23 July 2002. Stockholm. Today at the 8th International Conference on Alzheimer’s Disease and Related Disorders, Monique Breteler and colleagues at Erasmus Medical Center in Rotterdam, Netherlands, reported the surprising result of a reanalysis of their previous paper, which had sparked controversy (see related news item, see NSAID discussion transcript). Last fall, Bas In’t Veld et al. reported that NSAID consumption protected against AD, and that this applied to all NSAIDs that were taken by the 7,983 study participants of their community-based cohort. Shortly thereafter, Sascha Weggen et al. reported in Nature that, in vitro, the NSAIDs ibuprofen, indomethacin, and sulindac, but not naproxen or diclofenac, work by lowering Aβ42 production. This triggered debate about whether the experimental data was applicable to human AD, and about which NSAIDs would work in clinical treatment trials.
Prompted by the Weggen et al. paper, Breteler then reanalyzed their data by grouping the Aβ42-lowering and non-lowering NSAIDs in separate groups. Their poster showed that the protective effect they previously reported for all NSAIDs lumped together is actually restricted mostly to those that lowered Aβ in the Weggen et al. experiments. The adjusted risk ratio for ibuprofen, indomethacin, and sulindac decreased down to 0.62 depending on how long study participants took the drugs. However, the adjusted risk ratio for diclofenac and naproxen remained around one. “Our reanalysis agrees completely with the Weggen et al. data,” said Breteler. The data is not yet published -Gabrielle Strobel.