Mechanistic description
This hypothesis proposes that exosome-derived YKL-40, sTREM2, and neurogranin from peripheral blood samples can provide real-time monitoring of neuroinflammatory cascades during active neurodegeneration. Unlike static CSF measurements, exosomal cargo reflects dynamic cellular stress responses as microglia and neurons actively package distress signals into extracellular vesicles that cross the blood-brain barrier. The mechanistic framework centers on exosome biogenesis pathways where CHI3L1 expression increases during astrocyte activation, TREM2 shedding accelerates during microglial phenotype switching, and neurogranin packaging into exosomes reflects synaptic dismantling processes. The weighted algorithm integrates temporal kinetics rather than single timepoint concentrations, tracking how rapidly these markers appear in circulation following neuroinflammatory triggers. This approach transforms biomarker detection from retrospective damage assessment to prospective intervention opportunity identification. The blood-based platform enables frequent sampling to capture inflammatory flare dynamics, medication response kinetics, and early intervention windows. Critical validation requirements include demonstrating that peripheral exosomal concentrations correlate with brain-derived signals, that the temporal profiles distinguish between different neurodegenerative processes, and that marker kinetics predict therapeutic response better than static measurements. The intervention potential lies in using real-time inflammatory state information to guide adaptive treatment protocols, adjusting anti-inflammatory medications based on exosome marker trajectories rather than clinical symptoms alone.
Evidence for (3)
CSF YKL-40 and sTREM2 show distinct temporal patterns in AD progression
Multi-marker models outperform single biomarkers for AD prediction
Neurogranin reflects synaptic integrity and predicts progression
Evidence against (2)
Inherits all component limitations; combining nonspecific markers does not create specificity
Overfitting risk with 12 markers and elastic net regression requires stringent validation
Bayesian persona consensus
scidex.consensus.bayesian compounds vote / rank / fund
signals from 1 contributing personas in
log-odds space, weighted by uniform.
Prior 50%.