Quantifying the heterogeneity of cognitive functioning in Alzheimer's disease to extend the placebo-treatment dichotomy: Latent class analysis of individual-participant data from five pivotal randomized clinical trials of donepezil

By: Levine S.Z., Goldberg Y., Yoshida K., Samara M., Cipriani A., Iwatsubo T., Leucht S., Furawaka T.A.
Published in: European psychiatry : the journal of the Association of European Psychiatrists
SDGs : SDG 03  |  Units: Social Welfare & Health Sciences  | Time: 2021 |  Link
Description: BACKGROUND: The extent and profiles of heterogeneity in cognitive functioning among participants in clinical trials of a ntidementia medication are unknown. We aimed to quantify and identify profiles of heterogeneity of cognition in Alzheimer's disease. METHODS: Individual-level participant data were analyzed from five pivotal clinical trials of donepezil for Alzheimer's disease (N = 2,919). Based on Alzheimer's Disease Assessment Scale-Cognitive Subscale total scores from baseline up to week 12, a latent class model was used to identify heterogeneous groups. A logistic regression model was used to examine factors associated with group membership. Sensitivity analysis was conducted, restricted to the donepezil, and then the placebo arm. RESULTS: The latent class model identified three classes labeled as low scorers (i.e., least cognitive impairment; N = 1,666, 76.04%), improvers (N = 27, 1.23%), and high scorers (N = 498, 22.73%). Logistic modeling showed that donepezil compared to placebo was significantly (p < 0.05) positively associated with membership in the improvers class (OR = 6.88, 95% CI = 2.03, 42.95), and negatively with high scorers (OR = 0.79, 95% CI = 0.64, 0.98). Sensitivity analysis restricted to the placebo, then donepezil arms replicated similar heterogeneity patterns. CONCLUSIONS: Our results inform clinicians regarding the extent of heterogeneity in cognitive functioning during treatment and contribute to trial design considerations.