Introduction

Anti-nuclear autoantibodies (ANA) are present in many systemic autoimmune rheumatic diseases (SARDs), and the detection of ANA and antibodies to extractable nuclear antigens (ENA) is commonly used in the diagnostic workup of SARDs and autoimmune liver diseases1. Traditionally, ANA were detected by indirect immunofluorescence (IIF) assay on HEp-2 cells. However, manual IIF has been challenged as the standard method for ANA testing due to the lack of automation, standardization, modern data management, and human bias in the interpretation of IIF patterns2. In the past decades, novel automated IIF interpretation systems enabled the analysis of IIF patterns by designated pattern recognition algorithms, strengthening the position of IIF as a screening technique of ANA3,4.

According to the current testing algorithms, it is recommended that a positive ANA result, obtained by manual or automated IIF, should be followed-up by a confirmatory test for disease-specific autoantibodies also referred to as reflex test5.

A possible alternative for ANA IIF includes the use of high throughput multiplex immunoassays that allow simultaneous detection of a panel of specific ANA6. These diagnostic platforms include enzyme-linked immunosorbent assay (ELISA), chemiluminescence immunoassay (CIA), fluorescent-enzyme immunoassay (FEIA), line immunoassays (LIA), and other methods. However, these methods were evaluated as a first-line screening methods in the workup of SARDs diagnosis7, and discrepancies between different methods were observed as well as difficulty to interpret the results in different clinical settings.

In 2014, a set of recommendations for the appropriate assessment and interpretation of ANA detected by different methods was proposed by an international workgroup of experts representing 15 European countries8. According to this expert panel, methods other than IIF might be used for ANA screening. However, when the results of these methods are negative and the clinical suspicion is strong, IIF should be used for the definitive serological diagnosis. Moreover, it has been widely demonstrated that a combination of traditional IIF and alternative assays may improve the diagnostic accuracy of ANA testing9.

In light of this, in 2015 a technique for testing ANA and ANA-specific autoantibodies by combining both screening and confirmation assays has been introduced, by the name of CytoBead10,11,12,13. The method enables automated or manual ANA screening by cell-based IIF, which is combined with a confirmatory test that is based on size- and color-encoded microbeads (9 to 15 μm in diameter) at fixed areas coated with specific antigens. Notably, both steps are performed in one reaction environment, in contrast to a two-tier approach. The antigen-coated beads that are used for ANA differentiation cover the following antigens: double-stranded DNA (dsDNA), Jo-1, Sm/ribonucleoprotein (Sm/RNP), scleroderma antigen 70 kDa (Scl-70, topoisomerase I), Sm, Sjogren’s antigen A 52 kDa (SS-A/Ro52), SS-A/Ro60, and Sjogren’s antigen B (SS-B/La).

In addition to ANA detection, the CytoBead system may also be used to detect the dense-fine speckled (DFS) pattern specifically caused in general by anti-DFS70 autoantibodies (anti-DFS70)13. In IIF testing, this pattern is characterized by a dense fine-speckled staining of the interphase chromatin and a speckled staining of the metaphase plate14, but other specific ANAs including anti-dsDNA autoantibodies (anti-dsDNA) and ENA were also reported causing this pattern15. Anti-DFS70 target a 70 kDa protein also known as lens epithelium-derived growth factor (LEDGF), a product of the PSIP1 gene16. Previously, it has been found that they may account for positive ANA results in healthy patients, and they rarely appear in patients with SARDs or autoimmune liver diseases17.

Earlier, a good agreement was found between CytoBead ANA 2 and ANA IIF testing followed by ELISA in patients with SARDs and non-SARDs11. In the present study, we assessed the performance of CytoBead ANA 2 in the detection of ANA and ANA-specific autoantibodies in routine serum samples, and compared it to the performance of ANA IIF and BioPlex™ 2200 (a multiplexed assay for the detection of ANA-specific autoantibodies)18. In addition, we evaluated the ability of Cytobead ANA DFS-70 to identify the DFS pattern associated with anti-DFS70 antibodies in non-SARDs patients, using one reaction environment.

Materials and methods

Serum samples

This is a single-center laboratory study that was performed in the autoimmune laboratory of the Zabludowicz Center for Autoimmune Diseases Sheba Medical Center. This study assessed the performance of CytoBead ANA 2 and CytoBead ANA DFS-70 in two groups of samples that were sent to the laboratory for the assessment of ANA and ANA-specific autoantibodies.

