An Indicator Cell Assay-based Multivariate Blood Test for Early Detection of Alzheimer's Disease

Area of Science:

• Biomarker Discovery
• Neurodegenerative Diseases
• Next-Generation Diagnostics

Background:

• Current Alzheimer's disease (AD) diagnostics often lack sensitivity for early detection.
• There is a need for blood-based tests that can identify AD at mild cognitive impairment/mild dementia stages.
• Novel biosensor technologies are required to amplify subtle disease signals in patient samples.

Purpose of the Study:

• To develop and evaluate an Alzheimer's disease indicator cell assay platform (AD-iCAP) for early AD detection.
• To assess the performance of a machine learning-based classifier trained on AD-iCAP readouts.
• To explore the potential of AD-iCAP as a complementary diagnostic tool and for precision medicine applications.

Main Methods:

• Patient plasma was incubated with standardized neurons to generate gene-expression readouts.
• Multivariate disease classifiers were trained using machine learning on these gene-expression data.
• Systems biology analyses (GSEA, PCA, correlation/network analyses) were used for optimization and validation.

Main Results:

• The AD-iCAP model achieved an AUC of 0.64 on an external validation set and 0.77 on a blind set.
• Assay readouts were enriched for Alzheimer's-relevant pathways, including cholesterol biosynthesis and synaptic function.
• The assay demonstrated orthogonality to amyloid and tau measurements, suggesting a complementary role.

Conclusions:

• AD-iCAP is a promising novel approach for early Alzheimer's disease detection using blood samples.
• The assay's ability to detect multivalent cellular responses offers potential for patient stratification and precision medicine.
• Further prospective validation in larger cohorts is warranted to confirm its clinical utility.