Abstract
Background:
Indirect immunofluorescence (IIF) assays are recommended as the gold standard method for the detection of antinuclear antibodies (ANAs). This study aimed to investigate the reliability of an automated system.
Methods:
We compared 3745 serum samples using NOVA View archived images with manual analysis via microscopy. A custom cutoff value was established to distinguish ANA titers and was validated in two clinical laboratories. The automatic ANA pattern recognition system was evaluated, and all ANA-positive sera were subjected to two commercial ANA IIF kits to compare the consistency of the pattern interpretation results. For inconsistent patterns, a third ANA IIF testing kit was utilized.
Results:
Agreement of the interpretation of the ANA IIF test using the platform of NOVA View and manual microscopy was 96.9%. The local cutoff value to discriminate ANA titers in four main ANA patterns was calculated based on 1390 serum samples. In our laboratory, the titer prediction accuracy was superior to the preset cutoff in NOVA View (p<0.01); the performance was similar in another laboratory (p=0.11). The automatic pattern recognition accuracies of speckled, homogeneous, centromere, nucleolar and nuclear dot patterns were 62.7%, 57.4%, 92.6%, 30.5% and 27.3%, respectively. The consistency of the pattern interpretation results between INOVA and MBL kits was 95.3%.
Conclusions:
It is necessary to establish a custom value-added ANA report. However, confirmation of the digital immunofluorescence images by expert technicians was essential, and suspect results of an ANA pattern should be reconfirmed by another commercial ANA IIF kit to achieve more reliable results.
Introduction
An antinuclear antibody (ANA) screening test is a basic tool in the diagnosis and management of patients with connective tissue diseases (CTD) [1], [2]. Recently, numerous methods have been implemented in ANA screening tests, including enzyme-linked immunosorbent immunoassay (ELISA), automated chemiluminescent immunoassay [3], microarray systems [4] and indirect immunofluorescence (IIF) assays. Although some high-throughput methods have the advantage of rapid detection, IIF using HEp-2 (human epidermoid laryngeal carcinoma) cells remains the reference technique for ANA screening, which is the initial step in diagnosing systemic autoantibodies in conditions such as systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), Sjögren’s syndrome (SjS) and undifferentiated connective tissue disease (UCTD) [5].
The main limitations of IIF include the relatively low standardization and automation levels, subjective evaluation, intra- and interlaboratory variability and the requirement of professional morphologists [6], [7], [8]. For laboratories, particularly nonspecialized laboratories that employ manual interpretation, the results obtained among different users may vary [9], [10].
Therefore, optimization of ANA testing, including both ANA slide processing and microscopic analysis, is particularly important. Recently, many studies focused on the comparisons between automatic and manual microscope reading or discrepancies between various computer-aided ANA pattern diagnostic systems [11], [12], [13], [14], [15], [16]. However, few studies evaluated the reliability of the autotiter evaluation system [17]. Could it really provide a quantitative reading, which was correlated with the end-point titer?
In this study, we assessed the performance of a fully automated IIF-HEp-2 recognition system, including both the pattern interpretation ability and the autotiter evaluation system. Moreover, a custom cutoff value (the immunofluorescence intensity value) of different ANA titers according to various patterns is discussed and evaluated at two different laboratories.
Materials and methods
Patient serum sample and clinical information collection
A total of 3745 consecutive and non-duplicated serum samples were collected from different patients undergoing immunological tests to measure routine autoantibody levels in a comprehensive teaching hospital in Shanghai, China (Renji Hospital affiliated with Jiaotong University), from February to March 2016. This hospital is located in the center of Shanghai and is a large (1600 beds) teaching hospital that manages approximately 8000 admissions per day.
Anti-extractable nuclear antigen (ENA) antibodies were identified by blot immunoassay (Euroimmun AG, Lübeck, Germany) comprising an IgG autoantibody panel for seven different antigens: SSA (60-kDa natural subunits), SSB (La), Sm, nRNP/Sm (U1-RNP), Jo-1, Scl-70 and ribosomal P. Antinucleosome antibodies (ANuA) (Euroimmun AG, Lübeck, Germany) and anti-double-stranded DNA (dsDNA) (Trinity Biotech, Wicklow, Ireland) were measured using an ELISA method according to the manufacturer’s instructions and cutoff value. Moreover, the clinical diagnoses of the patients were recorded to determine the clinical significance of the different HEp-2-cell-IIF patterns and titers.
Ethical approval and consent to participate
All patient samples collected in this study were approved by Institutional Review Board of the Renji Hospital. No consent was needed in this study.
