[P2–414]: CHARACTERISING THE PROGRESSION OF ALZHEIMER's DISEASE SUBTYPES USING SUBTYPE AND STAGE INFERENCE (SUSTAIN)
Year: 2017
Type: article
Abstract: modelled regional declines in [ 18 F]FDG over 2 years as a function of baseline [ 18 F]florbetapir amyloid-b positron emission tomography (PET) and cerebrospinal phosphorylated tau thresholds.Rsquared and F-test served as the basis of the goodness-of-fit, while the Akaike information criterion compared linear and non-linear functions.The non-linear associations were assumed as sigmoidal: Y¼a+(b/(1+e^((-(x-c))/d))).The thresholds predictive of imminent metabolic decline were defined as the value where the curvatures changed signs (X¼c).We performed, for the first time, a voxelwise power analysis calculation to define the sample size needed for a disease-modifying therapy test an estimated 25% drug effect with 80% of power in changes in [ 18 F]FDG over 2 years across biomarker groups.Results: Sample size calculation, using averaged change in mediobasal temporal [ 18 F]FDG revealed that studies using [ 18 F]florbetapir plus p-tau thresholds of imminent neurodegeneration would require 87 individuals per trial arm for testing a drug effect of 25% over 24 months.For test the same drug effect, a study using preclinical AD individuals with standard thresholds would require 422 individuals for arm, whereas using all individuals of the population 756 individual per trail arm (Fig. 1).Conclusions: These results highlight the concept of combined amyloid-b and tau thresholds of imminent neurodegeneration as a framework with a high statistical power for testing the effects of anti-amyloid and anti-tau therapies in regional changes in [ 18 F]FDG over a typical 2-year clinical trial.
Source: Alzheimer s & Dementia
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4242394686