Can Cannabis Defend Against Neurodegenerative Diseases?

Cannabinoids are being studied for symptoms associated with neurological diseases, but cannabis has not been established as a way to prevent, reverse or slow neurodegeneration in people. Evidence for an individual symptom, such as agitation or spasticity, should not be confused with evidence that a treatment protects the brain or changes the underlying disease.

Symptoms and Disease Progression Are Different Questions

Alzheimer’s disease, Parkinson’s disease, Huntington’s disease and amyotrophic lateral sclerosis (ALS) have different causes, biological processes and treatment needs. Multiple sclerosis (MS), which is also frequently discussed in cannabinoid research, is an immune-mediated disease involving damage to myelin and nerve cells. Findings in one condition cannot simply be transferred to another.

A treatment might reduce a particular symptom without slowing nerve-cell loss. Demonstrating disease modification requires human studies designed to measure progression over an appropriate period. Changes in a laboratory marker, a patient’s sleep or a caregiver’s impression do not by themselves answer that question.

Alzheimer’s Disease: A Signal for Agitation, Not a Cure

A randomized trial published online in 2025 and in a 2026 journal issue tested dronabinol against placebo for three weeks in 75 people with agitation associated with Alzheimer’s disease. Dronabinol is a pharmaceutical form of THC, not an interchangeable substitute for every cannabis product.

The trial found a statistically significant benefit on the Pittsburgh Agitation Scale. The difference on its other co-primary agitation measure was not statistically significant, and secondary outcomes, including cognition and activities of daily living, did not differ between groups. Sleepiness was more frequent with dronabinol. The short trial supports further investigation of symptom management; it does not demonstrate reversal of dementia or prevention of Alzheimer’s disease.

An earlier observational study of 19 people with severe dementia reported improvements during treatment with a combination of THC and CBD. It was not a CBD-only study: average daily exposure was approximately 12.4 mg THC and 24.8 mg CBD. Without a randomized comparison group, its findings cannot establish that the treatment caused the changes or justify using those amounts as a home dosing guide.

Parkinson’s Disease: Findings From Controlled Trials

A 2024 randomized trial involving 61 participants tested a relatively high-CBD, low-THC oral extract for two weeks. Motor scores improved in both groups, but the difference between the cannabinoid and placebo groups was not statistically significant. Improvement from a person’s own baseline is not sufficient evidence of benefit when the comparison group also improves.

A 2026 phase II randomized trial enrolled 101 people with Parkinson’s disease and chronic pain; 87 completed the nine-week study. The tested CBD/THC oil did not improve the primary pain outcome or other nonmotor measures compared with placebo. No serious adverse events were reported, but that short-term result does not establish long-term safety or address every possible formulation.

These trials do not support confident claims that cannabis broadly controls Parkinson’s symptoms. They also did not demonstrate that cannabinoids slow the disease. Patients should continue established neurological care rather than replace prescribed treatment on the strength of anecdotal reports.

Multiple Sclerosis: A More Specific Symptom Application

Evidence is more developed for certain cannabinoid preparations used for MS-related spasticity. NCCIH’s evidence summary reports that nabiximols, a standardized THC/CBD mouth spray, probably reduces spasticity severity in the short term compared with placebo. Adverse effects and uncertainty about other outcomes remain relevant.

NICE guidance provides for a supervised four-week trial of THC/CBD spray in eligible adults with moderate to severe MS spasticity when other pharmacological treatments have not been effective. Continued treatment depends on a meaningful measured response. This is a product-specific symptom recommendation, not evidence that cannabis prevents MS progression or that an unstandardized extract has the same effects.

CBN and Neuroprotection: What the Laboratory Work Shows

The CBN research highlighted in earlier coverage was preclinical. In work reported by the Salk Institute in 2024, scientists designed four CBN-inspired compounds and studied protection against oxidative cell death in nerve-cell cultures. One compound, CP1, also showed promising results in a fruit-fly model of traumatic brain injury.

These chemically designed analogs were research compounds, not four types of retail CBN oil. Cell survival and a fruit-fly injury model do not establish an effective treatment for Alzheimer’s disease, Parkinson’s disease or ALS in humans. The findings are a starting point for drug development, with further testing needed before clinical conclusions.

The Endocannabinoid System Does Not Prove a Deficiency

The body has cannabinoid signaling pathways that researchers investigate in neurological disease. Their existence does not show that neurodegeneration is caused by an endocannabinoid deficiency, or that adding THC or CBD restores a missing balance. THC and CBD also have different pharmacology; they should not be described as acting identically at cannabinoid receptors.

Risks and Questions for the Care Team

People with neurological illness may already have difficulties with alertness, balance or memory. Drowsiness, dizziness or confusion can therefore have substantial practical consequences. NCCIH also cautions about cannabinoid adverse effects, including CBD-related drug interactions and liver problems. “Non-intoxicating” does not mean free of risk.

If a clinician considers a cannabinoid for symptom management, define the symptom, product, monitoring plan and stopping criteria. Review all medicines with the prescriber or pharmacist. In dementia care, consent and caregiver involvement also need attention. Eligibility under a medical-cannabis program is not proof that a product changes the disease.

The useful question is whether a particular preparation offers a worthwhile, measurable benefit for a particular symptom with acceptable risks. The current evidence does not support presenting cannabis as a proven neuroprotective treatment.