Abstract
Objective Children with juvenile dermatomyositis (JDM) and antibodies to antimelanoma differentiation-associated gene 5 (anti-MDA5) are at increased risk of severe disease complications, including interstitial lung disease (ILD). Data regarding treatment of disease complications in this patient population are limited. In this study, we examined the disease course of children with JDM and anti-MDA5 antibodies before and after treatment with rituximab (RTX).
Methods Patients aged 2-21 years and seen at the Children’s Hospital at Montefiore between July 2012 and August 2021, with a diagnosis of JDM, positive anti-MDA5 antibodies, and evidence of ILD, and who were treated with RTX were eligible for inclusion. Retrospective clinical and laboratory data were reviewed.
Results Five of 8 patients with positive anti-MDA5 antibodies had evidence of ILD (62.5%). Four patients had data available for review. All patients received at least 5 courses of RTX infusions, with discontinuation of steroids by an average of 12 months after starting RTX and a decrease to fewer than 2 concurrent medications by the fifth course of RTX. Indicators of ILD on high-resolution computed tomography and pulmonary function tests either improved or fully resolved over the course of RTX treatment for all patients. Patients also demonstrated resolution of active cutaneous manifestations and musculoskeletal disease activity.
Conclusion To our knowledge, this is the first study to examine the use of RTX in children with JDM and anti-MDA5 antibodies, with notable improvements in ILD, cutaneous, and musculoskeletal manifestations. Further studies are needed to better understand the efficacy of RTX for JDM disease-related complications.
Juvenile dermatomyositis (JDM) is an autoimmune condition characterized by the inflammation of skin and muscle. JDM accounts for 80% of juvenile idiopathic inflammatory myopathies (JIIMs) among children, with a reported incidence of 2.5 to 4.1 cases per million children per year in the USA.1,2 The majority of patients with JDM present with fatigue, muscle weakness, heliotrope rash, and Gottron papules.1,3-6 However, patients can have a varied clinical presentation, with differing degrees of gastrointestinal, musculoskeletal (MSK), mucocutaneous, and cardiopulmonary involvement.
In recent years, scientific investigations have explored the association of antibodies with clinical phenotypes to better predict outcomes. Myositis antibodies include both myositis-specific autoantibodies (MSAs) and myositis-associated autoantibodies (MAAs). MAAs can be seen in patients with autoimmune disease, with or without JIIMs. MSAs are found in up to 40% to 60% of children with a specific diagnosis of JIIM.7,8 Antibodies against melanoma differentiation-associated gene 5 (MDA5) are a type of MSA and have been reported in 6% to 7.4% of juvenile patients with JIIMs, and 7% to 33% of patients with JDM.8-10 Studies have found that in patients with JIIMs, anti-MDA5 antibodies are associated with rapidly progressive interstitial lung disease (ILD), but with lower creatinine kinase levels and more mild muscle disease.10 ILD is rare in children with JIIMs, but is a potential and serious complication. Children with ILD will typically have mild to absent respiratory symptoms at the time of diagnosis. Diagnosis is made by pulmonary function tests (PFTs) and high-resolution computed tomography (HRCT).
The mainstay of treatment in JDM is high-dose corticosteroids and methotrexate (MTX). Given that B cells are believed to play an immunopathogenic role in JDM, rituximab (RTX), a B cell–depleting anti-CD20 monoclonal antibody has been suggested for the treatment of JDM. The Rituximab in Myositis (RIM) trial was a multicenter, randomized, double-blinded, placebo-phase trial that failed to demonstrate a difference in time to response between early and late treatment arms; however, it found that 83% of the patients randomized to RTX decreased their corticosteroid dose by week 44 of evaluation and that juvenile onset disease predicted better efficacy of RTX.11 Other studies have reported that RTX use in dermatomyositis results in improvement in cutaneous disease activity.12 Two case series report clinical response after RTX in patients with JDM, but no improvement in calcinosis.13,14 Limited data from case series and case reports have also suggested potential efficacy of RTX in the treatment of patients with adult dermatomyositis (DM) or polymyositis and rapidly progressive ILD.13 To date, there are no studies that we know of that specifically examine the use of RTX in the treatment of children with JDM and positive for anti-MDA5 autoantibodies. This study aims to review the clinical features of this specific population before and after the use of RTX to better understand the role of RTX in the treatment of this population.
