Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
Contents lists available at ScienceDirect
Best Practice & Research Clinical Rheumatology journal homepage: www.elsevierhealth.com/berh
8
Therapeutic advancements in juvenile idiopathic arthritis Elizabeth A. Kessler, MD, MS 1, Mara L. Becker, MD, MSCE * Division of Rheumatology, Department of Pediatrics, Children’s Mercy Hospitals and Clinics, 2401 Gillham Road, Kansas City, MO 64108, USA
a b s t r a c t Keywords: Juvenile idiopathic arthritis DMARD TNF-a inhibitors Biologic Treatment strategy
The treatment of juvenile idiopathic arthritis (JIA) has substantially evolved over the past two decades. Research has been conducted and is ongoing on how therapies can best be utilized either as monotherapy or in combination for enhanced efficacy. The introduction of biologic therapies that selectively target specific cytokines has changed the acceptable clinical course of childhood arthritis. In addition to the development and utilization of new therapeutic agents, the pediatric rheumatology community has made vital progress toward defining disease activity, developing validated outcome measures, and establishing collaborative networks to assess both clinical outcomes and the long-term side effects related to therapeutics for juvenile arthritis. In this chapter, we will discuss the therapeutic evolution in JIA over the past two decades. Although the largest strides have been made with biologic agents, and these newer drugs have more rigorous data to support their use, select commonly used non-biologic therapies are included, with the discussion focused on more recent updated literature. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction Juvenile idiopathic arthritis (JIA) is the most common pediatric autoimmune musculoskeletal condition, estimated to affect one in 1000 children [1]. It is classified based on age of onset, number and type of joints involved, and the presence of serologic markers, systemic signs, and symptoms. As the field has * Corresponding author. Tel.: þ1 816 234 3686; fax: þ1 816 234 3082. E-mail addresses: [email protected] (E.A. Kessler), [email protected] (M.L. Becker). 1 Tel.: þ1 816 234 3686; fax: þ1 816 234 3082.
http://dx.doi.org/10.1016/j.berh.2014.03.005 1521-6942/Ó 2014 Elsevier Ltd. All rights reserved.
294
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
developed, significant work has been invested in the development of an agreed-upon classification system for juvenile arthritis worldwide. The current JIA subtypes are characterized as follows: systemic, oligoarticular, polyarticular, enthesitis-related, psoriatic, and undifferentiated arthritis [2] (Table 1). Despite this relatively rare condition with a heterogeneous disease phenotype and poorly understood pathophysiology, the treatment strategies for JIA have evolved tremendously over the past two decades. Early treatment options for JIA were limited to nonsteroidal anti-inflammatory drugs (NSAIDs) including aspirin, corticosteroids, and non-biologic disease-modifying antirheumatic drugs (DMARDs) like D-penicillamine, gold, sulfasalazine (SSZ), and methotrexate (MTX). Historically, few treatment options were available for children with disease refractory to these medications, and the risk-to-benefit ratio for long-term corticosteroid use and many of the early DMARDs was not favorable, resulting in
Table 1 Definition of subtypes of juvenile idiopathic arthritis.a Category
Characteristics
Systemic
Arthritis in 1 joint(s) with or preceded by quotidian fever of 2 weeks’ duration documented for 3 days and accompanied by 1 of the following: 1. Evanescent erythematous rash 2. Generalized lymphadenopathy 3. Hepatomegaly and/or splenomegaly 4. Serositis
Oligoarticular
Arthritis in 1–4 joints during the first 6 months of disease A. Persistent – affecting 4 joints throughout disease course B. Extended – affecting >4 joints after the first 6 months of disease
Exclusions: a, b, c, d.
Polyarticular, RF negative Polyarticular, RF positive
Psoriatic
Enthesitis-related
Undifferentiated Exclusions
Exclusions: a, b, c, d, e. Arthritis in 5 joints during first 6 months of disease, negative RF Exclusions: a, b, c, d, e. Arthritis in 5 joints during first 6 months of disease, positive RF on 2 separate occasions at least 3 months apart Exclusions: a, b, c, e. Arthritis and psoriasis, or arthritis and 2 of the following: 1. Dactylitis 2. Nail pitting or onycholysis 3. Psoriasis in a first-degree relative Exclusions: b, c, d, e. Arthritis and enthesitis, or arthritis or enthesitis with 2 of the following: 1. Presence of or history of sacroiliac joint tenderness and/or inflammatory lumbosacral pain 2. Presence of HLA-B27 antigen 3. Onset of arthritis in male >6 years old 4. Acute anterior uveitis 5. AS, ERA, sacroiliitis with IBD, Reiter’s syndrome, or acute anterior uveitis in a first-degree relative Exclusions: a, d, e. Arthritis that fulfills criteria for no category or 2 categories a. Psoriasis or history of psoriasis in the patient or first-degree relative b. Onset of arthritis in an HLA-B27-positive male >6 years old c. AS, ERA, sacroiliitis with IBD, Reiter’s syndrome, or acute anterior uveitis in a first-degree relative d. Positive RF on two separate occasions at least 3 months apart e. SJIA in the patient
RF, rheumatoid factor; HLA-B27, Human Leukocyte Antigen B27; AS, ankylosing spondylitis; ERA, enthesitis-related arthritis; IBD, inflammatory bowel disease; SJIA, systemic JIA. a Modified from the International League of Associations for Rheumatology [2].