The first group consisted of 112 routine sera samples from patients with suspected SARDs that were evaluated by ANA IIF and subsequently by BioPlex™ 2200 for the determination of ANA-specific autoantibodies. These samples were encoded and tested independently and in a blinded fashion (without prior knowledge of the results of ANA IIF or BioPlex™ 2200) by CytoBead ANA 2.

Following the assessment by CytoBead ANA 2, data regarding age, gender and diagnosis (presence or absence of SARDs) were collected from the digital records of the Sheba Medical Center. In patients with SARDs, the specific disease name was documented (Table 1).

Table 1 Demographic data and diagnoses of the study population evaluated by cytobead ANA 2.
Full size table

The second group included 29 ANA-positive samples exhibiting a DFS pattern by ANA IIF that were obtained from non-SARDs patients and further tested by CytoBead ANA DFS-70. Six samples were a subset of the samples evaluated in the first part of the study. Additional samples were selected from a group of routine samples that were previously submitted to our laboratory for routine ANA testing and were found to be positive for ANA with DFS pattern.

The study fulfilled the ethical guidelines of the most recent declaration of Helsinki and received approval by the local ethical committees (Edinburgh, 2000). The application number is 6849-20-SMC.

Diagnostic assays

Determination of ANA by indirect immunofluorescence (screening test)

All samples were tested by NOVA View instrument (Inova Diagnostics, San Diego, CA, USA), an automated fluorescent microscope routinely used in the autoimmune laboratory of the Sheba Medical Center.

The NOVA View® Automated Fluorescence Microscope is an automated system consisting of a fluorescence microscope and software that acquires, analyzes, stores, and displays digital images of stained indirect immunofluorescent slides. It is intended as an aid in the detection and classification of ANA by indirect immunofluorescence.

NOVA View determines the result (positive or negative including cytoplasmic staining) and performs ANA pattern interpretation. After pattern confirmation by the operator, NOVA View calculates a pattern-specific endpoint titer. Results are recorded electronically in a transcription free and paperless work environment and all digital images are archived for future review19.

Detection of specific ANAs by bioplex™ 2200 (confirmatory/reflex test)

All serum samples were evaluated by BioPlex™ 2200 in order to evaluate the presence of ANA-specific autoantibodies. Bio-Rad Bioplex™ 2200 system is a multiplexed assay that employs a flow immunoassay for the detection and identification of multiple antibodies to different antigens in a single incubation. The specific antigens that are used for autoantibody detection include dsDNA, centromere B (CENP-B), chromatin, Jo-1, ribosomal P, RNP 68 kDa, RNP A, Scl-70, Sm, Sm/RNP, SS-A/Ro52, SS-A/Ro60, SS-B/La18. For the current study, we have used the results of Sm/RNP when comparing the performances of BioPlex™ 2200 and by CytoBead ANA2.

Multiplexed detection of ANA by cytobead ANA2

All sera samples were analyzed with manual CytoBead ANA 2 assay for the detection of ANA on HEp-2 cells and detection of specific autoantibodies against dsDNA, Jo-1, SS-B/La (anti-SS-B/La), Sm (anti-Sm), Sm/RNP (anti-Sm/RNP), Scl-70 (anti-SCL70), SS-A/Ro52 (anti-SS-A/Ro52), and SS-A/Ro60 (anti-SS-A/Ro60).

CytoBead ANA 2 (GA Generic Assays) is a multiplex IIF test that combines screening of ANA on HEp-2 cells and a confirmatory assay based on multiplexed microparticle immunoassay using 9 μm and 15 μm red fluorescent microbeads (excitation 610 nm/emission 690 nm). Fluorescent microbeads were covalently coated with SS-B/La, Sm, Sm/RNP, dsDNA, Jo-1, Scl-70, SS-A/Ro60, and SS-A/Ro52. Furthermore, glass slides with multi-compartment wells were employed to immobilize HEp-2 cells in the central compartment (Fig. 1). For simultaneous confirmatory testing, antigen-coated microparticles were immobilized in the four compartments around the central part of the well. In addition, a reference microbead population of 12 μm size emitting green fluorescence and not coated with antigen was immobilized in each peripheral compartment to aid in the discrimination of the antigen-coated microbead populations.

