Sage Journals: Discover world-class research

Abstract

Background

Measurement of antithyroglobulin antibodies (TgAb) is important in patients with differentiated thyroid carcinoma (DTC) with total thyroidectomy. These patients are monitored based on serum thyroglobulin (Tg) levels. TgAb is known to interfere with the measurement of Tg by immunoassay. This study evaluates a new methodology for the measurement of TgAb Alinity Abbott® and the concordance with other methods (first and second generation Advia Centaur Systems Siemens® and Phadia 250 Thermofisher®).

Methods

The technical characteristics of Alinity TgAb measurement methodology were analysed, with imprecision and repeatable studies. In order to assess concordance, a minimum of 69 and a maximum of 76 samples from patients with DTC and total thyroidectomy were processed in parallel by several TgAb measurement methodologies. Agreement rates were determined using kappa statistics. The correlation between the four methods was examined pairwise using McNemar test analysis.

Results

The coefficients of variation (standard deviation as a percentage of the mean % CV) for the Alinity Abbott kit reagent TgAb were within 10% included the functional sensitivity. On the other hand, the concordance analysis with the kappa index concluded substantial agreement. The McNemar test showed a significant difference between Alinity versus Centaur second generation (difference 8.33%, CI 95% 0.68–8.33, p = .0313).

Conclusion

The new methodology for the measurement of TgAb meets the imprecision standards while presenting an adequate concordance agreement with other methodologies available in laboratories. It is important to define the functional sensitivity when reporting results so that they are as reliable as possible.

Keywords

Thyroglobulin antibodies differentiated thyroid carcinoma

Introduction

Thyroglobulin (Tg) is the most sensitive tumour marker for the follow-up of patients with differentiated thyroid carcinoma (DTC) after total thyroidectomy.^1,2 However, its determination can be affected by different types of interference, among which the presence of antithyroglobulin (TgAb) antibodies could lead to an underestimation of its measurement by current chemiluminescence methods.³

In patients with DTC treated and follow-up with Tg, TgAb bind to circulating Tg, decreasing its free concentration, thus providing falsely decreased results. On the one hand, it is essential to measure Tg and TgAb together;⁴ if TgAb are positive could compromise the determination of Tg, and TgAb would be used as tumour markers in the follow-up of these patients.^5,6 On the other hand, it is very important to define two different thresholds for TgAb: firstly, the interfering cut-off for the effects on Tg measurement and, secondly, positive cut-off for the diagnosis of autoimmune thyroid disease.⁷ Interfering cut-off needs to calculate the functional sensitivity of these assays, since this concentration is considered critical for the interpretation of the result.

The prevalence of the presence of these TgAb in patients with DTC has been extensively studied by different authors, which some estimate around 20–30%.^4,5,8

Guidelines require that TgAb should be measured in all specimens submitted for serum Tg testing because the qualitative status of TgAb (positive or negative) determines the risk of interference with Tg assay and serial quantitative TgAb measurements can be used as a surrogate tumour marker.^1,9

Different assays for the measurement of TgAb are marketed, with a non-negligible variability. Some authors confirm that the degree of correlation of TgAb assays is low and the conversion of results from different methods is unreliable.^6,8,10 Therefore, TgAb should be measured by the same assay during the follow-up of DTC patients.⁷ Additionally, the laboratory must verify the data supplied by the manufacturer. For this reason, a comparison and interchangeability study must be carried out.

The purpose of this study is to verify the technical specifications of a new methodology for the measurement of TgAb in Alinity systems (Abbott® Laboratory), especially imprecision and cut-off. In addition, the interchangeability and concordance with other previously implemented methodologies (TgAb first and second generation Advia Centaur® Siemens and Phadia 250 Thermofisher®) will be assessed.

Material and methods

Samples

In order to verify the technical specifications of the new Alinity Abbot kit reagent TgAb, assay imprecision and functional sensitivity are studied. An imprecision study was carried out,¹¹ processing in triplicate for 5 days, two controls supplied by Abbott, positive control (mean 159.8 kUI/L) and negative control (mean 0.5 kUI/L). To complete the measurement range and to assess imprecision, a pool of samples with a concentration of 8.3 kUI/L was prepared. According to the manufacturer’s instructions and from this concentration, different dilutions were made with Calibrator A (TgAb-free), to obtain solutions with concentrations close to the cut-off of the assay (4.11 kUI/L). Furthermore, all these concentrations were measured in triplicate for 5 days to obtain their coefficient of variation and corresponding mean.

