Targeted proteomic biomarker profiling using NULISA in a cohort enriched with risk for Alzheimer's disease and related dementias

Abstract

Introduction: Targeted proteomic assays may be useful for diagnosing and staging Alzheimer's disease and related dementias (ADRD). We evaluated the performance of a 120-marker central nervous system (CNS) NUcleic acid Linked Immuno-Sandwich Assay (NULISA) panel in samples spanning the Alzheimer's disease (AD) spectrum.

Methods: Cross-sectional plasma samples (n = 252) were analyzed using NULISAseq CNS panel from Alamar Biosciences. Receiver-operating characteristic (ROC) analyses demonstrated the accuracy from NULISAseq-tau phosphorylated at threonine 217 (pTau217) in detecting amyloid (A) and tau (T) positron emission tomography (PET) positivity. Differentially expressed proteins were identified using volcano plots.

Results: NULISAseq-pTau217 accurately classified A/T PET status with ROC areas under the curve of 0.92/0.86; pTau217 was upregulated in A+, T+, and impaired groups with log₂-fold changes of 1.21, 0.57, and 4.63, respectively, compared to A-. Of interest, TAR DNA-binding protein 43 (TDP-43) phosphorylated at serine 409 (pTDP43-409) was also upregulated in the impaired group and correlated with declining hippocampal volume and cognitive trajectories.

Discussion: This study shows the potential of a targeted proteomics panel for characterizing brain changes pertinent to ADRD. The promising pTDP43-409 findings require further replication.

Highlights: The NULISAseq pTau217 assay was comparable to the Simoa pTau217 assay, both utilizing the ALZpath antibody, in detecting amyloid positron emission tomography (PET) positivity, each with areas under the curve greater than 90%. Nineteen proteins were differentially expressed in participants with mild cognitive impairment (MCI) compared to those who were unimpaired. Markers of non-AD proteinopathies such as pTDP43-409, oligomeric alpha-synuclein, and huntingtin (HTT), were among those upregulated in MCI. High levels of plasma pTDP43-409 were associated with worsening hippocampal atrophy and cognitive decline, clinical indicators of limbic-predominant age-related TDP-43 encephalopathy (LATE), compared to those with low pTDP43-409.

Keywords: Alzheimer's disease and related dementias; NULISA; pTDP43; pTau217; plasma biomarkers.

FIGURE 1

Performance comparison between NULISAseq and Simoa for subset of n = 218 with both plasma assays and amyloid PET visual read. (A) Correlation between NULISAseq 2NPQ and Simoa pg/mL concentrations. Dashed lines indicate the Youden's cutoff for the NULISAseq (vertical line) and Simoa (horizontal line) assays, derived from their respective ROCs for amyloid visual read (red) and tau visual read (purple). (B, C) ROC curves from NULISAseq (blue) and Simoa (yellow), based on amyloid visual read (B) and tau visual read (C). Aβ, amyloid beta; NPQ, NULISA protein quantification; NULISA, NUcleic acid Linked Immuno‐Sandwich Assay; PET, positron emission tomography; ROC, receiver‐operating characteristic; Simoa, Single molecule array; VR, visual read.

FIGURE 2

Proteinomics analysis across groups of interest. We performed comparisons between binary groups of interest across all biomarkers in the CNS120 panel. We only considered as analytes of interest those that showed a log₂ fold change between categories of at least 0.5 (vertical dashed lines), and that maintained a p‐value < 0.05 after FDR correction (horizontal dashed line). Analytes meeting these criteria for each of the tests are highlighted in red. We stratified participants according to four different variables: (A) Cognitive diagnosis status (CU vs MCI); (B) amyloid status (A−/T− vs A+/T−); (C) tau status (A+/T− vs A+/T+); (D) synSAA status (synSAA− vs synSAA+). See Table S1 for details about the proteins differentially expressed in each comparison. CU, cognitively unimpaired; FDR, false discover rate; MCI, mild cognitive impairment.

FIGURE 3

Distribution of non‐AD biomarkers by categories of interest. We first compared plasma biomarker expression in participants with results from the CSF Amprion seed amplification assay (synSAA), classified as synSAA− (n = 68) and synSAA+(n = 17). (A) plasma Oligo‐SNCA and (B) plasma monomeric alpha‐synuclein (Mono‐SNCA). Given the results from the proteinomics analysis identifying expression of plasma biomarkers between CU and MCI individuals, we additionally looked at the distribution of plasma pTDP43 (C) and plasma TARDBP (D) among cognitive diagnosis statuses. We additionally show the A/T PET status in distinct colors in all these panels. CSF, cerebrospinal fluid; CU, cognitively unimpaired; MCI, mild cognitive impairment; PET, positron emission tomography; TARDBP, transactive response DNA‐binding protein; VR, visual read.

FIGURE 4

Estimated and observed longitudinal trajectories among pTDP43 groups. (A, B) Predicted slopes across time in each of the pTDP43 groups depending on their A/T status, for the PACC3‐TMTB and memory composite scores, respectively. (C) Predicted slopes for hippocampal volume across time in each of the pTDP43 groups depending on their A/T status. Solid dark lines represent the marginal effects across time for each pTDP43 group; shaded areas correspond to the SE. Dots and lines in the background represent the observed values from everyone in the sample. A, amyloid; PACC‐3‐TMTB, Preclinical Alzheimer's Cognitive Composite/Trail Making Test B; T, Tau.