Introduction

Antinuclear antibodies (ANA) testing is the most widely used method for the diagnosis of several autoimmune diseases, particularly systemic and autoimmune liver diseases [1, 2]. The clinical value of ANA testing has been well-established [3]. Specifically, the latest 2019 EULAR/ACR classification criteria for systemic lupus erythematosus establish positive ANA as the only obligatory entry criterion [4], which underscores the importance of ANA determination in the study of systemic autoimmune diseases. The most common ANA screening method is indirect immunofluorescence (IIF) on HEp­2 cells [5]. However, several variables can influence the results of IIF, including the HEp-2 substrate, the serum screening dilution ratio, and the operator’s interpretation. Guidelines on ANA testing by IIF recommend reporting both the titer and the staining patterns observed in HEp-2 cells, including not only nuclear patterns (e.g., homogeneous or speckled), but also cytoplasmic and mitotic patterns [6, 7]. The pattern is important because it helps to determine follow-up testing for specific autoantibodies. It is worth noting that solid phase assays (SPA) are increasingly being used in clinical laboratories to screen for ANA, although they do not provide information on patterns [2]. HEp-2 pattern recognition by IIF has undergone a revolution in recent years, mainly due to the development of automated computer-assisted diagnosis (CAD) platforms. Despite significant advances in this area, we are still far from achieving our ultimate goal of fully automated testing because although CAD-based technology is useful for initial screening, it is still insufficient to detect all the expert-level patterns defined by the International Consensus on ANA Patterns (ICAP) and mixed HEp-2 patterns [8, 9].

In 2015, the ICAP made a comprehensive effort to develop an alphanumeric nomenclature to harmonize the names and descriptions of HEp-2 patterns [10]. This report represents a major milestone in the standardization of this technique. Although the use of this nomenclature has been steadily increasing worldwide, many laboratories have not yet adopted this system.

Despite the routine use of ANA testing for autoimmune disease, no clear, well-established criteria are available on how to perform the technique and interpret the results. For this reason, the Spanish Group on Autoimmune Diseases (GEAI) from the Spanish Society of Immunology (SEI) considers it necessary to determine how autoimmunity laboratories in Spain perform and interpret ANA testing and how they determine related antigen specificities. The GEAI-SEI was created to promote closer contact among healthcare professionals specializing in the diagnosis of autoimmune diseases. This group carries out a range of activities, including workshops on autoimmune diseases, which provide useful information not only to laboratory specialists but also to clinicians, thereby improving the management and interpretation of autoantibody detection [11].

In this context, we surveyed autoimmunity laboratories in Spain. The GEAI committee then met to assess the results of the survey in order to establish consensus-based recommendations for the determination of ANA and antigen specificities.

Materials and methods

Participants

The survey was sent to a total of 65 centres, including members of the GEAI and the autoimmunity group of the Spanish Society of Clinical Chemistry (SEQC). Of these, 50 autoimmunity laboratories, uniformly distributed throughout Spain, completed the survey.

Survey

The survey (Table 1) consisted of 36 multiple choice questions to assess the following:

  • ANA testing methodology: screening; substrate; screening, final and estimated dilutions; use or not of an automated computer-assisted system.

  • Reporting of results: titer, pattern, and ICAP nomenclature.

  • Criteria for performing (or not) ANA-related antigen specificities, such as anti-dsDNA (double-strand DNA), anti-ENA (extractable nuclear antigen antibodies), and anti-nucleolar and anti-cytoplasmic antibodies.

  • Frequency of test repetition for ANA and antigen specificities.

  • Methods used to identify ANA-related antigen specificities.

  • Patterns associated with autoimmune liver diseases.

Table 1 Survey questions, results and responses
Full size table

Statistical analysis

Statistical analysis was performed with the IBM-SPSS statistical software, version 24. For each survey question, consensus was defined when ≥ 75% of all participating laboratories selected the same response option.

