MDA5 antibody positive amyopathic dermatomyositis

Respiratory Medicine xxx (2015) 1e7

Contents lists available at ScienceDirect

Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed

Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM-140/MDA5 antibody positive amyopathic dermatomyositis Toshinori Takada a, b, *, Ami Aoki a, Katsuaki Asakawa a, Takuro Sakagami a, Hiroshi Moriyama a, Ichiei Narita a, Shinji Sato c a b c

Division of Respiratory Medicine, Niigata University Medical and Dental Hospital, Niigata, Japan Uonuma Institute of Community Medicine, Niigata University Medical and Dental Hospital, Niigata, Japan Division of Rheumatology, Department of Internal Medicine, Tokai University School of Medicine, Isehara, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 March 2015 Received in revised form 28 June 2015 Accepted 6 July 2015 Available online xxx

Background: Patients with amyopathic dermatomyositis (ADM) sometimes develop rapidly progressive interstitial lung disease (ILD) predominantly in Asia. Although anti-CADM-140/MDA5 antibody titer could correlate with disease activity and predict the course of ILD associated with ADM, it is not clear how this antibody is involved in the pathogenesis of ILD in ADM. Methods: We retrospectively collected clinical records and preserved serum before treatment of consecutive patients with ADM-ILD treated in the Niigata University Medical and Dental Hospital since 2000. We measured anti-CADM-140/MDA5 antibody titer and compared it between survivors and nonsurvivors. Serum cytokine/growth factor protein concentration was measured using a multiplex immunoassay system. The associations between anti-CADM-140/MDA5 antibody titer and each cytokine/ growth factor protein concentration were evaluated. Results: Thirteen patients were enrolled into the study. Among them, four patients did not respond to intensive immunosuppressive therapy and died. The mean anti-CADM-140/MDA5 antibody titer was significantly higher in patients who did not responded to therapy than in those who survived (p < 0.05). Relationship analyses between the antibody titer and each cytokine/GF protein concentration revealed that Spearman's rank correlation coefficients were more than 0.4 in thirteen cytokine/GF proteins. In particular, the strongest correlation was found between anti-CADM-140/MDA5 antibody titer and CX3CL1 (r ¼ 0.8897). Conclusions: These results confirmed that anti-CADM-140/MDA5 antibody levels could predict outcomes of ADM-ILD. Relationship analyses suggested that CX3CL1 might be involved in the pathogenesis of antiCADM-140/MDA5 antibody positive ADM-ILD. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Dermatomyositis Interstitial lung disease Cytokine Chemokine CX3CL1

Abbreviations: Ab, antibody; ADM, amyopathic dermatomyositis; CT, computed tomography; CX3CL1, chemokine (C-X3-C motif) ligand 1; EGF, epidermal growth factor; FGF-2, fibroblast growth factors-2; Flt-3, FMS-like tyrosine kinase-3; G-CSF, granulocyte-colony stimulating factor; GM-CSF, granulocyte macrophage-colony stimulating factor; GRO, growth related oncogene; IFN, interferon; IL, interleukin; ILD, interstitial lung disease; IP-10, IFN-g inducible protein 10; IPF, idiopathic pulmonary fibrosis; IVCY, intravenous cyclophosphamide; MCP-1, monocyte chemotactic protein-1; MDA-5, melanoma differentiation-associated gene 5; MDC, macrophage-derived chemokine; MIP1a, macrophage inflammatory protein-1a; MMF, mycofenolate mofetil; PM, polymyositis; DM, dermatomyositis; PMX-DHP, direct hemoperfusion with polymyxin Bimmobilized fiber; RLR, RIG-I-like receptor; SSc, systemic sclerosis; TGF-a, transforming growth factor-a; TNF-a, tumor necrosis factor-a; UIP, usual interstitial pneumonia; VEGF, vascular endothelial growth factor. * Corresponding author. Uonuma Institute of Community Medicine, Niigata University Medical and Dental Hospital, 1-754 Asahimachi-dori, Niigata 951-8520, Japan. E-mail address: [email protected] (T. Takada). http://dx.doi.org/10.1016/j.rmed.2015.07.004 0954-6111/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

