Abstract
Immunoblot (IB) methods are widely used to detect myositis-specific autoantibodies (MSAs); however, false-positive results are common. In this study, we aimed to determine whether associating the anti-nuclear antibody (ANA) IIF pattern may help to improve the specificity of MSA detection by IB in patients with idiopathic inflammatory myositis (IIM).
Serum samples from 104 patients presenting with muscle weakness/myalgia and positive to at least one MSA by IB (MYOS12 Diver and MIOS7 Diver, D-tek) were tested for ANAs on HEp-2000 cells (Immuno Concepts). The chi-square test was used to analyse the concordance of the MSA result and its corresponding pattern by ANA testing between patients with and without IIM.
Eighty-three of the 104 patients had a diagnosis of definite IIM, while in 21 cases, patients were affected by other autoimmune diseases or various non-systemic diseases. Forty nine of 83 (59%) patients in the IIM group and 4/21 (19%) in the non-IIM group showed a concordance between ANA pattern and MSAs by IB (P < 0.001). MSA monopositivity was significantly associated with IIM (91.6%) compared with 61.9% in the non-IIM group (P = 0.0005).
Considering both the MSA result and its corresponding pattern by ANA testing may help to improve the specificity of MSA detection by IB and to confirm the diagnosis of MSA-associated IIM. The monopositivity of MSAs is an important additional tool to validate IB results.
Concordance with the ANA-immunofluorescence pattern may improve the clinical specificity of immunoblot tests.
An immunoblot result inconsistent with the ANA pattern may be a false positive.
A multi-positive myositis-specific autoantibody finding by immunoblot should be carefully interpreted.
Introduction
Idiopathic inflammatory myopathies (IIM) are a wide and heterogeneous group of connective tissue diseases whose classification has frequently evolved due to difficulty in establishing homogeneous groups. In recent years, a better characterization of myositis-related antibodies has improved definition of IIM-related groups [1]. Hence, while muscle biopsy is still essential for the diagnosis of most IIM cases, the use of myositis-related antibodies has become increasingly prominent [2, 3] and the combined use of clinical, histological and serological criteria is now recommended as standard practice for IIM diagnosis and classification [4].
Myositis-related antibodies are classified into two groups called myositis-specific antibodies (MSAs) and myositis-associated antibodies (MAAs). Whereas MSAs are more closely associated to IIM, MAAs can also be detected in other autoimmune or infectious conditions as well as in healthy subjects. Myositis-related antibodies are found in ∼60% of IIM patients, with prevalence rates varying depending on the antibody group: MAAs can be found in up to 80% of patients whereas MSAs are usually present in 30–50% of IIM [5].
Currently, several methods are available to test for MSAs, with variable sensitivity, specificity, costs, complexity and feasibility in clinical and research settings.
Anti-nuclear antibodies (ANAs) on HEp-2 cells by IIF lack sensitivity for MSAs [6] and even when they are positive, they do not allow clinicians to address any specific disease. However, they may suggest the presence of MSAs, as our group recently showed [7] by demonstrating the efficiency of a reflex algorithm for IIF cytoplasmic patterns followed by myositis immunoblot (IB) profiling. Thus, cytoplasmic staining should be looked for and reported in such patients, which would potentially lead to earlier diagnosis.
Immunoprecipitation (IP) is still the reference method as it evaluates the binding of the autoantibodies to the RNA and protein complexes in their native conformation, yielding the best sensitivity and specificity [8]. However, since this method is only available in a limited number of research laboratories, in practice the presence of these antibodies in serum is commonly assessed by IB assays, such as line immunoassay (LIA) and dot-blot (DB), which allow simultaneous detection of most of the MSAs and MAAs described so far [9].
