🧬 Aging Wiki

semaglutide

Properties1
aliasesOzempic, Wegovy, Rybelsus, semaglutide, GLP-1 RA

14 min read

Semaglutide

A synthetic GLP-1 receptor agonist approved for type 2 diabetes (Ozempic, 2017) and chronic weight management (Wegovy, 2021). The largest cardiovascular outcomes trial in obesity to date (SELECT, n=17,604) showed a 20% relative reduction in major adverse cardiovascular events (MACE) over ~3 years in overweight/obese adults without diabetes 1. The compound has attracted geroprotective interest because obesity-driven metabolic dysfunction and chronic inflammation overlap substantially with aging hallmarks — but no trial has yet tested semaglutide against a primary aging or biological-age endpoint. needs-human-replication

Identity

FieldValue
WHO INNsemaglutide
PubChem CID56843331
InChIKeyDLSWIYLPEUIQAV-CCUURXOWSA-N
ChEMBL IDCHEMBL2107778
Molecular formulaC187H291N45O59
Molecular weight4,114 Da (peptide)
ClassSynthetic GLP-1 analogue — 34-aa peptide
RouteSubcutaneous injection (weekly); oral tablet (daily, Rybelsus)
Half-life~160 h (~7 days) subcutaneous

Note on biologic status: Semaglutide is a peptide biologic, not a small molecule. PubChem CID 56843331 tracks the free-acid conjugate form; no InChI-compatible SMILES is available at typical lengths due to peptide size. ChEMBL CHEMBL2107778 records the bioactivity data.

Structural basis for long half-life

Native GLP-1(7–36) amide has a plasma half-life of 1–2 minutes due to DPP-IV cleavage and renal clearance. Semaglutide achieves ~7-day dosing through three modifications:

  1. Aib8 substitution — α-aminoisobutyric acid at position 8 blocks DPP-IV cleavage.
  2. Lys34Arg substitution — removes a secondary DPP-IV site.
  3. C18 fatty diacid side chain at Lys26 (via a linker) — mediates tight reversible binding to serum albumin, dramatically reducing renal filtration and receptor-mediated clearance 2.

The combined effect extends t½ from minutes to ~160 h, enabling once-weekly subcutaneous dosing 2.

Mechanism of action

Peripheral

Pancreatic β-cells: GLP-1R activation → Gαs → adenylyl cyclase → cAMP ↑ → PKA and Epac2 → glucose-dependent insulin secretion (GDIS). Critically, the insulin response is glucose-dependent — minimal effect at fasting glucose, sharply potentiated above ~5 mM. This is the mechanistic basis for the low intrinsic hypoglycemia risk.

Pancreatic α-cells: GLP-1R on α-cells → suppressed glucagon release, reducing hepatic glucose output postprandially.

GI tract: GLP-1R on vagal afferents and enteric neurons → slowed gastric emptying → blunted postprandial glucose excursions and earlier mechanoreceptor-driven satiety.

Central

Hindbrain (area postrema / NTS): GLP-1R expressed on neurons in circumventricular organs not protected by the blood-brain barrier; direct peptide access drives nausea/satiety signaling.

Hypothalamic ARC: GLP-1R on POMC/CART neurons → increased POMC expression and α-MSH release → MC4R activation → reduced food intake. Concurrent inhibition of NPY/AgRP neurons removes orexigenic tone [^unsourced-cns-mechanism]. unsourced — direct human neuroimaging or optogenetic confirmation of POMC-specific mechanism not yet established; most evidence from rodent intracranial injection studies.

Clinical indications and trial landscape

STEP-1 trial — weight loss in obesity without diabetes

Wilding et al. randomized 1,961 adults with BMI ≥30 (or ≥27 with comorbidity) and no diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo for 68 weeks (both groups with lifestyle intervention) 3.

  • Mean body weight loss: 14.9% semaglutide vs 2.4% placebo (difference −12.4 pp, 95% CI −13.4 to −11.5, p<0.001).
  • ≥5% weight loss: 86.4% semaglutide vs 31.5% placebo.
  • ≥15% weight loss: 32.0% semaglutide vs 1.7% placebo.
  • GI adverse events (nausea/vomiting/diarrhea): more frequent with semaglutide (~80% incidence of any GI AE vs ~50% placebo); mostly mild-to-moderate and front-loaded in first 20 weeks.
DimensionStatus
Pathway conserved in humans?yes (GLP-1R is a human target)
Phenotype conserved in humans?yes
Replicated in humans?yes (STEP 1–4, SUSTAIN series)

SELECT trial — cardiovascular outcomes in obesity without diabetes

Lincoff et al. enrolled 17,604 adults (age ≥45) with BMI ≥27, established CVD, and no diabetes and randomized to semaglutide 2.4 mg weekly or placebo; median follow-up ~33.4 months 1.