2

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

1. Introduction Polymyositis and dermatomyositis (PM-DM) are forms of idiopathic inflammatory myositis. DM is identified by a characteristic rash accompanying muscle weakness, whereas, PM is defined as a subacute myopathy without the skin rash seen in DM. In addition, if a patient has the typical DM rash but no or little muscle weakness, the clinical diagnosis is amyopathic DM (ADM) [1]. Interstitial lung disease (ILD) in PM-DM is increasingly recognized as a serious complication and a major cause of death in this disease [2]. Patients with ADM sometimes develop rapidly progressive ILD, which is often resistant to intensive therapy including high dose corticosteroids and immunosuppressive agents, resulting in fatal respiratory failure [3]. Rapidly progressive ILD in ADM has been reported predominantly in Asia, including Japan, Hong Kong, Taiwan, and Korea [4e7]. An autoantibody which was associated with ADM and rapidly progressive ILD with significantly higher frequency was identified and termed anti-CADM-140 antibody [8]. Since the antibody recognizes an antigen of an RNA helicase encoded by melanoma differentiation-associated gene 5 (MDA-5), the antibody is called anti-CADM-140/MDA5 antibody at present [9]. Several reports demonstrated that the antibody titers correlated with disease activity and predicted the course of ILD associated with ADM [10e15]. The antibody may be also a diagnostic and predictive marker for rapidly progressive ILD associated with juvenile DM [16e18]. In non-Japanese cohorts, anti-CADM-140/MDA5 antibody is frequently found in patients with classic DM or severe vasculopathy [19,20]. MDA5 is part of the RIG-I-like receptor (RLR) family and functions as a pattern recognition receptor that typically recognizes dsRNA that is over 2000nts in length. Cytoplasmic pattern recognition receptors recognize the RNA of internalized viruses and mediate production of type-1 IFNs and antiviral immune responses. MDA5 in humans is encoded by the IFIH1 gene. A recent study showed that gain-of-function mutations in IFIH1 lead to upregulated type-1 IFN responses [21]. Individuals with these mutations exhibit phenotypes consistent with autoimmune diseases,
res syndrome and systemic lupus erytheincluding Aicardi-Goutie matosus. Although anti-CADM-140/MDA5 antibody recognizes MDA-5, how this antibody plays a part in the pathogenesis of ILD associated with ADM is still unknown. In order to clarify the role of the antibody in the pathogenesis of the disease, we analyzed cytokine/growth factor protein profiles in the serum before treatment from patients with ADM-ILD using Luminex bead technology.

clinical data including the results of anti Jo1 antibody and other autoantibodies according to the Guidelines for Epidemiological Studies from The Ministry of Health, Labour and Welfare, Japan. The Committee of Ethics, Niigata University, approved the study protocol (2013, #1650) and all living participants provided written consent. 2.2. Anti-CADM-140/MDA5 antibody titer measurement An ELISA for anti-CADM-140/MDA5 using recombinant MDA5 established previously was used to measure the titer of anti-CADM140/MDA5 antibodies in preserved serum available before treatment of the patients with ADM-ILD [9]. Briefly, 96-well polyvinyl plates (Sumilon multi-well plates, H type; Sumitomo Bakelite, Tokyo, Japan) were coated with purified recombinant MDA5 protein dissolved in phosphate-buffered saline (PBS, 0.5 mg/mL) at 4  C for 12 h, followed by incubation with patients' sera diluted 1:250. All samples were examined in duplicate, and the antibody units were calculated from optical density at 450 nm by reference to a standard curve constructed using serial concentrations of a serum sample containing a high titer of the anti-CADM-140/MDA5 antibody. 2.3. Cytokine/growth factor protein level measurement A panel of cytokines was measured using the Milliplex Map Human Cytokine/Chemokine Kit (Merck Millipore, Darmstadt, Germany), in accordance with the instructions of the manufacturer. All samples were diluted by the addition of an equal amount of saline, and 15 mL of the diluted samples were used for this assay. The assay is a novel multiplexed, particle-based, flow-cytometric assay named Luminex systems (Luminex Corporation, Austin, TX, USA) which utilizes anti-cytokine monoclonal antibodies linked to microspheres incorporating distinct proportions of two fluorescent dyes. The assay was customized to detect and quantify epidermal growth factor (EGF), eotaxin, fibroblast growth factors-2 (FGF-2), FMS-like tyrosine kinase-3 (Flt-3) ligand, chemokine (C-X3-C motif) ligand 1 (CX3CL1), granulocyte-colony stimulating factor (G-CSF), granulocyte macrophage-CSF (GM-CSF), growth related oncogene (GRO), interferon-a2 (IFN-a2), IFN-g, interleukin-1 a (IL1a), IL-1b, IL-1ra, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-15, IL-17A, IFN-g inducible protein 10 (IP-10), monocyte chemotactic protein-1 (MCP-1), MCP-3, macrophage-derived chemokine (MDC), macrophage inflammatory protein-1a (MIP-1a), MIP-1b, sCD40L, sIL-2Ra, transforming growth factor-a (TGF-a), tumor necrosis factor-a (TNF-a), TNF-b, and vascular endothelial growth factor (VEGF).

2. Materials and methods 2.4. Statistical analysis 2.1. Patients We retrospectively collected clinical records and preserved serum available before treatment of the consecutive patients with ADM-ILD treated in the Niigata University Medical and Dental Hospital from 2000 to 2013. ADM was diagnosed by characteristic rashes of DM confirmed by dermatologists, within mildly elevated creatinine kinase levels (less than 700 IU/L), and normal or only mildly reduced muscle strength consistent with age, sex, and severity of systemic illness. ILD was diagnosed by clinical findings as follows: fine crackles, exertional dyspnea, nonproductive cough, and reticular shadow on chest radiographs or ground-glass opacity on chest high-resolution computed tomography (CT). Patients who also exhibited other connective tissue diseases, malignancies and infections, who had been treated in other hospitals with high-dose corticosteroids, or whose preserved serum before treatment were not available, were excluded from this study. We collected the

Differences in the levels of anti-CADM-140/MDA5 antibody between patients groups were performed using ManneWhitney U test. Spearman's rank correlation coefficient was used to examine the relationship between the titer of anti-CADM-140/MDA5 antibody and the concentration of each cytokine/growth factor protein in the patient serum. The difference between groups was considered significant when P < 0.05. 3. Results 3.1. Clinical features and comparison of anti-CADM-140/MDA5 antibody titer between survivors and dead patients Thirteen patients with ADM-ILD consisting of four males and nine females with the mean age of 53.7 years old were enrolled into the study (Table 1).

