A Randomized, Double-Blind, Placebo-Controlled Trial of Infliximab in Refractory Polymyositis and Dermatomyositis

Abstract

Objective:

To investigate in a pilot study the safety and efficacy of infliximab in patients with refractory dermatomyositis (DM) and polymyositis (PM).

Methods:

Participants received infusions of either placebo or infliximab 5 mg/kg at 0, 2, 6 and 14 weeks in blinded manner. The primary outcome was a >15% Manual Muscle Strength (MMT) improvement at Week 16 compared to Week 0. The secondary outcome measures were improvement defined by the International Myositis Assessment and Clinical Studies Group (IMACS) criteria. At Week 16, responders in each arm had the option of either continuing the same treatment or changing to the non-responder treatment for that study arm. Non-responders in the 5 mg/kg infliximab arm were increased to infliximab 7.5 mg/kg for week 22, 30, 38. Non-responders in the placebo arm at week 16 received infliximab 5 mg/kg at weeks 16, 18, 22, 30, 38. Outcomes were reassessed at Week 40.

Results:

Twelve subjects completed the study to week 16. Six of the twelve subjects received infliximab treatment at the dose of 5 mg/kg with only one subject meeting the responder criteria at that dose. Of the remaining five subjects on infliximab, three crossed over to the infliximab 7.5 mg/kg dose. One of those three subjects responded. All six patients in the placebo arm crossed over to the 5 mg/kg dosing regimen after Week 16, and two of those responded to infliximab.

Conclusions:

Infliximab therapy for patients with refractory PM and DM was well tolerated and may benefit a subset of patients.

Introduction:

Polymyositis (PM) and dermatomyositis (DM) are chronic inflammatory disorders traditionally characterized by progressive proximal muscle weakness. Glucocorticoids combined with other immunomodulating agents, commonly methotrexate (MTX), azathioprine (AZA) and/or intravenous immunoglobulin (IVIG), are mainstays of treatment for these potentially life-threatening diseases. Currently available therapy often results in a suboptimal response and/or toxicity, thus highlighting the need for new safe and effective treatments.

TNF-α is a potential therapeutic target in PM and DM, as it has been shown to accelerate skeletal muscle catabolism, cause contractile dysfunction and disrupt myogenesis (1, 2, 3). Published case reports and small series indicated potential benefits of TNF-α antagonism in PM and DM, such as improved muscle strength and decreased levels of muscle-associated enzymes (4). In addition, there is also strong evidence for the pathogenic role of TNF-α expression in PM and DM. A rabbit model of myositis has shown that during the inflammatory process, TNF-α mRNA levels were elevated in the inflammatory infiltrates in the effected muscle (5). In-situ hybridization of muscle biopsy tissue from PM and DM patients has also shown an increased expression of TNF-α in degenerating tissue(6). Other studies have also noted the increased expression of TNF-receptor and TNF-α in DM and PM sera as well as in muscle biopsy samples (7-15). Studies of the cytokine signatures in whole blood transcripts from a population of patients with PM and DM showed a correlation with type I interferon signatures and TNF-α (16). Disease course and susceptibility have also been associated with mutations affecting TNF -α. In patients with juvenile dermatomyositis (JDM), DM, and PM, the A- G polymorphism in the tumor necrosis alpha (TNF-α) −308 promoter region (TNF-α −308 A) is associated with prolonged disease course and increased production of TNF -α by peripheral blood mononuclear cells(17, 18).

Thus, the increased presence of TNF-α at both the protein and mRNA level in target tissues, as well as the increase in serum levels of soluble TNF receptor, imply a role for TNF-α in the pathogenesis of PM and DM.

This study evaluated whether treatment with infliximab, a chimeric IgG1 kappa monoclonal antibody against TNF-α, would increase muscle strength in patients with DM and PM who may have partially responded to prior therapy, but continue to have evidence of active disease. This pilot study also attempted to assess the safety and efficacy of infliximab in patients with DM and PM receiving corticosteroids and MTX or AZA, as well as, to determine the effective dose of infliximab.