METHODS
Patient population. Children and adolescents aged 2-21 years and seen in the pediatric rheumatology clinic at the Children’s Hospital at Montefiore between July 2012 and August 2021 with a diagnosis of probable or definite JDM (as defined by Bohan and Peter criteria), positive anti-MDA5 antibodies (as determined by the Oklahoma Medical Research Foundation clinical immunology laboratory), and evidence of ILD, and who were treated with RTX were defined as eligible for inclusion into the study.
Data collection. Retrospective clinical and laboratory data were reviewed using the electronic medical record system at time of JDM diagnosis, at the time of each RTX infusion, and at the most recent visit in the pediatric rheumatology clinic. Data obtained closest in time prior to the first RTX infusion are referred to as baseline. Data reviewed included demographics, date of diagnosis, duration of disease, medication usage, side effects of medication usage, and disease manifestations as evident by clinical, laboratory, and imaging findings. The Childhood Myositis Assessment Scale (CMAS) has good reliability, construct validity, and responsiveness, and is a valid instrument for assessment of muscle strength and endurance in children with JIIMs.15 At our center, a CMAS is obtained at every patient visit and was included in clinical assessment.
Pulmonary manifestations. Full PFTs were obtained on all patients with JDM, with values assessed including total lung capacity (TLC; abnormal < 80%), forced vital capacity (FVC; abnormal < 80%), and diffusing capacity for carbon monoxide (DLCO; suggestive of ILD < 75%). All patients additionally had at least 1 evaluation using HRCT. Prior to treatment with RTX, patients were identified by the treating physician as having ILD if they were noted to have abnormal PFTs (reduced DLCO ± reduced FVC or TLC) and/or abnormal findings on HRCT scan. For the purposes of this study, HRCT scans were subsequently interpreted by a fellowship-trained pediatric radiologist with 10 years of experience (JK), who was blinded to the underlying clinical information and timing of radiographic evaluation, to further elaborate on findings. Imaging was reviewed for findings of consolidation, ground glass opacities, traction bronchiectasis, reticulation, linear opacities, nodular opacities, cystic changes, honeycombing, air trapping, overall distribution, subpleural involvement, consistency with interstitial pneumonia, and change from prior imaging. The location of findings was also reviewed.
Ethics. Due to the retrospective nature of this study, no ethics approval was required according to the authors’ institutions.
RESULTS
Demographics and baseline laboratory evaluation. Eight patients were identified with JDM and anti-MDA-5 antibodies. Of those patients, 5 patients had evidence of ILD. Unfortunately, 1 young patient presented at diagnosis in critical condition with respiratory failure and passed away just as therapy was initiated, and thus was not included in this study. Ultimately, 4 patients were identified who met inclusion criteria, with a median age at diagnosis of 8 (range 3-12) years. Two of the patients were female and 2 were male. Three of the 4 patients were Hispanic, and 1 patient was non-Hispanic Black. Three of the 4 patients were noted to have elevated antinuclear antibodies. One patient was positive for anti-Ro/SSA antibodies (patient D), and 1 patient was positive for anti-RNP antibodies (patient A); however, these patients met Bohan and Peter criteria for JDM and did not demonstrate any additional clinical findings suggestive of overlapping connective tissue disease.
Medications. All patients received at least 5 courses of RTX infusions, with an average difference of 6.5 months between RTX infusions (Table). All courses of RTX were defined as 750 mg/m2/dose for 2 doses, given 2 weeks apart (maximum dose = 1000 mg). Patients’ CD19 counts and CD19 percentages were observed to decrease after starting RTX. Patient B had an anaphylactic reaction during her sixth infusion of RTX, after which RTX infusions were stopped. No other adverse side effects of RTX were noted. The patients’ concurrent medications are listed in the Table. All patients started out at baseline on at least 3 concurrent medications and decreased to fewer than 2 medications by the fifth course of RTX. All patients discontinued steroids an average of 12 months after starting RTX (patient A: 16 months; patient B: 20 months; patient C: 3 months; patient D: 9 months).
Clinical and laboratory biomarkers over the duration of RTX therapy (n = 4).