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
295
children with long-term morbidities from both their inadequately treated arthritis and also the therapies that were used as treatment. Major advances in the field, including the establishment of validated outcome measures, the organization of collaborative research networks, and the exploration of translational approaches to identify biomarkers of response and toxicity have been important contributors to therapeutic development in JIA. Outcome measures in JIA (see chapter 10) [118] In order to effectively compare treatment options and ultimately choose the best therapy in daily clinical practice, it is imperative that we have standardized and validated outcome measures. Definitions of disease flare, minimal disease activity, inactive disease, and clinical remission have been developed and validated, and are now used regularly in clinical trials and translational research studies. Tools for assessment of quality of life, an important component of health care for children with chronic disease like JIA, have also been developed (Table 2). In the past 10–20 years, there have been significant advancements in the creation of validated clinical outcome measures in JIA. The American College of Rheumatology Pediatric (ACR Pedi) response
Table 2 (see chapter 10) [118] Definitions of commonly used outcome measurements in pediatric rheumatology. Pediatric response criteria ACR Pedi 30
Minimum of 30% improvement from baseline in at least three of the six core set criteria, with a worsening of 30% in no more than one variable
ACR Pedi 50, 70, 90, and 100
Minimum improvement of 50%, 70%, 90%, and 100%, respectively, from baseline in at least three of the six core set criteria with a worsening of 30% in no more than one variable ACR pediatric core set criteria 1. Physician global assessment of disease activity (10 cm visual analog scale) 2. Parent/patient assessment of disease activity (10 cm visual analog scale) 3. Functional ability (CHAQ) 4. Number of joints with active arthritis 5. Number of joints with limited range of motion 6. ESR JADAS Composite score based on the arithmetic sum of the four following measures: 1. Physician global assessment of disease activity (VAS) 2. Parent/patient assessment of disease activity (VAS) 3. Number of joints with active arthritis based on 27-joint count 4. ESR Definitions of disease activity [113,114] Clinical inactive disease 1. No joints with active arthritis (all six conditions must be met) 2. No fever, rash, serositis, splenomegaly, or generalized lymphadenopathy related to JIA 3. No active uveitis 4. ESR or CRP level (or both if tested) within normal limits or, if elevated, not related to JIA 5. Physician global assessment of disease activity of lowest possible on the scale used 6. Duration of morning stiffness 15 min Clinical remission on medication Inactive disease on medication must persist for 6 months Clinical remission off medication Inactive disease for 12 months after discontinuation of all JIA therapy Quality-of-life measurement tools CHAQ [115] A disease-specific instrument that measures functional ability in activities of daily living PedsQL [116] Measures health-related quality of life via both child self-report and parent proxy report scales using both generic and rheumatology-specific measures CHQ [117] Generic instrument that contains 14 domains to evaluate the physical, emotional, and social components of health and provides physical and psychosocial summary scores ACR, American College of Rheumatology; CHAQ, Childhood Health Assessment Questionnaire; VAS, visual analog scale; ESR, erythrocyte sedimentation rate; sJIA, systemic juvenile idiopathic arthritis; JADAS, Juvenile Arthritis Disease Activity Score; CRP, C-reactive protein; JIA, juvenile idiopathic arthritis; PedsQL, Pediatric Quality of Life; CHQ, Child Health Questionnaire.
