An Exploratory Six-Probe Blood RNA Signature for Predicting 12-Month Cognitive Decline Along the Alzheimer's Disease...

Area of Science:

• Neuroscience
• Genomics
• Biomarker Discovery

Background:

• Predicting Alzheimer's disease (AD) progression is crucial for patient care but remains challenging.
• Existing blood transcriptomic studies lack reproducibility and clarity on signal origin (transcriptional vs. immune cell shifts).
• Generalizability of blood-based biomarkers across different platforms is not well-established.

Purpose of the Study:

• To develop and validate a reproducible blood-based RNA panel for predicting 12-month cognitive decline in Alzheimer's disease (AD).
• To confirm the transcriptional origin of the predictive signal and assess cross-platform generalizability.
• To establish conditions for prospective validation of the AD biomarker.

Main Methods:

• Applied a five-step analytical process to 96 ADNI-GO whole-blood microarray samples.
• Utilized PyImpetus Markov Blanket selection, Elastic Net with leave-one-out cross-validation (LOOCV), SHAP attribution, and MCP-counter cell-type deconvolution.
• Performed cross-platform mapping to AddNeuroMed dataset (n=329) and external validation in ADNI-2 (n=91).

Main Results:

• Identified a consistent six-probe panel (AQP7, RPS5, CHD2, SNX5, ASS1, uncharacterized chr12q15) across four independent runs.
• Achieved LOOCV Mean Absolute Error (MAE) of 1.388 and R-squared (R²) of 0.247, outperforming the full-probe baseline.
• Confirmed transcriptional origin and validated SNX5 replication in an independent cohort; external validation highlighted the importance of visit-timepoint matching.

Conclusions:

• An exploratory six-probe blood RNA panel reliably predicts 12-month cognitive decline in Alzheimer's disease (AD) with a transcriptional basis.
• Cross-platform replication of SNX5 and successful immune cell correction support the panel's robustness.
• Generalizability is dependent on visit-timepoint matching, underscoring the need for specific conditions for clinical utility and prospective validation.