The FLOW Trial (Perkovic 2024, NEJM) Proved That Semaglutide Protects the Kidneys in Type 2 Diabetes — Reducing the Composite of Kidney Disease Progression or Cardiovascular Death by 24%, Slowing eGFR Decline From −2.2 to −1.8 mL/min/Year, and Cutting ESKD Events by 47% — Establishing GLP-1 Receptor Agonists as Organ-Protective Drugs That Slow Diabetic Nephropathy Through Mechanisms Beyond Glucose Control

the definitive kidney outcomes data: Perkovic V et al. (2024, NEJM): FLOW (Evaluate Renal Function with Semaglutide Once Weekly); DESIGN: double-blind, placebo-controlled RCT; N=3,533 adults with T2DM + CKD (eGFR 50–75 mL/min/1.73m² AND UACR 100–10,000 mg/g, i.e., moderate-severe albuminuria); INTERVENTION: semaglutide 1mg SC weekly vs placebo (in addition to standard of care including RAAS blockade [ACEi/ARB], and allowed SGLT2 inhibitor use); follow-up: median 3.4 years; PRIMARY COMPOSITE ENDPOINT: sustained ≥50% decline in eGFR OR ESKD (dialysis, transplant, or eGFR <15) OR kidney-specific death OR cardiovascular death; RESULT: 331 events in semaglutide arm vs 410 in placebo (HR 0.76; 95% CI 0.66–0.88; p=0.0003); a 24% reduction; SECONDARY ENDPOINTS: eGFR slope (chronic): −1.8 vs −2.2 mL/min/1.73m²/year (acute initial drop at start of semaglutide is expected and hemodynamic — analogous to the initial SGLT2 dip — not a sign of renal harm); ESKD events: −47% (HR 0.53); major cardiovascular events (MACE): −18% (HR 0.82); cardiovascular death: −20%; all-cause death: −20%; UACR: −25% from baseline (significant reduction in albuminuria); STOPPING: the trial was stopped early (originally planned for 5 years) due to overwhelming efficacy at the pre-specified interim analysis; SIGNIFICANCE: FLOW is the first GLP-1 trial specifically powered for kidney outcomes; it establishes semaglutide as a renoprotective drug independent of (and in addition to) its glucose-lowering and weight-loss effects; the HR of 0.76 is comparable to SGLT2 inhibitor kidney trials (CREDENCE: 0.66; DAPA-CKD: 0.61; EMPA-KIDNEY: 0.72)

how GLP-1 agonists protect the kidney — multiple distinct mechanisms: MECHANISM 1 — REDUCED INTRAGLOMERULAR PRESSURE: diabetic kidneys hyperfilter due to efferent arteriole dilation and afferent arteriole constriction → elevated intraglomerular pressure → mechanical damage to the glomerular filtration barrier → podocyte loss → proteinuria; GLP-1R is expressed on afferent arterioles and mesangial cells; GLP-1R activation → vasodilation of the afferent arteriole → reduced intraglomerular pressure → reduced glomerular capillary shear stress; this is analogous to (and additive with) the efferent arteriole constriction mechanism of SGLT2 inhibitors (SGLT2i contract efferent arterioles via tubuloglomerular feedback); MECHANISM 2 — ANTI-INFLAMMATORY EFFECTS ON TUBULAR EPITHELIUM: GLP-1R is expressed on proximal tubular epithelial cells; GLP-1R activation in the tubule: reduces NF-κB activity → reduced IL-6, TNF-α, MCP-1 secretion; reduces mitochondrial oxidative stress in tubular cells; reduces TGF-β1 (the primary driver of tubulointerstitial fibrosis) → less fibrosis progression; MECHANISM 3 — WEIGHT LOSS REDUCING HYPERFILTRATION: obesity drives glomerular hyperfiltration via increased renal plasma flow, sodium retention, and RAAS activation; weight loss of −4 to −12% from GLP-1 therapy → reduced renal plasma flow → reduced hyperfiltration; MECHANISM 4 — GLUCOSE LOWERING: reduced HbA1c → less glycation of tubular proteins → less tubular cell apoptosis; reduced glucose toxicity in proximal tubular cells (AGE formation, PKC activation); MECHANISM 5 — ALBUMINURIA REDUCTION: the −25% UACR reduction seen in FLOW reflects improved glomerular filtration barrier integrity (podocyte preservation + reduced intraglomerular pressure) → less albumin leak; albuminuria is itself a driver of tubular inflammation (filtered albumin is inflammatory to tubular cells) — reducing it provides additional protection