Pulmonary. Three patients had PFT testing during the study period (Table). Patient C was too young to properly complete PFT testing and had no results to review. All 3 patients with available PFT data were noted to have either restrictive lung defects or diffusion defects, which are indicators of ILD. All 3 patients had at least 1 TLC < 80%, which was suggestive of a restrictive lung defect. Only patient A was noted to have a persistent restrictive lung defect. Two patients were noted to have a diffusion defect (patients B and D). PFT abnormalities, as measured by TLC, FVC, and DLCO were noted to improve over time. The radiologists’ impressions of patients’ HRCT are reported in the Table. Three patients had findings suggestive of ILD at baseline, but complete resolution of ILD on HRCT after RTX courses. One patient developed signs of ILD on HRCT while receiving RTX, but there was subsequent improvement in findings of ILD on HRCT at the most recent visit (patient D). An example of pulmonary changes noted on HRCT in patient C before and after RTX is demonstrated in Figure 1.
Axial noncontrast chest computed tomography images in Patient C, demonstrating (A) ground glass opacities at the lung bases before RTX, which (B) fully resolved after RTX. RTX: rituximab.
Cutaneous. Clinicians’ impressions of patients’ active cutaneous findings are reported in the Table. Three patients had heliotrope rash and 3 patients had Gottron papules at time of diagnosis. Other cutaneous findings included ulcers/infarcts, palmar papules, eyelid telangiectasias, and annular lesions. All 4 patients had resolution of active cutaneous findings over time, but 1 patient developed psoriasis. All patients had abnormal nailbed capillaries at baseline, but only 1 patient had persistently abnormal nailbed capillaries by completion of the fifth course of RTX.
MSK. Each patients’ muscle enzyme levels and CMAS scores surrounding courses of RTX are included in Figure 2. All 4 patients were noted to have elevated lactate dehydrogenase (LDH) at diagnosis, and 3 patients were noted to have elevated LDH at baseline (patients A, C, and D). Two patients had persistently elevated LDH up until the fifth course of RTX (patients C and D), but only 1 patient continued to have elevated LDH up until the most recent visit (patient C). Creatine kinase (CK) total was within normal limits for all 4 patients at time of diagnosis, baseline, and by the end of the fifth course of RTX infusion. Total elevated CK was noted only as an isolated incident for 2 patients, both during the third infusion (patients B and C).
Serial assessments of muscle strength, using the CMAS, and muscle enzymes (including creatine kinase, lactate dehydrogenase, and aspartate aminotransferase) demonstrating improvement in musculoskeletal manifestations throughout the duration of RTX therapy. “Most...” refers to the most recent visit. CMAS: Childhood Myositis Assessment Scale; RTX: rituximab.
DISCUSSION
In this study, we have reported on 4 patients positive for anti-MDA5 autoantibodies and pulmonary findings suggestive of ILD on HRCT or PFT, consistent with previous studies that have reported a higher prevalence of ILD in children positive for anti-MDA5 autoantibodies.8,16-19 Fortunately, indicators of ILD on HRCT and PFTs improved or resolved over the course of RTX treatment for all 4 patients, which is consistent with smaller case reports in children with JDM and case series in adult patients with DM. In a case report of a 2-year-old child with rapidly progressive ILD secondary to anti-MDA5–positive JDM, who failed treatment with methylprednisolone, cyclosporine, cyclophosphamide (CYC), and immunoglobulins, and was ultimately treated with extracorporeal membrane oxygenation, there was a notable response observed after treatment with RTX.20 In a separate case report of a 16-year-old female with anti-MDA5–positive JDM and rapidly progressive ILD, improvement was observed only after a combination of RTX and CYC after failed treatment with pulse methylprednisolone, oral prednisolone, tacrolimus, mycophenolate mofetil (MMF), MTX, and monthly intravenous immunoglobulin (IVIG).21 Similarly, a large case series of 21 adults with anti-MDA5–positive DM, 38% of whom had rapidly progressive ILD, found that CYC, MMF + a calcineurin inhibitor, or RTX were all efficacious treatments.18 The study also performed a literature review of 28 cases of adults with anti-MDA5–positive DM with rapidly progressive ILD treated with RTX and found that 15 patients remained stable or improved after RTX, while 13 patients died.13 Likewise, a retrospective study on RTX use for the treatment of refractory rapidly progressive ILD in 4 adults with anti-MDA5–positive amyopathic DM demonstrated improvement in respiratory symptoms in all 4 adults. This study noted improvement in HRCT for 3 patients, and significant improvements in lung function tests, recalcitrant vasculitic rashes, and average daily prednisolone dose overall.18
In our study, 1 patient with JDM, anti-MDA-5 antibodies, and ILD was excluded. This patient was a 2-year-old female who presented with atypical cutaneous findings, autoimmune hepatitis, and myositis. She was noted to have subcutaneous nodules with biopsy showing fat necrosis and microcalcifications. She was admitted to the intensive care unit shortly after the JDM diagnosis with acute respiratory failure. Treatments included methylprednisolone, IVIG, and 1 dose of RTX. Bronchial lavage fluid obtained prior to immunosuppression ultimately resulted in a concurrent diagnosis of pneumocystis jiroveci pneumonia. Her course was further complicated by pneumomediastinum, and the patient succumbed to respiratory failure only 1.5 months after symptom onset. An MSA panel resulted postmortem and was notable for positive anti-MDA5 antibodies. Lung pathology postmortem showed evidence of underlying ILD. Because of the limited follow-up data, this patient was not included in this study; however, her case demonstrates the need for prompt diagnosis and aggressive management of ILD.
