Randomized Trial of Low-dose Naproxen in Cognitively Intact Persons at Risk of Alzheimer's Dementia

The trial enrolled 195 cognitively normal persons aged 60+ with either a parental history of AD or a history of two or more affected first-degree relatives. Persons aged 55-59 were admitted if their current age was <= 15 years younger than AD onset in their index relative. Such persons are believed to be at approximately 3-fold increased risk of AD dementia. We expected a majority of them to show evidence of progressive pre-clinical AD. Participants were randomized 1:1 to receive the common non-steroidal anti-inflammatory drug (NSAID) naproxen in over-the-counter dosage (naproxen sodium 220 mg) or identical-appearing placebo tablets twice daily. At baseline and at three follow-up visits (3 months, 12 months and 24 months after randomization) they were tested for cognitive abilities and undergo brain imaging with both structural and functional MRI. They are also tested for sensori-neural capacities in olfactory identification and in the ability to discern spoken language in a distracting environment (to test central auditory processing). About 2/3 of participants also volunteered to undergo a series of lumbar punctures for donation of cerebrospinal fluid (CSF), which was assayed for several biochemical markers of AD that are now understood to be present for a decade or longer before the onset of symptoms. As well, their plasma and CSF are assayed for presence of naproxen and for numerous markers of inflammatory processes (cytokines and chemokines). The central hypothesis was that administration of naproxen would not only suppress these inflammatory markers but would also slow or reverse the progress of change in cognition and in biomarkers of the pre-clinical stage of AD. The analysis plan followed the principle of modified Intent-to-Treat, considering outcomes for all persons who had at least one follow-up examination while on-protocol. After completion of two years of treatment, these participants are being followed for a further two years to observe whether treatment-related changes are sustained -- indicating that the treatment effects represent modification of the disease process itself, as opposed to a temporary change in brain function.

Inclusion criteria

• good physical health including normal hemoglobin and hematocrit
• history or documentation of AD dementia in at least one parent, or in two siblings
• cognitive performance without diagnosable deficit such as dementia, "mild cognitive impairment"
• must have spouse or companion able to accompany participant for clinic visits
• six or more years of formal education
• fluent in either English or French
• provision of informed consent

Exclusion criteria

• no current peptic ulcer disease
• no history of prior peptic ulcer with bleed, perforation, intestinal obstruction
• no major psychiatric disturbance
• no regular use (4 or more doses per week) of aspirin, other non-steroidal anti-inflammatory drug (NSAID), opiate or other pain medication
• no use, present or past, of acetylcholinesterase inhibitors or memantine
• no regular use of vitamin E at dosage of 600 i.u.
• no drug or alcohol dependence
• no allergy to NSAIDs or sulfa antibiotics