- Roughly 1 in 4 people carry at least one APOE4 allele. One copy is associated with a 3 to 4-fold higher risk of late-onset Alzheimer's disease; two copies with roughly 8 to 12-fold.
- Researchers describe four APOE4-linked processes in the brain: lower BDNF signalling, mitochondrial dysfunction, neuroinflammation and impaired lipid transport to synapses.
- Cerebrolysin has the most human trial data of the seven: a 2015 meta-analysis of six randomised trials in mild-to-moderate Alzheimer's disease reported modest cognitive benefit. It is not FDA-approved.
- One study (PMC4926802) reported larger BDNF rises in APOE4 carriers given Cerebrolysin with donepezil. It has not been replicated, and no study has shown that APOE4 status changes clinical response to any peptide.
- Most of the evidence for the other six is preclinical or from small trials. None is approved for preventing or treating Alzheimer's disease.
APOE4 is the strongest common genetic risk factor for late-onset Alzheimer's disease. It is also a magnet for supplement and peptide marketing, much of which claims more than any study has shown. This page goes through the seven peptides that come up most in APOE4 discussions and reports what the research on each one measured, in whom, and what it did not test. None of them is approved for preventing or treating Alzheimer's disease, and nothing here is a recommendation to use any of them.
Roughly one in four people carry at least one copy of APOE4. Among people who develop late-onset Alzheimer's disease, around 65 percent carry it.
APOE is a protein that carries cholesterol and other fats around the brain. The APOE4 version does that job less well. Researchers link it to four processes: lower BDNF signalling, mitochondrial strain in neurons, overactive immune cells in the brain (microglia), and poorer delivery of fats to synapses. The compounds below were studied against one or more of those processes.
Four processes researchers link to APOE4
What the research on each compound shows, strongest evidence first
Evidence grades below describe what research exists: A means multiple human randomised trials; D means preclinical only. The order is by strength of evidence, not a recommendation.
1. Cerebrolysin
Evidence: A, in Alzheimer's disease generally. Cerebrolysin is a mixture of peptides and neurotrophic factors purified from brain tissue. A 2015 meta-analysis in Dementia and Geriatric Cognitive Disorders pooled six randomised controlled trials in mild-to-moderate Alzheimer's disease and reported a significant cognitive effect at four weeks (standardised mean difference 0.40) and global clinical benefit at six months (odds ratio 4.98). The effect size was modest.
For APOE4 specifically, one study (PMC4926802) of Cerebrolysin combined with donepezil reported larger BDNF rises in APOE4 carriers than in APOE3/E3 carriers. That is a single study of a blood marker, not a replicated clinical outcome. It does not show that APOE4 carriers benefit more clinically.
APOE4 carriers receiving Cerebrolysin combined with donepezil showed higher BDNF elevation compared to APOE3/E3 carriers receiving the same protocol.
Clinical study, PubMed Central PMC4926802
Cerebrolysin is approved in a number of countries for cognitive disorders. It is not FDA-approved. More is on the Cerebrolysin reference page.
2. SS-31 (elamipretide)
Evidence: B for safety, none for the brain. SS-31 binds cardiolipin in the inner mitochondrial membrane. It received FDA approval in September 2025 for Barth syndrome, a rare inherited disorder of cardiolipin. That program generated long-term human safety data in that population. No study has tested SS-31 in APOE4 carriers or in Alzheimer's disease. The SS-31 analysis covers the trials in detail.
3. Semax
Evidence: B-C. Semax is a synthetic analogue of ACTH(4-10), registered in Russia since 1994 for cerebrovascular conditions. A 2025 study in Acta Naturae reported cognitive effects in a preclinical Alzheimer's model, through BDNF/TrkB signalling and reduced oxidative stress. Russian clinical programs report rises in BDNF. There is no randomised trial of Semax in Alzheimer's disease or in APOE4 carriers. It is not FDA-approved. The Semax vs Selank comparison covers both compounds.
4. Selank
Evidence: C. Selank is a synthetic analogue of the immune peptide tuftsin, registered in Russia for anxiety. A 2020 review in Current Reviews in Clinical and Experimental Pharmacology described reductions in inflammatory signals (IL-6, TNF-alpha, IL-1beta) in preclinical stress models. No study has tested it in Alzheimer's disease or APOE4 carriers. It is not FDA-approved.
5. NAP (davunetide)
Evidence: C. NAP is an eight-amino-acid fragment of activity-dependent neuroprotective protein (ADNP). A 2025 review in Advanced Drug Delivery Reviews (PMID 40185278) describes how it stabilises microtubules and reduces tau hyperphosphorylation in laboratory work. A 2024 study in Translational Psychiatry of intranasal davunetide in amnestic mild cognitive impairment reported sex-dependent cognitive effects. An earlier late-stage trial of davunetide in progressive supranuclear palsy, a tau disorder, did not meet its endpoints. No APOE4 subgroup data has been published.