Results

The completed survey was returned by 50 of the 65 laboratories (77% response rate). Table 1 shows the 36 multiple choice questions and the number and frequency of each response. For many of the questions, more than one response was possible.

The most common screening methodology to determine ANA was IIF (84% of centres; question [Q]1, Fig. 1A) on HEp-2 cells and their variants (Q2, Fig. 1B). The laboratories that did not use IIF as the primary screening method (16% of laboratories) mainly used SPA (12%), such as ELISA (enzyme-linked immunosorbent assay) with cell extracts and recombinant antigens (8%), or ELISA, FEIA (fluoroenzyme immunoassay), or CLIA (chemiluminescence) with antigenic mixtures. In cases with a positive SPA screening test result, all of the laboratories indicated that they perform IIF on HEp-2 to characterize these ANA (Q3). Only two laboratories (4%) used multiplex bead technology as a screening methodology (Q2).

Fig. 1
figure 1

A Initial screening method for antinuclear antibodies (ANA) detection. (B) Substrate used for indirect immunofluorescence (IIF) assay

Full size image

Most laboratories (80%) do not use different ANA methods for initial screening (Q4). Among the laboratories that do use a different method, the choice is influenced by the requesting physician and/or the patient’s diagnosis (Q5). Most of the responding laboratories (90%) use a screening dilution ratio of 1:80 or 1:160 (38% and 52%, respectively; Q6). However, there was no consensus on whether to dilute, estimate, or perform serial twofold dilutions of positive sera (Q7). Similarly, there was no agreement among respondents regarding the maximum dilution ratio (Q8); however, the most commonly reported ratio was 1/1280 (40% of respondents). More than half (56%) of the participating laboratories do not use automated computer-assisted interpretation (Q9). The laboratories that use this methodology do so only to screen for positive vs. negative results (14%), or for screening and titer (4%), or for screening, titer, and pattern (22%).

Most laboratories (90%) report ANA test results as either negative or positive, together with titer and pattern (Q10). However, 20% of these laboratories do not titrate positive samples with specific patterns (Q11). Nearly half of the laboratories (48%) use the ICAP nomenclature for pattern designation (Q12). In the laboratories (n = 10) that use SPA for screening, 80% (8/10) report quantitative results (Q13).

Given that ANA and related autoantibody tests are often requested at the same time, the laboratories must decide whether to perform the autoantibody tests or not. One of the survey questions (Q14) asked the respondents to indicate whether they performed specificities for related antigens in cases with a negative ANA test. Responses to this question were highly variable, as follows: 18% of respondents perform autoantibody detection in all cases, 14% do not perform antigen specific tests in this case, and the remaining laboratories (68%) base their decision on the requested specificity and on the clinical information.

In the presence of a positive ANA test, laboratory practices were highly variable with regard to testing for related antigen specificities (Q15). However, for most laboratories (86%), the decision to perform these specificities is based on the ANA pattern (Q16). In these cases, the first step for 58% of laboratories is to use a screening test containing mixtures of the clinically relevant antigens (Q17). If this first test is positive, the result is then confirmed by a different methodology (i.e., ELISA, FEIA, CLIA, or line blot) using individualized antigens (Q18).

The minimum time interval for repeating ANA testing (e.g., for the initial or definitive diagnosis) was highly heterogeneous (questions 19 to 22). Similarly, the protocols used to test for related antigens, including anti-dsDNA, anti-ENA, or cytoplasmic antigens, were also highly variable (Q23 and Q24).

