Before you read the numbers
No drug on this page stops or reverses Alzheimer's. The newest ones slow the rate of
decline by a measured amount over 18 months. Whether that difference is noticeable in daily life is genuinely
debated among specialists: lecanemab and donanemab, what the trials and the
independent reviews found. The trial numbers, the FDA decisions and the ARIA rates for all three approved
antibodies are side by side on the anti-amyloid evidence page.
The drugs need amyloid confirmed first; what the new blood tests for
Alzheimer's measure, and how accurate they are, is on its own page.
We do not list doses and we never will. Dosing, eligibility and monitoring are decisions
for the prescribing doctor, who knows the whole medical history. Nothing here should be used to start, stop
or change a medicine.
ARIA stands for amyloid-related imaging abnormality — swelling or
small bleeds in the brain, seen on MRI, which can occur with the anti-amyloid antibodies. It is usually
symptom-free but can be serious, which is why these drugs require repeat scans.
Lecanemab (Leqembi)
Basis not stated
Approved
Accelerated approval
Needs biomarker confirmation
- How it works
- Humanized monoclonal antibody that selectively binds soluble amyloid-beta protofibrils, promoting their clearance
- Who it is for
-
early AD (MCI or mild dementia); biomarker-defined (confirmed amyloid pathology)
- Main trial
- Clarity AD (NCT03887455)
- What was measured
- CDR-SB change from baseline at 18 months
- What was found
- -0.45 point difference vs placebo, a 27% slowing of clinical decline (p<0.0001)
- Monitoring required
- Confirm amyloid beta pathology before treatment; obtain a recent baseline brain MRI, then MRI after 1, 2, 3 and 6 months of treatment (e.g. before the 3rd IV dose or 5th SC dose for the 1-month scan); enhanced clinical vigilance for ARIA during the first 14 weeks; ApoE e4 testing should be performed before initiation to inform ARIA risk (current label wording; the stored 'prior to the 5th, 7th and 14th infusions' schedule is from the 2023 label and is no longer in the label).
- In Europe
- approved 2025-04-15 (EU indication restricted to ApoE4 non-carriers or heterozygotes with confirmed amyloid pathology)
Initially approved via accelerated pathway 2023-01-06; converted to traditional approval 2023-07-06 following confirmatory Clarity AD phase 3 results. ARIA-E rate reported 12.6% (vs 1.7% placebo); ARIA-H 17.3% (vs 9.0% placebo); combined/any-ARIA figure of ~21.3% reported in some analyses -- figures vary slightly by source/subgroup (e.g., ApoE4 genotype), cite Clarity AD NEJM publication and Eisai ARIA analyses for exact breakdowns. Subcutaneous maintenance dosing approved later (2025) per Biogen press release. Listed price year: 2023
2026 list price detail: Eisai Inc. Product price list effective 2026-07-13: Leqembi IV 500 mg/5 mL vial $679.60, 200 mg/2 mL vial $271.83 (WAC per vial); LEQEMBI IQLIK 360 mg/1.8 mL autoinjector $385.00; IQLIK 250 mg/1.25 mL carton of two $770.00 (effective on or about 2026-08-24). Eisai IQLIK pricing statement: WAC $385 per autoinjector. Annual IV cost is weight-based, so no single 2026 annual list figure is published by Eisai; $26,500 is the 2023 launch statement figure.
Sources:
fda.gov · drugs.com · eisai.com · alzforum.org · bioarctic.com
Donanemab (Kisunla)
Basis not stated
Approved
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion (monthly)
- FDA: 2024-07-02
- Trial size: 1,736
- ARIA in 16%
ARIA-E on the currently recommended modified titration (Dosing Regimen 2, TRAILBLAZER-ALZ 6, n=212, through 12 months). The original regimen in TRAILBLAZER-ALZ 2 gave ARIA-E 24%, any ARIA 36%; with the modified titration these are 16% and 29%.
- List price $32,000/year
- How it works
- Humanized IgG1 monoclonal antibody targeting pyroglutamate Aβ(p3-42), a form aggregated in amyloid plaques; clears existing plaque
- Who it is for
-
early symptomatic AD (MCI or mild dementia); biomarker-defined (confirmed amyloid pathology via PET or CSF)
- Main trial
- TRAILBLAZER-ALZ 2 (NCT04437511)
- What was measured
- iADRS change from baseline at 18 months (integrated Alzheimer's Disease Rating Scale)
- What was found
- 35% slowing of decline vs placebo in low/medium tau population; 22% slowing in combined (overall) population
- Monitoring required
- Brain MRI required at baseline and prior to 2nd, 3rd, 4th, and 7th infusions to monitor for ARIA; ApoE4 genotype testing recommended prior to treatment to inform ARIA risk; treatment can be stopped once amyloid plaques are cleared
- In Europe
- approved 2025-09-24 (EMA: CHMP opinion adopted 24/07/2025 after re-examination; marketing authorisation issued 24/09/2025; EMEA/H/C/006024; indication restricted to ApoE4 non-carriers or heterozygotes with confirmed amyloid pathology; controlled access programme) - stored date 2025-09-25 is off by one day
Approval label states an approximately 21% ARIA-E rate and 17% ARIA-H rate vs 9% each in placebo (FDA label); alzforum separately cites ARIA-E ~25% and ARIA-H ~one-third in trial-wide safety data -- reported ARIA figures vary by source/subgroup analysis. Enrollment figure varies slightly across sources (1736 randomized vs ~1800 total screened/enrolled); used 1736 as the treated/randomized population commonly cited in trial reporting; primary endpoint result also commonly cited as 35% in low/medium tau subgroup specifically, per FDA and Lilly. Listed price year: 2024
list price detail: Lilly press release 2024-07-02: 'The price of each vial of Kisunla is $695.65'; course-of-therapy cost $12,522 (6 months, 6 infusions), $32,000 (12 months, 13 infusions), $48,696 (18 months, 19 infusions). Lilly's pricinginfo.lilly.com/kisunla page (CMAT-08352 06/2026) renders the current WAC dynamically and returned no price text to curl/jina/WebFetch, so the 2026 WAC could not be captured; 32,000 remains a 2024 figure.