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

3

In most cases, respiratory symptoms and/or signs developed in one month before the diagnosis. Anti-Jo1 antibody was not detected in any cases. Anti-Scl-70 antibody and anti-SS-A antibody were positive in two patients, but they did not fulfill diagnostic criteria of €gren's syndrome. All of them were systemic sclerosis (SSc) or Sjo treated with intensive immunosuppressive therapy including steroid pulse therapy followed by high-dose corticosteroids and oral cyclosporin. Intravenous cyclophosphamide (IVCY), mycofenolate mofetil (MMF), and direct hemoperfusion with polymyxin Bimmobilized fiber (PMX-DHP) were also added empirically depending on personal experience and the relative efficacy/safety ratio. Among them, 4 patients did not respond to the therapy and died. The mean anti-CADM-140/MDA5 antibody titer before treatment was significantly higher in patients who did not responded to therapy than in those who survived (p < 0.05, Fig. 1). 3.2. Relationship between anti-CADM140/MDA antibody titer and each cytokine/growth factor protein We then analyzed the relationship between the anti-CADM140/MDA5 antibody titer and each cytokine/GF protein concentration measured by the Milliplex Map Human Cytokine/Chemokine Kit. Case 1 was excluded from the analysis for lack of any more serum sample before treatment to measure cytokine/GF protein concentration. We did not evaluate the relationship when the cytokine/GF protein concentration in more than four samples were out of measurement ranges; the relationships between the antiCADM-140/MDA5 antibody titer and seven cytokine/GF protein concentrations were not evaluated. Of 32 cytokine/GF proteins, relationship analyses between the antibody titer and each cytokine/GF protein concentration revealed that Spearman's rank correlation coefficients were more than 0.4 in thirteen cytokine/GF proteins (Table 2). In particular, very strong correlations were found between antiCADM-140/MDA5 antibody titer and CX3CL1, and TGF-a; r ¼ 0.8897 and 0.7110, respectively (Fig. 2). We then compared the concentration of each cytokine/GF protein, but did not find significant differences between the survivors and the dead group. 3.3. Changes of anti-CADM140/MDA antibody titer and cytokine/ growth factor protein concentrations over the treatment duration Anti-CADM140/MDA antibody titer and concentrations of cytokine/growth factor protein were serially monitored over the

Fig. 1. Comparison of anti-CADM-140/MDA5 antibody titer before initiation of treatment between survivor and dead patients group. The mean titer of anti-CADM-140/ MDA5 antibody before treatment was significantly lower in those who survived (survivor group, n ¼ 9) than in those who died (the dead group, n ¼ 4).

treatment duration in two patients (case 9 and 12 in Table 1) with ADM who developed RP-ILD. Both of the patients had similarly elevated anti-CADM-140/MDA5 autoantibody levels, 182.4 and 149.9 units before treatment. Case 9 was treated by steroid pulse therapy followed by high-dose corticosteroids, oral cyclosporin, and MMF, but the lung disease finally progressed to be fatal (Fig. 3). Case 12 was treated by high-dose corticosteroids, oral cyclosporin, IVCY and PMX-DHP added on the first day of pulse therapy. His respiratory function was deteriorated in the first seven days, but the patient finally survived in about 10 weeks (Fig. 4). Tables 3 and 4 show the titer of anti-CADM140/MDA antibody and cytokine/growth factor protein concentrations which were correlated with the antibody with Spearman's rank coefficients more than 0.4 over the treatment duration. Although the antiCADM140/MDA antibody titer in both cases decreased gradually during the treatment, the concentrations of CX3CL1, IL-1a, IL-10, and IP-10 increased in the dead patient (Table 3). Whereas, of these cytokine/growth factor proteins, the concentrations of IL-10 and IP-10 decreased in the survived patient (Table 4).

Table 1 Clinical characteristics of the patients with interstitial lung disease associated with amyopathic dermatomyositis. Case

1 2 3 4 5 6 7 8 9 10 11 12 13

Age, yr

78 31 63 56 64 65 49 29 72 65 28 52 46

Sex

F M F F F F F F M F M M F

Symptoms, Moa Lung

Skin

1 1 1 2 1 1 1 1 1 1 1 1 1

2 4 13 5 4 44 3 4 1 3 2 4 5

Anti-Jo1 Ab (unit)

e 5.0 <3.0 <3.0 <5.0 <5.0 <5.0 <5.0 <5.0 <5.0 <5.0 <5.0 <5.0

Anti-CADM140/MDA5 Ab (unit)

32.6 28.0 90.2 49.2 294.1 160.4 75.1 91.1 182.4 49.8 23.4 149.9 31.5

Other auto Ab

Scl-70

SS-A

Treatment CS

CyA

þ þ þ þ þ þ þ þ þ þ þ þ þ

þ þ þ þ þ þ þ þ þ þ þ þ þ

Outcome Others

IVCY IVCY IVCY IVCY MMF MMF PMX IVCY,PMX

Alive Alive Dead Alive Dead Dead Alive Alive Dead Alive Alive Alive Alive

Ab, antibody; CS, Corticosteroids; CyA, Cyclosporin; IVCY, intravenous cyclophosphamide; MMF, mycofenolate mofetil; PMX, direct hemoperfusion with polymyxin Bimmobilized fiber. a Duration (months) of symptoms of each organ, lung or skin, before the diagnosis of amyopathic dermatomyositis was shown.