Evaluation of the NOVA View ANA-IIF pattern reading system
Microscope slide (INOVA Diagnostics, San Diego, CA, USA) processing, from sample dilution to the final wash steps, was completely performed using Quantlyser instrument (INOVA Diagnostics, San Diego, CA, USA). All ANA IIF testing slides were manually analyzed by two technicians using an immunofluorescent (EUROStar-III, Euroimmun AG, Lübeck, Germany) microscope; the slides were also automatedly evaluated by NOVA View with software version 1.0.4.3 by a certified medical technologist. According to the manufacturer’s instructions, a value of 49 measured light intensity units (LIU) was selected as the cutoff for NOVA View. Agreement between the manual and the NOVA View assessments was recorded.
Training and validation of custom ANA titers in the calculation of standards for four basic patterns
To further establish a standardized workflow for ANA titer analysis, 1390 ANA-positive sera, including four basic patterns (speckled, homogeneous, centromere and nucleolar), were applied to establish a cutoff LIU value of different ANA titers according to various patterns. All positive serum samples were diluted from 1:80 to 1:2560 in phosphate-buffered saline and automatedly analyzed using NOVA View to confirm the end-point titer. Samples with a positive ANA titer greater than 1:1280 were recorded as >1:1280. An LIU value of 49 was identified as negative for each pattern.
To validate the applicability and stability of the custom ANA titer cutoff value, an additional 94 ANA-positive samples were independently processed using a Quantlyser instrument and analyzed by digital IIF, which was captured by NOVA View in two different clinical laboratories of two comprehensive teaching hospitals in Shanghai (clinical laboratories of Renji Hospital and Ruijin Hospital). Accuracy was calculated as the difference in the evaluated and end-point titers within plus and minus one dilution. In addition, the ANA titer distribution and anti-ENA profiles of the 3745 serum samples from patients with definitive diagnoses of SLE, RA, SjS, and UCTD were analyzed.
Quality control of ANA with the automated ANA test system
Quality control was vital throughout the entire course of the ANA screening test. In this study, quality control, including positive and negative standards provided in the ANA test kit (INOVA Diagnostics, San Diego, CA, USA), and a moderate ANA level (homogeneous, end-point titer 1:320) were tested in parallel with the patient’s samples. The coefficients of variation (CVs) of the LIU values for each control were calculated.
ANA pattern recognition with NOVA View
Five basic ANA IIF patterns, including speckled, homogeneous, centromere, nucleolar and nuclear dots, were identified based on the software algorithm in the NOVA View reports system. The patterns distinguished by both the automated ANA pattern recognition system and the digital images confirmed by technologists were compared. In addition, all ANA-positive serum samples were manually processed using the commercial FluoHEpana Test kit (MBL, Tokyo, Japan) at the same dilution at 1:80 to compare the consistency of the two products. If an inconsistent pattern was obtained, a third ANA IIF testing kit (Euroimmun AG, Lübeck, Germany) was used (1:80 dilution).
Statistical analysis
Agreement between the two detection methods, including the manual/NOVA View system and various ANA IIF test kits, was analyzed via contingency tables and κ statistics with the 95% confidence interval (95% CI). The cutoff values of each ANA titer for the different patterns were calculated using receiver operating characteristic (ROC) curves with the maximum value of the sum of specificity and sensitivity. SPSS (IBM-SPSS, Inc., Armonk, NY, USA) was used for statistical calculations. Two-sided p<0.05 was considered significant.
Results
Comparison of positive and negative findings between automated and visual ANA testing
A total of 3745 consecutive serum samples were enrolled to assess agreement between NOVA View-aided diagnostic system and visual microscope interpretation. Among 3745 samples, 292 patients were clearly diagnosed with CTD, including 140 SLE, 65 UCTD, 39 SjS, 35 RA, four systemic sclerosis, three Raynaud’s phenomenon, three dermatomyositis, two mixed CTD and one polyarteritis nodosa patients, whereas 347 patients had non-CTD disease and enrolled as disease control group. In addition, for all the other samples, clear clinical diagnosis were unacquirable.
Among all serum samples, NOVA View reported 1697 (45.3%) positive and 2048 (54.7%) negative, compared to 1642 (43.8%) positive and 2103 (56.2%) negative reported by a visual microscope interpretation the ANA results. The results are presented in Table 1. The findings indicate that the total agreement of the two methods is 96.9% (95% CI 96.6%–97.2%), with a positive agreement of 98.1% (95% CI 97.8%–98.4%) and a negative agreement of 95.9% (95% CI 95.5%–96.3%). The strength of agreement was excellent (κ=0.937, 95% CI 0.925–0.949).
