Sodium-glucose cotransporter 2 (SGLT2) inhibitors are widely recognized within geroscience and preventive cardiology for robust cardiometabolic risk mitigation. However, clinical translation has long relied on indirect comparisons and meta-analyses derived from trials on high-risk diabetic cohorts.
A target trial emulation published in the Journal of the American Heart Association directly contrasts the three primary clinical agents—empagliflozin, dapagliflozin, and canagliflozin—in a real-world cohort of 137,232 adults with type 2 diabetes and moderate cardiovascular risk. The study offers rigorous causal-inference data on whether these molecules differ in hard survival endpoints or if longevity practitioners should treat them as clinically interchangeable.
The primary study text can be referenced via its digital object identifier at https://doi.org/10.1161/JAHA.125.046238.
Comparative Mortality and Hard Endpoint Trajectories
The analysis tracked weighted cohorts initiating canagliflozin (n = 42,877), dapagliflozin (n = 17,871), or empagliflozin (n = 76,485), with median follow-up periods exceeding 1,000 days across all arms.
All-Cause Mortality Contrasts (Intention-to-Treat)
-
Empagliflozin vs Canagliflozin: Hazard Ratio (HR) 0.86 (95% CI, 0.80–0.94; Wald P = 0.001), reflecting a 14% relative reduction in overall death risk.
-
Empagliflozin vs Dapagliflozin: HR 0.95 (95% CI, 0.86–1.05), showing no statistically significant mortality divergence.
-
Dapagliflozin vs Canagliflozin: HR 0.91 (95% CI, 0.83–1.01), failing to reach significance.
Three-Year Cumulative Mortality Incidence
-
Year 1: Canagliflozin 0.7%, Dapagliflozin 0.7%, Empagliflozin 0.6%.
-
Year 2: Canagliflozin 1.9%, Dapagliflozin 1.9%, Empagliflozin 1.6%.
-
Year 3: Canagliflozin 3.3%, Dapagliflozin 3.4%, Empagliflozin 2.8%.
Composite Major Adverse Cardiovascular Events (MACE)
-
3-Point MACE (Myocardial Infarction, Ischemic Stroke, All-Cause Mortality): Empagliflozin demonstrated lower hazard versus canagliflozin (HR 0.92; 95% CI, 0.87–0.97; P = 0.007). No significant difference was detected between empagliflozin and dapagliflozin (HR 0.94; 95% CI, 0.87–1.01) or dapagliflozin and canagliflozin (HR 0.98; 95% CI, 0.91–1.05).
-
Cumulative 3-Year MACE Rate: 6.2% for empagliflozin versus 6.8% for canagliflozin and 7.2% for dapagliflozin.
Non-Proportional Hazards and Dynamic Survival Curves
The proportional hazards assumption was formally violated across cardiovascular outcomes, revealing significant time-dependent dynamics. Rather than a constant treatment effect, the relative risk profile inverted over prolonged follow-up:
-
Early Phase (≤1 Year): Empagliflozin exhibited an early elevation in MACE risk relative to canagliflozin (HR 1.17; 95% CI, 1.03–1.32) and dapagliflozin (HR 1.18; 95% CI, 1.02–1.36), driven largely by an increased early signal for ischemic stroke (empagliflozin vs canagliflozin HR 1.36; 95% CI, 1.07–1.74) and arterial revascularization (HR 1.14; 95% CI, 1.07–1.22).
-
Intermediate Phase (1 to 2 Years): Risk profiles neutralized, with empagliflozin vs canagliflozin MACE hazard dropping to HR 0.93 (95% CI, 0.82–1.05).
-
Extended Phase (>2 Years): A divergence emerged favoring empagliflozin, with MACE reduced against canagliflozin (HR 0.87; 95% CI, 0.81–0.94) and dapagliflozin (HR 0.90; 95% CI, 0.82–1.00), and all-cause mortality reduced against canagliflozin (HR 0.84; 95% CI, 0.76–0.92).
This late divergence in cumulative incidence curves mirrors legacy effects observed in trials like EMPA-KIDNEY, where cardiovascular separation manifests primarily over extended timelines.
Pathological Phenotypes and Safety Profile
The study tracked macrovascular events and metabolic crises to assess comparative toxicity:
-
Heart Failure Hospitalization: Incidence remained consistently low across cohorts (0.9% to 1.0% at 3 years), with no between-drug variations (Wald P = 0.73).
