Your Kidneys on GLP-1: What the FLOW Trial Found (and What It Means for You)

A trial stopped early — because the kidney numbers were too good to keep going

When a clinical trial's data safety monitoring board stops a study ahead of schedule, it usually means something went wrong. People got hurt. The drug caused unexpected harm. The trial gets shut down.

The FLOW trial—semaglutide versus placebo in people with type 2 diabetes and chronic kidney disease—stopped early too. But not for that reason.

It stopped because the results were so decisively in favor of semaglutide that continuing to give half the participants a placebo was no longer ethically defensible. At a median of 3.4 years into a planned longer trial, the data safety monitoring board saw the numbers and said: enough. We know.

What did they know? That semaglutide 1.0 mg weekly—the Ozempic dose, not the higher Wegovy dose—cut the risk of major kidney disease events by 24% compared to placebo. The full results landed in the New England Journal of Medicine in May 2024. If you have type 2 diabetes and CKD, these numbers are worth understanding.

What "major kidney disease events" actually means

The FLOW trial wasn't measuring vague outcomes. The primary endpoint was a composite of four specific, hard events: a sustained drop of 50% or more in eGFR (a key measure of how well your kidneys filter), a sustained eGFR below 15 (kidney failure territory), starting dialysis or getting a kidney transplant, or dying from kidney or cardiovascular causes.

These aren't soft endpoints. These are the things nephrologists and patients spend years trying to prevent. The trial enrolled 3,533 adults with both type 2 diabetes and CKD—defined as an eGFR between 25 and 75 mL/min/1.73m²—which means it covered a wide range, from modestly reduced kidney function all the way to near-end-stage disease.

The hazard ratio came back at 0.76, with a 95% confidence interval of 0.66 to 0.88, and a p-value below 0.001. That's not a borderline result. The confidence interval doesn't get close to 1.0.

Beyond the headline event reduction, the trial tracked eGFR decline over time. In the semaglutide group, the rate of kidney function decline slowed by approximately 1.16 mL/min/1.73m² per year compared to placebo. That might sound small in isolation, but kidney function doesn't regenerate. Slowing the rate of decline is the ballgame.

The FLOW trial results at a glance

The weight loss number deserves a brief note: people in the semaglutide arm lost roughly 10% of body weight, which is meaningful on its own. That raises a reasonable question—how much of the kidney benefit is just from losing weight and controlling blood sugar better? Probably some of it. But the trial data and mechanistic research suggest something more is happening, which is why the "how it works" question matters.

How GLP-1s may protect the kidneys

The short version: it's probably multiple mechanisms hitting simultaneously, not one clean pathway.

Blood sugar control is the oldest part of the story. Chronically elevated glucose damages the tiny blood vessels inside the kidney's filtering units (the glomeruli), a process called glycotoxicity. GLP-1 receptor agonists lower blood sugar—that's their original FDA-approved reason to exist—so some kidney protection follows naturally from better glucose management.

Blood pressure reduction is the second piece. Lower blood pressure means less mechanical stress on the glomeruli, which reduces the progressive scarring that CKD patients spend years trying to slow. Semaglutide modestly reduces blood pressure; every millimeter helps when the kidneys are already under strain.

Then there's weight. Excess weight raises intraglomerular pressure—the pressure inside the kidney's tiny filters—independently of blood sugar. The 10% weight loss observed in FLOW matters for the kidneys on its own terms, not just as a marker of overall health improvement.

But here's where it gets more interesting: GLP-1 receptors exist in kidney tissue. They've been found in the proximal tubule cells and in mesangial cells. This opens the door to direct anti-inflammatory and antioxidant effects that are separate from what's happening with weight, blood sugar, or blood pressure. The research on exactly how these direct effects play out is still being worked out, but the presence of GLP-1 receptors in the kidneys is established. Something is happening locally, not just systemically.

Albuminuria—protein leaking into the urine, a marker of kidney damage—also decreased in the semaglutide arm of FLOW. That's a secondary finding, but albuminuria reduction has historically predicted slower progression to kidney failure, and it aligns with the primary endpoint data.

