Editorial: The efficacy of rituximab in refractory myositis: The jury is still out


Over the last decade, there has been a paradigm shift in the classification of the idiopathic inflammatory myopathies (IIMs). Bohan and Peter established diagnostic criteria for polymyositis (PM) and dermatomyositis (DM) in 1975 that are still widely used (1). However, those criteria have low specificity and therefore fail to distinguish IIMs from sporadic inclusion-body myositis (IBM) and noninflammatory myopathies, including limb muscular dystrophies with a similar distribution of weakness (e.g., dysferlinopathies) that may be associated with cellular infiltrates (2).

A different approach was taken in the 1980s. Classification was based on the results of elegant (immuno)histopathologic studies (3). It was noted that in both PM and sporadic IBM, non-necrotic muscle fibers are injured by autoinvasive CD8+ T cells that act in concert with CD4+ T lymphocytes, plasmacytoid dendritic cells, and macrophages, whereas in DM, infiltration of B lymphocytes, CD4 helper T cells, and macrophages can be found in perimysial areas of muscle fascicles and around small blood vessels (3).

The existence of PM as defined by Arahata and Engel was recently challenged (4, 5). In these studies, PM was found to be rare among the Dutch and French Canadian population and instead, another disease entity was recognized and labeled as nonspecific myositis (4) or overlap myositis (5). This myositis variant had the clinical characteristics of PM, but lacked the canonical histologic features described by Arahata and Engel. In contrast, nonspecific/overlap myositis had a histologic picture resembling DM. As in DM, patients with nonspecific/overlap myositis were frequently found to have an associated connective tissue disorder (CTD) or myositis-specific antibodies (4, 5). In addition, another IIM was recognized: necrotizing autoimmune myopathy (4, 6, 7), characterized by subacute or insidious onset, progressive symmetric proximal weakness, no skin abnormalities, and a grossly elevated serum creatine kinase value. Necrotizing autoimmune myopathy is distinguished from the other inflammatory myopathies by the absence of prominent inflammatory infiltrates and with macrophages rather than T cells being the effector cells (8). Necrotizing autoimmune myopathy has been found to be associated with CTD, cancer, and the use of statins. Because of the potential of necrotizing autoimmune myopathy to be amenable to treatment, it is important to distinguish it from other causes of muscle necrosis, such as rhabdomyolysis, muscular dystrophies, endocrinopathies, medications, and toxins.

There is a clear need for a new perspective on the treatment of the IIMs. The prognosis is not well known, since long-term outcome and prognostic factors vary widely. Favorable long-term outcome ranges from 18% to 90%. Predictors of poor outcome are the duration at disease onset, the presence of cancer (9, 10), and possibly also male sex (9), dysphagia, longstanding symptoms prior to diagnosis or start of therapy, subset of myositis, skin ulcers, delay in diagnosis or in start of therapy, various types of myositis, pulmonary (especially interstitial lung disease) and cardiac involvement, the presence of low total protein and albumin levels, and antisynthetase or anti–signal recognition particle autoantibodies (10).

A monocyclic disease course was seen in 15–48% of patients (9, 10). Over the long term, myositis has a chronic continuous or polycyclic disease course, with major effects on perceived disability and quality of life, despite regained muscle strength (9), although other investigators have reported that after a median of 7.5 years of followup, most patients (86%) had no disease activity, and 83% had no disability (10). It is of note that all these outcome studies were performed using the conventional classification criteria for PM and DM.

Despite the lack of a randomized controlled trial (RCT), high-dose steroids are considered first-line treatment in PM and DM. However, in a RCT comparing 2 regimens of steroids (daily oral high-dose prednisone versus 4 weekly cycles of high-dose oral dexamethasone) in adult patients with newly diagnosed myositis, a substantial proportion of the patients (∼55%) had to discontinue either type of steroid treatment early because of a lack of improvement and/or severe side effects (11).

In clinical practice, many immunosuppressants are added to the prednisone regimen, particularly if the patients do not respond adequately. However, a Cochrane review on treatment of DM and PM concluded that there was insufficient evidence from the available RCTs to confirm the value of immunosuppressants (other than prednisone) in myositis. This conclusion appears to contradict the experience of many clinicians (12). Novel therapies, such as biologic agents in the form of monoclonal antibodies or fusion proteins, are emerging. To date, these have not been investigated for use in patients with IIM in an adequate RCT, and results of case studies have been disappointing, showing only moderate improvement.

For example, in a 52-week pilot RCT of etanercept compared to placebo in patients with newly diagnosed DM and in those with refractory DM, no statistically significant differences between treatment groups were found with regard to muscle strength and motor functions, but there was a possible steroid-sparing effect in the etanercept-treated patients (13). Rituximab, a B cell–depleting agent, has been used in small series of patients with refractory myositis with reportedly favorable outcomes (14, 15). Encouraged by these results and based on the critical role of B cells in the initiation and propagation of the immune response in the pathogenesis of myositis, Oddis et al (16) initiated a multicenter, randomized, double-blind, placebo-phase trial in adult and pediatric myositis to assess the safety and efficacy of rituximab, the Rituximab in Myositis (RIM) study. Their findings are presented in this issue of Arthritis & Rheumatism.

