Abstract
Objective. To compare anti-PM/Scl autoantibody-associated interstitial lung disease (ILD) with anti-aminoacyl-tRNA synthetases (anti-ARS) autoantibody-associated ILD.
Methods. We retrospectively studied 21 patients with ILD from a department of respiratory medicine, including 9 with anti-PM/Scl autoantibodies (6 women, median age 55 yrs, followup 5.5 yrs) and 12 with anti-ARS autoantibodies (6 women, median age 59 yrs, followup 2.3 yrs).
Results. Pulmonary manifestations in patients with anti-PM/Scl autoantibody-associated ILD usually followed the extrapulmonary manifestations of the connective tissue disease (CTD) (7/9 cases). The predominant imaging features on initial high resolution computed tomography were ground-glass attenuation and reticular opacities, and mainly suggested nonspecific interstitial pneumonia in both groups. CTD was classified as dermatomyositis (DM; 2), undifferentiated CTD (2), cutaneous limited systemic sclerosis (2), rheumatoid arthritis (RA; 1), and overlap syndrome (1) in the anti-PM/Scl group; and polymyositis (4), undifferentiated CTD (5), DM (1), amyopathic DM (1), and RA (1) in the anti-ARS group. Frequencies of arthralgia, Raynaud phenomenon, cutaneous rash, and mechanic’s hands were comparable in both groups. Myalgia or muscle weakness was present in 0/9 PM/Scl and 5/12 ARS patients (p < 0.05). More than 1 autoantibody was present in 11 patients. ILD worsened despite treatment in 4 patients with anti-PM/Scl autoantibodies and 2 with anti-ARS autoantibodies, and included 1 death.
Conclusion. Anti-PM/Scl and anti-ARS antibodies are associated with similar clinical manifestations, with the exception only of more overt myositis in the latter, therefore challenging the clinical specificity of the antisynthetase syndrome.
- INTERSTITIAL LUNG DISEASE
- MYOSITIS
- SYSTEMIC SCLERODERMA
- ANTI-AMINOACYL-tRNA SYNTHETASE ANTIBODIES
- ANTI-PM/Scl ANTIBODIES
The anti-aminoacyl-tRNA synthetase antibodies (anti-ARS) include anti-Jo1, the most common, and anti-PL7, anti-PL12, anti-OJ, anti-EJ, anti-KS, anti-YRS, and anti-Zo antibodies. Several studies have shown an increased risk of interstitial lung disease (ILD) in patients with myositis and anti-ARS antibodies, an association often referred to as the so-called antisynthetase syndrome, encompassing arthritis, mechanic’s hand, fever, and Raynaud phenomenon in addition to myositis and ILD1–3.
Anti-PM/Scl antibodies have been described in a subgroup of patients diagnosed with myositis and/or systemic sclerosis (SSc)4,5. Interestingly, the clinical features associated with anti-PM/Scl antibodies overlap with those of the antisynthetase syndrome1,6. Further, anti-PM/Scl and anti-ARS antibodies are both associated with a conserved, Caucasian haplotype known as the HLA 8.1 ancestral haplotype, containing HLA-DRB1 *03/DQA1 *05/DQB1 *02 haplotypes3,4,7–9. Despite the high frequency of ILD in patients with anti-PM/Scl antibodies, little is known about its imaging pattern, chronology relative to extrapulmonary manifestations, and outcome. Anti-ARS-related ILD has been studied more extensively and has prognostic implications10. The aims of our study were to describe the clinical, imaging, and functional characteristics and outcome of patients with ILD associated with anti-PM/Scl antibodies, and to compare them to ILD associated with anti-ARS antibodies.
MATERIALS AND METHODS
Patient selection
We retrospectively selected patients evaluated between January 1995 and September 2008 in a department of respiratory medicine specialized in rare pulmonary disease, especially ILD, on the basis of the detection of anti-PM/Scl or anti-ARS autoantibodies, and the presence of ILD as defined by diffuse infiltrative opacities on high resolution computed tomography (HRCT) of the chest. Anti-PM/Scl and anti-ARS antibodies were systematically assessed in all patients routinely evaluated for ILD, and results were prospectively registered in a dedicated database in the department of immunology that performs autoimmunity analysis for the whole institution. One author (JCL) reviewed the medical records, pulmonary function tests, and laboratory tests of all patients at diagnosis, at 1-year followup, and at last followup.
