Frontier scope (kept peripheral, not deepened) — semaglutide is a widely-prescribed standard drug, so its efficacy on a hard outcome is admissible (the Pharmacotherapy taper); but this page sits at the prescriber edge because the population is a specific diagnosed disease (type 2 diabetes with established CKD, on maximal-dose RAS inhibition), not one of the reasonably-healthy or prevalent-lifestyle-modifiable strata the wiki centres — and the finding, by the page’s own reading, does not transport to the obese primary-prevention case. Held for the standard-drug efficacy signal and the surrogate-vs-hard-outcome lesson; not deepened (no dosing, agent-selection, or combination cardiorenal-therapy content, and no further FLOW-class kidney-trial acquisition).
The GLP-1 case elsewhere in this cluster is about weight and cardiovascular events in obesity (Semaglutide for Cardiovascular Risk in Obesity) and the non-cardiometabolic safety ledger (GLP-1 Non-Cardiometabolic Effects and Safety). FLOW opens a third, distinct benefit on a different endpoint in a different stratum: a hard kidney outcome in type 2 diabetes with established chronic kidney disease. It is the first dedicated hard-outcome kidney trial of a GLP-1 receptor agonist — «previous dedicated trials addressing clinically important kidney outcomes, such as kidney failure or a substantial decline in the eGFR, have been lacking» (Perkovic et al., 2024).
The stratum fixes what the number means
FLOW enrolled a narrow, high-risk population, not the obese primary-prevention strata most of the wiki’s other GLP-1 pages address: type 2 diabetes + CKD (eGFR 25-75 ml/min/1.73m² with elevated albuminuria), already on a maximal-dose RAS inhibitor, mean eGFR 47, 68% at very-high KDIGO risk, mean age 66.6, mean BMI 32. Semaglutide 1.0 mg SC weekly (the diabetes/glycaemic dose, not the 2.4 mg obesity dose) vs placebo; n=3533; median follow-up 3.4 years; stopped early for efficacy. This is a secondary-prevention CKD stratum — the benefit below transports to that profile, and the trialists flag it «may not be generalizable to other populations, such as those at lower risk».
What FLOW proved — hard outcomes, absolute and relative
| Outcome | Effect (HR, 95% CI) | Absolute | Hard or surrogate |
|---|---|---|---|
| Primary: major kidney disease events (composite) | 0.76 (0.66-0.88), P=0.0003 | 5.8 vs 7.5 events/100 pt-yr; NNT 20 over 3 yr (14-40) | mixed — hard failure + surrogate threshold + death |
| Kidney-specific composite | 0.79 (0.66-0.94) | — | mixed |
| Death from cardiovascular causes | 0.71 (0.56-0.89) | — | hard |
| Major cardiovascular events (MACE) | 0.82 (0.68-0.98), P=0.029 | NNT 45 over 3 yr | hard |
| Death from any cause | 0.80 (0.67-0.95), P=0.01 | NNT 39 over 3 yr | hard |
| Total eGFR slope | +1.16 ml/min/1.73m²/yr (0.86-1.47), P<0.001 | −2.19 vs −3.36 /yr | surrogate |
- «331 first events [5.8 per 100 patient-years of follow-up] vs. 410 first events [7.5 per 100 patient-years]), which resulted in a 24% lower relative risk of the primary outcome in the semaglutide group (hazard ratio, 0.76; 95% confidence interval [CI], 0.66 to 0.88; P = 0.0003)».
- «The number of persons who would need to be treated over 3 years to prevent one primary-outcome event was 20 (95% CI, 14 to 40).»
- «The risk of death from any cause was 20% lower in the semaglutide group than in the placebo group (227 vs. 279 events; hazard ratio, 0.80; 95% CI, 0.67 to 0.95, P = 0.01)».
Read the composite carefully — it is not all hard kidney failure. The primary endpoint pools kidney failure (dialysis/transplant/eGFR<15 — patient-important) with a ≥50% eGFR decline (a surrogate threshold, though a validated one) and death. The eGFR slope is a surrogate for future function loss -> Surrogate Outcomes. What is unambiguously hard-and-patient-important here is the all-cause and CV mortality reduction — and «the trial was not powered to separately detect effects on kidney failure». So the honest headline is reduced a composite of kidney events and reduced death, not prevented dialysis as a standalone proven claim.
Safety was reassuring, and the mechanism is weight-independent
Serious adverse events were lower on semaglutide (49.6% vs 53.8%), driven by fewer serious infections and CV disorders — the same net-favourable serious-AE direction seen in the high-background-risk SELECT population, and the mirror of the class GI tolerability tax (GI-driven discontinuation 4.5% vs 1.1%) -> GLP-1 Non-Cardiometabolic Effects and Safety. The kidney benefit was «unrelated to changes in body weight», and creatinine- and cystatin-C-based eGFR agreed — so this is not merely weight loss lowering serum creatinine. Mechanism is marked as such: «direct effects of GLP-1 receptor agonists on the kidney may include decreases in inflammation, oxidative stress, and fibrosis», with GLP-1 receptors on intrinsic kidney and immune cells (Perkovic et al., 2024).