296
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
criteria are a composite score based on improvement in six core outcome variables [3] (Table 2). Limitations of the ACR Pedi response criteria include the inability to directly compare absolute response between patients or measure change in absolute disease activity. Additionally, betweenstudy comparisons cannot easily be made [4]. As a result, the Juvenile Arthritis Disease Activity Score (JADAS) was developed, which provides continuous measurement of response, modeled after the Disease Activity Score in adult patients with rheumatoid arthritis (RA) (Table 2) [5]. Limitations of the JADAS include lack of meaningful cutoff points defining disease activity and a potential lack of generalizability to patients with less severe oligoarticular disease [5,6]. Collaborative networks The formation of the Pediatric Rheumatology Collaborative Study Group (PRCSG) was a milestone in the advancement of therapeutics for pediatric rheumatology. This group developed a standard methodology for the design, conduct, and analysis of drug trials in children with rheumatic diseases, which has played a major role in the investigation of new biologics such as etanercept, adalimumab, and abatacept in children [7]. The PRCSG has also spearheaded collaborative working agreements with groups such as the Pediatric Rheumatology International Trials Organization (PRINTO) to combine efforts across the globe to advance research and therapy for children with rheumatic disease, illustrated by the recent phase III study of canakinumab in JIA [8]. The Childhood Arthritis and Rheumatology Research Alliance (CARRA) is a North American organization of pediatric rheumatologists committed to investigator-initiated research, with important contributions including, but not limited to, the establishment of the largest multicenter registry of children with rheumatic diseases, and the CARRA consensus treatment plans (CTPs), which were developed to identify effective therapies in clinical settings through comparative effectiveness research. CARRA and PRINTO have also become part of vital pharmacovigilance projects to evaluate the longterm safety of the therapeutic agents currently used in JIA. CARRA has collaborated with the US Food and Drug Administration (FDA), biopharmaceutical industry representatives, and the Duke Clinical Research Institute to form a pharmacosurveillance model (CARRA Consolidated Safety Registry (CoRe)) that utilizes the longitudinal multicenter registry, which currently has data on approximately 7000 children with JIA [9]. Pharmachild is a pharmacovigilance program involving participating centers of PRINTO, the Pediatric Rheumatology European Society (PReS), and pharmaceutical companies consisting of twice-yearly registration of clinical data and standardized adverse events [10]. International expert opinion and collaboration were used to establish the 2011 ACR-published recommendations for the treatment of JIA [11]. This was the first time the ACR has made a concerted effort to establish treatment recommendations for a pediatric rheumatic disease. These recommendations take into account patient prognostic factors, as well as disease activity levels, and are based on expert opinion and available published studies. An update of these recommendations was completed in 2013 specifically focusing on the treatment of systemic JIA (SJIA) and macrophage activation syndrome (MAS), recognizing the importance of this severe subtype and the rapid advancement in therapeutics targeting key cytokines in this condition [12]. These recommendations represent important progress toward having evidence-based guidelines for the treatment of JIA by providing guidance in both initial and subsequent therapies for JIA, as well as recommendations for medication safety monitoring. Utilization of translational approaches Translational approaches to investigate or predict treatment response in JIA have provided the potential for more precise investigative tools to use in heterogeneous conditions such as JIA, where the disease course is difficult to predict, and it is challenging to know when it is appropriate to escalate or discontinue therapy. Cellular, genetic, and even radiologic biomarkers have been explored as more sensitive measures to characterize drug response and identify risk of future flare. Pharmacogenomics may identify genetic variation associated with drug response, and, like in adults, has been explored in pediatric rheumatic diseases. These modalities are more frequently being used to enhance our current investigations, and they have the potential for use in mainstream clinical practice.
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
297
In this chapter, we will discuss the evolution of therapeutics for JIA over the past two decades. Studies conducted prior to the newest International League of Associations for Rheumatology (ILAR) classification criteria for JIA utilized a variety of terms to classify childhood arthritis; thus, we will discuss the results of these studies utilizing the terminology used at the time the study was conducted. As there are subtle differences in these criteria, we feel it is more accurate to interpret study data with a clear understanding of the subgroups of juvenile arthritis that were included. Although the largest strides have been made in biologic agents in the past two decades, select commonly used non-biologic therapies are included, with the discussion focused on more recent, updated literature, and any new developments that may aid in our therapeutic management. Non-biologic DMARDs Methotrexate MTX is a folic acid analog and a potent competitive inhibitor of several enzymes in the folate pathway resulting in interruption of purine and pyrimidine synthesis, and an accumulation of adenosine, which is thought to be a large contributor to the site-specific anti-inflammatory effects of MTX [13]. MTX is the cornerstone of therapy for childhood arthritis and remains the most commonly used non-biologic DMARD in the treatment of JIA. Its efficacy and safety in juvenile arthritis has been reported in several prospective uncontrolled studies in children and was clearly demonstrated over two decades ago in a double-blind randomized controlled trial (RCT) that required multinational collaborative efforts [14]. Despite a relatively long track record in children with rheumatic disease, there remains unpredictable variability in response to this drug, and extensive efforts in the past two decades have focused on identifying its optimal dose and route as well as cellular, serum, and genetic biomarkers to aid in individualizing therapy in this population of patients. Extensive variability in dosing preferences in JIA has prompted work to identify the optimal dose and route of MTX. Recent work has shown no significant therapeutic advantage and higher risk of transaminase elevation with higher initial MTX doses [15]. For nonresponders to conventional MTX dosing, there are reports supporting the safety and efficacy of escalating to higher dosing regimens (25–30 mg/m2/week) [16]. However, a multicenter prospective study in JIA patients who were nonresponders to initial low-dose MTX suggests otherwise. [17] Nonresponders were randomized to intermediate-dose (15 mg/m2/week) or high-dose (30 mg/m2/week) MTX; both groups achieved similar ACR Pedi outcomes. The ACR Pedi 30 response rate was 62.5% in the intermediate-dosed group and 57.5% in the high-dosed group. Similar outcomes in intermediate-dosed and high-dosed groups were observed for the ACR Pedi 50 (57.5% and 55%) and ACR Pedi 70 (45% and 47.5%) response, respectively. The route of administration of MTX also continues to be debated. There is altered oral bioavailability