Reader response: INTREPAD: A randomized trial of naproxen to slow progress of presymptomatic Alzheimer disease

I read with interest the Null Hypothesis article by Meyer et al.¹ In 2001, Weggen et al.² suggested that flurbiprofen and ibuprofen had an effect on gamma secretase not shared by other cyclooxygenase inhibitors. This finding, along with several other failed studies of anti-inflammatory mechanisms in Alzheimer disease (AD), suggests that the widely replicated epidemiologic findings of nonsteroidal anti-inflammatory drug (NSAID) benefit for decreasing AD risk could be related to the effect on gamma-secretase, not inflammation. A flurbiprofen study failed to show a benefit for patients with AD,³ but a considerable number of studies suggest that the NSAID-related benefit requires years of treatment before dementia develops, which the study by Meyer et al.¹ was appropriately targeting. However, the question remains as to whether ibuprofen is the drug to test, not naproxen.

Accurate measurement in early AD is also problematic. The failure to find an effect with the Alzheimer Progression Score, with quite large variability bars, suggests that tools for much more accurate assessments of early AD effects are needed.

• © 2020 American Academy of Neurology