head-to-head comparison and combination: SGLT2 INHIBITORS IN CKD (established): CREDENCE (canagliflozin, N=4,401, T2DM+DKD): primary composite −30% (HR 0.70); DAPA-CKD (dapagliflozin, N=4,304, CKD with or without T2DM): primary composite −39% (HR 0.61); EMPA-KIDNEY (empagliflozin, N=6,609, CKD): primary composite −28% (HR 0.72); SGLT2 inhibitors are currently first-line for DKD protection due to earlier and more definitive data; they may be slightly more effective at preventing progression than semaglutide (CREDENCE HR 0.70 vs FLOW HR 0.76), though direct head-to-head has not been done; SGLT2 ADVANTAGES: oral (vs injection); daily dosing; proven benefit extends to non-diabetic CKD (DAPA-CKD included 33% non-diabetic CKD); robust data at lower eGFR (can be used down to eGFR ~20); SGLT2 LIMITATIONS: genital mycotic infections; rare DKA in T2DM; not beneficial for primary cardiovascular outcomes in eGFR <30; GLP-1 ADVANTAGES IN CKD: significant weight loss (adds metabolic benefit beyond renal protection); proven cardiovascular benefit (SELECT: −20% MACE even without diabetes); no UTI/genital infection risk; safe at very low eGFR (semaglutide not renally cleared — no dose adjustment needed to eGFR ≥15); GLP-1 LIMITATIONS IN CKD: injection required (Ozempic/semaglutide 1mg SC weekly); nausea may be more challenging in CKD patients (uremia-related nausea adds to GLP-1 GI effects); COMBINATION GLP-1 + SGLT2i: FLOW allowed SGLT2 inhibitor co-use; approximately 15% of FLOW patients were on SGLT2 inhibitors; subgroup analysis showed benefit was consistent regardless of SGLT2i use; the combination is mechanistically additive (different primary mechanisms — afferent dilation vs efferent constriction + tubuloglomerular feedback); current ADA/KDIGO guidelines recommend both when tolerated

post-FLOW clinical positioning: KDIGO (Kidney Disease: Improving Global Outcomes) 2022 guidelines on diabetic kidney disease (pre-FLOW, updated principles): first-line: ACEi or ARB (RAAS blockade) + SGLT2 inhibitor if eGFR ≥20; second-line: GLP-1 receptor agonist with proven CV benefit (at the time: liraglutide, semaglutide, dulaglutide); POST-FLOW UPDATE (anticipated 2024–2025 KDIGO revision): semaglutide (and GLP-1 agonists with CV benefit) are expected to move from second-line to co-first-line with SGLT2 inhibitors for T2DM+CKD with established CVD; for patients who are not on SGLT2 inhibitor (due to low eGFR, poor tolerance, or contraindication), GLP-1 as the primary add-on to ACEi/ARB is appropriate; ADA STANDARDS OF CARE 2024: in T2DM+CKD: (1) RAAS blockade (ACEi/ARB); (2) SGLT2 inhibitor if eGFR ≥20; (3) finerenone (non-steroidal MRA — FIDELIO-DKD/FIGARO-DKD) if not on SGLT2i or additional benefit needed; (4) GLP-1 receptor agonist with proven CV benefit; the "four pillars" framework — all four agents have complementary mechanisms and additive benefit in T2DM+CKD; GLP-1 DOSING IN CKD: semaglutide is not renally cleared (hepatic elimination); no dose adjustment for any level of CKD or ESKD; safe to use through dialysis (few data, but pharmacokinetically sound); acute eGFR dip at initiation (similar to SGLT2): hemodynamic, not nephrotoxic — reassure patient; monitor eGFR and electrolytes at 4 weeks after initiation in patients with advanced CKD