🧬 Aging Wiki

empagliflozin

Properties1
aliasesJardiance, BI 10773, empagliflozin

10 min read

Empagliflozin

A sodium-glucose cotransporter 2 (SGLT2) inhibitor marketed as Jardiance (Boehringer Ingelheim / Eli Lilly). FDA-approved for type 2 diabetes (2014), heart failure with reduced ejection fraction / HFrEF (2021), heart failure with preserved ejection fraction / HFpEF (2021), and chronic kidney disease / CKD (2023). The first glucose-lowering drug to demonstrate cardiovascular mortality benefit in a prospective outcome trial. Mechanistically overlaps with caloric-restriction mimicry (caloric loss, ketogenesis, AMPK activation) — making it a compound of interest for aging biology despite the absence of aging-specific trial endpoints to date.

Identity

  • PubChem CID: 11949646
  • ChEMBL: CHEMBL2110563
  • DrugBank: DB09038
  • InChIKey: OBWASQILIWPZMG-QZMOQZSNSA-N
  • Molecular formula: C23H27ClO7
  • Molecular weight: 450.9 g/mol
  • Class: SGLT2 inhibitor (gliflozin); small-molecule glucoside derivative
  • Approved dose: 10 mg or 25 mg oral once daily (10 mg for HF/CKD; 25 mg add-on for T2D glycemic control)

Mechanism of action

Primary: renal glucose excretion

Empagliflozin selectively and reversibly inhibits SGLT2, the sodium-glucose cotransporter responsible for ~90% of renal glucose reabsorption in the proximal tubule S1/S2 segment. Inhibition causes obligate glucosuria (~70–80 g glucose/day at 25 mg dose in T2D patients), producing:

  1. Caloric loss (~280–320 kcal/day) — functionally analogous to mild caloric restriction; sustained negative energy balance without dietary change.
  2. Osmotic diuresis and natriuresis — reduces plasma volume and cardiac preload/afterload; lowers blood pressure by ~3–4 mmHg systolic without reflex tachycardia.
  3. Reduction of plasma glucose — modest HbA1c reduction (~0.5–0.8%); not the primary driver of CV benefit (insulin-matched comparators had no CV benefit).

Secondary: ketogenesis and metabolic reprogramming

Glucosuria drives a compensatory shift to fat oxidation and ketogenesis — circulating beta-hydroxybutyrate (BHB) rises 2–3x during SGLT2 inhibition even in non-diabetic patients. BHB is a signaling molecule with multiple downstream effects:

  • HDAC inhibitor — BHB inhibits class I and IIa HDACs, upregulating antioxidant genes (FOXO3A, MT2) and modulating histone acetylation patterns overlapping with those seen in caloric-restriction and fasting [^unsourced-bhb-hdac]. unsourced — mechanism inferred from BHB HDAC literature; needs primary citation on empagliflozin specifically.
  • NLRP3 inflammasome inhibition — BHB suppresses nlrp3-inflammasome activation independently of HDAC effects, reducing IL-1beta and IL-18 secretion. needs-replication in the empagliflozin context.
  • Cardiac fuel switch — the heart preferentially oxidizes BHB over fatty acids and glucose during SGLT2 inhibition, improving myocardial oxygen efficiency (~28% more ATP per O2 consumed) [^unsourced-cardiac-fuel]. unsourced — mechanistic inference; needs direct myocardial substrate flux measurement in humans.

Tertiary: AMPK activation and autophagy

Glucose deprivation and caloric loss activate ampk via rising AMP:ATP ratio. AMPK phosphorylation suppresses mtor (TORC1), inducing autophagy and mitochondrial-biogenesis. Empagliflozin activates the AMPK/SIRT-1/PGC-1alpha axis in cardiac tissue in preclinical models 1. This mechanistic overlap with metformin and caloric-restriction raises the hypothesis that SGLT2 inhibitors are partial caloric-restriction mimetics.