For the concordance and interchangeability study, serum samples were collected from patients with a diagnosis of DTC, who had been previously undergone a Total Thyroidectomy (TT), with undetectable Tg levels, and who had been routinely measured both parameters, Tg and TgAb. The samples were selected between January and June 2021 and processed in parallel by the four assays indicated in Laboratory Methods (Assay 1, 2, 3 y 4). All four TgAb assays were standardized against the WHO reference serum 65/93 and performed according to the manufacturer’s instructions. All samples that do not meet these criteria will be excluded. In addition, according to the consensus document of the SEQC (Spanish Society of Clinical Chemistry),¹² a minimum of 40 samples is recommended as long as the measurement range to be studied is covered. Therefore, a minimum of 70 samples and a maximum of 76 samples were selected, depending on the sample volume available.

Laboratory methods

TgAb assays: The methods for the measurement of TgAb were defined as follows:

Assay 1. First generation reagent TgAb in Advia Centaur XP Siemens® equipment based on a chemiluminescent competitive immunoassay. The ADVIA Centaur TgAb assay is a competitive immunoassay using direct chemiluminescent technology. Autoantibody against thyroglobulin in the patient sample competes with polyclonal human TgAb bound to polyclonal goat anti-human antibody covalently coupled to paramagnetic particles in the solid phase for a limited amount of acridinium ester-labelled human thyroglobulin in the Lite Reagent.¹³

Assay 2. Second generation reagent TgAb in Advia Centaur XP Siemens® kit based on a chemiluminescent competitive immunoassay. The ADVIA Centaur aTgII assay is a 1-step, fully automated, dual analyte capture immunoassay using acridinium ester chemiluminescent technology for the detection of autoantibodies to thyroglobulin. This assay uses human thyroglobulin in the lite reagent and solid phase. In the Lite reagent, Tg is labelled with acridinium ester. In the solid phase, Tg is labelled with biotinyl and bound to streptavidin-coated paramagnetic particles. The TgAb from the patient sample binds to the Tg in the Lite reagent and the solid phase, thereby forming a bridge.¹⁴

Assay 3. Phadia 250 Termofisher® kit reagent TgAb, corresponding to a sandwich fluorescent enzyme immunoassay. TgAb are coated with human thyroglobulin antigen. If present in the patient’s sample, antibodies to Tg are attached to the specific antigen. After removal of the unbound antibodies, enzyme-labelled antibodies directed against human IgG antibodies (EliAIgG Conjugate) are added to form antibody-conjugate complexes.¹⁵

Assay 4. Alinity Abbott® kit reagent TgAb with chemiluminescent immunoassay technology for microparticle chemiluminescent immunoassay. This assay is a 2-step immunoassay for the quantitative determination of IgG class antithyroglobulin (TgAb) autoantibodies in human serum and plasma using chemiluminescent microparticle immunoassay (CMIA) technology. The sample, Tg-coated paramagnetic microparticles and assay diluent are combined and incubated. The TgAb present in the sample binds to the Tg-coated microparticles. The mixture is washed. Acridinium-labelled human IgG antibody conjugate is added to create a reaction mixture and incubated. After a wash cycle, the preactivator and activator solutions are added.¹⁶

Tg assay: The measurement of serum thyroglobulin was performed on Roche® Cobas e6000 electrochemiluminescence, ultrasensitive method assay. Undetectable Tg levels are considered <0.1 ng/mL.

Data collection and statistical analysis

All statistics analyses were performed using R commander v 2.6–2 y Medcalc v 11.4.2.0. The correlation between the four methods was examined pairwise using McNemar test analysis, where p < .05 was considered statistically significant. The agreement rates between four assays were determined by using the kappa statistics with 95% confidence intervals (CI) and with k values of <0, 0.01–0.2, 0.21–0.40, 0.41–0.60, 0.61–0.80 and 0.81–0.99, which represent less than chance agreement, slight, fair, moderate, substantial and excellent agreement, respectively.¹⁷

Results

The results obtained from all two-by-two comparisons are summarized in Table 1.

Table 1.