Correction: Stored notes say 'Approval label states an approximately 21% ARIA-E rate and 17% ARIA-H rate vs 9% each in placebo (FDA label)' - this does not match the Kisunla label; those numbers resemble the Leqembi label. Kisunla label (revised 07/2025) figures are given in the ARIA risk note. The label also states: FDA approved the modified titration (350/700/1050/1400 mg) via efficacy supplement S-004 on 2025-07-01 (Lilly press release 2025-07-09).
Sources:
fda.gov · alzforum.org · pricinginfo.lilly.com · alzint.org
Donepezil (Aricept)
Basis not stated
Approved
- Small molecule
- Given: oral (tablet, orally disintegrating tablet, oral solution); transdermal patch (Adlarity) approved 2022
- FDA: 1996-11-25
- How it works
- Reversible acetylcholinesterase inhibitor; increases synaptic acetylcholine availability
- Who it is for
-
mild, moderate, and severe AD; clinical
- Main trial
- multiple registration trials (pre-1996); not a single named pivotal trial
- What was measured
- ADAS-Cog and CIBIC-plus (varies by trial)
- What was found
- Symptomatic cognitive benefit vs placebo across pivotal trials; not disease-modifying
- In Europe
- not centrally authorised: no EMA EPAR exists for donepezil/Aricept (EMA medicines list of centrally authorised products, generated 27/09/2026, has no single-agent donepezil entry; EPAR URL returns 404). Donepezil is authorised nationally in EU member states, so no EU-wide approval date exists.
Symptomatic cholinesterase inhibitor, not a disease-modifying/anti-amyloid therapy. Generic since 2010. Indication expanded to severe AD in 2006/2007. Annual cost not meaningfully comparable given widespread generic availability at low cost; left blank rather than citing a potentially outdated generic price. The 1996-11-25 approval date is consistently corroborated across multiple independent secondary sources (alzforum, alzheimersnewstoday, drugs.com) but the underlying FDA approval-letter PDF (accessdata.fda.gov, NDA 20-690) is a scanned image that could not be parsed as text to directly confirm the exact day.
Sources:
alzforum.org
Rivastigmine (Exelon)
Basis not stated
Approved
- Small molecule
- Given: oral (capsule, solution); transdermal patch
- FDA: 2000-04-21
- How it works
- Reversible inhibitor of both acetylcholinesterase and butyrylcholinesterase
- Who it is for
-
mild to moderate AD (also approved for Parkinson's disease dementia); clinical
- Main trial
- multiple registration trials (pre-2000); not a single named pivotal trial
- What was measured
- ADAS-Cog and CIBIC-plus (varies by trial)
- What was found
- Symptomatic cognitive benefit vs placebo across pivotal trials; not disease-modifying
- In Europe
- approved centrally: Exelon EU marketing authorisation issued 11/05/1998 (EMEA/H/C/000169, opinion adopted 28/01/1998; indication: symptomatic treatment of mild to moderately severe Alzheimer's dementia and of dementia in idiopathic Parkinson's disease); Prometax (same substance) authorised 04/12/1998 (EMEA/H/C/000255). Stored 'exact original date not verified' can be replaced.
Symptomatic cholinesterase inhibitor. Also approved for Parkinson's disease dementia. Generic since ~2011 (capsules) / later for patch. The 2000-04-21 approval date is corroborated by the filename/metadata of the FDA approval letter (accessdata.fda.gov, NDA 20-823, '20823_Exelon_Approv.pdf') and multiple secondary sources, but the PDF itself is a scanned image that could not be parsed as text for direct in-document confirmation.
Sources:
accessdata.fda.gov · alzforum.org
Galantamine (Razadyne)
Basis not stated
Approved
- Small molecule
- Given: oral (tablet, extended-release capsule, oral solution)
- FDA: 2001-02-28
- How it works
- Reversible acetylcholinesterase inhibitor and nicotinic receptor modulator
- Who it is for
-
mild to moderate AD; clinical
- Main trial
- multiple registration trials (pre-2001); not a single named pivotal trial
- What was measured
- ADAS-Cog and CIBIC-plus (varies by trial)
- What was found
- Symptomatic cognitive benefit vs placebo across pivotal trials; not disease-modifying
- In Europe
- not centrally authorised: no EMA EPAR for galantamine (Reminyl/Razadyne) - absent from the EMA medicines list generated 27/09/2026 and the EPAR URL returns 404; authorised nationally in EU member states, so no EU-wide date exists
Originally marketed as Reminyl; renamed Razadyne in 2005 in the US to avoid confusion with the diabetes drug Amaryl. FDA label 'Initial U.S. Approval' field confirms only the year 2001; the exact month/day could not be confirmed from any accessible primary source (accessdata.fda.gov approval letter PDFs for this NDA are scanned images that could not be parsed as text), so the FDA approval date is left blank rather than guessing a day.
