Proteomics: Plasma Panel Flags Mild Cognitive Impairment
Most people with amnestic mild cognitive impairment (aMCI) are first seen in community clinics, where amyloid PET is rarely an option and plasma p-tau217 turns positive only once pathology has built up. Shan and colleagues tested whether deep plasma proteomics could see earlier. From 100 microliters of plasma per person in 84 adults, they identified 4,268 proteins by data-independent acquisition (DIA) mass spectrometry and distilled a 12-protein panel. It beat Simoa amyloid and tau markers handily.
Key Takeaways
Nanoparticle bead enrichment lets plasma proteomics reach more than 4,000 proteins from 100 microliters, deep enough to see low-abundance signals of early cognitive decline.
Many patients with clinically diagnosed amnestic MCI show no measurable amyloid or tau change in blood yet, so pathology-based plasma assays miss them.
A 12-protein plasma panel for amnestic MCI looks promising but comes from one 84-person cohort and still needs independent validation with absolute quantification.
How Deep Plasma Proteomics Reached Low-Abundance Proteins
Albumin and immunoglobulins dominate plasma and bury the tissue-derived proteins that early disease releases. Here, nanoparticle magnetic beads captured proteins through noncovalent binding, an idea related to the protein-corona enrichment of Blume and colleagues (2020), ahead of DIA protein profiling on an Orbitrap Exploris 480. Controls (n = 44) were cognitively normal and p-tau217 negative. The 53 proteins that were both differentially abundant and inside a disease-linked coexpression module fed a support vector machine (SVM) evaluated on a held-out 25% test set.
Key Findings
Depth without obvious separation: PCA did not split the groups, but 292 proteins differed significantly, 257 of them lower in aMCI.
Twelve proteins carried the signal: The SVM panel, led by GNG10, SVIP, and LSM3, reached an AUC of 0.992 in training and 0.933 in the 21-sample test set.
Simoa markers missed most cases: Plasma amyloid-beta 42/40, p-tau217, and p-tau181 correctly flagged 4 of 40 aMCI patients (24.4% overall accuracy) versus 93.9% for the panel.
A trio that falls short alone: The three top-weighted proteins reached an AUC of 0.884 without the other nine.
Figure 1. Deep plasma proteome of cognitively normal (CN) and aMCI participants. Panel A shows overlapping groups by principal component analysis, while Panel B shows clear separation by PLS-DA. Panel C is a volcano plot marking down-regulated (blue) and up-regulated (red) proteins at a fold change of at least 1.5 in either direction and adjusted p below 0.05. Adapted from Shan et al. (2026), Journal of Proteome Research.
Why Amyloid and Tau Assays Fell Short
Only 40% of aMCI patients were positive for any Simoa marker, which the authors attribute to clinical aMCI including pre-amyloid stages that blood amyloid and tau assays are not built to detect. The comparison is not like for like, though. A classifier trained on clinical labels will naturally beat a pathology assay scored against those same labels. Controls were also nearly six years younger (59.6 vs 65.3 years) and better educated, so part of the proteomic signal may track age rather than cognition.
From DIA Discovery to a Targeted PRM Assay
A parallel reaction monitoring (PRM) method covering 27 peptides detected all 12 proteins in plasma, and most kept their discovery-phase direction of change. Peptide quantification was relative, since stable isotope-labeled standards were not yet available. Biologically, the panel mixes synaptic (NPTN, KIDINS220), metabolic (GFPT2, MT-ND1, GSTZ1), and autophagy-related (SVIP, SEL1L) proteins; GHDC is essentially unannotated.
Planning the Handoff From Discovery to Assay
Plasma depth only pays off if the winning peptides can later be measured against heavy-labeled standards in a routine lab. We hit the same pressure point when scoping plasma studies at Dalton, and it's why we push for age-matched controls and a draft PRM transition list before the first discovery injection. Building that handoff into LC-MS proteomics early is the step most multi-omics biomarker discovery programs underestimate.
Frequently Asked Questions
What is plasma proteomics used for in biomarker discovery?
Plasma proteomics measures hundreds to thousands of circulating proteins at once, usually by mass spectrometry, to find proteins that track disease, drug response, or risk. Promising hits then move into targeted assays and larger cohorts for validation.
Can a blood test detect mild cognitive impairment before amyloid or tau changes?
In this 84-person study, a 12-protein plasma panel identified clinically diagnosed amnestic MCI with 93.9% accuracy, including patients with negative Simoa results. Replication in larger, age-matched cohorts is still needed.
How much plasma does a proteomics biomarker study need?
This study used 100 microliters per person for bead enrichment, which most biobank aliquots can spare. Discovery cohorts of 50 to 100 samples can nominate candidates, but validation needs larger independent cohorts and a targeted assay such as PRM.
Conclusion
It's now believable that bead-enriched plasma proteomics picks up something in clinically diagnosed aMCI that amyloid and tau assays don't, at least in one Shenzhen cohort. Whether that signal is Alzheimer's-specific, survives age-matched controls, or transfers to another site hasn't been shown. Treat the panel as a candidate list for multicenter PRM validation, not a diagnostic.
Related Reading
See how we run these analyses in one lab: Dalton's multi-omics CRO services.
Citation
Shan, H., Lu, G., Yuan, Y., Dong, Q., Fang, C., Yu, K., Chen, L., Wei, Z., Chen, H., Hu, L., Wu, T., Tao, S., Chen, Y., Luo, J., Qi, Y., Hu, J., Cheng, G., Fan, X., & Yin, Y. (2026). Deep plasma proteomics-based diagnostic panel for early detection of amnestic mild cognitive impairment. Journal of Proteome Research, 25(8), 4237-4250. https://doi.org/10.1021/acs.jproteome.6c00231
Note
This blog post summarizes findings from the above-cited research. Figures are adapted from the original publication. For full details, please refer to the source article.
By Seungjun Yeo, CEO at Dalton Bioanalytics. Specializing in multi-omics mass spectrometry for drug discovery and biomarker research.