Summary of data obtained with each of the four analytical platforms.

	Alinity		Centaur 1° generation	Centaur 2° generation	Phadia 250
Sample size	76		76	70	75
Lowest value (KUI/L)	0.65		15.0	0.0	2.0
Highest value (KUI/L)	116.41		257.00	14.4	269
Percentile 25 (CI 95%)	1.27 (1.15–1.58)		16.0 (15.0–17.20)	0 (0–0)	6.52 (5.81–7.26)
Percentile 75 (CI 95%)	4.61 (3.23–9.84)		24.15 (22.77–33.40)	0.70 (0–2.93)	15.75 (11.19–20.82)
Mean (CI 95%) (KUI/L)	6.67 (2.99–10.35)		29.29 (21.28–27.31)	1.09 (0.45–1.70)	16.66 (9.29–24.04)
Median (CI 95%) (KUI/L)	2.10 (1.63–2.92)		20.25 (17.79–22.50)	0 (0–0)	8.9 (7.76–10.46)
Cut-off positive (KUI/L)	0.33	4.11	25.0	1.40	12.0
Positive (n)	76	21	16	14	26

Imprecision

The coefficients of variation (standard deviation as percent of mean % CV) for Assay 4, Alinity Abbott kit reagent TgAb, were within 10%, except pool diluted n°5 (15.06%). The coefficients of variation are summarized in Table 2.

Table 2.

Imprecision study in Ramon y Cajal Hospital laboratory. The average obtained was shown with the Coefficients of variation (standard deviation as percent of mean % CV) for Assay 4, Alinity Abbott kit reagent TgAb, in different concentrations.

	Mean ((kUI/L)	Intra-assay CV (%)	Inter-assay/total CV (%)
Positive control	157.8	9.91	9.95
Negative control	0.6	2.44	4.07
Pool	8.3	5.00	5.93
Pool diluted 1	4.3	3.95	5.30
Pool diluted 2	2.3	4.89	5.76
Pool diluted 3	1.3	4.64	7.18
Pool diluted 4	0.8	4.32	8.81
Pool diluted 5	0.5	8.92	15.06

Not all concentrations evaluated in the current study have been tested by the manufacturer, as a result of it only the levels of the positive control and the concentration of Pool Diluted 1 (4, 3 kUI/L) can be compared. The results of the comparison are shown in Table 3.

Table 3.

Imprecision comparative results in manufacturer´s laboratory and Ramon y Cajal Hospital laboratory, using a control Positive provided by the manufacturer and a Panel 1 with concentration 4.67 kUI/L. Panel 1 consists of a pool of samples prepared by the manufacturing laboratory and whose concentration (4.67 kUI/L) is similar to that obtained in the Pool Diluted 1 prepared in our laboratory (4.30 kUI/L).

	Manufacturer laboratory			Ramon y Cajal hospital laboratory
	Mean (kUI/L)	CV Intraserial (%)	CV Intralaboratory (%)	CV intraserial (%)	CV Intralaboratory (%)
Positive control	159.83	1.9	2.1	9.91	9.95
Panel 1	4.67	2.6	2.8	3.95	5.30

Concordance and interchangeability study

The concordance analysis was performed with the kappa index. The TgAb results obtained by each of the four methodologies were dichotomized into ‘positive’ and ‘negative’, according to the cut-off points established in clinical practice:

• Assay 1 First generation reagent TgAb in Advia Centaur XP Siemens: 25 kUI/L. This is the Limit of Quantification calculated by Ramon y Cajal Hospital laboratory. Is defined as the lowest concentration that can be measured with a coefficient of variation less than or equal to 20%.

• Assay 2 Second generation reagent TgAb in Advia Centaur XP Siemens: 1.4 kUI/L. This is the Limit of Quantification calculated by manufacturer laboratory. Is defined as the lowest concentration that can be measured with a coefficient of variation less than or equal to 20%.

• Assay 3 Phadia 250 Termofisher kit reagent TgAb: 12 kUI/L. This is the Limit of Quantification calculated by manufacturer laboratory. Is defined as the lowest concentration that can be measured with a coefficient of variation less than or equal to 20%.