DimensionStatus
AMPK pathway conserved in humans?yes
Cardiac AMPK activation demonstrated in humans?partial — inferred from metabolic outcomes; direct AMPK phosphorylation not measured in human myocardium
Replicated in humans (AMPK mechanism)?in-progress

Uric acid reduction

Empagliflozin reduces serum urate by ~10–15% via competitive inhibition of URAT1 (uric acid transporter) in the proximal tubule — a mechanism independent of glucose excretion. Lower urate reduces oxidative stress and endothelial dysfunction, plausibly contributing to CV benefit. no-mechanism — relative contribution of urate reduction to overall CV protection is not established.

Clinical trial evidence

EMPA-REG OUTCOME (Zinman 2015) — T2D + established CVD

n=7,020 adults with T2D and established cardiovascular disease randomized to empagliflozin 10/25 mg vs placebo on top of standard care; median follow-up 3.1 years 2.

  • Primary endpoint (3P-MACE): HR 0.86 (95% CI 0.74–0.99, p=0.04 superiority) — 14% relative risk reduction.
  • CV mortality: HR 0.62 (95% CI 0.49–0.77, p<0.001) — 38% RRR. First positive CV mortality result for any glucose-lowering drug.
  • HF hospitalization: HR 0.65 (95% CI 0.50–0.85, p<0.001) — 35% RRR.
  • All-cause mortality: HR 0.68 (95% CI 0.57–0.82, p<0.001).
  • Glycemic effect (HbA1c −0.5%) was too small to explain the CV benefit; diuresis/volume unloading hypothesis and metabolic reprogramming proposed as primary drivers.

EMPEROR-Preserved (Anker 2021) — HFpEF

n=5,988 adults with HFpEF (LVEF >40%, NYHA class II–IV) randomized to empagliflozin 10 mg vs placebo; median follow-up 26.2 months 3.

  • Primary endpoint (CV death + HF hospitalization): HR 0.79 (95% CI 0.69–0.90, p<0.001) — 21% RRR.
  • HF hospitalization: HR 0.73 (95% CI 0.61–0.88).
  • First positive trial for any agent in HFpEF — a condition affecting the majority of older patients with heart failure, with no previously approved therapy.
  • eGFR decline slowed by ~1.4 mL/min/1.73m2/year vs placebo.

EMPEROR-Reduced (Packer 2020) — HFrEF

n=3,730 adults with HFrEF (LVEF ≤40%, NYHA class II–IV) randomized to empagliflozin 10 mg vs placebo; median follow-up 16 months 4.

  • Primary endpoint (CV death + HF hospitalization): HR 0.75 (95% CI 0.65–0.86, p<0.001) — 25% RRR.
  • Total HF hospitalizations: RR 0.70 (0.58–0.85) — also significant for recurrent events.
  • eGFR decline slowed significantly.
  • Established empagliflozin as a pillar of HFrEF guideline-directed therapy.

EMPA-KIDNEY (Herrington 2023) — CKD

n=6,609 adults with CKD (eGFR 20–45 or eGFR 45–90 with urinary ACR ≥200) randomized to empagliflozin 10 mg vs placebo; median follow-up 2.0 years; trial stopped early for efficacy 5.

  • Primary endpoint (kidney disease progression or CV death): HR 0.72 (95% CI 0.64–0.82, p<0.001) — 28% RRR.
  • Benefit extended to patients with eGFR as low as 20 mL/min/1.73m2 and to non-diabetic CKD.
  • Expanded label to CKD regardless of T2D status.

Aging biology relevance

Empagliflozin’s mechanistic profile intersects with multiple aging hallmarks and longevity pathways:

MechanismHallmark overlapEvidence grade
Caloric loss (~280 kcal/day) → CR-like signalingderegulated-nutrient-sensingStrong (human, indirect)
BHB ketogenesis → HDAC inhibition, FOXO3A activationepigenetic-alterations, deregulated-nutrient-sensingPreclinical; needs-human-replication
AMPK activation → mTOR suppression → autophagydisabled-macroautophagy, deregulated-nutrient-sensingPreclinical 1; needs-human-replication
NLRP3 inflammasome suppression via BHBchronic-inflammationPreclinical; needs-human-replication
Mitochondrial biogenesis (AMPK/SIRT-1/PGC-1alpha)mitochondrial-dysfunctionPreclinical 1
Uric acid reductionchronic-inflammationHuman (mechanistic)
Cardiac unloading + volume reductioncardiovascular-agingHuman (mechanistic, from trial hemodynamics)

