Researchers from the University of the Free State in South Africa have published a comprehensive review of the research on cannabinoids and Alzheimer’s disease, arguing that compounds from the Cannabis sativa plant could open a new angle on treating the disease. The work appeared in Frontiers in Pharmacology and explains how compounds from marijuana might help the brain resist one of the most dangerous diseases of our time.
A fresh laboratory study, published at the end of July 2026, adds a different piece to the puzzle. Israeli chemists Asif Shitrit, Ali Khattib and Soliman Khatib tested whole cannabis flower extracts against three enzymes linked to Alzheimer’s at the same time, and reported the results in the Journal of Cannabis Research. Several extracts blocked one of those enzymes about four times more effectively than galantamine, a drug already approved for Alzheimer’s. The catch is that all of it happened in test tubes and in rat cells, not in people.
Key Takeaways on Cannabinoids and Alzheimer’s
Cannabinoids act on several things that go wrong in an Alzheimer’s brain at once: inflammation, oxidative damage, the build-up of toxic proteins and the cell’s own waste-disposal system, known as autophagy. Most of that evidence comes from cell cultures and mice. The new flower extract study fits the same pattern: four extracts from different cannabis chemovars inhibited acetylcholinesterase, butyrylcholinesterase and beta-secretase (BACE1), and the most active fractions were rich in the acidic forms of cannabinoids, CBDA and CBGA, rather than in THC and CBD. The strongest human signal so far is a small Brazilian trial in which a very low daily dose of THC and CBD was linked to better scores on a standard memory and thinking test after 26 weeks, with 28 patients completing it. That is encouraging, but it is a single small trial. No cannabinoid has been shown to slow or reverse Alzheimer’s in people, and none is approved to treat it.
Why Current Medications Don’t Work
Alzheimer’s disease is the leading cause of dementia worldwide. Toxic proteins (amyloid-beta and tau protein) accumulate in the brain, killing nerve cells. People gradually lose memory, thinking ability, and the capacity to care for themselves.
For years there were only two types of medications: acetylcholinesterase inhibitors (donepezil, galantamine) and memantine. They provide some symptom relief but don’t stop the disease. The problem is they only address consequences, not the underlying causes of brain destruction. A newer class, anti-amyloid antibodies such as lecanemab, has since been approved in some regions. Even so, the authors of the new Journal of Cannabis Research paper sum up the situation bluntly: current treatments ease symptoms but fail to halt the disease, and they often cause side effects.
Moreover, most attempts to create drugs targeting amyloid plaques have failed in clinical trials. This has pushed scientists to explore alternative approaches, and cannabinoids have emerged as a promising option worth testing.
How Cannabinoids May Protect the Brain
The Cannabis sativa plant contains over 100 active compounds, but two have been studied most extensively: THC (tetrahydrocannabinol) and CBD (cannabidiol). They share the same chemical formula but have different structures, and therefore work differently.
THC is what makes marijuana psychoactive. It binds to CB1 receptors in the brain. CBD doesn’t cause euphoria and works through other mechanisms (our cannabidiol guide explains these mechanisms in detail). Both substances have shown the ability to protect nerve cells from destruction in laboratory and animal studies.
What they actually do:
- Reduce brain inflammation
- Decrease oxidative stress (cell damage from free radicals)
- Help clear toxic proteins
- Restore mitochondrial function (cellular power plants)
Autophagy: The Brain’s Self-Cleaning System
The key mechanism is autophagy, the process by which cells digest their own “garbage.” Imagine cells have a waste disposal system: it collects damaged proteins and old organelles, packages them in a special vesicle (autophagosome), and sends them for recycling.
In Alzheimer’s patients, this system breaks down. Waste isn’t cleared but accumulates, and that is how those amyloid plaques form. Research shows that cannabinoids activate autophagy, helping the brain eliminate toxic proteins.
CBD works through the PI3K/Akt/mTOR pathway, an important system regulating cellular metabolism. It blocks mTOR (a protein that inhibits autophagy), and the self-cleaning process activates. Our separate look at CBD research in Alzheimer’s explains why these mechanisms are still a laboratory finding, and where human trials stand today.
Cannabis Flower Extracts and Alzheimer’s Enzymes: What the New Lab Study Adds
The autophagy story is about clearing waste. The new Israeli study looks at a different problem: the enzymes that drive Alzheimer’s in the first place. Think of them as three pairs of scissors. Two of them, acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), cut up acetylcholine, a chemical messenger the brain needs for learning and memory. Blocking them leaves more of the messenger available, which is exactly how donepezil and galantamine work. BChE matters more as the disease advances, because its activity stays high or even rises while AChE falls. The third, beta-secretase or BACE1, makes the first cut in a larger protein that eventually yields amyloid beta, the material of plaques.
