Models of Neuroimaging, Biomarkers, and Cognitive in Alzheimer's Disease. Implications for Clinical Trial Design.

Objectives: Identify a window for early treatment by estimating the time course of early pathophysiological changes in Alzheimer's disease, clarify the relationship between emerging pathology and symptom onset as well as estimate the time to clinically meaningful decline in order to inform clinical trial design.Methods: The participants included in the analyses of the five papers were drawn from four cohorts: ADNI, AIBL, BioFINDER, A4.Repeated measures of longitudinal MRI, PET, CSF and cognitive responses were modeled using (1) mixed-effects regression with a random intercept and slope or (2) generalized least squares.Nonlinearity in longitudinal responses was captured using restricted cubic splines.Clinical trial scenarios were simulated to estimate the power to detect assumed drug effects.Results: Clinical trials in preclinical AD are generally underpowered to detect a plausible treatment effect.Optimal composites to capture decline in the observed preclinical AD population were equal weight composites across all available cognitive and functional measures.Estimates of several major milestone events of AD progression include changes in CSF Aβ42 29 years before Aβpositivity, an increase in regional Aβ PET deposition 15 years before, increases in tau pathology 7-8 years before, and signs of cognitive dysfunction 4-6 years before Aβ-positivity.Cognitively unimpaired Aβ+ participants approach early MCI cognitive performance levels on general cognition six years after baseline.To achieve 80% power to detect a 25% treatment effect, 2,000 participants/group for a 4-year trial and 600 participants/group for a 6-year trial are required.Discussion: Including a large number of components in a cognitive/functional composite endpoint may smooth over aberrations in scores in a particular assessment from visit to visit within a subject, thus lowering the withinsubject variance and improving signal to noise.In later stage preclinical AD, suitable power for a phase III trial can be achieved with considerably lower sample sizes while capturing both cognitive and functional change to demonstrate a clinically meaningful drug effect-both while initiating treatment in subjects who are still cognitively unimpaired.Small but meaningful increases in levels of CSF tau and temporoparietal tau are observed years before the current threshold for Aβ-positivity.In the context of secondary prevention trials, tau levels in these participants would already have been increasing for several years, likely more.These data support the use of primary prevention trials against Aβ where treatment is initiated years before the current threshold for Aβ-positivity.The separation between cognitively unimpaired participants and early MCI was just over one SD on the PACC, suggesting that one point of additional decline in Aβ+ participants compared to Aβ-participants could be taken as an approximate benchmark for clinically meaningful decline.Based on the PACC estimates, a treatment effect of 40%-50% would be required to delay the cognitive decline of a group of Aβ+ participants from reaching the one SD milestone by three years.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.