LIA and DB sensitivity for IIM diagnosis varies [10, 11] depending on the nature of the antigen involved, and because some of them are susceptible to protein denaturation and/or degradation during purification and coating phases. Indeed, the clinical accuracy of each antibody can range from very good for anti-Jo1, anti-Mi-2α/β, anti-TIF1γ, anti-MDA5, anti-SAE1 and anti-PM/Scl100, to less satisfactory for anti-SRP, anti-NXP2, anti-PM/Scl75 and anti-Ku antibodies. Despite IB methods representing significant progress in MSA detection, these assays are not standardized and assay harmonization among manufacturers is poor [12, 13]. Vulsteke et al. observed differences in specificity among manufacturers and among individual antibodies. For example, 2.9% and 2.4% of controls tested positive for anti-Jo1 by Euroimmun LIA and Trinity LIA, respectively, compared with 0.4% by Alphadia DB [14]. Overall, Euroimmun LIA and Trinity LIA showed more reactivity in controls than Alphadia DB, except for anti-SAE for which Euroimmun LIA showed a lower rate of false-positive results in controls. Besides controls, differences in reactivity among manufacturers were also observed in IIM patients, with the most pronounced discrepancies for anti-TIF-1γ (2.1% with Alphadia DB vs 12.4% with Euroimmun LIA and 11% with Trinity LIA). It should be noted that even for well-established markers such as anti-Jo1 antibodies, differences between manufacturers were observed in patients with IIM.
These discrepancies were also observed when comparing IB assays to the immunoprecipitation method [6]. Since IB assays may also suffer from low specificity, results with IB assays require careful evaluation and validation of their performance to decipher the extent to which these assays can contribute to clinical practice.
Indeed, a false-positive result for MSA may have a considerable clinical impact, i.e. misdiagnoses, misguided therapies, a cascade of costly and unnecessary clinical referrals and patient anxiety. Furthermore, a suspected false-positive for a cancer-associated MSA (e.g. anti-TIF1γ antibodies) may have ethical implications [15].
To this end, Picard et al. evaluated the hypothetical importance of analysing anti-SRP antibodies both by IIF on HEp-2 cells and dot immunoassay to ensure clinically specific and relevant identification [16]. Indeed, according to the positivity of the dot immunoassay alone, 40% of the patients would have received a false diagnosis of IIM. Strikingly, by adding the characteristic finely granular staining of the cytoplasm on HEp-2 cells, they found a high association (90%) with clinical myositis.
Similar to the Picard study, which was, however, restricted to only anti-SRP antibodies, our study, including a wider number of MSAs, aimed to establish whether considering both the MSA result and its corresponding pattern through ANA testing may help to improve the specificity of MSA detection by IB.
Methods
Patients and sera
Serum samples from 104 patients (F: M ratio 3.9: 1; mean age 59 years) presenting with muscle weakness/myalgia and positive to at least one MSA in the IIM IB profiles (MYOS12DIV-24 and SYN10DIV-24, BlueDiver Dot kits, D-tek, Mons, Belgium) were tested for ANA by IIF on HEp-2000 cells (Immuno Concepts, Buffalo, NY, USA) at a starting dilution of 1: 80. Interpretation of the ANA-IIF test was done independently by two expert physicians and ANA patterns were classified according to the International Consensus on ANA Patterns (ICAP) nomenclature [17] that states:
The nuclear fine speckled pattern (AC-4) appears as fine tiny speckles across all nucleoplasm, with the nucleoli stained or not stained, and the mitotic cells having the chromatin mass not stained. Antibodies displaying this pattern include anti-TIF1γ and anti-Mi2. The cytoplasmic dense fine speckled pattern (AC-19) appears as a cloudy, almost homogeneous, speckled pattern throughout the cytoplasm. Antibodies associated with this pattern include PL-7 and PL-12. In case of the cytoplasmic fine speckled/speckled pattern (AC-20), small speckles are scattered in the cytoplasm mostly with homogeneous or dense fine speckled background. Possible antibodies are against aminoacyl-tRNA-synthetases, mainly Jo-1 (histidyl-tRNA synthetase).
