Abstract
Objectives
Autoantibodies recognizing cytosolic 5′-nucleotidase 1A (NT5C1A) are found in adult patients with myositis and other autoimmune diseases. They are especially prevalent in adults with inclusion body myositis (IBM), in which they are associated with more severe weakness and higher mortality. This study was undertaken to define the prevalence and clinical features associated with anti-NT5C1A autoantibodies in juvenile myositis.
Methods
We screened sera from 380 juvenile myositis patients, 30 patients with juvenile idiopathic arthritis (JIA), and 92 healthy control children for anti-NT5C1A autoantibodies. Clinical characteristics were compared between myositis patients with and without anti-NT5C1A autoantibodies.
Results
Anti-NT5C1A autoantibodies were present in 102 of 380 (27%) patients with juvenile myositis and in 11 of 92 (12%) healthy control children (p=0.002) and 27% of children with JIA (p=0.05 vs. controls). Eighty-three of 307 (27%) juvenile dermatomyositis and 16 of 46 (35%) juvenile overlap myositis patients' sera were anti-NT5C1A autoantibody positive (p<0.01 vs. healthy controls for each), but only 3 of 27 (11%) sera from juvenile polymyositis patients were anti-NT5C1A positive. Juvenile myositis patients with and without anti-NT5C1A autoantibodies had similar clinical phenotypes. However, anti-NT5C1A autoantibody positive myositis patients had greater pulmonary symptoms at diagnosis (p=0.005), more frequent hospitalizations (p=0.01) and required a larger number of medications (p<0.001).
Conclusion
Anti-NT5C1A autoantibodies are present in more than one-quarter of children with juvenile myositis and JIA compared with only 12% of healthy children, suggesting NT5C1A is myositis-associated in children. As in adults with IBM, juvenile myositis patients with anti-NT5C1A autoantibodies have more severe disease.
Introduction
Myositis is a diverse group of autoimmune diseases that includes polymyositis (PM), dermatomyositis (DM), and inclusion body myositis (IBM).[1] Myositis patients frequently have autoantibodies associated with distinct clinical phenotypes.[2] Autoantibodies found exclusively in myositis patients are known as myositis-specific autoantibodies (MSAs), whereas those that are also found in other autoimmune conditions are known as myositis-associated autoantibodies (MAAs). Interestingly, while the same autoantibodies found in adult myositis are also common in juvenile myositis, they are not always seen in the same frequency or associated with the same clinical features in both age groups. For example, autoantibodies against p155/140 (transcriptional intermediary factor1; TIF-1) are highly associated with malignancy in adults, but not in children.[3]
Autoantibodies recognizing cytosolic 5′-nucleotidase 1a (NT5C1A) were initially described in adults with IBM. [4, 5] In these and subsequent studies, the reported prevalence of anti-NT5C1A autoantibodies in IBM patients has ranged from 33% to 80%, depending on the patient population, type of assay used, and the cut-offs chosen to define a positive result. [4-12]. Importantly, since all studies have shown that less than 10% of adults with PM are positive for anti-NT5C1A, detection of this autoantibody may be a potentially useful biomarker for distinguishing these two forms of myositis. Subsequent studies demonstrated that anti-NT5C1A autoantibodies are also found in about 10-15% of adult DM patients and in 4-36% of adult patients with lupus or Sjogren's syndrome.[8, 9, 12] Given that they are found in myositis as well as in other autoimmune diseases, anti-NT5C1A can be defined as an adult MAA.
In adults with DM, no distinguishing clinical features have been identified for patients with anti-NT5C1A autoantibodies.[8, 9] However, one study has shown that anti-NT5C1A autoantibody positive IBM patients are more likely to have dysphagia, facial weakness, reduced forced vital capacity, and to require assistive devices than those without these autoantibodies. [7] Consistent with the possibility that anti-NT5C1A autoantibodies confer a more severe disease phenotype, another recent paper showed that IBM patients with this serologic feature had a higher adjusted mortality risk than autoantibody negative IBM patients.[13]
As anti-NT5C1A autoantibodies have not been previously described in children with myositis, the purpose of the present study was to define the prevalence and clinical features of anti-NT5C1A autoantibodies in a large cohort of patients with juvenile myositis. We also examined whether anti-NT5C1A autoantibodies are present in another pediatric rheumatologic condition, juvenile idiopathic arthritis (JIA), towards determining if they are myositis-specific or -associated autoantibodies in children.
Patients and Methods
Patients and serum samples
Patients from the Childhood Myositis Heterogeneity Collaborative Study with probable or definite myositis by Bohan and Peter criteria [14] with a serum sample available for NT5C1A autoantibody testing were included in the study. Serum samples, stored at -80 degrees Celsius from 1-27 years, were available from 380 children with myositis, 30 with JIA and 92 healthy control children. The myositis patients included 307 (81%) with juvenile DM (JDM), 27 (7%) with juvenile PM (JPM) and 46 (12%) with juvenile connective tissue disease-myositis overlap (JCTM) syndromes. The JCTM subgroup included 7 with juvenile idiopathic arthritis, 14 with juvenile systemic lupus, 11 with systemic sclerosis, and 14 with various other rheumatic conditions. Healthy control children enrolled in the same studies and were often age, gender and race-matched to myositis patients. They had no family history of autoimmune disease in first degree relatives, no history of infections or immunizations within the two months prior to enrollment, and no history of chronic inflammatory diseases.