Our patients demonstrated improvement not only in pulmonary measures of disease activity but also in measures of cutaneous and MSK disease activity, along with decreased steroid and concurrent medication usage over the course of RTX treatment. All patients had improvement in nailbed capillaries, with 3 patients achieving complete normalization. This is suggestive of improved disease activity, as studies have reported greater nailfold capillary changes in active JDM as compared to inactive.22-25 All patients had elevated LDH at diagnosis, but demonstrated a decrease in LDH over time, with 2 patients achieving normalization of LDH levels. CMAS scores also improved; 3 patients had initial scores suggestive of mild-moderate physical disability, but after 5 courses of RTX all patients had normalization of CMAS (score ≥ 50), suggesting no physical disability. The normalization of CMAS and muscle enzymes, along with limited physical disability is likely a combination of the milder nature of anti-MDA5–positive myositis, prolonged disease duration, and medical treatment. Children with anti-MDA5–positive JDM tend to have more mild muscle disease, and studies have reported that muscle enzyme levels are more likely to be normal with prolonged disease duration prior to treatment, particularly beyond 4 months.10,26
Our study supports previous studies in which clinical improvement is observed over the course of RTX, but adds valuable insight as to how anti-MDA5–positive JDM presents in a population of underrepresented children, given that prior studies have reported that Hispanic and non-Hispanic Black children make up a minority of children with JDM (representing 14% and 11% of children with JDM in the US, respectively).2 Our site has also previously described potentially worse clinical outcomes in Black or Hispanic patients with positive anti-MDA5 antibodies and evidence of ILD in 5 of 8 patients (62.5%), which is higher than previously reported in cohorts in North America and the UK.8,27 This highlights that the clinical improvement seen with RTX in our study population may be more impactful in a high-risk patient population.
Given the retrospective nature of this study and the small sample size of our study population, there are several limitations to be noted. First, available data were inconsistent among the 4 patients. The data used to identify the clinical picture surrounding a RTX infusion often varied in interval from RTX infusion, and intervals had variability across patients. Clinical and laboratory data were not always obtained at the same time, making the clinical picture surrounding a course of RTX a mixture of multiple points in time, rather than a single point in time. Additionally, there is no comparison group to accurately determine how RTX treatment affected disease course.
To our knowledge, this is the first study to specifically examine the use of RTX in an anti-MDA5–positive subset of children with JDM. In our small patient sample, patients who underwent treatment with 5 courses of RTX exhibited improvements in pulmonary, MSK, and cutaneous-related manifestations of JDM. Additionally, all patients had a reduction in number of concurrent medications and cessation of steroid use. Given the small sample size, retrospective nature of this study, and lack of a comparator group, true causality cannot be determined; however, this study provides encouragement that RTX has potential to improve outcomes in children with anti-MDA5–positive JDM. Further studies are warranted to better understand the role of RTX in management of JDM.
Footnotes
M. Peskin and M. Mostowy contributed equally to this work.
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication August 23, 2023.
- Copyright © 2024 by the Journal of Rheumatology