Comparison of NOVA View and visual interpretation of ANA testing.
| Visual interpretation | |||
|---|---|---|---|
| Positive | Negative | Number (percentage) | |
| NOVA View interpretation | |||
| Positive | 1611 | 86 | 1697 (45.3%) |
| Negative | 31 | 2017 | 2048 (54.7%) |
| Number (percentage) | 1642 (43.8%) | 2103 (56.2%) | 3745 (100%) |
A vital characteristic of an automated ANA detection system is the ability to reliably identify negative results. In our study, 31 serum samples recognized as positive were not identified by NOVA View. Among these 31 samples, five were identified as negative because of an unclear pattern, whereas the LIU values were greater than 48 according to NOVA View. One sample that was interpreted as negative (LIU=15) by the automated system and reported as positive with cytoplasmic fluorescence tested as Jo-1-positive by blot immunoassay. For the remaining 30 serum samples, antibodies against ENA were not found, or there was no definitive CTD diagnosis. In addition, no dsDNA and ANuA were detected in all the 31 samples.
In addition, 86 serum samples with positive results by NOVA View were identified as negative by manual immunofluorescent microscopy. The average LIU of the 86 samples was 104.6±43.1 (ranging from 49 to 211). However, among these 86 samples, for which visual interpretation showed negativity, anti-SSA Ro 60 was detected in 2; one sample was Ro60+U1RNP positive, one sample was Ro60+ribosomal P and two samples were dsDNA positive. With respect to clinical diagnosis, among these 86 samples, four patients were diagnosed with RA, two patients with SLE and one patient with SjS.
Training and validation of the cutoff value of the automated ANA titer system
Totally 1390 serum samples with ANA-positive results, including 622 speckled, 523 homogeneous, 163 centromere and 82 nucleolar patterns, were utilized in the training set. The details of the average intensity obtained by NOVA View, the optimal cutoff value of the different patterns in each end-point titer, which ranged from 1:80 to 1:1280, and the area under the ROC curve with the 95% CI are provided in Figure 1 and Table 2.
Fluorescence intensity distribution for the four basic ANA patterns.
(A) Distribution of the light intensity obtained by NOVA View in four main ANA patterns with different end-point titers. (B) Distribution of the cutoff value of the four basic ANA patterns using the automatic ANA titer system.
Cutoff value training results using the NOVA View diagnostic system.
| Pattern | End-point titer | Sample number | Average intensity (95% CI) | LIU cutoff | Area under ROC (95% CI) |
|---|---|---|---|---|---|
| S | 1:80 | 79 | 134.3 (117.5–151.0) | 49–253 | 0.96 (0.95–0.98) |
| 1:160 | 107 | 316.4 (281.1–351.8) | 254–504 | 0.96 (0.94–0.97) | |
| 1:320 | 99 | 586.8 (524.1–649.4) | 505–694 | 0.96 (0.95–0.97) | |
| 1:640 | 98 | 1090.9 (940.5–1241.3) | 695–1216 | 0.96 (0.94–0.97) | |
| 1:1280 | 95 | 1891.1 (1685.1–2097.2) | 1217–1518 | 0.95 (0.93–0.96) | |
| >1:1280 | 144 | 3129.9 (2955.2–3304.6) | ≥1519 | – | |
| H | 1:80 | 128 | 153.6 (111.0–196.2) | 49–287 | 0.97 (0.95–0.98) |
| 1:160 | 121 | 429.1 (380.3–477.8) | 288–639 | 0.95 (0.94–0.97) | |
| 1:320 | 91 | 745.1 (682.3–807.8) | 640–1063 | 0.97 (0.95–0.98) | |
| 1:640 | 62 | 1166.3 (1073.4–1259.1) | 1064–1387 | 0.97 (0.97–0.99) | |
| 1:1280 | 57 | 1944.3 (1750.8–2137.8) | 1388–1397 | 0.95 (0.93–0.96) | |
| >1:1280 | 64 | 2155.9 (2040.6–2271.2) | ≥1398 | – | |
| ACA | 1:80 | 3 | 70.0 (28.21–111.8) | 49–114 | 0.99 (0.98–1.00) |
| 1:160 | 3 | 243.0 (225.6–260.4) | 115–253 | 0.98 (0.96–1.00) | |
| 1:320 | 21 | 325.2 (269.3–381.2) | 254–528 | 0.96 (0.93–0.99) | |
| 1:640 | 34 | 618.5 (529.1–707.9) | 529–715 | 0.92 (0.89–0.96) | |
| 1:1280 | 67 | 1409.0 (1193.0–1625.0) | 716–1015 | 0.80 (0.73–0.87) | |
| >1:1280 | 35 | 1743.3 (1463.5–2023.1) | ≥1016 | – | |
| N | 1:80 | 17 | 124.2 (89.4–158.9) | 49–269 | 0.95 (0.91–1.00) |
| 1:160 | 16 | 399.4 (190.2–608.6) | 270–401 | 0.93 (0.86–1.00) | |
| 1:320 | 22 | 581.1 (488.1–674.2) | 402–786 | 0.92 (0.87–0.98) | |
| 1:640 | 13 | 780.9 (642.7–919.2) | 787–904 | 0.98 (0.95–1.00) | |
| 1:1280 | 9 | 1276.9 (1026.7–1527.1) | 905–1522 | 0.99 (0.98–1.00) | |
| >1:1280 | 5 | 2039.8 (1577.6–2502.0) | ≥1523 | – |
S, speckled; H, homogeneous; ACA, centromere; N, nucleolar; LIU, light intensity unit. The average intensity was obtained at a dilution of 1:80.