The two most commonly used testing methods for anti-dsDNA antibodies were CLIA (42%) and FEIA (22%) (Q25, Fig. 2A). Most laboratories (70%) confirm positive anti-dsDNA results by another technique (Q26, Fig. 2B), mainly IIF in Crithidia luciliae cells (CLIFT) (Q27). Even though Farr radioimmunoassay is the reference method for anti-dsDNA antibodies testing, clinical laboratories rarely use this method because it requires radioactive materials, which has numerous drawbacks: it entails risks for personnel and the environment, and it requires custom-built spaces, highly qualified personnel, and high costs. Interestingly, a substantial proportion of the laboratories (38%) verify negative anti-dsDNA results (mainly in patients with suggestive symptomatology or in the presence of hypocomplementemia due to consumption by immune complexes) by performing another test using a different technique (Q28). To detect centromere antibodies, 54% of laboratories use only line blot assays while 24% use only IIF on HEp-2 cells (Q29).

Fig. 2
figure 2

A Initial method for anti-dsDNA antibodies detection. (B) Confirmation of positive anti-dsDNA antibodies by another technique

Full size image

The autoantibodies tested depend on the specific HEp-2 pattern. For cytoplasmic patterns AC-19 and AC-20, most of the laboratories (94%) perform specificities for related antigens, even if not requested by the clinician (Q30). However, 38% of the respondents only determine specificities in cases with a negative anti-ENA screening test and the presence of suggestive symptoms (Fig. 3A). The specificities analysed were highly variable among the laboratories (Q31). In nucleolar patterns (AC-8, 9, 10), determination of specificities is conditioned by the titer (question 33, Fig. 3B). For cytoplasmic and nucleolar specificities, most laboratories use line blot (64% and 72%, respectively; Q32 and 34).

Fig. 3
figure 3

A Determination of related antigenic specificities in cytoplasmic indirect immunofluorescence (IIF) HEp-2 patterns (AC-19, AC-20). (B) Determination of related antigenic specificities in nucleolar IIF HEp-2 patterns (AC-8, AC-9, AC-10)

Full size image

Finally, in cases presenting nuclear or cytoplasmic patterns associated with autoimmune liver diseases, 94% of the laboratories determine the antigenic specificities associated with these patterns, even if not requested (Q35 and 36).

Discussion

ANA autoantibody tests provide highly valuable data for the diagnostic work-up, which is why these are the most commonly requested tests by physicians in patients when autoimmune disease is suspected [12]. Moreover, the combination of antibody levels and patterns can provide valuable information to help establish the diagnosis [13].

The primary aim of this survey was to assess the procedures and practices utilised by laboratories in Spain to detect ANA and related antigens. A secondary aim was to disseminate the survey findings, as well as recommendations based on those results, to promote greater harmonization among laboratories in Spain.

Overall, we found that, for most of the key items in this survey, the participating laboratories generally follow the same testing procedures. The screening methodology for ANA used by most of the laboratories (84%) is HEp-2 IIF (Fig. 1A). In addition, 90% of the laboratories also report the titer and ANA pattern together with the test results, as recommended by international guidelines and expert consensus statements (Fig. 2B) [6, 14, 15]. Similarly, the substrates used for IIF were HEp-2 cells and their variants (Fig. 1B). Although only a minority (12%) of laboratories reported using SPA as the screening methodology (rather than IIF on HEp-2 cells), the use of SPA for ANA determination continues to grow, despite the fact that the equivalence of these two testing methodologies has not yet been fully established [2]. In this regard, it has been reported that SPA without HEp-2 extracts have a low sensitivity and high specificity, particularly in non-selected populations [16]. Consequently, in cases with a positive SPA screening test, the finding should be confirmed by IIF on HEp-2 cells, which is exactly the approach taken by the laboratories in this survey.

We found that the ANA pattern conditioned follow-up testing for specific antigen-related antibodies in most laboratories. Follow-up testing is essential for specific patterns (e.g., nucleolar, cytoplasmic, centromere, or liver disease associated patterns) that have a strong association with specific pathologies. Moreover, although both pattern and titer are directly associated with the likelihood of disease [13, 17], a substantial proportion of the laboratories (20%) do not titrate samples with these specific patterns, but instead directly perform the associated antigenic determination. Our data show that the methods used to detect these specificities varied widely among the participating laboratories. In any case, suspected antigen-related specificities should be characterized with individualized antigenic tests.