Sources:
alzforum.org
Memantine (Namenda)
Basis not stated
Approved
- Small molecule
- Given: oral (tablet, extended-release capsule, oral solution)
- FDA: 2003-10-16
- How it works
- Low-to-moderate affinity NMDA receptor antagonist; reduces excitotoxicity from excess glutamate signaling
- Who it is for
-
moderate to severe AD; clinical
- Main trial
- multiple registration trials (pre-2003); not a single named pivotal trial
- What was measured
- SIB (Severe Impairment Battery) and CIBIC-plus (varies by trial)
- What was found
- Symptomatic cognitive/functional benefit vs placebo across pivotal trials; not disease-modifying
- In Europe
- approved 2002 (EU approval preceded US approval)
NMDA receptor antagonist, mechanistically distinct from cholinesterase inhibitors; often used in combination with donepezil in moderate-severe disease. Namenda XR (extended-release) approved later (2010). The 2003-10-16 approval date is corroborated by the filename/metadata of the FDA approval letter (accessdata.fda.gov, NDA 21-487, '21-487_Namenda_Approv.pdf') and multiple secondary sources, but the PDF itself is a scanned image that could not be parsed as text for direct in-document confirmation. EMA detail: approved 2002-05-15 (Ebixa, H. Lundbeck, EMEA/H/C/000463) and 2002-05-17 (Axura, Merz, EMEA/H/C/000378) - EU approval preceded US approval, as stored
Sources:
alzforum.org
Donepezil/Memantine combination (Namzaric)
Basis not stated
Approved
- Small molecule
- Given: oral (extended-release capsule)
- FDA: 2014-12-23
- How it works
- Fixed-dose combination of NMDA receptor antagonist (memantine ER) and acetylcholinesterase inhibitor (donepezil)
- Who it is for
-
moderate to severe AD (in patients already stabilized on both individual components); clinical
- Main trial
- bioequivalence/pharmacokinetic studies supporting combination product; not a de novo efficacy pivotal trial
- In Europe
- not authorised in the EU: two centrally submitted memantine/donepezil fixed-dose combinations, Acrescent (EMEA/H/C/002424) and Balaxur (EMEA/H/C/002708), carry status 'Refused' in the EMA medicines list; no Namzaric EPAR exists (URL 404). Stored 'not filed / not found' understates this - applications were filed and refused.
Approved as a fixed-dose combination convenience product based on the established efficacy of its two individual components, not a new pivotal efficacy trial. Developed by Adamas/Forest/Actavis.
Sources:
prnewswire.com
Being in a late-stage trial is not evidence that a drug works. Most drugs that reach this
point still fail. The discontinued list below is the proof.
Remternetug
Basis not stated
Phase 3
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion and subcutaneous injection. Subcutaneous self-injection is already in use within TRAILRUNNER-ALZ 1, and TRAILRUNNER-ALZ 3 and the DIAN-TU prevention arm are subcutaneous; a Phase 1 autoinjector-versus-prefilled-syringe PK study (NCT07056309) completed in December 2025.
- Trial size: 1,667
- How it works
- Monoclonal antibody targeting pyroglutamated amyloid-beta aggregated in plaques (next-generation successor to donanemab, engineered for faster plaque clearance and subcutaneous dosing)
- Who it is for
-
TRAILRUNNER-ALZ 1: early symptomatic Alzheimer's disease (MMSE 20-30, ages 60-85). TRAILRUNNER-ALZ 3: early disease with no-to-minimal cognitive or functional impairment (ages 55-80). Also dominantly inherited Alzheimer's disease in the DIAN-TU primary prevention trial (age 18+).; biomarker-defined: amyloid PET for TRAILRUNNER-ALZ 1; a phosphorylated-tau blood result consistent with amyloid pathology (Alzforum specifies plasma p-tau217) for TRAILRUNNER-ALZ 3
- Main trial
- TRAILRUNNER-ALZ 1 (NCT05463731; Phase 3; completed 2026-05-12); TRAILRUNNER-ALZ 3 (NCT06653153; Phase 3; active, not recruiting; estimated primary completion April 2029); plus a remternetug arm in the DIAN-TU primary prevention trial (NCT06647498, n=280 estimated)
- What was measured
- TRAILRUNNER-ALZ 1: proportion of participants reaching amyloid plaque clearance on amyloid PET at Week 52 (the registered sole primary outcome). TRAILRUNNER-ALZ 3: time to clinically meaningful progression on the Clinical Dementia Rating, up to Week 255.
- What was found
- No TRAILRUNNER-ALZ 1 primary-endpoint result is posted on ClinicalTrials.gov and none is reported by Alzforum. The reported amyloid-clearance result is Phase 1 interim data: at 2,800 mg every participant fell below the amyloid-positivity PET threshold within three months.
- Monitoring required
- MRI monitoring for ARIA (protocol-specified; exact schedule not published)
Eli Lilly's next-generation anti-amyloid antibody. The 600 previously recorded as TRAILRUNNER-ALZ 1 enrolment was the placebo-controlled cohort only; the registered actual enrolment is 1,667, with an additional open-label safety cohort. TRAILRUNNER-ALZ 3 (estimated 1,400, subcutaneous only, estimated primary completion April 2029) is testing at-home self-administration. Also in the Knight Family DIAN-TU primary prevention trial (estimated n=280). Interim Phase 1 safety data showed 10 ARIA-E and 7 ARIA-H cases in early cohorts, all ARIA-E in APOE4 carriers (not a population-level percentage). No Phase 3 efficacy or amyloid-clearance readout has been reported.