Study Subjects and Methods

Study Subjects.

Study subjects were patients recruited from the National Institutes of Health (NIH) Clinical Center over a period of 4 years from September 2003 to April 2007. Inclusion criteria were: Age >18 years old; probable or definite PM or DM by Bohan and Peter criteria and proposed International Myositis Classification Criteria Project (IMCCP) criteria, proximal muscle weakness and no signs of distal muscle weakness (19-21). Patients were required to have proximal manual muscle test (MMT) scores between 80-120 (out of 160, from 8 bilateral muscle groups, scored on a 0-10 point scale)(22), and active myositis (evidenced by weakness, elevated CK, LDH , aldolase or MRI findings consistent with inflammation) despite adequate treatment with at least one disease-modifying antirheumatic drug. A muscle biopsy was either performed or the slides of previous biopsies were reviewed to assure that all subjects had muscle pathology consistent with PM or DM that the muscle biopsies did not show evidence of necrotic muscle fibers or rimmed vacuoles. Patients were on at least 4 weeks of a stable dose of prednisone (<0.5 mg/kg/d) and methotrexate (>7.5 mg/kg/week) or azathioprine (100-150 mg/day). Subjects were not allowed to escalate their therapy during the study period.

Patients were excluded if they were determined to have cytopenias, history of chronic or serious infections, history of clotting/bleeding disorders, inclusion body myositis, malignancy-related myositis, drug-induced myopathy, systemic lupus erythematosus, cardiac abnormalities or had previously received anti-TNF therapies.

Informed consents were obtained from all enrolled subjects. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by National Institute of Diabetes and Digestive and Kidney Diseases/National Institute of Arthritis, Musculoskeletal and Skin Diseases Institutional Review Board (IRB).

Study Design.

The study was a randomized, double-blind, placebo-controlled crossover study. Study participants and care providers were blinded to the intervention. Participants were randomized by pharmacy staff using a random number table in blocks of 4, with 2 each receiving either drug or placebo. As illustrated in Figure 1, patients were randomized to receive either placebo or infliximab (Remicade; Centocor, Malvern, PA) 5 mg/kg. The study was conducted in two phases where the outcomes were analyzed at Week 16 and Week 40. In phase I of the study infusions were given at Weeks 0, 2, 6 and 14 in a double-blinded fashion.

Figure 1:

Patients were randomized to receive 4 infusions of placebo vs treatment (infliximab). Response was defined as ≥ 15% improvement in strength based on the Manual Muscle Test (MMT*). Outcomes were assessed at 16 and 40 weeks. Responders were continued on 5mg/kg of infliximab. Non- responders were started on infliximab (5 mg/kg) in the placebo arm and increased to 7.5 mg/kg of infliximab in the treatment arm.

*The MMT-8 score is a summative value with a scoring range of 0 to 150. The 8 muscle groups included in the MMT-8 were tested using a 10-point scale. Muscle groups tested bilaterally were the shoulder abductors, elbow flexors, wrist extensors, knee extensors, ankle dorsiflexors, hip abductors, and hip extensors. Muscle groups tested unilaterally were the neck flexors.

In the second phase of the study, based on assessment of the primary outcome (MMT improvement by ≥ 15% at 16 weeks), patients were assigned into new groups. Responders continued in their original treatment allocation. Non-responders in the placebo arm were given 5 mg/kg of infliximab in an open-label fashion at 16, 18, 22, 30 and 38 weeks. The doses given to non-responders at weeks 16 and 18 were part of the induction therapy. Non-responders in the infliximab arm (5mg/kg body weight of infliximab) who failed to meet improvement criteria by Week 16 received a higher infliximab dose of 7.5 mg/kg in an open label fashion at weeks 22, 30 and 38. Safety assessments were performed at six-week intervals to evaluate long-term drug effects.

Study Assessments, Procedures and Endpoints.