All subjects were enrolled in investigational review board-approved natural history studies from 1990 to 2016, as previously described,[15] and all provided informed consent. A standardized physician questionnaire captured demographics, clinical features, laboratory features, environmental exposures at illness onset or diagnosis, as well as therapeutic usage and responses.[15] Seven organ system symptom scores at diagnosis, defined as the number of symptoms present divided by the number of symptoms assessed, and an overall clinical symptom score as the average of the seven individual organ symptom scores, were calculated as previously described.[16, 17] Complete clinical response and remission were defined as at least 6 months of inactive disease on or off therapy, respectively.[17] A course of treatment was defined as a single episode from beginning of administration of a given medication to the termination of treatment with that medication, or combination of medications, in each patient. The majority of patients had verification of the data via medical record review. HLA typing of DRB1 and DQA1 alleles was performed as previously described [18]. Sera from a control group of healthy children was obtained in the same protocols, and sera from 39 patients with JIA were obtained from the NIEHS Twin Sibling study. [19]
Myositis autoantibody assays
Anti-NT5C1A autoantibody detection
As previously described, lysates of HEK 293 cells transfected with NT5C1A and non-transfected HEK 293 cell lysates were electrophoresed on sodium dodecyl sulfate–polyacrylamide gels, transferred to nitrocellulose membranes, and immunoblotted with either a positive control rabbit polyclonal antibody recognizing NT5C1A (Applied Biological Materials) or human sera diluted 1:1,000 for 1 hour. To achieve uniformity between assays, each assay run included a positive control (i.e., rabbit anti-NT5C1A immunoblotting non-transfected vs. NT5C1A-transfected HEK 293 lysates). [8] Exposures of immunoblots in which the positive control lanes were of equivalent intensity were used for scoring. Human sera that recognized the 43-kd NT5C1A protein in NT5C1A transfected cells but not in untransfected cells were considered to be positive for anti-NT5C1A autoantibodies. All blots were independently scored by two readers (RY and ALM, who were blinded to sample identity) as being positive or negative for anti-NT5C1A reactivity. The interrater reproducibility for the positivity of this study was excellent, with an agreement of 96.1% and a Cohen's kappa coefficient of 0.90. In the few cases where there was disagreement, a third blinded reader (IPF) adjudicated.
Other myositis autoantibodies were tested by validated methods, including protein and RNA immunoprecipitation (IP) using radiolabeled HeLa or K562 cell extracts and double immunodiffusion.[15] For anti-p155/140, anti-MJ (Nuclear Matrix Protein 2, NXP2), and anti-melanoma-differentiation-associated gene 5 (MDA5) autoantibodies, serum samples were screened by IP, with confirmation testing by IP-immunoblotting.[15] Anti-3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) autoantibodies were screened by enzyme-linked immunosorbent assay (ELISA) and confirmed by immunoprecipitation using 35S-methionine-labeled HMGCR protein produced by in-vitro transcription and translation as previously described.[20]
Analysis
Dichotomous variables were expressed as percentages and absolute frequencies, and continuous features were reported as means and standard deviations (SD). Pairwise comparisons for categorical variables between groups were made using chi-square test or Fisher's exact test, as appropriate, while continuous variables were compared using Student's t-test. Logistic and linear regression were used to adjust the comparisons for possible confounding variables, including the year of diagnosis, length of follow-up, and myositis autoantibodies. Creatine kinase, a highly positively skewed variable, was expressed as median, first, and third quartile for descriptive purposes and transformed through a base-10 logarithm for analysis. All statistical analyses were performed using Stata/MP 14.1. As this was an exploratory study, a 2-sided p value of 0.05 or less was considered statistically significant except for the HLA analyses, in which the Benjamini and Hochberg method to correct for multiple comparisons was performed.
Results
Anti-NT5C1A autoantibodies were more prevalent in juvenile myositis patients than in healthy control children (27% vs. 12%; p=0.002) (Table 1). Sera from 27% of JDM, 11% of JPM, and 35% of JCTM patients had anti-NT5C1A autoantibodies. Among these clinical subgroups, the prevalence of anti-NT5C1A autoantibodies was significantly increased in JDM (p=0.003) and JCTM patients (p=0.001), including those with lupus-myositis overlap (p=0.02), compared to healthy controls. Patients with JCTM had a higher prevalence of anti-NT5C1A autoantibodies than those with JPM (p=0.03); otherwise there were no significant differences in autoantibody prevalences among the JDM, JPM, and JCTM subgroups. Of note, the prevalence of anti-NT5C1A autoantibodies was significantly increased in the children with JIA compared to healthy controls (27% vs. 12%; p=0.02) (Table 1).
