Pilot Study of Etanercept in Patients With Refractory Juvenile Dermatomyositis

Authors


Abstract

Objective

To evaluate the efficacy of etanercept in patients with juvenile dermatomyositis (DM) refractory to standard treatment.

Methods

Nine patients with juvenile DM prospectively received etanercept 0.4 mg/kg subcutaneous twice weekly concurrently with baseline medications for 12 weeks. Patients were reevaluated 12 weeks (week 24) after stopping etanercept. Outcome measures included a validated Disease Activity Score (DAS), serum muscle enzymes, Childhood Myositis Assessment Scale (CMAS), and nailfold capillaroscopy (NFC).

Results

Six patients completed all visits; 2 patients completed through week 12 and 1 patient stopped after the fifth etanercept dose due to marked worsening of a rash. At week 12, 7 patients had a mild decrease in DAS and 1 patient noted worsening of the DAS. At week 24, 1 patient remained stable, 2 patients had worsening of the DAS, and 3 patients had improvement of the DAS (1 patient with inactive disease), including the patient who worsened while receiving etanercept. This patient and the patient who stopped (worsening rash) both had the tumor necrosis factor α (TNFα) 308A allele. There was a trend of worsening NFC at week 12, while at week 24 improvement of NFC was noted. There was no appreciable change in serum muscle enzymes or CMAS throughout the study.

Conclusion

In this trial of patients with refractory juvenile DM, etanercept did not demonstrate appreciable improvement and some patients noted worsening of disease. Caution should be taken when recommending TNF receptor inhibitors to patients with active symptoms of juvenile DM, and close followup is warranted. Further investigation of the interaction of the TNFα-308A polymorphism and type I interferon is needed to define the mechanism of TNF blockade in juvenile DM.

Introduction

Juvenile dermatomyositis (DM) is the most common inflammatory myopathy of childhood. Although the etiology is unknown, proinflammatory cytokines are implicated in the pathogenesis of inflammatory myopathies ([1]). Specifically in juvenile DM, tumor necrosis factor α (TNFα) was expressed in the muscle fibers of children with untreated disease ([2]). The finding that the TNFα-308A polymorphism was associated with dystrophic calcifications ([3]) was the impetus for this study.

The medications given to children with juvenile DM are primarily immunosuppressive. Despite a wide variation in the treatment of juvenile DM, corticosteroids remain the cornerstone of therapy ([4]). Additional treatments include methotrexate and intravenous gamma globulin. However, there are multiple toxicities related to corticosteroids, and not all patients improve on these medications. There is a need to identify therapies for patients refractory to traditional treatments. Reports have suggested improvement in inflammatory myopathies with the introduction of etanercept ([5]) and infliximab ([5, 6]) in patients who were unresponsive to traditional therapies. A study of 5 patients with resistant juvenile DM reported improvement in all children with the initiation of infliximab ([7]). The Muscle Study Group conducted a pilot trial of etanercept in 16 adult patients with DM (11 received etanercept and 5 received placebo) and found that nearly half of the etanercept group was tapered off of prednisone, whereas all subjects in the placebo arm failed treatment ([8]).

The objective of our study was to evaluate the efficacy of etanercept in patients with juvenile DM refractory to standard medications. Additionally, we sought to identify associations with TNFα-308 polymorphisms and response to therapy.

Box 1. Significance & Innovations

  • Clinical trials of therapeutic agents in juvenile dermatomyositis (DM) are lacking.
  • Tumor necrosis factor α (TNFα) has been implicated in the pathogenesis of juvenile DM; therefore, blockage of TNF has been considered as a therapeutic approach.
  • Minimal improvement in disease activity was noted in this trial of etanercept in juvenile DM, with some patients demonstrating worsening disease.
  • Anti-TNF agents for the treatment of juvenile DM should be used with caution.

Patients and methods

Patients

Nine patients diagnosed as having juvenile DM by a single pediatric rheumatologist (LMP) were included. Age-appropriate consent/assent was obtained from all participants. The institutional review board approved this study. Patient clinical characteristics are described in Table 1. All of the following criteria were met for inclusion: probable or definite diagnosis of juvenile DM according to the Bohan and Peter criteria ([9, 10]); disease onset prior to age 16 years, with the age range at the time of the study from 6–27 years; must have had the disease for 12 months and failed to achieve adequate disease control, as defined by steroids for ≥4 months (10–30 mg/kg/dose, maximum 1,000 mg) of intravenous (IV) methylprednisolone given at the start of treatment until creatinine kinase levels were ≤1,000, and thereafter up to 3 times each week, with up to 0.5 mg/kg/dose oral prednisone given on days when IV methylprednisolone is not administered (or 1–2 mg/kg/dose oral prednisone); and ≥1 other immunosuppressive agent (methotrexate 0.3–1 mg/kg/dose given each week by injection [subcutaneous or IV], cyclophosphamide 500–1,000 mg/m2 IV given every 3–4 weeks, azathioprine 0.5–1 mg/kg/dose given by mouth daily, or cyclosporine 2–5 mg/kg/day given by mouth twice daily). Subjects were excluded if an immunosuppressive agent had been initiated within 30 days prior to the beginning of the screening period, the leukocyte count was <3,000/mm3, the platelet count was <50,000/mm3, serum creatinine was above the upper limit of normal, they had a history of poor adherence to prior therapies, or they had active infection, including tuberculosis.

