ACR 2006 Highlights from Johns Hopkins University

Myositis Highlights

Lisa Christopher-Stine, M.D.

Abstract # 2053 IL-17 and IFN-Regulated Genes and Chemokines as Biomarkers of Disease Activity in Inflammatory Myopathies

Authors:

Hatice Bilgic, Steven R. Ytterberg, Kelly T. McNallan, Joseph C. Wilson,  Thearith Koeuth1, Jason W. Bauer, Kevin G. Moder, Shreyasee Amin, Clement J. Michet,  Erik J.Peterson,, Emily C. Baechler, Ann M. Reed.

Background:

Adequate assessment of disease activity in idiopathic inflammatory myopathies (IIM) remains problmatic. This study investigate a gene signature that will hopefully aid in clinical care and lead to novel models of disease pathogenesis. The Interferon-alpha (IFN) signature is a candidate biomarker in dermatomyositis. IL-17 and Th17 pathway have been implicated in several autoimmune diseases, but a link with IIM has not yet been established. We hypothesized that the IFN signature and other pro-inflammatory cytokines, including IL-17 and IL-6, may be biomarkers for disease activity in IIM, including adult (DM) or juvenile dermatomyositis (JDM) and polymyositis (PM).

Methods:

Peripheral blood samples and clinical/laboratory data were obtained from 64 patients who met the Bohan and Peter criteria for IIM (33 DM, 14 JDM, 17 PM). The whole blood IFN gene expression signature was defined by expression levels of 2 IFN-regulated genes (IFIT1, G1P2) measured by quantitative real-time RT-PCR. Multiplexed sandwich immunoassays were used to quantify serum levels of 4 IFN-regulated chemokines (I-TAC, IP-10, MCP-1, MCP-2) and other pro-inflammatory cytokines. Spearman’s rank correlation coefficient was used to describe the relation between biomarkers and disease activity (Physician’s Global Visual Analogue Score).

Results:  

In pooled DM/JDM cases, disease activity was positively correlated with the IFN gene expression signature and serum levels of IFN-regulated chemokines. The investigators also detected significant correlations between disease activity and serum levels of IL-17 and IL-6. Adiitionally, tight correlation was detected between IL-17 levels and  theIFN gene expression signature, and between IL-17 and IFN-regulated chemokine levels. 
In PM cases: The IFN gene expression signature did not correlate with disease activity, and MCP-1 was the only IFN-regulated chemokine that was significantly correlated with disease activity. However, serum IL-17 was significantly correlated with the IFN signature and IFN-regulated chemokine levels.

Conclusions:

IL-17 and IFN gene/protein signature are candidate biomarkers for disease activity in DM/JDM. If these results are validated, measurement of these candidate biomarkers may permit more refined disease activity evaluations in IIM. In addition, observed co-regulation of IFN-driven chemokines and Th-17-related cytokines suggests a novel pathogenic linkage between these pro-inflammatory pathways in IIM patients.

Editorial comment:

IL-17 and Th17 pathway have been implicated in several autoimmune diseases, but until now, a link with IIM has not yet been established. IL-17 is a very hot topic as a potential therapeutic target in many inflammatory autoimmune diseases. Pharmaceutical companies have already developed an oral anti IL-17 antibody that will likely go into phase I clinical trial for IIM this coming year.  In addition to IL-17 as a potential therapeutic target due to its upregulation ad tight correlation with INF alpha gene expression, this may also be a surrogate marker for disease activity if it were readily available and could be easily measured commercially.  The tight correlation between INF gene expression and IL-17 levels makes one wonder whether the previously upegulated INF gene signature is directly related to increased IL-17 and whether IL-17 either alone or in combination with INF, in fact, may be a more potent therapeutic target than INF alone.

Abstact # 2054 Role of Adenine Nucleotide Catabolic Enzymes In Mediating Muscle Weakness in Myositis

Authors:

Gouri S. Pandey, Richard L. Sabina, Robert L. Wortmann,  Emidio Pistilli, Jack VanderMeulen, Rashmi Rawat, Nina Raben, Paul H. Plotz, Kanneboyina Nagaraju.