Table 1. Prevalence of anti-NT5C1A in the sera of pediatric patients.
| Total (n=502) % (N) |
|
|---|---|
|
| |
| Juvenile myositis (n=380) | 27% (102) ** |
| Juvenile Dermatomyositis (n= 307) | 27% (83) ** |
| Juvenile Polymyositis (n= 27) | 11% (3) |
| Juvenile myositis overlap syndromes (n=46) | 35% (16) ** |
| Juvenile myositis overlapping with Juvenile systemic lupus erythematosus (n=14) | 36% (5) * |
| Juvenile myositis overlapping with Juvenile systemic sclerosis (n=11) | 27% (3) |
| Juvenile myositis overlapping with Juvenile idiopathic arthritis (n=7) | 14% (1) |
| Juvenile myositis overlapping with other pediatric autoimmune diagnosis (n=14) | 50% (7) ** |
| Juvenile idiopathic arthritis (n=30) | 27% (8) * |
| Healthy pediatric controls (n=92) | 12% (11) |
<0.05;
<0.01;
<0.001
%= percentage positive for anti-NT5C1A autoantibodies, N=number of sera testing positive for anti-NT5C1A autoantibodies.
There were no significant differences in gender or race between juvenile myositis patients with and without anti-NT5C1A autoantibodies (Table 2). There was an association between the presence of anti-NT5C1A and anti-p155/140 autoantibodies, and fewer of the anti-NT5C1A autoantibody positive patients were negative for other MSAs. Specifically, 34% of the anti-p155/140-positive patients were also positive for anti-NT5C1A compared with just 24% of anti-p155/140-negative patients (p=0.04). Conversely, patients with anti-NT5C1A autoantibodies more frequently had co-existing anti-p155/140 autoantibodies than those who were negative for anti-NT5C1A autoantibodies (42% vs. 31%, p= 0.04). Of note, anti-NT5C1A autoantibody positive patients were more likely to have shorter follow-up times (4.4 vs. 6.2 years, p=0.01) compared to anti-NT5C1A autoantibody negative patients (Table 2). Given these differences, subsequent multivariate analyses were adjusted for follow-up duration and the presence of myositis autoantibodies (including anti-p155/140 autoantibodies) as well as the year of diagnosis (considering that treatment strategies may have changed over time).
Table 2. Demographic and myositis autoantibody features of juvenile myositis patients according to anti-NT5C1A autoantibody status.
| Anti-NT5C1A Ab+ (n=102) | Anti-NT5C1A Ab- (n=278) | p-value | |
|---|---|---|---|
|
| |||
| Age at diagnosis (years) | 9.5 (4.4) | 8.8 (4.3) | 0.2 |
| Age at enrollment (years) | 11.8 (6.0) | 12.7 (7.4) | 0.3 |
| Delay to diagnosis (years) | 0.69 (1.00) | 0.71 (1.23) | 0.9 |
| Follow-up (years) | 4.4 (4.6) | 6.2 (6.8) | 0.01 |
| Female | 70% (71/102) | 72% (200/278) | 0.7 |
| Race | |||
| White | 65% (66/102) | 65% (182/278) | 0.9 |
| Black | 12% (12/102) | 17% (47/278) | 0.2 |
| Hispanic | 7% (7/102) | 6% (17/278) | 0.8 |
| Other races | 17% (17/102) | 12% (32/278) | 0.2 |
| Myositis autoantibodies | |||
| Anti-p155/140 | 42% (42/100) | 31% (82/268) | 0.04 |
| Anti-NXP2 | 22% (22/101) | 21% (58/274) | 0.9 |
| Anti-MDA5 | 10% (10/102) | 8% (22/275) | 0.6 |
| Anti-synthetase autoantibodies | 4% (4/96) | 4% (11/273) | 1.0 |
| Anti-SRP | 0% (096) | 3% (7/273) | 0.2 |
| Anti-Mi2 | 4% (4/96) | 3% (9/267) | 0.8 |
| Anti-HMGCR | 1% (1/102) | 1% (3/278) | 1.0 |
| MSA negative | 19% (19/102) | 29% (77/269) | 0.05 |
Dichotomous variables were represented as percentage (count/total) and continuous variables as mean (SD).
In the multivariate analysis comparing their clinical features, anti-NT5C1A autoantibody-positive patients were more likely to have Raynaud's phenomenon and V-sign or shawl rashes compared to anti-NT5C1A autoantibody-negative patients (Table 3). The prevalences of dysphonia and photosensitivity were increased among anti-NT5C1A autoantibody positive patients in the univariate analysis but not in the multivariate analysis, and there was a trend towards more frequent interstitial lung disease in those with anti-NT5C1A autoantibodies (Table 3). There were no other significant differences in the prevalences of the main muscle, lung, gastrointestinal, constitutional involvement, or other cutaneous manifestations between these two groups in either univariate or multivariate analyses (Table 3).