• Assay 4 Alinity Abbott kit reagent TgAb. For this method, a concordance study was carried out using two cut-off points. Firstly, 4.11 kUI/L corresponds to the concentration within the 97.5% percentile of the population studied by manufacturer. Secondly, 0.33 kUI/Lis the limit of quantification calculated on more than 60 replicate samples and whose coefficient of variation does not exceed 20%. Kappa values have been calculated by using both cut-off points.

The McNemar test was used to compare the different methodologies versus the new Alinity test. The results obtained are expressed as a percentage of difference with 95% CI and are shown in Table 4.

Table 4.

The McNemar test showed the difference between paired methodologies. The table provides the difference between proportions (expressed as a percentage) with 95% CI. When the p-value is less than the conventional 0.05, there is a significant difference between the two methodologies.

Methodologies	Difference (%)	CI 95%	p-value
Centaur 1° generation versus Alinity	6.67	−3.22 to 12.9	0.2266
Centaur 2° generation versus Alinity	8.33	0.68–8.33	0.0313
Phadia 250 vs Alinity	6.58	−1.45 to 9.14	0.1250

Discussion

The high prevalence of TgAb in patients with CDT (8) makes their measurement in the follow-up of these patients indisputable. Therefore, when changing the measurement methodology, it is necessary to conduct a study of both, imprecision and interchangeability, with respect to the technique in use. The Alinity reagent for the measurement of TgAb has an acceptable inaccuracy at all concentrations tested, including those close to the cut-off point indicated by the manufacturer (4.11 kUI/L). In a large number of cases, these cut-off points are optimized for the diagnosis of autoimmune thyroid disease, but not to detect interference in the measurement of thyroglobulin, and they are usually especially high.⁵

In this current case, the manufacturer indicates a Limit of Detection LoD = 0.11 kUI/L and a Limit of Quantification LoQ = 0.33 kUI/L, with a cut-off point of 4.11 kUI/L. On the one hand, the imprecision calculated in Ramon y Cajal Hospital laboratory for the lowest concentration (0.5 kUI/L) was less than 20%, which could indicate that the functional sensitivity could be lower than 4.11 kUI/L. On the other hand, if a concordance study is performed by using the cut-off point of 0.33 kUI/L, almost all the patients would have the presence of antibodies, which would not comply with the results of other methodologies (kappa index = 0.00 (Table 5). Thus, using the cut-off point of 4.11 kUI/L, it is found a concordance of 0.61, 0.71 and 0.78 with Centaur second generation, Centaur first generation and Phadia, respectively, which indicates a substantial agreement.

Table 5.

Kappa index for the different methodologies and the different cut-off points in Alinity.

Methodologies	Kappa index [CI 95%] cut-off point 4.11 kUI/L	Kappa index [CI 95%] cut-off point 0.33 kUI/L
Centaur 1° generation versus Alinity	0.61 [0.40–0.81]	0.00
Centaur 2° generation versus Alinity	0.71 [0.52–0.91]	0.00
Phadia 250 vs Alinity	0.78 [0.63–0.93]	0.00

The McNemar test showed only one significant difference between Alinity versus Centaur second generation (difference 8.33%, IC95% 0.68–8.33, p = .0313)

Conclusion

Overall, despite the differences between the methodologies provided by various authors,^10,18,19 the concordance of this new Alinity assay is acceptable as long as the cut-off point of 4.11 kUI/L is used. In view of the calculated data on the inaccuracy of this TgAb assay, further studies are considered to assess a possible lower functional sensitivity than that detailed by the manufacturer.

Our study had a number of limitations: A very limited number of samples; in order to reduce the degree of uncertainty of our results, it would be convenient to increase the sample size. On the other hand, our study showed concordance between the results of Alinity thyroglobulin antibody and the other methodologies, but could not ensure the presence of interference by thyroglobulin antibodies in the measurement of thyroglobulin, which would have required the measurement of this parameter by mass spectrometry.

Footnotes

Acknowledgements

The support of all the participants who have agreed to participate in this study is greatly appreciated.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The study was evaluated and approved by the Ethical committee of Hospital Ramón y Cajal (approval date: 14/06/2021 Acta 415).

Guarantor

AMGC.