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

4

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

Table 2 Relationship between anti-CADM140/MDA antibody titer and each cytokine/growth factor protein. Cytokine/GF protein

Coefficient

p value

Cytokine/GF protein

Coefficient

p value

EGF Eotaxin FGF-2 Flt-3 ligand CX3CL1 G-CSF GM-CSF GRO IFN-a2 IFN-g IL-1a IL-1b IL-1ra IL-2 IL-3 IL-4 IL-5 IL-6 IL-7 IL-8

0.1678 0.2797 0.3077 0.1175 0.8897 0.5734 0.6294 0.0350 0.2242 0.5874 0.6643 NA 0.2102 0.4155 NA 0.6025 0.0459 0.1888 0.3853 0.2098

0.6021 0.3786 0.3306 0.7162 0.0001 0.0513 0.0283 0.9141 0.4837 0.0446 0.0185 NA 0.5121 0.1792 NA 0.0382 0.8875 0.5567 0.2161 0.5128

IL-9 IL-10 IL-12 (p40) IL-12 (p70) IL-13 IL-15 IL-17A IP-10 MCP-1 MCP-3 MDC MIP-1a MIP-1b sCD40L sIL-2Ra TGF-a TNF-a TNF-b VEGF

NA 0.5385 NA 0.4476 0.3433 0.2697 0.0420 0.4336 0.3776 0.3916 0.2587 0.4126 0.0420 NA NA 0.7110 0.2727 NA 0.6713

NA 0.0709 NA 0.1446 0.2747 0.3966 0.8970 0.1591 0.2262 0.2081 0.4168 0.1826 0.8970 NA NA 0.0095 0.3911 NA 0.0168

GF, growth factor; EGF, Epidermal Growth Factor; FGF, Fibroblast growth factors; FLT, FMS-like tyrosine kinase; CX3CL1, chemokine (C-X3-C motif) ligand 1; G-CSF, granulocyte-colony stimulating factor; GM-CSF, Granulocyte Macrophage colony-stimulating Factor; GRO, Growth Related Oncogene; IFN, Interferon; IL Interleukin; IP-10, IFN-g inducible protein 10; MCP, Monocyte Chemotactic Protein; MDC, macrophage-derived chemokine; MIP, macrophage inflammatory protein; TGF, transforming growth factor; TNF, tumor Necrosis Factor; VEGF, vascular endothelial growth factor; NA, not applicable.

Fig. 2. Scatterplots showing the titer of anti-CADM-140/MDA5 antibody and the concentration of CX3CL1 (A) or TGFa (B) before initiation of treatment. Both panels show positive correlation: r ¼ 0.8897 and 0.7110.

4. Discussion DM is an uncommon inflammatory disease marked by muscle weakness and a distinctive skin rash [22]. ADM is a form of DM in which patients have characteristic skin findings of DM with little or no evidence of myositis [1]. Sato et al. identified an autoantibody, which recognized a 140kD antigen of an RNA helicase encoded by MDA-5 and was almost specifically found in the patients with ADM [8,9]. Thus, the antibody was termed anti-CADM-140/MDA5 antibody. The pathogenesis of DM involves complex immunologic interactions of both the innate and adaptive systems. The majority of the current evidence suggests that the genes and proteins induced by IFN-Is consisting of IFNa and IFNb and potential mechanisms by which these downstream products relate to clinical disease activity [23]. Gono et al. evaluated the serum cytokine profiles of 38 patients with PM-DM by multiplex assay using the Milliplex MAP Human Cytokine/Chemokine Panel [24]. They concluded that IL-6,

IL-8, TNF-a and IP-10 are associated with disease activity in PMDM and that Serum IL-8 is a useful predictor for fatal outcomes due to ILD with PM-DM. Although the number of patients in the study was limited, relationship analyses between the antibody titer and each cytokine/GF protein concentration revealed that Spearman's rank correlation coefficients were more than 0.4 in thirteen cytokine/GF proteins. In particular, we found the strongest correlation between anti-CADM-140/MDA5 antibody titer and CX3CL1. MDA5 is a RIG-Ilike receptor dsRNA helicase enzyme encoded by the IFIH1 gene in humans. As part of the RIG-I-like receptor family, it functions as a pattern-recognition receptor recognizing dsRNA over 2000nts in length as a sensor for viruses [25]. Binding of viral RNA to patternrecognition receptors usually triggers signaling cascades that result in type I IFN production and the expression of inhibitory IFNstimulated genes to control virus replication and spread. Inappropriate or sustained activation of these signaling pathways, however,
res can cause autoimmune diseases such as Aicardi-Goutie

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

5

Table 3 Changes of anti-CADM140/MDA antibody titer and cytokine/growth factor protein concentrations of a dead patient. Values

PaO2/FiO2 ratio anti-CADM140/MDA Ab titer, Unit CX3CL1, pg/mL G-CSF, pg/mL GM-CSF, pg/mL IFN-g, pg/mL IL-1a, pg/mL IL-2, pg/mL IL-4, pg/mL IL-10, pg/mL IL-12 (p70), pg/mL IP-10, pg/mL MIP-1a, pg/mL TGFa, pg/mL VEGF, pg/mL

Fig. 3. Chest radiographs and computed tomography of a dead patient (case 9) with rapidly progressive interstitial lung disease associated with amyopathic dermatomyositis. Very slight interstitial lung disease taken on the day one (A, B) deteriorated in about four weeks despite the treatment of high dose corticosteroid, cyclosporin and mycofenolate mofetil (day 26, C, D).

