Rituximab in the treatment of dermatomyositis: An open-label pilot study

Authors


Abstract

Objective

To test the hypothesis that B cells play a role in the pathophysiology of dermatomyositis (DM) by examining the effect of B cell depletion in patients with symptomatic DM. Patients were treated with rituximab, a CD20+ B cell–depleting monoclonal antibody.

Methods

This was an open-label uncontrolled pilot trial in 7 adult patients with DM, 6 of whom had longstanding illness that was responding inadequately to a number of currently available immunosuppressive agents. All patients received 4 intravenous infusions of rituximab given at weekly intervals. Patients were followed up for up to 1 year without further treatment with rituximab. One patient was lost to followup. The principal efficacy outcome was muscle strength, measured by quantitative dynomometry.

Results

All 6 evaluable patients exhibited major clinical improvement, with muscle strength increasing over baseline by 36–113%. Maximal improvements in muscle strength occurred as early as 12 weeks after the initial infusion of rituximab. CD20+ B cells were effectively depleted in all patients by 12 weeks. Four patients experienced a return of symptoms that coincided with the return of B cells before the 52-week end point. Two patients maintained their increased muscle strength at 52 weeks, and 1 of these patients maintained this strength even after the return of B cells. Other symptoms of DM, including rash, alopecia, and reduced forced vital capacity, improved markedly in patients with these symptoms. Rituximab was well tolerated, with no treatment-related severe or serious adverse events during the observation period of this study.

Conclusion

This small open-label study of DM patients treated with rituximab provided sufficiently encouraging results to justify a more formal evaluation of the value of B cell depletion therapy in the treatment of DM.

Dermatomyositis (DM) is a chronic and debilitating inflammatory disease of skin and muscle that affects people of all ages. It has an annual reported incidence of 1 per 100,000 persons and can lead to disability and death (1–5). Patients usually present with progressive muscle weakness, which is often accompanied by an erythematous rash over the extensor surfaces of the joints and across the face. Articular, cardiac, pulmonary, and gastrointestinal manifestations occur in up to half of the patients (6–8), and the illness can be associated with severe morbidity. Consistent with the pathologic skin changes associated with other autoimmune connective tissue diseases, such as systemic lupus erythematosus (SLE), there are perivascular infiltrates of inflammatory cells in the skin, with an interface dermatitis consisting of lymphocytes causing ballooning degeneration at the dermoepidermal junction (9, 10). Diagnosis is confirmed by a muscle biopsy, which shows several pathognomonic changes, including perivascular inflammatory cells, immune complex deposition in the walls of intramuscular arteries and veins, and perifascicular muscle fiber atrophy (11).

Current evidence suggests that the idiopathic inflammatory myopathies, including DM, may be humorally mediated (12) because they are often associated with a panel of myositis-specific autoantibodies (MSAs) (6). In some patients these MSAs may be helpful in defining clinically homogeneous patient subsets and may predict the presence of associated rash or interstitial lung disease. The presence of autoantibodies lends support to the humoral basis of this illness (13). In addition, immunophenotyping has revealed that B cells are the most abundant inflammatory cells at the perivascular sites (11).

The traditional treatment approach to DM is immunosuppressive therapy, usually beginning with corticosteroids. However, up to 70% of patients treated with corticosteroids show an incomplete response, including 10% who are unresponsive (14). Corticosteroid treatment is also associated with important unwanted effects, such as weight gain, glucose intolerance, cataracts, osteoporosis, steroid-induced myopathy, and growth arrest in children. In patients who have an inadequate response to, or intolerable side effects with, these treatments, alternative approaches to treatment, including the use of azathioprine (15–18), methotrexate (17–20), cyclophosphamide (21), and/or cyclosporin A (20, 22), have been investigated in small-scale clinical trials, but with limited success. Human intravenous immunoglobulin (IVIG) may provide some benefit in refractory disease (22–31), although the effect is short-lived.

Consistently effective treatment regimens for DM remain elusive, and this illness therefore continues to pose a management challenge. In the search for alternative approaches, 2 lines of reasoning have led to the hypothesis that B cell depletion may be effective in the treatment of DM. First, in humorally mediated autoimmune peripheral neuropathies, the use of the CD20+ B cell–depleting monoclonal antibody rituximab directed against the CD20 antigen resulted in the depletion of circulating B cells with an accompanying significant improvement in muscle strength and without significant side effects (32, 33). Second, anecdotal experience in the use of rituximab in 2 patients, ages 11 and 26 years with refractory DM, seemed promising (34). Prior to treatment with rituximab, neither of these patients could walk unassisted, and both had a marked rash. Within 4 weeks of treatment with rituximab, each patient noted significant improvement in the rash and muscle strength, and by 3 months, both patients were walking without assistance.

This report presents the response data for the first 7 patients recruited into an open-label pilot study of rituximab in DM patients.