The two most commonly reported screening dilution ratios in this survey were 1:160 (52% of laboratories) and 1:80 (38%). By contrast, two recent surveys, one conducted by the ICAP [14, 18] and the other by the European Autoimmune Standardization Initiative (EASI) [14, 18], found that most laboratories use the 1:80 screening dilution ratio (80% and 60.5%, respectively). We found that laboratory practices related to diluting, estimating, and serializing positive sera to obtain the final titer were highly heterogeneous. Similarly, the frequency of repeat testing for ANA and related antigen specificities was also highly variable. This clear lack of consensus on these items is probably due to the absence of specific recommendations in current guidelines. The appropriate time interval for test repetition should be established by the consensus between the requesting physician and the testing laboratory, based on the pathogenicity of the antibodies and their importance for the diagnosis and/or follow-up of the specific disease.

Our survey showed that slightly less than half (48%) of the laboratories use the ICAP nomenclature. Given the importance of harmonizing data reporting, this finding is highly relevant. In this regard, we hope that the results of the present survey will encourage all laboratories in Spain to incorporate this nomenclature into their routine practice. Curiously, the proportion of laboratories using this nomenclature in our survey is substantially higher than other surveys, such as the one conducted by the ICAP, where only 27% of laboratories either incorporate AC pattern descriptions or are in the process of modifying their reporting practices to include AC codes in the final ANA report [19]. Unfortunately, the survey conducted by the EASI did not report this percentage, but the authors did mention that the use of the ICAP nomenclature would help to improve the standardization of methodologies, tests, and interpretation of results [18].

We found wide variability in the techniques used to detect specificities for ANA-related antigens. The participating laboratories applied the same procedures for only a few items in this area, such as testing for anti-dsDNA antibodies, with most respondents using a confirmatory method (mainly CLIFT). In addition, the most common option for cytoplasmic and nucleolar specificities was line blot. These findings further underscore the need for clear recommendations to harmonize testing procedures.

Among the limitations of our study, it should be noted that the presence of low-titer ANAs, which can be present in up to 40% of healthy individuals and act as a confounding factor, was not considered in the survey.

Finally, although we did not assess clinician-laboratory communication in the present survey, we strongly believe that close communication between the laboratory and the requesting clinician is essential to ensure that the appropriate tests and to enable proper interpretation of the results. In this regard, the patient’s clinical condition, together with the results of ANA testing, should determine the need to test for antigen-related specificities.

Conclusions

The results of this survey provide a real-world picture of how detection of antinuclear antibodies and related antigens is performed in autoimmunity laboratories in Spain. In most of the laboratories surveyed, the recommendations of experts and international guidelines are followed, but there is still some variability, probably due to the lack of clear recommendations for certain procedures.

Based on the findings of this survey, the GEAI committee held a series of meetings to reach consensus on best practices and to draw up specific conclusions and recommendations (Table 2).

Table 2 Conclusions and recommendations
Full size table

Overall, this survey shows that most autoimmune laboratories in Spain use a similar approach to ANA testing. However, the survey also reveals a clear need to further standardize testing and reporting protocols. In this regard, surveys such as this can provide valuable data to promote greater harmonization among laboratories. We believe that this survey should be repeated approximately 3 years from now to ascertain whether the recommendations developed by the GEAI committee have been implemented. Greater harmonization of testing procedures would provide many benefits, including optimization of the resource use. More importantly, this would provide better quality and more reliable results, which would ultimately benefit patients.

ANA, antinuclear antibodies; IIF, indirect immunofluorescence; ELISA, enzyme-linked immunosorbent assay; FEIA, fluoroenzyme immunoassay; CLIA, chemiluminescence; SPA, solid phase assays; dsDNA, double-strand DNA; ENA, extractable nuclear antigen antibodies; CLIFT, IIF in Crithidia luciliae cells; RIA, radioimmunoassay; AC, anti-cell.