Cannabis-Based Spray Shows Early Safety Promise for Alzheimer’s Agitation, But Proof of Efficacy Still Awaited

A small UK feasibility trial found nabiximols (Sativex) was well-tolerated with zero adverse reactions in nursing home residents with Alzheimer’s dementia and agitation, though the study was explicitly not designed or powered to test whether the intervention actually reduces agitation symptoms.

Why This Matters

Agitation affects up to 70% of people living with Alzheimer’s dementia and remains one of the most distressing symptoms for patients, families, and care staff alike. Current pharmacological options, primarily antipsychotics, carry serious safety risks including increased mortality in this population, leaving a substantial unmet clinical need. Cannabinoid-based therapies have plausible mechanistic pathways through the endocannabinoid system’s role in mood regulation and neuroinflammation, making rigorous early-phase testing both timely and necessary. This feasibility trial represents a critical preliminary step in determining whether a larger confirmatory study is warranted.

Clinical Summary

The STAND trial, published in Age and Ageing in 2025, is a double-blind, placebo-controlled, randomised feasibility trial examining whether nabiximols (Sativex), an oromucosal spray containing a standardized combination of THC and CBD, can be practically and safely delivered to nursing home residents with probable Alzheimer’s dementia and clinically significant agitation. The mechanistic rationale centers on the endocannabinoid system’s involvement in modulating neuroinflammation, stress responses, and emotional regulation, pathways increasingly implicated in the neuropsychiatric symptoms of dementia. The intervention was up-titrated over four weeks to a target dose of 10.8 mg THC and 10 mg CBD per day, administered by care home nursing staff.

The trial randomised 29 of a planned 60 participants between November 2021 and July 2022, with the recruitment shortfall attributed primarily to COVID-19 pandemic restrictions affecting UK care homes. Despite this, the study met its key feasibility benchmarks: 100% participant retention at four weeks, 100% medication adherence across both arms, and zero reported adverse reactions to the intervention. Preliminary effect-size estimates on the Cohen-Mansfield Agitation Inventory (CMAI) were calculated but are explicitly exploratory, as the trial was neither designed nor powered to detect treatment effects. The authors appropriately conclude that these results support progression to a larger phase 3 confirmatory trial but do not themselves constitute evidence of efficacy.

Dr. Caplan’s Take

What this trial gets right is asking the right question first. Before you can test whether something works, you need to know whether you can actually deliver it to a vulnerable population in a real-world setting. For a cannabis-based medicine in advanced dementia patients living in care homes, that is not a trivial question. Families ask me regularly about cannabis for agitation in their loved ones with Alzheimer’s, and what they need to hear is honest: we now have early evidence that it can be given safely in this setting, but we genuinely do not yet know if it helps.

In practice, I do not recommend nabiximols or other cannabinoid products for dementia-related agitation based on current evidence. What I do is ensure that non-pharmacological approaches, including environmental modifications, structured activities, and caregiver training, are fully optimized before any pharmacological intervention is considered. When medication is necessary, I discuss the limited options and their risk profiles transparently. This trial is a reason for cautious interest, not a reason to change prescribing.

Clinical Perspective

This study sits squarely in the early feasibility phase of the research arc for cannabinoid-based interventions in dementia-related agitation. It confirms that oromucosal delivery of nabiximols is logistically manageable in the care home environment and that the compound appears well-tolerated at the doses tested. However, it resolves nothing about clinical efficacy, dose-response relationships, or how nabiximols might compare to existing treatments. The favorable safety signal is encouraging but must be interpreted within the context of a very small sample followed for only four weeks of active treatment, a duration insufficient to capture longer-term safety concerns or delayed adverse effects.

