Table 4.

Comparison between general EVs [42] vs speculative CNS-enriched EVs for diagnosing patients with Parkinson’s disease (PD) from healthy controls using a bivariate and hierarchal summary receiver operating characteristics (HSROC) model. Reproduced from Taha et al. [42]

EV source	Mean sensitivity (95% CI)	Mean specificity (95% CI)	Pooled AUC (partial AUC))	Mean DOR ± SE (95% CI)
General EVs [42]	84.4% (77.7–90.7%)	79.1% (72.5–84.0%)	0.852 (0.672)	21.6 ± 1.3 (12.0–38.9)
CNS-enriched EVs	72.5% (64.4–79.3%)	75.9% (69.2–81.4%)	0.800 (0.692)	8.3 ± 2.1 (5.0–13.8)

The sensitivity, specificity, pooled area under the curve (AUC) and partial AUC, focusing on a specific range of false-positive rates (FPR), are obtained using the bivariate model. The diagnostic odds ratio (DOR) is obtained from the HSROC model

EV extracellular vesicles; CNS central nervous system; SE standard error