Novel CSF β-synuclein-specific assays signal early synaptic degeneration in Alzheimer’s disease

Bayoumy, Sherif; Goossens, Julie; De Rocker, Charlotte; Sie, Senna Y.; Barrett, Nolan J.; van der Flier, Wiesje M.; Teunissen, Charlotte E.; Vanmechelen, Eugeen; Verberk, Inge M. W.

doi:10.1186/s13195-025-01716-8

Alzheimer's Research & Therapy

Table 1 Analytical characteristics of the β-syn ELISAs

From: Novel CSF β-synuclein-specific assays signal early synaptic degeneration in Alzheimer’s disease

		N-terminus β-syn			Mid-region β-syn			C-terminus β-syn
Assay	Platform	ELISA			ELISA			ELISA
	Status	Prototype			Prototype			Prototype
	Developers	Ulm University Hospital			Amsterdam UMC / ADx NeuroSciences			Amsterdam UMC / ADx NeuroSciences
	Biofluid	CSF			CSF			CSF
	Fold sample dilution	4			4			4
Calibration curve	Type	Recombinant			Recombinant			Recombinant
	No. of calibrator points	10			10			10
	Range, pg/mL	4–1000			4–1000			4–1000
	Curve fit	1/y²-weighted 5PL			1/y²-weighted 5PL			1/y²-weighted 5PL
Antibodies	Capture	EP1646Y			ADx- β-syn1			ADx- β-syn2
Antibodies	Detector	EP1537Y			EP1537Y			EP1537Y
Clinical samples measurements (Amsterdam Dementia Cohort)	Number	160			160			160
	Range concentration, pg/mL	60.0–976.0			13.2–252.8			22.7–636.9
	Range, CV%	0–33.0%			0–26.0%			0–32.6%
	Average CV%	4.9%			3.9%			7.5%
	n measured < LLOQ	0			0			0
	n measured > 20%CV	2			1			11
Analytical validation results
Sensitivity	Functional LLoD, pg/mL	2.15			0.74			2.42
Concentrations of QC panels	QC1: high, pg/mL	NA			57.8			23.8
	QC2: intermediate, pg/mL				90.3			179.8
	QC3: low, pg/mL				154.0			278.6
Precision of QCs	Average Intra-assay %CV	NA			7.2			16.7
Precision of QCs	Average Inter-assay %CV	NA			31.8			17.4
Parallelism	Average slope of samples	NA			0.78			0.70
	Range of slopes of samples				0.87–70			0.64–0.76
	Average slope of calibrator				0.82			0.67
	Parallelism, %				94%			105%
Dilution linearity	Spiked concentration, pg/ml				1000			1000
		Df (x)	Mean %L		Mean %L			Mean %L
	Dilution linearity factor with mean %Linearity	1	NA		-			-
		4			144			58
		16			103			99
		64			139			96
		256			87			73
Recovery	Spiked concentration (pg/mL) With mean %Recovery	Spike	Mean %R		Spike (pg/mL)	Mean %R		Spike (pg/mL)	Mean %R
		NA			500	67		500	63
					125	64		250	64
					32	67		125	57
Stability	Freeze/ thaw cycles	Average response (%)		Standard deviation	Average normalized response (%)		Standard deviation	Average response (%)	Standard deviation
	0 F/T	NA			100.0		2	100.0	15
	1 F/T	NA			100.0		2	100.0	10
	2 F/T	NA			100.0		4	105.2	15
	3 F/T	NA			100.0		2	100.0	11

The analytical performance of the ELISAs was assessed based on sensitivity, precision, parallelism, dilution linearity and recovery. The sensitivity was determined by calculating the detection limits of each assay. We utilized a commercially available recombinant truncated human β-syn (residues 1–122, UniProt: P37840), lacking the 12 C-terminal amino acids of the full-length 134-residue protein. The recombinant protein was untagged. The functional LLoD was calculated as the mean signal of 16 blanks plus 10 times the SD multiplied by the dilution factor of four times. Quality control (QC) samples were made of remnant CSF samples. The mean intra- and inter-assay variation was determined by measuring the QC panels over three independent runs on three days. The 160 clinical samples were measured in duplicate. Precision was evaluated by calculating the intra-assay and inter-assay coefficients of variation (CV%) of these measurements. Parallelism was evaluated to ensure the consistency of assay performance across different sample dilutions, and dilution linearity confirmed the linear response of the assays to diluted samples. For parallelism, four samples were measured after being serially diluted four times 2-fold, starting with a 2-fold dilution that reached 32a-fold). For dilution linearity, three samples were spiked with 1000 pg/mL recombinant β-syn and subsequently measured undiluted and diluted 256-fold. Recovery tests were used to assess the accuracy of the assays in quantifying known amounts of β-syn. PL: polynomial; LLoD: lowest limit of detection; QC: quality control; CV: coefficient of variation; %L: % linearity; %R: % recovery, F/T: freeze / thaw cycles

Back to article page

ISSN: 1758-9193

Contact us

General enquiries: journalsubmissions@springernature.com