Semaglutide and Kidney Disease: The FLOW Trial Results Explained

When a clinical trial is stopped early because the treatment is working so well it would be unethical to continue giving participants a placebo — that is called stopping at a pre-specified interim analysis for overwhelming benefit. It is rare. It happened with the FLOW trial in 2023, and the results were published in the New England Journal of Medicine in May 2024 by Perkovic and colleagues.

For the nephrology community, FLOW is a landmark trial. For the 37 million Americans with chronic kidney disease (CKD), it is potentially practice-changing. And for anyone prescribed semaglutide who has impaired kidney function, it is essential reading. This article explains exactly what FLOW found, why the results matter, how semaglutide protects kidneys at a mechanistic level, and what patients and their physicians should monitor when combining semaglutide with CKD management.

What the FLOW Trial Was: Design and Population

FLOW (Semaglutide in Patients with Type 2 Diabetes and Chronic Kidney Disease) was a multicenter, randomized, double-blind, placebo-controlled trial. It enrolled 3,533 adults with type 2 diabetes and chronic kidney disease, defined by:

• Estimated glomerular filtration rate (eGFR) of 50–75 mL/min/1.73m² with urine albumin-to-creatinine ratio (UACR) ≥ 300 mg/g, OR
• eGFR of 25–50 mL/min/1.73m² with UACR ≥ 100 mg/g

Participants received either subcutaneous semaglutide 1.0 mg weekly or matching placebo, in addition to standard-of-care CKD treatment — which allowed background use of RAAS inhibitors (ACE inhibitors or ARBs) and SGLT2 inhibitors. This is critical context: the benefit seen in FLOW was on top of existing kidney-protective therapies, not instead of them.

The primary composite endpoint was a kidney disease progression event, defined as: sustained ≥50% decrease in eGFR from baseline, end-stage kidney disease (ESKD), kidney-related death, or cardiovascular death. Secondary endpoints included cardiovascular outcomes, all-cause mortality, and measures of eGFR trajectory.

Why the Trial Stopped Early

The Data Safety Monitoring Board (DSMB) reviewed an unblinded interim analysis in 2023 and recommended early termination. At that point, the semaglutide group had crossed the pre-specified boundary for overwhelming benefit — meaning the magnitude of kidney protection was so large that it was statistically unlikely the result would reverse if the trial continued to its planned completion. The ethical principle is straightforward: once you know a treatment works this convincingly, withholding it from placebo participants is no longer justified.

The Key Results: What FLOW Found

The FLOW trial results were published in the New England Journal of Medicine in May 2024 (Perkovic et al.). The headline numbers:

• Primary composite endpoint (kidney progression or cardiovascular death): 24.4% relative risk reduction — hazard ratio 0.76 (95% CI: 0.66–0.88), p<0.001
• ESKD or kidney-related death specifically: also significantly reduced (HR 0.80)
• Cardiovascular death: 20% relative reduction
• All-cause mortality: 18% lower in the semaglutide group (HR 0.82)
• eGFR decline rate: semaglutide group lost eGFR more slowly — approximately 1.4 mL/min/1.73m² per year slower than placebo

To put the primary endpoint number in perspective: a 24% relative risk reduction in a population already receiving maximal background therapy (RAAS inhibitors, many on SGLT2 inhibitors) is a substantial additional benefit. Most new CKD drugs in the past two decades have delivered incremental improvements. Semaglutide's effect size is in the same territory as SGLT2 inhibitors — which are already considered practice-changing in nephrology.

How GLP-1 Drugs Protect Kidneys: The Mechanisms

The size of the benefit seen in FLOW demands a mechanistic explanation. Why would a drug originally developed for glucose control and weight loss produce such substantial kidney protection? The answer is multifactorial — and the mechanisms are complementary rather than redundant.

1. Reduction of Intraglomerular Pressure

The glomerulus — the kidney's filtration unit — is damaged in diabetic CKD in part through chronically elevated filtration pressure. This hyperfiltration causes physical stress on the capillary walls, triggering inflammation and progressive scarring. GLP-1 receptor agonists appear to reduce intraglomerular pressure through two mechanisms: vasodilation of the afferent arteriole (the vessel feeding the glomerulus) and reduction of angiotensin II activity. The net effect is lower mechanical stress on the filtration barrier.

2. Reduction of Systemic Blood Pressure

Semaglutide consistently reduces systolic blood pressure by approximately 3–5 mmHg in clinical trials — a modest but meaningful effect. In CKD, every mmHg of blood pressure reduction matters. Hypertension is both a cause and consequence of CKD progression; semaglutide's antihypertensive effect provides continuous kidney protection above and beyond its glycemic effects.

3. Direct Anti-inflammatory Effects in the Kidney

GLP-1 receptors are expressed in renal tubular cells — not just in the glomerulus — and activation of these receptors produces local anti-inflammatory effects. Specifically, GLP-1R signaling appears to suppress NF-κB pathway activation and reduce production of pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) in kidney tissue. This matters because CKD progression is driven as much by inflammation-mediated fibrosis as by hemodynamic injury.

