Results of a US study of elderly men and women with a family history of Alzheimer’s suggest that the nonsteroidal anti-inflammatory drugs (NSAIDs) naproxen and celecoxib did not improve cognitive function and the investigators recommended against using them to prevent Alzheimer’s.
Previous research, based on observational studies, has suggested that risk of Alzheimer’s is lower in people who use NSAIDs, supporting the idea that inflammatory processes are involved in the development of the neurodegenerative disease and in the decline of cognitive function such as thinking, memory and learning.
The longitudinal study is the work of the ADAPT research group and was published in the online ahead of print issue of the Archives of Neurology on 12th May. The purpose was to investigate the effect of two NSAIDs, naproxen (brands include Aleve, Synflex, Naprelan) and celecoxib (brands include Celebrex and Celebra) on cognitive ability in the elderly.
ADAPT stands for the Alzheimer’s Disease Anti-inflammatory Prevention Trial, a randomized, placebo-controlled, multicenter, primary prevention trial sponsored by the National Institute on Aging. The trial ran from March 2001 to December 2004 and recruited over 2,500 participants who were aged 70 and over, from areas surrounding the 6 trial sites: Baltimore, Maryland; Boston, Massachusetts; Rochester, New York; Seattle, Washington; Sun City, Arizona; and Tampa, Florida.
The participants had to have a history of at least 1 first-degree relative with Alzheimer’s-like dementia, and not have cognitive impairment at the time of recruitment (this was tested using a battery of cognitive tests). Anyone using NSAIDs regularly was excluded, but anyone using up to 81 mg a day of aspirin was included. 2,117 participants completed the trial and had follow up cognitive assessments.
The participants were randomly assigned to three groups. One group, comprising around 600 participants, took Celecoxib (200 mg twice daily), a second group, also comprising about 600 of the participants took naproxen sodium (220 mg twice daily), and the third group, comprising around 900 of the participants, had the placebo.
The main outcome measures were seven tests of cognitive skills, and an overall summary score of all the tests, measured each year. The study finished early, in December 2004, because a previous study showed a link between celecoxib and increased risk of cardiovascular problems.
The results showed that:
- The global summary scores over time were significantly lower for naproxen (but not for celecoxib) compared with placebo (-0.05 standard deviations; p= .02).
- For one specific cognitive function test, the Modified Mini-Mental State Examination (modified MMSE), often used to screen for dementia, both treatment groups showed lower scores over time compared with placebo (-0.33 points for celecoxib, p= .04; and -0.36 points for naproxen, p= .02).
- Removing people with dementia from the analyses reduced the differences among the three groups.
- Limiting analyses to scores obtained while participants were being treated with the study drugs showed similar results to the intention to treat analyses.
An intention to treat analysis is where everyone who starts on the trial is treated, in the figures, as part of the trial, even if they dropped out. It’s a statistical method used by epidemiologists to overcome distortion that can creep in when people drop out of trials (which effectively undermines the randomization of subjects among treatment groups).
The ADAPT researchers concluded that:
“Use of naproxen or celecoxib did not improve cognitive function. ”
They also said that, “There was weak evidence for a detrimental effect of naproxen”.
When they looked at the figures including six months after treatment finished, they found the ADAPT cognitive test results “did not show a protective effect use of NSAIDs and may suggest that cognitive scores are lower”.
The researchers said there could be several explanations for why these results clash with previous observational studies.
Firstly, observational studies do not randomly assign patients to treatment groups and then monitor the effects of interventions; they follow and analyze the behaviour or symptoms of groups that are already formed, and additional factors that were not measured may have disturbed the results of such previous studies, suggested the authors.
Secondly, the results of this trial may not apply to the NSAIDs such as ibuprofen, only to celecoxib and naproxen, and thirdly, perhaps NSAIDs only have protective powers when taken several years before cognitive function would start to fall.
They recommended continued follow up of participants after treatment, in order to find out if there are any further delayed timing effects. However, in the meantime they recommended that “naproxen and celecoxib should not be used for the prevention of Alzheimer’s disease“.
“Cognitive Function Over Time in the Alzheimer’s Disease Anti-inflammatory Prevention Trial (ADAPT): Results of a Randomized, Controlled Trial of Naproxen and Celecoxib.”
ADAPT Research Group.
Arch Neurol. 2008; 65 (7), Early Release Article, posted May 12, 2008.
DOI:10.1001/archneur.2008.65.7.nct70006.
Sources: Journal abstract, JAMA press release.
Written by: Catharine Paddock, PhD