6. Humanin and HNG
Evidence: D. Humanin is a small peptide encoded in mitochondrial DNA. Hashimoto and colleagues identified it in 2001 (Science) as protecting neurons in cell models from death triggered by familial Alzheimer's disease genes. Circulating humanin has been reported to be lower in APOE4 carriers. HNG is a modified, more potent form used in laboratory work. There are no human trials.
7. BPC-157
Evidence: D for the brain. BPC-157 comes up in APOE4 discussions for cerebrovascular reasons. A 2021 review in Neural Regeneration Research (PMC8504390) describes neuroprotective effects in preclinical models of brain injury and ischaemia. There is no human brain study. BPC-157 is not FDA-approved and is not lawfully compoundable in the US.
Evidence side by side
| Compound | Human data | APOE4-specific data | US status (2026) |
|---|---|---|---|
| Cerebrolysin | 6 RCTs in Alzheimer's disease (2015 meta-analysis) | One BDNF marker study, not replicated | Not FDA-approved |
| SS-31 | Barth syndrome trials | None | FDA-approved for Barth syndrome only |
| Semax | Russian clinical programs | None | Not FDA-approved |
| Selank | Russian clinical programs (anxiety) | None | Not FDA-approved |
| NAP (davunetide) | MCI study (2024); failed late-stage PSP trial | None | Not FDA-approved |
| Humanin / HNG | None | Lower levels reported in carriers | Not FDA-approved |
| BPC-157 | No brain studies | None | Not FDA-approved; not lawfully compoundable |
One copy of APOE4 is associated with a 3 to 4 times higher risk of late-onset Alzheimer's disease compared with APOE3/E3; two copies with roughly 8 to 12 times. Risk is not destiny: many carriers never develop the disease.
The approved drugs that do name APOE4
The FDA approved lecanemab (Leqembi) in July 2023 and donanemab (Kisunla) in July 2024, both anti-amyloid antibodies for early Alzheimer's disease. Both labels carry boxed warnings about amyloid-related imaging abnormalities (ARIA) that name APOE4 carriers, and especially homozygotes, as higher risk, and recommend APOE4 testing before starting treatment. That is where APOE4 status currently changes a real prescribing decision, and it is a decision for a neurologist.
What the genetics can and cannot tell you
Knowing your APOE genotype tells you about risk, which is a statement about populations. It does not tell you that any peptide would protect you, and no study has shown that APOE4 status changes clinical response to any of the seven compounds above. If you learn you carry APOE4, the people to talk to are your doctor or a genetic counsellor; the APOE4 result itself can be confirmed by a clinical test.
The short version: of the seven peptides discussed for APOE4, only Cerebrolysin has randomised trials in Alzheimer's disease, with modest effects, and the APOE4-specific evidence is a single unreplicated marker study. The rest rely on preclinical work or trials in other conditions. If you want to see your APOE, BDNF and related variants and what the research on each does and does not show, your report reads the DNA file you already have against the published research.
Frequently asked questions
Is any peptide proven to protect APOE4 carriers from Alzheimer's disease?
No. No peptide is approved for preventing or treating Alzheimer's disease, and no study has shown that APOE4 status changes clinical response to any peptide. Cerebrolysin has randomised trials in mild-to-moderate Alzheimer's disease with modest effects, but it is not FDA-approved.
Does Cerebrolysin work better in APOE4 carriers?
One study (PMC4926802) reported larger BDNF rises in APOE4 carriers given Cerebrolysin with donepezil than in APOE3/E3 carriers. That is a single, unreplicated study of a blood marker. It does not show a larger clinical benefit.
What is Semax and what has it been studied for?
Semax is a synthetic analogue of ACTH(4-10), registered in Russia since 1994 for cerebrovascular conditions. A 2025 preclinical study reported cognitive effects in an Alzheimer's model through BDNF signalling. There is no randomised trial in Alzheimer's disease or APOE4 carriers, and it is not FDA-approved.
Which approved drugs mention APOE4?
The anti-amyloid antibodies lecanemab (Leqembi) and donanemab (Kisunla) carry boxed warnings about amyloid-related imaging abnormalities (ARIA) that name APOE4 carriers, especially homozygotes, as higher risk. The labels recommend APOE4 testing before treatment. That decision belongs to a neurologist.
What is humanin?
Humanin is a small peptide encoded in mitochondrial DNA. A 2001 Science paper (Hashimoto et al.) found it protected neurons in cell models from death triggered by familial Alzheimer's disease genes. Lower circulating levels have been reported in APOE4 carriers. There are no human trials, and it is not an approved drug.
How much does APOE4 raise Alzheimer's risk?
One copy is associated with a 3 to 4 times higher risk of late-onset Alzheimer's disease compared with APOE3/E3; two copies with roughly 8 to 12 times. Many carriers never develop the disease, and a consumer chip result should be confirmed by a clinical test before any medical decision.
This article is for informational and educational purposes only. It reports what published research found; it is not medical advice, it does not diagnose, treat, cure or prevent any disease, and nothing in it is a dose or a recommendation to use any compound. Talk to a licensed clinician about any medical decision.