4. Reduction of Albuminuria

Albuminuria — protein leaking through a damaged glomerular filter into urine — is both a marker and a driver of CKD progression. In FLOW, semaglutide significantly reduced the urine albumin-to-creatinine ratio (UACR), indicating a direct improvement in glomerular barrier function. This is consistent with findings from the SUSTAIN and PIONEER diabetes trials, which showed albuminuria reductions as a secondary endpoint.

5. Glycemic and Metabolic Improvements

Improved glycemic control reduces the formation of advanced glycation end-products (AGEs) — proteins damaged by excess glucose — which accumulate in the kidney matrix and contribute to fibrosis. Lower HbA1c reduces this AGE-mediated damage. Additionally, the weight loss from semaglutide reduces adipokine dysregulation (particularly resistin and TNF-α from visceral fat), which has independent nephrotoxic effects in type 2 diabetes.

GLP-1 Receptor Expression in the Kidney: The Direct Evidence

A key question in GLP-1 nephrology is whether kidney protection is purely mediated through systemic effects (blood pressure, glucose, weight) or whether there is a direct renal action via GLP-1 receptors expressed in kidney tissue. The evidence now points to both.

GLP-1 receptors have been identified in:

• Proximal tubule cells — which handle the bulk of sodium and water reabsorption
• Distal tubule segments
• Vascular smooth muscle of afferent arterioles
• Glomerular mesangial cells (the structural cells within the glomerulus)

Sattar et al.'s review of the mechanisms underlying GLP-1 renal protection identifies a particularly important proximal tubule effect: GLP-1R activation in tubular cells reduces sodium-hydrogen exchanger 3 (NHE3) activity, decreasing sodium reabsorption in the proximal tubule. This is mechanistically similar — though distinct — from the SGLT2 inhibitor mechanism of action, which also reduces proximal tubule sodium reabsorption. The parallel pathways help explain why both drug classes improve kidney outcomes, and why their mechanisms appear complementary rather than redundant when combined.

What FLOW Means for Standard of Care in CKD

Before FLOW, the evidence base for GLP-1 drugs in CKD was derived from secondary analyses of cardiovascular outcome trials — LEADER (liraglutide), SUSTAIN 6 and CREDENCE-adjacent analyses (semaglutide), DECLARE (dapagliflozin). These analyses consistently showed kidney benefit signals, but none were powered or designed to test it as a primary hypothesis.

FLOW changed that. With a dedicated trial showing 24% relative risk reduction in a population specifically selected for CKD, semaglutide's place in the nephrology armamentarium shifted from "may offer kidney benefit" to "demonstrated kidney benefit in a primary trial." The clinical implication is that type 2 diabetic CKD patients who are not on a GLP-1 receptor agonist now have evidence-based grounds to discuss starting one — not just for cardiovascular or glycemic reasons, but for kidney protection as a primary indication.

The major nephrology and endocrinology societies are expected to formally update their guidelines to reflect FLOW. The American Diabetes Association (ADA) Standards of Medical Care already recommend GLP-1 RAs for cardiovascular risk reduction in diabetic CKD; FLOW provides the foundation for a more specific kidney-protection recommendation.

Comparison with CREDENCE: Complementary Mechanisms

CREDENCE (2019) was the landmark kidney outcomes trial for SGLT2 inhibitors, showing canagliflozin produced a 30% relative risk reduction in CKD progression in a type 2 diabetic CKD population. The FLOW results are strikingly similar in magnitude to CREDENCE — which raises the obvious question of whether combining both drug classes amplifies benefit.

The mechanisms suggest they should be complementary:

• SGLT2 inhibitors: primary mechanism is SGLT2 blockade in proximal tubule → reduced sodium reabsorption → tubuloglomerular feedback → afferent arteriolar constriction → lower intraglomerular pressure. Also reduce metabolic acidosis and have direct anti-fibrotic effects.
• GLP-1 RAs: primary renal mechanism includes NHE3 inhibition (distinct sodium transporter), direct anti-inflammatory effects via GLP-1R in tubular cells, reduction of systemic inflammation, blood pressure lowering, and weight loss-mediated reduction of adipokine nephrotoxicity.

In FLOW, approximately 35% of participants were already on an SGLT2 inhibitor at baseline. The kidney benefit from semaglutide was evident regardless of background SGLT2 inhibitor use — suggesting additive (or at minimum non-redundant) effects. This is now one of the strongest arguments for combining both drug classes in type 2 diabetic CKD patients: the mechanisms are genuinely different, and the trial data is consistent with additive benefit.

