Peptide Vaccines for Alzheimer's Disease
Neurodegenerative Peptides
53% risk reduction
In the ABvac40 phase 2 trial, vaccinated patients had a 53% reduced risk of meaningful cognitive decline at 24 months compared to placebo.
Pascual-Lucas et al., Alzheimer's & Dementia, 2025
Pascual-Lucas et al., Alzheimer's & Dementia, 2025
View as imageThe idea behind peptide vaccines for Alzheimer's disease is straightforward: train the immune system to produce antibodies against amyloid-beta (Abeta), the peptide fragment that accumulates in brain plaques. If the body can clear these plaques through its own antibody response, it might slow or prevent the cognitive decline that defines the disease. The first attempt (AN1792 in 2001) was halted when 6% of patients developed brain inflammation. Twenty years of iterative design later, second-generation peptide vaccines using shorter Abeta fragments have reached clinical trials with promising results. ABvac40, targeting the C-terminus of Abeta40, showed reduced cognitive decline in a phase 2 trial published in 2025. For context on why amyloid-beta is central to Alzheimer's research, see our pillar article on amyloid-beta.
Key Takeaways
- The amyloid hypothesis, now over 30 years old, posits that accumulation of amyloid-beta peptide fragments drives Alzheimer's disease pathology (Selkoe & Hardy, 2016)
- AN1792, the first Abeta vaccine (full-length Abeta1-42), was halted in 2002 when 6% of patients developed T-cell-mediated meningoencephalitis
- ABvac40 (targeting Abeta33-40) reduced the risk of meaningful cognitive decline by 53% at 24 months with no cases of ARIA-E or meningoencephalitis (Pascual-Lucas et al., 2025)
- Second-generation vaccines use shorter Abeta peptide fragments (Abeta1-15 or Abeta33-40) to generate antibodies while avoiding the T-cell epitopes that caused inflammation
- Multiple therapeutic strategies targeting amyloid exist: active vaccines (peptide-based), passive immunotherapy (antibody infusions like lecanemab), and peptide inhibitors of aggregation (Jeremic et al., 2021)
- LL-37 antimicrobial peptide fragments modulate amyloid-beta aggregation, revealing unexpected interactions between innate immunity and neurodegeneration (Wang et al., 2025)
The Amyloid Hypothesis: Why Target This Peptide?
Amyloid-beta is not a foreign invader. It is a natural peptide produced by the normal cleavage of amyloid precursor protein (APP), a transmembrane protein expressed throughout the body. In healthy brains, Abeta is produced and cleared continuously. In Alzheimer's, this balance breaks down: production increases, clearance decreases, or both, leading to accumulation of Abeta into oligomers, protofibrils, and eventually the dense amyloid plaques visible on brain imaging and autopsy.
Selkoe and Hardy (2016) reviewed 25 years of evidence supporting the amyloid hypothesis.[1] The strongest genetic evidence comes from familial Alzheimer's disease: every known genetic cause involves mutations that increase Abeta production or shift the ratio of Abeta42 (more aggregation-prone) to Abeta40 (less aggregation-prone). Conversely, a naturally occurring APP mutation that reduces Abeta production protects against Alzheimer's in the Icelandic population.
The hypothesis has faced criticism because many anti-amyloid therapies failed clinically. But the recent approvals of lecanemab and donanemab (monoclonal antibodies that clear amyloid plaques and modestly slow cognitive decline) have reinvigorated interest. If passive immunotherapy works, could active immunization with peptide vaccines achieve the same result more cheaply and durably?
AN1792: The First Attempt and Its Failure
In 2001, researchers administered AN1792, a vaccine consisting of full-length synthetic Abeta1-42 peptide with QS-21 adjuvant, to patients with mild-to-moderate Alzheimer's. The approach was based on dramatic results in mouse models where vaccination cleared amyloid plaques and improved behavior.
The phase 2 trial enrolled 372 patients but was halted in 2002 when 18 patients (6%) developed meningoencephalitis, a T-cell-mediated brain inflammation. Post-mortem studies revealed that some vaccinated patients had extensive plaque clearance, confirming the vaccine's biological activity. Some patients who generated antibody responses showed slower cognitive decline. But the inflammatory side effect was unacceptable.