Six mechanisms, working in parallel:

1. Blood sugar control (reduces glycotoxicity)
2. Blood pressure reduction (reduces glomerular mechanical stress)
3. Weight loss (lowers intraglomerular pressure)
4. Direct anti-inflammatory effects via kidney GLP-1 receptors
5. Reduced albuminuria (less protein leaking through damaged filters)
6. Reduced oxidative stress (cellular-level damage control)

The SGLT2 inhibitor question — because your nephrologist will bring it up

If you have CKD and you're talking to a kidney specialist in 2026, you've almost certainly heard about SGLT2 inhibitors. Dapagliflozin (Farxiga) and empagliflozin (Jardiance) both have FDA-approved CKD indications. They're not off-label for kidney disease—they're actually, formally approved for it. The DAPA-CKD trial (dapagliflozin, 2020) showed a 39% reduction in kidney disease progression or death among people with CKD, including those without diabetes. CREDENCE (canagliflozin) and EMPA-KIDNEY (empagliflozin) added more supporting data across the same years.

This is the current standard of care. SGLT2 inhibitors came first, have the clearest regulatory approval, and are what most nephrologists reach for first.

The FLOW data for semaglutide doesn't erase that. It adds to it. The question being asked at nephrology conferences right now isn't "SGLT2i or GLP-1?"—it's "should high-risk patients be on both?" Early combination data is emerging. The two drug classes work through largely different mechanisms, and they may stack.

If your nephrologist isn't already on top of SGLT2 inhibitors for your CKD, that's worth raising specifically. The FLOW data for semaglutide is compelling, but SGLT2i approval came first, and it's where most kidney-specific conversations start.

About 850 million people have CKD — and most don't know it

Chronic kidney disease affects roughly 850 million people globally. In the United States, about 37 million adults—around 15% of the adult population—have CKD. The majority don't know. Kidney disease is famously silent until it's not: eGFR can decline for years with no symptoms, and by the time people feel sick, the damage is substantial.

Diabetic kidney disease is the number one cause of end-stage renal disease (ESRD) worldwide. In the US, diabetes accounts for about 38% of new ESRD cases each year. This is why FLOW is such a charged result: the intersection of type 2 diabetes and CKD is enormous, the progression to dialysis or transplant is devastating and expensive, and any drug that slows that trajectory has outsized population-level impact.

For context: starting dialysis changes your life in ways that are hard to overstate. Three to four sessions per week, each lasting three to five hours. Strict fluid and dietary restrictions. A significant increase in cardiovascular mortality. Average life expectancy on dialysis in the US is 5–10 years. The goal is always to delay or avoid it. A 24% reduction in the events leading there is not a footnote.

The realistic picture — no approval yet, cost barriers are real

There's no sugarcoating this part: as of May 2026, no GLP-1 medication has FDA approval for chronic kidney disease. Not semaglutide, not tirzepatide, not any of them. The FLOW data is Phase 3 and it's published in the NEJM—that's the quality of evidence that typically supports an FDA filing—but Novo Nordisk hasn't received a CKD indication yet.

What that means practically: if you and your doctor decide semaglutide makes sense for your kidneys, it's off-label prescribing. Your doctor can absolutely do this—it's legal, and physicians prescribe off-label constantly—but your insurance coverage situation gets complicated.

Ozempic is usually covered when diabetes is the diagnosis. If your plan covers it for T2D and you also have CKD, the CKD benefit may come along for free, so to speak. But if you're trying to get GLP-1 coverage specifically citing CKD as the indication, that's a harder conversation with your insurer, and prior authorization denials are common.

On the cost side: Ozempic lists at roughly $900–$1,000 per month without insurance. Wegovy is around $1,349. Without coverage, that's the number. Medicare has been expanding GLP-1 coverage for cardiovascular and obesity indications—changes taking effect in mid-2026—but CKD is not yet in that expansion.

GoodRx and NovoCare (Novo Nordisk's patient assistance program) can reduce costs significantly for people who qualify. The eligibility thresholds vary, so it's worth checking directly rather than assuming you don't qualify.