Diagnostic criteria for CTD
Criteria for the diagnosis of possible, probable, or definite polymyositis (PM) and dermatomyositis (DM) were those described by Bohan and Peter11. Criteria for other CTD included the 1987 revised American College of Rheumatology criteria for rheumatoid arthritis (RA)12, the preliminary criteria for SSc13, the 1997 revised criteria for systemic lupus erythematosus14, and the European/American consensus criteria for Sjögren’s syndrome15. Amyopathic DM was diagnosed when a patient had heliotrope rash and/or Gottron sign, and no muscle weakness or abnormalities (normal muscle enzyme levels, electromyography, muscle biopsy, and muscle magnetic resonance imaging)16. Early SSc was diagnosed according to the 2001 criteria proposed by Leroy and collaborators17. Patients with CTD-associated antibodies and symptoms suggestive of a CTD but not fulfilling the above criteria were classified as having undifferentiated CTD (UCTD).
Laboratory test results
Biochemical assessment was performed as part of routine evaluation of the diagnosis of ILD and included C-reactive protein (CRP; normal < 5 mg × l−1), creatine phosphokinase (normal < 175 U × l−1), aldolase (normal < 8.5 U × l−1), lactic dehydrogenase (LDH; normal < 248 U × l−1), and gammaglobulins (normal < 12.5 g × l−1). All patients with ILD underwent systematic immunologic investigation including antinuclear antibodies (ANA), anti-PM/Scl, anti-Mi2, autoantibodies against the major anti-extractable nuclear antigens [SSA/Ro, SSB/La, Sm, U1-ribonucleoprotein, topoisomerase-1 (Scl70), Jo1], anti-double stranded DNA (anti-dsDNA) antibodies, anti-cyclic citrullinated peptide antibodies, rheumatoid factor, and antineutrophil cytoplasmic antibodies (ANCA). Anti-ARS antibodies other than anti-Jo1 (anti-PL7, PL12, OJ, and EJ) were assessed in case of positive cytoplasmic staining of HEp-2 cells and/or suggestive clinical features. Anti-fibrillarin, anti-RNA polymerase-I, and anti-Th/To antibodies were assessed in case of a nucleolar pattern of ANA. Anti-RNA polymerase III antibodies were detected in case of granular or nucleolar pattern of ANA. Techniques of antibody detection are described in Table 1. All sera were stored at −80°C in a sera collection that was approved by the Institutional Review Board of the Hospices Civils de Lyon.
Characteristics of ILD
Lower limits of diffusing capacity of the lung for carbon monoxide (DLCO), vital capacity, and total lung capacity were defined as < 80% of the predicted values. Hypoxemia was defined by room-air arterial oxygen tension (PaO2) < 10 kPa (75 mm Hg). Normal reference values for bronchoalveolar lavage (BAL) differential cell count were < 15% lymphocytes, < 5% neutrophils, and < 1% eosinophils18. Lung HRCT scans at diagnosis were read by 2 chest physicians expert in ILD (VC, JFC) without knowledge of serological or histopathological information. The categories assessed were ground-glass attenuation, consolidation, parenchymal micronodules, reticulation, honeycombing, and traction bronchiectasis. Imaging patterns were classified as usual interstitial pneumonia (UIP) pattern (combination of reticular opacities, traction bronchiectasis, and honeycombing with basal and subpleural predominance, and minimal ground-glass opacities); nonspecific interstitial pneumonia (NSIP) pattern (combination of ground-glass attenuation and reticulation, with little if any honeycombing); organizing pneumonia (OP) pattern (patchy consolidation); or unclassifiable pattern19. All lung biopsy specimens were evaluated by consensus by 3 of the authors experienced in ILD, and were classified using published guidelines19. All patients underwent routine echocardiographic assessment at the diagnosis of ILD and during followup. Right-side heart catheterization was performed in patients with elevated systolic right ventricular pressure (> 45 mm Hg) or right ventricular dysfunction. Pulmonary arterial hypertension was defined according to guidelines20.
Treatment and outcome
ILD response to therapy was classified into 3 categories (improvement, stable, and failure) according to the American-European International Consensus Statement19. The response of myositis to therapy was evaluated according to the objective improvement of muscle strength and change in muscle enzyme levels.
