Semaglutide Improves Kidney Outcomes in Type 2 Diabetes

Start with the event rates

My notebook has 18.7% and 23.2% written side by side.

Those are the rates of FLOW’s primary outcome in the semaglutide and placebo groups. The gap is 4.5 percentage points, which works out to an approximate number needed to treat of 22 over the trial’s median follow-up of 3.4 years. That estimate needs some restraint: FLOW enrolled people at high risk, it was driven by the number of events, and it stopped early.

The figure that received more attention was 24%, the relative reduction in the primary outcome. I wrote that above the two event rates, though I find myself returning to the percentages below it because they show what happened in each group rather than giving only the relative comparison.

FLOW included 3,533 adults with type 2 diabetes and albuminuric chronic kidney disease. Participants received once-weekly semaglutide 1.0 mg or matching placebo in addition to standard care. The primary outcome combined kidney failure, a sustained eGFR reduction of at least 50%, and death from kidney or cardiovascular causes.

The hazard ratio was 0.76, with a 95% confidence interval of 0.66 to 0.88 and a P value of 0.0003.

This changes the evidence around semaglutide. Earlier diabetes and cardiovascular trials had reported encouraging kidney findings with glucagon-like peptide-1 receptor agonists, particularly less albuminuria and fewer composite kidney events, but progression of kidney disease was usually a secondary question. FLOW was designed around it.

What else moved in FLOW

The primary result was not carried by cardiovascular deaths alone. A kidney-specific composite covering kidney failure, a sustained eGFR decline of at least 50%, or kidney-related death also favored semaglutide, with a hazard ratio of 0.79. Cardiovascular death had a hazard ratio of 0.71.

Major cardiovascular events occurred less often among participants assigned semaglutide. Death from any cause was lower, too. These outcomes matter in diabetic kidney disease, where declining kidney function may be only one part of a person’s medical history and where cardiovascular illness can shape treatment decisions just as much as a lab result does.

Serious adverse events were less common overall with semaglutide than with placebo. More participants in the semaglutide group stopped treatment because of adverse events, however, with gastrointestinal effects contributing to that difference. Investigators found no new kidney-specific safety pattern, and the broader safety findings were consistent with previous experience with this drug class.

Where it might fit, and where FLOW is quiet

FLOW did not compare semaglutide directly with an SGLT2 inhibitor. It did not compare it with finerenone either.

That matters more than the tidy 24% in my notebook might suggest. SGLT2 inhibitors already have extensive evidence for slowing chronic kidney disease progression, along with evidence involving heart-failure outcomes, and their kidney data extend to people who do not have type 2 diabetes. Finerenone has shown kidney and cardiovascular benefits in adults with type 2 diabetes and albuminuric chronic kidney disease who were receiving renin-angiotensin system blockade.

Only a minority of FLOW participants were taking an SGLT2 inhibitor when they entered the trial, reflecting the years in which enrollment occurred. Use of nonsteroidal mineralocorticoid receptor antagonists was limited as well. FLOW therefore tells us much more about semaglutide versus placebo on a background made up largely of renin-angiotensin system blockade than it tells us about adding semaglutide to the fuller treatment mix some patients receive now.

The drugs act through different pathways, and their roles may turn out to be complementary. FLOW leaves the sequence unsettled. It also leaves open the size of any added benefit when these treatments are combined, along with questions about tolerability and cost.

Semaglutide now has a dedicated kidney outcome trial as well as established cardiovascular evidence and effects on glucose and weight. SGLT2 inhibitors bring their own broad kidney and heart-failure record. Finerenone has evidence in persistent albuminuric diabetic kidney disease, while potassium and kidney function require attention during treatment. Those facts do not arrange themselves into one obvious order.

Who entered the trial

FLOW was randomized, double-blind, and placebo-controlled. Participants had type 2 diabetes and chronic kidney disease defined through combinations of reduced eGFR and substantial albuminuria.

People could generally qualify with an eGFR from 25 to less than 50 mL/min/1.73 m² and a urinary albumin-to-creatinine ratio above 100 mg/g. For those with an eGFR from 50 to 75, the ratio needed to be above 300 mg/g. The upper albuminuria limit was 5,000 mg/g.

Participants were generally expected to be receiving renin-angiotensin system blockade at the highest tolerated labeled dose unless they could not take it. Semaglutide and placebo were added to that care, which is an important part of reading the result: the study did not test semaglutide as a stand-alone replacement for established kidney-protective treatment.

The closest match to the trial population is therefore an adult with type 2 diabetes, an eGFR of at least 25 mL/min/1.73 m², and albuminuric chronic kidney disease who is generally receiving tolerated renin-angiotensin system blockade. FLOW gives less certainty for people whose kidney disease does not include substantial albuminuria. It also does not establish the same benefit in type 1 diabetes, after a kidney transplant, during dialysis-dependent kidney failure, or below the trial’s eGFR entry threshold.