Eligible patients included adults with a diagnosis of definite or probable refractory DM or PM and patients at least 5 years of age or older with definite or probable juvenile DM according to specific criteria. The definition of improvement chosen for this trial was based on the International Myositis Assessment and Clinical Studies Group (IMACS) preliminary validated top-ranked response criterion of a ≥20% improvement in 3 of any 6 core set measures with no more than 2 worsening by ≥25%. A definition of worsening was also specified.

Patients were randomly assigned to a rituximab early (active drug at weeks 0 and 1, with placebo at weeks 8 and 9) or rituximab late (placebo at weeks 0 and 1, with active drug at weeks 8 and 9) arm. An equal number of adult PM, adult DM, and juvenile DM patients received drug either at the beginning of the trial or 8 weeks later (placebo-phase duration agreed upon by consensus of the Steering Committee). Week 8 represented the end point of the placebo-controlled trial since the rituximab late group had not yet received study drug. The corticosteroid dosage was held constant until week 16; if patients met the definition of improvement (or experienced complications), a dosage reduction was begun at no more than 20% of the existing dose every 4 weeks. The primary end point was the time to achieving the definition of improvement, which was compared between the rituximab early and rituximab late groups. Secondary end points were the time to achieving a 20% improvement in manual muscle testing scores on 2 consecutive visits, as compared between the 2 groups, and the proportion of patients achieving the definition of improvement at week 8. Of 236 patients screened, 200 were randomized. Prior to screening, diagnostic accuracy was adjudicated in all PM patients (86 muscle biopsy samples reviewed; 44 subsequent exclusions, 14 for IBM, 29 for undetermined myopathy but not PM or DM, and 1 for excessive muscle damage).

Eighty-three percent of the patients met the definition of improvement by week 44, with no between-group difference in the time to achieving the definition of improvement. Results for the 2 secondary outcomes were also similar. There was a nonsignificant difference between the early and late rituximab arms in the juvenile DM group. A prednisone-sparing effect was not a defined outcome measure, but most of the patients were able to reduce their prednisone dosage irrespective of being in the early or late rituximab group. Rituximab was tolerated rather well. There were 67 serious adverse events in 64 patients, 26 of which were drug-related, and the majority were of infectious origin. No difference in adverse events was observed at week 8, the randomized placebo-controlled time point.

The authors are to be commended for performing a large clinical trial encompassing 200 patients with a rare disease such as myositis. The RIM Study is the first prospective, randomized double-blind trial in myositis to enroll both pediatric and adult patients and is the largest trial ever performed in patients with IIMs. It represents the first collaboration between adult and pediatric rheumatologists and neurologists for the study of myositis, showing that both specialist groups were able “to share the same planet” (17), and this holds promise for future collaborative trials. The investigators showed that improvement was measurable, albeit after 20 weeks, in both treatment groups in adult myositis patients, that adding rituximab led to a steroid-sparing effect, and that the drug was relatively safe.

Several reasons may explain why the RIM Study failed to achieve its primary efficacy end point. The investigators mention the following issues: the power calculation based on the postulated effect of rituximab by 8 weeks, the selection of a placebo phase of 8 weeks, and the core set of measures and the definition of improvement.

The Steering Committee had assumed an effect of rituximab at week 8 in more than half of the patients based on data reported in the literature, but this was seen at week 20 in the adult myositis group, similar to another reported observation in patients with adult myositis who were treated with rituximab (15). This led to underestimation of the anticipated placebo rate. The choice of an 8-week placebo phase was mainly determined based on ethical considerations. The core set of measures was partially validated and agreed upon by consensus, but has not recently been used in a prospective clinical trial such as the RIM Study.

Another important issue is to identify which patients with PM and DM would be the most likely to benefit from receiving treatment with rituximab or any other of the novel therapies. Selection of the patients should be based on the most recent classification criteria (18). The authors ruled out myositis patients who had an overlap with connective tissue disorders and those with concomitant cancer, thus excluding patients with nonspecific myositis and necrotizing autoimmune myopathy, both of which are known to be amenable to treatment. Since not all muscle biopsy samples were available for review, it is still possible that cases were misdiagnosed as “true PM” when in fact they were IBM. Whether the reviewers of the muscle biopsies adhered to strict histopathologic criteria for the diagnosis of IBM, requiring the presence of rimmed vacuoles, is not explicitly mentioned. It could well be that this led to the inclusion of patients who may not to be responsive to treatment. The juvenile DM group, in which diagnosis is rather straightforward, did show a trend toward a difference between both treatment arms, but the sample size was too small to draw solid conclusions.

Finally, muscle imaging and, in particular, magnetic resonance imaging may be helpful in further selection of suitable myositis patients. Muscle edema indicating active inflammation can be demonstrated by showing areas of high signal intensity on STIR and fat-suppressed T2-weighted sequences, even in clinically asymptomatic muscles, and on T1-weighted sequences, replacement of skeletal muscle by fat can be assessed (19). These analyses may help select the patient subgroups that should be included in future clinical trials.

In conclusion, Oddis et al have proved that large treatment trials are possible in this difficult disease. Future trials will benefit from the experience obtained in the RIM Study.


Dr. de Visser drafted the article, revised it critically for important intellectual content, and approved the final version to be published.