Table 3. Clinical features according to anti-NT5C1Aautoantibody status.
| Ever present | ||||
|---|---|---|---|---|
|
| ||||
| Anti-NT5C1A Ab+ (n=102) | Anti-NT5C1A Ab - (n=278) | univariate p-value | multivariate p-value | |
|
| ||||
| Muscle involvement | ||||
| Proximal weakness | 99% (101/102) | 100% (277/278) | 0.5 | 0.6 |
| Distal weakness | 50% (51/101) | 45% (122/271) | 0.3 | 0.4 |
| Muscle atrophy | 35% (36/102) | 38% (105/274) | 0.6 | 0.8 |
| Myalgia | 66% (67/101) | 63% (172/271) | 0.6 | 0.5 |
| Falling | 44% (45/102) | 45% (122/274) | 0.9 | 0.8 |
| Dysphonia | 40% (40/101) | 29% (80/275) | 0.05 | 0.2 |
| Lung involvement | ||||
| Interstitial lung disease | 13% (13/102) | 7% (20/276) | 0.09 | 0.06 |
| Dyspnea on exertion | 35% (36/102) | 27% (74/273) | 0.1 | 0.2 |
| Joint involvement | ||||
| Arthritis | 57% (58/102) | 49% (137/277) | 0.2 | 0.5 |
| Arthralgia | 66% (67/102) | 64% (177/276) | 0.8 | 0.7 |
| Joint contractures | 68% (69/102) | 57% (159/277) | 0.07 | 0.4 |
| Skin involvement | ||||
| Heliotrope rash | 84% (85/101) | 77% (214/277) | 0.1 | 0.6 |
| Gottron's papules | 87% (89/102) | 81% (223/277) | 0.1 | 0.5 |
| Calcinosis | 27% (28/102) | 31% (86/278) | 0.5 | 0.5 |
| Raynaud's phenomenon | 17% (17/102) | 14% (39/276) | 0.5 | 0.03 |
| Mechanic's hands | 9% (9/100) | 7% (18/275) | 0.4 | 0.2 |
| V or Shawl sign rash | 43% (44/102) | 26% (72/276) | 0.001 | 0.02 |
| Malar rash | 81% (83/102) | 66% (184/278) | 0.004 | 0.1 |
| Photosensitivity | 62% (61/99) | 43% (116/272) | 0.001 | 0.09 |
| Linear extensor erythema | 42% (42/101) | 34% (91/271) | 0.2 | 0.5 |
| Gastrointestinal involvement | ||||
| Dysphagia | 44% (45/102) | 39% (109/277) | 0.4 | 0.5 |
| Regurgitation | 25% (26/102) | 19% (52/277) | 0.2 | 0.4 |
| Systemic involvement | ||||
| Weight loss | 44% (44/101) | 42% (115/277) | 0.7 | 0.9 |
| Fever | 30% (30/100) | 31% (83/267) | 0.8 | 0.6 |
| Early symptom scores | 0.27 (0.12) | 0.23 (0.11) | 0.005 | 0.1 |
| Total | ||||
| Muscle | 0.38 (0.19) | 0.38 (0.20) | 0.8 | 0.9 |
| Joint | 0.52 (0.41) | 0.43 (0.42) | 0.06 | 0.4 |
| Cutaneous | 0.29 (0.15) | 0.24 (0.13) | 0.002 | 0.3 |
| Gastrointestinal | 0.09 (0.13) | 0.07 (0.10) | 0.03 | 0.06 |
| Pulmonary | 0.13 (0.17) | 0.08 (0.15) | 0.004 | 0.005 |
| Cardiac | 0.03 (0.08) | 0.03 (0.08) | 0.8 | 0.6 |
| Constitutional | 0.41 (0.26) | 0.39 (0.27) | 0.4 | 0.8 |
Abbreviations: NT5C1A, cytosolic 5′-nucleotidase 1A; Ab, autoantibody. Dichotomous variables were represented as percentage (count/total), continuous variables as mean (SD) and the creatine kinase was presented as median (Q1-Q3).
A number of differences between anti-NT5C1A autoantibody-positive and -negative patients suggested that patients with these autoantibodies had more severe disease. In the multivariate analysis, anti-NT5C1A-positive patients had higher pulmonary symptom scores (p=0.003) and showed a trend towards higher total symptom score (p=0.09) at the time of diagnosis (Table 3). Anti-NT5C1A autoantibody-positive patients had a higher mean number of hospitalizations (1.6 vs. 1.1, p=0.01) than those without this autoantibody. Furthermore, anti-NT5C1A autoantibody-positive patients required a greater number of medication treatments per year (4.8 vs. 3.6, p=<0.001) with a higher percentage of patients requiring intravenous pulse steroids (78% vs. 47%, p<0.001), intravenous immunoglobulin (67% vs. 24%, p<0.001), and use of other immunosuppressive medications (33% vs. 20%, p=0.04) (Table 4). Anti-NT5C1A autoantibody-positive patients had a trend for a more severe functional class at last assessment, but this did not reach statistical significance in the multivariate analysis (29% vs. 18% functional class 2, p=0.08).