Screening and baseline assessments included standard laboratory testing (complete blood counts with differential, chemistries, prothrombin time/partial thromboplastin time, chest x-ray, electrocardiogram, purified protein derivative and aldolase). For each potential patient, a muscle biopsy was either performed or the slides of previous biopsies were reviewed by an Armed Forces Institute of Pathology (AFIP) pathologist with expertise in muscle biopsies to assure that all subjects had muscle pathology consistent with PM or DM. Additional outcomes that were evaluated include Patient and Physician global disease activity, muscle associated enzymes - creatinine kinase, LDH, AST, ALT, aldolase, CLINHAQ, and MRI short tau inversion recovery images.

The Myositis Disease Activity Assessment Tool (MDAAT https://www.niehs.nih.gov/research/resources/imacs/diseaseactivity/) was also completed on participants (23).

At each visit, the patients underwent MMT testing (by a licensed physical therapist). Two physical therapists with experience in examining Idiopathic Inflammatory Myopathy (IIM) patients (24) administered the MMT using standardized instructions, positioning and stabilization techniques established by Kendall et al(25). An 8 hour in person MMT training tutorial was provided by F.P. Kendall and attended by both examiners prior to data collection. The examiners were blinded to each other. They performed MMT on consecutive days for 7 study participants with a resulting ICC (2,1) of 0.90. The Kendall 10-point MMT was selected because inter- rater reliability was established in myositis (26).

The primary outcome (MMT > 15% improvement) and secondary outcome (International Myositis Assessment and Clinical Studies Definition of Improvement) [IMACS DOI]) were assessed at weeks 16 and 40(22). The study was terminated early due to changes of standard of care suggesting additional agents be tried prior to infliximab.

Statistical Analysis

SAS enterprise version 4.3 was used for statistical analysis. Fisher’s exact test was performed on outcome results to assess differences in the placebo and infliximab groups between responders and non-responders. Chi square test was performed on IMACS data at week 40.

Initial study was powered at 25 people, assuming 80% power with 2- sided alpha of 0.05 to detect 15% change from baseline MMT for 5mg/kg treatment group and zero change in placebo group. The power calculation from post hoc analysis is 16.9.

Results

Baseline characteristics, patient disposition and concomitant therapy.

Eighteen subjects were screened for eligibility, of which 13 qualified. There were six subjects allocated to the placebo arm and seven subjects to the infliximab arm. One withdrew very early due to partial response. Twelve subjects completed the study to week 16. Two patients withdrew from the infliximab arm after week 16 due to partial or inadequate response. There was one death from a cause unrelated to the study drug and therefore 9 patients were included in the analysis at week 40.

The randomization process resulted in similar demographic, laboratory and clinical features in the two treatment groups (Table 1). There were 11 subjects with polymyositis and 1 subject with dermatomyositis. There were no significant differences between the treatment and placebo groups in the primary outcome and secondary outcome measures. Only two patients with polymyositis tested positive for Jo-1 antibody and one of them was a responder.

Table 1.

Baseline patient characteristics by treatment group.

CHARACTERISTIC	PLACEBO (N = 6)	INFLIXIMAB (5 mg/kg) (N = 7)a	p value
Female	5 (83.3)	6 (100)	0.70
N (%)
Age (years)	41.8 ± 9.3	49.0 ± 11.3	0.35
Ethnicity (%)			0.61
African American	16.7	83.3
Caucasian	66.7	17.7
Asian	16.7	0
PM (%)*	83.3	100	0.70
DM (%)	16.7	0
Disease Duration (years)	3.8 ± 1.5	7.3 ± 5.1	0.14
Medication at baseline	14.1 ± 10.8	20.5 ± 10.7	0.33
CS (mean mg/kg/d)	16.2 ± 11.1	12.5 ± 14.0	0.62
[N=12]	0.6	2.2	0.59
MTX (mean mg/wk)
[N=9]
AZA (mean mg/kg/d)
[N=3]
MMT	104.2 ± 9.9	104.3 ± 1.3	0.98
CK	608 ± 428.3	1695 ± 1520.9	0.12
HAQ	1.10 ± 0.44	1.46 ± 0.40	0.17
Physician Global Activity Assessment	63.5 ± 20.3	74.7 ± 11.9	0.27
Patient Global Activity Assessment	62 6 ±13 1	73 6 ± 17 9	0.25

^a=Infliximab group had 7 patients, 46 y/o Caucasian woman with polymyositis withdrew from the study after the first visit due to worsening of the disease .