Sources:
clinicaltrials.gov · clinicaltrials.gov · clinicaltrials.gov · clinicaltrials.gov · alzforum.org
Trontinemab
Basis not stated
Phase 3
Needs biomarker confirmation
- How it works
- Anti-amyloid-beta monoclonal antibody fused to a transferrin-receptor-binding Fab fragment (Brainshuttle technology) to enhance blood-brain-barrier penetration, enabling deep plaque clearance at low doses
- Who it is for
-
early symptomatic AD (also planned for preclinical/secondary prevention); biomarker-defined
- Main trial
- TRONTIER 1 (NCT07169578; phase 3; started 2025-09-17; 1101 enrolled, actual; active, not recruiting; primary completion 2028-02-16) and TRONTIER 2 (NCT07170150; phase 3; started 2025-11-12; 1060 enrolled, actual; active, not recruiting; primary completion 2028-02-16); PrevenTRON (NCT07717411; phase 3 in cognitively unimpaired individuals at risk, ages 55-80; status RECRUITING; estimated enrollment 1600; registry start date 2026-11-30 (estimated); primary completion 2032-09-03)
- What was measured
- TRONTIER 1 and 2: change from baseline to week 72 in CDR-SB (sole registered primary outcome). PrevenTRON: time to progression, defined as confirmed CDR-Global Score >0, up to about 6 years. Amyloid PET clearance was the primary outcome only of the phase 1b/2a Brainshuttle AD study (NCT04639050, Part 3).
- What was found
- In Phase Ib/IIa data, 91% of participants in the 3.6 mg/kg cohort achieved amyloid PET negativity by 28 weeks; 72% achieved deep clearance below 11 centiloids
- Monitoring required
- MRI monitoring for ARIA (protocol-specified)
Roche/Genentech program; ARIA-E rate reported as low as ~1% at the highest dose in early-phase data, a notable safety advantage cited versus lecanemab (~13%) and donanemab (~24%) in company/conference presentations -- these are early-phase (Ib/IIa), not phase 3, safety figures and may not hold at larger scale. Phase 3 TRONTIER trials (800 patients each) began September 2025; efficacy/safety phase 3 readout not yet available. The pivotal-trial size basis: 1101 + 1060 actual enrolment across TRONTIER 1 and 2 (registry); stored 800 was the planned per-trial figure
indication basis: TRONTIER 1/2 enrol ages 50-90 with early symptomatic AD; PrevenTRON ages 55-80 cognitively unimpaired. Stored 'PrevenTRON recruitment began September 2026' is not supported by the registry, which lists an estimated start of 2026-11-30 while showing status RECRUITING (record first posted 2026-07-21, updated 2026-09-10).
Sources:
alzforum.org · roche.com
E2814 (etalanetug)
Basis not stated
Phase 2/3
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion
- Trial size: 197
- How it works
- Anti-tau monoclonal antibody targeting the microtubule-binding region (MTBR) of tau, intended to block cell-to-cell propagation of pathological tau
- Who it is for
-
dominantly inherited Alzheimer's disease (DIAN-TU Tau NexGen, symptomatic and asymptomatic cohorts) AND sporadic early Alzheimer's disease (Eisai phase 2 with concurrent lecanemab, NCT06602258, 105 enrolled actual, started 2024-09-30, active not recruiting, primary completion 2026-12-03) - the stored 'not yet tested in sporadic late-onset AD' is out of date; biomarker-defined
- Main trial
- DIAN-TU Tau NexGen (Phase II/III; DIAN-TU-001 extension), administered alone or with lecanemab
- What was measured
- Tau NexGen: change from week 24 to weeks 104 and 208 in tau PET in the symptomatic population (Cohort 1). Sporadic-AD phase 2 (NCT06602258): change from baseline in CSF MTBR-tau243 at 6 months. Phase 1b/2 Study 103 (NCT04971733, n=8, completed 2024-05-24, results posted): primary outcomes were safety (TEAEs, labs, vitals).
- What was found
- Phase Ib/II data (presented Dec 2025) showed etalanetug reduced plasma eMTBR-tau243 and stabilized/trended toward decreased tau-PET signal; full clinical efficacy results not yet reported
Developed by Eisai/Denali (originally University of Washington/Genentech-licensed). First subject enrolled in the Phase II/III DIAN-TU study Jan 2022. Note this program is being studied specifically in dominantly inherited (autosomal-dominant, early-onset) AD, not yet as a phase 3 program in typical sporadic late-onset AD; labeled here as phase-2 to reflect the biomarker-stage data available, though the DIAN-TU trial is nominally Phase II/III. No clinical (cognitive) efficacy readout located as of this research. Status basis: DIAN-TU Tau NexGen (NCT05269394) is registered as Phase 2/Phase 3, 197 enrolled actual, active not recruiting, primary completion April 2028; plus a separate phase 2 in sporadic early AD (NCT06602258).
Sources:
eisai.com · alzforum.org
ALZ-801 / Valiltramiprosate
Basis not stated
Phase 3 failed
Needs biomarker confirmation
- Small molecule
- Given: oral
- Trial size: 325
- How it works
- Oral small molecule that inhibits amyloid-beta oligomerization/aggregation (prodrug of tramiprosate)
- Who it is for
-
early AD (MCI or mild dementia), specifically APOE4/4 homozygotes in pivotal trial; biomarker-defined (APOE4/4 genotype required in pivotal trial)
- Main trial
- APOLLOE4 (NCT04770220; phase 3; 325 enrolled actual; completed, primary completion 2024-05-29; results posted). Long-term extension NCT06304883 (163 enrolled) TERMINATED - 'Sponsor business decision' (registry updated 2026-06-25).
