Skip to content
GLP Loss
← Research
Evidence

GLP-1 and Kidney Disease: What the FLOW Trial Found

FLOW randomized 3,533 adults with type 2 diabetes and albuminuric kidney disease. Major kidney events occurred in 331 on semaglutide against 410 on placebo, a hazard ratio of 0.76.

Owen Castellanos8 min read
FLOW: major kidney disease events over 3.4 years3,533 adults with type 2 diabetes and chronic kidney diseaseSemaglutide 1.0 mg331Placebo410Hazard ratio 0.76, 95% CI 0.66 to 0.88.331 first events on semaglutide against 410 on placebo.Annual eGFR decline was slower by 1.16 mL per minute.

Kidney disease is slow, silent and expensive, and the treatments that genuinely slow it are few. Semaglutide joined that short list in 2024 on the strength of a trial designed around kidney endpoints rather than reporting them as a secondary curiosity. That evidence sits behind the molecule sold across the semaglutide sellers, and it is narrower than the headlines suggested.

Who FLOW enrolled

Participants had type 2 diabetes plus chronic kidney disease, defined by two entry routes. One required an estimated glomerular filtration rate of 50 to 75 alongside a urinary albumin-to-creatinine ratio above 300. The other required a rate of 25 to below 50 with a ratio above 100.[1]Both routes demand measurable protein in the urine. This was not a trial of people whose kidneys were merely at risk.

A total of 3,533 participants were randomized, 1,767 to subcutaneous semaglutide at 1.0 mg weekly and 1,766 to placebo. Median follow-up reached 3.4 years, and the trial was stopped early after a prespecified interim analysis recommended it.[1]

What the trial found

The primary outcome combined the onset of kidney failure, a fall of at least 50% in filtration rate from baseline, and death from kidney-related or cardiovascular causes. First events occurred in 331 participants on semaglutide against 410 on placebo, a hazard ratio of 0.76 (95% CI, 0.66 to 0.88).[1]

The relative figure is the one that traveled. In absolute terms those counts are 18.7% against 23.2%, a gap of roughly 4.5 percentage points over a median 3.4 years. On those rates, about 22 participants were treated for that period for each first event avoided, which is a strong number for a chronic disease with few effective treatments and a smaller promise than a quarter fewer events sounds.

Stripping the composite back to its kidney-specific components alone still favored semaglutide, at a hazard ratio of 0.79 (95% CI, 0.66 to 0.94). Death from cardiovascular causes carried a hazard ratio of 0.71 (95% CI, 0.56 to 0.89), major cardiovascular events 0.82 (95% CI, 0.68 to 0.98) and death from any cause 0.80 (95% CI, 0.67 to 0.95).[1]

The mechanical figure underneath all of that is the filtration slope. The mean annual decline was less steep by 1.16 mL per minute per 1.73 m² on semaglutide. Serious adverse events were reported in 49.6% of the semaglutide group against 53.8% on placebo, which is a sicker-population number and a reminder that everyone here was unwell at baseline.[1]

The class, not only the molecule

A meta-analysis of randomized trials pooled 11 studies and 85,373 participants, 29,386 of them female. Among the 67,769 with type 2 diabetes, GLP-1 receptor agonists reduced a composite kidney outcome by 18% against placebo (hazard ratio 0.82; 95% CI, 0.73 to 0.93) and kidney failure by 16% (hazard ratio 0.84; 95% CI, 0.72 to 0.99).[2]

Adding the cardiovascular outcome trial run in participants without diabetes barely moved those numbers, at 0.81 and 0.84 respectively, with no evidence of heterogeneity between that trial and the diabetes studies.[2] Serious adverse events did not differ between drug and placebo (risk ratio 0.95; 95% CI, 0.90 to 1.01), while stopping because of adverse events was half again as likely (risk ratio 1.51; 95% CI, 1.18 to 1.94).

