.
What the slope does not show is recovery. Filtration still fell in both groups, just more slowly on semaglutide. FLOW is a story about delaying progression in a progressive disease, not about restoring kidney function that has already been lost, and no result in the trial supports a claim of reversal.
Heart events and deaths fell as well
People with type 2 diabetes and chronic kidney disease die of cardiovascular causes more often than of kidney failure itself, which is why FLOW built cardiovascular death into its primary outcome and tracked heart events and survival as confirmatory secondaries. All three favored semaglutide.
Death from cardiovascular causes gave a hazard ratio of 0.71 (95% CI, 0.56 to 0.89), a 29% lower risk. Major adverse cardiovascular events, the standard composite of cardiovascular death, nonfatal heart attack, and nonfatal stroke, gave a hazard ratio of 0.82 (95% CI, 0.68 to 0.98; P=0.029), an 18% lower risk. Death from any cause gave a hazard ratio of 0.80 (95% CI, 0.67 to 0.95; P=0.01), a 20% lower risk over the trial's follow-up.
Keep the hierarchy in mind when quoting these numbers. The kidney composite was the primary outcome the trial was sized to measure. The cardiovascular and mortality results were prespecified and tested in a fixed sequence, which makes them confirmatory rather than exploratory, but they rest on fewer events than the primary outcome and their confidence intervals are wider. They strengthen the overall picture; they are not three separate trials' worth of proof.
Read this next Semaglutide research profile
On safety, serious adverse events were reported in 49.6% of the semaglutide group and 53.8% of the placebo group. That high overall rate reflects the population: older adults with diabetes and advanced kidney disease accumulate hospitalizations and serious illnesses at a high rate regardless of treatment. The FLOW paper does not present the trial as evidence about rarer harms of the drug class; the class safety record has to be read from the product label and the much larger diabetes and weight-management programs. Our semaglutide thyroid warning explainer covers the best-known label warning and what the human evidence behind it shows.
How FLOW sits next to the other kidney evidence
Three comparisons put the FLOW result in context. Each one gets its own evidence, because the trials differ in drug, population, and design and cannot be ranked like a single head-to-head study.
Earlier GLP-1 trials: kidney signals, not kidney trials. Before FLOW, the kidney findings for this drug class came from cardiovascular outcome trials, where kidney outcomes were secondary and the results were driven mainly by albuminuria, the amount of protein leaking into the urine, rather than by kidney failure events. Those signals were consistent enough to justify a dedicated trial, and FLOW was that trial: the first in this class designed and powered with major kidney events as the primary outcome. The step from a secondary albumin signal to a primary kidney failure composite is the step from suggestive to tested.
SGLT2 inhibitors: proven separately, never compared head to head. A different drug class, the SGLT2 inhibitors, earned kidney indications from its own dedicated outcome trials (CREDENCE, DAPA-CKD, and EMPA-KIDNEY) in overlapping but not identical populations, and these medicines were already standard kidney-protective care when FLOW ran. FLOW tested semaglutide added to standard care against placebo added to standard care; it never randomized anyone between semaglutide and an SGLT2 inhibitor. Whether the two classes add to each other is a separate question, covered below, and no cross-trial subtraction of hazard ratios can answer it.
Tirzepatide: no kidney outcome trial yet. Tirzepatide, the other widely discussed drug in this space, has large trials for diabetes, weight management, and sleep apnea, and our tirzepatide sleep apnea review covers its SURMOUNT-OSA program. It has no FLOW equivalent: in the sources reviewed as of October 2026, no dedicated kidney outcomes trial has reported results for tirzepatide, so any kidney claim for it rests on secondary measures such as albuminuria, not on kidney failure events. That is a gap in the record, not evidence that tirzepatide does or does not protect kidneys.
The combination question FLOW could not answer
The most important open question in this area is whether semaglutide adds kidney protection for people already taking an SGLT2 inhibitor, the population most clinicians now treat. FLOW was designed before SGLT2 inhibitors became universal in this population, and only 550 of its 3,533 participants, about 16%, were taking one at enrollment.