Fig. 1
figure 1

CytoBead slide format used for the simultaneous screening and confirmation of ANA-specific autoantibodies. HEp-2 cells are located in the middle compartment of the wells. Antigen-coated, red fluorescent beads are localized in the smaller compartments around the large well, which differ in size and coated antigen: Compartment I contains SS-B/La (9 μm) and Jo-1-coated beads (15 μm) Compartment II contains Sm (9 μm) and Sm/RNP-coated beads (15 μm). Compartment III contains dsDNA (9 μm) and Scl-70-coated beads (15 μm). Compartment IV contains SS-A/Ro60 (9 μm) and SS-A/Ro52-coated beads (15 μm). In addition, size and location reference (SLR) beads are immobilized in all the compartments listed above (adapted from manual Cytobead ANA 2 Instruction).

Full size image

In the first step, autoantibodies in patient samples diluted 1/80 react specifically with antigens of the HEp-2 cells and beads fixed onto the slide surface. Unbound components are removed through a wash step, following 30 min incubation at room temperature. Bound autoantibodies react specifically in a second reaction step with anti-human antibodies coupled to fluorescent molecules. Excess conjugate molecules are removed through a further wash step, following 30 min incubation at room temperature. After covering, slides are read under a fluorescence microscope (excitation wavelength 490 nm, emission wavelength 520 nm). The slides were examined by 2 independent experienced laboratory physicians. Specific fluorescence patterns are given according to the histological arrangement of antigens in the HEp-2 cells. Presence of a centromere (CENP) pattern in the central compartment containing HEp-2 cells was recognized as the sample being positive for anti-CENP antibodies.

Detection of dense-fine speckled pattern by CytoBead® ANA DFS-70 test

In total, 29 ANA positive samples that exhibited DFS pattern by IIF and obtained from non-SARDs patients were evaluated by manual CytoBead® ANA DFS-70. CytoBead® ANA DFS-70 is an indirect immunofluorescence test for the determination of ANA. This test allows to determine the presence of the DFS pattern by the use of DFS70 antigen-coated beads (Fig. 2)13.

Fig. 2
figure 2

Illustration of the slide well used for determination of autoantibodies to dense-fine speckled 70 kDa antigen (DFS70). HEp-2 cells are located in the middle compartment. DFS-70 (lens-derived growth factor)-coated beads (15 μm − 2) are localized in the left compartment together with size and location reference (SLR) beads (13 μm − 1). SLR beads always show a uniform green fluorescence, which is due to polymerization of the fluorescent dye with the microbead substance.

Full size image

For the validation of manual CytoBead ANA DFS-70, we obtained 13 anti-DFS70-positive sera from non-SARDs patients that were evaluated by CIA using BIO-FLASH and described elsewhere20. In addition, ten ANA IIF-negative sera samples were used as negative control.

Statistical analysis

The Shapiro-Wilk test was used to test for normal distribution of data. Continuous variables were presented as mean ± standard deviation (sd), categorical variables were presented as counts and/or percentages, where appropriate. Inter-assay agreement between the different assays was assessed by Cohen’s kappa (with 95% confidence interval) using inter-rater statistics analysis. Kappa values were interpreted as following 0.01–0.20 - None-slight agreement, 0.21–0.40 - Fair agreement, 0.41–0.60 - Moderate agreement, 0.61–0.80 - Substantial agreement, 0.81–1.00 - almost perfect agreement21. Fisher’s test was employed to calculate a p value for a 2 × 2 frequency table. To test the difference between paired proportions, p-values were calculated by McNemar’s test. P-values less than 0.05 were considered significant. The sensitivity and specificity of CytoBead ANA 2 for the detection of ANA were calculated with the results of ANA IIF as a reference. Calculations were performed using Medcalc statistical software (Medcalc, Mariakerke, Belgium).

Results

A total of 112 sera samples from 104 patients were assessed by CytoBead ANA 2. BioPlex™ 2200 ANA results were available for all 112 serum samples, and ANA IIF results were available for 103 samples.

