Identification of preclinical dementia according to ATN classification for stratified trial recruitment: A machine learning approach

INTRODUCTION

The Amyloid/Tau/Neurodegeneration (ATN) framework was proposed to identify the preclinical biological state of Alzheimer's disease (AD). We investigated whether ATN phenotype can be predicted using routinely collected research cohort data.

METHODS

927 EPAD LCS cohort participants free of dementia or Mild Cognitive Impairment were separated into 5 ATN categories. We used machine learning (ML) methods to identify a set of significant features separating each neurodegeneration-related group from controls (A-T-(N)-). Random Forest and linear-kernel SVM with stratified 5-fold cross validations were used to optimize model whose performance was then tested in the ADNI database.

RESULTS

Our optimal results outperformed ATN cross-validated logistic regression models by between 2.2% and 8.3%. The optimal feature sets were not consistent across the 4 models with the AD pathologic change vs controls set differing the most from the rest. Because of that we have identified a subset of 10 features that yield results very close or identical to the optimal.

DISCUSSION

Our study demonstrates the gains offered by ML in generating ATN risk prediction over logistic regression models among pre-dementia individuals.

Autosomal dominant and sporadic late onset Alzheimer's disease share a common in vivo pathophysiology

The extent to which the pathophysiology of autosomal dominant Alzheimer's disease corresponds to the pathophysiology of 'sporadic' late onset Alzheimer's disease is unknown, thus limiting the extrapolation of study findings and clinical trial results in autosomal dominant Alzheimer's disease to late onset Alzheimer's disease. We compared brain MRI and amyloid PET data, as well as CSF concentrations of amyloid-β42, amyloid-β40, tau and tau phosphorylated at position 181, in 292 carriers of pathogenic variants for Alzheimer's disease from the Dominantly Inherited Alzheimer Network, with corresponding data from 559 participants from the Alzheimer's Disease Neuroimaging Initiative. Imaging data and CSF samples were reprocessed as appropriate to guarantee uniform pipelines and assays. Data analyses yielded rates of change before and after symptomatic onset of Alzheimer's disease, allowing the alignment of the ∼30-year age difference between the cohorts on a clinically meaningful anchor point, namely the participant age at symptomatic onset. Biomarker profiles were similar for both autosomal dominant Alzheimer's disease and late onset Alzheimer's disease. Both groups demonstrated accelerated rates of decline in cognitive performance and in regional brain volume loss after symptomatic onset. Although amyloid burden accumulation as determined by PET was greater after symptomatic onset in autosomal dominant Alzheimer's disease than in late onset Alzheimer's disease participants, CSF assays of amyloid-β42, amyloid-β40, tau and p-tau181 were largely overlapping in both groups. Rates of change in cognitive performance and hippocampal volume loss after symptomatic onset were more aggressive for autosomal dominant Alzheimer's disease participants. These findings suggest a similar pathophysiology of autosomal dominant Alzheimer's disease and late onset Alzheimer's disease, supporting a shared pathobiological construct.