With specific and truly effective treatments for Alzheimer's still years away, non-steroidal anti-inflammatory drugs (NSAIDs) today are among the most promising treatments to delay the onset of Alzheimer's disease. Yet studies to date have had mixed results, some finding a protective effect, some not. Most of those studies have in common the problem that asking elderly people about which doses of what drugs they have taken in the past is error-prone. After all, people with incipient Alzheimer's forget such things. A paper in the November 22 New England Journal of Medicine moves the NSAID question a step forward by observing prospectively, for seven years, a population-based cohort of 6,989 initially nondemented elderly people whose NSAID prescriptions were fully documented in pharmacy records.

Bas In'T Veld, Bruno Stricker, and others at the Erasmus Medical Center, Rotterdam, the Netherlands, found that the risk of developing Alzheimer's decreased with increasing NSAID intake. Compared to non-users, those who took NSAIDs for up to two years had a 0.83 relative risk of developing AD. For those who took the drugs even longer, the relative risk dropped markedly, to 0.2.

(This link did not vary with age or ApoE status; NSAIDs did not protect against vascular dementia in this cohort; and the use of corticosteroids and non-narcotic analgesics did not correlate with AD risk. Most people used NSAIDs for osteo- or rheumatoid arthritis.)

Interestingly, the authors also found that it made no difference which NSAID one was taking. This finding contradicts a recent study by Todd Golde and Edward Koo, reporting that some, but not all, NSAIDs preferentially reduce Aβ42 production (see ARF related news story). The study went on to suggest that NSAIDs may treat AD by directly affecting brain amyloid pathology rather than by inhibiting cyclooxygenase (COX) activity, as has been widely assumed. The authors found effects on Aβ generation for ibuprofen, indomethacin, and sulindac sulphide but not naproxen. However, in the Rotterdam study, naproxen was the third-most commonly used NSAID and was as effective as all other NSAIDs.

How to reconcile this discrepancy? Pat McGeer, who has investigated the inflammatory response and the use of Cox inhibitors in AD (McGeer & McGeer, 1995; McGeer, 2000) says that the Golde/Koo data may not be applicable to humans because they came from experiments with APP-transfected cells exposed to NSAID concentrations several orders of magnitude higher than what humans can tolerate. "On the bases of everything we know so far, all NSAIDs are equally effective except for selective Cox2 inhibitors, which are less effective," said McGeer.

Definitive answers on what works in humans must await the completion of clinical trials. Several treatment trials and at least one prevention trial of some NSAIDs are currently ongoing (see NIH trial information).—Gabrielle Strobel


  1. Nice study, but confirmatory nonetheless.

    View all comments by Edward Koo
  2. I believe this study has the potential to accelerate active AD prevention. I'm puzzeled that the re-analysis of data (see 23 July 2002 news story) discerning between NSAIDs is taking so long to be published. This has to be seen as a CORRECTION, and an ALERT and promptly made available.

    View all comments by Larry Nault

Make a Comment

To make a comment you must login or register.


News Citations

  1. Anti-inflammatory Drugs Side-Step COX Cascade to Target Aβ

External Citations

  1. see NIH trial information

Further Reading

Primary Papers

  1. . Nonsteroidal antiinflammatory drugs and the risk of Alzheimer's disease. N Engl J Med. 2001 Nov 22;345(21):1515-21. PubMed.