  • Primary endpoint (3-point MACE: CV death, non-fatal MI, non-fatal stroke): HR 0.80 (95% CI 0.72–0.90, p<0.001) — a 20% relative risk reduction.
  • CV death HR 0.85 (95% CI 0.71–1.01); non-fatal MI HR 0.72 (95% CI 0.61–0.85); non-fatal stroke HR 0.82 (95% CI 0.61–1.10 — imprecise).
  • The mechanistic attribution is uncertain: 14.9% weight reduction likely explains part, but the MACE benefit appeared before substantial weight loss; reduced inflammation (hsCRP) has been proposed as an independent pathway.
  • hsCRP: semaglutide reduced hsCRP by ~40% from baseline in the STEP-HFpEF program 4; SELECT sub-analyses are ongoing.

The SELECT trial is the strongest current human evidence for a clinically meaningful cardiometabolic benefit of semaglutide in overweight adults without diabetes.

FLOW trial — kidney outcomes in T2D + CKD

Perkovic et al. enrolled 3,533 adults with type 2 diabetes and CKD (eGFR 24–75, UACR ≥300 mg/g); randomized to semaglutide 1.0 mg weekly or placebo 5.

  • Primary composite (≥50% eGFR decline, ESKD, kidney- or CV-death): HR 0.76 (95% CI 0.66–0.88, p<0.001) — 24% relative risk reduction.
  • Kidney-specific composite HR 0.79 (95% CI 0.66–0.94).
  • eGFR slope: semaglutide −2.19 mL/min/1.73m²/yr vs −3.36 mL/min/1.73m²/yr (placebo), a 35% slower decline.
  • This extends semaglutide’s evidence base to kidney protection; mechanism may overlap with inflammation and intraglomerular pressure reduction.

FDA approvals

IndicationBrandApproval yearDose
Type 2 diabetes (SC)Ozempic20170.5/1.0/2.0 mg weekly
T2D (oral)Rybelsus20197/14 mg daily
Chronic weight managementWegovy20212.4 mg weekly
Reduction of CV events (obesity/overweight)Wegovy20242.4 mg weekly

Geroprotective angle — aging biology hypotheses

Semaglutide does not have a dedicated aging trial. The geroprotective case rests on four converging lines:

1. Obesity as accelerated aging. Excess adiposity and metabolic syndrome amplify multiple aging hallmarks — chronic-inflammation (elevated IL-6, TNF-α, CRP), deregulated-nutrient-sensing (hyperinsulinemia, mTORC1 overactivation), and mitochondrial-dysfunction. Reversing these via 15% body weight reduction could theoretically reduce biological age. A phase 2b RCT published in Nature Communications (n=84; 32 weeks; HIV-associated lipohypertrophy population) found semaglutide 1.0 mg SC weekly produced concordant reductions across multiple epigenetic aging clocks as a secondary/exploratory endpoint: PCGrimAge −3.08 yr (p=0.007), PhenoAge −4.90 yr (p=0.004), DunedinPACE −0.09 units/yr (~9% slower pace, p=0.01), after ANCOVA adjustment for baseline BMI, hsCRP, and sCD163 6. Seven of eleven organ-system clocks reached significance (most prominently inflammation −5.01 yr, brain −4.99 yr, heart −4.34 yr). Critical caveats: (1) the population is diseased (HIV lipohypertrophy, elevated baseline inflammation); (2) epigenetic aging was not the primary endpoint and no multiple-comparisons correction was applied to the clock battery; (3) three of seven authors are affiliated with TruDiagnostic, the company whose clocks constitute the study endpoints — a material conflict of interest (see 6). This does not establish biological-age benefit in a lean, healthy, metabolically normal individual. No trial has yet tested a validated biological-age clock as a primary endpoint in a non-diseased population. needs-healthy-population-replication

2. Direct anti-inflammatory signaling. GLP-1R is expressed on macrophages and other immune cells; semaglutide appears to suppress NF-κB-driven cytokine production partly weight-independently. Reductions in hsCRP (~40%) and IL-6 have been observed in clinical trials 4; the psoriasis RCT (n=50) documented significant serum IL-6 reduction 7. However, which anti-inflammatory effects are weight-loss-mediated vs direct GLP-1R remains contested. no-mechanism

3. Cardiovascular risk reduction. The SELECT MACE benefit (HR 0.80) implies prolonged event-free survival in middle-aged to older adults with CVD — a direct healthspan effect. Whether this translates to reduced overall mortality or slower biological aging over longer time horizons is unknown.