PATIENTS AND METHODS

Patients.

This was an open-label uncontrolled trial of rituximab (Rituxan; Genentech, South San Francisco, CA) in adult (≥18 years of age) patients with DM that was diagnosed based on 5 major criteria (35). Patients had failed to respond to at least 1 previous standard therapy for DM (corticosteroids, cyclosporin A, methotrexate, IVIG, or other immunosuppressive therapy) or had an average muscle strength of <75% of normal as measured by quantitative muscle dynomometry performed by an independent examiner. In addition, patients had to have normal hematologic parameters (absolute neutrophil count >1,000/mm3, platelets >100,000/mm3, and hemoglobin >7 gm/dl), adequate renal function (serum creatinine <2 mg/dl), and adequate liver function (total bilirubin <2 mg/dl, aspartate aminotransferase and/or alanine aminotransferase <2 times the upper limit of normal). The study received local Institutional Review Board approval, and all patients were required to provide written informed consent prior to their participation.

Patients were excluded if they had a known hypersensitivity to murine proteins or any component of rituximab, a preexisting known malignancy requiring treatment, other disorders producing severe weakness or sensory loss, were pregnant, or were breastfeeding. Patients were also excluded if they had another clinically significant medical condition(s) that may have had an impact on the course of the study, including active infection, human immunodeficiency virus (HIV) positivity, hepatitis B positivity, or previous or active Lyme disease.

Assessments.

After screening, a medical and medication history was taken, and a complete physical and neurologic examination, and an evaluation of muscle strength by manual testing and quantitative myometry (36) were conducted. An electromyogram necessarily demonstrating active myositis and a muscle biopsy consistent with a diagnosis of DM (although not performed on all patients as a baseline assessment if they had these assessments documented previously) were obtained. A complete blood cell count, electrolyte levels, creatine phosphokinase (CPK) levels, MSAs, quantitative immunoglobulins, CD3 and CD19 levels, HIV and hepatitis B status, and human antichimeric antibody (HACA) levels were recorded. (At the time of this report, HACA levels had not yet been assayed. Since HACA data are important in enabling re-treatment decisions, these data will be reported separately in the future.) Baseline forced vital capacity (FVC) was also measured and recorded.

Patients were assessed postbaseline at weeks 4, 12, 24, 36, and 52, and all patients were followed up for up to 52 weeks or until there were signs or symptoms of relapse. The end point of this study was a 1-year followup period without the need for further treatment or a relapse that required additional treatment. The primary efficacy outcome parameter in this study was the maximal muscle strength measured by quantitative dynomometry in 18 different muscle groups. This was then compared with standardized normal levels for age and sex to derive a percentage of normal strength (36). Treatment was considered to be effective if muscle strength was improved by >12% at 24 and 52 weeks. Laboratory, physical, and verbal safety assessments were obtained at each interval (or sooner, if clinically appropriate).

Administration of rituximab.

Eligible patients received 4 intravenous infusions of rituximab given at weekly intervals on days 1, 8, 15, and 22. Premedication with acetaminophen and diphenhydramine was given at each infusion to attenuate possible infusion-related reactions. The first 3 patients received rituximab at a dose of 100 mg/m2/infusion as a safety prerequisite in order to meet regulatory requirements. Only after this dose resulted in no unexpected reactions in these patients were subsequent patients treated with a dose of 375 mg/m2/infusion. In all cases, the first infusion of rituximab was administered intravenously at an initial rate of 50 mg/hour, and in the absence of hypersensitivity or infusion-related events, the infusion rate was increased in 50-mg/hour increments every 30 minutes to a maximum of 400 mg/hour. Subsequent infusions of rituximab were administered at an initial rate of 100 mg/hour and increased in 100-mg/hour increments at 30-minute intervals to a maximum of 400 mg/hour, as tolerated. Previous medication was continued as concomitant medication, provided that dosages had been stable for at least 2 months before trial enrollment, although concomitant medication could be tapered according to clinical need at the discretion of the investigating physician at any time during the trial period.

RESULTS

Patient population.

Data from 7 patients (5 women and 2 men, age range 21–64 years) are reported. Their baseline demographics, symptoms, and treatment history are shown in Table 1. Six patients had longstanding illness (4–15 years) and had received treatment with at least 3 or more drugs consisting of corticosteroids, azathioprine, methotrexate, cyclosporin A, cyclophosphamide, IVIG, and/or etanercept, before entry into the current trial (Table 1). One patient was newly diagnosed and had not received previous treatment with traditional therapy.