-
Atherosclerotic Events: Rates of myocardial infarction (1.5% across all cohorts at 3 years; Wald P = 0.88) and ischemic stroke (1.1% to 1.3% at 3 years; Wald P = 0.34) showed no drug-specific variation over the total follow-up.
-
Arterial Revascularization Paradox: Empagliflozin initiation was associated with an elevated hazard of arterial revascularization procedures compared to canagliflozin (overall HR 1.06; 95% CI, 1.01–1.12; Wald P = 0.03). This excess risk remained prominent in as-treated sensitivity models.
-
Genitourinary Infections: Severe infections accrued steadily, reaching 9.0% for canagliflozin and 9.2% for both dapagliflozin and empagliflozin at year 3.
-
Acute Metabolic Crises: Severe hypoglycemic crises (0.4% at 3 years) and hyperglycemic crises/euglycemic ketoacidosis (0.4% at 3 years) remained rare and showed no pairwise divergence.
Translation Barriers and Methodological Limitations
Extrapolating these findings to healthy lifespan extension protocols requires addressing substantial methodological caveats:
-
As-Treated Discrepancies and Adherence Censoring: While the intention-to-treat analysis indicated an all-cause mortality benefit for empagliflozin, the as-treated sensitivity analysis—censoring patients at therapy discontinuation—attenuated the MACE difference to non-significance (HR 1.08; 95% CI, 1.01–1.16 for revascularization, but no MACE reduction). High real-world discontinuation rates (60% to 70%) introduce informative censoring concerns.
-
Black-Box Mortality Attributions: Claims databases cannot reliably link to detailed death certificates, precluding cause-specific mortality classification (e.g., differentiating sudden cardiac death from non-cardiovascular deaths like cancer or infection).
-
Absence of Intermediate Biomarkers: The database lacked granular biometric adjustments for serial HbA1c, ambulatory blood pressure, renal filtration slope (eGFR), or baseline lipid subfractions.
-
Moderate Risk vs. Normoglycemic Translation: The study cohort comprised individuals with existing metabolic disease (mean age 65.7 years, 82% concurrent metformin use, 72% on renin-angiotensin system inhibitors). Translating an 8% to 14% relative hazard reduction from an older, diabetic population to healthy adults targeting longevity is unvalidated, where baseline mortality hazards are substantially lower and the absolute risk-reduction denominator approaches zero.
Evidence Chain
-
Statement: In an emulated target trial of 137,232 adults with type 2 diabetes at moderate cardiovascular risk, empagliflozin was associated with an 8% lower risk of 3-point MACE and a 14% lower risk of all-cause mortality compared to canagliflozin, but exhibited no statistically significant difference compared to dapagliflozin.
Source: https://doi.org/10.1161/JAHA.125.046238 -
Statement: A network meta-analysis of 14 randomized controlled trials identified an all-cause mortality reduction for empagliflozin compared with dapagliflozin among diabetic cohorts, but found no divergence between canagliflozin and dapagliflozin.
Source: https://doi.org/10.1186/s12933-023-02035-8 -
Statement: In a large US population-based cohort, pairwise comparisons of canagliflozin versus empagliflozin and dapagliflozin versus empagliflozin demonstrated comparable risks for myocardial infarction and stroke, with all-cause mortality analysis restricted by data availability.
Source: https://doi.org/10.1001/jamainternmed.2024.7357 -
Statement: A Korean nationwide observational cohort of type 2 diabetes patients demonstrated lower risks of heart failure hospitalization and cardiovascular death with dapagliflozin compared to empagliflozin.
Source: https://doi.org/10.1186/s12933-023-01911-7 -
Statement: An emulated target trial assessing second-line glucose-lowering therapies established that SGLT2 inhibitors confer superior cardiovascular and renal protection compared to sulfonylureas and DPP-4 inhibitors in moderate-risk diabetes.
Source: https://doi.org/10.1038/s44161-024-00453-9 -
Statement: Long-term post-trial follow-up from the EMPA-KIDNEY trial indicates that cardiovascular and renal event reductions persist after active therapy cessation, illustrating enduring physiological legacy effects.
Source: https://doi.org/10.1056/NEJMoa2409183