The average age in the SARDs group was 48.1 ± 18.9 years, and 19.2% were males. The average age in the non-SARDs group was 56.7 ± 19.9 years, and 40.0% were males. Retrospective collection of clinical data showed that 44 patients did not have SARDs, 52 suffered from SARDs, and data was not available for 8 patients (Table 1). The most common SARD observed was systemic lupus erythematosus (SLE) (42.3%), followed by Sjogren’s syndrome (19.2%) and undifferentiated connective tissue disease (UCTD) (17.3%). Lower frequency SARDs included systemic sclerosis (SSc), central nervous system (CNS) vasculitis, polymyositis, rheumatoid arthritis (RA) and mixed connective tissue disease (MCTD). Demographic data were not available for 14 patients from the non-SARDs group.

ANA and specific autoantibody analysis

The prevalence of ANA and specific autoantibodies by the different assays (CytoBead ANA 2, classical ANA IIF and Bioplex) in the SARDs (n = 52) and non-SARDs (n = 44) groups is presented in Table 2. Analysis by both ANA IIF and CytoBead ANA 2 showed that ANA were more prevalent in the SARDs group (92.0% by ANA IIF, 98.1% by CytoBead ANA 2) than in the non-SARDs group (17.5% and 27.3%, respectively, p < 0.001 for both methods). The prevalences the of disease-specific autoantibodies anti-SS-A/Ro, anti-Sm and anti-Sm/RNP were also significantly higher in the SARDs group (for anti-Sm/RNP only by CytoBead anti-Sm/RNP, p < 0.05). The antibodies with the highest prevalence by CytoBead ANA 2 in the SARDs group were autoantibodies to SS-A/Ro (63.5%), SS-B/La (15.4%), Sm/RNP (15.4%), dsDNA (13.5%), and Sm (13.5%). In the same group, the antibodies with the highest prevalence by BioPlex™ 2200 ANA were anti-SS-A/Ro (57.7%), anti-Sm (21.2%), anti-SS-B/La (13.5%), anti-dsDNA (11.5%), and anti-Sm/RNP (9.6%). The results obtained by CytoBead ANA 2 are shown in Fig. 3, both for detecting ANA and ANA-specific autoantibodies.

Table 2 Prevalence of antinuclear antibodies (ANA) and ANA-specific autoantibodies in SARDs and Non-SARDs groups according to different methods (Cytobead ANA 2, ANA IIF, bioplex™ 2200).
Full size table
Fig. 3
figure 3

CytoBead ANA 2 fluorescence images of antinuclear antibody (ANA) and ANA-specific autoantibody detection. (A) Negative HEp-2 cells (negative control) (B) Cytobead negative control demonstrating only fluorescence of evenly stained reference beads (13 μm) (upper arrow) with unstained SS-A/Ro-coated beats (lower arrow). (C) Positive HEp-2 cells (positive control). (D) Positive CytoBead ANA 2 determination of anti-SS-A/Ro antibody revealed by peripheral fluorescence staining of SS-A/Ro-coated microbeads (15 μm) (arrow).

Full size image

Comparison of classical and cytobead ANA 2 testing

The comparison between CytoBead ANA 2 and classical ANA IIF for the detection of ANA and the comparison between CytoBead ANA 2 and BioPlex™ 2200 ANA for the detection of specific autoantibodies is shown in Table 3. The sensitivity and specificity for the detection of SARDs by ANA IIF were 92.0% (95% CI: 80.8 − 97.8%) and 82.5% (95% CI: 67.2 − 92.7%) and by CytoBead ANA 2 were 98.1% (95% CI: 89.7 − 100.0%) and 72.7% (95% CI: 57.2 − 85.0%).

Table 3 Comparison of different methods for the determination of antinuclear antibodies (ANA) and ANA-specific autoantibodies.
Full size table

A substantial agreement was observed between classical ANA IIF and CytoBead ANA 2 (k = 0.74, 95% CI 0.61–0.88). However, there was a significant difference of 9.71% (95% CI 3.39–16.03, p = 0.006) between both methods by McNemar’s test. This difference was mainly due to 11 samples that were positive by CytoBead ANA 2 and negative by ANA IIF. Notably, 8 out of those 11 samples were positive both by CytoBead ANA 2 and BioPlex™ 2200 for specific autoantibodies, including anti-SS-A/Ro (5 samples, 62.5%), anti-Scl70 (3 samples, 37.5%), and anti-SS-B/La (2 samples, 25%).

