Home › Brain News
By StudyFinds Analysis
•
Reviewed by John Anderer
Research led by Dr. Chu Chen and Teresa Lozano (University of Texas Health Science Center at San Antonio)
Jan 29, 2026
(© Peter Kim – stock.adobe.com)
Scientists have discovered that combining a low dose of THC, the active ingredient in marijuana, with celecoxib (an arthritis drug sold as Celebrex) helps protect against memory loss and brain damage in mouse models of Alzheimer’s disease. The pairing works better than either medication alone, and because both are already FDA-approved, this approach could reach human testing faster than experimental drugs still in development.
Researchers at the University of Texas Health Science Center tested the combination in mice bred to develop Alzheimer’s-like symptoms. Animals receiving both THC and celecoxib performed better on memory tests and showed less brain deterioration than untreated mice. The treatment also shifted disease patterns inside individual brain cells, pushing them back toward a healthier state.
The findings, published in Aging and Disease, offer hope for a disease affecting more than 6.9 million Americans. Both medications have been prescribed safely for decades: THC formulations like Marinol treat nausea in cancer patients, while celecoxib reduces arthritis inflammation. This existing safety record could accelerate the path from laboratory discovery to clinical trials.
THC presents a paradox for researchers. High doses impair memory and cognition, but previous studies suggested lower amounts might actually protect aging brains. The challenge was finding a dose that’s just right; high enough to provide benefits without causing marijuana’s side effects.
The team tested doses ranging from 0.1 to 3.0 milligrams per kilogram in mice. The lowest doses barely made a dent. At 1.0 mg/kg, THC started reducing harmful brain proteins, but effects remained modest. The 3.0 mg/kg dose delivered better results, slashing levels of toxic amyloid-beta proteins by more than half and boosting proteins needed for healthy brain connections.
The dose was considered low in the study, and THC is already used medically in prescription form, but the right dose for people would need to be tested in clinical trials. At the levels used in mice, the animals showed brain-protective benefits without the cognitive impairment and intoxication associated with higher THC exposures.
Mice received daily treatment for 30 days starting at five months old, just as early symptoms were beginning to appear. In Alzheimer’s, two abnormal proteins accumulate in the brain: amyloid-beta forms sticky plaques between neurons, while tau protein tangles inside cells. Both kill neurons and disrupt communication. THC treatment dramatically reduced both toxic proteins while also decreasing brain inflammation and cell death.
While THC reduced Alzheimer’s damage, it simultaneously triggered inflammation in the brain, potentially undermining its own benefits. Even at the lowest dose tested, THC increased production of inflammatory molecules that contribute to brain damage when chronically elevated.
Earlier research showed that THC’s memory-impairing effects involve activation of COX-2, an enzyme driving inflammation. celecoxib specifically blocks COX-2. The researchers bet that pairing these medications might shut down THC’s inflammatory side effects while keeping its protective actions.
It worked. Mice receiving both drugs showed far less brain inflammation than those given THC alone. celecoxib neutralized THC’s tendency to activate inflammatory pathways without interfering with its brain-protecting effects. The anti-inflammatory drug was given 30 minutes before THC each day at a very low human-equivalent dose.
Using advanced genetic analysis, researchers examined individual brain cells to understand what was happening at the molecular level. They looked at cells from the hippocampus, the brain region most involved in memory and most damaged in Alzheimer’s.
In Alzheimer’s mice, genes controlling connections between brain cells had gone haywire: some were overactive, others suppressed. THC treatment reversed many of these changes, pushing gene activity back toward normal. The combination with celecoxib worked even better. Genes that had been shut down in disease were turned back on. Genes involved in clearing toxic proteins from the brain were restored. Inflammatory genes that had spiked were brought back down.
The treatment shifted many of those cellular signals back toward a healthier pattern, essentially normalizing the disease’s genetic fingerprint inside brain cells.
To test whether the molecular changes translated to actual cognitive improvement, researchers used the Morris water maze: a pool where mice must remember the location of a hidden platform just beneath the water’s surface.
Untreated Alzheimer’s mice struggled over seven training days, swimming in confused circles and taking a long time to find the platform. Mice receiving the combination treatment performed almost as well as healthy controls, steadily improving and reaching the platform quickly by the final day.