The critical lesson from AN1792 was that the full-length Abeta1-42 peptide contains both B-cell epitopes (the parts that generate antibodies, located at the N-terminus) and T-cell epitopes (the parts that activate T cells, located at the C-terminus). The brain inflammation was caused by T-cell activation, not the antibody response. This insight directly informed the design of every subsequent peptide vaccine for Alzheimer's.
Second-Generation Vaccines: Removing the T-Cell Problem
After AN1792, researchers redesigned Abeta vaccines to include only the antibody-generating portions of the peptide, removing the T-cell epitopes that caused encephalitis. Jeremic et al. (2021) reviewed the full landscape of anti-amyloid therapeutic strategies, including second-generation active vaccines.[2]
ABvac40 uses a synthetic peptide corresponding to amino acids 33-40 of Abeta40, conjugated to keyhole limpet hemocyanin (KLH) carrier protein and formulated with aluminum hydroxide adjuvant. By targeting the C-terminus of Abeta40 (rather than the full sequence), ABvac40 generates antibodies against Abeta40 without activating the T-cell epitopes that caused problems with AN1792.
The phase 1 trial (2018) demonstrated safety and immunogenicity: the vaccine generated anti-Abeta40 antibodies with no serious adverse events. The phase 2 trial (Pascual-Lucas et al., 2025) randomized patients with mild cognitive impairment or very mild Alzheimer's to ABvac40 or placebo. At 24 months:
- Vaccinated patients had a 53% reduced risk of meaningful within-patient cognitive changes (HR 0.47, 95% CI 0.27-0.85, P = 0.012)
- No cases of ARIA-E (amyloid-related imaging abnormalities, a brain swelling side effect common with antibody therapies) were observed
- No meningoencephalitis occurred
- ARIA-H (microhemorrhages) rates were similar between groups
ACI-24 took a different approach, using the Abeta1-15 sequence (the N-terminal portion containing B-cell epitopes) displayed on a liposome platform. By truncating the peptide at position 15, the vaccine avoids the C-terminal T-cell epitopes entirely. ACI-24 entered clinical testing but the original formulation was withdrawn to undergo optimization. An improved version (ACI-24.060) was subsequently licensed by Takeda for further development.
Active vs Passive Immunotherapy: Why Vaccines Matter
Lecanemab and donanemab (passive immunotherapy) have demonstrated that clearing amyloid plaques can slow cognitive decline. But they require intravenous infusions every 2-4 weeks at costs of approximately $26,500/year, and 20-35% of patients develop ARIA (brain swelling or microhemorrhages) that requires monitoring with serial MRI scans.
Active peptide vaccination offers theoretical advantages:
- Cost: A vaccine series (3-5 injections over months) would cost a fraction of ongoing antibody infusions
- Durability: The immune system maintains antibody production through memory B cells, potentially providing sustained protection
- Safety signal: ABvac40 showed no ARIA-E, possibly because the vaccine generates a different antibody profile than monoclonal antibodies, with a more gradual, physiological rate of plaque clearance
- Accessibility: Vaccines can be stored and administered in primary care settings, while antibody infusions require specialized infusion centers
The disadvantage is control. With passive immunotherapy, the dose and timing of antibody exposure are precisely controlled. With a vaccine, the immune response varies between individuals and cannot be easily titrated or stopped if problems arise.
The Tau Problem: Why Amyloid Vaccines May Not Be Enough
Alzheimer's pathology involves two hallmark protein aggregates: amyloid-beta plaques and tau neurofibrillary tangles. Anti-amyloid vaccines address only the first. Tau pathology correlates more closely with cognitive decline than amyloid plaque burden, and tau spreads through the brain in a stereotypical pattern that tracks with symptom progression.
This raises a fundamental question about peptide vaccine strategy: even if a vaccine successfully clears amyloid plaques, will it halt the tau-driven neurodegeneration that may already be self-sustaining? The lecanemab and donanemab trials showed that amyloid clearance slowed but did not stop cognitive decline, consistent with the idea that tau and other downstream processes continue independently.
Combination approaches targeting both amyloid and tau are being considered. Anti-tau peptide vaccines are in earlier stages of development, with several candidates in preclinical testing. The challenge with tau vaccines is identifying which tau epitopes are pathogenic (tau exists in many forms, and most are normal) and ensuring that anti-tau antibodies can access intracellular tau aggregates, since tau pathology is primarily inside neurons while antibodies are outside. The ideal future Alzheimer's peptide vaccine might combine amyloid and tau epitopes in a single formulation, addressing both major pathological hallmarks simultaneously.