Contributorship

All authors contributed to the study conception and design. The first draft of the manuscript was written by AMGC and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

ORCID iD

Ana M Garcia Cano

References

Latrofa

Ricci

Montanelli

, et al. Thyroglobulin autoantibodies in patients with papillary thyroid carcinoma: comparison of different assays and evaluation of causes of discrepancies. J Clin Endocrinol Metab 2012; 97(11): 3974–3982.

Pacini

Schlumberger

Dralle

, et al. European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur J Endocrinol 2006; 154(6): 787–803.

Spencer

Bergoglio

Kazarosyan

, et al. Clinical impact of thyroglobulin (Tg) and Tg autoantibody method differences on the management of patients with differentiated thyroid carcinomas. J Clin Endocrinol Metab 2005; 90(10): 5566–5575.

Haugen

Alexander

Bible

, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: The American thyroid association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016; 26(1): 1–133.

Kim

Yoon

Kim

, et al. Change of serum antithyroglobulin antibody levels is useful for prediction of clinical recurrence in thyroglobulin-negative patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab 2008; 93(12): 4683–4689.

Spencer

Petrovic

Fatemi

. Current Thyroglobulin Autoantibody (TgAb) assays often fail to detect interfering TgAb that can result in the reporting of falsely low/undetectable serum Tg IMA values for patients with differentiated thyroid cancer. J Clin Endocrinol Metab 2011; 96(5): 1283–1291.

Latrofa

Ricci

Sisti

, et al. Significance of low levels of thyroglobulin autoantibodies associated with undetectable thyroglobulin after thyroidectomy for differentiated thyroid carcinoma. Thyroid 2016; 26(6): 798–806.

Spencer

Takeuchi

Kazarosyan

, et al. Serum thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement, and prognostic significance in patients with differentiated thyroid carcinoma. 1998; 83(4): 7.

Cooper

Doherty

Haugen

, et al. Revised American thyroid association management guidelines for patients with thyroid nodules and differentiated thyroid cancer: the American thyroid association (ATA) Guidelines taskforce on thyroid nodules and differentiated thyroid cancer. Thyroid 2009; 19(11): 1167–1214.

10.

Taylor

Parkington

Bradbury

, et al. Concordance between thyroglobulin antibody assays. Ann Clin Biochem Int J Lab Med 2011; 48(4): 367–369.

11.

Reverter

Juliani

Poblador

, et al. Comisión de Metrología y Sistemas Analíticos, 6.

12.

Prada de Medio

Blázquez Sánchez

Perich Alsina

, et al. Verificación de la intercambiabilidad de resultados entre equipos duplicados. Rev Lab Clínico 2014; 7(1): 17–24.

13.

Anti-Tg (aTG) Advia Centaur XP [Internet]. https://doclib.siemens-healthineers.com/rest/v1/view?document-id=712911 (accessed 14 April 2021).

14.

Anti-Thyroglobulin II (aTgII) Advia Centaur XP [Internet]. https://doclib.siemens-healthineers.com/rest/v1/view?document-id=690709 (accessed 14 April 2021).

15.

Anti-TG Phadia Thermofisher.pdf [Internet]. https://dfu.phadia.com/Data/Pdf/5fa52ed889c23208405eb478.pdf (accessed 19 May 2022).

16.

Technical Library | Abbott core laboratory [Internet]. https://www.corelaboratory.abbott/tl/ous/technical-library.html (accessed 1 July 2021).

17.

Viera

Garrett

. Understanding interobserver agreement: the kappa statistic. Fam Med: 4 2005; 37(5): 360–363.

18.

Nygaard

Bentzen

Laurberg

, et al. Large discrepancy in the results of sensitive measurements of thyroglobulin antibodies in the follow-up on thyroid cancer: a diagnostic dilemma. Eur Thyroid J 2012; 1(3): 193–197.

19.

Spencer

Fatemi

. Thyroglobulin antibody (TgAb) methods – Strengths, pitfalls and clinical utility for monitoring TgAb-positive patients with differentiated thyroid cancer. Best Pract Res Clin Endocrinol Metab 2013; 27(5): 701–712.

Evaluation of the new Abbott®Alinity assay for the measurement of thyroglobulin antibodies in patients with differentiated thyroid carcinoma

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Material and methods

Samples

Laboratory methods

Data collection and statistical analysis

Results

Imprecision

Concordance and interchangeability study

Discussion

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

Ethical approval

Guarantor

Contributorship

ORCID iD

References