Retrospective Analysis of a New Antigen Assay to Detect Anti-Transglutaminase Antibodies for Diagnosis and Monitoring of Celiac Disease

Abstract

BACKGROUND

The improved diagnosis and monitoring of celiac disease (CD) requires increasingly sensitive and specific serological markers. The development of new antigens (neo-epitopes) able to detect different sets of anti-transglutaminase antibodies requires studies to demonstrate their advantages.

OBJECTIVE

The goal of this study was to evaluate a new antigen for the detection of anti-tTG antibodies that combines tissue transglutaminase (tTG) and gliadin peptides. This protein complex occurs under physiological conditions in vivo at the time of deamination.

PATIENTS AND METHODS

In this retrospective study, serum samples collected from patients from January to December 2018 were analyzed. Results of the determinations for neo-epitope anti-tTG antibodies (manufacturer's cut-off values: negative <12 U/mL, indeterminate 12-18 U/mL; positive >18 U/mL) were recorded along with age, sex, and duodenal biopsy results. Neo-epitope assay results were correlated with clinical and laboratory data and final CD diagnoses, using the manufacturer's cut-off value and a proposed alternate cut-off value.

RESULTS

A total of 3820 neo-epitope anti-tTg determinations were analyzed. The percentage of values in the indeterminate zone was low (3.6%), with very few values (0.9%) at the exact manufacturer's cut-off limit (18 U/mL). This indicated a good resolution capacity for positives and negatives. The percentage of false positives was lower compared to a previous study (2013 study: 3.6%, vs current study (2018): 0.5%). A new positivity cut-off value was proposed: 20 U/mL, which increased the positive predictive value of the test.

CONCLUSIONS

The neo-epitope assay is a more precise tool than recombinant human transglutaminase assay for the diagnosis and monitoring of celiac patients. It has a greater ability to resolve positive and negative results (“minimum indeterminate area”). Its greater sensitivity could detect situations (presentation and/or dietary transgression) where conventional techniques show negative or weakly positive serology.

Keywords

Celiac disease anti-transglutaminase autoantibodies autoimmunity

Introduction

The multisystemic nature of celiac disease (CD), the growing tendency to use non-invasive diagnostic methods, and the importance of early and timely detection for both diagnosis and monitoring, have created a need for increasingly efficient serological markers. This requires reliable kits that detect the specific autoantibodies of CD with greater sensitivity and specificity. These kits would be adaptable to the variable presentation of this pathology and the multiple medical specialties that currently must include CD in their list of differential diagnoses.

The detection of antitransglutaminase antibodies (anti-tTG) by ELISA using recombinant human tissue transglutaminase (tTG) as the capturing antigen is the most widely used screening test for CD. It has high sensitivity, specificity and high correlation with duodenal biopsy and is therefore considered the gold standard for diagnosis of CD.^1,2 In recent years, the role of serology has been assessed and both the European Society for Pediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) 2012 criteria and the latest ESPGHAN 2020 guidelines confirm that determination of anti-tTG IgA and total IgA are the first choice for the diagnosis of CD due to high sensitivity and specificity, availability and automated methodology (ELISA). If there is an IgA deficiency, a second diagnostic step is necessary – analysis of IgG isotype antibodies. Elevated anti-tTG IgA titers (greater than 10 times the cut-off value) predict with great specificity the existence of injury to the intestinal villi. Some false positive results have been described, usually at low values due to cross-reaction with antibodies resulting from autoimmune diseases, liver diseases or infections. Although the determination of anti-tTG IgA is not standardized, most trials for this assay show high accuracy, especially at high titers. However, variations have been observed between different types of tests and between laboratories using the same anti-tTG assay.