- What was measured
- ADAS-Cog13 change from baseline to week 78 (co-primary with safety/TEAEs per registry)
- What was found
- Primary endpoint not met: overall APOE4/4 population ADAS-Cog13 LS mean difference -0.51 (ALZ-801 3.89 vs placebo 4.40; MMRM p=0.6058, ClinicalTrials.gov posted results) = 11% slowing, p=0.607 (N=320) in the publication; key secondary A-IADL difference 0.01 (p=0.9966). Hippocampal atrophy slowed 18% vs placebo (p=0.017, N=290) - not '~28% relative to an external ADNI control' as stored. Prespecified MCI stratum (MMSE >26, N=125): ADAS-Cog13 52% slowing (nominal p=0.041), DAD 96% (nominal p=0.016), CDR-SB 102% (nominal p=0.053), hippocampal atrophy slowing 26% (p=0.004); no clinical effect in mild AD (MMSE <=26, N=195). No increased brain edema or microhaemorrhage; nausea, vomiting and decreased appetite more than double placebo.
- Monitoring required
- No increased vasogenic edema/ARIA-type risk reported (oral small molecule, not an anti-amyloid antibody)
Alzheon's oral candidate specifically targeting APOE4/4 homozygotes (~15% of AD patients, at highest genetic risk and often excluded from anti-amyloid antibody trials due to ARIA risk). Topline results presented at AD/PD Vienna, April 1, 2025; peer-reviewed publication announced October 2025. Company continues to pursue regulatory discussions based on subgroup/biomarker findings rather than having met the primary endpoint outright; status reflects an ongoing phase-3-stage program without a clean primary endpoint win. Status basis: 'phase-3' is defensible only as 'phase 3 completed, primary endpoint not met'; the phase 3 extension was terminated for a sponsor business decision in 2026 and no regulatory submission was verified from any page loaded.
Publication: Abushakra S et al. Drugs 2025;85(11):1455-1472 (PMID 41015981, doi 10.1007/s40265-025-02250-5); plasma biomarker analysis Hey JA et al. Drugs 2026;86(10):1705-1721 (PMID 42678603)
Sources:
alzheon.com · neurologylive.com
Kept on the page deliberately. If you read a hopeful headline about an Alzheimer's drug a few
years ago and wondered what happened to it, the answer is often here. "Failed" means the pivotal
trial finished and missed its primary endpoint.
Aducanumab (Aduhelm)
Basis not stated
Discontinued
Accelerated approval
Needs biomarker confirmation
- How it works
- Human monoclonal antibody targeting aggregated soluble oligomers and insoluble fibrils of amyloid-beta
- Who it is for
-
early AD (MCI or mild dementia); biomarker-defined (confirmed amyloid pathology)
- Main trial
- EMERGE (NCT02484547) and ENGAGE (NCT02477800)
- What was measured
- CDR-SB change from baseline at 78 weeks
- What was found
- EMERGE met primary endpoint at high dose (significantly slower decline vs placebo); ENGAGE, an identically designed trial, did not meet its primary endpoint -- discordant results were central to controversy over approval
- Monitoring required
- Obtain a recent (within one year) brain MRI prior to initiating treatment; obtain MRIs prior to the 5th infusion (first 6 mg/kg dose), 7th (first 10 mg/kg dose), 9th and 12th infusions (label revised 8/2023) - stored 'prior to 7th and 12th infusions' omits the 5th and 9th
- In Europe
- application withdrawn: CHMP recommended refusal in December 2021; Biogen requested re-examination but withdrew the application on 20 April 2022 before re-examination concluded (EMA withdrawn-applications page; EMA medicines list status 'Application withdrawn', EMEA/H/C/005558)
Approved via accelerated pathway 2021-06-07 based on amyloid-lowering as a surrogate marker, despite an FDA advisory committee having voted against approval; highly controversial approval. Launch price was $56,000/year, cut ~50% to $28,200 in December 2021 amid backlash and restricted CMS coverage. Biogen announced discontinuation on 2024-01-31, ending the confirmatory ENVISION trial; stated reason was business reprioritization toward lecanemab and other pipeline assets, explicitly not safety or efficacy concerns (per Biogen statement), though analysts widely attribute the decision to poor commercial uptake following restrictive CMS coverage. Patients on prescription could continue treatment through November 1, 2024. ARIA-E rate reported ~35% vs 3% placebo; ARIA-H microhemorrhage 19% vs 7%; ARIA-H superficial siderosis 15% vs 2% (Wikipedia summary of trial data, cross-referenced against alzforum figures of ~one-third ARIA-E and 17% ARIA-H). Listed price year: 2022
the pivotal-trial size basis: label section 14: Study 1 (EMERGE) 1638 randomised 1:1:1 and Study 2 (ENGAGE) 1647 randomised 1:1:1 = 3285 across the two phase 3 trials; stored 1638 is EMERGE alone while the pivotal-trial entry names both trials status basis: Biogen discontinued development and commercialisation on 2024-01-31 and terminated the confirmatory ENVISION study; accelerated approval was never converted. Drugs@FDA still lists BLA 761178 products with marketing status 'Prescription' and no withdrawal action, so 'accelerated-then-withdrawn' would overstate a formal regulatory withdrawal; the product is no longer on DailyMed.