In addition, we assessed the automated ANA titer system preset in NOVA View. The rates of perfect match, indicating complete agreement between the automated dilution and the end-point titer, were 44.1%, 45.9%, 33.7% and 47.6% for the speckled, homogeneous, centromere and nucleolar patterns, respectively. Because plus or minus one dilution was acceptable, the automated dilution recommended system showed an accuracy of 98.4%, 98.3%, 94.5% and 96.3%, respectively, for these four patterns.
To test the performance of the cutoff value established in our laboratory, 94 independent ANA-positive samples, including 47 speckled, 36 homogeneous, nine centromere and two nucleolar ANA patterns with end-point titers ranging from 1:80 to 1:2560, were utilized for validation. The prediction performance in the two clinical laboratories was analyzed and compared to the automated titer system of NOVA View (Table 3). With regard to exact accuracy, indicating that the recommended dilution was the same as the end-point titer, the laboratory at Renji Hospital showed better performance than the automated titer system (57.4% vs. 43.7%, respectively, p<0.01), although this was not the case for Ruijin Hospital. Regarding the acceptable accuracy rate, indicating that the prediction dilution was plus or minus one dilution, the laboratory of Renji Hospital performed as well as the automated titer system. However, the acceptable accuracy rate of the laboratory at Ruijin Hospital was inferior to that of NOVA View software (86.2% vs. 97.8%, respectively, p<0.01).
Comparison of the performance between the two clinical laboratories and the NOVA View automatic diagnosis system according to the established cutoff value.
| Accuracya (95% CI) | p-Valueb | Acceptable accuracyc (95% CI) | p-Valued | |
|---|---|---|---|---|
| Laboratory Ae | 57.4% (47.5–67.4%) | <0.01 | 98.9% (96.9–100%) | 0.45 |
| Laboratory Bf | 47.9% (37.8–58.0%) | 0.11 | 86.2% (79.2–93.1%) | <0.01 |
| Automatic titer systemg | 43.7% (41.1–46.3%) | – | 97.8% (97.0–98.5%) | – |
aAccuracy indicates that the prediction dilution is exactly the end-point titer. bχ2-test of the accuracy between the laboratory and automated titer system. cAcceptable accuracy indicates that the prediction dilution is plus or minus one dilution. dχ2 test of the acceptable accuracy between the laboratory and automated titer system. eLaboratory of Renji Hospital. fLaboratory of Ruijin Hospital. gThe automated titer system included the automated ANA titer system built into the software of NOVA View.
Quality control for the automated ANA quantitative assay
Quality control was tested in parallel with the patient samples 24 times using the same lot number of the test kit. The line chart in Supplemental Figure S1 indicates the intensity of the quality control. The average values were equal to 2037.9, 5.9 and 708.9, SD 275.3, 1.1 and 50.3, and the CV values (%) were 13.5%, 18.6% and 7.1% for the positive standards, negative standards and weakly positive pooled sera (homogeneous, end-point titer 1:320), respectively.
Antibody determination in CTD and disease control group patients
In 292 patients clearly diagnosed as CTD, the antibody determination results of 140 SLE, 65 UCTD, 39 SjS and 35 RA patients were analyzed (Supplemental Figure S2 and Supplemental Table S1). One hundred and thirteen patients (113/327, 32.6%) in disease control group showed positive ANA results. The median ANA titers for these SLE, UCTD, SjS, RA patients and 327 non-CTD patients in disease control group were 1:640, 1:640, 1:640, 1:320 and negative, respectively. The most frequent ANA pattern of the four diseases was speckled, with the exception of RA patients, who more commonly showed a homogeneous pattern (H vs. S: 68.6% vs. 22.9%, respectively).
