For the roughly one in three adults with type 2 diabetes who also carries chronic kidney disease, the question of whether a GLP-1 receptor agonist can protect the kidneys regardless of underlying heart status has major clinical weight. A definitive answer reshapes how nephrologists and cardiologists stratify treatment priorities for one of medicine's most complex comorbidity clusters.
The FLOW trial enrolled 3,533 participants with type 2 diabetes and chronic kidney disease, randomizing them to once-weekly subcutaneous semaglutide 1.0 mg or placebo. This pre-specified subgroup analysis stratified outcomes by three cardiovascular phenotypes: established atherosclerotic cardiovascular disease (ASCVD, present in 33.9%), heart failure (19.2%), and high predicted cardiovascular risk without established disease — defined by a 10-year PREVENT score ≥20%, present in 66.5% of that subset. The composite primary endpoint captured serious renal deterioration events including a ≥50% eGFR decline, eGFR falling below 15 mL/min/1.73 m², dialysis initiation, transplantation, and kidney or cardiovascular death. Hazard ratios consistently favored semaglutide across all subgroups — approximately 0.80 in the ASCVD group and 0.74 in those without, with a non-significant interaction p-value of 0.62 — indicating the benefit was statistically uniform. Heart failure patients showed a particularly pronounced hazard ratio near 0.67.
This analysis is consequential for several reasons. The GLP-1 class has already demonstrated cardiovascular mortality benefits in SUSTAIN-6 and SELECT trials, but kidney-specific protection as a primary endpoint was less established until FLOW's main results. Now, the subgroup consistency strengthens the biological plausibility of a mechanism — likely a combination of hemodynamic glomerular pressure reduction, anti-inflammatory signaling, and metabolic offloading — that operates independently of baseline cardiovascular architecture. The heart failure signal warrants particular attention, as this population has historically been underserved by glucose-lowering agents. Key limitations include that subgroup analyses, even pre-specified ones, carry elevated Type I error risk, and the trial's single dose and subcutaneous delivery limit generalizability to oral formulations. Overall, this is confirmatory and clinically reinforcing rather than paradigm-shifting, but it meaningfully expands the evidence base for semaglutide in high-risk cardiorenal patients.