Table 1. Patient characteristics*
PatientAge, yearsSexRaceDisease duration, yearsPrior treatmentConcurrent treatmentTNF allele
  1. TNF = tumor necrosis factor; IV = intravenous.
112FWhite7.3
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • Cyclosporine
  • Methotrexate
  • Hydroxychloroquine
GG
228MAsian24.6
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
MethotrexateGG
313FWhite4.1
  • IV methylprednisolone
  • Oral prednisone
  • Hydroxychloroquine
  • IV cyclophosphamide
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
GG
47MWhite4.8
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • IV gamma globulin
AA
514FAfrican American10.8
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • IV gamma globulin
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
GG
618FWhite3.5
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
GG
714FWhite3.3
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
GG
817FWhite7.2
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
AG
916FWhite7.5
  • IV methylprednisolone
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
  • Oral prednisone
  • Methotrexate
  • Hydroxychloroquine
AA

Study design

This open-label trial occurred over 6 months, with patient evaluations at baseline and weeks 4, 12, and 24. The study was conducted from February 2000 to January 2002. Etanercept was administered at 0.4 mg/kg/dose (maximum of 25 mg) subcutaneous twice weekly for the initial 12 weeks of the study (total of 24 doses). During the study time period, this was standard pediatric dosing. Etanercept was stopped during weeks 13 through 24, with the final study visit occurring at week 24. Patients' families were called weekly during weeks 1 through 12 and reminded to keep a diary of drug administration and adverse events. Other medications were maintained at constant doses throughout the study. The small number of patients in this study precluded formal statistical analysis; however, descriptive analysis was utilized.

Laboratory data

Laboratory assessment of the patients at each study visit included complete blood cell count and differential, muscle enzymes (creatinine kinase, lactate dehydrogenase, aspartate transaminase, and aldolase), erythrocyte sedimentation rate, von Willebrand factor antigen (vWF:Ag), neopterin, and absolute natural killer cell count by flow cytometry. Additionally, at the baseline visit, TNFα-308 allele typing was obtained, as previously described.

Outcome measures

A validated juvenile DM Disease Activity Score (DAS), which provides subscores for skin and muscle involvement, was the primary outcome measure ([11]). Secondary outcome measures included the Childhood Myositis Assessment Scale (CMAS) and nailfold capillaroscopy (NFC). NFC studies were obtained by freeze-frame video microscopy of each of 8 fingers (thumbs excluded). Each image was analyzed for the total number of end row loops (ERLs) per mm over the 8 digits, with the mean ERL (total ERLs divided by 8) calculated for each patient.

Statistical analysis

Significance of change in the DAS at each visit was tested using Wilcoxon's rank sum test. Spearman's correlation analysis was used to evaluate the correlations for change in the DAS with NFC, CMAS, muscle enzymes, and vWF:Ag.

Results

Six patients completed all study medication and visits; 2 patients completed the first 4 and 12 weeks of etanercept but did not return for the 24-week visit, and 1 patient stopped the study after the fifth dose of etanercept due to marked worsening of a rash that was diffuse (shawl sign, extremities), increased over joints, and had an intense malar flush. At week 4, 5 patients had a mild improvement in the total DAS. At week 12, 7 patients had a mild decline in the total DAS ranging from 0.5–2 points, which included similar improvements in skin and muscle subscores. The eighth patient had worsening of the DAS by 3 points because of muscle disease. Interestingly, this latter patient and the patient who stopped the study due to a worsening rash each were positive for the TNFα-308A polymorphism. At week 24 (i.e., off etanercept), the patient who worsened on etanercept noted improvement in the DAS by 4 points, which included improvement in both muscle and skin subscores. Two patients had worsening of the DAS by 2–4 points. Two patients had improvement of the DAS following discontinuation of etanercept by 1.5–4 points (Figure 1). However, no statistically significant decrease in the DAS was observed in any of the 3 followup visits (P = 0.33 for week 4, P = 0.09 for week 12, and P = 0.21 for week 24 by Wilcoxon's test), possibly due to the small sample in the current study.

Figure 1.

Change in Disease Activity Score (DAS) before etanercept (baseline), following exposure to etanercept (weeks 4 and 12), and following the washout phase (week 24).

NFC was obtained at each study visit in 5 patients. Four patients noted decline (i.e., worsening) of mean ERL at week 12, while at week 24 these patients had improvement in NFC. There were no appreciable trends noted in muscle enzymes or the CMAS at week 12 versus week 24. There was no significant correlation between the change of DAS and the change of NFC, CMAS, or muscle enzymes in any of the followup visits. However, at week 12, all of the patients had concordant changes in both the DAS and vWF:Ag (decrease in 5 patients and increase in 1 patient; the other 2 patients had no vWF:Ag measures).