Table 4. Disease outcomes and medications received according to anti-NT5C1A autoantibody status.
| Anti-NT5C1A Ab+ (n=102) | Anti-NT5C1A Ab – (n=278) | Univariate p-value | Multivariate p-value | |
|---|---|---|---|---|
|
| ||||
| Disease Course | ||||
| Monocyclic course | 16% (12/76) | 23% (53/229) | 0.2 | 0.2 |
| Polycyclic course | 18% (14/76) | 24% (56/229) | 0.3 | 0.6 |
| Chronic continuous course | 66% (50/76) | 52% (120/229) | 0.04 | 0.2 |
| Steinbrocker functional class at final assessment (mean) | 1.5 (0.8) | 1.4 (0.8) | 0.6 | 0.6 |
| Muscle enzymes | ||||
| Peak creatine kinase, IU/L | 1010 (296-3971) | 672 (252-5460) | 0.7 | 0.2 |
| Peak aldolase, IU/L | 16.9 (23.9) | 20.8 (37.2) | 0.3 | 0.7 |
| Severity at onset (mean, 0-4) | 2.1 (1.4) | 2.2 (0.9) | 0.4 | 0.3 |
| Mortality | 4% (4/102) | 3% (9/278) | 0.8 | 0.3 |
| Hospitalized | 65% (62/96) | 55% (148/268) | 0.1 | 0.1 |
| Number of hospitalizations | 1.6 (2.3) | 1.1 (1.7) | 0.04 | 0.01 |
| Wheelchair use | 21% (21/100) | 18% (47/268) | 0.4 | 0.1 |
| Response to treatment | ||||
| Complete clinical response | 22% (19/87) | 33% (74/225) | 0.06 | 0.6 |
| Remission | 15% (13/88) | 27% (63/232) | 0.02 | 0.5 |
| Total # of medications used | 4.8 (2.0) | 3.6 (2.0) | < 0.001 | < 0.001 |
| Medications received | ||||
| Oral steroids | 99% (87/88) | 99% (230/232) | 1.0 | 0.6 |
| Intravenous pulsed steroids | 78% (69/88) | 47% (110/232) | < 0.001 | < 0.001 |
| Methotrexate | 83% (73/88) | 71% (164/232) | 0.03 | 0.4 |
| Intravenous immunoglobulin | 67% (59/88) | 24% (55/232) | < 0.001 | < 0.001 |
| Other DMARDs | 33% (29/88) | 20% (46/232) | 0.01 | 0.04 |
Abbreviations: NT5C1A, cytosolic 5′-nucleotidase 1A; Ab, autoantibody; DMARDs, disease modifying anti-rheumatic drugs.
Dichotomous variables were represented as percentage (count/total), continuous variables as mean (SD) and the creatine kinase was presented as median (Q1-Q3).
The multivariate analyses were adjusted for the presence or absence of each MSA. Consequently, our finding that anti-NT5C1A autoantibodies are associated with more severe disease is independent of the MSA status of the patient. Furthermore, when analyzed separately, MSA negative, anti-p155/140-positive, and anti-NXP2-positive patients all had evidence of more severe disease when they were also anti-NT5C1A-positive (data not shown).
To determine whether there is an immunogenetic association with anti-NT5C1A autoantibodies, we compared the prevalence of HLA DRB1 and DQA1 alleles between Caucasian juvenile myositis patients with and without this autoantibody. We also compared Caucasian patients with anti-NT5C1A autoantibodies to healthy Caucasian subjects and all anti-NT5C1A-positive patients to all anti-NT5C1A-negative patients. However, after multiple correction adjustment, no statistically significant associations between anti-NT5C1A autoantibodies and these class II MHC alleles were found in any of these comparisons.
Discussion
In adults, autoantibodies recognizing NT5C1A are considered to be MAAs rather than MSAs because they are found not only in patients with myositis, but also in patients with lupus and Sjogren's syndrome.[8, 9] In this study, we show that anti-NT5C1A autoantibodies are also MAAs in children since they are found not only in juvenile myositis patients, but also in those with JIA. As expected for a MAA, we found that anti-NT5C1A autoantibodies were frequently found in association with other MSA, especially with anti-p155/140 autoantibodies.
The prevalence of anti-NT5C1A autoantibodies varies among adults with different myositis subtypes. They occur most frequently in those with IBM, less frequently in DM, and least frequently in those with PM. Here, we demonstrate that the prevalence of these autoantibodies in children also varies among myositis subtypes, with 27% of JDM but only 11% of JPM patients testing positive for anti-NT5C1A autoantibodies. In the current study, we also found that anti-NT5C1A autoantibodies are present in 35% of children with myositis-overlap syndromes. Although the number of patients in each myositis overlap subgroup was small, the autoantibodies were found most commonly in those with myositis associated with juvenile lupus (36%), less frequently in myositis associated with juvenile systemic sclerosis (27%), and least frequently in those with myositis associated with juvenile idiopathic arthritis (14%) patients. In adults, the prevalence of anti-NT5C1A autoantibodies was less than 5% in adults with myositis-scleroderma overlap,[9] but future studies will be required to define the prevalence of anti-NT5C1A in adults with other forms of myositis-overlap.
