Endogenous anti-streptavidin antibodies causing erroneous laboratory results more common than anticipated

Figures & data

Figure 1. Flowchart showing included samples, number of samples allocated for investigation of cause of interference and number of samples analysed with alternative methods. TFT: thyroid function tests; HAMA: human anti-mouse antibodies; ASA: anti-streptavidin antibodies.

Table 1. Comparison of Roche TFT immunoassays with alternative methodology.

Analyte	Passing–Bablok regression equation	Pearson correlation coefficient, r	Reference range Roche method	Reference range alternative methods
TSH	Roche = (1.2 × DELFIA) − 0.04	0.99	0.27–4.2 mIU/L	0.5–3.6 mIU/L
FT4	Roche = (1.3 × DELFIA) − 2.4	0.98	12.8–20.4 pmol/L	8.0–21 pmol/L
FT3	Roche = (1.3 × Immulite ) − 1.0	0.93	4.0–6.8 pmol/L	2.8–6.5 pmol/L
FT3	Roche = (0.88 × DELFIA) − 0.35	0.96		3.5–8.3 pmol/L

Figure 2. (A) Roche TSH sandwich assay, simplified. Biotin labelled monoclonal antibody from mouse against human TSH and ruthenium-labelled chimer mouse/human antibody against human TSH form an immune complex with TSH in the patient sample. Streptavidin coated paramagnetic beads (SCB) bind the biotinylated immune complex and attach to a magnet. After a washing procedure, tripropylamine (TPA) is added. TPA emits electrons. Ruthenium is excited and emits photons. The chemiluminescence signal will be proportional to the TSH level in the serum sample. (B) Possible method-dependent interference in the TSH assay. Antibodies against ruthenium or the ruthenium labelled antibody may either decrease or enhance chemiluminescent signal and lead to a false decreased or elevated TSH result. Human antibodies against mice (HAMA) may make a bridge between the two mouse-derived antibodies in the assay and thereby enhance the chemiluminescent signal and lead to a false elevated TSH result. HAMA in high titers may block the formation of an immune complex with TSH, and lead to a false reduced TSH result. Biotin treatment or endogenous antibodies against streptavidin (ASA) in the sample may prevent the binding of the biotinylated immune complex to the streptavidin coated paramagnetic beads. The decreased chemiluminescent signal will lead to a falsely reduced TSH result. (C) Roche competitive FT3 method, simplified. FT3 in the patient sample and biotin labelled FT3 analog compete for ruthenium labelled sheep antibody against human FT3. SCB bind biotin labelled FT3 analog and attach to a magnet. After a washing procedure, tripropylamine (TPA) is added. TPA emits electrons. Ruthenium is excited and emits photons. The chemiluminescence signal will be inversely proportional to the FT3 level in the serum sample. (D) Possible method-dependent interference in the FT3 assay. Antibodies against ruthenium or the ruthenium labelled antibody may either decrease or enhance chemiluminescent signal and lead to a false elevated or decreased FT3 result. Biotin treatment or endogenous ASA in the sample may prevent the binding of biotin labelled FT3 analogue to SCB. The decreased chemiluminescent signal will lead to a false elevated FT3 result.

Table 2. Sera with confirmed interference to ruthenium or the ruthenium labeled FT3 antibody.

Method	Roche ECLIA (results are not method – corrected)					Auto – DELFIA		Siemens – Immulite			Cause of interference
Parameter	TSH mIU/L	FT4 pmol/L	FT3 pmol/L	Anti-TPO kIU/L	TRAb IU/L	TSH mIU/L	FT4 pmol/L	FT3 pmol/L	Anti-TPO kIU/L	TSI IU/L
RR adults Gender – age	0.27–4.2	12.8–20.4	4.0–6.8	<34	<1.8	0.5–3.6	8.0–21.0	2.8–6.5	<35	<0.55
Female – 3 years	0.96	5	<1.5			1.85	14.3	7			Antibodies against ruthenium
Female – 16 years	2.60	28	12.7	45	14.8	2.75	13.4	5.6	12	<0.10	Antibodies against ruthenium
Female – 71 years	5.00	9.7	>50			3.69	11.3	7.07			Antibodies against the ruthenium labelled FT3 antibody

Results of immunoassays for three samples with method-dependent interference caused by antibodies against ruthenium or the ruthenium labeled FT3 antibody.