The IIM IB profiles MYOS12DIV-24 and SYN10DIV-24 include the following antigens: Jo1, PL-7, PL-12, EJ, SRP, Mi-2, MDA-5, TIF1γ, HMGCR, Ro52, SAE1/2, NXP-2, KS, HA, ZO and OJ. In this study, EIF-3 was also tested thanks to a Research Use Only kit (D-tek). Eighty-three of the 104 patients had a diagnosis of definite IIM, while in 21 cases, patients were affected by other autoimmune diseases or various non-systemic diseases. The diagnosis of IIM was based on the Bohan and Peter criteria: symmetric muscle weakness, increased serum muscle enzymes, myopathic changes on electromyography, histological findings on muscle biopsy, and typical dermatological manifestations. Sera were collected in nine centres belonging to the Study Group on Autoimmune Diseases of the Italian Society of Clinical Pathology and Laboratory Medicine. All patients gave their informed consent to this retrospective study according to the Declaration of Helsinki and Italian legislation (Authorization of the Privacy Guarantor No. 9, 12 December 2013).
Statistical analysis
The chi-square test was used to analyse the concordance of the MSA result and its corresponding pattern by ANA testing between patients with and without IIM. A P-value <0.05 was considered significant.
Results
Sixty seven of 83 (80.7%) sera in the IIM group and 16/21 (76.2%) in the non-IIM group displayed a positive ANA pattern at the IIF test on HEp-2 cells (homogeneous, speckled, nucleolar, cytoplasmic, mixed). The speckled pattern (AC 2, 4, 5) was more prevalent in IIM patients, while the nucleolar pattern (AC 8, 9, 10) was more prevalent in non-IIM; of note, cytoplasmic patterns (AC 18, 19, 20) were equally represented. Thirty-seven out of the 83 IIM patients (44.6%) and three out of the 21 (14.3%) non-IIM patients that were positive to at least one MSA by IB assay displayed their corresponding pattern on HEp-2 cells (P = 0.0002) (Table 1). Furthermore, when analysing the correlation between each single MSA detected by IB and its expected IIF pattern on HEp-2 cells, we showed a statistically significant difference between patients with and without IIM only for anti-Jo1, anti-Mi2 and anti-SRP antibodies (Table 2). The frequency of one positive MSA marker was 91.6% in the IIM group and 61.9% in the non-IIM group, while the positivity of three or more markers was 0% in the IIM group and 14.3% in the non-IIM group (P = 0.0005) (Fig. 1).
Prevalence of overall MSA results with and without its corresponding pattern by ANA testing
| IB/IIF POSITIVE (n = 83) | IIM (n = 67) | non IIM (n = 16) |
|---|---|---|
| IIF positive with corresponding pattern | 59.0% (37/67) | 19.0% (3/16) |
| IIF positive without a corresponding pattern | 41.0% (21/67) | 81% (13/16) |
| IB/IIF POSITIVE (n = 83) | IIM (n = 67) | non IIM (n = 16) |
|---|---|---|
| IIF positive with corresponding pattern | 59.0% (37/67) | 19.0% (3/16) |
| IIF positive without a corresponding pattern | 41.0% (21/67) | 81% (13/16) |
ANA: anti-nuclear antibody; IB: immunoblot; IIF: indirect immunofluorescence; IIM: idiopathic inflammatory myositis; MSA: myositis-specific autoantibody.
Prevalence of overall MSA results with and without its corresponding pattern by ANA testing
| IB/IIF POSITIVE (n = 83) | IIM (n = 67) | non IIM (n = 16) |
|---|---|---|
| IIF positive with corresponding pattern | 59.0% (37/67) | 19.0% (3/16) |
| IIF positive without a corresponding pattern | 41.0% (21/67) | 81% (13/16) |
| IB/IIF POSITIVE (n = 83) | IIM (n = 67) | non IIM (n = 16) |
|---|---|---|
| IIF positive with corresponding pattern | 59.0% (37/67) | 19.0% (3/16) |
| IIF positive without a corresponding pattern | 41.0% (21/67) | 81% (13/16) |
ANA: anti-nuclear antibody; IB: immunoblot; IIF: indirect immunofluorescence; IIM: idiopathic inflammatory myositis; MSA: myositis-specific autoantibody.