Background:

The major histocompatibility complex class I (MHC class I) transgenic mouse model of myositis models several features of myositis patients including muscle weakness. It was proposed by Dr. Sabina and colleagues in 1991 that acquired deficiency of AMP deaminase (AMPD) may be responsible for muscle weakness in myositis. Muscle weakness starts early in these mice which affords the evaluation of  expression and activity of adenine nucleotide catabolic enzymes in the skeletal muscle fibers of control and transgenic mice.

Methods:

Gene expression profiling of type I (soleus) and type II (gastrocnemius) muscles fibers and RT-PCR of selected genes; AMPD enzyme activity, immunohistochemical staining for myosin heavy chain, pathway specific enzymes and in vitro muscle force and in vivo behavioral activity were measured.

Results:

This study demonstrated that several genes of adenine nucleotide catabolic pathway (adenosine kinase; AMPD 1 and 3; adenosine deaminase)were differentially expressed in type 1 and type 2 muscles of control and transgenic mice. AMPD enzyme activities as well as AMPD1 mRNA expression (RT-PCR) was significantly decreased in type II fibers of MHC class I mice in comparison to control mice (Type 2 muscle AMPD activity (mU/mg soluble protein) (n=6); control (3799±1087) vs. transgenic (790±834 p<0.001). Preliminary data also indicated that decrease in AMPD activity precedes disease symptoms. Conversely, the investigators did not find a significant difference in AMPD activity between Type 1 muscle fibers of control and MHC mice (enzyme activity (mU/mg) (n=6); control (190±136) vs. transgenic (362±205; p=0.12). Additionally, no decrease in AMPD activity in other models of muscle disease (mdx and A/J) was observed. The authors provide further evidence that the decrease in AMPD activity may be due to active fiber type conversion from type 2 to type 1 muscle in the transgenic mice. These biochemical and histological data are also consistent with decreased behavioral (grip strength, locomotor activity) and in vitro muscle function parameters such as specific muscle force, twitch force, time to peak tension and slope in these mice.

Conclusion:

This study demonstrated that the adenine nucleotide catabolic pathway is differentially expressed in type 1 and type 2 muscle fibers and those differences from control animals correlate with in vivo and in vitro muscle weakness in the MHC class I mice. Since these differences appear to be specific to myositis, the study proposes that AMPD may be an early marker of the skeletal muscle pathology in myositis. Thus, understanding the molecular mechanisms controlling this pathway may provide clues for treating muscle weakness for myositis.

Editorial comment:

There are several mechanisms by which myositis potentially develops – both immune mediated and non-immune mediated. This study explores the arm of a potentially non-immune mediated mechanism, drawing from findings noted 17 years ago but not followed up further. AMP deficiency is a common fining in the general population, affecting up to 2% of us, but for most, there is little clinical significance. This provocative study explores the phenomenon of AMP deficiency as a functional mediator of weakness n myositis mice. It will be interesting to see if these findings are translated to human myositis muscle.

Abstract # 2055 The Role of The Alarmin HMGB-1 In The Pathogenesis Of Inflammatory Myopathies

Authors:

Ingrid E. Lundberg, Cecilia Grundtman, Joseph Bruton, Therese Östberg, David S. Pisetsky, Helena Erlandsson Harris, Ulf Andersson, Håkan Westerblad.

Idiopathic inflammatory myopathies (IIMs) are characterized by muscle weakness and inflammation, although the mechanisms damaging muscle are unknown. High mobility group box chromosomal protein-1 (HMGB-1) is a ubiquitously expressed nuclear protein that can serve as alarmin and display potent pro-inflammatory activity when released from cells. Cytoplasmic HMGB-1 expression in infiltrating inflammatory cells, vascular endothelial cells and muscle fibers as well as extracellular staining surrounding inflammatory cells was previously observed in IIM patients. (CHECK PUB MED)

Purpose:

To investigate whether HMGB-1 in muscle fibers can translocate from nucleus to cytoplasm of muscle fibers and impair muscle fiber contractility.