Discussion

This is the first reported prospective study of anti-TNF therapy in patients with juvenile DM. Although the majority of patients demonstrated mild improvement in disease activity, 2 patients noted marked worsening while taking etanercept. Most interesting is that both patients had polymorphisms in the TNFα-308 position, in which there was a G to A substitution. In 2000, an open-label study of infliximab reported that 7% of patients developed antibodies to double-stranded DNA after exposure to the treatment ([12]). Subsequently, caution has been advised in the administration of anti-TNF medications to patients with lupus, a disease known to be associated with elevated type I interferon (IFN). Type I IFN is classically involved in viral defense and associated with multiple proinflammatory mechanisms. The interplay of TNF and IFN remains under investigation. At the time of this study, the possible role of type I IFN in the pathogenesis of juvenile DM was being elucidated; therefore, IFN was not measured in our patients. A plausible mechanism for worsening symptoms in our study is that the TNF-308A polymorphism was associated with increased TNFα production, which, when inhibited by etanercept, led to perturbations in type I IFNs and more active disease. This conjecture is supported by studies that show that IFNα can potentiate TNF-induced apoptosis ([13]), but the role and timing of administration of TNF inhibitors in this dynamic interaction has not yet been characterized in children with juvenile DM.

Treatment of 2 adult patients ([6]) and 5 children ([7]) with inflammatory myopathy with infliximab resulted in improvement. Seven of the patients in this study noted mild improvements in both skin and muscle findings. One of the limitations of this study is that the treatment phase lasted 3 months, and it may take up to 5 months to attain an optimum response from anti-TNF therapy. However, during the washout phase, 2 of the patients continued to demonstrate improvement in disease activity, while 1 patient remained stable and 2 patients had an increase in the DAS. It is unclear from our limited number of patients what this variation in washout response suggests about anti-TNF therapy in patients with juvenile DM.

Similar to the 2 patients who deteriorated in this trial, a series of 5 adult patients with DM receiving etanercept 25 mg subcutaneously twice weekly had disease exacerbation, with continued rash and worsening muscle involvement ([14]). Although the Muscle Study Group demonstrated disease improvement in nearly half of the patients receiving etanercept for adult DM, 5 patients developed a worsening rash while receiving this treatment ([8]). An open-label study of infliximab in 13 adults with refractory idiopathic inflammatory myopathy demonstrated poor improvement ([15]). In fact, of the 9 patients who completed this open-label study, at followup, 5 patients demonstrated muscle inflammation by magnetic resonance imaging. Type I IFN was analyzed in the sera of 10 patients; compared to baseline, there was a significant increase following infliximab. A limitation of this study is a lack of IFN measurement, since these data would have been most interesting in the patients whose condition worsened. A recent case report described a 7-year-old girl initially diagnosed with psoriasis who subsequently developed proximal muscle weakness, orbital edema, and elevated muscle enzymes consistent with juvenile DM ([16]). Despite treatment with etanercept, her skin and muscle symptoms worsened. Improvement was noted with discontinuation of etanercept and administration of IV/oral corticosteroids and subcutaneous methotrexate. Type I IFN pathways also have been implicated in the pathogenesis of psoriasis.

Overall, therapy with a biologic agent has tremendously enhanced the treatment of rheumatic disease, but studies in pediatric patients are limited. To date, only 1 clinical trial involving patients with juvenile DM has been reported ([17]). In this randomized, placebo-phase trial of rituximab in the treatment of refractory inflammatory myopathy, 48 patients with juvenile DM were included. Although the primary end point was not achieved, 83% of study participants, including patients with juvenile myositis, met the definition of improvement after a median of 20.2 weeks in those receiving rituximab late and 20.0 weeks in those receiving rituximab early. The authors concluded that further studies involving B cell–depleting therapy are needed with reconsideration of trial design. This highlights a limitation of our study because patients received etanercept therapy for only 12 weeks, which may have been insufficient to detect either improvement or worsening of disease activity. A recent report of abatacept and sodium thiosulfate administered to a 14-year-old female with resistant juvenile DM involving calcinosis and ulceration resulted in marked disease improvement ([18]). Future clinical trials of abatacept, rituximab, and other therapies with a biologic agent in children with juvenile DM are clearly warranted.

In conclusion, our study suggests that anti-TNFα therapies be used with caution in children with active type I IFN–driven symptoms of juvenile DM in whom traditional therapies have failed to achieve disease control. Data concerning efficacy for active juvenile DM symptoms, rash, weakness, and arthritis have yet to be acquired. Further understanding of the pathophysiology, role of cytokines in inflammatory myopathies, and pharmacogenomics should guide future clinical trials in patients with juvenile DM.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Pachman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Pachman.

Acquisition of data. Pachman.

Analysis and interpretation of data. Rouster-Stevens, Ferguson, Morgan, Huang, Pachman.

Ancillary