Table 1. Demographic characteristics, baseline symptoms, and treatment history in 7 dermatomyositis patients treated with rituximab*
PatientAge/sexDisease duration, yearsBaseline CPK, units/literSymptomsPrevious treatmentCurrent treatment
  • *

    Patients 1–3 received rituximab at a dose of 100 mg/m2/infusion. Patients 4–6 received rituximab at a dose of 375 mg/m2/infusion. CPK = creatinine phosphokinase; IVIG = intravenous immunoglobulin; ILD = interstitial lung disease; HCQ = hydroxychloroquine.

164/M0.31,200RashNoneNone
221/F7200RashCorticosteroids, azathioprineIVIG
348/F4910ILDCorticosteroids, cyclophosphamide, azathioprineCorticosteroids, cyclophosphamide, azathioprine
453/F155,600Rash, ILDCorticosteroids, methotrexate, cyclosporin A, etanercept,IVIG, azathioprine
538/F13334Rash, alopeciaMethotrexate, HCQ, IVIGCorticosteroids
653/F15128Rash, alopecia, calcinosisMethotrexate, etanerceptCorticosteroids, azathioprine
745/M1057NoneEtanercept, methotrexate, azathioprine, corticosteroidsNone

All patients were symptomatic, with rash and reduced muscle strength, and required additional and/or alternative treatment. Following treatment with rituximab, 2 patients were followed up for a minimum of 24 weeks, 2 patients for 36 weeks, and 2 patients for at least 52 weeks before signs of relapse emerged that necessitated re-treatment. One patient was lost to followup immediately after treatment with rituximab.

Muscle strength.

At baseline, patients exhibited reduced muscle strength, ranging from 39% to 60% of normal. All 6 evaluable patients experienced an increase in muscle strength following treatment with rituximab (Figure 1). Improvements were not distinguishable by rituximab dose in this small sample. At the time of maximal improvement, muscle strength ranged from 68% to 102% of normal strength, which represented a 36–113% increase over baseline (Figure 2). Thus, all patients exceeded the prospectively defined minimum criterion of effectiveness (>12% increase in muscle strength). Improvements were evident as early as 4 weeks after initial rituximab infusion and were maximal at 12 weeks in 2 patients and at 24 weeks in 2 patients. Thereafter, muscle strength began to wane in these 4 patients. In the other 2 patients, maximal improvement in muscle strength was attained at 36 weeks and continued beyond this time point. One of these patients was evaluated at week 52, whereas the other was not reevaluated until week 90. In both of these patients, maximal improvement in muscle strength noted at week 36 was maintained thereafter.

Figure 1.

Time course of changes in muscle strength in 6 patients following treatment with rituximab. Patients 1–3 received rituximab at a dose of 100 mg/m2/infusion. Patients 4–6 received rituximab at a dose of 375 mg/m2/infusion.

Figure 2.

Maximum improvement from baseline strength and time to maximum improvement in 6 patients treated with rituximab. Patients 1–3 received rituximab at a dose of 100 mg/m2/infusion. Patients 4–6 received rituximab at a dose of 375 mg/m2/infusion. Values above the bars are the percentages.

Other signs and markers of improvement.

As a marker of disease activity in DM, all patients showed a decrease in CPK levels following treatment with rituximab that was consistent with improved muscle strength. CPK levels fell from a baseline range of 128–5,600 units/liter (Table 1) to a minimum range of 57–1,168 units/liter.

In addition, all patients with rash before treatment with rituximab showed an improvement in this symptom. Furthermore, the hair of 2 patients with alopecia regrew, and 1 formerly emaciated patient gained weight. In 3 patients with impaired pulmonary function at baseline (including 2 who had interstitial lung disease), FVC improved by week 12 (Table 2). The other 3 evaluable patients had a normal FVC at baseline. Two of the 3 patients with impaired FVC were positive for Jo-1 autoantibodies at baseline; their antibody status did not change. Improvements associated with rituximab also facilitated a reduction in concomitant medications.

Table 2. Improvements in forced vital capacity (FVC) in dermatomyositis patients treated with rituximab*
PatientJo-1 statusFVC, % predictedMaximum improvement, %
Before rituximab12 weeks36 weeks52 weeks
  • *

    FVC was normal at baseline in 3 patients. Patient 3 received rituximab at a dose of 100 mg/m2/infusion. Patients 4 and 6 received rituximab at a dose of 375 mg/m2/infusion. Patients 3 and 4 presented with interstitial lung disease diagnosed at baseline. ND = not determined (or patient had withdrawn from the study by this time point).

3+576482ND44
4+4565NDND44
65472NDND33

B cell depletion.

Following rituximab infusions, all 6 patients exhibited a total depletion of B cells, as measured by flow cytometric analysis of CD19 levels. At 12 weeks, there was total depletion of B cells in 5 patients whose levels were measured at this time point. Postbaseline B cell measurements for the remaining patient started at week 24. By 24–36 weeks, 4 patients began to experience a decrease in muscle strength, which coincided with the return of circulating B cells (Figure 3). In the other 2 patients, improvement in muscle strength persisted up to the 1-year study end point and, interestingly, 1 of these patients remained in remission following the documented return of B cells. In the other patient who remained in remission, B cell status was unavailable between weeks 36 and 52.