Methods:

HMGB-1 expression was assessed in differentiated muscle fibers from wild-type mice after incubation with interferon-gamma (IFN-.). Muscle fibers from wild-type mice and from receptor for advanced glycated end products (RAGE)-/- mice were incubated with recombinant HMGB-1. Contractility was measured indirectly as intracellular Ca2+ release; major histocompatibility complex (MHC) class I expression was also determined. Muscle tissue from IIM patients were investigated by immunohistochemistry.

Results:

IFN-. induced up-regulation of HMGB-1 in fiber nuclei and translocation of HMGB-1 to the fiber cytoplasm. Recombinant HMGB-1 induced MHC class I expression and impaired Ca2+ release from the sarcoplasmic reticulum during induction of fatigue in a concentration dependent manner in muscle fibers from wild-type and RAGE-/- mice. No expression of RAGE in muscle fibers in RAGE-/- or wild-type mice was noted. In patients with short disease duration and without inflammatory muscle infiltrates, HMGB1co-localized with MHC class I expression in the cytoplasm of muscle fibers; but more fibers were positive for HMGB-1 than for MHC class I. In patients with inflammatory infiltrates,  HMGB-1 and MHC class I were expressed in the cytoplasm of the same muscle fibers. 

Conclusions:

HMGB-1 may be translocated from muscle fiber nuclei to cytoplasm upon stimulation by pro-inflammatory stimuli. HMGB-1 may serve as an inducer of MHC class I in muscle fibers as supported by the clinical data from patients with early disease. Furthermore, HMGB-1 may promote muscle weakness by modulating Ca2+ release; this effect was independent of RAGE. Together, these findings point to HMGB-1 as a disease mediator and potential target for therapy in IIMs.

Editorial comments:  

There are few disease-specific therapeutic targets currently known. HMGB-1 inhibition may be one such unique target, as reduction in HMGB-1 would theoretically reduce the burden of clinical muscle disease by increasing muscle strength by prvention of calcium release as well as inhibiting MHC class I induction, known to be upregulated in myocytes of patients with inflammatory myopathies and not in normal, disease-free myocytes.

Abstract # 2056 MDA5 (Melanoma-Differentiation-Associated Gene 5) as an Autoantigen Recognized by Anti-CADM-140 Antibody in Patients with Clinically Amyopathic Dermatomyositis

Authors:

Shinji Sato1, Kana Hoshino, Takashi Satoh, Akira Suwa, Michito Hirakata, Shinichi Inada, Masataka Kuwana.

Background:

This group of investigators previously identified an autoantibody against a 140 kDa polypeptide (anti-CADM-140 antibody), which is detected preferentially in patients with clinically amyopathic dermatomyositis (C-ADM) with rapidly progressive interstitial lung disease (RP-ILD). The antigenic target and its association with pathogenesis, however remained  unclear. This study has identified the CADM-140 autoantigen using a series of molecular techniques.

Methods:

Anti-CADM-140 prototype serum was used to screen HeLa cells. Positive clones were further examined for their reactivity to nine anti-CADM-140 positive sera with C-ADM and twelve  anti-CADM-140 negative sera (2 with C-ADM, 2 with polymyositis, one with juvenile DM, one with systemic lupus erythematosus, one with systemic sclerosis and 7 with normal individuals) using a phage expression system. Candidate clones were subjected to immunoprecipitation (IP) of in vitro transcribed and translated products. Nucleotide sequence of the selected clone was determined and applied to homology search using the NCBI database. Finally, goat anti-MDA5 (melanoma-differentiation-associated gene 5) polyclonal antibody was used for IP-blotting to confirm whether anti-CADM-140 positive sera react with MDA5.

Results:

Nine clones were selected by the cDNA library screening using anti-CADM-140 prototype serum. Of these, one clone was recognized by eight anti- CADM-140 positive sera but not by any control sera. Anti-CADM-140 positive sera immunoprecipitated in vitro transcribed and translated products of that clone but no control sera. Nucleotide sequence of the clone was identical to the carboxyl terminal potion of MDA5. Moreover, goat anti-MDA5 polyclonal antibody reacted with approximate 140kDa immunoprecipitates of anti-CADM-140 positive sera from HeLa cell extracts.