In adults with IBM, anti-NT5C1A autoantibodies have been associated with more severe muscle disease [7] and increased mortality.[13] However, no evidence for an association between anti-NT5C1A autoantibodies and more severe disease in adults with other forms of myositis has been reported. Here, we demonstrate that in patients with juvenile myositis these autoantibodies are associated with more frequent hospitalizations, higher pulmonary symptoms, and increased number of medications used. The underlying reasons for the association of anti-NT5C1A autoantibodies with more severe disease remain unclear, but may relate to a direct effect of the autoantibodies on myofiber protein degradation, as demonstrated in a study of passive transfer of anti-NT5C1A autoantibodies into mice. [21] Another possibility is that a more severe inflammatory response in juvenile myositis predisposes patients to the development of additional immunoreactivities such as that against NT5C1A.
In some instances, individual autoantibodies have significant associations with specific HLA alleles. However, in this study, we did not find an association between the presence of anti-NT5C1A autoantibodies and any class II HLA alleles. This is consistent with a recent report by Limaye, showing no HLA associations in adult IBM patients with anti-NT5C1A autoantibodies. [11]
In the current study, we were surprised to find that 11 out of 94 (12%) healthy children were positive for anti-NT5C1A autoantibodies detected by a previously validated immunoblotting assay. [7] We have demonstrated that 61% of adult IBM patients and only 5% of adult PM patients were anti-NT5C1A autoantibody positive using the immunoblotting method, findings which are in agreement with an established dot blot technique [4] and an established immunoprecipitation technique. [5] In our previous study, we also found that 5% of healthy adult controls were anti-NT5C1A-positive by immunoblotting. [7] While another study reported that none of 32 adult healthy controls were anti-NT5C1A positive by immunoprecipitation [5], the prevalence of anti-NT5C1A autoantibodies in healthy adult controls has not been reported for the dot blot assay [4] or either of two different ELISA assays.[6] Nonetheless, taken together, these comparisons indicate that the anti-NT5C1A immunoblot assay performs as well as other detection methods and does not have an unacceptably high false positive rate. Of note, high titer autoantibodies against Ro52, another common target of the immune system in various systemic autoimmune diseases, are also found in healthy controls as well as in patients with more severe disease manifestations.[22]
When comparing the prevalence of anti-NT5C1A autoantibodies in children and adults within the same clinical groups using the same immunoblotting detection method, younger subjects consistently are more likely to have reactivity to these autoantibodies. For example, 27% of JDM but only 15% of adult DM [7] patients are anti-NT5C1A autoantibody positive (p = 0.03). Similarly, 11% of JPM but only 5% of adult PM patients [7] have these autoantibodies. Given that the same pattern is also observed among healthy children and healthy adults, we hypothesize that in both diseased and healthy groups, immunoreactivity against NT5C1A may decrease with age.
The current study has several limitations. First, the cohort of juvenile myositis patients had some data collected retrospectively, resulting in some missing data, and was collected over more than 20 years, with potential chronology bias. However, we adjusted the variables of this study for the year of diagnosis and tested the distribution of missing values across groups and did not find evidence of a significant bias (data not shown). Second, we only screened a small number of patients with one other systemic autoimmune disease, JIA; the small sample size in this group of patients did not allow us to study differences in severity between patients with and without anti-NT5C1A autoantibodies in a reliable way. Future studies will be needed to determine the full range of pediatric rheumatologic conditions in which anti-NT5C1A autoantibodies are found and to determine whether their presence correlates with disease severity or other clinical features. Third, there is no widely accepted “gold standard” for detecting anti-NT5C1A autoantibodies. However, we have developed and validated an immunoblotting detection method. This technique has a sensitivity (61%) and specificity (95%) for detecting anti-NT5C1A autoantibodies in adult IBM patients that is consistent with most other methods that utilize full-length NT5C1A protein (sensitivity range 47-80% and specificity range 95-100%). [4-12] Of note, one published study utilizing an ELISA method detected anti-NT5C1A autoantibodies in only 37% of IBM patients. However, this assay used 3 short peptides rather than the full length NT5C1A protein and the authors of the study acknowledged that it may have poor sensitivity since it cannot detect reactivity to conformational epitopes. [9]
These limitations notwithstanding, this study shows that anti-NT5C1A autoantibodies are present in approximately one-quarter of patients with juvenile myositis as well as JIA. Furthermore, as shown for adults with IBM, juvenile myositis patients with anti-NT5C1A autoantibodies have more severe disease than those without these autoantibodies. Additional studies will be required to confirm the association with disease severity in JM and to determine whether anti-NT5C1A autoantibodies are a result or a cause of the more severe clinical manifestations seen in the juvenile myositis and adult IBM patients who have them.