Clinical data was available for 8 patients out of those 11 samples: 4 were in the non-SARDs group and 4 were in the SARDs group. Interestingly, 3 out of 4 patients with SARDs had matching results by CytoBead and by BioPlex™ 2200 ANA (2 were positive for anti-SS-A/Ro and were diagnosed with UCTD and Sjogren’s syndrome, whereas 1 was positive for anti-SS-B/La and anti-Scl70 by both assays and was diagnosed with systemic sclerosis).

With regard to the 4 non-SARDs patients with positive ANA by CytoBead and negative ANA by ANA IIF, 2 of those patients had positive specific antibodies by BioPlex™ 2200 (both were positive for anti-SS-A/Ro and one patient was positive for anti-Scl70).

In addition, the patient that had a negative result by CytoBead ANA and a positive result by ANA IIF was classified as having UCTD and had a positive anti-CENP-B result by BioPlex™ 2200.

Comparison of bioplex™ 2200 ANA and cytobead testing for specific autoantibodies

With regard to comparison of specific autoantibody testing by inter-rater statistics, corresponding Kappa values ranged from 0.70 to 0.90 (Table 3), which means a substantial to almost perfect agreement between BioPlex™ 2200 ANA and CytoBead ANA 2. In particular, an almost perfect agreement was observed for anti-SS-B/La (k = 0.90), anti-CENP-B (k = 0.90), anti-SS-A/Ro (k = 0.87), and anti-Jo-1 (k = 0.88), whereas a substantial agreement was revealed for anti-Scl70 (k = 0.79), anti-Sm (k = 0.76), anti-dsDNA (k = 0.72), and anti-Sm/RNP (k = 0.70). According to McNemar’s statistics, there was no significant difference for specific autoantibody testing by both methods (p > 0.05 for all tested antibodies).

However, a sub-analysis of discrepant results showed, for example, that a higher (albeit non-significant) percentage of anti-dsDNA results was detected by CytoBead ANA 2 in contrast to BioPlex™ 2200 ANA (12.5% vs. 8.9%). This finding was more prominent in the non-SARDs group (9.1% vs. 2.3%, Table 2). Thus, four patients within the non-SARDs group had positive anti-dsDNA by CytoBead ANA 2; One had a positive anti-dsDNA by BioPlex™ 2200 ANA as well, and 3 had a negative anti-dsDNA by BioPlex™. Interestingly, 2 of the 3 patients that showed a positive CytoBead ANA 2 result for anti-dsDNA but a negative anti-dsDNA result by BioPlex™ 2200 ANA were also positive for ANA by classical IIF.

Assessment of DFS pattern by CytoBead® ANA DFS-70 test

Manual Cytobead ANA DFS-70 was used to evaluate 13 anti-DFS70-positive sera from non-SARDs patients (previously evaluated by CIA using BIO-FLASH). Positive results were obtained for all 13 samples. Negative results were obtained for all 10 ANA IIF-negative sera samples used as controls.

In addition, manual CytoBead ANA DFS-70 was used to assess 29 samples obtained from non-SARDs patients. All these samples were positive by classical ANA IIF and exhibited a DFS70 pattern. After evaluation with CytoBead ANA DFS-70, positive results (suggesting the presence of anti-DFS70) were obtained for 20 samples (68.9%). Representative results of the simultaneous determination of the DFS pattern and anti-DFS70 by CytoBead ANA-DFS70 is presented in Fig. 4.

Fig. 4
figure 4

Representative results of the simultaneous detection of the DFS pattern on HEp-2 cells and anti-DFS70 autoantibodies on microbeads by CytoBead ANA DFS70. (A) DFS pattern on fixed HEp2 cells with positive staining of the metaphase plates. (B) Confirmation test on the same slide for the presence of anti-DFS70 autoantibodies detected by positive fluorescent staining of DFS70 antigen (lens-derived growth factor)-coated 15 μm microbeads (arrow).

Full size image

Discussion

The currently accepted methodology for ANA testing includes a two-tier setup; an initial assessment by ANA IIF on HEp-2 cells and confirmatory/reflex testing by, for example, multiplexed assays detecting specific autoantibodies. However, different high-throughput multiplexed immunoassays that allow the detection of a panel of ANA-specific autoantibodies have been suggested as an alternative to classical ANA IIF6,7.

