Fig. 3: Clinical performance of IP-MS plasma Aβ1-42/Aβ1-40 ratio in the IGNITE cohort of cognitively normal older adults (n = 317 biological replicates). | Nature Communications

Fig. 3: Clinical performance of IP-MS plasma Aβ1-42/Aβ1-40 ratio in the IGNITE cohort of cognitively normal older adults (n = 317 biological replicates).

From: Streamlined resource-efficient plasma amyloid-beta mass spectrometry assay has improved biomarker performance in preclinical Alzheimer’s disease

Fig. 3: Clinical performance of IP-MS plasma Aβ1-42/Aβ1-40 ratio in the IGNITE cohort of cognitively normal older adults (n = 317 biological replicates).

A, B Box-and-whisker plots depict the separation of Aβ PET-positive and Aβ PET-negative groups using the original (PAβ V1.0) and the improved (PAβ V2.0 IS) assays, respectively. The plots show the median, 25th–75th percentiles (box), and whiskers extending to 1.5 × IQR, with individual data points overlaid. Group differences were evaluated using the two-sided Wilcoxon Rank Sum test, with p-values provided. The preliminarily derived optimal cutoff values for each assay, determined by the Youden method, are marked in red. C ROC curves with AUC values and confidence intervals showing the classification accuracies of the Aβ1-42/Aβ1-40 ratio measured with either the IP-MS assay to distinguish Aβ-PET positive vs. negative groups (positivity determined as a CL value of > 24). To better understand plasma Aβ1-42/Aβ1-40 ratio changes according to stepwise brain Aβ pathology, the same IGNITE cohort dataset was further divided into three CL groups (PET-negative, transition, PET-positive). The Box-and-whisker plots for the V1.0 assay (D), V2.0 assay (E) were generated using the same statistical tests and plotting format as in panels (A and B). The ROC curves to identify an abnormal Aβ PET scan with the transition (i.e., intermediate) group excluded (F) are also shown.

Back to article page