Acknowledgments
The authors thank Mariana Kaplan, David Amici, Cassie Parks, Rebecca de Lorenzo, Maria Casal-Dominguez, Assia Derfoul, and Jose Milisenda for their critical reading and suggestions on the manuscript. This manuscript is based on work previously presented and published as a conference abstract at (a) The Global Conference on Myositis in Potomac, Maryland, USA (May 5-8, 2017), (b) The 2017 American College of Rheumatology Annual Meeting in San Diego, California, USA (October 19, 2017), and (c) the European League Against Rheumatism Annual European Congress of Rheumatology, Madrid, Spain (June 14-17, 2017).
Funding: This work was supported in part by the National Institute of Environmental Health Sciences and the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health. Iago Pinal-Fernandez and Takayuki Kishi were supported by research fellowships from The Myositis Association. Ira Targoff is a consultant to the Oklahoma Medical Research Foundation Clinical Immunology Laboratory regarding myositis autoantibody testing.
Footnotes
Competing Interests: None.
Contributorship: All authors were involved in drafting the article or revising it critically for important intellectual content and approved the final manuscript to be submitted and agreed to be accountable for all aspects of the work. Conception and design of the study: RMY, IPF, FWM, LGR, and ALM. Acquisition of data: RMY, IPF, TK, KP, INT, and LGR. Analysis and interpretation of data: RMY, IPF, TK, INT, FWM, LGR, and ALM. There is no one else who fulfils the criteria that has been excluded as an author.
Ethical Approval: All subjects were enrolled in investigational review board-approved natural history studies from 1990 to 2016. The parents of the pediatric patients in this study signed a consent form approved by the Institutional Review Board at the National Institutes of Health.
Data Sharing Agreement: No unpublished data related to this study are publicly available.
Contributor Information
Richard M. Yeker, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD.
Iago Pinal-Fernandez, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD.
Takayuki Kishi, Environmental Autoimmunity Group, National Institutes of Environmental Health Sciences, NIH, Bethesda, MD.
Katherine Pak, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD.
Ira N. Targoff, U.S. Department of Veterans Affairs Medical Center, University of Oklahoma Health Sciences Center, and Oklahoma Medical Research Foundation, Oklahoma City, OK.
Frederick W. Miller, Environmental Autoimmunity Group, National Institutes of Environmental Health Sciences, NIH, Bethesda, MD.
Lisa G. Rider, Environmental Autoimmunity Group, National Institutes of Environmental Health Sciences, NIH, Bethesda, MD.
Andrew L. Mammen, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD.
References
- 1.Dalakas MC. Inflammatory Muscle Diseases. The New England journal of medicine. 2015 Jul 23;373(4):393–394. doi: 10.1056/NEJMc1506827. [DOI] [PubMed] [Google Scholar]
- 2.Rider LG, Miller FW. Deciphering the clinical presentations, pathogenesis, and treatment of the idiopathic inflammatory myopathies. JAMA. 2011 Jan 12;305(2):183–190. doi: 10.1001/jama.2010.1977. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gunawardena H, Wedderburn LR, North J, Betteridge Z, Dunphy J, Chinoy H, et al. Clinical associations of autoantibodies to a p155/140 kDa doublet protein in juvenile dermatomyositis. Rheumatology (Oxford) 2008 Mar;47(3):324–328. doi: 10.1093/rheumatology/kem359. [DOI] [PubMed] [Google Scholar]
- 4.Larman HB, Salajegheh M, Nazareno R, Lam T, Sauld J, Steen H, et al. Cytosolic 5′-nucleotidase 1A autoimmunity in sporadic inclusion body myositis. Ann Neurol. 2013 Mar;73(3):408–418. doi: 10.1002/ana.23840. [DOI] [PubMed] [Google Scholar]
- 5.Pluk H, van Hoeve BJ, van Dooren SH, Stammen-Vogelzangs J, van der Heijden A, Schelhaas HJ, et al. Autoantibodies to cytosolic 5′-nucleotidase 1A in inclusion body myositis. Ann Neurol. 2013 Mar;73(3):397–407. doi: 10.1002/ana.23822. [DOI] [PubMed] [Google Scholar]
- 6.Greenberg SA. Cytoplasmic 5′-nucleotidase autoantibodies in inclusion body myositis: Isotypes and diagnostic utility. Muscle & nerve. 2014 Oct;50(4):488–492. doi: 10.1002/mus.24199. [DOI] [PubMed] [Google Scholar]