Sources:
alzforum.org · alz.org · en.wikipedia.org
Semaglutide (Alzheimer's disease program) (Rybelsus / Ozempic / Wegovy (approved brands in other indications; not AD-specific))
Basis not stated
Discontinued
Needs biomarker confirmation
- Small molecule (peptide)
- Given: oral
- Trial size: 1,855
- How it works
- GLP-1 receptor agonist; hypothesized neuroprotective/anti-inflammatory effects unrelated to amyloid or tau clearance
- Who it is for
-
early-stage symptomatic AD (MCI or mild dementia); biomarker-defined (confirmed amyloid positivity)
- Main trial
- EVOKE (NCT04777396) and EVOKE+ (NCT04777409)
- What was measured
- CDR-SB change from baseline at week 104
- What was found
- No significant difference vs placebo: mean CDR-SB change 2.3 in both semaglutide and placebo arms in EVOKE; estimated difference -0.08 (95% CI -0.35 to 0.20), p=0.57 -- trial did not meet primary endpoint
Novo Nordisk announced topline failure of both EVOKE and EVOKE+ phase 3 trials on 2025-11-24; combined enrollment across both trials exceeded 3,800 participants. Oral semaglutide 14mg did not slow clinical decline on CDR-SB despite producing statistically significant (~10%) reductions in some neuroinflammation-related biomarkers -- explicitly not disease-modifying at the clinical level. Not an amyloid- or tau-targeted therapy; ARIA not applicable/not reported as this is not an anti-amyloid antibody. Status marked 'discontinued' for the AD indication specifically; semaglutide remains approved for diabetes/obesity under other brand names.
Sources:
alz.org · clinicaltrials.gov · clinicaltrials.gov
Bepranemab
Basis not stated
Discontinued
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion
- Trial size: 466
- How it works
- Anti-tau monoclonal antibody targeting mid-region tau, intended to block extracellular tau propagation
- Who it is for
-
mild cognitive impairment / mild-to-moderate AD; biomarker-defined
- Main trial
- TOGETHER (Phase 2)
- What was measured
- CDR-SB change from baseline at 80 weeks
- What was found
- Primary endpoint not met at either the 45 mg/kg or 90 mg/kg dose; tau-PET showed up to 58% slowing of tau accumulation vs placebo (secondary/exploratory endpoint) and some cognitive benefit signal in ApoE4 non-carriers with low baseline tau
Originally partnered between UCB and Roche/Genentech (Roche paid $120M upfront, up to $2B in milestones in 2020); Roche/Genentech terminated the collaboration and returned global rights to UCB around October 2024, shortly before the Phase 2 TOGETHER primary endpoint miss was announced at CTAD (Clinical Trials on Alzheimer's Disease) 2024. Despite the primary endpoint failure, UCB highlighted positive secondary/exploratory tau-PET and subgroup cognitive findings; status marked 'discontinued' reflecting the Roche exit and primary endpoint failure, though UCB indicated it may continue analyzing data.
Sources:
fiercebiotech.com · alzforum.org · alzforum.org
Semorinemab
Basis not stated
Discontinued
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion
- Trial size: 457
- How it works
- Anti-tau monoclonal antibody targeting the N-terminal region of all six tau isoforms
- Who it is for
-
prodromal to mild AD (TAURIEL); moderate AD (LAURIET); biomarker-defined
- Main trial
- TAURIEL (Phase 2)
- What was measured
- CDR-SB change from baseline over 73 weeks
- What was found
- Primary endpoint not met: no significant slowing of decline on CDR-SB vs placebo; secondary endpoints (ADAS-Cog13, ADCS-ADL) also not met; safety endpoint (adverse event rate similar to placebo) was met
Developed by AC Immune/Genentech (Roche). TAURIEL (prodromal-to-mild AD) failed its primary endpoint. A second Phase 2 trial in moderate AD (LAURIET) also did not demonstrate clinically meaningful benefit on cognitive measures, though it reportedly showed slowing on some functional/behavioral measures in exploratory analyses; Genentech discontinued further development of semorinemab for AD following these results.
Sources:
alzheimersnewstoday.com · neurologyadvisor.com
ALZ-801 / Valiltramiprosate
Basis not stated
Phase 3 failed
Needs biomarker confirmation
- Small molecule
- Given: oral
- Trial size: 325
- How it works
- Oral small molecule that inhibits amyloid-beta oligomerization/aggregation (prodrug of tramiprosate)
- Who it is for
-
early AD (MCI or mild dementia), specifically APOE4/4 homozygotes in pivotal trial; biomarker-defined (APOE4/4 genotype required in pivotal trial)
- Main trial
- APOLLOE4 (NCT04770220; phase 3; 325 enrolled actual; completed, primary completion 2024-05-29; results posted). Long-term extension NCT06304883 (163 enrolled) TERMINATED - 'Sponsor business decision' (registry updated 2026-06-25).