4. CNS and neurodegeneration hypotheses. GLP-1R is expressed in dopaminergic neurons and astrocytes; preclinical models suggest neuroprotection (α-synuclein clearance, reduced neuroinflammation). Observational studies suggest reduced Parkinson’s disease incidence in GLP-1R agonist users (Liraglutide > semaglutide data available).

5. Cartilage protection in osteoarthritis — weight-loss-independent (R34 update 2026-05-08). Qin et al. (Cell Metab 2026) reported that semaglutide produces chondroprotective effects in obesity-associated osteoarthritis mouse models that are NOT explained by weight loss or appetite suppression 8. Using a precise diet-controlled experimental setting (likely pair-feeding; specific protocol pending PDF verification), the authors dissociated chondroprotection from semaglutide’s classical metabolic actions. The proposed mechanism — the GLP-1R-AMPK-PFKFB3 axis — reprograms chondrocyte metabolism from glycolysis to oxidative phosphorylation under inflammatory conditions, reducing cartilage degeneration, osteophyte formation, synovial lesion, and pain sensitivity. A randomized pilot clinical study (ChiCTR2200066291) is reported to support these findings; n, dose, and primary endpoint are not in the abstract. This is the first major demonstration of a weight-loss-independent chondroprotective mechanism for any GLP-1 RA in OA, and provides the contemporary counterpoint to the negative Gudbergsen 2021 systemic-liraglutide OA RCT (cited in 9). The paper adds mitochondrial-dysfunction (chondrocyte fuel-switch) and deregulated-nutrient-sensing (AMPK as primary nutrient-sensor) as new hallmark intersections for the indication. PDF NOT yet verified — quantitative claims (mouse n’s, dose, mg/kg, mechanism specifics including whether AMPK and PFKFB3 modulation was genetic vs pharmacologic, OXPHOS measurement modality) require PDF cross-check before downstream-claim use. needs-pdf-verification needs-human-replication.

EVOKE / EVOKE+ Phase 3 in early Alzheimer’s disease — NEGATIVE (R34 update 2026-05-08). Cummings et al. (Lancet 2026) reported the first Phase 3 RCTs of oral semaglutide 14 mg flexible-dose vs placebo in early-stage symptomatic AD (mild cognitive impairment or mild dementia, amyloid-confirmed) — n=1,855 (evoke) and n=1,953 (evoke+); 156-week treatment period; primary endpoint change in CDR-SB at week 104 10. Both trials missed the primary endpoint: mean CDR-SB change at week 104 was 2.3 (semaglutide) vs 2.3 (placebo) in evoke (estimated difference −0.08, 95% CI −0.35 to 0.20, p=0.57) and 2.2 vs 2.1 in evoke+ (difference 0.10, 95% CI −0.17 to 0.38, p=0.46). Both trials were discontinued due to negative clinical outcome. Treatment-emergent adverse events were higher with semaglutide (91.2% vs 84.8%) but consistent with the known semaglutide profile. The negative result substantially weakens the GLP-1-RA-as-AD-therapy hypothesis derived from prior observational and preclinical signals; it does NOT rule out potential benefit at earlier disease stages or in non-AD aging-cognition contexts, but it removes the strongest near-term rationale for AD-specific use. This is the first large-scale aging-rejuvenation-relevant Phase 3 endpoint failure for the class.

Effects on aging hallmarks

HallmarkProposed effectEvidence quality
chronic-inflammationReduces hsCRP ~40%, IL-6; weight-loss + possible direct GLP-1R effect on macrophagesClinical RCT (STEP-HFpEF, psoriasis)
deregulated-nutrient-sensingReduces hyperinsulinemia; indirect mTORC1 deactivation via caloric restriction effectMechanism inferred; no direct mTOR assay in humans
mitochondrial-dysfunctionAdipose tissue mitochondrial function improves with weight loss; chondrocyte glycolysis-to-OXPHOS shift via GLP-1R-AMPK-PFKFB3 in OA mice 8Mouse model + pilot RCT; PDF unverified
cellular-senescenceNo direct trial data; obesity increases senescent cell burden; weight loss hypothesized to reduce p16+ cellsNot tested unsourced
deregulated-nutrient-sensing (chondrocyte)AMPK activation in chondrocytes downregulates PFKFB3 → restores OXPHOS metabolism in OA mice 8Mouse + pilot RCT; mechanism specifics PDF-unverified

