Development of practice and consensus-based strategies including a treat-to-target approach for the management of moderate and severe juvenile dermatomyositis in Germany and Austria

Juvenile dermatomyositis (JDM) is the most common inflammatory myopathy of childhood and a major cause of morbidity and mortality among patients with pediatric rheumatic diseases [1]. The management of JDM is highly variable internationally [2, 3]. Few controlled trials have been performed on the treatment of JDM, e.g. on methotrexate (MTX), cyclosporin A (CSA) and rituximab (RTX) [4, 5]. The type of glucocorticoid regimen used is especially variable, with many centers using high-dose oral glucocorticoid therapy while others use intermittent intravenous methylprednisolone pulse therapy [6‐10]. Multiple smaller open label or retrospective studies have been performed, e.g. on cyclophosphamide (CYC), tumor necrosis factor inhibitors (TNFi), high-dose intravenous immune globulins (IVIG), or mycophenolate mofetil (MMF) [9, 11‐13]. Frequently, treatment is based on expert opinion [1, 6, 14]. Notably, only few formally approved treatment options exist for the treatment of dermatomyositis (DM), in Germany, including various glucocorticoids (methylprednisolone, prednisolone, prednisone, dexamethasone, triamcinolone), azathioprin [AZA], and, under certain conditions, IVIG) [15]. However, none of these medications have been specifically approved for the use in JDM. Recently, consensus treatment plans have been established in North America by the Childhood Arthritis & Rheumatology Research Alliance (CARRA) and consensus-based treatment recommendations were developed by the Single Hub and Access point for pediatric Rheumatology in Europe (SHARE) initiative [16, 17]. The role of biologic agents in the management of JDM is less well defined, even though they are frequently used as outlined by a recent survey among CARRA members [18]. However, there is evidence that treatment approaches for JDM in Germany vary in several aspects from those used in other countries, including the choice of glucocorticoid therapy regimens [19]. Lately, the concept of treat-to-target and tight control has been a cornerstone of current treatment recommendations for rheumatoid arthritis, systemic lupus erythematosus and gout, and has been discussed for pediatric rheumatic diseases as well [20‐24].

The PRO-KIND (PROjekte zur Klassifikation, Überwachung und Therapie in der KINDerrheumatologie; projects for the classification, monitoring and therapy in pediatric rheumatology) initiative is a sub-committee of the Society for Pediatric Rheumatology (GKJR) in Germany and Austria and aims to define consensus-based plans to harmonize diagnostic and treatment approaches. This initiative was started since it was perceived that children with juvenile rheumatic diseases in Germany and Austria often are either treated too late or not treated with the most up-to-date therapeutic regimens. To meet this challenge, the goal of the PRO-KIND initiative is to foster the use of harmonized diagnostic and treatment plans with defined targets [25, 26]. Previously, the JDM working group had identified current practice patterns in Germany and Austria among pediatric rheumatologists and neurologists [19].

As a next step, the goal of the PRO-KIND working group on JDM was to harmonize previously identified patterns into plans and statements.

Based on data on the current clinical practice in Germany we defined consensus-based statements and strategies supporting the diagnosis and the management of JDM. The expert panel consisted both of pediatric neurologist and rheumatologists with all of them having practical experience in managing JDM. The group supports the use of “adapted” Bohan and Peter criteria for the diagnosis of JDM, including the use of specific MRI findings as a Bohan and Peter “equivalent” criterion, similar to what others have proposed in the past [2]. There are a variety of diagnostic tests that are frequently employed in Germany to assist in the diagnosis and assessment of patients with JDM, and therefore are represented here. More recent developments, such as the discovery of myositis-specific antibodies (MSAs) that may be rather specific to JDM when compared to adult DM are also included, and it is felt that these antibodies may have diagnostic and prognostic value [36‐39]. We defined, to our knowledge for the first time, an explicit treat-to-target strategy for the management of JDM using the recently developed ACR/EULAR response criteria indicating absent, minimal, moderate or major improvement for formal assessment of treatment response [35]. Notably, others have shown that aggressive therapy of JDM using treatment targets may lead to improved outcome [8]. Similarly, the North American CARRA plans guide treatment decisions based on whether patients are unchanged, worsening, or experiencing medication side effects or disease complications [17]. The European SHARE recommendations focus on whether improvement is present or absent to guide treatment decisions but does not precisely delineate time frames for improvement [16]. The treat-to-target concept focusses on strategy rather than on specific treatment modalities which, for example, has been proven superior to conventional therapy in rheumatoid arthritis [40]. While the overall goal of therapy was defined as clinical inactive disease, we did not define this further during the consensus conference. However, we think that it would be reasonable to apply the validated set of criteria by either IMACS or PRINTO [41, 42].

