# Comparison of NULISAseq and SIMOA methods in plasma biomarker analysis in two memory clinic and research cohorts

## TL;DR

This study compares two plasma biomarker analysis methods, NULISAseq and SIMOA, in Alzheimer's disease diagnosis, finding that NULISAseq shows more distinct group differences.

## Contribution

The study provides a direct comparison of newly commercialized NULISAseq with established SIMOA methods in Alzheimer's biomarker analysis.

## Key findings

- NULISAseq and SIMOA showed strong agreement for most biomarkers (Spearman's rho >= 0.8).
- NULISAseq produced more distinct group differences in pTau217 and GFAP levels between MCI Aβ+ and MCI Aβ-.
- pTau231 and pTau181 showed more pronounced differences with NULISAseq in the AD spectrum.

## Abstract

The diagnostic work‐up of Alzheimer's disease (AD) requires reliable and accurate diagnostic biomarkers to aid clinicians in making the correct diagnosis. Plasma biomarker analysis methods are emerging as cost‐effective and possibly reliable tools. The NULISAseq (Alamarbio) CNS panel has recently been commercialized and has not yet been extensively tested with previously used assays. We aimed to compare a selection of biomarkers assayed with NULISAseq against more established SIMOA methods in a memory clinic and a research cohort from Karolinska.

A total of 176 subjects from a memory clinic cohort and a research cohort were included in this study. The clinical cohort from the Karolinska Hospital Memory Clinic included 126 patients (70W/56M, mean age 65±8 years, range 42‐86), who, after extensive investigations (including Amyloid‐PET imaging), were assigned to the following diagnostic groups: MCI Aβ+(n = 19), MCI Aβ‐(n = 29), AD(n = 51), other dementia(n = 23) and no dementia(CU)(n = 4). Further details on the general characteristics are presented in Table 1. The research cohort (ROADAD) included 30 men and 20 women (mean age 64±13 years, range 28‐86). For all 176 cases, plasma samples, taken during the investigations, were processed with NULISAseq to obtain pTau217, pTau231, pTau181, GFAP, NFL, Aβ42 and Aβ40 levels. Additionally, SIMOA assays (Quanterix‐N4PE and AlzPath) were also applied. SIMOA measurements for pTau231 and pTau181 were not available for the ROADAD cohort.

In the entire sample, the two methods generally showed strong agreement (Figure 1) (Spearman's rho>=0.8 for pTau isoforms and GFAP). pTau217 vs pTau231 had rho=0.7 for SIMOA, while rho=0.86 for NULISA. When the two panels of measurements were compared in the clinical cohort to evaluate differences between diagnoses(Figure 2), pTau217 and GFAP levels were significantly different between MCI Aβ‐ and MCI Aβ+ regardless of the method. Comparing the diagnostic groups, pTau231 was more similar to pTau217 with the NULISA than with SIMOA method. pTau181 showed more pronounced differences within the AD spectrum with NULISA than SIMOA.

Both methods show similar results and trends, but NULISAseq produced more distinct group differences. It remains to be understood if the closer relation between pTau217 and pTau231 with NULISA is a benefit or a disadvantage compared with SIMOA methods.