Peptides That Block Amyloid Aggregation
Beyond vaccination, peptides are being developed to directly prevent Abeta from forming toxic aggregates. Kalita et al. (2020) designed peptidomimetics using tail-to-side-chain stapling that inhibit Abeta amyloid fibrillization.[3] These stapled peptides bind to Abeta monomers and prevent them from assembling into the oligomers and fibrils that are toxic to neurons.
Wang et al. (2025) made an unexpected discovery: fragments of the antimicrobial peptide LL-37 can modulate Abeta aggregation dynamics.[4] LL-37 truncated fragments interact with Abeta and alter its aggregation pathway, potentially redirecting it toward less toxic forms. This finding connects innate immune defense peptides to neurodegenerative disease in ways previously unrecognized.
These aggregation-inhibiting approaches complement vaccination strategies. A vaccine clears existing plaques through antibody-mediated phagocytosis. An aggregation inhibitor prevents new plaques from forming. Used together, they could potentially address both existing pathology and ongoing Abeta production.
Neuroprotective Peptides: Protecting Neurons While Clearing Plaques
Even if amyloid clearance succeeds, neurons damaged by years of Abeta exposure may not recover without additional support. Neuroprotective peptides represent a complementary approach.
Humanin, first identified by Tajima et al. (2002), is a 24-amino-acid peptide that protects neurons against Abeta toxicity.[5] Karachaliou et al. (2023) reviewed the accumulating evidence for humanin and its analogs as potential Alzheimer's therapeutics.[6] Humanin blocks Abeta-induced apoptosis through multiple mechanisms, including interaction with BAX (a pro-apoptotic protein) and activation of survival signaling pathways. Humanin analogs with improved potency and stability have been developed for preclinical testing.
NAP peptide (davunetide), derived from activity-dependent neuroprotective protein (ADNP), is another neuroprotective peptide that reached clinical trials for neurodegeneration. Gozes et al. (2009) described its development for addressing Alzheimer's tangles (tau pathology), representing a peptide approach that targets the other major pathological hallmark of the disease.[7] For a detailed look at this peptide, see our article on NAP peptide (davunetide).
Where Peptide Vaccines Stand in the Alzheimer's Pipeline
The field has matured from the AN1792 failure to a position where peptide vaccines are demonstrating clinical signals:
ABvac40 has the strongest data: a positive phase 2 trial with a 53% reduction in cognitive decline risk and no ARIA-E. Phase 3 trials would need to confirm these results in a larger, longer study.
ACI-24.060 is being developed by Takeda/AC Immune with an optimized formulation designed to improve the antibody response consistency.
Combination strategies pairing active vaccination with passive antibody therapy, anti-tau approaches, or neuroprotective peptides are being considered to address multiple disease mechanisms simultaneously.
The approval of lecanemab and donanemab established that amyloid clearance has clinical benefit, removing the biggest uncertainty about the vaccine approach. The remaining questions are whether a vaccine can generate sufficient antibodies, whether the antibody profile is safe, and whether the effect size is clinically meaningful. ABvac40's phase 2 data suggests the answers may be yes on all three counts. The vaccine approach also aligns with the growing recognition that Alzheimer's prevention (treating people before symptoms appear) may be more effective than treatment of established disease. A safe, inexpensive peptide vaccine could potentially be administered to at-risk individuals identified through genetic testing or biomarker screening years before cognitive symptoms develop, a strategy that would be impractical with costly antibody infusions. For context on how other peptide approaches to Alzheimer's are progressing, see our dedicated pipeline article.
The Bottom Line
Peptide vaccines for Alzheimer's have evolved from the failed AN1792 (full-length Abeta1-42, halted due to encephalitis) to second-generation designs using shorter fragments that generate antibodies without activating inflammatory T cells. ABvac40 (targeting Abeta33-40) showed a 53% reduction in cognitive decline risk at 24 months with no ARIA-E in a phase 2 trial. Complementary peptide strategies include aggregation inhibitors that prevent plaque formation and neuroprotective peptides (humanin, NAP) that protect neurons from Abeta toxicity.