A prespecified subgroup analysis (Mann et al., Nat Med 2024) split the results by that baseline use. Among participants not taking an SGLT2 inhibitor, the primary outcome hazard ratio was 0.73 (95% CI, 0.63 to 0.85). Among the 550 who were, it was 1.07 (95% CI, 0.69 to 1.67). The test for whether those two results truly differ, the interaction test, was not statistically significant (P=0.109), and the eGFR slope point estimate favored semaglutide in both subgroups, although the interval in the small SGLT2 subgroup (0.75; 95% CI, -0.01 to 1.5) included no effect.
Read that subgroup honestly. A hazard ratio of 1.07 in 550 people comes with a confidence interval wide enough to include both a large benefit and moderate harm; it cannot show that the combination fails, and the overall trial cannot show that the combination adds benefit on top of an SGLT2 inhibitor. The correct summary is the unsatisfying one: the added-benefit question is still open, and it needs a trial designed to answer it. We did not find reliable data in the sources reviewed that settles it, and that does not mean the answer is no.
What the US and Canadian approvals actually cover
United States. On January 28, 2025, the FDA approved Ozempic (semaglutide) to reduce the risk of kidney disease worsening, kidney failure (end-stage kidney disease), and death due to cardiovascular disease in adults with type 2 diabetes and chronic kidney disease; the product label phrases the same indication as sustained eGFR decline, end-stage kidney disease, and cardiovascular death. Novo Nordisk described it as the first approval of a GLP-1 medicine for this kidney indication. The approval is for the branded diabetes product and that population; it is not an approval of the higher weight-management dose for kidney outcomes, and it says nothing about compounded semaglutide, which has no kidney outcome trial behind it. Our compounded semaglutide versus Ozempic comparison covers how compounded products differ from the studied product.
Canada. Health Canada approved the same kidney indication in August 2025: Ozempic to reduce the risk of sustained eGFR decline, end-stage kidney disease, and cardiovascular death in adults with type 2 diabetes and chronic kidney disease, again a first for this drug class in Canada. As in the US, the approval rests on FLOW and matches the population FLOW studied.
Two limits are built into both approvals. The population is adults with both type 2 diabetes and chronic kidney disease: the approvals do not extend to kidney disease without diabetes, to type 1 diabetes, or to children. And in both countries this is a prescription medicine for a diagnosed condition, where kidney staging by eGFR and urine testing is part of the diagnosis, not an optional detail.
What FLOW did not test
The untested list is as important as the results list. FLOW did not enroll people with kidney disease who did not have type 2 diabetes, people with type 1 diabetes, anyone under 18, or people already receiving dialysis. It tested the 1.0 mg weekly diabetes dose, not the 2.4 mg weekly weight-management dose, so the kidney results cannot be quoted as results of the weight-loss product. It did not compare semaglutide against an SGLT2 inhibitor or any other active treatment, and, as covered above, it could not measure the added benefit of combining the two classes.
The trial also says nothing about what happens after treatment stops. Participants were followed while the trial ran, not after a planned discontinuation, so whether kidney protection persists, fades, or reverses off the drug is untested in this program. Weight regain after stopping this drug class is documented in other trials, which is a reason for caution about assuming permanence, not a measured kidney result. See what the discontinuation studies found for the weight side of that record.
How each semaglutide kidney outcomes claim is graded
Grades below reflect the Evidence Research Team's internal evidence assessment, not formal rubric sign-off. Applying the site's grading approach to each finding separately: in rubric terms, the moderate grades below reflect low risk of bias and direct measurement in the studied population, downgraded because a single trial leaves consistency unassessable, the trial stopped early, and a single manufacturer funder means publication bias cannot be excluded.