Day 1

25

37

315 182.4 406.6 100.2 38.7 30.4 41.5 2.33 5.81 21.0 10.2 2144.7 362.5 18.0 500.1

226 92.7 3876.0 84.3 48.3 8.75 63.6 <0.64 4.97 22.0 36.4 7345.7 614.6 30.9 179.0

56 60.9 6638.9 640.0 28.9 9.86 75.7 <0.64 19.4 160.7 17.0 12292.7 127.7 8.1 166.9

Ab, antibody; CX3CL1, chemokine (C-X3-C motif) ligand 1; G-CSF, granulocytecolony stimulating factor; GM-CSF, Granulocyte Macrophage colony-stimulating Factor; IFN, Interferon; IL Interleukin; IP-10, IFN-g inducible protein 10; MIP, macrophage inflammatory protein; TGF, transforming growth factor; VEGF, vascular endothelial growth factor.

Table 4 Changes of anti-CADM140/MDA antibody titer and cytokine/growth factor protein concentrations of a survived patient. Values

PaO2/FiO2 ratio anti-CADM140/MDA Ab titer, units CX3CL1, pg/mL G-CSF, pg/mL GM-CSF, pg/mL IFN-g, pg/mL IL-1a, pg/mL IL-2, pg/mL IL-4, pg/mL IL-10, pg/mL IL-12 (p70), pg/mL IP-10, pg/mL MIP-1a, pg/mL TGFa, pg/mL VEGF, pg/mL

Day 1

8

67

251 149.9 287.8 66.3 63.5 44.1 179.3 21.0 67.0 35.4 58.4 4090.4 27.3 30.6 579.6

165 42.2 1037.8 110.6 302.5 74.9 224.8 117.0 342.4 23.2 388.8 2883.5 340.4 55.6 1280.1

422 35.9 106.3 52.6 31.8 15.5 18.0 3.13 7.54 11.7 10.3 1159.4 1601.2 18.4 256.8

Ab, antibody; CX3CL1, chemokine (C-X3-C motif) ligand 1; G-CSF, granulocytecolony stimulating factor; GM-CSF, Granulocyte Macrophage colony-stimulating Factor; IFN, Interferon; IL Interleukin; IP-10, IFN-g inducible protein 10; MIP, macrophage inflammatory protein; TGF, transforming growth factor; VEGF, vascular endothelial growth factor. Fig. 4. Chest radiographs and computed tomography of a survived patient (case 12) with high titer of anti-CADM140/MDA antibody. Subpleural consolidation mostly in lower lobes taken on the day one (A, B) improved in about ten weeks by the treatment of high dose corticosteroid, cyclosporin and intravenous cyclophosphamide with addition of direct hemoperfusion with polymyxin B-immobilized fiber on the first day of steroid pulse therapy (day 38, C, D).

syndrome and systemic lupus erythematosus [21,26]. Nakano et al. reported that type I IFN induced higher CX3CL1 mRNA expression and protein secretion in pulmonary arterial vascular endothelial cells than those in umbilical vein [27]. Aberrant function of MDA in the patients with ADM-ILD could cause elevated CX3CL1 expression in the serum. CX3CL1 is a large cytokine protein of 373 amino acids, which contains multiple domains and is the only known member of the CX3C chemokine family [28]. CX3CL1 is primarily expressed on endothelial cells and promotes strong adhesion of leukocytes to activated endothelial cells by interacting with the chemokine receptor CX3CR1 [29,30]. A soluble version of CX3CL1 has also been

observed, which potently chemoattracts T cells and monocytes [31]. CX3CL1 has been found to play an important role in rheu€gren's syndrome, sysmatic diseases; rheumatoid arthritis, Sjo temic lupus erythematosus, and SSc [32e36]. Suzuki et al. reported that serum soluble CX3CL1 was significantly higher in PM-DM patients than in healthy controls and was correlated with serum creatinine kinase levels, manual muscle testing scores, and alveolar-arterial oxygen pressure difference. They concluded that serum CX3CL1 level could be a surrogate marker of disease activity in PM-DM [37]. While CX3CL1 is involved in the recruitment of cells into tissues undergoing inflammatory responses by induction of both adhesion and migration of leukocytes, the role of CX3CL1 in the pathogenesis of pulmonary fibrosis is still unknown. In vitro, lung fibroblasts are a prominent cellular source of CX3CL1 in response to the combination of IL-1b and IFN-g [38]. The secretion and expression of lung fibroblast-derived CX3CL1 were markedly