What Labs to Watch: eGFR Monitoring on GLP-1 with CKD

Patients with CKD starting semaglutide need structured laboratory monitoring. Here is the evidence-based framework for what to track and how to interpret changes:

The Dehydration Risk: Acute Kidney Injury and GLP-1 Drugs

The kidney-protective narrative of FLOW coexists with a legitimate concern: semaglutide causes nausea, vomiting, and reduced oral intake — especially during dose escalation. In CKD patients who are already vulnerable to volume depletion (often on diuretics, sometimes in warm climates, frequently elderly), this nausea-driven dehydration can cause acute kidney injury (AKI).

The FDA added a warning to GLP-1 RA labeling regarding acute kidney injury following case reports. The mechanism is not nephrotoxic — it is pre-renal: reduced fluid intake combined with vomiting causes hypovolemia, which reduces renal perfusion pressure, which reduces eGFR acutely. In a CKD patient with limited renal reserve, this can trigger a meaningful acute-on-chronic decline.

Practical dehydration prevention for CKD patients on semaglutide:

• Set a daily fluid target with your physician (often 2L+ for non-fluid-restricted CKD patients)
• Use electrolyte supplements to replace sodium and potassium lost through reduced intake and any GI symptoms — particularly important when nausea reduces oral intake
• Dose titration should be slow: do not advance dose until GI symptoms from the current dose have resolved
• Check serum creatinine if GI symptoms persist more than 24–48 hours to rule out acute eGFR decline
• Hold nephrotoxic co-medications (NSAIDs, contrast agents) during any acute GI illness period

At-Home Kidney Monitoring: Creatinine Testing

For CKD patients who want to monitor their kidney function more actively between medical appointments, at-home creatinine testing kits have become increasingly available. While these cannot replace a full laboratory panel (which also measures cystatin-C, electrolytes, CBC, and provides more precise eGFR calculation), a home creatinine reading can provide early warning of a significant acute change.

When is home creatinine testing most useful for semaglutide users with CKD?

• During the first 6 months of therapy, when GI symptoms (and thus dehydration risk) are highest
• After any episode of prolonged nausea or vomiting lasting more than 24 hours
• When laboratory visits are inconvenient or infrequent (e.g., rural patients, travel)
• As a trend-tracking tool alongside formal lab visits to detect gradual changes between appointments

Who Should (and Should Not) Be on Semaglutide with CKD

Strong Candidates

Based on FLOW and the broader evidence base, CKD patients most likely to benefit from semaglutide are those who match the FLOW population:

• Type 2 diabetes with moderately reduced eGFR (25–75 mL/min/1.73m²) and significant albuminuria
• Overweight or obese patients where weight loss provides additional metabolic benefit
• Patients with cardiovascular disease or high cardiovascular risk (where SELECT and SUSTAIN 6 data also applies)
• Patients already on RAAS inhibitors who need additional kidney protection beyond what RAAS blockade alone provides

Patients Requiring Careful Assessment

• Advanced CKD (eGFR <25): FLOW enrolled down to eGFR 25; patients below this threshold have limited trial data. The dehydration-driven AKI risk increases as renal reserve diminishes. Individualized risk-benefit assessment is essential.
• Patients on dialysis: ESKD patients on dialysis were not enrolled in FLOW. Pharmacokinetic data for semaglutide in dialysis-dependent patients is limited.
• Type 1 diabetes with CKD: FLOW was exclusively type 2 diabetes. GLP-1 receptor agonists have a different benefit-risk profile in type 1 (including diabetic ketoacidosis risk in some settings).
• Patients with severe GI motility disorders: gastroparesis and severe GI dysmotility are contraindications to semaglutide; in CKD patients where gastroparesis is common (diabetic autonomic neuropathy), this requires assessment before prescribing.

The Bottom Line: A New Standard of Care

FLOW is one of the most significant nephrology trials of the past decade. A 24% relative risk reduction in kidney disease progression — on top of existing standard-of-care treatment including RAAS inhibitors and, in many cases, SGLT2 inhibitors — is a clinically meaningful and trial-validated benefit. The mechanisms are coherent, the effect size is large, and the mortality benefit makes this a drug that prolonged lives in the trial population.

The practical takeaways for patients and clinicians:

• Type 2 diabetic CKD patients should discuss semaglutide with their nephrologist and endocrinologist — not just for glucose control, but as a kidney-protective therapy
• Combining semaglutide with SGLT2 inhibitors is supported by complementary mechanisms and trial data showing benefit regardless of background SGLT2 use in FLOW
• The dehydration/AKI risk during dose escalation is real but manageable with structured monitoring, electrolyte supplementation, and slow titration
• eGFR and UACR monitoring should be structured and proactive — a stable or improving UACR trajectory is a strong signal that semaglutide is working for the kidneys
• The trial was in type 2 diabetes specifically; the evidence does not currently extend to CKD from other causes

For the first time in nephrology, there is now a dedicated kidney outcomes trial proving a GLP-1 receptor agonist prevents kidney disease progression. That is not a secondary finding buried in a cardiovascular trial. It is the primary result of a trial designed to answer exactly this question — and the answer was definitive enough to stop the trial early.