This is the highest-grade neurological evidence that exists for palmitoylethanolamide, and it is also the study most likely to be misrepresented.
A Phase 2 randomized, double-blind, placebo-controlled trial found that PEA combined with luteolin slowed decline on the primary clinical measure in frontotemporal dementia over 24 weeks. That is a real finding in a disease with no approved pharmacological treatment.
It is also a finding about frontotemporal dementia specifically. Not Alzheimer’s disease, not age-related memory changes, and certainly not brain fog in an otherwise healthy person. Most secondary outcomes in the trial, including the standard cognitive screening tests, showed no treatment effect at all.
Key Takeaways
- A Phase 2, monocentric, randomized, double-blind, placebo-controlled trial randomized 48 patients with probable frontotemporal dementia to co-ultramicronized PEA plus luteoline 700 mg + 70 mg twice daily, or placebo, for 24 weeks [1].
- On the primary outcome, mean decline was 0.53 points in the treatment group and 1.39 in the placebo group, a difference of 0.86 (95% CI 0.28 to 1.45, P = 0.005) [1].
- Activities of daily living and a language screening measure also favoured treatment, but no effect was found on the other secondary outcomes, including MMSE and the Addenbrooke Cognitive Examination Revised [1].
- The trial was single-centre with 48 patients, 45 of whom completed [1]. The authors describe the efficacy as promising and call for further investigation [1].
- Alzheimer’s evidence for PEA is a different and much weaker tier: the leading study is a rat model of amyloid injection [2], and reviews of neurodegenerative applications are explicit that most work runs from rodents rather than humans [3].
What Frontotemporal Dementia Is
Frontotemporal dementia is a neurodegenerative disorder for which no pharmacological treatments have been approved, and neuroinflammation plays a central role in driving the mechanisms underlying it [1]. It typically strikes younger than Alzheimer’s disease, and it tends to present with changes in behaviour, personality or language rather than with memory loss first.
The absence of any approved treatment is the reason a supplement-grade compound got a properly designed Phase 2 trial here. When nothing is approved, the bar for testing something well tolerated is lower, and the need for a result is higher.
Co-ultramicronized PEA combined with luteoline had emerged as a candidate in neurodegenerative disorders pathogenically related to frontotemporal dementia, on the basis of demonstrated anti-inflammatory and neuroprotective properties [1]. Reviews of PEA in neuroinflammation frame the compound as a way of keeping inflammatory signalling within physiological boundaries rather than suppressing it outright [4].
How the Trial Was Built
The study was a Phase 2, monocentric, randomized, double-blind, placebo-controlled trial evaluating safety and efficacy in frontotemporal dementia patients [1]. Forty-eight patients with a diagnosis of probable frontotemporal dementia were randomly assigned one to one to receive co-ultramicronized PEA combined with luteoline as an oral suspension at 700 mg plus 70 mg twice daily, with 25 in the treatment arm and 23 on placebo, for 24 weeks [1]. Mean age was 63.2 years and 47.9% were female [1].
Research note: Before the larger placebo-controlled trial, a review reported a study from its authors’ own group testing co-ultraPEALut (PEA plus luteolin) for four weeks in seventeen patients with frontotemporal dementia [7].
The trial was funded by the Italian Ministry of Health and partially by Epitech Group, the supplement’s maker [1].
The primary efficacy outcome was the change at 24 weeks in the Clinical Dementia Rating Dementia Staging Instrument with the National Alzheimer Coordinating Center frontotemporal lobar degeneration modules, sum of boxes, usually shortened to CDR plus NACC FTLD-SoB [1]. That is the disease-specific staging instrument for this condition, not a generic cognitive test, and choosing it as the primary endpoint is a sign the trial was designed by people who work in this area.
Secondary measures included the Frontal Assessment Battery, Screening for Aphasia in Neurodegeneration, the Alzheimer’s Disease Cooperative Study Activities of Daily Living scale, the Neuropsychiatric Inventory, the Mini-Mental State Examination and the Addenbrooke’s Cognitive Examination Revised [1]. Of the 48 randomized, 45 completed the study, a 93% completion rate [1].
