NL
DE
Neurologist Edo Richard, PhD student Melina den Brok and Pim van Gool (neurologist at AmsterdamUMC) have used a Bayesian meta-analysis of published randomized controlled trials to show that there is strong evidence of absence of an effect of anti-amyloid immunotherapy in Alzheimer’s disease. Their systematic review is published in Alzheimer’s & Dementia, on May 31st 2021
This goes well beyond the much weaker conclusion resulting from frequentist analysis, which only allows to conclude there is absence of evidence. They show that it is extremely unlikely that a new RCT, even if it does show an effect on the progression of Alzheimer’s Disease, will change the overall conclusion of the absence of an effect. In addition to this, all previously reported effects, none of which was statistically significant, were well below an effect that would suggest clinical relevance. Their work illustrates the value of Bayesian meta-analysis, particularly in the case of repeated small non-significant effects in individual trials. They argue that other lines of research into treatment of Alzheimer’s disease should be pursued, which have more potential to bear fruit. This is particularly salient in the context of the recent FDA decision to conditionally approve the anti-amyloid drug aducanumab .
Publication
Edo Richard, Melina den Brok, Pim van Gool. Bayes analysis supports null hypothesis of anti-amyloid beta therapy in Alzheimer’s Disease (2021). Alzheimer’s & Dementia. doi: 10.1002/alz.12379.