If you're waiting for full FDA approval before bringing this up with your doctor, you might wait years. The data is out there now. Showing your nephrologist or endocrinologist the FLOW trial and asking "does this apply to me?" is a reasonable thing to do at your next appointment.

What to check before starting — the practical list

Before semaglutide makes sense for someone with CKD, a few baseline numbers matter.

eGFR (estimated glomerular filtration rate): This is your kidney function score. FLOW enrolled people with eGFR 25–75. If your eGFR is below 15, you're in kidney failure territory, and the prescribing conversation is different. If it's above 60, your CKD is early-stage and the urgency may be lower. The key number is where you are on the slope—how fast is your eGFR declining year over year?

UACR (urine albumin-to-creatinine ratio): This measures how much protein is leaking into your urine, which reflects how damaged your kidney filters are. Elevated UACR above 30 mg/g is abnormal; above 300 mg/g is in the high-risk zone. Semaglutide reduced albuminuria in FLOW, so knowing your baseline UACR helps you and your doctor track the drug's kidney effects over time.

Current CKD medications: Are you already on an SGLT2 inhibitor (Farxiga, Jardiance)? An ACE inhibitor or ARB for blood pressure? Your nephrologist will want to see the full picture before adding a GLP-1.

HbA1c and blood sugar control: FLOW required T2D enrollment. If your diabetes is poorly controlled, that's the primary argument for semaglutide, and the kidney benefit is additive.

Blood pressure: Both elevated BP and the blood pressure reduction from semaglutide matter. If you're already on multiple antihypertensives, adding a drug that further lowers BP requires monitoring.

Insurance and formulary status: Before you get attached to a treatment plan, find out where semaglutide sits on your plan's formulary. A quick call to your insurance's pharmacy benefits line, or checking through your patient portal, will tell you what tier it's on and whether a prior authorization is needed.

Questions to bring to your doctor

These are specific, not generic. The goal is a conversation, not a pamphlet.

• "My eGFR has been declining at roughly X points per year. Given the FLOW trial data, does adding semaglutide make sense for slowing that progression?"
• "I'm currently on [SGLT2i/ACE inhibitor/ARB]. Is there evidence supporting combining any of those with semaglutide for CKD?"
• "What UACR level should I aim for, and would you re-check it after three to six months on semaglutide to see if it's moving?"
• "Would you prescribe semaglutide off-label for kidney protection, or do we need to wait for an FDA CKD indication? What's your read on the FLOW data?"
• "My insurance requires prior authorization for Ozempic. Can we list type 2 diabetes as the primary indication and document CKD as a secondary motivation?"
• "What are the signs that semaglutide isn't working well for my kidneys, and what would trigger a change in plan?"
• "Is there any reason my level of CKD would make the nausea side effects harder to manage?" (Kidney function affects drug clearance, and that's worth asking about directly.)

"I've been following my kidney numbers for four years. The FLOW trial was the first time I felt like there might be a drug that actually slows this down rather than just managing symptoms."

— Shared in an online CKD patient community, early 2025

The SELECT trial adds another data point

FLOW was the headliner for kidney disease, but it wasn't the only trial looking at semaglutide's kidney effects. The SELECT trial—which tested semaglutide 2.4 mg (the Wegovy dose) in people with obesity and established cardiovascular disease, but without diabetes—also tracked kidney function as a secondary endpoint. eGFR declined more slowly in the semaglutide group compared to placebo, even in people without diabetes.

That's a meaningful extension of the FLOW story. FLOW showed kidney protection in people with type 2 diabetes and CKD. SELECT suggests the kidney signal might be there in people with obesity alone—no diabetes required. It's not the primary endpoint of that trial, and it shouldn't be read as equivalent to FLOW's primary evidence, but it points in a consistent direction.

The full picture of who benefits most from GLP-1 kidney protection is still being drawn. Phase 3 data in non-diabetic CKD populations would close a lot of open questions.

This fits a pattern — GLP-1s keep turning up in unexpected places

If you've been paying attention to the GLP-1 research coming out over the past three years, the kidney data fits a pattern. These drugs were approved for blood sugar control. Then weight loss. Then cardiovascular risk reduction (the SOUL trial and related CV data showed a 20% reduction in major cardiac events). Then fatty liver disease data—three separate NEJM papers in 2024. And now kidneys.