Data analysis
Two-by-two tables were constructed based on dichotomous data, including clinical, radiological, and biological variables with thresholds provided by manufacturers. Statistical analysis of dichotomous variables was performed using 2-sided Fisher’s exact test. Mann-Whitney test was used for continuous variables. Analyses were performed with SPSS 16.0 for Windows with significance established at p < 0.05.
RESULTS
Patients
Nine patients with anti-PM/Scl and 12 with anti-ARS antibodies and ILD were studied. Anti-ARS antibodies consisted of 7 anti-Jo1, 3 anti-PL7, 1 anti-EJ, and 1 double-positive for anti-EJ and anti-OJ antibodies. One patient diagnosed with undifferentiated CTD and both anti-PM/Scl and anti-Jo1 antibodies developed chronic respiratory failure despite immunosuppressive therapy; manifestations of the CTD were limited to mildly elevated serum muscle enzyme at diagnosis; lung transplant was performed, and the patient remained asymptomatic until last followup; this patient was excluded from the comparison between the anti-ARS and anti-PM/Scl groups. Paraneoplasic CTD was diagnosed in none of the patients. Silicosis and hypersensitivity pneumonitis were ruled out in all cases.
Respiratory involvement and manifestations
The general characteristics and respiratory manifestations of patients with anti-PM/Scl and anti-ARS antibodies were compared (Table 2, Table 3). The prevalence of dyspnea, cough, and crackles did not significantly differ between the 2 groups. Pulmonary function test results were available in all patients.
Eleven of 12 patients with anti-ARS antibodies and 7 of 9 patients with anti-PM/Scl antibodies had reduced vital capacity, total lung capacity, or DLCO (nonsignificant; NS). The mean vital capacity was 71.6% ± 16.2% and 71.5% ± 18.5%, mean total lung capacity 67.9% ± 14.9% and 65.7% ± 14.8%, and mean DLCO 62.6% ± 19.9% and 65.1% ± 16.1% in the anti-PM/Scl and anti-ARS groups, respectively (NS). Two of 9 patients with anti-PM/Scl antibodies and 4 of 12 patients with anti-ARS antibodies were hypoxemic at rest at diagnosis (NS). Differential cell counts of BAL performed in 5 patients with anti-PM/Scl and 11 patients with anti-ARS showed no significant difference between the groups (Figure 1).
Chest computed tomography
The predominant features on the initial HRCT were ground-glass attenuation and reticulation in both groups, with frequent association of some areas of consolidation (Table 4, Figure 1). Traction bronchiectases, micronodules, and honeycombing were uncommon. In the anti-PM/Scl patients, the HRCT pattern was mostly suggestive of NSIP in 8 cases and OP in 1 case. In anti-ARS patients, HRCT was suggestive of NSIP pattern in 9 cases and of UIP and OP patterns in 1 case each, and were unclassifiable in 1 case (Table 3, Figure 2). One current smoker patient with anti-ARS antibodies had typical imaging features of combined pulmonary fibrosis and emphysema. Imaging findings and HRCT patterns were not statistically different between the groups.
Characteristics of the CTD
Extrapulmonary manifestations of CTD among anti-PM/Scl and anti-ARS patients were present in all patients, except one with anti-PM/Scl antibodies (Table 5, Table 6). Remarkably, no patient with anti-PM/Scl antibodies had muscle weakness or pain throughout followup, compared to 5 of 12 patients with anti-ARS antibodies (p = 0.045). All patients with electromyographic evidence of myositis had elevated muscle enzyme levels. Arthralgia principally involved the small joints of the hands (6/6 anti-PM/Scl patients, 7/9 anti-ARS patients; NS), as well as wrists, elbows, shoulders, knees, and feet. No patient had articular deformities. One patient with anti-ARS antibodies had typical features of Sjögren syndrome with sicca syndrome, positive Schirmer test, confirmatory salivary gland biopsy, and splenomegaly. Another patient with anti-ARS antibodies had nephrotic syndrome and microscopic hematuria without renal insufficiency; renal biopsy revealed minimal change of glomerulopathy. The CTD in patients with anti-PM/Scl were DM (n = 2), limited cutaneous SSc (2), RA (1), UCTD (2), and overlap syndrome with coexistence of amyopathic DM, limited cutaneous SSc and Sjögren syndrome (1). Among the anti-ARS group, CTD were classified as PM (n = 4), UCTD (5), amyopathic DM (1), DM (1), and RA (1). Confidence for the diagnosis of DM/PM was definite in 4 cases, probable in 2 cases, and possible in 3 cases.