Since there is a relative lack of data from randomized controlled clinical trials (RCTs), and since it is unlikely that large-scale RCTs for JDM will take place, one possible solution to gather more information on the efficacy or inefficacy of various treatments may be to collect “real life” data on disease outcome, i.e. to pursue comparative effectiveness research [43]. An important prerequisite for comparative effectiveness research is the application of uniform and harmonized treatments. Therefore, the harmonization of existing clinical practice has been an important goal for this project. Furthermore, the regular collection of a dataset in patients with JDM is of utmost importance to allow comparison of clinical outcomes, not only nationally but also internationally. Consequently, our statements include the minimal dataset agreed upon by international experts [29].

These harmonized consensus strategies overall fit rather well into the above-mentioned framework set up by the CARRA consensus treatment plans and the SHARE recommendations which had been developed previously or simultaneously, respectively. However, there are important differences in clinical practice in Germany when compared to this framework. For example, practitioners in Germany have a strong preference to choose a glucocorticoid regimen consisting of ongoing intermittent IVMP therapy in combination with low-to-moderate dose glucocorticoids instead of conventional daily high-dose glucocorticoid regimens [19]. There is surprisingly little data available on this important question. Proponents of intermittent IVMP pulse therapy suspect similar efficacy to high-dose daily glucocorticoids but less adverse effects, whereas other data suggest it may be less efficacious [7, 44, 45]. However, there are no definitive data that demonstrate superiority of one glucocorticoid regimen over another. Regarding treatment choices in case of severe or refractory JDM, there is a strong preference for IVIG, similar what is outlined in the CARRA treatment plans. Finally, the group agrees that patients with JDM should be managed in centers with an expertise in JDM. Furthermore, it is of course assumed that treatment decisions should always be shared between patients, families and providers. Since centers participating in the final consensus meeting manage more than half of all registered patients with JDM in Germany, it can be assumed that the consensus reached here is broadly generalizable in Germany. We have carefully worded the statements so that they are not confused with treatment guidelines or strict treatment recommendations.

There are several limitations to the PRO-KIND consensus strategies for JDM. First, the concepts outlined here may not represent the optimal way to manage JDM but rather represent current clinical practice among JDM experts in Germany and are based on a low level of evidence. Second, the statements and resulting strategies do not represent treatment protocols or “To Do”-lists. Management of patients with JDM will remain highly variable. Third, the field is constantly evolving: Some MSAs that were only recently reported in children, for example, anti-3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) antibodies, are not included in our list but may be useful to determine in some patients, especially in case of refractory myositis [37, 46‐48]. Fourth, some of the statements are rather vague, for example tapering of glucocorticoids, the choice of DMARD therapy beyond MTX, DMARD dosing regimens and the definitions of clinical inactive disease or disease flare. This is in part due to the difficulties achieving consensus on more precise statements which, in turn, results from the lack of convincing evidence for the preference of one option over another. We attempt to circumnavigate these issues by the implementation of a stringent treat-to-target strategy as outlined above. Fifth, many of the treatment options outlined in our strategies are not legally approved for the treatment of JDM, including the standard treatment with MTX. Therefore, treatment choices for individual patients may need to be adjusted based on local regulations for the off-label use of certain drugs. Sixth, some of the treatment options may not be widely available, again either due to regulation or due to cost.

In summary, we have developed consensus-based strategies for the diagnosis, monitoring and treatment of JDM by the harmonization of current clinical practice with the overarching goal is to improve the outcome of patients with JDM. Concerning therapy, we are placing an emphasis on a treat-to-target strategy using internationally accepted improvement criteria rather than on individual medications, fitting well into existing frameworks. As others have done previously, we define a minimal dataset to regularly collect in patients with JDM in order to allow comparative effectiveness research. We conclude that, due to the rarity of the condition, international collaboration will be critical in order to improve the outcome of patients with JDM. We believe that this work will add to the armamentarium of providers managing patients with JDM worldwide.