DOI-10-1016-S0140-6736-26-00459-9 — EVOKE / EVOKE+ Primary Results (Lancet 2026)

[DRAFT — authored 2026-04-19. Citation verified against Lancet and Novo Nordisk press release 2026-04-19. PubMed PMID may not yet be assigned at time of authoring; DOI filename is used per task spec.]

Citation (working)

Johannsen P, Cummings JL, Atri A, Feldman HH, Hansson O, Hendrix SB, Ismail Z, Mummery CJ, Stubbings V, Aggarwal A, Liu-Seifert H, Waldemar G, Kivipelto M; EVOKE and EVOKE+ Investigators. Efficacy and safety of oral semaglutide 14 mg (flexible dose) in early-stage symptomatic Alzheimer's disease (evoke and evoke+): two phase 3, randomised, placebo-controlled trials. Lancet. 2026 (online ahead of print). doi: 10.1016/S0140-6736(26)00459-9.

External URL: DOI resolver

Study Design

Design: Two parallel Phase 3 randomized, double-blind, placebo-controlled trials (evoke and evoke+)
N: 3,808 total across both trials (NCT04777396 evoke + NCT04777409 evoke+)
Population: Adults 55-85 years with early-stage symptomatic AD (MCI or mild dementia) + biomarker-confirmed amyloid positivity
Intervention: Oral semaglutide 14 mg daily (flexible dose) vs placebo
Treatment duration: 104 weeks (+52 week extension planned)
Primary endpoint: Change in Clinical Dementia Rating – Sum of Boxes (CDR-SB)

Key Findings

Primary endpoint (CDR-SB)

Null finding: Oral semaglutide 14 mg was NOT efficacious in slowing clinical progression as measured by CDR-SB
Trials did not confirm superiority of semaglutide over placebo on the primary endpoint

Secondary findings

hsCRP: Semaglutide reduced hsCRP with estimated treatment ratios 0.76 (95% CI 0.64-0.90) in EVOKE and 0.71 (95% CI 0.62-0.82) in EVOKE+ — anti-inflammatory biomarker signal consistent with class
AD-related biomarkers: Some biomarker improvements observed — but did not translate into clinical cognitive benefit
Cognitive secondary endpoints: Broadly consistent with primary — no clinically meaningful separation from placebo

Safety profile

Safety and tolerability of semaglutide in early AD consistent with profile in other indications (obesity, T2D)
GI events as expected; no new AD-specific safety signals

Regulatory / program implications

1-year extension discontinued based on the primary-endpoint failure
Novo Nordisk announced: will not pursue further AD development

Clinical Relevance

EVOKE / EVOKE+ is the landmark Phase 3 evidence that semaglutide (and by extrapolation, the GLP-1 class at approved doses) does not slow clinical progression in early-stage Alzheimer's disease as measured by CDR-SB. Core teaching points:

Null primary endpoint is definitive — the largest-to-date Phase 3 GLP-1 RA trial in AD, biomarker-confirmed population, adequately powered.
Biomarker improvements do not guarantee clinical efficacy — hsCRP and AD biomarkers responded to semaglutide, but CDR-SB did not. An important lesson in biomarker-to-clinical-outcome translation.
Implication for prescribing: Clinicians should NOT prescribe semaglutide for AD-related cognitive indications. Observational / real-world signals suggesting GLP-1 benefit for cognition (e.g., in T2D cohorts) do not survive Phase 3 scrutiny at approved doses in an AD-specific population.
Does not exclude benefit in non-AD cognitive contexts — e.g., diabetic encephalopathy, vascular cognitive impairment, or earlier-stage cognitive decline. But AD trials are negative.

Linked Peptides

Semaglutide

Related Lessons

Lesson 5.2 — Semaglutide Deep Dive (Investigational / Cognition section — null finding framing)
Lesson 5.4 — Emerging indications pipeline

Related Studies

PMID-39780249 – EVOKE EVOKE+ Design Oral Semaglutide Alzheimer — design paper companion

Orchestrator Notes

Citation filed under DOI prefix per task spec; PMID may be assigned after full Lancet 2026 PubMed indexing. Update filename and frontmatter to PMID when available.

DOI-10-1016-S0140-6736-26-00459-9 – EVOKE Primary Results Johannsen Lancet 2026