Conclusions: 

These results indicate that anti-CADM-140 autoantibody in sera from C-ADM patients recognizes MDA5. Given the fact that MDA5 is involved in innate immune defense against viruses through the detection of viral dsRNA, these investigators postulate that virus infection may play an important role in development of C-ADM with RP-ILD.

Editorial comments:

This new antigenic target differs from most previously-associated myositis related antigens. The phenotype is quite specific and often deadly; thus, early identification of this subtype would be helpful.  Perhaps commercially available antibodies to this antigenic target could be made available to help to screen for this phenotype. Early identification may lead to streamline therapy and better prognostic outcomes. A viral insult as a trigger for myositis has often been suggested but never substantiated. The fact that the autoatigen here is involved in the innate immune system to defend one from viruses re-visits the potential pathogenic link of myositis and viral etiologies. It will be important to replicate this findings in other ethnic myositis patients, as to date, this has only been studied in patients of Japanese descent.

Presentation Number: 2057 Novel Autoantibodies Targeting a p140 Protein are a Major Autoantigen System in Juvenile Dermatomyositis and a Marker of Calcinosis

Authors:

Harsha Gunawardena, Lucy R. Wedderburn, Hector Chinoy, Zoe E. Betteridge, Jean North, William ER Ollier, Robert G. Cooper, Athimalaipet V. Ramanan, Joyce E. Davidson, Neil J.

Background:

To date, there are few antibody-phenotype correlations in children with JDM, in contrast to the well defined and emerging antibody-phenotype correlations in adult myositis patients. Identification of novel autoantigens in juvenile dermatomyositis (JDM) may lead to the identification of homogeneous subsets, further insights into pathogenesis and have prognostic implications. The aim of this study was to demonstrate that autoantibodies target a p140 protein and this is a major autoantigen in JDM. It describes the associations of anti-p140 autoantibodies in children recruited to the JDM National Registry (UK and Ireland).

Methods:

Clinical data and sera were collected from the first 156 children recruited. Serum was screened by immunofluorescence (IF) and immunoprecipitation (IPP). Immunodepletion was performed to establish whether p140 is a different target to p155/140 which is also recognized in this cohort (bt is seemingly not related to cancer, as postulate din the adult myositis cohort) . DNA from 100 children was genotyped for HLA-DRB1 and compared to 864 randomly selected UK Caucasian controls.

Results:

21% of children were positive for anti-p140 on IPP, with a weak non-specific nuclear pattern or negative ANA on IF. There were no other antibodies noted in the  anti-p140 cases. Immunodepletion confirmed that p140 and p155/140 are different autoantigens. There was no significant difference in sex ratio, age at diagnosis and disease duration between anti-p140 positives and negatives. No significant difference was observed in the type or distribution of rash when comparing anti-p140 positives with negatives except for more rash on the trunk in negative cases (p=0.017). Calcinosis, however,  was significantly more frequent in anti-p140 positives (52%) compared to negatives (13%) (p<0.001, OR 7.1 95% CI 3-17). When comparing anti-p140 and anti-p155/140 cases; cutaneous edema (p=0.04), rash over the trunk (p=0.002) and small joints (p=0.013) was more frequent with antip155/ 140. Calcinosis in anti-p140 remained a significant feature compared to anti-p155/140 (p=0.005, OR 6.4 95% CI 2-22). HLA-DRB1*08 was a significant risk factor in antip140 positives (25%) compared to controls (5%) (OR 5.8 95% CI 2-18, p(corr)=0.002).

Conclusion :

Anti-p140 found in this cohort is likely to be the same as anti-MJ, described to target nuclear matrix protein NXP-2 (described previously by Targoff, Oddis, and colleagues). Further confirmation is required, however. Anti-p140 is a major autoantibody subset in JDM and is associated with HLA-DRB1*08. Further characterization may provide insights into the pathophysiology of calcinosis and JDM. 

Editorial comments:

Because calcinosis remains quite recalcitrant to therapy in both children and adults, it would be helpful to further elucidate the pathogenesis or associated autoatibodies. To date, calcinosis does not appear to respond to immuosuppressive therapy, and sometimes disease activity and calcinosis development and progression may be discordant. Perhaps early identification of this subset of JDM (and perhaps adult DM) patients could lead to early, potentially prophylactic therapy for calcinosis and reduce the burden of disease for these patients.