Chronology of pulmonary and CTD manifestations
Pulmonary symptoms preceded CTD manifestations by 4 months in 1 patient with anti-PM/Scl, and by a median interval of 8 months in 4 patients with anti-ARS (NS). Pulmonary symptoms were concomitant with the CTD manifestations in 1 anti-PM/Scl case and 5 anti-ARS cases (NS). Pulmonary manifestations occurred after CTD features in 5 anti-PM/Scl patients (median interval 84 mo) and in 3 anti-ARS patients (median interval 72 mo) (NS). The information was not available for one anti-PM/Scl patient.
Echocardiographic findings
Two patients had asymptomatic pericardial effusion (1 with anti-Jo1 antibodies, 1 with anti-PL7 antibodies), which regressed with immunosuppressive therapy. Pulmonary hypertension was present at echocardiography in 3 patients with anti-PM/Scl and 1 patient with anti-ARS antibodies, and was confirmed by right-side heart catheterization in all cases. One further patient with anti-PM/Scl antibodies had right ventricular dilatation without pulmonary hypertension at right-side heart catheterization. No patient had left ventricular failure.
Lung pathology
Video-assisted thoracoscopic lung biopsy was performed in 8 cases. The pathological patterns were 1 UIP and 2 NSIP in the anti-PM/Scl group, and 1 UIP, 1 NSIP, and 1 diffuse alveolar damage in the anti-ARS group. In addition, 2 anti-ARS patients had unclassifiable interstitial pneumonia. Transbronchial biopsy specimens showed typical features of OP in 1 patient with anti-ARS antibodies.
Laboratory results
Serum CRP, serum LDH, and hypergammaglobulinemia were not significantly different between the 2 groups (Table 7). ANA were present in all 12 patients with anti-ARS antibodies, with a characteristic cytoplasmic pattern in 10 patients (7 also had a homogeneous, speckled, or nucleolar pattern). Seven of 9 patients with anti-PM/Scl antibodies had ANA, 6 with a homogeneous nucleolar pattern (Table 6, Table 7). Rheumatoid factor was found in 4 patients with anti-ARS antibodies and 3 patients with anti-PM/Scl antibodies. Three patients with anti-ARS antibodies and 1 with anti-PM/Scl antibodies had anti-cyclic citrullinated peptide antibodies; all of them had arthralgia and 1 patient fulfilled the diagnostic criteria for RA. One patient within each group had mixed cryoglobulinemia. Among patients with anti-PM/Scl antibodies, associated autoantibodies included anti-topoisomerase-1 (n = 1 patient), anti-fibrillarin (1), anti-Ro (1), anti-dsDNA (2), and anticentromere antibodies (1). Anti-dsDNA, ANCA with proteinase-3 specificity, anti-La, and anti-Mi2 antibodies were detected in anti-ARS-positive patients (1 each). Eight patients had 2 autoantibodies, including 5 with anti-PM/Scl and 3 with anti-ARS, and 3 patients had at least 3 autoantibodies (1 with anti-PM/Scl, 2 with anti-ARS).
Treatment and outcomes
All patients received oral corticosteroids, starting with prednisone 0.5–1 mg × kg−1 × day−1 equivalent dose. The proportions of patients with improvement or stable or worsening disease did not differ (Table 3, Table 8). The proportion of patients requiring immunosuppressive therapy was higher in the anti-ARS group (p = 0.02), because of severe myositis (in 2 patients also treated with intravenous immunoglobulin, alone or associated with plasma exchanges and rituximab) or arthritis (in 2 anti-ARS-positive patients also treated with methotrexate followed by leflunomide). Methotrexate was required in only 1 anti-PM/Scl-positive patient for extrapulmonary involvement, i.e., corticosteroid-resistant arthritis.
The median duration of followup was 5.5 years (range 3.8–15.3) for the anti-PM/Scl group and 2.4 years (range 0.5–21.4) for the anti-ARS group. The mean interval from onset of pulmonary symptoms to the last followup (or death) was 5.0 years (range 1.0–11.0) for the anti-PM/Scl group and 2.3 years (range 0.5–21.4) for the anti-ARS group. One patient with anti-ARS antibodies died from subacute respiratory failure 4 months after the onset of ILD. Treatment and outcome did not significantly differ between the 2 groups at last visit (Table 8). Two patients in each group needed longterm oxygen therapy. Muscle manifestations improved with immunosuppressive treatment in all patients. Severe drug-related side effects occurred in 3 patients (hematologic toxicity with cyclophosphamide in 2, methotrexate-induced pneumonitis in 1 patient).