Independence — a new endpoint, not new backing for semaglutide works
(inferred from Perkovic et al., 2024) FLOW is the same molecule (semaglutide) and the same sponsor (Novo Nordisk) as the held SELECT (Lincoff) and STEP (Wilding) trials. So it is NOT an independent line of evidence for the general claim that semaglutide works — three sponsor-run semaglutide trials are one programme, not three routes (the symmetric-standards / laundered-independence guard). What FLOW does add is a new endpoint (hard kidney outcomes) in a new population (T2D+CKD) that none of the held trials measured — so it is genuine new content, not a type-E corroboration of the CV/weight findings. The value is a filled gap (a hard kidney outcome), not extra confidence in an already-held claim.
Class-level context — FLOW inside the GLP-1 CVOT meta-analysis (F-refinement)
(Badve et al., 2025) FLOW is one trial in the class-level meta-analysis (Badve 2024; 11 trials, 85,373), which pools it with the CV-outcome programme -> GLP-1 Receptor Agonists and Cardiovascular and Kidney Outcomes. Two things that page settles bear on FLOW’s reading here (kept to the efficacy signal — no dosing, agent-selection, or combination-therapy content, per this page’s frontier scope):
- The class moves a HARD kidney outcome, and this partly cashes FLOW’s kidney-failure gap. The prior class MA (Sattar 2021, not held) reported a 21% kidney reduction «mainly driven by new-onset macroalbuminuria, a surrogate outcome not validated for clinical kidney disease outcomes». Badve re-runs it on a composite excluding macroalbuminuria — composite kidney HR 0·82 (0·73-0·93, high-certainty) — and shows kidney failure «for the first time» (HR 0·84, 0·72-0·99, moderate- certainty) at the class level (Badve et al., 2025). FLOW carried 49% of that kidney-failure weight, so this is largely FLOW being confirmed at class scale rather than an independent route — but the standalone kidney-failure signal FLOW could not confirm alone is now moderate-certainty across the class.
- The muscle-loss/creatinine confound, and its reassurance, generalize. Badve names the same worry FLOW addressed with cystatin-C: GLP-1 weight loss «may in turn reduce creatinine generation due to reduced muscle mass and therefore artificially improve creatinine-based eGFR measurements», and the kidney-failure reduction (not creatinine-confoundable) «provides important reassurance that the kidney benefits of GLP-1 receptor agonists are real». (inferred from Badve et al., 2025)
Absolute class kidney benefit is modest outside high kidney risk (class NNT 164 vs FLOW’s 20), because most pooled trials enrolled near-normal kidney function — so FLOW’s large absolute benefit is a high-baseline-risk feature (route (a)), not the class norm. (inferred from Badve et al., 2025; Perkovic et al., 2024)
Decision relevance
- T2D + CKD (eGFR 25-75, albuminuric), on RAS inhibition: semaglutide is now a strong incremental option for kidney and survival protection — NNT 20 (kidney composite) / 39 (all-cause death) over 3 years, on top of RAS inhibition, with a net-favourable serious-AE profile. It joins RAS inhibitors, SGLT2 inhibitors and finerenone as guideline-relevant cardiorenal therapy; the trialists frame it as a candidate for combination, though FLOW could not test combination (only 15.6% were on SGLT2i).
- This does NOT transport to the obese primary-prevention stratum. The population is diabetic, albuminuric, low-eGFR, and old — a high absolute baseline risk. Absolute benefit scales with that baseline (Baseline Risk and the Relative-Absolute Split); a lower-risk person gets a proportionally smaller absolute return even if the relative effect held, and the relative effect itself is untested outside CKD.
- In the Layer-1 ranking (Layer 1 - Ranking Interventions for a Stratum) for a T2D+CKD stratum, this is a large, certain lever on patient-important outcomes — near the top for that specific profile.
Gaps
- Combination therapy untested — SGLT2i/finerenone uptake was low; the order and additivity of the four cardiorenal therapies is unresolved. G.
- Not powered for kidney failure alone (FLOW) — the standalone dialysis/transplant benefit is suggested by FLOW, not confirmed within it; now partly cashed at class level (Badve 2024: kidney failure HR 0·84, 0·72-0·99, moderate-certainty), though FLOW carried ~half that pooled weight. G (narrowed).
- Generalizability — mostly White; effect in lower-risk, non-diabetic, or non-albuminuric CKD unknown. G.