RR: reference range.

Figure 5. TSH (A), FT4 (B) and FT3 (C) concentrations analysed by the Roche method and alternative methods in 42 samples investigated for the cause of interference and in 51 samples suspected of method dependent interference. Concentrations above and below the measuring limits were given the measuring limit values. Values obtained by the Roche method were adjusted for method dependent differences according to the equations listed in Table 1. *Results above and below the measuring limit. P: patient treated with carbimazol; H: interference probably due to HAMA.

Table 3. Sera with confirmed ASA before and after incubation with SCB.

Patient Gender – age	Female – 33 years			Female – 21 years			Female – 5 years			Female – 40 years			Female – 32 years
	Initial	After SCB	Difference from initial	Initial	After SCB	Difference from initial	Initial	After SCB	Difference from initial	Initial	After SCB	Difference from initial	Initial	After SCB	Difference from initial
Competitive
Estradiol (nmol/L)	NS			0.56	0.09	−84%	NR			0.19	0.06	−67%	NS
Cortisol (nmol/L)	344.3	327.0		1362	884.9	−35%	186.5	159.9		396.3	356.4		318.2	280.8
Progesterone nmol/L)	1.98	1.45	−27%	19.3	<1	> −95%	NR			4.2	0.5	−89%	NS
25(OH)Vit D (nmol/L)	61.0	59.0		99.4	104.0		56.7	59.3		53.4	55.0		49.0	49.7
Vit B12 (pmol/L)	366.9	358.6		320.1	280.6		550.2	467.5	−15%	273.7	265.1		NS
Sandwich
AMH (pmol/L)	9.8	9.3		13.8	24.2	75%	NR			<0.7	<0.7		50.1	49.5
C-peptide (pmol/L)	689.3	880.8	28%	761.9	1026	35%	523.7	626.9	20%	1687	1563		NS
P1NP (µg/L)	40.1	47.7	19%	48.3	55.8	16%	423.5	540.6	28%	13.4	13.2		75.5	85.3
PTH (pmol/L)	NS			NS			<0.13	1.2	>815%	1.9	2.9	51%	<0.13	1.9	>1346%
NTproBNP (ng/L)	NS			416.4	380.3		837.6	785.8		1830	1653		NS
PRL (mIU/L)	144.3	155.7		402.9	391.4		183.8	164.6		232.9	204.9		299.2	288.9
SHBG (nmol/L)	46.3	50.3		121.7	115.3		110.6	98.5		70.2	63.1		NS
Troponin T (ng/L)	NS			232.6	296.4	27%	344.6	320.9		302.8	294.8		NS

Results of immunoassays measured by Roche assays before (Initial) and after incubation with streptavidin coated paramagnetic beads (After SCB). Five sera containing endogenous anti-streptavidin antibodies (ASA) were spiked with sera containing high concentrations of Troponin T and NTproBNP. Interference was considered present (numbers in bold) if a difference from initial concentration of ≥20% was observed after incubation with SCB. A difference from initial concentration of ≥15% is indicated.

25(OH)VitD: 25-hydroxyvitamin D; Vit B₁₂: vitamin B₁₂; AMH: anti-Müllerian hormone; P1NP; n-terminal propeptide of type 1 collagen; PTH: parathyroid hormone; NTproBNP: N-terminal pro-brain natriuretic peptide; PRL: prolactin; SHBG: sex hormone-binding globulin; NS: no specimen; NR: not relevant.

Analyses were performed by Oslo University Hospital on Roche Modular E170, Cobas e601 and Cobas e602.