The team took flowers from four cannabis chemovars, labelled C1 to C4, heated the ground material to 100 °C for one hour to partly convert acidic cannabinoids into their neutral forms, and extracted them with ethanol. C1, C3 and C4 carried THC and CBD in roughly balanced amounts, with THC-to-CBD ratios of about 0.8 to 1.3. C2 was a CBD-dominant plant with a ratio of about 0.05. C3 and C4 also contained CBGA and CBNA, which were not detected in the other two.
All four extracts inhibited AChE at levels comparable to galantamine. The real difference showed up with BChE. Scientists measure strength with a number called IC50, the concentration needed to cut an enzyme’s activity in half; the lower the number, the stronger the effect. Extracts C2, C3 and C4 reached it at 9.9, 7.19 and 9.06 micrograms per millilitre, against 35.93 for galantamine. C1, at 35.81, did no better than the drug. All four extracts also inhibited BACE1, which the authors describe as among the first evidence of this kind for cannabis extracts: by 7.9 to 21.3 percent at 10 micrograms per millilitre, 26.6 to 51 percent at 25, and 47.9 to 65.3 percent at 50.
When the researchers split extract C3 into six fractions, the most active ones were not the THC and CBD fractions. At 1 microgram per millilitre, fractions f1 to f4 blocked AChE by about 40 percent, similar to galantamine, while f5 and f6, which contained CBDA and CBGA, blocked it by 85 percent. Against BChE, fractions f1, f2 and f5 reached half-inhibition at about 1.4, 3.5 and 2.5 micrograms per millilitre, compared with 7.2 for the whole C3 extract. Only f5 and f6 cut BACE1 activity by about half at 25 micrograms per millilitre. That fits earlier work from Italy, where CBDA and CBGA inhibited the same three enzymes and improved memory problems caused by amyloid beta in mice.
Extract C3 also helped rat PC12 cells, a common stand-in for neurons, survive exposure to amyloid beta and to hydrogen peroxide, with the best protection at 10 micrograms per millilitre. Fraction f5 raised survival too, but not significantly. One detail stands out: PC12 cells generally lack CB1 and CB2 receptors, so whatever protected them did not go through the classic cannabinoid docking points. For the full chemistry, our step-by-step report on the cannabis extract enzyme study walks through every experiment.
What the Enzyme Study Cannot Tell Us
“Four times stronger than galantamine” needs an asterisk. Galantamine is one purified molecule; each extract is a mixture of dozens of compounds, compared by total weight. The authors used different concentration ranges for different tests, so the targets can’t be ranked against each other directly. The BACE1 estimates rest on only three concentration points, and the team says plainly that they should be read with caution. Purified CBDA and CBGA were not tested, so their role is an inference from chemistry, not a demonstration. Nothing is known yet about whether these compounds reach the brain, how the body processes them, or how extracts containing THC would interact with the many medicines elderly patients usually take. BACE1 has also been an unforgiving target: several drugs that blocked it well in the lab failed in human trials.
Breakdown of Key Experiments
To see the real-world impact, researchers analyzed data from petri dishes to human trials. Here is a breakdown of the findings:
| Study Type | Compound Used | Key Findings |
|---|---|---|
| In Vitro (Lab Cells) | CBD | Reduced toxic protein levels; protected neurons from cell death induced by amyloid-beta. |
| In Vivo (Mice) | CBD | Improved performance in spatial memory tests; reduced neuroinflammation; slowed plaque accumulation. |
| In Vitro (Lab Tests) | THC | Blocked the enzyme acetylcholinesterase (similar to standard drugs) and prevented amyloid clumping. |
| In Vitro (Enzymes, Rat Cells) | Whole flower extracts | Inhibited AChE, BChE and BACE1; CBDA- and CBGA-rich fractions most active; protected PC12 cells. |
| Clinical (Humans) | THC + CBD | 2025 Trial: Higher memory-test scores than placebo after 26 weeks; no difference in adverse events. |
What Experiments Showed
In laboratory conditions:
- CBD reduced levels of toxic proteins in brain cells
- Protected neurons from amyloid-beta-induced death
- Restored mitochondrial function
- Reduced inflammation
In animal models:
- Mice with Alzheimer’s given CBD performed better on memory tests
- Brain inflammation decreased
- Amyloid plaque accumulation slowed
- Spatial learning improved
THC showed similar effects in laboratory tests: it blocks the enzyme acetylcholinesterase (like conventional Alzheimer’s drugs) and prevents amyloid from clumping into plaques.