Evaluation of automated pattern recognition
There were 1401 in 1611 serum samples within five basic patterns, 403 (28.8%) were reported as “unrecognized” by NOVA View software. For each pattern, the recognition accuracies of the speckled, homogeneous, centromere, nucleolar and nuclear dot patterns were 62.7%, 57.4%, 92.6%, 30.5% and 27.3%, respectively.
Moreover, to compare the consistency of the two products, in all 1611 ANA-positive samples, 76 serum (4.7%) samples showed varying pattern recognition results. The main difference was related to discrimination between the speckled and homogeneous patterns and the staining of the nucleolar pattern. The details of the disagreement are listed in Table 4. After confirmation using a third commercial kit (Euroimmun), two samples demonstrated varying results in the three different commercial kits (Table 4 and Figure 2). When assessed by a single technologist, one patient (Figure 2A–C) was classified with cytoplasmic speckled, nucleolar and speckled patterns using the INOVA, MBL and Euroimmun ANA kits, respectively, whereas another patient (Figure 2D–F) was classified with homogeneous, homogeneous nucleolar and speckled patterns, respectively. For the other 74 samples, following confirmation with the third ANA IIF test commercial kit, agreement between INOVA and Euroimmun kits was 59.5% (44/74), whereas agreement between MBL and Euroimmun was 40.5% (30/74).
Discrimination between 74 samples by three commercial kits.
| Num | INOVA pattern | MBL pattern | Euroimmun pattern (n) |
|---|---|---|---|
| 13 | H | S | H (1), S (12) |
| 32 | S | H | H (5), S (27) |
| 6 | N (Pos) | N (Neg)a | N (Pos) (5), N (Neg) (1)a |
| 18 | N (Neg)b | N (Pos) | N (Pos) (7), N (Neg) (11)b |
| 3 | Dots (Pos) | Dots (Neg)c | Dots (Pos) (3), Dots (Neg) (0)c |
| 1 | Dots (Neg)d | Dots (Pos) | Dots (Pos) (0), Dots (Neg) (1)d |
| 1 | Golgi-like (Pos) | Golgi-like (Neg)e | Golgi-like (Pos) (1)e |
n, sample number; H, homogeneous; S, speckled; N, nucleolar; Dots, nuclear dots. aSix samples with nucleolar pattern by INOVA were five homogeneous, one speckled by MBL and five nucleolar, one homogeneous by Euroimmun. bEighteen samples with nucleolar pattern by MBL were 14 speckled, four homogeneous by INOVA and seven nucleolar, eight speckled, three homogeneous by Euroimmun. cThree samples with nuclear dots pattern by INOVA were two speckled, one nucleolar by MBL and three nuclear dots by Euroimmun. dOne sample with nuclear dots pattern by MBL was speckled by both INOVA and Euroimmun. eOne sample with Golgi-like pattern by INOVA was speckled by MBL and Golgi-like pattern by Euroimmun.
ANA pattern of 2 samples using three commercial kits.
(A–C) originating from one patient using INOVA, MBL and Euroimmun, respectively; (D–F) originated from another patient using INOVA, MBL and Euroimmun, respectively. All the samples were diluted at 1:80.
Discussion
IIF using HEp-2 cells is recommended as the gold standard reference method to detect the presence of ANAs [18], [19], [20]. The scope of our research, which processed and analyzed more than 10,000 fluorescence tests in a routine laboratory setting, was (1) to compare the positive and negative results between the automated and conventional visual approaches, (2) to build the cutoff LIU for different patterns as well as validate and compare the custom cutoff with an automated autotiter system in two different clinical laboratories according to the end-point titer, (3) to evaluate the performance of NOVA View in patients with a clear diagnosis of CTD, and (4) to assess the automatic ANA pattern recognition system and compare the pattern from HEp-2 cell slides produced by different manufacturers.
In this study, high-quality digital images led to excellent consistency between the automated IIF analysis and the manual microscopy results, as positive and negative agreement levels between the two methods were 98.1% and 95.9%, respectively (κ=0.973). Similar to the results in our study, many studies have examined agreement between automated and conventional ANA IIF analyses. Regardless of the platform on which they are based, such as EUROPattern, AKLIDES software or NOVA View, automated interpretation systems all exhibit reliable discrimination between positive and negative results [11], [13], [16], [21], [22], [23].