of the drug with higher doses, and early studies suggested an increased effectiveness of subcutaneous MTX administration in patients who did not respond or tolerate initial oral dosing [18]; however, oral administration of the drug is still prevalent and has been shown to be effective in some studies. In a recently published large observational German MTX Registry, over half the patients (63%) received oral MTX exclusively, and this group had similar rates of ACR Pedi 30, 50, and 70 response as well as toxicity to subjects treated via the subcutaneous route [19]. To aid in the characterization and prediction of drug response, in the last decade, biomarkers for MTX efficacy and toxicity have been extensively investigated. Serum levels of cellular proteins indicative of subclinical disease activity such as calprotectin (MRP 8/14) have been identified as plausible predictors of successful MTX withdrawal [20]. Polyglutamated forms of MTX and folate have been studied for correlation with drug response, and may be useful in the future to predict optimal or poor responders to the drug [21–23]. The effect of genotype upon drug response has been extensively studied in adults with RA and recently more in JIA. Genes in the purine synthesis portion of the folate pathway have been investigated, based on the biologic plausibility that enzyme variability may directly affect the formation of adenosine and thus the anti-inflammatory properties of MTX [24,25]. Single nucleotide polymorphisms (SNPs) in MTX transporter genes can also affect intracellular drug levels and alterations in response and efficacy [26,27]. A predictive response model was developed based on the presence of an elevation in the erythrocyte sedimentation rate (ESR) and SNPs in methionine synthase
298
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
reductase (MTRR), ATP-binding cassette, sub-family B, member 1 (ABCB1), ATP-binding cassette, subfamily C, member 1 (ABCC1), and the proton-coupled folate transporter (PCFT). If no factors were present, the probability of nonresponse, defined by an inability to reach an ACR Pedi 70, was 0.98, whereas if all factors were present, the probability of nonresponse was reduced to 0.42 [28]. The application of pharmacogenomics in the clinical setting has enormous potential, and the studies above reveal the importance of genotype on individual drug response; however, additional work needs to be done to 1) validate these findings, 2) understand the effect of ontogeny or development upon gene expression, and 3) develop rapid, cost-effective testing methods that can incorporate this knowledge into routine clinical care. MTX is generally considered safe and well tolerated. With its effect upon purine and pyrimidine synthesis, cells that require DNA for rapid turnover tend to be affected most, such as those in the gastrointestinal (GI) tract and bone marrow. The most common adverse effects involve the GI tract and include nausea, abdominal pain, and stomatitis within 24–48 h of the dose. However, in recent years, in addition to the physical GI symptoms, conditioned responses that result in anticipatory and associative GI symptoms with MTX have been recognized and termed MTX “intolerance,” [29] and these occur at much higher rates, potentially contributing to the poorer adherence as compared to biologic DMARDS [30]. Preventing these GI side effects may be the most effective way to prevent MTX “intolerance,” and hence utilizing daily folic acid and even antinausea medications such as ondansetron may prevent symptoms that can later become conditioned. Mild acute hepatotoxicity can occur and may require dose reduction, but hepatic fibrosis and cirrhosis have not been described in children [31]. Hematologic toxicity manifested as macrocytic anemia, leucopenia, and thrombocytopenia are rare, and reported infections are usually mild common bacterial infections. Utilizing a dose of 15 mg/m2/week has become standard in recent trials [32], and the 2011 ACR Recommendations for Treatment of JIA assumes MTX dosing to be 15 mg/m2/week administered via the parenteral route [11]. Current guidelines recommend the use of MTX in oligoarticular JIA unresponsive to NSAIDS or intra-articular glucocorticoid injections (IASIs), or if poor prognostic features are present, and in polyarticular and SJIA with active arthritis [11]. Data support MTX as an effective therapy in extended oligoarticular JIA more than several other JIA subtypes, for example the systemic subtype [33,34]. Although MTX is also included in one of the CTPs for new-onset SJIA [35], recently updated ACR recommendations for the treatment of SJIA suggest utilizing MTX for mild or moderate arthritis in SJIA rather than for the treatment of systemic features or MAS, where it has been shown to be less effective [12,33,34,36]. MTX has not been studied specifically for use in children with enthesitisrelated arthritis (ERA); however, a long-term open-label study has shown successful treatment of axial but not peripheral joint involvement in adults with ankylosing spondylitis (AS) [37]. The authors’ personal experience is to use MTX at a dose of 15 mg/m2/week preferably in the subcutaneous form as first-line therapy for extended oligoarticular or polyarticular JIA and after IASI or NSAID failure in persistent oligoarticular disease. We utilize MTX as an adjunctive therapy in SJIA for persistent arthritic features, but lean more heavily on interleukin-1 (IL-1) or IL-6 inhibitors for active systemic features. Leflunomide Leflunomide inhibits de novo pyrimidine synthesis by inhibiting the enzyme dihydroorotate dehydrogenase through its active metabolite. Few studies exist in children; however, a multinational randomized clinical trial comparing leflunomide to MTX demonstrated a similar mean percent improvement index between groups, though the rate of ACR Pedi 30 responses was significantly higher in the MTX group at week 16 (89% vs. 68%; p ¼ 0.02) [38]. At week 48, there were no significant differences in the mean percent improvement index and the rate of ACR Pedi 30 responses between groups. Interestingly, children
Contents lists available at ScienceDirect
Best Practice & Research Clinical Rheumatology journal homepage: www.elsevierhealth.com/berh
8
Therapeutic advancements in juvenile idiopathic arthritis Elizabeth A. Kessler, MD, MS 1, Mara L. Becker, MD, MSCE * Division of Rheumatology, Department of Pediatrics, Children’s Mercy Hospitals and Clinics, 2401 Gillham Road, Kansas City, MO 64108, USA
a b s t r a c t Keywords: Juvenile idiopathic arthritis DMARD TNF-a inhibitors Biologic Treatment strategy
The treatment of juvenile idiopathic arthritis (JIA) has substantially evolved over the past two decades. Research has been conducted and is ongoing on how therapies can best be utilized either as monotherapy or in combination for enhanced efficacy. The introduction of biologic therapies that selectively target specific cytokines has changed the acceptable clinical course of childhood arthritis. In addition to the development and utilization of new therapeutic agents, the pediatric rheumatology community has made vital progress toward defining disease activity, developing validated outcome measures, and establishing collaborative networks to assess both clinical outcomes and the long-term side effects related to therapeutics for juvenile arthritis. In this chapter, we will discuss the therapeutic evolution in JIA over the past two decades. Although the largest strides have been made with biologic agents, and these newer drugs have more rigorous data to support their use, select commonly used non-biologic therapies are included, with the discussion focused on more recent updated literature. Ó 2014 Elsevier Ltd. All rights reserved.