The common thread is that GLP-1 receptors are not only in the pancreas and brain. They're distributed through the body—in the heart, liver, kidneys, and brain. Drugs that target these receptors seem to have downstream effects that nobody fully anticipated when semaglutide was being developed as a diabetes drug. Whether this represents a class effect or a molecule-specific effect is still being sorted out, but the consistency of the signals across organ systems is hard to dismiss.

For people wondering whether they qualify for GLP-1 treatment in the first place, CKD context may be increasingly relevant to that conversation—even though it's not yet an FDA-approved indication.

What to realistically expect if you start

If you have type 2 diabetes and CKD and your doctor prescribes semaglutide, here's what the first few months typically look like:

The dose starts low. Semaglutide is initiated at 0.25 mg weekly and titrated upward slowly—typically to 0.5 mg after four weeks, then 1.0 mg after another four weeks. The FLOW trial used 1.0 mg, which is the standard Ozempic maintenance dose, not the higher Wegovy doses.

Nausea is the most common early side effect. It hits roughly 20–30% of people in the first four to six weeks, then usually subsides. Eating smaller meals, avoiding high-fat foods, and not lying down immediately after eating all help. For most people, it's manageable. A minority find it severe enough to pause or stop.

eGFR can initially dip slightly when starting semaglutide, similar to what happens with SGLT2 inhibitors. This is generally not a true decline—it reflects a hemodynamic change in kidney blood flow—and it typically stabilizes. But it's worth flagging with your doctor so they're not alarmed by a short-term eGFR blip on your next lab panel.

The kidney benefits observed in FLOW took time to accumulate. The trial ran 3.4 years before stopping early. You won't see a dramatic eGFR improvement in three months. The goal is a slower rate of decline, tracked over years—not a reversal you'll notice in a single lab draw.

Tracking your kidney numbers while on GLP-1

If you're using Blueshot to log your medication, there's a practical case for also tracking the numbers your nephrologist follows—not to replace their monitoring, but to give you a clearer picture of your own trajectory over time.

The core numbers to know:

• eGFR: Target is stability, or slower-than-expected decline. A rough benchmark is that CKD typically progresses at 1–5 mL/min/1.73m² per year; slower than that on treatment is a win.
• UACR: Anything moving from high toward lower categories is a positive signal.
• HbA1c: Better glucose control typically correlates with slower CKD progression.
• Blood pressure (systolic): Below 130 mmHg is the current target for most people with CKD and diabetes, per ADA/KDIGO guidelines.

Logging your medication timing alongside your labs gives you a long-term picture that's easy to review with your nephrologist, rather than reconstructing your history from memory at each appointment.

The FLOW trial stopped early because the evidence was clear enough. Semaglutide reduced the risk of major kidney events by 24%—dialysis, kidney failure, death from kidney or cardiovascular causes—in people with type 2 diabetes and CKD, with a confidence interval that doesn't touch 1.0.

That result doesn't come with an FDA approval stamp for kidney disease yet. It comes with off-label complexity, prior authorization battles, and a monthly list price that makes plenty of people's eyes water. SGLT2 inhibitors are still the first-line kidney-protective drug for most people, and rightly so—they got there earlier with clearer regulatory approval.

But the data is out there, it's NEJM-quality, and it's worth a real conversation with your nephrologist or endocrinologist—not just a passing mention. If you've been watching your eGFR decline for years, the FLOW trial is the most direct evidence available that a GLP-1 drug might slow that trajectory.

Bring the trial to your next appointment. Ask the specific questions. The kidney damage clock doesn't wait for the FDA to catch up.

This article is for informational purposes only and does not constitute medical advice, diagnosis, or treatment. All GLP-1 medications discussed are prescription drugs — do not start, stop, or change any medication without consulting your doctor. Individual results vary. For the most current prescribing information, refer to the FDA-approved labeling for each drug.

Start managing your GLP-1 with Blueshot

AI coaching, injection scheduling, and weight tracking in one app