- What was measured
- ADAS-Cog13 change from baseline to week 78 (co-primary with safety/TEAEs per registry)
- What was found
- Primary endpoint not met: overall APOE4/4 population ADAS-Cog13 LS mean difference -0.51 (ALZ-801 3.89 vs placebo 4.40; MMRM p=0.6058, ClinicalTrials.gov posted results) = 11% slowing, p=0.607 (N=320) in the publication; key secondary A-IADL difference 0.01 (p=0.9966). Hippocampal atrophy slowed 18% vs placebo (p=0.017, N=290) - not '~28% relative to an external ADNI control' as stored. Prespecified MCI stratum (MMSE >26, N=125): ADAS-Cog13 52% slowing (nominal p=0.041), DAD 96% (nominal p=0.016), CDR-SB 102% (nominal p=0.053), hippocampal atrophy slowing 26% (p=0.004); no clinical effect in mild AD (MMSE <=26, N=195). No increased brain edema or microhaemorrhage; nausea, vomiting and decreased appetite more than double placebo.
- Monitoring required
- No increased vasogenic edema/ARIA-type risk reported (oral small molecule, not an anti-amyloid antibody)
Alzheon's oral candidate specifically targeting APOE4/4 homozygotes (~15% of AD patients, at highest genetic risk and often excluded from anti-amyloid antibody trials due to ARIA risk). Topline results presented at AD/PD Vienna, April 1, 2025; peer-reviewed publication announced October 2025. Company continues to pursue regulatory discussions based on subgroup/biomarker findings rather than having met the primary endpoint outright; status reflects an ongoing phase-3-stage program without a clean primary endpoint win. Status basis: 'phase-3' is defensible only as 'phase 3 completed, primary endpoint not met'; the phase 3 extension was terminated for a sponsor business decision in 2026 and no regulatory submission was verified from any page loaded.
Publication: Abushakra S et al. Drugs 2025;85(11):1455-1472 (PMID 41015981, doi 10.1007/s40265-025-02250-5); plasma biomarker analysis Hey JA et al. Drugs 2026;86(10):1705-1721 (PMID 42678603)
Sources:
alzheon.com · neurologylive.com
Gantenerumab
Basis not stated
Discontinued
Needs biomarker confirmation
- How it works
- Fully human monoclonal antibody targeting aggregated amyloid-beta (both fibrillar and plaque forms)
- Who it is for
-
early AD (prodromal to mild); biomarker-defined
- Main trial
- GRADUATE I and GRADUATE II (Phase 3)
- What was measured
- CDR-SB change from baseline at 116 weeks
- What was found
- Neither trial reached statistical significance: GRADUATE I showed an 8% relative reduction in decline (-0.31 points, p=0.0954); GRADUATE II showed a 6% relative reduction (-0.19 points, p=0.2998) -- both trials failed their primary endpoint
Roche/Genentech announced negative topline GRADUATE I/II results on 2022-11-14; combined N=1,965 across both trials. Drug removed roughly half the amyloid plaque expected based on earlier-phase data, which was cited as a likely contributor to the efficacy miss. Pooled ARIA-E incidence was 25% (mostly asymptomatic, few discontinuations). Roche subsequently discontinued the gantenerumab clinical development program for sporadic AD (a related trial in dominantly inherited AD, DIAN-TU, also did not show benefit).
Sources:
roche.com · alzforum.org
Solanezumab
Basis not stated
Discontinued
Needs biomarker confirmation
- Monoclonal antibody
- Given: IV infusion
- Trial size: 2,129
- How it works
- Humanized monoclonal antibody targeting soluble monomeric amyloid-beta
- Who it is for
-
mild dementia due to AD; biomarker-defined (confirmed amyloid accumulation, required in EXPEDITION3)
- Main trial
- EXPEDITION3
- What was measured
- ADAS-Cog14 change from baseline at 80 weeks
- What was found
- Did not meet primary endpoint; difference vs placebo not statistically significant (p=0.095); this was the third consecutive Phase 3 failure for solanezumab across EXPEDITION, EXPEDITION2, and EXPEDITION3
Eli Lilly discontinued regulatory plans for solanezumab in mild AD dementia following the EXPEDITION3 failure (announced Nov 2016). Targets soluble monomeric amyloid-beta rather than aggregated/plaque forms, a mechanistic distinction from lecanemab/donanemab that may partly explain its lack of efficacy. Also separately tested (and failed) in the DIAN-TU dominantly-inherited AD prevention trial and the A4 preclinical-AD prevention trial.
Sources:
alzforum.org
Crenezumab
Basis not stated
Discontinued
Needs biomarker confirmation
- Monoclonal antibody
- Given: subcutaneous injection / IV infusion
- Trial size: 1,500
- How it works
- Humanized IgG4 monoclonal antibody engineered for reduced effector function, targeting aggregated amyloid-beta oligomers
- Who it is for
-
early/prodromal AD (CREAD); also tested in preclinical autosomal-dominant AD (API-ADAD); biomarker-defined (confirmed amyloid pathology)
- Main trial
- CREAD 1 and CREAD 2 (Phase 3)
- What was measured
- CDR-SB change from baseline
- What was found
- Trials stopped early (January 2019) after an interim futility analysis found they were unlikely to meet the primary endpoint of slowing decline on CDR-SB; no new safety signals identified
Genentech/Roche and AC Immune program; CREAD 1 (started 2016) and CREAD 2 (started mid-2017) used doses 4x higher than in earlier Phase 2 testing. Terminated for futility, not safety. A related Phase 2 trial in carriers of an autosomal-dominant AD mutation (API-ADAD, in Colombia) also failed to meet its primary endpoint (reported 2022), though some exploratory biomarker signals were later highlighted by AC Immune.
