Work

O4‐01‐05: Use of biomarkers as endpoint in trials with subjects with preclinical and prodromal Alzheimer's disease: An ADNI study

Year: 2012
Type: article
Abstract: Alzheimer's disease (AD) modifying drugs are probably most effective when given in subjects with preclinical or prodromal AD. However, reliable outcome measures in this population are still under investigation. Cognitive outcomes are less useful because cognitive decline is slower than in demented subjects. Our aim was to compare different biomarkers and cognitive tests as outcome measure in subjects with preclinical AD and prodromal AD. We investigated changes in biomarkers and cognitive markers over time and determined sample size needed to find a significant treatment effect on these markers in a fictive trial. We selected subjects with preclinical AD (normal cognition and CSF Aß 1-42 (Aß1-42) <193pg/l) and prodromal AD (MCI and Aß1-42 <193pg/l) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Biomarker outcomes were CSF abeta1-42, CSF tau, metabolism on FDG-PET, and whole brain, hippocampal and ventricle volume on MRI. Cognitive outcomes were ADAS-Cog and MMSE scores. We estimated slopes of decline over a period up to 4 years with mixed models. Markers that declined significantly (P<0.05) were used to determine sample size. Total sample size was calculated for a fictive 3-year trial, showing a decline of 30% with a power of 80%, a 2-sided alpha of 5% and an annual dropout of 10%. Forty-five preclinical AD and 148 prodromal AD subjects were included. Subjects with preclinical AD showed significant decline on Aß1-42 (P = 0.028), tau (P = 0.05) and all MRI measures (P = 0.000). Subjects with prodromal AD declined on all markers, except Aß1-42. In comparison to subjects with preclinical AD, subjects with prodromal AD showed more decline on FDG-PET (P = 0.002), whole brain volume (P = 0.03), ventricular volume (P = 0.000), hippocampal volume (P = 0.000) and the ADAS-Cog and MMSE scores (P = 0.000). Sample sizes for each outcome measure are summarized in the table. In subjects with preclinical AD only CSF and MRI markers declined over time. Change in MRI marker would be the best outcome measure for treatment trials, at least for proof of concept studies. In subjects with prodromal AD all markers except amyloid declined over time. Also in this population change in brain atrophy on MRI would be the best outcome measure.
Language: English
Field: Medicine
Open Access status: bronze
OpenAlex ID: W2015329082
FWCI: 0.179
Cites:
Cited by: 1
Related to: 10