Safety profile of Alzheimer's disease populations in Alzheimer's Disease Neuroimaging Initiative and other 18-month studies.

Background: Demonstration of a disease-modifying effect of a therapeutic agent on Alzheimer's disease (AD) requires a trial lasting for at least 18 months. An understanding of expected rates of adverse events (AEs), overall discontinuations, and discontinuations due to AEs, serious AEs, and deaths would be useful in planning such trials.

Methods: We examined safety information for patients taking placebo from five published 18-month AD trials and for patients from the Alzheimer's Disease Neuroimaging Initiative study.

Results: AEs reported consistently across multiple studies were dyspnea (occurring in 5.3%-5.8% of patients), headache (4.0%-5.5%), constipation (4.3%-4.7%), nausea (2.0%-5.8%), joint swelling (3.6%-3.7%), vomiting (3.6%-3.7%), and anxiety (3.2%-3.6%). Larger multinational studies, as compared with smaller studies with fewer sites and geographies, demonstrated greater overall discontinuations (24.6%-33.0% vs 8.2%-21.0%) and greater discontinuations due to AEs (9.5%-11.6% vs 2.7%-3.2%). Rates of death (1.8%-2.4%) and SAEs (19.9%-21.2%) were consistent across 18 month published studies and in ADNI; fall was the most common SAE (2.6%-4.0%) where SAEs were reported.

Conclusions: In general, comparable types of AEs, frequency of deaths, and serious AEs were seen for patients taking placebo in five randomized, controlled 18-month AD trials and in Alzheimer's Disease Neuroimaging Initiative, whereas rates of discontinuations were more variable. Evaluation across studies was complicated by inconsistent methods of reporting safety information. Evaluation of large databases of placebo patients from therapeutic AD trials is needed to further enhance the understanding of expected safety outcomes in clinical trials of AD patients.