However, discrepancies between different methods have previously been reported regarding the assessment and interpretation of ANA and specific ANA-related autoantibodies8. For example, anti-Jo-122, anti-ribosomal P23 or anti-SS-A/Ro24 may be detected by multiplexed assays in patients who are negative for ANA detected by IIF. An additional example is the presence of anti-dsDNA detected by ELISA in patients with a negative ANA result by IIF, which raises difficulties regarding the interpretation of such results in the context of SLE diagnosis. Antinuclear antibody positivity has recently been introduced as an entry criterion for SLE underlining the importance for a sensitive assay for this parameter25. Moreover, several studies have reported that ANA-negative SLE patients have anti-dsDNA antibodies suggesting that specific dsDNA epitopes may be not detected by HEp-2 IIF screening tests26,27.

Accordingly, in patients with a high clinical suspicion for SARDs, if the results of the alternative assays (such as multiplexed assays or ELISA) are negative or a discrepancy exists between the results of classical ANA IIF and alternative assays, it is recommended to perform a confirmation/reflex test by a different method7. Thus, there is clearly a demand for advanced methods that will be able to reliably evaluate both the presence of ANA and the presence of specific autoantibodies in one reaction environment.

In this study, we assessed the manual version of the CytoBead platform, which simultaneously can perform classical ANA analysis on HEp-2 cells and multiplexed detection of autoantibodies by microbead immunoassay. According to our results, a substantial agreement was observed between classical ANA IIF and manual CytoBead ANA 2 (k = 0.74), yet a significant discordance between the two methods by McNemar’s statistical test. This discrepancy was mainly associated with 11 samples that were positive by CytoBead ANA 2 and negative by ANA IIF. As outlined in the results section, 8 samples (72.7%) had matching disease-specific autoantibodies by CytoBead ANA 2 and by BioPlex™ 2200, which raises the possibility of false negative ANA IIF results. The most common detected autoantibody in ANA IIF-negative samples was anti-SS-A/Ro (5/8, 62.5%), which is consistent with the published literature11,24,28. For instance, it was reported that the 36% of samples positive for SS-A/Ro52, and 41% for SS-A/Ro60 were not be detected when IIF was used as the only screening method28. In an earlier study, which tested 1840 consecutive serum samples, anti-SS-A/Ro was detected in 11 (0.6%) patients with negative ANA IIF results. Notably, 6 of those patients were diagnosed with rheumatoid arthritis, SLE, or Sjogren’s syndrome24.

In our cohort, analysis of available clinical data showed that in 3/4 patients with SARDs (UCTD, Sjogren’s syndrome and systemic sclerosis), who had negative ANA by classical IIF, matching results for specific autoantibodies were obtained by CytoBead ANA 2 and BioPlex™ 2200 ANA (anti-SS-A/Ro in two patients, and anti-SS-B/La together with anti-Scl70 in one patient). These results are in line with a previous comprehensive study, which evaluated the diagnostic performance of CytoBead ANA 2 in 697 serum samples obtained from SARDs and non-SARDs patients as well as from blood donors11. Thus, four ANA IIF-negative patients with RA and Sjogren’s syndrome exhibited anti-SS-A/Ro or anti-CENP-B detected by CytoBead ANA 2. Thus, the simultaneous testing approach of CytoBead ANA 2 managed to identify patients with SARDs in whom a negative ANA IIF result was initially obtained.

With regard to specific autoantibody testing, a generally good agreement was found between CytoBead ANA 2 and BioPlex™ 2200 with kappa values ranging from 0.70 to 0.90 and no statistically significant discrepancies. Remarkably, an almost perfect agreement was observed for anti-SS-B/La, anti-CENP/CENP-B, anti-Jo-1 and anti-SS-A/Ro (k > 0.80, respectively). A substantial agreement was observed for anti-Scl70, anti-Sm, anti-dsDNA, and anti-Sm/RNP (k ≥ 0.70).