The ANA titer is an important concern in a routine clinical laboratory. There is evidence to indicate that a high ANA titer is more related to autoimmune diseases [2], [24], [25]. However, with respect to clinical efficacy, the end-point titer of ANA is not commonly manipulated in clinical laboratories in routine work. The automated system, which has the ability to provide the recommended autotiter according to a quantitative reading of the immunofluorescence intensity, may provide an objective value-added report.
Our results showed a significant association between the LIU value and the end-point titer at a dilution of 1:80, which was consistent with the report of S. Schouwers et al. [17], [25]. However, we further proved that the custom cutoff was superior to the preset cutoff in NOVA View (local vs. preset cutoff: 57.4% vs. 43.7%, respectively, p<0.01) in our laboratory. However, when the test validation was performed in the other laboratory, the exact prediction was approximately the same (p=0.11), whereas the acceptable accuracy increased with the preset cutoff value (p<0.01). Therefore, it is important to establish a local cutoff to increase the accuracy of the value-added reporting. Nevertheless, the preset cutoff was more credible than the findings originating from other laboratories without a custom cutoff.
In addition, the median ANA titers for the patients with a clear diagnosis of SLE, UCTD, SjS and RA were 1:640, 1:640, 1:640 and 1:320, respectively, which showed that these patients had a relatively increased ANA titer compared with disease control group with the median ANA titers as negative. However, because of the insufficient sample size, the local cutoff for the basic pattern of nuclear dots was not established; this will be investigated further in a future study.
Another vital component of the automated digital immunofluorescence system is pattern recognition. In our study, NOVA View correctly identified 61.9% of the samples, which has the similar performance that reported by Bizzaro et al. [14]. Optimal performance in pattern recognition was found for the centromere type (92.6%), whereas the lowest performance was for the nuclear dot pattern (27.3%). The most common patterns of speckled and homogeneous were identified in 62.7% and 57.4% of cases, respectively. Such situations may occur, in part, because some investigators tend to recognize a dense granular pattern as homogeneous. In addition, the computer-aided system, which recognizes a pattern based on an algorithm, may be affected by cytoplasmic fluorescence with a mixed pattern. Nevertheless, the automated pattern-determining system was not satisfactory. Thus, confirmation of digital immunofluorescence images by expert technicians is suggested to be an essential prerequisite of high-quality ANA reports.
As there existed heterogeneity in the performance of various HEp-2 assay kits, what is the difference ratio between INOVA and some other commercial kits? To our knowledge, most literatures focused on reporting positive or negative rate on single devices [11], [12], [26], [27], [28], even less data attempted to compare patterns between different assay kits [9], [29]. Therefore, we further evaluated the discrepancy caused by the HEp-2 assay kits in our clinical laboratory. In our research, 1611 ANA-positive serum samples were processed using two kits, and 76 samples (4.7%) were found to have different pattern results; in these cases, a third commercial kit for ANA detection was used to further confirm the ANA pattern. The principal problem remains the ability to distinguish between the speckled and homogeneous patterns. Another key issue was with regard to the staining of the nucleolar pattern. For 24 samples that showed different ANA patterns between INOVA and MBL commercial kits, following confirmation by Euroimmun ANA slides, the numbers of nucleolar-positive samples were 6, 18 and 7 for INOVA, MBL and Euroimmun, respectively. Therefore, when encountering paradoxical IIF patterns according to anti-ENA profiles or existing discrepancies with past ANA patterns in other laboratories, different commercial ANA IIF kits may be utilized to confirm the pattern.
Conclusions
The automated antinuclear immunofluorescence antibody analysis system of NOVA View provided not only high-accuracy ANA-positive/negative identification compared to manual microscopy recognition but also reliable, digital immunofluorescence intensity that enabled value-added reports. However, with respect to automated ANA pattern recognition, expert technicians were obliged to validate the positive results by NOVA View outside the dark room. Suspect results could be confirmed by another commercial ANA IIF kit to achieve more reliable results.
Acknowledgments
The authors thank Lu Yu, PhD, and Zhou Zhenyuan, PhD, for the assistance in the manuscript draft.