Pharmacokinetics and dosing

  • Bioavailability (SC): ~89%; absorption is slow (tmax ~1–3 days).
  • Bioavailability (oral): ~1% (Rybelsus); requires fasting + large water volume absorption-enhancement.
  • Half-life (SC): ~160 h, enabling once-weekly dosing.
  • Protein binding: >99% albumin (via C18 fatty diacid side chain).
  • Metabolism: proteolytic cleavage of peptide backbone and fatty acid side chain oxidation; no meaningful CYP450 involvement.
  • Renal clearance: minimal; dose adjustment not required for CKD.
  • Titration: dose escalation over 16–20 weeks is required (0.25 → 0.5 → 1.0 → 1.7 → 2.4 mg for Wegovy) to minimize GI AEs.

Adverse effects and safety concerns

Adverse effectFrequency / notes
Nausea~44% semaglutide vs 16% placebo (STEP-1) — most common, peaks in weeks 4–12
Vomiting~24% semaglutide vs 6% placebo (STEP-1)
Diarrhea~30% semaglutide vs 16% placebo (STEP-1)
Gallbladder diseaseIncreased risk (~2.6% vs 1.2%; rapid weight loss increases bile lithogenicity)
PancreatitisRare; causal relationship debated; contraindicated in history of pancreatitis
Medullary thyroid carcinomaBlack-box warning (rodent signal; human risk not established but contraindicated in MEN2 / family history MTC)
Muscle mass lossWeight loss includes ~25–40% lean mass loss, not all adipose — a clinically significant concern for older adults where sarcopenia risk is elevated long-term-unknown
Diabetic retinopathyWorsening observed in early SUSTAIN trials in T2D patients with pre-existing retinopathy (transient, mechanism unclear)
Injection-site reactionsMild, common

Muscle-loss caveat for aging context: In STEP-1, body composition data showed lean mass declined alongside fat mass during rapid weight loss. For older adults (>65) where baseline muscle mass and function may already be compromised, rapid semaglutide-induced weight loss may worsen or unmask sarcopenia. This is an unresolved safety signal for any aging-specific application. long-term-unknown

Translation gap

human-evidence-strong — robust RCT evidence for cardiometabolic benefit exists (SELECT MACE, FLOW kidney endpoints). The gap is aging-specific:

  • No trial has enrolled by chronological or biological age as primary criterion.
  • No trial has used a validated biological-age clock as a primary endpoint; the Corley et al. (Nature Communications 2026) HIV lipohypertrophy RCT measured several clocks as secondary/exploratory endpoints 6, but in a diseased population only, and with material COI from TruDiagnostic-affiliated co-authors.
  • No trial has enrolled healthy older adults without diabetes or established CVD — the population most relevant to primary aging prevention.
  • The muscle-loss safety concern in older adults with borderline sarcopenia has not been prospectively addressed.

Gaps and open questions

  • no-mechanism — Relative contribution of weight loss vs direct GLP-1R signaling to cardiovascular and anti-inflammatory benefits in SELECT not yet parsed.
  • long-term-unknown — Consequences of 10–15% lean mass loss in older adults on long-term functional outcomes (grip strength, gait speed, fall risk) not reported in existing trials.
  • needs-human-replication — CNS neuroprotective claims (Parkinson’s, Alzheimer’s) rest largely on preclinical data + observational signals; Phase 2 RCTs ongoing.
  • needs-healthy-population-replication — Biological-age clock response (GrimAge, DunedinPACE, PhenoAge) measured in HIV lipohypertrophy RCT as a secondary/exploratory endpoint 6, but not yet tested as a primary endpoint or in a healthy non-diseased population. Corley et al. (Nat Commun 2026) provides signal in a diseased cohort with material TruDiagnostic COI; a dedicated aging RCT in healthy older adults using an independent epigenetic platform is needed for confirmation.
  • dose-response-unclear — Optimal dose for aging-specific benefit (if any) unknown; 2.4 mg weekly chosen for maximal weight loss, not for any aging-primary endpoint.