The Primary Result
Patients in the treatment group showed less decline on the primary outcome than patients on placebo [1]. The estimated mean change from week 0 to week 24 was 0.53 in the treatment group, with a 95% confidence interval of 0.12 to 0.94, against 1.39 for placebo with a confidence interval of 0.96 to 1.82 [1]. The estimated mean difference between groups was 0.86, with a confidence interval of 0.28 to 1.45 and P = 0.005 [1].
Over 24 weeks, patients on PEA plus luteolin declined less than those on placebo
In a small phase 2 trial, the PEA plus luteolin combination slowed decline on two measures, but not on the others.
Bar length shows size of decline on each scale; this is PEA plus luteolin, not PEA alone. Source: Assogna et al., Brain communications 2025.
Data table
| Item | Value |
|---|---|
| Primary outcome (CDR plus NACC FTLD), PEA plus luteolin | 0.53 |
| Primary outcome, placebo | 1.39 |
| ADCS-ADL, PEA plus luteolin | 1.80 |
| ADCS-ADL, placebo | 7.39 |
Embed this chart
Free to reuse on your site. Copy the code below; please keep the credit link.
Read the direction carefully. Both groups declined. The treatment group declined less. In a progressive neurodegenerative disease, slowing decline is the realistic goal, and it is what the trial reports. Nobody improved back toward health.
Two secondary measures moved in the same direction. Estimated mean change on the activities of daily living scale was -1.8 for treatment against -7.39 for placebo, and on the aphasia screening measure -3.987 against -10.35 [1]. Function and language, both clinically meaningful domains in this disease.
The Part That Gets Left Out
No effect of treatment was found on the other secondary outcome measures [1].
That includes the Mini-Mental State Examination and the Addenbrooke’s Cognitive Examination Revised, the two tests a general reader is most likely to recognise as cognitive measures, plus the Frontal Assessment Battery and the Neuropsychiatric Inventory [1].
A trial where the disease-specific staging instrument, daily function and language all move while the general cognitive screens do not is a plausible pattern, since those screens are blunt instruments in frontotemporal dementia. It is still a mixed result, and it means the honest description is that one carefully chosen primary outcome and two secondaries favoured treatment while several others did not.
Scale matters too. A single centre, 48 patients, 24 weeks. Phase 2 exists to decide whether a Phase 3 is worth funding, and this one argues yes. The authors themselves conclude that the findings warrant further investigation and offer potential for developing therapeutic strategies [1], which is the language of a promising start rather than a proven treatment.
Why This Is Not an Alzheimer’s Result, and Not a Nootropic One
The temptation to generalise from this trial is strong and should be resisted on the evidence.
For Alzheimer’s disease, the most cited PEA work is preclinical. Systemic PEA was given to adult male rats after intrahippocampal injection of beta amyloid, and it restored biochemical markers related to reactive gliosis, amyloid processing and tau hyperphosphorylation through PPAR-alpha involvement, with a matching reduction in a mild memory deficit in a water maze task [2]. That is a rat model, and its own authors frame it as disclosing therapeutic potential rather than demonstrating clinical benefit [2].
Reviews covering PEA in neurodegenerative disease describe the arc explicitly as running from rodents to humans, which is a candid acknowledgement of where the weight of evidence sits [3].
Human data in Alzheimer’s disease are limited to one patient with amnestic mild cognitive impairment given high-dose PEA plus luteolin for nine months. A preliminary meta-analysis of three human studies in 282 patients with neurocognitive disorders found a pooled MMSE change of 3.80 points, but the 95% confidence interval ran from -0.16 to 7.75, so it included no effect [7][8].
And nothing in this trial speaks to healthy cognition. Every participant had probable frontotemporal dementia [1]. A drug that slows decline in a neurodegenerative disease has no established relationship to focus, memory or mental clarity in someone who does not have that disease. Treating this trial as a nootropic endorsement inverts what it measured.