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

6

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

reduced by specific inhibitors of the signal transducer and activator of transcription 1 transcription factor. Hasegawa et al. reported that soluble CX3CL1 levels were significantly raised in sera and were associated with severity of pulmonary fibrosis in the patients with SSc [39]. On the other hand, an analysis using a panel of 8 serum cytokines including CX3CL1 demonstrated that only serum IL-6 was an independent predictor of DLCO decline in both idiopathic pulmonary fibrosis (IPF) and SSc-ILD [40]. IPF and SSc-ILD are pathologically classified as a usual interstitial pneumonia (UIP) pattern. The natural history of UIP/IPF was thought to be a steady, gradual, and predictable decline in lung function over years. Although the most common finding in the pathologic appearance of ILD in DM is nonspecific interstitial pneumonia [41], rapidly progressive ILD in ADM is often associated with diffuse alveolar damage [4,42]. CX3CL1 might be involved in the pathogenesis of not UIP pattern but diffuse alveolar damage in ILD-ADM. Rapidly progressive ILD accompanied by ADM is occasionally resistant to aggressive immunosuppressive therapy to be fatal [4]. Patients with high titers of anti-CADM-140/MDA5 antibodies are reported to have a poorer prognosis than other patients with ADM [11,12,13]. In a previous report, four patients with ADM-ILD in the non-responder group died within three months after initiation of treatment with no significant reduction in the anti-CADM-140/ MDA5 antibody titer [11]. Although our results were derived from a case series of two patients, the anti-CADM-140/MDA5 antibody titer decreased gradually in both dead and survived patients. Interestingly, the concentrations of CX3CL1, IL-1a, IL-10, and IP-10 increased in the dead patient, whereas those of IL-10 and IP-10 decreased in the survived patient. These cytokines may be involved in the progression of anti-CADM-140/MDA5 antibody positive ADM-ILD. This study has several limitations. The patients were collected from a single center in a retrospective manner and enrolled in a study according to the availability of preserved serum before treatment. In the treatment of intensive immunosuppressive therapy, selection of immunosuppressive drugs was empirical and depended on physicians' experience and the relative efficacy/safety ratio. Cytokines which are not included in the Milliplex Map Human Cytokine/Chemokine Kit were not measured. Prospective studies with more patients are required to confirm our findings. In conclusion, the results of the present study reconfirmed that anti-CADM-140/MDA5 antibody levels could predict outcomes of ADM-ILD. Relationship analyses between the antibody titer and cytokine/GF protein concentrations suggested that CX3CL1 might be involved in the pathogenesis of anti-CADM-140/MDA5 antibody positive ADM-ILD. Author contributions TT had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis, including and especially any adverse effects. TT, AA, KA, TS, HM, IN, and SS contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript. Conflict of interest statement None of the authors has a financial relationship with a commercial entity with an interest in the subject of this manuscript. Acknowledgment This work was supported by JSPS KAKENHI Grant Number 25461151. The authors thank Dr. Takahiro Tanaka, Bioscience

Medical Research Center, Niigata University Medical and Dental Hospital, for useful discussion on statistical analyses.