Abbreviations, PM= Polymyositis;

^*DM=Dermatomyositis; CS= corticosteroids; MTX= methotrexate; AZA= azathioprine; MMT= manual muscle strength testing; CK= creatinine kinase; HAQ = health assessment questionnaire

Efficacy

At 16 weeks, one African-American female randomized to infliximab and met improvement criteria with a MMT changing from 84 to 97. Overall, there was one responder and five non-responders in the infliximab group and no responders in the placebo group at 16 weeks (p=0.2963 with a R² of 0.2146). At Week 16, non-responders in the placebo group (N = 6) were started on infliximab 5 mg/kg (Figure 1). At Week 22, non-responders in the infliximab group (N = 5) were increased to a dose of 7.5 mg/kg, while the responders (N = 1) remained at 5 mg/kg. No significant difference was seen in Week 16 MMT scores between the placebo group and the infliximab group (104.2 ± 9.9 and 104.3 ± 17.3 respectively p = 0.98). At Week 40, three subjects met the response criteria: a Caucasian man with DM whose MMT increased from 102 to 130; a Caucasian woman with PM whose MMT increased from 127 to 148; and an African American woman with PM whose MMT increased from 104 to 131.

IMACS DOI

Patients met IMACS DOI if 3 of any 6 measures improved by > 20% with no more than 2 worsening by > 25% and MMT did not worsen. In the first phase of the study at week 16, one subject met the IMACS DOI, four patients had no change and one patient worsened in the placebo group. Three patients met the criteria for improvement, two subjects had no change and one subject became worse in the infliximab arm.

In the second phase of the study at 40 weeks, four patients improved and two patients had no change in the placebo to infliximab group. Two patients met improvement criteria, one patient had no change, and three patients worsened in the infliximab group (p = 0.435).

Adverse events.

A 68 -year old woman with polymyositis in the infliximab arm died 2 months after the last infusion. Her death was deemed unrelated to the study drug. The patient was withdrawn from the study at Week 16 due to lack of treatment efficacy.

Adverse events (AEs) occurring in the placebo arm prior to crossover were myalgias (5 events), dyspnea on exertion (3), back pain (2), dysphagia (1), abdominal pain (1) and urinary tract infection (1).

The AEs occurring in the infliximab arm were back pain (1), fatigue (8) and myalgia (5), and while in the placebo arm after crossover AEs were fatigue (8), myalgias (7), back pain (1) and edema (1). One infusion reaction occurred at the sixth infusion for the last patient to complete the protocol. That patient received acetaminophen and diphenhydramine in subsequent infusions and experienced no additional infusion reactions. Outside of this, no study patient had received premedication.

DISCUSSION

Although many pharmacologic agents have been utilized in the treatment of DM and PM, few have been adequately evaluated in controlled trials (2, 3). A significant number of patients do not fully respond to or have intolerable adverse effects from current medications such as corticosteroids, methotrexate, azathioprine, IVIG, mycophenolate mofetil, tacrolimus, cyclophosphamide, or rituximab. Thus, there is a clear need for additional agents for treating refractory PM and DM.