DISCUSSION
A similar proportion of patients with either anti-ARS or anti-PM/Scl antibodies was found in our cohort evaluated for ILD in a tertiary care center. All anti-PM/Scl patients had symptomatic pulmonary involvement with a globally favorable prognosis comparable to that of patients with anti-ARS antibodies. In contrast to anti-ARS-related ILD, anti-PM/Scl-related ILD usually developed after extrapulmonary manifestations of CTD, and rarely preceded them or remained isolated. The extrapulmonary manifestations were characterized by a similar prevalence of Raynaud phenomenon, arthralgia, and mechanic’s hands. Manifestations associated with anti-PM/Scl and anti-ARS antibodies differed only by a significantly lower prevalence of clinical myositis.
Features of ILD associated with anti-PM/Scl antibodies have not previously been reported in detail5,21. The most common HRCT findings of anti-PM/Scl-related ILD were basilar irregular opacities with areas of ground-glass attenuation, reticulation, and little architectural distortion, a pattern suggestive of NSIP, and were consistent with previous series of DM/PM-associated ILD10,22–24. The prevalence of traction bronchiectases, which are common in idiopathic NSIP, varies across the series in DM/PM10,22,23,25. Honeycombing, which is considered a hallmark of UIP, was absent in patients with anti-PM/Scl antibodies, consistent with a NSIP pattern on HRCT and a favorable outcome22,23,26.
Interestingly, anti-PM/Scl-associated ILD in the absence of overt CTD manifestations and similar to conditions reported in patients with anti-ARS antibodies27,28 has not been reported previously. This condition has likely been underrecognized, because the anti-PM/Scl autoantibodies are often not measured in the absence of clinical myositis or SSc manifestations29. In addition, our series was characterized by the low prevalence of definite CTD, highlighting the need for careful assessment of patients presenting with ILD in search of manifestations of formes frustes of CTD, notably amyopathic DM (DM rash without myositis) and SSc sine scleroderma characterized by visceral organ involvement (lungs, kidneys, gastrointestinal tract) with abnormal nailfold capillaries and ANA30. Thus, recognition of the underlying CTD requires careful examination of patients, including Raynaud phenomenon, cutaneous changes suggestive of SSc or DM, arthralgia of hands, muscle symptoms, sicca symptoms, nailfold capillaroscopy, and measurement of muscle enzyme levels. Autoantibodies could be evaluated comprehensively, based on ANA pattern on immunofluorescence, especially in cases of the cytoplasmic pattern (suggestive of anti-ARS antibodies or anti-signal recognition particle antibodies) or nucleolar pattern (suggestive of anti-PM/Scl, anti-Th/To, anti-fibrillarin, or anti-RNA polymerase I antibodies). However, as some patients have no ANA and often no typical cytoplasmic or nucleolar pattern, specific antibodies should be analyzed in all suspected cases, independently of the detection of ANA31. Early diagnosis of an underlying CTD may lead to prompt identification of life-threatening manifestations that may follow diagnosis of ILD, such as scleroderma renal crisis or severe involvement of pharyngeal function in PM/DM. Further, diagnosing the CTD may contribute to adapting therapy and evaluating the prognosis of disease32. There are no evidence-based treatment recommendations for patients with ILD and anti-ARS or anti-PM/Scl antibodies, and management should be adapted individually, based on the severity and outcome of ILD regardless of the immunology profile. It is our practice to use oral corticosteroids as first-line treatment, with additional immunosuppressive therapy in patients with progressive disease despite corticosteroids or in whom disease control requires daily prednisone doses of 20 mg or higher.