Another test came 24 hours later when researchers removed the platform entirely. Healthy mice spent most of their time swimming in the quadrant where the platform used to be. They remembered. Untreated Alzheimer’s mice swam randomly, their memory gone.
The treatment worked in two different mouse models: one developing the amyloid-beta plaques seen in Alzheimer’s, another developing the tau tangles. Both types of mice showed improved memory and reduced brain damage, suggesting the combination tackles multiple aspects of the disease.
Another exciting aspect of these findings is the fact both medications are already FDA-approved. These drugs have been used safely among millions of patients for decades. Developing new drugs from scratch typically takes over a decade and costs billions, and most candidates fail in clinical trials due to unexpected side effects.
This combination could lower some of the usual barriers associated with human trials. The safety profiles are already established. The doses used were remarkably low, well below standard prescriptions for both medications. Such low doses should minimize side effects.
Previous trials testing anti-inflammatory drugs alone in Alzheimer’s patients failed. But those studies didn’t combine them with compounds specifically targeting the toxic proteins. The current research suggests both are needed. Something to attack the disease and something to control inflammation.
The researchers focused on prevention rather than cure, treating mice just as early symptoms were beginning. This mirrors mild cognitive impairment in humans, which is the stage where memory problems are noticeable but don’t yet severely impact daily life. Starting treatment before extensive brain damage occurs might preserve function in ways impossible once neurons have already died.
People with mild cognitive impairment could be ideal test candidates. The window between first symptoms and severe dementia typically lasts several years, providing an opportunity for early intervention. If this combination can halt or slow that progression in humans, it could potentially help some people stay in the mild stage longer rather than sliding into more severe dementia.
In summation, the findings position low-dose THC plus celecoxib as a potentially faster path to clinical testing. The combination targets multiple disease processes while leveraging decades of existing safety data, making it an unusually practical candidate for human trials aimed at preserving cognitive function in at-risk populations.
Disclaimer: This article is for general information only and is not medical advice. The findings described come from studies in mice, not people, and do not prove that THC, celecoxib, or their combination can prevent or treat Alzheimer’s disease in humans. Anyone concerned about memory problems or considering changes to medication should consult a qualified healthcare professional.
This study used a 30-day treatment period to test the concept. While sufficient to demonstrate brain-protective and anti-inflammatory effects, Alzheimer’s is a chronic disease developing over years or decades. Long-term studies are needed to assess sustained benefits, safety over extended periods, and potential side effects. The research was conducted in mouse models that replicate specific aspects of Alzheimer’s but may not fully capture the complexity of human disease. Translation to humans will require clinical trials to determine optimal dosing, treatment duration, and which patient populations would benefit most.
This work was supported by the National Institutes of Health (grant R01NS076815) and by startup and endowment funds from the Joe R. & Teresa Lozano Long School of Medicine at University of Texas Health Science Center at San Antonio. The National Institute on Drug Abuse Drug Supply Program provided Δ9-THC, and the NIMH Chemical Synthesis and Drug Supply Program provided celecoxib. The authors declared no competing financial interests.
Authors: Jian Zhang, Dexiao Zhu, Mei Hu, Mingzhe Pan, Chu Chen (corresponding author) | Affiliation: Department of Cellular and Integrative Physiology, Joe R. and Teresa Lozano Long School of Medicine, University of Texas Health Science Center at San Antonio, San Antonio, TX 78229, USA | Journal: Aging and Disease | Paper Title: “A Combination of Low-Dose Δ9-THC and Celecoxib as a Therapeutic Strategy for Alzheimer’s Disease” | DOI: http://dx.doi.org/10.14336/AD.2025.1206 | Date: Received September 24, 2025; Revised December 2, 2025; Accepted December 3, 2025; Early access date December 18, 2025 | Data Availability: Single-nucleus RNA sequencing data deposited in NCBI Gene Expression Omnibus under accession number GSE301659. Supplementary materials and code available through specified repositories.
About StudyFinds Analysis
Called “brilliant,” “fantastic,” and “spot on” by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.
StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.
Our Editorial Team
Steve Fink
Editor-in-Chief
Sophia Naughton
Associate Editor
January 29, 2026
January 29, 2026
January 29, 2026
January 29, 2026
As Seen On