Recently, a new antigen was developed that combines tTG plus gliadin peptides, obtained as the protein complex. This complex occurs under physiological conditions in vivo at the time of deamidation and will henceforth be called the neo-epitope. In a previous study, the sensitivity and specificity of this new antigen was demonstrated in relation to duodenal biopsy - considered the gold standard for diagnosing CD. This study also showed greater sensitivity, specificity, and better resolution of positive and negative results, providing a decrease in the “indeterminate zone"^3,5 and a better detection of dietary transgressions.⁵

The objective of the current retrospective study was to analyze the results obtained with the use of the new neo-epitope technique and to verify the conclusions obtained in the comparison study⁵ by analyzing a greater number of serum samples from a more heterogeneous population of patients.

Patients and Methods

Patients

A retrospective analysis was conducted for all anti-tTG determinations performed during the period from January to December 2018. This study was approved by the scientific ethics committee of the institution (ethics committee reference number 060802-13, Clínica Santa María, Santiago, Chile). Informed consent was obtained from each patient whose serum samples were used in this study.

The patients were selected by searching the general registry of laboratory tests, identifying and extracting all those with anti-tTG antibody determinations done in that period using the neo-epitope technique (AESKULISA TGt-A New Generation®, AESKU, Wendelsheim, Germany). No results were pre-selected or eliminated, nor were previous trials conducted with these data.

The neo-epitope assay

This complex can be conceptualized as three different antigens: tTG, gliadin peptides, and the structure that results from the binding of the two components. In other words, the complex exposes three different epitopes: tTG, deamidated gliadin peptides (DGP) and the complex formed by both at the time of deamidation.³ This new antigen could increase detection of antibodies to different auto-antigens in celiac patients.

The neo-epitope detects a wider group of autoantibodies, increasing the sensitivity of the assay and aiding diagnosis in those patients where the initial serology was negative. This would optimize the diagnosis of patients with silent and/or latent disease, help detect dietary transgressions (voluntary or involuntary) and even decrease the number of false positives by increasing specificity and sensitivity. Better resolution of positive and negative results, with values as far away as possible from the cut-off points, therefore decreasing the “indeterminate zone”.^3,5

Clinical Data Collection

The following variables were obtained from medical records: result of anti-tTG (numerical value), age, sex, duodenal biopsy, and definitive diagnosis of CD. If the patient had been diagnosed with CD prior to examination, other diagnoses or clinical conditions were also recorded, such as whether the patient was on a gluten-free diet, and if there had been dietary transgressions.

Positive or negative anti-tTG results were determined using the cut-off values established by the manufacturer: negative <12 U/mL, indeterminate between 12 and 18 U/mL, and positive >18 U/mL. For patients identified as false positives or with results close to the cut-off values, data on liver enzymes such as gamma-glutamyl-transferase (GGT), glutamic-pyruvic transferase (GPT), and bilirubin were reviewed from the laboratory examination registry to establish the possible relationship of false positive anti-tTG results with hepatic dysfunction.

Statistical Analysis

A descriptive analysis of sex, age group and outcome of anti-tTG antibody determinations was performed using frequency distributions. The proportion of false positives was obtained for the year 2018 and compared with those obtained in the analysis from 2013,⁵ using the Z test for comparison of proportions. For all statistical analyses, P < .05 was considered significant. Stata 13 statistical software (StataCorp LLC, College Station, Texas, USA) was used for all statistical analyses.

Results

A total of 3820 anti-tTG determinations were done during the study period. The demographic characteristics of the studied patients and the anti-tTG results are detailed in Table 1. Table 2 summarizes the distribution of results of the neo-epitope technique by age and by range of reference values as indicated by the manufacturer.

Table 1.

Patient characteristics and distribution of outcomes for samples collected in 2018.

Variable		N	%
Sex	Female	2227	58.3%
	Male	1593	41.7%
Age group	Children (0-16 years)	1261	33.0%
	Adults (>16 years)	2559	67.0%
Result	Negative (<12 U/mL)	3581	93.7%
	Indeterminate zone (12-18 U/Ml)	136	3.6%
	Positive (>18 U/mL)	103	2.7%

Table 2.

Distribution of neo-epitope anti-tTG results in children and adults.