Surprisingly, 11 of 21 patients had a combination of autoantibodies reported as specific for CTD subsets (e.g., anti-dsDNA and lupus, anti-cyclic citrullinated peptide antibody and RA, anti-Mi2 and DM, or anti-Scl70/anti-fibrillarin and diffuse cutaneous SSc) or mutually exclusive (e.g., double anti-ARS positivity, anti-Mi2 and anti-ARS, or presence of 2 antinucleolar autoantibodies)33. Extensive immunological assays have shown that patients with DM/PM are characterized by frequent association of autoantibodies with marked serological heterogeneity rather than single specificity34. Interpretation of immunological test results and the inference between the presence of autoantibodies and clinical symptoms may be hazardous in the absence of a study specifically dedicated to this complex condition35. To confirm the presence of true-positive anti-PM/Scl and exclude false-positives, we performed 2 ELISA using different antigens, i.e., a human recombinant PM/Scl-100 protein and the PM1-α antigen. All 9 patients with anti-PM/Scl autoantibodies were positive in both assays, including the 3 patients without nucleolar fluorescence on ANA. The specificity of the ELISA for anti-PM/Scl autoantibodies using the PM1-α antigen was reported to be 93% for the diagnosis of myositis/SSc overlap36. As the majority of our patients with anti-ARS still have limited followup experience, it cannot be excluded that some may eventually develop a more overt form of CTD.
The peculiar clinical pattern of patients with anti-ARS antibodies, i.e. myositis, ILD, arthritis, sclerodactyly, sicca syndrome, Raynaud phenomenon, and mechanic’s hands, was shared by patients with anti-PM/Scl antibodies. Our findings are supported by a previous study showing similar clinical features associated with anti-PM/Scl and anti-ARS antibodies except for sclerodactyly that was not assessed6. Reciprocally, patients with anti-ARS antibodies commonly have manifestations belonging to both the SSc and the PM/DM spectra that are usually described as related to anti-PM/Scl antibodies, including slight SSc-like cutaneous changes or involvement of the gastrointestinal tract2,8,27,37–41. We therefore consider that patients with anti-PM/Scl and anti-ARS antibodies belong to the same clinical diagnostic group. Of note, alleles of the 8.1 Caucasian ancestral haplotype including HLA-DRB1*03, DQA1*05, and DQB1*02 correlate closely with both autoantibodies3,4,8,42. HLA-DRB1*03 is implicated in the development of ILD among patients with CTD and predicted the occurrence of ILD in mixed CTD43, SSc44, and DM/PM, irrespective of myositis subtype or presence of autoantibodies in the latter42. HLA-DRB1*03 may represent a determinant factor in the development of systemic manifestations in patients with anti-ARS and anti-PM/Scl antibodies6. Other HLA markers also differentiate anti-PM/Scl from anti-ARS patients. Hence, HLA-DPB1*0101 allele, a gene separated from the 8.1 Caucasian ancestral haplotype, was shown to be associated only with anti-Jo1-positive PM/DM compared to those positive for anti-PM/Scl9. It is conceivable that the HLA haplotype correlates better than the subtype of autoantibody with the clinical phenotype.
There are several limitations to our study. First, not all patients had a lung and muscle biopsy, due to the retrospective study design. However, in the absence of muscle weakness or muscle enzyme elevation, systematic invasive testing would be difficult to warrant. Second, the small numbers of subjects in each group limited the statistical power of the comparison and did not allow testing for non-identity. A larger multicenter prospective study is needed to confirm our analysis. Third, multiple comparisons may statistically yield spurious false-positive results, and our study was exploratory and hypothesis-generating. Fourth, systematic screening for non-Jo1 ARS antibodies was not performed in our population of ILD patients, and it is conceivable that the frequency of anti-PL-7 and PL-12 antibodies may have been underestimated31. Fifth, this series represents recruitment from a tertiary department of respiratory medicine, and results may not be applicable in another setting.
Our hypothesis-generating study supports the notion that anti-PM/Scl and anti-ARS antibodies are associated with similar ILD.
We report an original series of patients with ILD associated with anti-PM/Scl antibodies, including a subset of patients who did not share the classical myositis-scleroderma syndrome, but presented with no or minimal manifestations of myositis. Our study showed that the populations delineated by anti-PM/Scl and anti-ARS antibodies are not as homogeneous as previously thought, and largely overlap, with the exception of more frequent overt myositis in patients with anti-ARS antibodies. Similar clinical and especially pulmonary manifestations are observed in patients with anti-PM/Scl or anti-ARS antibodies, challenging the clinical specificity of the antisynthetase syndrome.
Acknowledgments
We are grateful to Dr. Lara Chalabreysse and Dr. David Meyronet for help in reviewing lung histopathology. We thank Dr. Delphine Gamondes, Prof. Philippe Douek, and Prof. Didier Revel, who performed HRCT of the chest.
Footnotes
- Accepted for publication November 3, 2009.