Kapogiannis · FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2015 · case-control and longitudinal biomarker study · n=84 patients (26 AD, 20 DM2, 16 FTD, 22 longitudinal AD) plus matched controls

Dysfunctionally phosphorylated type 1 insulin receptor substrate in neural-derived blood exosomes of preclinical Alzheimer's disease.

Cited 352 times in the scientific literature.

Level 4 - case-series / case-control

Case-control biomarker study with a small longitudinal case cohort.

PubMed 25342129 · doi:10.1096/fj.14-262048 · record verified 2026-08-30

What was done

Researchers isolated neural-enriched plasma exosomes using immunoabsorption from patients with Alzheimer's disease (AD, n=26), type 2 diabetes (DM2, n=20), frontotemporal dementia (FTD, n=16), and matched controls. Longitudinal samples were also examined from 22 individuals at two time points: when cognitively normal and 1 to 10 years later upon AD diagnosis. ELISA was used to quantify phosphorylated serine 312-IRS-1, P-pan-tyrosine-IRS-1, and their ratio (termed insulin resistance factor, R), followed by stepwise discriminant modeling.

What was found

Exosomal levels of P-serine 312-IRS-1, P-pan-tyrosine-IRS-1, and R were significantly altered in AD and DM2 compared to controls, with R significantly higher in AD than in DM2 or FTD. Discriminant modeling correctly classified 100% of AD, 97.5% of DM2, and 84% of FTD cases. In the longitudinal cohort of 22 patients, marker alterations and R elevations were significantly different from controls 1 to 10 years prior to clinical AD diagnosis. Specific biomarker values and confidence intervals were not reported in the abstract.

Why it matters

The findings show that brain-derived insulin resistance signaling can be measured noninvasively in peripheral blood, suggesting a potential biomarker for identifying preclinical Alzheimer's disease years before symptom onset.

Limits

The study is limited by small sample sizes within each subgroup and a case-control design without an independent validation cohort. Total control numbers, absolute marker concentrations, and diagnostic test characteristics (such as sensitivity and specificity confidence intervals) were not reported in the abstract.

Cited by