The other number in the notebook

FLOW stopped early after a planned interim analysis crossed its prespecified threshold for efficacy. Mean eGFR declined by 2.19 mL/min/1.73 m² per year in the semaglutide group and 3.36 in the placebo group.

I boxed the difference: 1.16 mL/min/1.73 m² per year.

The eGFR slope gives a second view of the trial. The primary outcome counted major clinical events, while the slope indicates that filtration declined more slowly on average in participants assigned semaglutide. A slower average slope is not the same as showing that an individual patient avoided kidney failure, although its direction is consistent with the kidney event results.

The early stop still creates uncertainty about how large and durable the benefit would be over a longer period. The primary outcome also joined kidney disease progression with kidney death and cardiovascular death, meaning the 24% relative reduction cannot be assigned to each part of the composite. The separate kidney-specific and cardiovascular findings were favorable, but they are not interchangeable.

The manufacturer funded FLOW. That role and the investigators’ disclosures belong in the reading of the report. Replication and studies using routine clinical care could address questions that an industry-funded randomized trial, even a large one with a clear primary result, cannot settle on its own.

Nor can FLOW divide the observed kidney benefit into neat shares attributable to weight, glucose, blood pressure, inflammation, or direct kidney pathways. I left space beneath that question in the notebook. The trial measured outcomes, not a separate contribution from every possible mechanism.

Questions people ask

How much did semaglutide reduce kidney and cardiovascular risk in FLOW?

FLOW randomized 3,533 adults to weekly semaglutide or placebo alongside standard care. During a median follow-up of 3.4 years, the primary outcome occurred in 18.7% of the semaglutide group and 23.2% of the placebo group. That was a 24% relative reduction. The kidney-specific composite favored semaglutide, and average eGFR decline was slower.

Does semaglutide replace SGLT2 inhibitors or finerenone for chronic kidney disease?

FLOW did not test replacement and did not directly compare semaglutide with either treatment. Relatively few participants were already taking an SGLT2 inhibitor, while combined treatment with finerenone was limited. The trial supports semaglutide as another therapy with demonstrated kidney benefit in the population studied, but the preferred order and the effects of combining these drugs remain unsettled.

Which patients were included in FLOW?

The trial enrolled adults with type 2 diabetes, albuminuric chronic kidney disease, and an eGFR of at least 25 mL/min/1.73 m². Participants were generally expected to receive tolerated renin-angiotensin system blockade. Applying the result is less certain in nonalbuminuric disease and in populations excluded from the trial, including people receiving dialysis.

Questions clinicians ask

Does FLOW show that semaglutide prevents kidney failure?

FLOW found a lower risk of a composite outcome that included kidney failure, sustained major eGFR loss, and death from kidney or cardiovascular causes. Its kidney-specific composite also favored semaglutide.

That does not mean every patient receiving semaglutide will avoid kidney failure. It is not a lifetime estimate. Median follow-up was 3.4 years, and the study stopped early.

Should semaglutide come before an SGLT2 inhibitor?

The trial cannot answer that. Its comparator was placebo, not an SGLT2 inhibitor, and SGLT2 inhibitors remain supported by dedicated trials across broader chronic kidney disease populations, including substantial evidence on heart-failure outcomes.

Can the same benefit be assumed for someone already taking an SGLT2 inhibitor and finerenone?

No. FLOW included relatively few participants taking an SGLT2 inhibitor at enrollment, and use of finerenone or another nonsteroidal mineralocorticoid receptor antagonist was limited. A biological reason for therapies to fit together is not proof of the added benefit, cost, tolerability, or best order of the combination.

At the bottom of the notebook page, below the boxed eGFR slope difference and the open space about mechanism, the event rates remain where I started: 18.7% and 23.2%.

References

1. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes — The New England Journal of Medicine, 2024
2. Semaglutide in Participants With Type 2 Diabetes and Chronic Kidney Disease (FLOW) — ClinicalTrials.gov, 2024
3. Diabetes Management in Chronic Kidney Disease: A Consensus Report by the American Diabetes Association and Kidney Disease: Improving Global Outcomes — Diabetes Care, 2022
4. Dapagliflozin in Patients with Chronic Kidney Disease — The New England Journal of Medicine, 2020
5. Empagliflozin in Patients with Chronic Kidney Disease — The New England Journal of Medicine, 2023
6. Effect of Finerenone on Chronic Kidney Disease Outcomes in Type 2 Diabetes — The New England Journal of Medicine, 2020