Prevalence of each MSA with and without its corresponding ANA pattern in immunoblot-positive IIM patients
| Anti-Jo1 (n = 37), 91.9% | Anti-ARSa (n = 15), 66.7% | Anti-Mi2 (n = 12), 83.3% | Anti-SRP (n = 14), 57.1% | Anti-TIF1γ (n = 9), 44.4% | Anti-SAE (n = 19), 68.4% | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Immunoblot positive | IIM (n = 34) | P-value | IIM (n = 10) | P-value | IIM (n = 10) | P-value | IIM (n = 8) | P-value | IIM (n = 4) | P-value | IIM (n = 13) | P-value |
| IB positive with IIF corresponding pattern | 59.5% | 0.02 | 33.3% | n.s. | 75% | 0.003 | 35.7% | 0.0002 | 22.2% | n.s. | 36.8% | n.s. |
| IB positive without IIF corresponding pattern | 32.4% | 33.3% | 8.3% | 21.4% | 22.2% | 31.6% | ||||||
| Anti-Jo1 (n = 37), 91.9% | Anti-ARSa (n = 15), 66.7% | Anti-Mi2 (n = 12), 83.3% | Anti-SRP (n = 14), 57.1% | Anti-TIF1γ (n = 9), 44.4% | Anti-SAE (n = 19), 68.4% | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Immunoblot positive | IIM (n = 34) | P-value | IIM (n = 10) | P-value | IIM (n = 10) | P-value | IIM (n = 8) | P-value | IIM (n = 4) | P-value | IIM (n = 13) | P-value |
| IB positive with IIF corresponding pattern | 59.5% | 0.02 | 33.3% | n.s. | 75% | 0.003 | 35.7% | 0.0002 | 22.2% | n.s. | 36.8% | n.s. |
| IB positive without IIF corresponding pattern | 32.4% | 33.3% | 8.3% | 21.4% | 22.2% | 31.6% | ||||||
aAnti-synthetase antibodies include PL-7, PL-12, EJ, OJ, KS, Zo and HA. ANA: anti-nuclear antibody; ARS: anti-synthetase antibodies; IB: immunoblot; IIM: idiopathic inflammatory myositis; MSA: myositis-specific autoantibody; n.s.: not significant.
Prevalence of each MSA with and without its corresponding ANA pattern in immunoblot-positive IIM patients
| Anti-Jo1 (n = 37), 91.9% | Anti-ARSa (n = 15), 66.7% | Anti-Mi2 (n = 12), 83.3% | Anti-SRP (n = 14), 57.1% | Anti-TIF1γ (n = 9), 44.4% | Anti-SAE (n = 19), 68.4% | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Immunoblot positive | IIM (n = 34) | P-value | IIM (n = 10) | P-value | IIM (n = 10) | P-value | IIM (n = 8) | P-value | IIM (n = 4) | P-value | IIM (n = 13) | P-value |
| IB positive with IIF corresponding pattern | 59.5% | 0.02 | 33.3% | n.s. | 75% | 0.003 | 35.7% | 0.0002 | 22.2% | n.s. | 36.8% | n.s. |
| IB positive without IIF corresponding pattern | 32.4% | 33.3% | 8.3% | 21.4% | 22.2% | 31.6% | ||||||
| Anti-Jo1 (n = 37), 91.9% | Anti-ARSa (n = 15), 66.7% | Anti-Mi2 (n = 12), 83.3% | Anti-SRP (n = 14), 57.1% | Anti-TIF1γ (n = 9), 44.4% | Anti-SAE (n = 19), 68.4% | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Immunoblot positive | IIM (n = 34) | P-value | IIM (n = 10) | P-value | IIM (n = 10) | P-value | IIM (n = 8) | P-value | IIM (n = 4) | P-value | IIM (n = 13) | P-value |
| IB positive with IIF corresponding pattern | 59.5% | 0.02 | 33.3% | n.s. | 75% | 0.003 | 35.7% | 0.0002 | 22.2% | n.s. | 36.8% | n.s. |
| IB positive without IIF corresponding pattern | 32.4% | 33.3% | 8.3% | 21.4% | 22.2% | 31.6% | ||||||
aAnti-synthetase antibodies include PL-7, PL-12, EJ, OJ, KS, Zo and HA. ANA: anti-nuclear antibody; ARS: anti-synthetase antibodies; IB: immunoblot; IIM: idiopathic inflammatory myositis; MSA: myositis-specific autoantibody; n.s.: not significant.