Translation gap assessment: Unlike most aging-biology compounds, empagliflozin has abundant human evidence — but all four positive trials enrolled disease populations (T2D, HF, CKD). No RCT has tested empagliflozin in non-diabetic older adults with aging endpoints (grip strength, gait speed, biological age, functional independence, or all-cause mortality in the general population). The CR-mimetic and AMPK-activating mechanisms plausibly operate in non-diabetic older adults, but this is unverified. needs-human-replication for aging-specific endpoints.

SGLT2i in older adults with heart failure — meta-analysis (Khalid 2026)

R34 update (2026-05-08). Khalid et al. (Arch Gerontol Geriatr 2026) pooled 10 studies (4 RCTs + 6 cohorts; n=20,844) comparing SGLT2i vs control in adults ≥65 with HF 6:

  • All-cause mortality: HR 0.81 (95% CI 0.72–0.90, p<0.001)
  • CV death: HR 0.83 (95% CI 0.74–0.94, p=0.004)
  • First HF hospitalization: HR 0.73 (95% CI 0.66–0.80, p<0.001)
  • Composite CV death + HF hospitalization: HR 0.78 (95% CI 0.70–0.87, p<0.001)
  • Rehospitalization: HR 0.60 (95% CI 0.53–0.69, p<0.001)
  • Renal function decline slowed by 1.86 mL/min/1.73m²/yr (95% CI 1.15–2.58, p<0.001)
  • Genital infection RR 3.07 (95% CI 2.03–4.64); UTI RR 1.19 (95% CI 1.03–1.38)
  • Serious AE rate lower with SGLT2i (RR 0.92, 95% CI 0.89–0.95)

Class-level (not empagliflozin-specific) but pools the four major empagliflozin RCTs along with dapagliflozin trials. First well-powered meta-analysis directly stratifying for age ≥65 — partly addresses the “elderly-specific data” gap noted in the safety profile section. Mortality benefit and HF outcome benefits preserved in the elderly subset. Genital/UT infection signals consistent with single-trial data. The age-disparity-in-prescribing analysis from Yang 2026 (Diabetes Obes Metab, doi:10.1111/dom.70421) flags an under-prescription gap: only 13.7% of SGLT2i-eligible adults ≥80 receive the class despite the favorable evidence base.

Pharmacokinetics

  • Bioavailability: ~84% oral (high vs typical small molecules; no first-pass effect)
  • Half-life: ~12.4 h (supports once-daily dosing)
  • Protein binding: ~86.2%
  • Metabolism: primarily hepatic UGT1A3/1A8/1A9/2B7-mediated glucuronidation (not CYP-dependent); no major CYP3A4 interactions
  • Renal excretion: ~54% (urine); ~41% fecal
  • Dose adjustment: not required for mild-moderate renal impairment; reduced glucose-lowering efficacy at eGFR <45 but CV/renal protection persists per EMPA-KIDNEY

Safety profile

Well-characterized across >25,000 trial participants across four major RCTs.

Common adverse effects:

  • Genital mycotic infections (female ~6x, male ~4x increased risk) — mechanistic consequence of glucosuria; manageable, rarely serious
  • Urinary tract infections — modest increase
  • Polyuria/pollakiuria — osmotic diuresis effect
  • Volume depletion / orthostatic hypotension — most relevant in elderly on diuretics; requires monitoring

Serious but rare:

  • Diabetic ketoacidosis (DKA): Risk in T2D, especially type 1 DM (not approved for T1D); risk negligible in non-diabetics. Euglycemic DKA possible even with normal blood glucose. Relative risk ~3x in T2D trials but absolute rate low (<0.1%).
  • Fournier’s gangrene: Rare necrotizing genital infection; FDA black-box warning. Absolute incidence ~1 per 100,000 patient-years.
  • Limb amputation: Class warning from canagliflozin data; not confirmed in empagliflozin trials.