Sources:
gene.com · alzforum.org
Bapineuzumab
Basis not stated
Discontinued
- Monoclonal antibody
- Given: IV infusion
- How it works
- Humanized monoclonal antibody targeting the N-terminus of aggregated amyloid-beta
- Who it is for
-
mild to moderate AD; clinical, stratified by ApoE4 carrier status
- Main trial
- Studies 301 and 302 (Phase 3)
- What was measured
- Co-primary endpoints: cognitive (ADAS-Cog) and functional (DAD) change from baseline
- What was found
- Co-primary endpoints not met in either the ApoE4 carrier study (301, announced July 2012) or the ApoE4 non-carrier study (302, announced August 2012)
One of the earliest anti-amyloid monoclonal antibodies tested in AD; developed by Janssen AI (Johnson & Johnson) and Pfizer. Phase 3 IV development in mild-to-moderate AD discontinued August 2012 after both co-primary endpoints were missed in ApoE4 carriers (Study 301) and non-carriers (Study 302). Notable for establishing ARIA (then called vasogenic edema) as a class safety signal for anti-amyloid antibodies, particularly in ApoE4 carriers, that informed the design of all later programs (including lecanemab and donanemab). Exact trial N not found in accessible sources; left blank rather than estimating.
Sources:
jnj.com · pfizer.com
Verubecestat
Basis not stated
Discontinued
- Small molecule
- Given: oral
- How it works
- Oral BACE1 (beta-secretase) inhibitor; blocks the enzymatic step that generates amyloid-beta from APP
- Who it is for
-
mild-to-moderate AD (EPOCH); prodromal AD (APECS); clinical in EPOCH; biomarker-defined in APECS
- Main trial
- EPOCH (mild-to-moderate AD, Phase 3); APECS (prodromal AD, Phase 2/3)
- What was measured
- ADAS-Cog and CDR-SB (varies by trial)
- What was found
- EPOCH stopped February 2017 for futility (unlikely to show positive benefit-risk); APECS stopped early 2018 after showing worsening cognition (worse ADAS-Cog13 and CDR-SB scores) and hippocampal atrophy vs placebo at the 40mg dose
Merck's BACE1 inhibitor. EPOCH trial stopped for futility, not safety; the later APECS trial (prodromal/earlier-stage AD) was stopped after data showed dose-related cognitive worsening and hippocampal atrophy versus placebo -- an unexpected class-level concern for BACE inhibitors (excessive BACE inhibition may impair non-amyloid substrates important for synaptic function). No significant hepatotoxicity was reported for verubecestat specifically (distinguishing it from atabecestat's liver-toxicity-driven discontinuation).
Sources:
merck.com · alzres.biomedcentral.com
Lanabecestat
Basis not stated
Discontinued
Needs biomarker confirmation
- Small molecule
- Given: oral
- How it works
- Oral BACE1/BACE2 inhibitor; blocks amyloid-beta generation from APP
- Who it is for
-
early AD (AMARANTH); mild AD dementia (DAYBREAK-ALZ); biomarker-defined
- Main trial
- AMARANTH and DAYBREAK-ALZ (Phase 3)
- What was measured
- Cognitive/functional composite endpoints (varied by trial)
- What was found
- Both trials stopped for futility on recommendation of an independent data monitoring committee (June 12, 2018); neither the 20mg nor 50mg dose moved primary or secondary outcome measures relative to expected placebo decline
Jointly developed by AstraZeneca and Eli Lilly. Discontinued for futility (lack of efficacy), explicitly not for safety concerns, per the companies' joint statement on 2018-06-12.
Sources:
astrazeneca.com · investor.lilly.com
Atabecestat
Basis not stated
Discontinued
Needs biomarker confirmation
- Small molecule
- Given: oral
- How it works
- Oral BACE1 inhibitor; blocks amyloid-beta generation from APP
- Who it is for
-
preclinical (late-onset) AD; biomarker-defined
- Main trial
- EARLY (NCT02569398, Phase 2b/3)
- What was measured
- Safety/tolerability (liver enzymes) and cognitive composite
- What was found
- Trial halted before efficacy readout due to serious liver enzyme elevations (hepatotoxicity) in a subset of participants; post-hoc analysis of available cognitive data also suggested dose-related cognitive worsening
Janssen (Johnson & Johnson) program. Halted due to hepatotoxicity (serious liver enzyme elevations); later analysis of drug-induced liver injury cases suggested an immune-mediated mechanism (T cells responsive to atabecestat/its metabolites). This is the clearest liver-toxicity-driven BACE inhibitor discontinuation among the four BACE agents in this dataset.
Sources:
alzheimersnewstoday.com · alzforum.org
Elenbecestat
Basis not stated
Discontinued
Needs biomarker confirmation
- Small molecule
- Given: oral
- How it works
- Oral BACE inhibitor; blocks amyloid-beta generation from APP
- Who it is for
-
early AD (MCI due to AD and mild AD dementia); biomarker-defined
- Main trial
- MISSION AD1 (NCT02956486) and MISSION AD2 (NCT03036280), Phase 3
- What was measured
- CDR-SB change from baseline (design; trial stopped before readout)
- What was found
- Both trials discontinued on 2019-09-13 on recommendation of an independent data monitoring committee, which found an unfavorable overall risk-benefit ratio; efficacy readout was not completed
Jointly developed by Eisai and Biogen. Discontinued based on an unfavorable risk-benefit assessment by an independent data monitoring committee rather than a specific single adverse event category; the long-term extension of the earlier Phase 2 study (Study 202) was also stopped. Effectively ended the class-wide push for BACE inhibitors in AD alongside verubecestat, lanabecestat, and atabecestat.
Sources:
eisai.com · investors.biogen.com