Category	Children		Adults		Total
Category	n	%	n	%	n	%
Negative (<12 U/mL)	1197	94.9%	2384	93.2%	3581	93.7%
Indeterminate zone (12-18 U/mL)	37	3.0%	99	3.9%	136	3.6%
Positive (>18 U/mL)	27	2.1%	76	3.0%	103	2.7%
Total	1261	100%	2559	100%	3820	100%

Distribution and Analysis of Patients with Negative Results

During the study period, 3581 negative results were obtained (below 12 U/mL). This was the largest fraction of the patients tested (93.7%). The number of patients with results in the “indeterminate zone” (12-18 U/mL) was 136 (3.6%).

Special emphasis was placed on the analysis of 33 patients (0.9%) that had a result of exactly 18 U/mL, the cut-off limit determined by the manufacturer. In this group, nine patients (27.3%) had a final diagnosis other than CD, but with common specific characteristics such as elevated transaminases and/or GGT, fatty liver or alcohol abuse. Four (12.1%) were celiac patients in follow-up while on a gluten-free diet. In three of the patients (9.1%), the treatment process indicated re-evaluation and subsequent follow-up due to diagnostic doubt. Four of the patients (12.1%) had autoimmune thyroiditis. For the remaining 13 patients (39.4%), the diagnosis was not CD and they lacked any other common features with CD.

The remaining patients of this group presented values close to the lower limit of the indeterminate zone (12 U/mL), the vast majority with negative endomysial antibodies (EMA) or DGP results, or even negative biopsies; they are therefore not described in as much detail.

Distribution and Analysis of Patients with Positive Results

Patients with positive results (>18 U/mL) were analyzed separately (Table 3). For this analysis the results were divided into the following ranges: >18 to 20 U/mL (n = 9); >20 to 50 U/mL (n = 33); >50 to <300 U/mL (n = 23) and >300 U/mL (n = 38) (Table 3). Of these 103 patients, 77 were patients at the time of CD diagnosis or were in follow-up while on a gluten-free diet. Seven of the patients were being re-tested for an equivocal diagnosis, and 19 were found to be false positives. The analysis of this last group is described in the next section.

Table 3.

Distribution of positive results for the neo-epitope anti-tTG assay according to the ranges of values obtained, and percentage of false positives for each range.

Neo-epitope result ranges	Total		False positives
Neo-epitope result ranges	n	%	n	%
18-20 U/mL	9	8.7%	9	100.0%
21-50 U/mL	33	32.0%	10	30.3%
51-300 U/mL	23	22.3%	0	0.0%
>300 U/mL	38	36.9%	0	0.0%
Total	103	100%	19	18.4%

False Positive Percentage with the Neo-Epitope Technique in Routine Practice

When the results of neo-epitope positive patients with definitive diagnoses other than CD were considered, a total of 19 false positives (0.5%) were found. Four of these patients presented altered liver tests (21.1%) and three had other autoimmune pathologies, all of them with negative EMA (15.8%). The remaining 11 had normal duodenal biopsies and tests without identifiable interference such as normal liver tests. It should be noted that the described false positives were concentrated in the ranges of 18 to 20 U/mL and 21 to 50 U/mL (Table 3).

False positives were compared between the current study (2018: 0.5%) and the 2013 validation study⁽⁵⁾ (3.6%) to assess whether the same proportion of false positives was seen in these two studies. Both studies utilized the neo-epitope technique. The Z test for comparison of proportions indicated that the proportion of false positives is significantly lower in the 2018 study than in the 2013 validation study (P < .001 value).

New Cut-off Point of the Neo-Epitope Technique

Given the results obtained in this study, the results of the indeterminate zone obtained from the previous study⁵ and proposals by other authors,^6,9 the predictive capacity of this new antigen was examined using a cut-off point of 20 instead of 18 U/mL. The positive predictive value (PPV) was calculated for a cut-off point of 18 U/mL, set by the manufacturer and the proposed new cut-off point of 20 U/mL. The results were: 74.8% (77/103) and 81.9% (77/94), respectively, showing a tendency to increase the PPV with the proposed new cut-off.