Figure 3.

Improvement in muscle strength in relation to B cell levels in 5 evaluable patients treated with rituximab. Patient 1 (not shown) had a maximal increase in muscle strength over baseline of 73% at week 36 and a complete depletion of B cells (measured by CD19 levels) at week 12. CD19 levels between weeks 12 and 52 for this patient are not available. Muscle strength measurements at week 4 were not available for patient 4. CD19 levels at week 12 were not available for patient 6.

Safety.

Rituximab was well tolerated in this patient group. One patient who received 100 mg/m2/infusion of rituximab experienced some shortness of breath and hypertension associated with the third infusion, which improved when the infusion rate was slowed by half. Otherwise, all patients tolerated the infusions without difficulty or sequelae. Laboratory parameters never fell below normal values. One patient developed grade III cellulitis in which calcinosis broke through the skin. Although there were 2 similar episodes in this patient prior to this trial, a relationship to the study treatment could not be ruled out. The patient was treated with intravenous antibiotics, and the episode resolved within 10 days.

DISCUSSION

DM is a debilitating condition, which if untreated, is associated with serious consequences. Even with currently available treatment, such consequences are not always obviated. While corticosteroids are the most frequently used primary treatment for DM, most patients derive only limited benefit, necessitating the use of add-on treatments. However, the choice of such treatment remains essentially empirical because all are associated with some important limitations, particularly side effects. Any putative alternative treatment that is both effective and well tolerated in these patients can therefore be regarded as an advancement. The striking results of B cell depletion therapy on the symptoms of DM in this small series show promise in fulfilling these criteria, and although this is not a definitive trial, it is the first to document the potential utility of rituximab as an effective approach to the treatment of this condition.

Rituximab is a genetically engineered (chimeric murine/human) monoclonal antibody directed against the CD20 antigen. The CD20 surface antigen is widely expressed during B cell ontogeny but is not found on hemopoietic stem cells, pro–B cells, mature plasma cells, or other normal tissues (37–39). CD20 regulates early steps in activation and differentiation, and may function as a calcium ion channel (40, 41). Rituximab's property of depleting CD20+ B cell lines while sparing stem and plasma cells has been used successfully in the treatment of non-Hodgkin's lymphoma (NHL) (42) and other autoimmune disorders, such as rheumatoid arthritis (RA), SLE, and IgM-mediated neuropathies (32, 33, 43–45). In addition, the presence of myositis-specific antibodies points to B cell–mediated humoral abnormalities in DM (9–11).

This study of up to 1 year in duration has indicated that B cell depletion coincides with a decrease in the symptoms of DM, notably, improvements in muscle strength, rash, vital capacity, and enzyme markers. Furthermore, the data indicate a trend, in that 4 of the 6 patients experienced symptoms of relapse, beginning by 24–36 weeks, associated with the return of B cells. This adds support to the contention that B cells play a pivotal role in the pathophysiology of DM. In addition, the duration of B cell depletion and associated reduction in symptoms following a single course of treatment with rituximab in patients with DM is consistent with that seen in other conditions such as NHL and IgM-mediated neuropathies. It is interesting that 2 of the 6 patients in this study experienced a much more sustained response that lasted beyond the 52-week followup period, despite a documented complete return of B cells in 1 patient.

The good tolerability of rituximab in this series of patients with DM is consistent with the experience in RA and SLE (46, 47), in that there were no serious adverse events associated with its infusion. Such reactions appear to be less frequent in patients with autoimmune conditions compared with patients with NHL (45). In NHL patients, lysis of tumor burden may contribute the mostly mild-to-moderate transient infusion reactions seen in this group, the frequency of which is greater during the first infusion than in subsequent infusions (42, 48).

The results of this study, therefore, help to establish the critical role of B cells in the pathophysiology of DM and indicate that B cell depletion with rituximab appears to be effective in the treatment of this condition. These results suggest that this approach is well tolerated, convenient, safe, and without complications such as increased infection rates. However, the optimal therapeutic dose of rituximab in DM is currently unclear, and the schedule for re-treatment, based on the return of signs and symptoms of the illness, has not been evaluated in this pilot study. Therefore, further investigation to understand the processes involved and to ascertain the effect of re-treatment at relapse is warranted from larger controlled clinical trials.

In this small open-label study of patients with DM, rituximab consistently depleted B cell levels. This corresponded to an improvement in the symptoms of the illness. These results are sufficiently encouraging to prompt a more formal evaluation of the role of B cells in the pathophysiology of DM, and the value and place of rituximab in its treatment.

Acknowledgements

Editorial support for the manuscript was provided by Genentech.

Ancillary