Abstract # 167 Clinical Heterogeneity Within the Antisynthetase Syndrome

Authors:

Lisa Christopher-Stine, Grace Hong, Livia Casciola-Rosen, Sonye K. Danoff, Andrew L. Mammen, Allan C. Gelber.

Background:

The antisynthetase syndrome is characterized by the presence of one of the antisynthetase autoantibodies in addition to interstitial lung disease, inflammatory myopathy, fever, polyarthritis, Raynaud’s phenomenon, and mechanics’ hands. Anti-Jo-1 is the most common antisynthetase autoantibody(ASA); however, 8 other ASAs have been identified, all of which are directed against enzymes that acetylate tRNA.
Objective: We sought to determine whether the salient features of the antisynthetase syndrome differ in those patients with less common ASAs (anti-OJ, anti-PL-7, and anti-PL12) when compared to those seropositive for Jo-1.

Methods:

Sera from 191 consecutive myositis patients were evaluated for the presence of an ASA. Jo-1 antibody was assayed by ELISA. The other ASAs were assayed by immunoprecipitation. Clinical characteristics, including fever, arthritis, Raynaud’s phenomenon, mechanic’s hands, interstitial lung disease (defined radiographically by CT scan); probable or definite myositis defined by Bohan and Peter criteria, as well as MRI muscle edema were recorded. Associations were examined using a Fisher exact test; p<0.05 indicated statistical significance.

Results:

Specifically, 29 were positive (anti-Jo-1 (n=20); anti-PL-12 (n=5), anti-PL-7 (n=2), and anti-OJ (n=2). These seropositive patients were 86% female, 48% Caucasian, with a median age of 41.5 years at the time of onset of muscle symptoms.

Table 1. Clinical Characteristics of patients with Anti-Jo-1 vs. Other Antisynthetase Autoantibodies


Clinical Characteristic

Anti-Jo-1

Other synthetase

P-value

(% positive, N=20)

(% positive, N=9)

Fever

77

89

0.64

Arthritis

65

100

0.074

Raynaud's phenomenon

60

78

0.43

Mechanics hands

50

67

0.45

Interstitial lung disease

79

100

0.27

Bohan & Peter Myositis

89

33

0.0048

EMG +

82

29

0.021

MRI Edema

80

71

1

Muscle Bx +

88

25

0.0093

Elevated CPK (ever)

95

89

0.53

Elevated aldolase (ever)

88

100

0.54

Patients with ASAs other than Jo-1 were less likely to meet the Bohan & Peter criteria for a diagnosis of myositis than Jo-1 positive patients [33% vs 89%; (p=0.0048)]. In contrast, there was no difference between the groups with respect to the presence of an elevated CPK or aldolase level or of MRI muscle edema. However, patients with ASAs other than Jo-1 were less likely to have EMG electrodiagnostic evidence of myopathy [29% vs. 82%] (p=0.021) and less likely to have histologic evidence of inflammation [20% vs 88%] (p=0.0093). No other syndrome specific clinical characteristics differed in frequency between the groups.

Conclusion:

There is heterogeneity of clinical characteristics among the well-defined antisynthetase syndromes. Distinct autoantibodies appear to confer specificity to the degree of muscle involvement, as assessed electrodiagnostically and by MRI imaging. Most importantly, consideration of a diagnosis of the antisynthetase syndrome should not be ruled out in patients presenting with  a normal EMG or negative muscle biopsy in the presence of muscle weakness (overt or subtle), elevated muscle enzymes, and interstitial lung disease.

Editorial comments:

We undertook this analysis based on our experience in the Johns Hopkins Myositis Center. Often, patients would be referred for interstitial lung disease with an isolated (sometimes transiently elevated) elevated CPK, with or without weakness, with a negative or inconclusive muscle biopsy and a normal or non-irritable EMG which suggested that they did not have the traditionally accepted criteria for myositis according to Bohan and Peter and thus an autoimmune cause for their lung disease was removed from consideration on the differential diagnosis. This study helps to illustrate the clear difference between the Jo-1 and non-Jo-1 patients with regard to diagnostic utility of EMG and muscle biopsy in these subsets of patients.