Discussion

Laboratory techniques used as diagnostic support for CD should provide early detection of the different situations that occur in CD (diagnosis, follow-up, dietary transgressions) and achieve good resolution of positive and negative results. This should reduce the “indeterminate zone” or uncertainty, thereby avoiding the possibility of obtaining values close to the cut-off. These values can mislead the clinician due to the diversity of known pathologies related to gluten that make up a set of presumptive diagnoses.

The objective of this paper was to evaluate the incorporation of the neo-epitope assay, and test the hypothesis that the neo-epitope assay is the most useful tool for initial CD screening and follow-up for patients on a gluten-free diet.

In the comparison study,⁵ 3.6% false positives were obtained with the neo-epitope technique compared to 7.2% with the recombinant human anti-tTG technique in relation to biopsy results. In the current follow-up study, the proportion of false positives obtained when studying a larger number of samples was considerably reduced (19/3820 = 0.5%). This confirmed the advantages of this new antigen in testing for anti-tTG antibodies. These results could be due to the fact that the new sensitizing antigen exposes three different epitopes: tTG, gliadin deaminated peptides and the complex formed by these two components at the time of deamidation. Having three different epitopes could increase the chances of diagnosis of the different autoantibodies found in celiac patients. These results agree with those published by other authors who evaluated several kits that are currently on the market.^3,6,7

An important aspect of this technique, as stated in our previous study, is good resolution of values with results as far away as possible from the cut-off value of the technique and outside the indeterminate zone. With the results from the previous study,⁽⁵⁾ it was proposed that the use of the neo-epitope technique would lead to the reduction of the “indeterminate” or “uncertain” zone. This could provide more decisive results than the recombinant human anti-tTG technique (ImmuLisa™ Celiac anti-tTG IgA Antibody Enhanced ELISA IMMCO, Williamsville, NY, USA).

In the present study, it was proposed that cut-off points more reflective of the reality of the Chilean population could be established, potentially avoiding additional testing based on the diagnostic algorithm for patients with indeterminate results. During 2018, 3820 samples were analyzed, of which 93.7% were clearly negative and only 3.6% were recorded in the uncertainty zone. The analysis of the characteristics of the patients who had values ≥18–20 U/mL showed they were non-celiac patients. This led us to propose a cut-off value of 20 U/mL for our population. This would allow us to: (a) increase the PPV value from 74.8% (positive cut-off value >18) to 81.9% (positive cut-off value >20); (b) reduce the number of patients in the “indeterminate zone” and (c) decrease the number of false positives for other autoimmune diseases and hepatic dysfunction.^6,9 These advantages come without losing the greater capacity for detection of patients on a gluten-free diet that have not reversed their markers or have normalized their duodenal mucosa. This is fully in line with the proposal by Porcelli et al⁶ and Rozenberg et al¹⁰ who reference similar situations in their studies.

The results presented here using the new neo-epitope technique detected a significant number of patients with voluntary or involuntary dietary transgressions. It was also used to follow patients after dietary modification immediately post-diagnosis. The result showed the ability to detect that the mucosa had not yet normalized its architecture despite the gluten-free diet. Silvester et al¹¹ showed that the usual markers were not good indicators of histological normalization. It could be inferred that the new neo-epitope technique had a better correlation with histological improvement, or better yet, with the detection of histological damage to the mucosa, even if the patient is on a gluten-free diet.¹¹ It is important for the patient to rigorously comply with a gluten-free diet and to have a highly sensitive and specific technique that is capable of detecting minimal transgressions to alert the clinician to work with the patient to detect the foods that are potentially causing this transgression.

The results of the present study are in line with those of other authors regarding this new antigen and it has proven to be an adequate tool for routine implementation, both for diagnosis and follow-up. This technique also has the greatest screening utility by providing a “minimum indeterminate zone”.