Introduction Juvenile idiopathic arthritis (JIA) is the most common pediatric autoimmune musculoskeletal condition, estimated to affect one in 1000 children [1]. It is classified based on age of onset, number and type of joints involved, and the presence of serologic markers, systemic signs, and symptoms. As the field has * Corresponding author. Tel.: þ1 816 234 3686; fax: þ1 816 234 3082. E-mail addresses: [email protected] (E.A. Kessler), [email protected] (M.L. Becker). 1 Tel.: þ1 816 234 3686; fax: þ1 816 234 3082.
http://dx.doi.org/10.1016/j.berh.2014.03.005 1521-6942/Ó 2014 Elsevier Ltd. All rights reserved.
294
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
developed, significant work has been invested in the development of an agreed-upon classification system for juvenile arthritis worldwide. The current JIA subtypes are characterized as follows: systemic, oligoarticular, polyarticular, enthesitis-related, psoriatic, and undifferentiated arthritis [2] (Table 1). Despite this relatively rare condition with a heterogeneous disease phenotype and poorly understood pathophysiology, the treatment strategies for JIA have evolved tremendously over the past two decades. Early treatment options for JIA were limited to nonsteroidal anti-inflammatory drugs (NSAIDs) including aspirin, corticosteroids, and non-biologic disease-modifying antirheumatic drugs (DMARDs) like D-penicillamine, gold, sulfasalazine (SSZ), and methotrexate (MTX). Historically, few treatment options were available for children with disease refractory to these medications, and the risk-to-benefit ratio for long-term corticosteroid use and many of the early DMARDs was not favorable, resulting in
Table 1 Definition of subtypes of juvenile idiopathic arthritis.a Category
Characteristics
Systemic
Arthritis in 1 joint(s) with or preceded by quotidian fever of 2 weeks’ duration documented for 3 days and accompanied by 1 of the following: 1. Evanescent erythematous rash 2. Generalized lymphadenopathy 3. Hepatomegaly and/or splenomegaly 4. Serositis
Oligoarticular
Arthritis in 1–4 joints during the first 6 months of disease A. Persistent – affecting 4 joints throughout disease course B. Extended – affecting >4 joints after the first 6 months of disease
Exclusions: a, b, c, d.
Polyarticular, RF negative Polyarticular, RF positive
Psoriatic
Enthesitis-related
Undifferentiated Exclusions
Exclusions: a, b, c, d, e. Arthritis in 5 joints during first 6 months of disease, negative RF Exclusions: a, b, c, d, e. Arthritis in 5 joints during first 6 months of disease, positive RF on 2 separate occasions at least 3 months apart Exclusions: a, b, c, e. Arthritis and psoriasis, or arthritis and 2 of the following: 1. Dactylitis 2. Nail pitting or onycholysis 3. Psoriasis in a first-degree relative Exclusions: b, c, d, e. Arthritis and enthesitis, or arthritis or enthesitis with 2 of the following: 1. Presence of or history of sacroiliac joint tenderness and/or inflammatory lumbosacral pain 2. Presence of HLA-B27 antigen 3. Onset of arthritis in male >6 years old 4. Acute anterior uveitis 5. AS, ERA, sacroiliitis with IBD, Reiter’s syndrome, or acute anterior uveitis in a first-degree relative Exclusions: a, d, e. Arthritis that fulfills criteria for no category or 2 categories a. Psoriasis or history of psoriasis in the patient or first-degree relative b. Onset of arthritis in an HLA-B27-positive male >6 years old c. AS, ERA, sacroiliitis with IBD, Reiter’s syndrome, or acute anterior uveitis in a first-degree relative d. Positive RF on two separate occasions at least 3 months apart e. SJIA in the patient
RF, rheumatoid factor; HLA-B27, Human Leukocyte Antigen B27; AS, ankylosing spondylitis; ERA, enthesitis-related arthritis; IBD, inflammatory bowel disease; SJIA, systemic JIA. a Modified from the International League of Associations for Rheumatology [2].