Percentage of positive MSA markers in the IIM and in the non-IIM groups
IIM: idiopathic inflammatory myositis; MSA: myositis-specific autoantibody.
Discussion
The clinical heterogeneity of the IIM forms strongly needs reliable markers that may help to define the clinical phenotype [18]. In real life, on the one hand most clinicians still use Bohan and Peter diagnostic criteria that, however, lack specificity and are dated, and on the other hand they still rely on the ANA IIF test even if it is not very sensitive in detecting MSA. With cytoplasmic pattern, lack of sensitivity is worsened by the fact that there is no consensus as to whether cytoplasmic pattern is to be considered ANA positive or negative [17], and by the challenging interpretation of ANA cytoplasmic staining, which is often present at low titre [19].
Uncertainty in interpreting the ANA test has meant that in most immunology laboratories, the more sensitive IB test is commonly used today to search for MSA. However, IB does not always provide completely accurate results in terms of specificity, and false positives are common [8, 14].
Given the growing and widespread use of the IB method in the IIM diagnostic field, it is clear that improving its specificity is fundamental. This prompted us to design the present study, extending the cited French study on anti-SRP detection to a wider number of MSAs, combining IB results with the ANA IIF pattern.
Overall, we confirmed the importance of concordance between the ANA IIF pattern and the IB result to improve the clinical specificity of IB. Since in the non-IIM group, 81% of patients showed MSA positivity incompatible with the ANA IIF pattern, ANA testing may be a key factor when interpreting the detection of myositis-related antibodies by IB. In stratifying patients for each specific MSA cohort, we showed statistical significance in the association between an ANA pattern and IB for Jo1, Mi2 and SRP specificities; the low number of patients in each single antibody cohort limited the study’s statistical power for other MSAs.
Interestingly, 92% of patients in the IIM group had only one positive MSA and 8% had two MSAs. Conversely, MSA monopositivity was observed only in 62% of the non-IIM patients, whereas 14% were positive to three or more MSAs. Since MSAs are reported to be usually mutually exclusive, finding three or more positive MSAs by IB should be interpreted with caution [20]. A limitation of this study is that we did not perform immunoprecipitation because nowadays this technique is restricted to a limited number of research laboratories. However, the study maintains its practical value because it reproduced the daily diagnostic path used in most, if not all, clinical laboratories.
In conclusion, as indicated in the algorithm we propose (Fig. 2), being familiar with the corresponding ANA IIF pattern may represent an important element when validating an MSA result via IB. This means that when a specific autoantibody test result is inconsistent with the ANA pattern, a false-positive test result may be considered (e.g. positive anti-Mi2 with cytoplasmic staining can be considered inconsistent [21]). We can speculate that education and increased awareness among rheumatology and pulmonary specialists regarding ANA patterns will lead to improved care for patients with IIM [22, 23]. Finally, providing a report with customized comments may make clinicians aware of the advantages and limits of the currently available tests in the diagnostic workup of IIM.
ANA IIF pattern association (including the AC-ICAP nomenclature) with distinct MSA specificities
MSA: myositis-specific autoantibody.
Acknowledgements
This paper is dedicated to the memory of our colleague and friend Elio Tonutti, to honor his lifetime commitment to and enthusiasm for science, his vivacious wit and his brilliant intuition. The authors thank Alphadia and Immuno Concepts for kindly providing the reagents for autoantibody detection free of costs, and Dr Nicolas Bodart and Dr Benoit Autem of D-tek for the technical assistance and data analysis. Alphadia, Immuno Concepts and D-tek are distributed in Italy by Alifax S.r.l.
Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript.
Disclosure statement: The authors have declared no conflicts of interest.
Comments