Clinicians should be aware that nabiximols contains THC, which in older adults with cognitive impairment raises legitimate concerns about sedation, falls, delirium, and cardiovascular effects, none of which can be ruled out as risks based on a 29-person feasibility study. Drug interactions with commonly prescribed dementia medications, including cholinesterase inhibitors and psychotropics, have not been systematically evaluated in this population. The most actionable recommendation from this trial is not clinical but conversational: when patients or families raise cannabis-based treatments for dementia agitation, clinicians can acknowledge that structured research is underway and that safety signals are cautiously positive, while clearly communicating that efficacy remains unestablished.

Study at a Glance

Study Type

Randomised, double-blind, placebo-controlled feasibility trial (phase 2)

Population

Adults aged 55 to 95 with probable Alzheimer’s dementia and clinically significant agitation, residing in UK nursing homes

Intervention

Nabiximols (Sativex) oromucosal spray, up-titrated to 4 sprays per day (10.8 mg THC and 10 mg CBD daily) over 4 weeks

Comparator

Matched placebo oromucosal spray

Primary Outcomes

Four prespecified feasibility thresholds: recruitment rate, 75% or greater follow-up at 4 weeks, 80% or greater adherence, and Cohen’s d of 0.3 or greater on CMAI at week 4

Sample Size

29 randomised (target was 60)

Journal

Age and Ageing

Year

2025

Registration

ISRCTN71635621

Funding Source

Not specified in available text

What Kind of Evidence Is This

This is a registered phase 2 feasibility randomised controlled trial, not an efficacy trial. In the evidence hierarchy, feasibility studies sit below confirmatory RCTs and are designed to evaluate whether a full-scale trial is practical and safe to conduct, not whether the intervention works. The single most important inference constraint is that no conclusions about therapeutic efficacy are possible from this design, regardless of what direction any exploratory clinical outcome signals may point.

How This Fits With the Broader Literature

Research on cannabinoid-based therapies for neuropsychiatric symptoms of dementia remains in early stages, with a handful of small pilot studies and case series producing mixed results. A 2019 crossover pilot by Shelef and colleagues examined dronabinol in agitated dementia patients and reported modest improvements, though that study was similarly limited by small sample size and short duration. Systematic reviews of cannabinoids for dementia-related behavioral symptoms have consistently concluded that evidence is insufficient to support clinical recommendations. The STAND trial contributes by establishing a rigorous methodological framework and demonstrating practical delivery in the care home environment, which previous studies had not systematically addressed. It extends the literature primarily on feasibility rather than efficacy grounds.

Common Misreadings

The most likely overinterpretation is treating the zero adverse reactions finding as proof that nabiximols is broadly safe for people with advanced dementia. A sample of 29 participants followed for four weeks of active treatment is far too small and too brief to characterize the true safety profile of a THC-containing product in a cognitively impaired elderly population. Similarly, any preliminary effect-size estimates from this trial should not be cited as evidence that the drug reduces agitation. The study was not powered for efficacy, and treating exploratory signals as confirmatory findings would misrepresent the strength of the evidence.

Bottom Line

The STAND trial demonstrates that delivering nabiximols to care home residents with Alzheimer’s dementia and agitation is logistically feasible and well-tolerated in the short term. It does not establish whether the drug reduces agitation. These findings provide a credible foundation for designing an adequately powered phase 3 confirmatory trial, which remains the essential next step before any clinical recommendations can be considered.

References

1. Marchant NL, Sherif MA, Goss A, et al. Nabiximols for agitation in Alzheimer’s dementia (STAND): a randomised, double-blind, placebo-controlled feasibility trial. Age and Ageing. 2025. ISRCTN71635621.
2. Shelef A, Barak Y, Berger U, et al. Safety and efficacy of medical cannabis oil for behavioral and psychological symptoms of dementia: an open label, add-on, pilot study. Journal of Alzheimer’s Disease. 2016;51(1):15-19.
3. Liu CS, Chau SA, Bhutto A, et al. Cannabinoids for the treatment of agitation and aggression in Alzheimer’s disease. CNS Drugs. 2015;29(8):615-623.