That last figure is the practical one, and it reappears on the bill. Tolerability decides how long treatment lasts, which is why it belongs next to any plan priced on an annual prepayment — see sellers that hold one price.

The signal arrived earlier

Liraglutide's outcome trial reported prespecified secondary renal results from 9,340 randomized patients over a median 3.84 years. The renal composite occurred in 268 of 4,668 on liraglutide against 337 of 4,672 on placebo, a hazard ratio of 0.78 (95% CI, 0.67 to 0.92).[3]

That result was driven mainly by new persistent macroalbuminuria, which appeared in 161 participants against 215 (hazard ratio 0.74; 95% CI, 0.60 to 0.91). Renal adverse event rates were similar in both arms, at 15.1 and 16.5 events per 1,000 patient-years, with acute kidney injury at 7.1 and 6.2.[3] The semaglutide diabetes trial had also recorded lower rates of new or worsening nephropathy, alongside the retinopathy finding discussed in the cardiovascular article.[4]

What none of this establishes

The dose that produced the FLOW result was 1.0 mg weekly. The dose most readers are quoted for weight loss is 2.4 mg, and the kidney trial did not test it. Neither did it test tirzepatide, which has no reported kidney outcome trial of its own.

A composite endpoint also needs reading component by component. Death from kidney-related or cardiovascular causes sits inside the primary outcome here, so part of what moved the headline ratio is the same cardiovascular effect measured elsewhere in this class. The separate kidney-only ratio of 0.79 is the figure that answers a kidney question, and it is the weaker of the two.[1]

Nor does a kidney benefit in albuminuric disease mean the class is gentle on kidneys in general. Repeated vomiting and the dehydration that follows can precipitate acute kidney injury, which is a labeled concern for the whole class and one of the reasons the schedule in the titration article climbs slowly. Existing kidney disease is a prescriber conversation, not a checkout decision.

Every trial above studied branded product at labeled doses under supervision. Compounded vials, which is what most sellers on the review index dispense, have never been through that review, and a hazard ratio does not transfer by sharing a molecule name. How each figure here was checked is set out in the methodology.

Frequently asked

Does semaglutide protect the kidneys?
In FLOW it slowed kidney disease in people who already had it. Among 3,533 adults with type 2 diabetes and albuminuric chronic kidney disease, first major kidney events occurred in 331 on semaglutide against 410 on placebo, a hazard ratio of 0.76 over a median 3.4 years.
What dose produced the FLOW kidney result?
Semaglutide 1.0 mg once weekly, the diabetes dose. The 2.4 mg dose used for weight management was not tested in that trial, so the kidney hazard ratio should not be quoted as a property of a weight-loss prescription.
Do GLP-1 drugs ever harm the kidneys?
Acute kidney injury from dehydration is a labeled concern for the class, usually following repeated vomiting or diarrhea during dose escalation. In the liraglutide renal analysis, acute kidney injury rates were 7.1 and 6.2 events per 1,000 patient-years on drug and placebo.
Is the kidney benefit a class effect?
A meta-analysis of 11 trials and 85,373 participants found a composite kidney outcome reduced by 18% (hazard ratio 0.82) and kidney failure by 16% across GLP-1 receptor agonists in type 2 diabetes. Tirzepatide has no reported kidney outcome trial of its own.

Sources

  1. [1] Perkovic V, Tuttle KR, Rossing P, et al. (2024). Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes. N Engl J Med. PMID 38785209
  2. [2] Badve SV, Bilal A, Lee MMY, et al. (2025). Effects of GLP-1 receptor agonists on kidney and cardiovascular disease outcomes: a meta-analysis of randomised controlled trials. Lancet Diabetes Endocrinol. PMID 39608381
  3. [3] Mann JFE, Ørsted DD, Brown-Frandsen K, et al. (2017). Liraglutide and Renal Outcomes in Type 2 Diabetes. N Engl J Med. PMID 28854085
  4. [4] Marso SP, Bain SC, Consoli A, et al. (2016). Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. PMID 27633186

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence