Comparison of the analytical and clinical performances of two different routine testing protocols for antinuclear antibody screening

Abstract Background The diagnosis of systemic autoimmune rheumatic diseases (SARD) is based on the detection of serum antinuclear antibodies (ANA) for which indirect immunofluorescence (IIF) is the golden standard. New solid‐phase immunoassays have been developed to be used alone or in combination with the detection of extractable antinuclear antibodies (ENA) to improve SARD diagnosis. The purpose of this study was to compare the clinical performances of different ANA screening methods alone or in combination with ENA screening methods for SARD diagnosis. Methods A total of 323 patients were screened for ANA by IIF, EliA™ CTD Screen, and ELISA methods. Agreements were calculated between the methods. Then, EliA™ CTD Screen positive samples were screened for ENA by line immunoassay (LIA) and fluorescence enzyme immunoassay (FEIA). Results The diagnostic accuracy of EliA™ CTD Screen (79% sensitivity and 91% specificity) was better than that of ELISA or IIF. The combination of EliA™ CTD plus IIF had the highest sensitivity (93%). ENA determination revealed that Ro52 and Ro60 were the most prevalent specificities. The use of IIF alone was not able of detecting up to 36% of samples positive for Ro52, and 41% for Ro60. Conclusions EliA™ CTD Screen has a better diagnostic performance when compared to IIF and ELISA. The combined use of EliA™ CTD Screen and IIF clearly improves the rate and accuracy of SARD diagnosis. The use of EliA™ CTD Screen as first‐line screening technique allows the detection of antibodies, which could not be detected by IIF alone.


| INTRODUC TI ON
Systemic autoimmune rheumatic diseases (SARD), also known as connective tissue diseases (CTD), including all diseases triggered by the formation of immune complexes that enter the circulation, are then deposited in different tissues and organs, and cause damage. 1 The detection of antinuclear antibodies (ANA) and of antibodies to extractable nuclear antigens (ENA) is used in the diagnosis of SARD and in the prediction of an early onset of disease. 2 Indirect immunofluorescence (IIF) on HEp-2 cells (human epidermoid laryngeal carcinoma cells) has been defined as the reference screening method for ANA in the clinical laboratory routine. 3 However, IIF is a labor-intensive, time-consuming procedure and has poor reproducibility due to the subjective interpretation of results. 4 Therefore, in 2014, an international workgroup of experts representing 15 European countries developed a set of recommendations for the appropriate assessment and interpretation of ANA detected by different methods. According to this expert panel, alternative assays might be preferred for ANA screening when clinical suspicion is strong; and when the results of these alternative methods are negative, IIF should be used for the definitive diagnosis. 3 In this regard, various automated solid-phase immunoassays (such as enzyme-linked immunosorbent assay (ELISA) and fluorescence enzyme immunoassay-based (FEIA) assays) have been developed to be used as first-line screening methods in SARD diagnosis. 2 ELISA is a plate-based assay technique in which antigens of HEp-2 cell extracts are immobilized on a solid surface, whereas FEIA is designed as a sandwich assay where a mix of antigens is coated to the solid phase.
Antinuclear autoantibodies bind to these antigens, which are coupled to an enzyme-linked antibody, producing a fluorescent signal upon binding. 2 Such systems are attractive alternatives to IIF, not only because of the automated process but also because of the improved specificity compared to IIF. Moreover, it has been widely demonstrated that a combination of these two techniques may improve the diagnostic accuracy of ANA screening. [5][6][7][8] In routine laboratory testing, after a positive result is obtained on a screening platform, it is useful to determine the specificity of the antibody using different ENA testing platforms due to their prognostic and diagnostic power. Taking into account, the availability of different methods for SARD diagnosis, it is important to evaluate the performance of different tests in order to determine which ANA and ENA combination performs best and is more sustainable.
The aim of this study was the comparative analysis of three different ANA screening methods (EliA™ CTD Screen, IIF, and ELISA).
An analysis was carried out considering single or combination tests (EliA™ CTD Screen plus IIF vs. ELISA plus IIF). We also evaluated the potential ANA plus ENA combination testing in SARD diagnosis using two different ENA screening methods (LIA and FEIA).

| Antinuclear antibodies (ANA) Screening
ANA screening was performed with the following three techniques:  For combined testing (ie, when more than one of these tests were used), patients with a positive result in one of the tests were considered positive, and patients with negative results in both tests were considered negative.

| Extractable nuclear antigens (ENA) Screening
EliA ™ CTD Screen positive samples (n = 123) were analyzed for the following ENA specificities using two different methods:

| Statistical analysis
The differences between the results obtained from different methods were analyzed by chi-square, phi coefficient, and contingency coefficient tests. Cohen's kappa coefficients were used to estimate the measuring agreement among methods. Statistical analyses were performed using the software SPSS Statistics v25.

| Demographics
In this patient cohort, the prevalence of SARD was 41.2% (116 of 246 women, and 17 of 77 men). The median age of the patients was 56 ± 16 years. Table 1 describes the demographic data for each group.

| Comparison of ANA Screening platforms
The results of ANA screening with IIF, ELISA, and EliA™ CTD Screen were obtained and compared. As shown in Table 2 183 of the results were consistent (56.7%) and 140 results (43.3%) were contradictory between IIF and EliA™ CTD Screen techniques, whereas 251 of the results were consistent (77.7%) and 72 results (22.3%) were contradictory between ELISA Screen and EliA™ CTD Screen techniques.
Bivariate analysis between EliA™ CTD Screen, IIF, and ELISA techniques measured by Pearson's chi-square test revealed statistically significant differences between each test (p < 0.05) ( Table 3).
Phi coefficient and contingency coefficient were also calculated and presented in Table 3. Method agreement analysis (Kappa coefficient) revealed that the agreement between EliA™ CTD Screen and ELISA (moderate, 0.540) was higher than the agreement between EliA™ CTD and IIF (weak, 0.150) or between ELISA and IIF (weak, 0.203).
Importantly, the agreement between combined tests (EliA™ CTD plus IIF vs ELISA plus IIF) was stronger (strong, 0.814). Discrepant results are shown in Table4.
Next, SARD diagnosis performances of three screening platforms were compared (

| Specific ENA profile
The antigenic specificities were assessed using two methods: EliA™ and LIA. Both platforms displayed the highest positivity rates for Ro60 and Ro52 antibodies that are most commonly reported in the literature for SARD diagnosis. Therefore, the percentage of these antigens (as calculated from the total positive specificities of each method) was assessed in systemic autoimmune diseases (Table 6).

| DISCUSS ION
This study included patients with suspicion of SARD referred from primary care, rheumatology, nephrology, and internal medicine departments. It is important to distinguish these different groups of patients which generally are referred to as SARD or ANA-associated rheumatic diseases (AARD) as there is considerable variability among them. [9][10][11] In our study cohort, the percentages of patients detected by IIF (55%, n = 175) were similar to those obtained in the routine work of our laboratory and in other studies such as Bizzaro et al. 5 and Dellavance et al. 12 which reported prevalence rates of 46.7% and 44.3%, respectively, which also used a 1/80 dilution as a cut-off point as we used in this study.
The comparative analysis of the three screening methods revealed that the sensitivities and specificities of the IIF and ELISA methods were more similar to each other (69% sensitivity in both methods, and 56% and 74% specificity in IIF and ELISA, respectively), while EliA™ CTD Screen had higher sensitivity and specificity levels for SARD (79% sensitivity, 91% specificity). Positive and negative LRs (8.33 and 0.23, respectively) were also better for the two studied methods. 15 It should also be noted that the performance of IIF may vary among laboratories, and it is more consistent in fully automated tests. 16 When the diagnostic efficiency of combined tests was evaluated (EliA™ CTD Screen plus IIF, and ELISA plus IIF), sensitivity and negative LR values were observed to be increased compared to EliA™ CTD Screen or ELISA alone measurements. In particular, EliA™ CTD Screen plus IIF combination results were highly promising with high sensitivity and low negative LR (sensitivity = 93% and negative LR = 0.14).
Although combination tests had lower specificity and positive LR values compared to individual tests, sensitivity was higher and negative LR was below the limits considered clinically significant. Higher measuring agreement has been observed between EliA™ The study has three main limitations. The first is sample size.
The cohort size (n = 323) was limited; however, it can be considered representative of the routine work of a laboratory specialized in the diagnosis of SARD. The second relies on the fact that the analysis of antibodies against the different antigenic specificities associated with SARD was performed exclusively in patients with positive EliA™ CTD Screen following the usual protocol of the laboratory. Finally, neither disease duration nor concurrent therapies and disease activity were considered for the analysis, so its impact on the results has not been evaluated.
In summary, EliA™ CTD Screen presents better diagnostic performance results versus IIF and ELISA individual measurements.
The combined use of IIF with a screening test based on the use of solid-phase fixed antigens, such as EliA™ CTD Screen, clearly improves the detection capacity of SARD (sensitivity = 93% and LR− = 0.14). Furthermore, the use of EliA™ CTD Screen as firstline screening technique allows the detection of Ro52 and Ro60 antibodies (considered two of the most prevalent antibodies in SARD) which would not be detected by IIF alone, thus increasing the false-negative rate.

ACK N OWLED G EM ENT
The authors greatly acknowledge Thermo Fisher Scientific for providing EliA™ reagents, and also the Virgen Macarena University Hospital for funding.

CO N FLI C T O F I NTE R E S T
Dr. González has received a speaker honorarium from Thermo Fisher Scientific.

AUTH O R CO NTR I B UTI O N S
All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

E TH I C S A PPROVA L
This study was approved by the Ethics Committee of the University of Virgen Macarena Hospital (Approval reference number: 201932115312). Authors declare that the manuscript has not been submitted to any other journal for simultaneous consideration, has not been published previously and data have not been manipulated or copied from other authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.