- 7.Goyal NA, Cash TM, Alam U, Enam S, Tierney P, Araujo N, et al. Seropositivity for NT5c1A antibody in sporadic inclusion body myositis predicts more severe motor, bulbar and respiratory involvement. J Neurol Neurosurg Psychiatry. 2016 Apr;87(4):373–378. doi: 10.1136/jnnp-2014-310008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lloyd TE, Christopher-Stine L, Pinal-Fernandez I, Tiniakou E, Petri M, Baer A, et al. Cytosolic 5′-Nucleotidase 1A As a Target of Circulating Autoantibodies in Autoimmune Diseases. Arthritis Care Res (Hoboken) 2016 Jan;68(1):66–71. doi: 10.1002/acr.22600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Herbert MK, Stammen-Vogelzangs J, Verbeek MM, Rietveld A, Lundberg IE, Chinoy H, et al. Disease specificity of autoantibodies to cytosolic 5′-nucleotidase 1A in sporadic inclusion body myositis versus known autoimmune diseases. Ann Rheum Dis. 2016 Apr;75(4):696–701. doi: 10.1136/annrheumdis-2014-206691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kramp SL, Karayev D, Shen G, Metzger AL, Morris RI, Karayev E, et al. Development and evaluation of a standardized ELISA for the determination of autoantibodies against cN-1A (Mup44, NT5C1A) in sporadic inclusion body myositis. Auto Immun Highlights. 2016 Dec;7(1):16. doi: 10.1007/s13317-016-0088-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Limaye VS, Lester S, Blumbergs P, Greenberg SA. Anti- C N1A antibodies in South Australian patients with inclusion body myositis. Muscle & nerve. 2016 Apr;53(4):654–655. doi: 10.1002/mus.24989. [DOI] [PubMed] [Google Scholar]
- 12.Muro Y, Nakanishi H, Katsuno M, Kono M, Akiyama M. Prevalence of anti-NT5C1A antibodies in Japanese patients with autoimmune rheumatic diseases in comparison with other patient cohorts. Clin Chim Acta. 2017 Sep;472:1–4. doi: 10.1016/j.cca.2017.07.002. [DOI] [PubMed] [Google Scholar]
- 13.Lilleker JB, Rietveld A, Pye SR, Mariampillai K, Benveniste O, Peeters MT, et al. Cytosolic 5′-nucleotidase 1A autoantibody profile and clinical characteristics in inclusion body myositis. Ann Rheum Dis. 2017 Jan 25; doi: 10.1136/annrheumdis-2016-210282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bohan A, Peter JB. Polymyositis and dermatomyositis (second of two parts) The New England journal of medicine. 1975 Feb 20;292(8):403–407. doi: 10.1056/NEJM197502202920807. [DOI] [PubMed] [Google Scholar]
- 15.Shah M, Mamyrova G, Targoff IN, Huber AM, Malley JD, Rice MM, et al. The clinical phenotypes of the juvenile idiopathic inflammatory myopathies. Medicine (Baltimore) 2013 Jan;92(1):25–41. doi: 10.1097/MD.0b013e31827f264d. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Habers GE, Huber AM, Mamyrova G, Targoff IN, O'Hanlon TP, Adams S, et al. Brief Report: Association of Myositis Autoantibodies, Clinical Features, and Environmental Exposures at Illness Onset With Disease Course in Juvenile Myositis. Arthritis Rheumatol. 2016 Mar;68(3):761–768. doi: 10.1002/art.39466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Oddis CV, Rider LG, Reed AM, Ruperto N, Brunner HI, Koneru B, et al. International consensus guidelines for trials of therapies in the idiopathic inflammatory myopathies. Arthritis Rheum. 2005 Sep;52(9):2607–2615. doi: 10.1002/art.21291. [DOI] [PubMed] [Google Scholar]
- 18.Mamyrova G, O'Hanlon TP, Monroe JB, Carrick DM, Malley JD, Adams S, et al. Immunogenetic risk and protective factors for juvenile dermatomyositis in Caucasians. Arthritis Rheum. 2006 Dec;54(12):3979–3987. doi: 10.1002/art.22216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gan L, O'Hanlon TP, Lai Z, Fannin R, Weller ML, Rider LG, et al. Gene Expression Profiles from Disease Discordant Twins Suggest Shared Antiviral Pathways and Viral Exposures among Multiple Systemic Autoimmune Diseases. PLoS One. 2015;10(11):e0142486. doi: 10.1371/journal.pone.0142486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kishi T, Rider LG, Pak K, Barillas-Arias L, Henrickson M, McCarthy PL, et al. Anti-3-Hydroxy-3-Methylglutaryl-Coenzyme A Reductase Autoantibodies are Associated with DRB1*07:01 and Severe Myositis in Pediatric Myositis Patients. Arthritis Care Res (Hoboken) 2017 Jan 27; doi: 10.1002/acr.23113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tawara N, Yamashita S, Zhang X, Korogi M, Zhang Z, Doki T, et al. Pathomechanisms of anti-cytosolic 5′-nucleotidase 1A autoantibodies in sporadic inclusion body myositis. Ann Neurol. 2017 Apr;81(4):512–525. doi: 10.1002/ana.24919. [DOI] [PubMed] [Google Scholar]
- 22.Burbelo PD, Ching KH, Han BL, Bush ER, Reeves WH, Iadarola MJ. Extraordinary antigenicity of the human Ro52 autoantigen. Am J Transl Res. 2010 Mar 15;2(2):145–155. [PMC free article] [PubMed] [Google Scholar]