There is limited and variable experience with anti-TNF-α agents in the treatment of DM and PM. (27-29). There are case reports indicating rapid and sustained clinical response to infliximab in recalcitrant DM case with calcinosis (30). Case series, and small pilot studies, have suggested some patients may respond(31). In an open-label trial of infliximab plus methotrexate in new-onset DM and PM two of six patients showed improvement in MMT, patient physician visual analogue scale (VAS), physician VAS, and CK level. However, low incidence of drug naive patients and the high dropout rate, that resulted in premature cessation of the study, made analysis of the results difficult (32). A retrospective study looking at the role of anti-TNF agents in resistant DM and PM showed that patients with high CK levels at baseline tended to benefit from treatment. However, due to lack of standardized outcome measures and its retrospective design, interpretation of the study is difficult (33). Etanercept had earlier been found to improve muscle strength without change to CK levels in a case of refractory PM (34). Subsequently, a randomized, placebo controlled trial of 16 patients looking at etanercept (50 mg SQ weekly) revealed favorable steroidsparing effect of etanercept in DM. However, no improvement in function occurred in either group. The trial demonstrated failure to wean patients off prednisone in the placebo group. In the etanercept group, 5 of 11 patients were successfully weaned off prednisone (35).

In some studies, patients were found to have worsening of disease. An open-label study with infliximab (5mg/kg) over 14 weeks found an increase of muscle inflammation on MRI and an increase in type I IFN activity in DM and PM patients on infliximab treatment (36). Long-term follow up of two patients treated with infliximab (10mg/kg at 4 doses at 2 week intervals) for DM and PM showed an initial benefit after 12 weeks (31, 37). However, after 20 weeks, symptoms including myalgias, muscle weakness, and elevated CK began to occur (37).

This first randomized, double-blind placebo-controlled trial assessing the effect of infliximab in DM or PM resulted in no significant differences in the endpoints at any time during the initial 16- week study period, or during the later crossover periods. However, some patients showed improvements in MMT and other outcome variables at the 5 mg/kg and 7.5 mg/kg doses. Only one Jo- 1 positive patient with polymyositis showed a response out of two patients with this autoantibody. These numbers are too small to draw a conclusion that antibody status plays a role.

Attempts were made to assess possible predictors of response to infliximab, but the small sample sizes limited this effort. Nonetheless, there were no obvious associations with disease duration, severity, autoantibodies, organ involvement, diagnosis, age, sex, or race that could explain which patients would respond to infliximab at either dose.

In this study, several factors could have contributed to the lack of statistically significant improvement in MMT or the IMACS DOI. The study was underpowered due to having fewer participants then needed to fully power the study. The starting dose may have been too low for optimal effect. . Patients who only responded at higher doses of infliximab support this finding.

Infliximab may be increased up to 10mg/kg, as seen in Crohn’s disease. This pilot study was also underpowered to assess many factors that have been previously associated with poor responses in PM and DM like presence of interstitial lung disease or severity of muscle disease at initial presentation.

The use of anti-TNF-α agents in the treatment of DM and PM will continue to be debated. Both improvement in disease course and the development of flare has been reported. Some case reports describe the development of new onset myositis after use of an anti-TNF agent for another disease(38, 39).

This study demonstrated that some patients have an improvement with infliximab. However, studies with larger sample sizes are needed to investigate the role of anti- TNF therapy in patients with inflammatory myopathies as study size was a major limitation of our study. Future studies should consider using doses of infliximab around 7.5 mg/kg or higher as we demonstrated not only a higher response rate at the larger dose, but that some specific patients who did not respond at the lower dose, did respond at the higher dose. Some patients did respond after 16 weeks of active treatment suggesting this time frame may be adequate for future studies. Our study looked at outcome measures including IMACS and MMT. Future studies should also use the new ACR/EULAR consensus guidelines for definition of improvement.

This is the first randomized, double-blind, placebo controlled trial of infliximab in DM and PM patients and suggests that certain patients may respond to this therapy, however, more work is needed to identify its proper use.

Table 2.

Manual Muscle Test (MMT) and International Myositis Assessment and Clinical Studies Group (IMACS) changes seen in patients at 16 weeks.