Semaglutide reduces major kidney events in type 2 diabetes with chronic kidney disease: moderate certainty. One large phase 3 randomized, double-blind trial (human randomized evidence, 3,533 participants) with a prespecified primary composite found a statistically significant 24% relative reduction, and the kidney-specific components, the eGFR slope, and the mortality secondaries all pointed the same way. The grade stops at moderate because the entire result rests on a single manufacturer-funded trial with no independent replication yet, and early stopping for benefit can make an effect estimate somewhat optimistic.
Semaglutide slows the decline in kidney filtration: moderate certainty. The eGFR slope was a prespecified confirmatory secondary outcome, the difference of 1.16 units per year was statistically significant, and it is consistent with the event results in the same trial. It shares the same single-trial, single-funder limits as the primary grade.
Semaglutide reduces cardiovascular events and death in this population: moderate certainty. These were prespecified confirmatory secondaries tested in a fixed hierarchy and all favored semaglutide, but they rest on fewer events than the primary outcome and come from the same single trial program.
Semaglutide adds kidney benefit on top of an SGLT2 inhibitor: insufficient evidence. The relevant subgroup held 550 participants, its confidence interval was wide, and the interaction test was not significant. FLOW neither proves nor rules out an added benefit.
Semaglutide protects kidneys in people without diabetes, in type 1 diabetes, or at the weight-management dose: insufficient evidence. None of those populations or that dose were studied in FLOW, and no completed trial in the sources reviewed answers those questions.
See how we rank peptides by evidence strength for where semaglutide's trial record sits in the wider ranking, and the weight management research topic for the weight-loss side of the same compound's record.
Semaglutide kidney outcomes: the short version
In FLOW, 3,533 adults with type 2 diabetes and chronic kidney disease took semaglutide 1.0 mg weekly or placebo for a median of 3.4 years. Major kidney events occurred in 18.7% versus 23.2%, a 24% lower relative risk, kidney filtration declined more slowly by 1.16 units per year, and deaths from any cause fell 20%. The trial was stopped early for clear benefit and now underpins kidney approvals for Ozempic in the United States (January 2025) and Canada (August 2025).
What is still missing: any test in kidney disease without diabetes, a trial of the weight-management dose for kidney outcomes, a properly powered answer on adding semaglutide to an SGLT2 inhibitor, independent replication outside the manufacturer's program, and any data on whether the protection lasts after treatment stops.
Common questions
Does semaglutide protect the kidneys?
In the FLOW trial, 3,533 adults with type 2 diabetes and chronic kidney disease who took semaglutide had a 24% lower risk of a major kidney event (kidney failure, a fall of at least 50% in kidney filtration, or death from kidney or cardiovascular causes) than those on placebo over a median of 3.4 years. That result applies to the population studied: adults with type 2 diabetes and chronic kidney disease in the filtration and urine protein ranges FLOW enrolled. No trial has shown the same outcome benefit in kidney disease without diabetes.
How much did semaglutide lower kidney events in FLOW?
A primary-outcome event occurred in 331 of 1,767 participants on semaglutide (18.7%) and 410 of 1,766 on placebo (23.2%), a hazard ratio of 0.76, which is a 24% lower relative risk. Kidney function also declined more slowly: the filtration rate fell by 2.19 units per year on semaglutide versus 3.36 on placebo, a difference of 1.16 units per year.
Is Ozempic approved for kidney disease?
Yes, for a specific population. In January 2025 the US FDA approved Ozempic (semaglutide) to reduce the risk of kidney disease worsening, kidney failure, and death due to cardiovascular disease in adults with type 2 diabetes and chronic kidney disease. Health Canada approved the same kidney indication in August 2025. Both approvals rest on FLOW and cover adults with type 2 diabetes and chronic kidney disease, not kidney disease in general.
Did FLOW test semaglutide in people without diabetes?
No. Every FLOW participant had type 2 diabetes as well as chronic kidney disease. The trial says nothing direct about kidney disease in people without diabetes, type 1 diabetes, or children, and it did not enroll people already on dialysis or with the most advanced kidney failure. Benefits in those groups are untested, not disproven.
Does semaglutide add kidney benefit on top of an SGLT2 inhibitor?