Several studies have reviewed different combinations of autoantibodies and algorithms for diagnosis of celiac disease, in order to establish which is most appropriate in each clinical situation or based on the age of the patient.^12,14 As previously stated, this is a new antigen, for which it was considered that for its initial evaluations it was essential to interpret and correlate the results to the histological findings. Not until prospective studies were conducted that established its usefulness could it even be considered that a clinician would not perform a biopsy to confirm diagnosis, especially considering the new international guidelines. In this study, this assay was shown to be a serological marker with some advantages over conventional markers. However, it would not be appropriate, at least for now, to consider dispensing with a biopsy and histology by using this marker, alone or combined with other markers, for the diagnosis of celiac disease, certainly not until additional studies have been completed. The main utility of this technique, with the data that are currently available, would be as a more useful tool for initial screening and for monitoring of a gluten-free diet, mainly to find difficult-to-detect dietary transgressions.

It is important to note that the retrospective nature of our study is a limitation, as not all information was available for all patients. However, due to the lack of information in the literature relative to this new neo-epitope technique, it is vitally important to communicate these results which could be very useful for laboratories and specialists involved in treating CD patients.

Considering the importance of early diagnosis of CD to prevent complications and the association of CD with other autoimmune diseases, a CD patient may suffer from not being introduced to a gluten-free diet as early as possible.¹⁵ The idea of lowering the diagnostic dividing line and increasing the “tip of the iceberg” as expressed by Lerner et al should be adopted. This provides a new approach to the diagnostic algorithm in which the neo-epitope plays a leading role.¹⁶ All the findings above emphasize the importance of conducting a prospective study with this new antigen, in conjunction with other markers, to validate its usefulness as a first-choice assay in the diagnosis and monitoring of celiac disease.

Conclusion

Based on the current study, which is in agreement with publications by other authors, the neo-epitope technique can be considered a more precise tool than other assays for both diagnosis and monitoring of celiac patients. It is important to highlight its greater resolution of positive and negative results resulting in a “minimum indeterminate zone” and low proportion of results close to the cut-off value. These data strongly suggest consideration of an alternate (higher) cut-off value (20 U/mL). This study of a representative population with the neo-epitope technique in relation to biopsy results and final diagnoses suggests this alterative cut-off could significantly increase the PPV of the technique.

Key Points

The technique with the neo-epitope antigen could be considered a more precise tool than an assay using recombinant human transglutaminase both for diagnosis and for follow-up of celiac patients in pediatric and adult populations.

In this study, the neo-epitope-based assay had greater capacity to resolve positive and negative results, resulting in a “minimum indeterminate zone” and a low proportion of results close to the cut-off value.

The exercise of determining a cut-off value for each patient population, regionally, if possible, by conducting a study with a representative population of samples correlated with biopsy results and the final diagnoses could significantly increase the PPV of the technique.

Footnotes

Acknowledgements

We would like to express our gratitude to Alejandro Pavéz and Miguel Catalán for their valuable collaboration in the preparation of this manuscript. In addition, the authors thank Carles Masgrau PhD, Ana María Ortega, Michael K James PhD, and Jordi Bozzo PhD (Grifols) for their expert review, comments and suggestions provided in the preparation of this article.

Author Contributions

SV and AC originated the idea and prepared and reviewed the manuscript. AC carried out the statistical study. All authors approved the final version of the manuscript and agreed to the submission.

Declaration of Conflicting Interests

The author(s) received support from Grifols, the distributor of AESKU products for the USA, Chile, Mexico, Spain and Portugal, to prepare this article.

Funding

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

List of Abbreviations

References

Volta

Villanacci

. Celiac disease: diagnostic criteria in progress. Cell Molec Immunol. 2011;8(2):96-102. doi: 10.1038/cmi.2010.64.

Rozengerg

Lerner

Pacht

, et al. A new algorithm for the diagnosis of celiac disease. Cell Molec Immunol. 2011;8(2):146-149. doi: 10.1038/cmi.2010.63.