Elderly-specific: The main trials included adults with median age ~60–67 years. Orthostatic hypotension risk is higher in the very elderly (>80). Volume depletion monitoring is warranted.

Comparators within drug class

DrugApproved indicationsNotes
metforminT2DDifferent mechanism (AMPK via mitochondrial complex I); older evidence base; no HF/CKD approval
DapagliflozinT2D, HFrEF, HFpEF, CKDClass peer; similar outcomes across DAPA-HF, DELIVER trials
CanagliflozinT2D, HFrEF, CKDCREDENCE (renal), CANVAS (limb amputation signal)

Empagliflozin and dapagliflozin have near-parallel evidence profiles; class-effect vs drug-specific effects remain debated. Most cardiologists now treat as a class.

Limitations and gaps

  1. No aging-endpoint RCT. All positive trials enrolled cardiovascular or renal disease populations. Whether empagliflozin extends healthspan or delays functional decline in healthy older adults is unknown. needs-human-replication
  2. CR-mimetic claim is mechanistic inference. The caloric loss is real, but SGLT2 inhibitors don’t reproduce the full CR transcriptional signature. The comparison to caloric-restriction requires direct head-to-head metabolomic/transcriptomic validation.
  3. AMPK activation in humans not directly measured. Preclinical AMPK data 1 derive from disease models (doxorubicin cardiotoxicity, diabetes); whether equivalent activation occurs in healthy aging myocardium is unknown. needs-human-replication
  4. Biological-age biomarkers not measured in any large empagliflozin trial (no DNAm clock, no proteomic clock). A mechanistic aging-biology substudy in EMPEROR-Preserved or EMPA-KIDNEY would be tractable and high-yield. needs-replication
  5. BHB HDAC/epigenetic effects uncited for empagliflozin specifically. The signaling pathway (BHB → HDAC inhibition → FOXO3A) is established in fasting/ketosis literature; applying it to empagliflozin-induced BHB elevation is plausible but needs direct measurement. unsourced
  6. Long-term safety in non-diabetic elderly. All safety data come from T2D/HF/CKD populations. Genital mycotic infection risk and DKA risk in non-diabetic older adults are inferred, not directly measured. long-term-unknown

Cross-references

Footnotes

Footnotes

  1. doi:10.1093/toxres/tfad007 · Ahmed et al. 2023 · in-vitro/in-vivo · model: doxorubicin cardiotoxicity (rat); documents AMPK/SIRT-1/PGC-1alpha activation by empagliflozin ↩ ↩2 ↩3 ↩4

  2. zinman-2015-empa-reg-outcome · n=7,020 · rct · HR 0.86 (MACE), HR 0.62 (CV death) · model: T2D + established CVD, 3.1 yr follow-up ↩

  3. anker-2021-emperor-preserved · n=5,988 · rct · HR 0.79 (primary endpoint) · model: HFpEF LVEF >40%; PDF at ↩

  4. packer-2020-emperor-reduced · n=3,730 · rct · HR 0.75 (primary endpoint) · model: HFrEF LVEF ≤40% ↩

  5. herrington-2023-empa-kidney · n=6,609 · rct · HR 0.72 (primary endpoint) · model: CKD eGFR 20–90; trial stopped early for efficacy ↩

  6. doi:10.1016/j.archger.2026.106138 · pmid:41544527 · meta-analysis · 10 studies (4 RCTs + 6 cohorts; n=20,844) · Khalid A, Balach R, Rasool A, et al. · Arch Gerontol Geriatr 2026;143:106138 · SGLT2i vs control in HF patients ≥65 yr · all-cause mortality HR 0.81 (95% CI 0.72–0.90); CV death HR 0.83; HF hospitalization HR 0.73; rehospitalization HR 0.60; renal slope improved 1.86 mL/min/1.73m²/yr · class-level analysis (includes empagliflozin + dapagliflozin trials) ↩

Graph View

Backlinks