References [1] M.C. Dalakas, R. Hohlfeld, Polymyositis and dermatomyositis, Lancet 362 (2003) 971e982 [PMID: 14511932. [2] I. Marie, P.Y. Hatron, S. Dominique, et al., Short-term and long-term outcomes of interstitial lung disease in polymyositis and dermatomyositis: a series of 107 patients, Arthritis Rheum. 63 (2011) 3439e3447 [PMID: 21702020 10.1002/art.30513]. [3] R.D. Sontheimer, S. Miyagawa, Potentially fatal interstitial lung disease can occur in clinically amyopathic dermatomyositis, J. Am. Acad. Dermatol. 48 (2003) 797e798 [PMID: 12734516. [4] T. Suda, T. Fujisawa, N. Enomoto, et al., Interstitial lung diseases associated with amyopathic dermatomyositis, Eur. Respir. J. 28 (2006) 1005e1012 [PMID: 16837503. [5] N. Sakamoto, H. Mukae, T. Fujii, et al., Nonspecific interstitial pneumonia with poor prognosis associated with amyopathic dermatomyositis, Intern. Med. 43 (2004) 838e842 [PMID: 15497521. [6] W.A. High, J.B. Cohen, B.A. Murphy, et al., Fatal interstitial pulmonary fibrosis in anti-Jo-1-negative amyopathic dermatomyositis, J. Am. Acad. Dermatol. 49 (2003) 295e298 [PMID: 12894081. [7] E.H. Kang, E.B. Lee, K.C. Shin, et al., Interstitial lung disease in patients with polymyositis, dermatomyositis and amyopathic dermatomyositis, Rheumatol. Oxf. 44 (2005) 1282e1286 [PMID: 15972351. [8] S. Sato, M. Hirakata, M. Kuwana, et al., Autoantibodies to a 140-kd polypeptide, CADM-140, in Japanese patients with clinically amyopathic dermatomyositis, Arthritis Rheum. 52 (2005) 1571e1576 [PMID: 15880816 10.1002/ art.21023]. [9] S. Sato, K. Hoshino, T. Satoh, et al., RNA helicase encoded by melanoma differentiation-associated gene 5 is a major autoantigen in patients with clinically amyopathic dermatomyositis: association with rapidly progressive interstitial lung disease, Arthritis Rheum. 60 (2009) 2193e2200 [PMID: 19565506 10.1002/art.24621]. [10] S. Sato, M. Kuwana, T. Fujita, et al., Amyopathic dermatomyositis developing rapidly progressive interstitial lung disease with elevation of anti-CADM-140/ MDA5 autoantibodies, Mod. Rheumatol. 22 (2012) 625e629 [PMID: 22124544 10.1007/s10165-011-0558-9]. [11] S. Sato, M. Kuwana, T. Fujita, et al., Anti-CADM-140/MDA5 autoantibody titer correlates with disease activity and predicts disease outcome in patients with dermatomyositis and rapidly progressive interstitial lung disease, Mod. Rheumatol. 23 (2013) 496e502 [PMID: 22644102 10.1007/s10165-012-06634]. [12] Y. Muro, K. Sugiura, K. Hoshino, et al., Disappearance of anti-MDA-5 autoantibodies in clinically amyopathic DM/interstitial lung disease during disease remission, Rheumatol. Oxf. 51 (2012) 800e804 [PMID: 22210662 10.1093/ rheumatology/ker408]. [13] T. Koga, K. Fujikawa, Y. Horai, et al., The diagnostic utility of anti-melanoma differentiation-associated gene 5 antibody testing for predicting the prognosis of Japanese patients with DM, Rheumatol. Oxf. 51 (2012) 1278e1284 [PMID: 22378718 10.1093/rheumatology/ker518]. [14] T. Gono, S. Sato, Y. Kawaguchi, et al., Anti-MDA5 antibody, ferritin and IL-18 are useful for the evaluation of response to treatment in interstitial lung disease with anti-MDA5 antibody-positive dermatomyositis, Rheumatol. Oxf. 51 (2012) 1563e1570 [PMID: 22589330 10.1093/rheumatology/kes102]. [15] H. Cao, M. Pan, Y. Kang, et al., Clinical manifestations of dermatomyositis and clinically amyopathic dermatomyositis patients with positive expression of anti-melanoma differentiation-associated gene 5 antibody, Arthritis Care Res. 64 (2012) 1602e1610 [PMID: 22623119 10.1002/acr.21728]. [16] N. Sakurai, K. Nagai, H. Tsutsumi, et al., Anti-CADM-140 antibody-positive juvenile dermatomyositis with rapidly progressive interstitial lung disease and cardiac involvement, J. Rheumatol. 38 (2011) 963e964 [PMID: 21532064 10.3899/jrheum.101220]. [17] I. Kobayashi, Y. Okura, M. Yamada, et al., Anti-melanoma differentiationassociated gene 5 antibody is a diagnostic and predictive marker for interstitial lung diseases associated with juvenile dermatomyositis, J. Pediatr. 158 (2011) 675e677 [PMID: 21232756 10.1016/j.jpeds.2010.11.033]. [18] S.L. Tansley, Z.E. Betteridge, H. Gunawardena, et al., Anti-MDA5 autoantibodies in juvenile dermatomyositis identify a distinct clinical phenotype: a prospective cohort study, Arthritis Res. Ther. 16 (2014) R138 [PMID: 24989778 10.1186/ar4600]. [19] D. Fiorentino, L. Chung, J. Zwerner, et al., The mucocutaneous and systemic phenotype of dermatomyositis patients with antibodies to MDA5 (CADM140): a retrospective study, J. Am. Acad. Dermatol. 65 (2011) 25e34 [PMID: 21531040 10.1016/j.jaad.2010.09.016]. [20] N.S. Narang, L. Casciola-Rosen, S. Li, et al., Cutaneous ulceration in dermatomyositis: association with anti-melanoma differentiation-associated gene 5 antibodies and interstitial lung disease, Arthritis Care Res. 67 (2015) 667e672 [PMID: 25331610 10.1002/acr.22498]. [21] G.I. Rice, Y. del Toro Duany, E.M. Jenkinson, et al., Gain-of-function mutations in IFIH1 cause a spectrum of human disease phenotypes associated with