Author contributions: ZB and SX conducted the ANA-IIF automated analysis with quality control. ZB and LE manually and automatically analyzed the ANA slides. ZH performed the ANuA and dsDNA testing via ELISA. LJ compared the ANA-positive sera using different commercial ANA test kits. LM and ZJ participated in the study design and helped revise the draft. All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: This work was supported by the National Natural Science Foundation of China (grants 81601853, 81671975 and 81371875), Shanghai medical and health development foundation grant 2016-05.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
References
1. Solomon DH, Kavanaugh AJ, Schur PH. Evidence-based guidelines for the use of immunologic tests: antinuclear antibody testing. Arthritis Care Res 2002;47:434–44.10.1002/art.10561Search in Google Scholar PubMed
2. Bizzaro N, Wiik A. Appropriateness in anti-nuclear antibody testing: from clinical request to strategic laboratory practice. Clin Exp Rheumatol 2003;22:349–55.Search in Google Scholar
3. Bentow C, Lakos G, Rosenblum R, Bryant C, Seaman A, Mahler M. Clinical performance evaluation of a novel, automated chemiluminescent immunoassay, QUANTA Flash CTD Screen Plus. Immunol Res 2015;61:110–6.10.1007/s12026-014-8601-5Search in Google Scholar PubMed
4. Dillaerts D, De Baere H, Bossuyt X. Clinical autoantibody detection by microarray. Clin Chem Lab Med 2017;55:578–85.10.1515/cclm-2016-0533Search in Google Scholar PubMed
5. Tozzoli R, Bizzaro N, Tonutti E, Villalta D, Bassetti D, Manoni F, et al. Guidelines for the laboratory use of autoantibody tests in the diagnosis and monitoring of autoimmune rheumatic diseases. Am J Clin Pathol 2002;117:316–24.10.1309/Y5VF-C3DM-L8XV-U053Search in Google Scholar PubMed
6. Hiemann R, Büttner T, Krieger T, Roggenbuck D, Sack U, Conrad K. Challenges of automated screening and differentiation of non-organ specific autoantibodies on HEp-2 cells. Autoimmun Rev 2009;9:17–22.10.1016/j.autrev.2009.02.033Search in Google Scholar PubMed
7. Fenger M, Wiik A, Høier-Madsen M, Lykkegaard JJ, Rozenfeld T, Hansen MS, et al. Detection of antinuclear antibodies by solid-phase immunoassays and immunofluorescence analysis. Clin Chem 2004;50:2141–7.10.1373/clinchem.2004.038422Search in Google Scholar PubMed
8. Pham B, Albarede S, Guyard A, Burg E, Maisonneuve P. Impact of external quality assessment on antinuclear antibody detection performance. Lupus 2005;14:113–9.10.1191/0961203305lu2069oaSearch in Google Scholar PubMed
9. Copple SS, Giles SR, Jaskowski TD, Gardiner AE, Wilson AM, Hill HR. Screening for IgG antinuclear autoantibodies by HEp-2 indirect fluorescent antibody assays and the need for standardization. Am J Clin Pathol 2012;137:825–30.10.1309/AJCPICNFG7UCES1SSearch in Google Scholar PubMed
10. Sparic R, Buzadzic S, Argirovic R, Opalic J. Adnexal torsion during puerperium as a complication of delayed surgical treatment of ovarian tumor in pregnancy. Acta Chir lugosl 2012; 59:119–21.10.2298/ACI1201119SSearch in Google Scholar
11. Copple SS, Jaskowski TD, Giles R, Hill HR. Interpretation of ANA indirect immunofluorescence test outside the darkroom using NOVA view compared to manual microscopy. J Immunol Res 2014;2014:7.10.1155/2014/149316Search in Google Scholar PubMed PubMed Central
12. Krause C, Ens K, Fechner K, Voigt J, Fraune J, Rohwader E, et al. EUROPattern Suite technology for computer-aided immunofluorescence microscopy in autoantibody diagnostics. Lupus 2015;24:516–29.10.1177/0961203314559635Search in Google Scholar PubMed
13. Bonroy C, Verfaillie C, Smith V, Persijn L, De Witte E, De Keyser F, et al. Automated indirect immunofluorescence antinuclear antibody analysis is a standardized alternative for visual microscope interpretation. Clin Chem Lab Med 2013;51:1771–9.10.1515/cclm-2013-0016Search in Google Scholar PubMed
14. Bizzaro N, Antico A, Platzgummer S, Tonutti E, Bassetti D, Pesente F, et al. Automated antinuclear immunofluorescence antibody screening: a comparative study of six computer-aided diagnostic systems. Autoimmun Rev 2014;13:292–8.10.1016/j.autrev.2013.10.015Search in Google Scholar PubMed