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
295
children with long-term morbidities from both their inadequately treated arthritis and also the therapies that were used as treatment. Major advances in the field, including the establishment of validated outcome measures, the organization of collaborative research networks, and the exploration of translational approaches to identify biomarkers of response and toxicity have been important contributors to therapeutic development in JIA. Outcome measures in JIA (see chapter 10) [118] In order to effectively compare treatment options and ultimately choose the best therapy in daily clinical practice, it is imperative that we have standardized and validated outcome measures. Definitions of disease flare, minimal disease activity, inactive disease, and clinical remission have been developed and validated, and are now used regularly in clinical trials and translational research studies. Tools for assessment of quality of life, an important component of health care for children with chronic disease like JIA, have also been developed (Table 2). In the past 10–20 years, there have been significant advancements in the creation of validated clinical outcome measures in JIA. The American College of Rheumatology Pediatric (ACR Pedi) response
Table 2 (see chapter 10) [118] Definitions of commonly used outcome measurements in pediatric rheumatology. Pediatric response criteria ACR Pedi 30
Minimum of 30% improvement from baseline in at least three of the six core set criteria, with a worsening of 30% in no more than one variable
ACR Pedi 50, 70, 90, and 100
Minimum improvement of 50%, 70%, 90%, and 100%, respectively, from baseline in at least three of the six core set criteria with a worsening of 30% in no more than one variable ACR pediatric core set criteria 1. Physician global assessment of disease activity (10 cm visual analog scale) 2. Parent/patient assessment of disease activity (10 cm visual analog scale) 3. Functional ability (CHAQ) 4. Number of joints with active arthritis 5. Number of joints with limited range of motion 6. ESR JADAS Composite score based on the arithmetic sum of the four following measures: 1. Physician global assessment of disease activity (VAS) 2. Parent/patient assessment of disease activity (VAS) 3. Number of joints with active arthritis based on 27-joint count 4. ESR Definitions of disease activity [113,114] Clinical inactive disease 1. No joints with active arthritis (all six conditions must be met) 2. No fever, rash, serositis, splenomegaly, or generalized lymphadenopathy related to JIA 3. No active uveitis 4. ESR or CRP level (or both if tested) within normal limits or, if elevated, not related to JIA 5. Physician global assessment of disease activity of lowest possible on the scale used 6. Duration of morning stiffness 15 min Clinical remission on medication Inactive disease on medication must persist for 6 months Clinical remission off medication Inactive disease for 12 months after discontinuation of all JIA therapy Quality-of-life measurement tools CHAQ [115] A disease-specific instrument that measures functional ability in activities of daily living PedsQL [116] Measures health-related quality of life via both child self-report and parent proxy report scales using both generic and rheumatology-specific measures CHQ [117] Generic instrument that contains 14 domains to evaluate the physical, emotional, and social components of health and provides physical and psychosocial summary scores ACR, American College of Rheumatology; CHAQ, Childhood Health Assessment Questionnaire; VAS, visual analog scale; ESR, erythrocyte sedimentation rate; sJIA, systemic juvenile idiopathic arthritis; JADAS, Juvenile Arthritis Disease Activity Score; CRP, C-reactive protein; JIA, juvenile idiopathic arthritis; PedsQL, Pediatric Quality of Life; CHQ, Child Health Questionnaire.