Torsten

Pfeiffer

Selmi

Gershwin

. Diagnostic challenges in celiac disease and the role of the tissue transglutaminase-neo-epitope. Clinic Rev Allerg Immunol. 2010;38(2-3):298-301. doi: 10.1007/s12016-009-8160-z.

Rozenberg

Lerner

Pacht

, et al. A novel algorithm for the diagnosis of celiac disease and a comprehensive review of celiac disease diagnostics. Clinic Rev Allerg Immunol. 2012;42(3):331-341. doi 10.1007/s12016-010-8250-y.

Verbeke

Gutierrez

Catalán

Canals

. Estudio de un neo-epítope para detección de anticuerpos antitransglutaminasa en enfermedad celíaca. ARS MEDICA Revista de Ciencias Médicas. 2020;45(3):29-35. https://doi.org/10.11565/arsmed.v45i3.1650.

Porcelli

Ferretti

Vindigni

Terzuoli

. Assessment of a test for the screening and diagnosis of celiac disease. J Clin Lab Anal. 2016;30(1):65-70. doi: 10.1002/jcla.21816.

Lerner

Ramesh

Matthias

. Serologic diagnosis of celiac disease: new biomarkers. Gastroenterol Clin N Am. 2019;48(2):307-317. doi: 10.1016/j.gtc.2019.02.009.

Torrutti

Bizzaro

. Diagnosis and classification of celiac disease and gluten sensitivity. Autoimmun Rev. 2014;13(4-5):472-476. doi: 10.1016/j.autrev.2014.01.043.

Román Riechmanna

Castillejo de Villasante

Cilleruelo Pascuala

, et al. Aplicación racional de los nuevos criterios de la European society for paediatric gastroenterology, hepatology and nutrition (ESPGHAN) 2020 para el diagnóstico de la enfermedad celíaca. An Pediatr (Barc). 2020;92(2):110.e1-110.e9. https://doi.org/10.1016/j.anpedi.2019.12.001.

10.

Husby

Koletzko

Korponay-Szabó

, et al. European Society paediatric gastroenterology, hepatology and nutrition guidelines for diagnosing coeliac disease 2020. J Pediatr Gastroenterol Nutr. 2020;70(1):141-156. doi: 10.1097/MPG.0000000000002497.

11.

Silvester

Kurada

Szwajcer

Kelly

Leffler

Duerksen

. Test for serum transglutaminase and endomysial antibodies do not detect most patients with celiac disease and persistent villous atrophy on gluten-free diet: a meta-analysis. Gastroenterology. 2017;153(3):689-701.e1. doi: 10.1053/j.gastro.2017.05.015.

12.

Sugai

Moreno

Hwang

, et al.

Celiac disease serology in patients with different pretest probabilities: is biopsy avoidable?

World J Gastroenterol. 2010;16(25):3144-3152. doi: 10.3748/wjg.v16.i25.3144. PMID: 20593499; PMCID: PMC2896751.

13.

Oyaert

Vermeersch

De Hertogh

, et al. Combining antibody tests and taking into account antibody levels improves serologic diagnosis of celiac disease. Clin Chem Lab Med. 2015;53(10):1537-1546. doi: 10.1515/cclm-2013-1099. PMID: 25719330.

14.

Wolf

Petroff

Richter

, et al. Validation of antibody-based strategies for diagnosis of pediatric celiac disease without biopsy. Gastroenterology. 2017;153(2):410-419.e17. doi:10.1053/j.gastro.2017.04.023. Epub 2017 Apr 28. PMID: 28461188.

15.

Verbeke

Cruchet

Gotteland

, et al. Risk markers for insulin-dependent diabetes mellitus. Rev Med Chile. 2004;132(8):979-984. doi: 10.4067/s0034-98872004000800010

16.

Lerner

. Serological diagnosis of celiac disease – moving beyond the tip of the iceberg. Int J Celiac Dis. 2014;2(2):64-66. doi: 10.12691/ijcd-2-2-8.