FLOW cannot settle that question. Only 550 of 3,533 participants (about 16%) were taking an SGLT2 inhibitor when they entered the trial, which is too few to measure the combination effect precisely. The subgroup analysis found no statistically significant difference in how semaglutide performed with or without an SGLT2 inhibitor, but the confidence interval in the small subgroup was wide. SGLT2 inhibitors have their own kidney outcome trials and remain a separate, proven part of care for this population.
Was the dose in FLOW the weight-loss dose?
No. Researchers administered semaglutide 1.0 mg once weekly in FLOW, the diabetes dose range sold as Ozempic. The higher 2.4 mg weekly dose sold for weight management was not the dose tested for kidney outcomes, so the kidney results should not be quoted as results of the weight-loss dose or of compounded products, which were not studied in FLOW at all.
Sources
Perkovic V, Tuttle KR, Rossing P, et al.; FLOW Trial Committees and Investigators. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes. N Engl J Med. 2024;391(2):109-121. DOI 10.1056/NEJMoa2403347. PMID 38785209. Randomized, double-blind, placebo-controlled trial, n=3,533 (1,767 semaglutide, 1,766 placebo), 387 sites in 28 countries; semaglutide 1.0 mg weekly versus placebo, both added to standard care; median follow-up 3.4 years after early cessation recommended at a prespecified interim analysis; primary composite 331 versus 410 first events, hazard ratio 0.76 (95% CI, 0.66 to 0.88; P=0.0003); funded by Novo Nordisk; ClinicalTrials.gov NCT03819153.
Mann JFE, et al. Effects of semaglutide with and without concomitant SGLT2 inhibitor use in participants with type 2 diabetes and chronic kidney disease in the FLOW trial. Nat Med. 2024. DOI 10.1038/s41591-024-03133-0. Prespecified subgroup analysis: 550 participants using an SGLT2 inhibitor at baseline and 2,983 not using one; primary outcome hazard ratio 1.07 (95% CI, 0.69 to 1.67) in the SGLT2 subgroup and 0.73 (95% CI, 0.63 to 0.85) in the non-SGLT2 subgroup; interaction P=0.109; eGFR slope differences of 0.75 and 1.25 units per year in the two subgroups. Funded by Novo Nordisk as part of the FLOW program.
US Food and Drug Administration, via Novo Nordisk announcement. FDA approves Ozempic (semaglutide) as the only GLP-1 RA to reduce the risk of worsening kidney disease and cardiovascular death in adults with type 2 diabetes and chronic kidney disease. January 28, 2025. Approval to reduce the risk of sustained eGFR decline, end-stage kidney disease, and cardiovascular death, based on FLOW.
Rossing P, et al. The FLOW trial: rationale and design. Nephrol Dial Transplant. 2023. PMID 36651820. Design paper for FLOW; its rationale summarizes the earlier GLP-1 kidney evidence as secondary cardiovascular-trial outcomes driven mainly by albuminuria, which is the basis for the comparison section's account of the pre-FLOW record.
Tirzepatide kidney status: checked October 16, 2026, against PubMed and ClinicalTrials.gov. No dedicated tirzepatide kidney outcomes trial with reported results was identified in that search; the comparison section's statement rests on that dated search, not on a claim that such a trial can never report.
SGLT2 inhibitor kidney outcome trials referenced in the comparison section: CREDENCE (canagliflozin, N Engl J Med 2019; PMID 30990260), DAPA-CKD (dapagliflozin, N Engl J Med 2020; PMID 32970396), and EMPA-KIDNEY (empagliflozin, N Engl J Med 2023; PMID 36331190). Each tested its own drug in its own population; none compared an SGLT2 inhibitor with semaglutide.
Health Canada approval, announced by Novo Nordisk Canada, August 2025. Ozempic approved to reduce the risk of sustained eGFR decline, end-stage kidney disease, and cardiovascular death in adults with type 2 diabetes and chronic kidney disease, the first such approval for a GLP-1 medicine in Canada, based on FLOW.