The Formulation Question Again
As in the smell-loss trials covered elsewhere on this site, the intervention was PEA combined with luteoline in a fixed co-ultramicronized preparation, delivered as an oral suspension at 700 mg plus 70 mg twice daily [1]. Luteolin is not a bystander in the formulation, and the trial has no arm isolating either component.
The delivery form is also specific. An oral suspension of a co-ultramicronized preparation is not necessarily interchangeable with a standard capsule, but no head-to-head trial has compared micronized with unmicronized PEA, and a pharmacology review found no published information on oral bioavailability [5][6]. This site covers micronized versus ultramicronized forms separately for that reason.
What This Means for Families
For a family dealing with frontotemporal dementia, this is a real trial worth raising with the treating neurologist, in a disease where there is otherwise nothing approved to raise. The regimen, duration and outcome measures are all specific enough to discuss concretely.
For anyone else, including people concerned about Alzheimer’s risk or about their own memory, this trial does not apply. The evidence at that end of the spectrum is mostly rodent work and mechanism, plus one patient case report [7][2][3], and the distance between a rat model and a treatment is exactly the distance this Phase 2 trial had to travel for a different disease.
Frequently Asked Questions
Does PEA treat dementia?
A Phase 2 trial in 48 patients found PEA combined with luteoline slowed decline on the primary staging measure in frontotemporal dementia over 24 weeks [1]. It is one single-centre trial in one specific dementia, and several secondary outcomes showed no effect.
What dose was used?
Co-ultramicronized PEA plus luteoline, 700 mg plus 70 mg, twice daily as an oral suspension, for 24 weeks [1].
Does this apply to Alzheimer’s disease?
No. The trial enrolled only patients with probable frontotemporal dementia [1]. The main Alzheimer’s evidence for PEA is a rat model of amyloid injection [2].
Will PEA help brain fog or memory in a healthy person?
Nothing in this literature addresses that. The participants all had a neurodegenerative diagnosis, and the cognitive screening tests that a healthy person might recognise showed no treatment effect even within the trial [1].
How strong is the evidence overall?
It is the strongest neurological evidence for the compound, and it is still Phase 2: one centre, 48 randomized, 45 completers, 24 weeks, with the authors calling for further investigation [1].
References
- Assogna M, Di Lorenzo F, Bonnì S, et al. Phase 2 study of palmitoylethanolamide combined with luteoline in frontotemporal dementia patients. Brain Commun (2025). PMID 40046339
- Scuderi C, Stecca C, Valenza M, et al. Palmitoylethanolamide controls reactive gliosis and exerts neuroprotective functions in a rat model of Alzheimer disease. Cell Death Dis (2014). PMID 25210802
- Beggiato S, Tomasini MC, Ferraro L, et al. Effects of Palmitoylethanolamide on Neurodegenerative Diseases: A Review from Rodents to Humans. Biomolecules (2022). PMID 35625595
- Petrosino S, Schiano Moriello A. Palmitoylethanolamide: A Nutritional Approach to Keep Neuroinflammation within Physiological Boundaries. Int J Mol Sci (2020). PMID 33333772
- Linda Rankin et al. Palmitoylethanolamide for the treatment of pain: pharmacokinetics, safety and efficacy. British Journal of Clinical Pharmacology (2016). PMID 27220803
- Rankin L et al. The Basal Pharmacology of Palmitoylethanolamide. International journal of molecular sciences (2020). PMID 33114698
- Assogna M et al. Synaptic Effects of Palmitoylethanolamide in Neurodegenerative Disorders. Biomolecules (2022). PMID 36009055
- Colizzi M et al. Therapeutic effect of palmitoylethanolamide in cognitive decline: A systematic review and preliminary meta-analysis of preclinical and clinical evidence. Frontiers in psychiatry (2022). PMID 36387000
These statements have not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease. Content is for informational purposes only and is not medical advice; consult a qualified healthcare provider before starting any supplement. As an Amazon Associate we earn from qualifying purchases.