Abstract #  2058 A Randomized, Double-Blind, Placebo-Controlled Trial of Infliximab in Patients with Polymyositis and Dermatomyositis

Authors:

K. Coyle, A. Pokrovnichka, K. French, G. Joe, J. Shrader, L. Swan, I. Cabalar, M. Harris-Love, P. Plotz, F. Miller, M. Gourley.

Background/Purpose:

To investigate the safety and efficacy of infliximab in patients with refractory dermatomyositis (DM) and polymyositis (PM) receiving corticosteroids combined with methotrexate or azathioprine.

Methods:

This 40-week randomized, double-blind, placebo-controlled trial included subjects >18 years old who: fulfilled Bohan & Peter criteria; had evidence of active disease with muscle enzyme elevation, STIR MRI signal abnormalities (as measurd by a composite activity score designed by the musculoskeletal radiologist at the NIH), baseline manual muscle testing (MMT) score of 80-120 (out of 160); stable prednisone < 0.5 mg/kg/day and DMARD doses for 4 weeks prior to participation; and who failed previous immunosuppressive treatment or had intolerance to therapy. Exclusion criteria included malignancy, CNS disease, prior infliximab therapy, infection, failure to comply with adequate birth contrl methods,  lupus and heart failure.

Participants received four blinded infusions of either placebo or infliximab 5 mg/kg at 0, 2, 6 and 14 weeks. Un-blinding occurred at week 16 following assessment of improvement by the primary (=15% MMT improvement from week 0) and secondary outcomes (improvement as defined by the International Myositis Assessment and Clinical Studies Group (IMACS) criteria). Subjects in the infliximab arm who did not improve based on MMT criteria (nonresponders) were increased to open-label infliximab 7.5 mg/kg at weeks 22, 30, 38. Nonresponders in the placebo arm received open-label 5 mg/kg at weeks 16, 18, 22, 30, 38. Responders in each arm had the option of either continuing the same treatment or changing to the non-responder treatment for that study arm. Outcomes were reassessed at week 40.

Results:

Twelve subjects (11F/1M; 6 African American, 5 Caucasian, 1 Asian; 11 PM /1 DM) completed the study to week 16. For these subjects, the mean age was 45.4 + 10.9 years, mean disease duration 5.6 + 4.0 years, mean corticosteroid dose was 17.3 + 10.8 mg/day; mean methotrexate dose (N=9) was 16.4 + 11.3 mg/week and mean azathioprine dose (N=3) 50.0 + 86.1 mg/day. Mean baseline MMTs were 104.2 + 9.9 (placebo arm, N=6) vs. 104.2 + 17.3 (infliximab arm, N=6). Three subjects in the infliximab arm withdrew after week 16; 3 subjects completed the infliximab 7.5 mg/kg dosing regimen. All 6 patients in the placebo arm crossed over to the 5 mg/kg dosing regimen after week 16.

On infliximab 3/12 subjects improved by MMT and 7/12 subjects improved by IMACS criteria at 16 weeks and one additional subject improved by MMT at 40 weeks. With placebo treatment none of the 6 subjects improved for primary outcome and two improved for secondary outcome. Two SAEs, unrelated to study drug, occurred during the study.

Adverse events were otherwise mild. One patient experienced an infusion reaction which was well controlled with premedication with subsequent dosing.

Conclusions:

This first randomized, controlled clinical trial of infliximab reported in patients with myositis suggests that infliximab is well tolerated but has only limited efficacy in adult PM and DM.

Editorial comments:

Although safety was not an issue, there was lack of efficacy of Infliximab for IIM; however, there were fewer subjects enrolled than anticipated in the original trial design, decreasing the study’s statistical power to detect a difference in its main outcome measure, MMT, between the groups. Additionally, three of six patients withdrew from the infliximab arm at week 16, seemingly choosing not to be escalated to the higher dose of 7.5 mg. The most durable and robust response appeared occur in this patient.  This data does appear to suggest a dose-response curve. Perhaps higher initial infliximab dosing, such as 10 mg/kg, would achieve a faster durable therapeutic response.

AddThis Social Bookmark Button