Cross-references

Footnotes

Footnotes

  1. lincoff-2023-select-semaglutide-cvd · n=17,604 · rct · p<0.001 · model: overweight/obese adults with established CVD without diabetes (~33.4 mo) · doi:10.1056/NEJMoa2307563 ↩ ↩2

  2. doi:10.3390/medsci13040265 · review · Weiskirchen R, Lonardo A · Medical Sciences 2025 · bench-to-bedside review of semaglutide PK, molecular modifications, and clinical program · local copy: pending ↩ ↩2

  3. wilding-2021-step1-semaglutide-obesity · n=1,961 · rct · p<0.001 · model: obese/overweight non-diabetic adults (68 weeks, semaglutide 2.4 mg SC weekly vs placebo) · doi:10.1056/NEJMoa2032183 ↩

  4. doi:10.1016/j.jacc.2024.04.038 · rct (STEP-HFpEF sub-analysis) · n=529 · Kosiborod et al. · JACC 2024 · semaglutide reduced hsCRP and NT-proBNP across NYHA class in HFpEF · local copy: downloaded ↩ ↩2

  5. perkovic-2024-flow-semaglutide-ckd · n=3,533 · rct · p<0.001 · model: T2D + CKD adults (semaglutide 1.0 mg SC weekly vs placebo) · doi:10.1056/NEJMoa2403347 ↩

  6. corley-2025-semaglutide-epigenetic-aging · n=84 (45 semaglutide / 39 placebo) · rct (double-blind phase 2b, 32 wk) · model: HIV-associated lipohypertrophy adults; virologically suppressed HIV on ART · doi:10.1038/s41467-026-72861-3 · published Nature Communications 2026-05-19 (preprint doi:10.1101/2025.07.09.25331038 originally deposited on medRxiv 2025) · COI: 3/7 authors affiliated with TruDiagnostic (Corley MJ — scientific advisor; Dwaraka V, Smith R — employees); TruDiagnostic performed all clock analyses — material conflict as clock vendor co-authored study showing their own clocks respond to drug · epigenetic clocks were secondary/exploratory endpoints (parent trial primary = CT/DEXA adipose tissue); no multiple-comparisons correction across clock battery; population not generalizable to healthy non-diseased individuals ↩ ↩2 ↩3 ↩4 ↩5

  7. doi:10.3390/biom15010046 · open-label rct · n=50 · Petković-Dabić et al. · Biomolecules 2025 · semaglutide reduced serum IL-6 and hsCRP in obese T2D patients with psoriasis · local copy: downloaded ↩

  8. qin-2026-semaglutide-oa · doi:10.1016/j.cmet.2026.01.008 · pmid:41666927 · in-vivo (mouse OA + obesity, C57BL/6 male) + pilot RCT (ChiCTR2200066291) · Qin H, Yu J, Yu H, et al.; corresponding Tong L, Chen D, Speakman JR, Zhang HT · Cell Metab 2026;38:582–597.e6 · weight-loss-independent chondroprotection via GLP-1R-AMPK-PFKFB3 axis; chondrocyte glycolysis→OXPHOS shift · PDF NOT yet in a local paper archive (DOI not ingested as of 2026-05-08); quantitative claims unverified — use abstract-level claims only ↩ ↩2 ↩3

  9. meurot-2022-liraglutide-oa · doi:10.1038/s41598-022-05323-7 · companion liraglutide-OA paper; intra-articular delivery; verified-true (PDF read end-to-end 2026-05-08) · cited here for class-level GLP-1R cartilage-expression confirmation and the negative Gudbergsen 2021 systemic-liraglutide OA RCT counterpoint ↩

  10. doi:10.1016/S0140-6736(26)00459-9 · pmid:41865758 · Cummings JL, Atri A, Sano M, Zetterberg H, Scheltens P, Knop FK, Johannsen P, Wichmann CA, Abschneider RM, Leon T, Feldman HH · Lancet 2026 Mar 19 (online ahead of print) · two parallel Phase 3 RCTs (evoke NCT04777396, evoke+ NCT04777409) · n=3,808 total (1,855 evoke + 1,953 evoke+); amyloid-confirmed early-stage AD; ages 55–85 (mean 72.2); oral semaglutide 14 mg flexible-dose vs placebo for 156 weeks · PRIMARY ENDPOINT NEGATIVE — CDR-SB change at week 104 estimated difference −0.08 (95% CI −0.35 to 0.20, p=0.57) evoke; 0.10 (95% CI −0.17 to 0.38, p=0.46) evoke+ · safety consistent with prior semaglutide indications · trials discontinued for negative clinical outcome · funded Novo Nordisk · local copy: pending (Lancet OA via CC BY 4.0) ↩

Graph View

Backlinks