15. Daves M, Blecken J, Matthias T, Frey A, Perkmann V, Dall A, et al. New automated indirect immunofluorescent antinuclear antibody testing compares well with established manual immunofluorescent screening and titration for antinuclear antibody on HEp-2 cells. Immunol Res 2016. doi:10.1007/s12026-016-8874-y.10.1007/s12026-016-8874-ySearch in Google Scholar PubMed
16. Voigt J, Krause C, Rohwäder E, Saschenbrecker S, Hahn M, Danckwardt M, et al. Automated indirect immunofluorescence evaluation of antinuclear autoantibodies on HEp-2 cells. Clin Dev Immunol 2012;2012:651058.10.1155/2012/651058Search in Google Scholar PubMed PubMed Central
17. Roggenbuck D, Hiemann R, Schierack P, Reinhold D, Conrad K. Digital immunofluorescence enables automated detection of antinuclear antibody endpoint titers avoiding serial dilution. Clin Chem Lab Med 2014; 52:e9–e11.10.1515/cclm-2013-0543Search in Google Scholar PubMed
18. Meroni PL, Schur PH. ANA screening: an old test with new recommendations. Ann Rheum Dis 2010;68:1420–2.10.1136/ard.2009.127100Search in Google Scholar PubMed
19. González DA, de León AC, Pérez MdCR, Díaz BB, Hernández AG, García DG, et al. Efficiency of different strategies to detect autoantibodies to extractable nuclear antigens. J Immunol Methods 2010;360:89–95.10.1016/j.jim.2010.06.013Search in Google Scholar PubMed
20. Wiik A. Anti-nuclear autoantibodies: clinical utility for diagnosis, prognosis, monitoring, and planning of treatment strategy in systemic immunoinflammatory diseases. Scand J Rheumatol 2005; 34:260–8.10.1080/03009740500202664Search in Google Scholar PubMed
21. Mulliez S, Maenhout TM, Bonroy C. Impact of the routine implementation of automated indirect immunofluorescence antinuclear antibody analysis: 1 year of experience. Clin Chem Lab Med 2016;54:e183–e6.10.1515/cclm-2015-0900Search in Google Scholar PubMed
22. Kivity S, Gilburd B, Agmon-Levin N, Carrasco MG, Tzafrir Y, Sofer Y, et al. A novel automated indirect immunofluorescence autoantibody evaluation. Clin Rheumatol 2012; 31:503–9.10.1007/s10067-011-1884-1Search in Google Scholar PubMed
23. Tozzoli R, Antico A, Porcelli B, Bassetti D. Automation in indirect immunofluorescence testing: a new step in the evolution of the autoimmunology laboratory. Auto Immun Highlights 2012;3:59–65.10.1007/s13317-012-0035-2Search in Google Scholar PubMed PubMed Central
24. Selmi C, Ceribelli A, Generali E, Scirè CA, Alborghetti F, Colloredo G, et al. Serum antinuclear and extractable nuclear antigen antibody prevalence and associated morbidity and mortality in the general population over 15 years. Autoimmun Rev 2016;15:162–6.10.1016/j.autrev.2015.10.007Search in Google Scholar PubMed
25. Schouwers S, Bonnet M, Verschueren P, Westhovens R, Blockmans D, Mariën G, et al. Value-added reporting of antinuclear antibody testing by automated indirect immunofluorescence analysis. Clin Chem Lab Med 2014;52:547–51.10.1515/cclm-2013-0610Search in Google Scholar PubMed
26. Gorgi Y, Dhaouadi T, Sfar I, Haouami Y, Abdallah TB, Raso G, et al. Comparative study of human and automated screening for antinuclear antibodies by immunofluorescence on HEp-2 cells. Int J Stat Med Res 2015;4:270–6.10.6000/1929-6029.2015.04.03.4Search in Google Scholar
27. Shovman O, Agmon-Levin N, Gilburd B, Martins T, Petzold A, Matthias T, et al. A fully automated IIF system for the detection of antinuclear antibodies and antineutrophil cytoplasmic antibodies. Immunol Res 2015;61:135–40.10.1007/s12026-014-8588-ySearch in Google Scholar PubMed
28. Alsuwaidi M, Dollinger M, Fleck M, Ehrenstein B. The reliability of a novel automated system for ANA immunofluorescence analysis in daily clinical practice. Int J Rheumatol 2016;2016:5.10.1155/2016/6019268Search in Google Scholar PubMed PubMed Central
29. Infantino M, Meacci F, Grossi V, Manfredi M, Benucci M, Merone M, et al. The burden of the variability introduced by the HEp-2 assay kit and the CAD system in ANA indirect immunofluorescence test. Immunol Res 2016. doi:10.1007/s12026-016-8845-3.10.1007/s12026-016-8845-3Search in Google Scholar PubMed
Supplemental Material:
The online version of this article (https://dx.doi.org/10.1515/cclm-2017-0050) offers supplementary material, available to authorized users.
©2017 Walter de Gruyter GmbH, Berlin/Boston