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004

T. Takada et al. / Respiratory Medicine xxx (2015) 1e7

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

upregulated type I interferon signaling, Nat. Genet. 46 (2014) 503e509 [PMID: 24686847 10.1038/ng.2933]. A. Bohan, J.B. Peter, Polymyositis and dermatomyositis (first of two parts), N. Engl. J. Med. 292 (1975) 344e347 [PMID: 1090839. A. Arshanapalli, M. Shah, V. Veerula, et al., The role of type I interferons and other cytokines in dermatomyositis, Cytokine 73 (2015) 319e325 [PMID: 25541432 10.1016/j.cyto.2014.11.026]. T. Gono, H. Kaneko, Y. Kawaguchi, et al., Cytokine profiles in polymyositis and dermatomyositis complicated by rapidly progressive or chronic interstitial lung disease, Rheumatol. Oxf. 53 (2014) 2196e2203 [PMID: 24970922 10.1093/rheumatology/keu258]. H. Kato, O. Takeuchi, E. Mikamo-Satoh, et al., Length-dependent recognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-I and melanoma differentiation-associated gene 5, J. Exp. Med. 205 (2008) 1601e1610 [PMID: 18591409 10.1084/jem.20080091]. M. Funabiki, H. Kato, Y. Miyachi, et al., Autoimmune disorders associated with gain of function of the intracellular sensor MDA5, Immunity 40 (2014) 199e212 [PMID: 24530055 10.1016/j.immuni.2013.12.014]. M. Nakano, T. Fujii, M. Hashimoto, et al., Type I interferon induces CX3CL1 (fractalkine) and CCL5 (RANTES) production in human pulmonary vascular endothelial cells, Clin. Exp. Immunol. 170 (2012) 94e100 [PMID: 22943205 10.1111/j.1365-2249.2012.04638.x]. J.F. Bazan, K.B. Bacon, G. Hardiman, et al., A new class of membrane-bound chemokine with a CX3C motif, Nature 385 (1997) 640e644 [PMID: 9024663 10.1038/385640a0]. T. Imai, K. Hieshima, C. Haskell, et al., Identification and molecular characterization of fractalkine receptor CX3CR1, which mediates both leukocyte migration and adhesion, Cell 91 (1997) 521e530 [PMID: 9390561. H. Nomiyama, T. Imai, J. Kusuda, et al., Human chemokines fractalkine (SCYD1), MDC (SCYA22) and TARC (SCYA17) are clustered on chromosome 16q13, Cytogenet. Cell Genet. 81 (1998) 10e11 [PMID: 9691168. H. Umehara, E. Bloom, T. Okazaki, et al., Fractalkine and vascular injury, Trends Immunol. 22 (2001) 602e607 [PMID: 11698220. J.H. Ruth, M.V. Volin, G.K. Haines 3rd, et al., Fractalkine, a novel chemokine in rheumatoid arthritis and in rat adjuvant-induced arthritis, Arthritis Rheum. 44 (2001) 1568e1581 [PMID: 11465708 10.1002/1529-0131(200107)44: 7<1568::AID-ART280>3.0.CO;2-1]. T. Matsumiya, K. Ota, T. Imaizumi, et al., Characterization of synergistic induction of CX3CL1/fractalkine by TNF-alpha and IFN-gamma in vascular

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

7

endothelial cells: an essential role for TNF-alpha in post-transcriptional regulation of CX3CL1, J. Immunol. 184 (2010) 4205e4214 [PMID: 20231691 10.4049/jimmunol.0903212]. S.O. Moon, W. Kim, M.J. Sung, et al., Resveratrol suppresses tumor necrosis factor-alpha-induced fractalkine expression in endothelial cells, Mol. Pharmacol. 70 (2006) 112e119 [PMID: 16614140 10.1124/mol.106.022392]. T. Odai, M. Matsunawa, R. Takahashi, et al., Correlation of CX3CL1 and CX3CR1 levels with response to infliximab therapy in patients with rheumatoid arthritis, J. Rheumatol. 36 (2009) 1158e1165 [PMID: 19369458 10.3899/ jrheum.081074]. H. Okamoto, M. Yamamura, Y. Morita, et al., The synovial expression and serum levels of interleukin-6, interleukin-11, leukemia inhibitory factor, and oncostatin M in rheumatoid arthritis, Arthritis Rheum. 40 (1997) 1096e1105 [PMID: 9182920 10.1002/1529-0131(199706)40:6<1096::AIDART13>3.0.CO;2-D]. F. Suzuki, T. Kubota, Y. Miyazaki, et al., Serum level of soluble CX3CL1/fractalkine is elevated in patients with polymyositis and dermatomyositis, which is correlated with disease activity, Arthritis Res. Ther. 14 (2012) R48 [PMID: 22394569 10.1186/ar3761]. T. Isozaki, K. Otsuka, M. Sato, et al., Synergistic induction of CX3CL1 by interleukin-1beta and interferon-gamma in human lung fibroblasts: involvement of signal transducer and activator of transcription 1 signaling pathways, Transl. Res. J. Lab. Clin. Med. 157 (2011) 64e70 [PMID: 21256458 10.1016/j.trsl.2010.11.007]. M. Hasegawa, S. Sato, T. Echigo, et al., Up regulated expression of fractalkine/ CX3CL1 and CX3CR1 in patients with systemic sclerosis, Ann. Rheum. Dis. 64 (2005) 21e28 [PMID: 15608300 10.1136/ard.2003.018705]. A. De Lauretis, P. Sestini, P. Pantelidis, et al., Serum interleukin 6 is predictive of early functional decline and mortality in interstitial lung disease associated with systemic sclerosis, J. Rheumatol. 40 (2013) 435e446 [PMID: 23378460 10.3899/jrheum.120725]. W.W. Douglas, H.D. Tazelaar, T.E. Hartman, et al., Polymyositis-dermatomyositis-associated interstitial lung disease, Am. J. Respir. Crit. Care Med. 164 (2001) 1182e1185 [PMID:. Y. Matsuki, H. Yamashita, Y. Takahashi, et al., Diffuse alveolar damage in patients with dermatomyositis: a six-case series, Mod. Rheumatol. 22 (2012) 243e248 [PMID: 21904785 10.1007/s10165-011-0502-z].

Please cite this article in press as: T. Takada, et al., Serum cytokine profiles of patients with interstitial lung disease associated with anti-CADM140/MDA5 antibody positive amyopathic dermatomyositis, Respiratory Medicine (2015), http://dx.doi.org/10.1016/j.rmed.2015.07.004