296
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
criteria are a composite score based on improvement in six core outcome variables [3] (Table 2). Limitations of the ACR Pedi response criteria include the inability to directly compare absolute response between patients or measure change in absolute disease activity. Additionally, betweenstudy comparisons cannot easily be made [4]. As a result, the Juvenile Arthritis Disease Activity Score (JADAS) was developed, which provides continuous measurement of response, modeled after the Disease Activity Score in adult patients with rheumatoid arthritis (RA) (Table 2) [5]. Limitations of the JADAS include lack of meaningful cutoff points defining disease activity and a potential lack of generalizability to patients with less severe oligoarticular disease [5,6]. Collaborative networks The formation of the Pediatric Rheumatology Collaborative Study Group (PRCSG) was a milestone in the advancement of therapeutics for pediatric rheumatology. This group developed a standard methodology for the design, conduct, and analysis of drug trials in children with rheumatic diseases, which has played a major role in the investigation of new biologics such as etanercept, adalimumab, and abatacept in children [7]. The PRCSG has also spearheaded collaborative working agreements with groups such as the Pediatric Rheumatology International Trials Organization (PRINTO) to combine efforts across the globe to advance research and therapy for children with rheumatic disease, illustrated by the recent phase III study of canakinumab in JIA [8]. The Childhood Arthritis and Rheumatology Research Alliance (CARRA) is a North American organization of pediatric rheumatologists committed to investigator-initiated research, with important contributions including, but not limited to, the establishment of the largest multicenter registry of children with rheumatic diseases, and the CARRA consensus treatment plans (CTPs), which were developed to identify effective therapies in clinical settings through comparative effectiveness research. CARRA and PRINTO have also become part of vital pharmacovigilance projects to evaluate the longterm safety of the therapeutic agents currently used in JIA. CARRA has collaborated with the US Food and Drug Administration (FDA), biopharmaceutical industry representatives, and the Duke Clinical Research Institute to form a pharmacosurveillance model (CARRA Consolidated Safety Registry (CoRe)) that utilizes the longitudinal multicenter registry, which currently has data on approximately 7000 children with JIA [9]. Pharmachild is a pharmacovigilance program involving participating centers of PRINTO, the Pediatric Rheumatology European Society (PReS), and pharmaceutical companies consisting of twice-yearly registration of clinical data and standardized adverse events [10]. International expert opinion and collaboration were used to establish the 2011 ACR-published recommendations for the treatment of JIA [11]. This was the first time the ACR has made a concerted effort to establish treatment recommendations for a pediatric rheumatic disease. These recommendations take into account patient prognostic factors, as well as disease activity levels, and are based on expert opinion and available published studies. An update of these recommendations was completed in 2013 specifically focusing on the treatment of systemic JIA (SJIA) and macrophage activation syndrome (MAS), recognizing the importance of this severe subtype and the rapid advancement in therapeutics targeting key cytokines in this condition [12]. These recommendations represent important progress toward having evidence-based guidelines for the treatment of JIA by providing guidance in both initial and subsequent therapies for JIA, as well as recommendations for medication safety monitoring. Utilization of translational approaches Translational approaches to investigate or predict treatment response in JIA have provided the potential for more precise investigative tools to use in heterogeneous conditions such as JIA, where the disease course is difficult to predict, and it is challenging to know when it is appropriate to escalate or discontinue therapy. Cellular, genetic, and even radiologic biomarkers have been explored as more sensitive measures to characterize drug response and identify risk of future flare. Pharmacogenomics may identify genetic variation associated with drug response, and, like in adults, has been explored in pediatric rheumatic diseases. These modalities are more frequently being used to enhance our current investigations, and they have the potential for use in mainstream clinical practice.
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
297
In this chapter, we will discuss the evolution of therapeutics for JIA over the past two decades. Studies conducted prior to the newest International League of Associations for Rheumatology (ILAR) classification criteria for JIA utilized a variety of terms to classify childhood arthritis; thus, we will discuss the results of these studies utilizing the terminology used at the time the study was conducted. As there are subtle differences in these criteria, we feel it is more accurate to interpret study data with a clear understanding of the subgroups of juvenile arthritis that were included. Although the largest strides have been made in biologic agents in the past two decades, select commonly used non-biologic therapies are included, with the discussion focused on more recent, updated literature, and any new developments that may aid in our therapeutic management. Non-biologic DMARDs Methotrexate MTX is a folic acid analog and a potent competitive inhibitor of several enzymes in the folate pathway resulting in interruption of purine and pyrimidine synthesis, and an accumulation of adenosine, which is thought to be a large contributor to the site-specific anti-inflammatory effects of MTX [13]. MTX is the cornerstone of therapy for childhood arthritis and remains the most commonly used non-biologic DMARD in the treatment of JIA. Its efficacy and safety in juvenile arthritis has been reported in several prospective uncontrolled studies in children and was clearly demonstrated over two decades ago in a double-blind randomized controlled trial (RCT) that required multinational collaborative efforts [14]. Despite a relatively long track record in children with rheumatic disease, there remains unpredictable variability in response to this drug, and extensive efforts in the past two decades have focused on identifying its optimal dose and route as well as cellular, serum, and genetic biomarkers to aid in individualizing therapy in this population of patients. Extensive variability in dosing preferences in JIA has prompted work to identify the optimal dose and route of MTX. Recent work has shown no significant therapeutic advantage and higher risk of transaminase elevation with higher initial MTX doses [15]. For