Patient	Diagnosis	16 week placebo or drug	Baseline MMT	MMT at 16 weeks	% change in MMT at 16 weeks	IMCAS at 16 weeks	Anti-Jo1 antibody
1	PM	P	118	127	+7.63	N	Pos
2	PM	P	110	123	+11.82	I	Neg
3	DM	P	101	102	+0.90	N	Neg
4	PM	P	108	120	+11.11	N	Neg
5	PM	P	90	93	−3.33	N	Neg
6	PM	P	98	112	−14.28	I	Neg
7	PM	D	98	91	−7.14	N	Neg
8	PM	D	110	92	−16.36	N	Neg
9	PM	D	84	97	+15.48	I	Pos
10	PM	D	93	104	+11.83	I	Neg
11	PM	D	87	77	−11.49	N	Neg
12	PM	D	120	119	+0.83	I	Neg

Pos= Positive

Neg= Negative

PM= Polymyositis

DM=Dermatomyositis

P= Placebo

D= Drug

I =Met IMACS definition of improvement (3 of any 6 measures improved by > 20% with no more than 2 worsening by > 25% and MMT did not worsen)

N=No change or worsened by IMCAS definition of improvement

^*Infliximab dose: At week 40, placebo patients were started on 5 mg/kg infliximab and infliximab was increased to 7.5 mg/kg on the patients who were already on 5 mg/kg.

Table 3:

Manual Muscle Test (MMT) and International Myositis Assessment and Clinical Studies Group (IMACS) DOI changes seen in patients at 40 weeks.

Patient	Disease diagnosis	40 week normal or high dose*	MMT at 16 weeks	MMT at 40 weeks	% change in MMT at 40 weeks	IMCAS at 40 weeks
1	PM	L	127	148	+16.5	I
2	PM	L	123	121	+1.6	N
3	DM	L	102	130	+27.4	I
4	PM	L	120	130	+8.3	I
5	PM	L	93	98	+6.4	I
6	PM	L	112	119	+7.1	I
7	PM	H	91	Patient dropped out		N
8	PM	H	92	74	−19.5	N
9	PM	H	97	130	+23.7	I
10	PM	H	104	131	+25.9	I
11	PM	H	77	Patient dropped out		N
12	PM	H	119	124	+4.2	N

L= initial dose at 5 mg/kg, H- High dose at 7.5 mg/kg

PM= Polymyositis, DM=Dermatomyositis

I =Met IMACS definition of improvement , N=No change or worsened by IMCAS definition of improvement

Table 4.

IMACS DOI components at 16 weeks

Patients	Drug or Placebo	Patient Global	Physician Global	CK (20%)	CLIN HAQ	MMT	Sum Analog Scale	Outcome
1	P	N	N	N	I	N	I	N
2	P	N	N	I	I	N	Y	I
3	P	W	N	W	W	N	N	W
4	P	N	N	N	I	N	W	N
5	P	W	N	I	N	N	N	N
6	P	N	I	I	I	N	N	I
7	L	N	N	N	No data	N	No data	N
8	L	N	N	N	N	N	No data	N
9	L	I	I	I	I	N	I	I
10	L	I	I	I	I	I	I	I
11	L	W	N	W	W	N	W	W
12	L	I	N	W	I	N	N	N

P=Placebo

L = Initial dose of Infliximab at 5 mg/kg

I= Improved (improved by > 20%)

W= Worse (worsened by > 25%)

N= No change (Did not reach criteria for improvement or worsening)

Table 5.

IMACS DOI components from 16 to 40 weeks

Patients	Normal or High Dose	Patient Global	Physician Global	CK (20%)	CLIN HAQ	MMT	Sum Analog Scales	Outcome
1	L	I	I	W	I	N	I	I
2	L	W	W	W	W	N	W	W
3	L	N	N	W	I	N	N	N
4	L	I	I	I	I	I	I	I
5	L	I	I	I	N	N	I	I
6	L	I	N	W	I	N	N	N
7	H	I	N	I	I	N	I	I
8	H	I	N	I	I	N	I	I
9	H	N	N	N	No data	N	No data	N
10	H	N	W	I	N	N	N	N
11	H	I	I	I	I	N	I	I
12	H	I	I	I	I	N	No data	I

L=Infliximab at initial dose 5 mg/kg

H= High dose Infliximab at 7.5 mg/kg

I= Improved (improved by > 20%)

W= Worse (worsened by > 25%)

N= No change (Did not reach criteria for improvement or worsening)