Combination Works Better
One of the most interesting findings from mouse studies: combining THC and CBD can work more effectively than either substance alone. In 2025, a clinical trial concluded in which people aged 60 to 80 with Alzheimer’s dementia received low doses of THC+CBD (0.35 mg THC + 0.245 mg CBD daily). The results were published in the Journal of Alzheimer’s Disease.
Result: after 26 weeks, the cannabis group scored significantly higher on the Mini-Mental State Exam, a standard memory and thinking test, than the placebo group. Adverse events did not differ between the groups, but the trial’s secondary measures did not change either. The Frontiers review calls it the first robust clinical evidence for THC and CBD combinations. With 28 patients completing the trial, it is best read as a strong signal that justifies bigger studies, not as proof that cannabinoids treat Alzheimer’s.
Mouse experiments showed the combination reduces glutamate levels in the hippocampus (the brain region responsible for memory) and decreases neuronal hyperactivity, one factor in brain damage.
Not Just THC and CBD
Scientists are also studying other cannabinoids. For example, cannabigerol (CBG) is often called the “mother cannabinoid,” because its acid form, CBGA, is the starting material from which the plant builds THC, CBD and the rest. It shows promising neuroprotective properties and may enhance the effects of CBD and THC. The new flower extract study points the same way from another direction: CBGA, together with CBDA, sat in the fractions that blocked Alzheimer’s enzymes most strongly.
That matters for a practical reason. Acidic cannabinoids turn into their neutral forms when heated, which is what happens with smoking, vaping and baking. If they prove important, the right preparation will look nothing like a joint.
The Cannabis sativa plant contains over 500 bioactive compounds, many still unstudied. Perhaps a combination of different cannabinoids will prove most effective.
What’s Next
Several clinical studies of cannabinoids for Alzheimer’s are currently underway. Scientists are testing:
- Optimal doses and THC:CBD ratios
- Long-term safety
- Effectiveness at different disease stages
- Interactions with conventional medications
Researchers’ main conclusion: cannabinoids work across multiple pathways simultaneously, activating autophagy, reducing inflammation, protecting mitochondria, and helping clear toxic proteins. That makes them interesting research candidates, but the gap between laboratory and clinic remains wide. Our review of 107 studies on CBD in neurology shows how often promising lab results in Alzheimer’s and Parkinson’s have failed to translate into clear benefit for patients.
It’s important to understand we’re talking about medical use of precise doses, not smoking marijuana. Treatment research requires standardized extracts with known active compound content.
Frequently Asked Questions
No. There is no cure, and no human study has shown that CBD or any other cannabinoid slows or reverses the disease. Laboratory and animal research suggests CBD and THC can reduce brain inflammation, activate the brain’s waste-removal systems and block enzymes linked to the disease, and one small trial found better memory-test scores. Those are reasons for more research, not evidence of a cure.
The 2025 clinical trial used very low doses: 0.35 mg of THC and 0.245 mg of CBD a day, far below typical consumer products. At that level, adverse events were no more common than with placebo over 26 weeks. There is no established dose of CBD or THC for Alzheimer’s, and the trial was small. THC and CBD can also interact with other medications, so medical supervision is essential.
Standard drugs like donepezil and galantamine boost acetylcholine by blocking one enzyme. In laboratory studies, cannabinoids appear to act on several problems at once: they reduce oxidative stress and inflammation, help clear toxic proteins via autophagy and, according to the new flower extract study, block both cholinesterase enzymes and BACE1. None of this has yet been confirmed in patients.
It depends on the state. No cannabinoid product is approved by the FDA to treat Alzheimer’s disease. Since April 28, 2026, marijuana covered by a state medical marijuana license and FDA-approved drug products containing marijuana sit in Schedule III of the Controlled Substances Act, while a broader federal rescheduling proceeding is still under way. Whether a patient with Alzheimer’s or dementia can obtain medical cannabis is decided by state programs, which set their own qualifying conditions and rules. A physician who knows the state’s program is the right starting point.
Disclaimer
This article summarizes published preclinical and clinical research for informational purposes only. It is not medical or legal advice and does not recommend using cannabis, cannabis extracts or any cannabinoid product to prevent or treat Alzheimer’s disease or any other condition. The enzyme and cell findings described here come from laboratory experiments and have not been tested in humans. No cannabinoid is approved as a treatment for Alzheimer’s disease. Cannabis products can interact with prescription medicines, which is a particular concern for older adults. Laws on cannabis differ between countries and US states. Anyone considering a change in treatment should speak with a qualified healthcare professional.