A sub-analysis of discrepant results showed that a higher percentage of samples from non-SARDs patients were positive for anti-dsDNA by CytoBead ANA 2 in comparison with Bioplex™ 2200 (4 patients, 12.5% vs. 1 patient, 2.3%). Notably, 2/3 patients that were negative for anti-dsDNA by Bioplex™ 2200 tested positive for ANA on HEp-2 cells by IIF. This finding has raised the possibility that those patients may develop a SARD in the future. An alternative explanation could be a lower specificity of the anti-dsDNA assessment by the CytoBead ANA 2 in comparison with Bioplex™ 2200.

In the second part of our study, serum samples obtained from non-SARDs patients that were positive by ANA IIF and exhibited a DFS pattern were analyzed by the manual Cytobead ANA DFS-70 assay. It was found that 20 samples (68.9%) exhibited anti-DFS70 by microbead assay. According to the literature, the prevalence of anti-DFS70 antibodies was reported from 41 to 91% in subjects with DFS pattern on ANA IIF29,30,31. This wide range of reported prevalences seems to be mainly associated with intra-laboratory discrepancies and lack of standardization, possibly associated with heterogeneity of the antigen causing the DFS pattern or with different sources of anti-DFS70 assays antigens. A previous study to evaluate the performance of CytoBead ANA DFS-70 with regard to the identification of autoantibodies to DFS70 investigated a cohort of 541 individuals that were referred for ANA screening13. This study showed a high specificity of 95.8% but low sensitivity of 54.4% compared with the gold standard IIF test when the CytoBead HEp-2 substrate was only used for ANA testing. However, when the simultaneous anti-DFS70 detection by microbead assay was added, the sensitivity of the CytoBead test increased to 99.1% with respect to samples that showed a DFS pattern by classical ANA IIF confirmed by two independent anti-DSF70 tests.

In routine ANA testing, the prevalence of the DFS pattern ranges from 0.8 to 12.3%, depending on the study population and the diagnostic test performance32,33,34,35. The latter can be influenced by various HEp-2 substrates and fixation methods. A DFS pattern may be associated with the presence of anti-DFS70, which are common in ANA-positive non-diseased subjects and extremely rare in patients with SARDs15,32,36,37,38,39. Moreover, the presence of isolated anti-DFS70 may not only rule out the diagnosis of current SARDs, but also the development of future SARDs32,36,37,38,39. However, the DFS pattern may also be associated with specific ANA including anti-dsDNA and ENA observed in SARDs patients13. Therefore, when a DFS pattern is identified on ANA IIF screening assay in patients with a high clinical suspicion for SARDs, it is recommended to perform a confirmatory test for disease-specific autoantibodies such as anti-dsDNA. In contrast, patients with low clinical suspicion for SARDs and a positive DFS pattern by IIF may be directly tested for anti-DFS70 by a confirmatory assays (ELISA, CIA, LIA and dot-blot) to exclude the presence of a SARD13,29,33,38.

In the present study, anti-DFS70 was found in 68.9% of ANA IIF-positive serum samples with a DFS pattern using the single-step CytoBead ANA DFS-70 assay. Notably, all these samples were obtained from patients without known SARDs. These results show that the CytoBead ANA DFS-70 assay may potentially assist in the exclusion of SARD diagnosis in ANA positive patients and prevent unnecessary and costly laboratory tests and clinical follow up.

Our study has certain limitations. One such limitation is a probable selection bias. The study was performed on routine serum samples that were sent for the assessment of ANA. No pre-selection of patients was made and therefore the age and gender were not matched within the study cohorts.

However, it should be noted that the main objective of the study was to assess the performance of the different methods, and not to compare different cohorts of patients. Additionally, ANA IIF results were not available for 9 patients, so only comparison with Bioplex was performed for those patients. Finally, the number of participants in this study was limited by the inclusion of only 141 serum samples in the study, which limits the significance of the conclusions.

Conclusion

The present study indicates a good agreement between manual CytoBead ANA 2 and ANA IIF testing followed by BioPlex™ 2200 regarding ANA screening and detection of ANA-specific autoantibodies. In addition, CytoBead ANA DFS-70 is an effective and reliable test that allows the detection of anti-DFS70 associated with an ANA IIF DFS pattern. CytoBead ANA 2 is a novel one-step assay that enables simultaneous ANA screening and confirmation tests and, thus, represents a promising alternative approach to the time-consuming and costly two-tier ANA analysis. However, as the number of samples in this study was relatively small, additional studies may be warranted.