nonresponders to conventional MTX dosing, there are reports supporting the safety and efficacy of escalating to higher dosing regimens (25–30 mg/m2/week) [16]. However, a multicenter prospective study in JIA patients who were nonresponders to initial low-dose MTX suggests otherwise. [17] Nonresponders were randomized to intermediate-dose (15 mg/m2/week) or high-dose (30 mg/m2/week) MTX; both groups achieved similar ACR Pedi outcomes. The ACR Pedi 30 response rate was 62.5% in the intermediate-dosed group and 57.5% in the high-dosed group. Similar outcomes in intermediate-dosed and high-dosed groups were observed for the ACR Pedi 50 (57.5% and 55%) and ACR Pedi 70 (45% and 47.5%) response, respectively. The route of administration of MTX also continues to be debated. There is altered oral bioavailability of the drug with higher doses, and early studies suggested an increased effectiveness of subcutaneous MTX administration in patients who did not respond or tolerate initial oral dosing [18]; however, oral administration of the drug is still prevalent and has been shown to be effective in some studies. In a recently published large observational German MTX Registry, over half the patients (63%) received oral MTX exclusively, and this group had similar rates of ACR Pedi 30, 50, and 70 response as well as toxicity to subjects treated via the subcutaneous route [19]. To aid in the characterization and prediction of drug response, in the last decade, biomarkers for MTX efficacy and toxicity have been extensively investigated. Serum levels of cellular proteins indicative of subclinical disease activity such as calprotectin (MRP 8/14) have been identified as plausible predictors of successful MTX withdrawal [20]. Polyglutamated forms of MTX and folate have been studied for correlation with drug response, and may be useful in the future to predict optimal or poor responders to the drug [21–23]. The effect of genotype upon drug response has been extensively studied in adults with RA and recently more in JIA. Genes in the purine synthesis portion of the folate pathway have been investigated, based on the biologic plausibility that enzyme variability may directly affect the formation of adenosine and thus the anti-inflammatory properties of MTX [24,25]. Single nucleotide polymorphisms (SNPs) in MTX transporter genes can also affect intracellular drug levels and alterations in response and efficacy [26,27]. A predictive response model was developed based on the presence of an elevation in the erythrocyte sedimentation rate (ESR) and SNPs in methionine synthase
298
E.A. Kessler, M.L. Becker / Best Practice & Research Clinical Rheumatology 28 (2014) 293–313
reductase (MTRR), ATP-binding cassette, sub-family B, member 1 (ABCB1), ATP-binding cassette, subfamily C, member 1 (ABCC1), and the proton-coupled folate transporter (PCFT). If no factors were present, the probability of nonresponse, defined by an inability to reach an ACR Pedi 70, was 0.98, whereas if all factors were present, the probability of nonresponse was reduced to 0.42 [28]. The application of pharmacogenomics in the clinical setting has enormous potential, and the studies above reveal the importance of genotype on individual drug response; however, additional work needs to be done to 1) validate these findings, 2) understand the effect of ontogeny or development upon gene expression, and 3) develop rapid, cost-effective testing methods that can incorporate this knowledge into routine clinical care. MTX is generally considered safe and well tolerated. With its effect upon purine and pyrimidine synthesis, cells that require DNA for rapid turnover tend to be affected most, such as those in the gastrointestinal (GI) tract and bone marrow. The most common adverse effects involve the GI tract and include nausea, abdominal pain, and stomatitis within 24–48 h of the dose. However, in recent years, in addition to the physical GI symptoms, conditioned responses that result in anticipatory and associative GI symptoms with MTX have been recognized and termed MTX “intolerance,” [29] and these occur at much higher rates, potentially contributing to the poorer adherence as compared to biologic DMARDS [30]. Preventing these GI side effects may be the most effective way to prevent MTX “intolerance,” and hence utilizing daily folic acid and even antinausea medications such as ondansetron may prevent symptoms that can later become conditioned. Mild acute hepatotoxicity can occur and may require dose reduction, but hepatic fibrosis and cirrhosis have not been described in children [31]. Hematologic toxicity manifested as macrocytic anemia, leucopenia, and thrombocytopenia are rare, and reported infections are usually mild common bacterial infections. Utilizing a dose of 15 mg/m2/week has become standard in recent trials [32], and the 2011 ACR Recommendations for Treatment of JIA assumes MTX dosing to be 15 mg/m2/week administered via the parenteral route [11]. Current guidelines recommend the use of MTX in oligoarticular JIA unresponsive to NSAIDS or intra-articular glucocorticoid injections (IASIs), or if poor prognostic features are present, and in polyarticular and SJIA with active arthritis [11]. Data support MTX as an effective therapy in extended oligoarticular JIA more than several other JIA subtypes, for example the systemic subtype [33,34]. Although MTX is also included in one of the CTPs for new-onset SJIA [35], recently updated ACR recommendations for the treatment of SJIA suggest utilizing MTX for mild or moderate arthritis in SJIA rather than for the treatment of systemic features or MAS, where it has been shown to be less effective [12,33,34,36]. MTX has not been studied specifically for use in children with enthesitisrelated arthritis (ERA); however, a long-term open-label study has shown successful treatment of axial but not peripheral joint involvement in adults with ankylosing spondylitis (AS) [37]. The authors’ personal experience is to use MTX at a dose of 15 mg/m2/week preferably in the subcutaneous form as first-line therapy for extended oligoarticular or polyarticular JIA and after IASI or NSAID failure in persistent oligoarticular disease. We utilize MTX as an adjunctive therapy in SJIA for persistent arthritic features, but lean more heavily on interleukin-1 (IL-1) or IL-6 inhibitors for active systemic features. Leflunomide Leflunomide inhibits de novo pyrimidine synthesis by inhibiting the enzyme dihydroorotate dehydrogenase through its active metabolite. Few studies exist in children; however, a multinational randomized clinical trial comparing leflunomide to MTX demonstrated a similar mean percent improvement index between groups, though the rate of ACR Pedi 30 responses was significantly higher in the MTX group at week 16 (89% vs. 68%; p ¼ 0.02) [38]. At week 48, there were no significant differences in the mean percent improvement index and the rate of ACR Pedi 30 responses between groups. Interestingly, children
