Abstract
Tumour necrosis factor inhibitors (TNFis) can trigger autoantibody formation and, in a small proportion of patients, result in symptomatic autoimmune diseases, including polymyositis and dermatomyositis. Focal myositis presents with localized muscle pain and is diagnosed based on magnetic resonance imaging (MRI), electromyography (EMG) and muscle histopathology. We report the first case of TNFi-induced focal myositis in the left thigh of a patient with ulcerative colitis on infliximab. A 30-year-old lady presented with left thigh pain with no other systemic manifestation or muscle weakness. The immunology profile showed positivity of antinuclear antibody (ANA), high titre of anti-ds-DNA, and anti-Mi-2 without any features of systemic lupus erythematosus or dermatomyositis. Discontinuation of the TNFi and commencement of glucocorticoids resulted in the resolution of the myositis.
Introduction
Blockade of tumour necrosis factor (TNF) is used in a variety of autoimmune diseases including rheumatoid arthritis, inflammatory bowel disease (IBD) and ankylosing spondylitis. It is also a treatment option for refractory myositis. Paradoxically, tumour necrosis factor inhibitors (TNFis) can induce autoantibodies and rarely trigger autoimmune disease. We present a case of focal myositis in a patient treated with infliximab, a murine-human chimeric monoclonal antibody directed against TNF for IBD.
Case report
A lady in her 30s with ulcerative colitis since the age of 12 was started on 6-weekly infliximab infusions at a dose of 5 mg/kg for 2 years prior with concurrent azathioprine (AZA) 200 mg daily (2 mg/kg). There were no known extra-intestinal manifestations. At the time of presentation to the rheumatology department of a major tertiary referral hospital, she had subacute onset of atraumatic left anterolateral thigh pain. The pain was severe, rendering her unable to ambulate. No other limbs were involved and there were no constitutional symptoms. Of note, there were no clinical features of lupus, dermatomyositis or other connective tissue diseases. Physical examination revealed a markedly tender left thigh to palpation, especially over the lateral aspect without accompanying swelling or erythema. While there was a global restriction in left hip movement due to intense pain there was no evidence of proximal myopathy or neurological deficits.
Blood investigations showed a positive antinuclear antibody (ANA) in 1:640 with speckled and homogenous patterns, anti-double-stranded DNA (anti-ds-DNA) 254 IU/mL (normal, 0–27 IU/mL) and anti-Mi-2 antibodies (Immunoblot EUROLINE panel). Baseline ANA result prior to initiation of infliximab was not available. Other myositis specific antibodies (MSA), creatine kinase, inflammatory markers, complements, extractable nuclear antigens, anti-histone antibodies, blood cultures, infective serologies and renal and liver profiles were normal. Magnetic resonance imaging (MRI) of both thighs demonstrated abnormally increased signal intensity within the left tensor fascia lata, iliotibial band and gluteus minimus muscle and tendon in keeping with focal myositis (Figure 1). No hip joint pathology was noted. Muscle biopsy showed regenerating myofibres and foci of endomysial inflammation with lymphocytes and macrophages. No granulomatous inflammation and tissue cultures were negative. A diagnosis of TNFi-related focal myositis was made. MRI hip and thigh T2 fat-suppressed views (1a coronal, 1b axial) showing abnormally increased signal intensity within the left (*) iliotibial band, tensor fascia lata muscle and tendon and gluteus minimus muscle, in keeping with myositis.
Prednisolone at a dose of 25 mg daily was commenced with almost immediate improvement in her pain and mobility. Infliximab was ceased, and she was switched to vedolizumab, a colon-specific α4β7 integrin inhibitor infusion. Prednisone was continued for 5 months and ceased.
A follow-up thigh MRI after 5 months of prednisone showed a marked improvement in the muscle inflammation (Figure 2). A subsequent 8-month interval MRI following commencement of prednisone and 3 months following its cessation showed complete resolution of myositis. Repeated serologic testing showed persistence of the ANA (1:160) however, dsDNA and anti-Mi-2 had become negative. MRI hip and thigh T2 fat-suppressed views (1a coronal, 1b axial) showing resolution of previously seen signal intensity within the left (*) iliotibial band, tensor fascia lata and gluteus minimus muscle.
Discussion
Autoantibodies can develop in patients treated with TNFis, but these are usually not associated with clinical disease.1,2 The most common TNFi-induced autoantibodies are ANA; in up to 50% of patients and anti-ds-DNA in 15% of patients.1,2 Other antibodies associated with TNFi use that have been reported include PL-7, PL-12, Jo-1, anti-PM-Scl and U1-RNP. 1 Beigel et al. 2 looked at 180 patients with IBD treated with a TNFi, and ANA and anti-ds-DNA positivity were seen in 80 (44%) and 28 (15.6%) of patients respectively. 2 However, mild lupus-like syndrome was seen in 16 (8.9%) and only two patients (1.1%) had severe symptoms. 2
Tumour necrosis factor inhibition has rarely been associated with a range of autoimmune diseases including polymyositis, dermatomyositis, anti-synthetase syndrome, demyelinating disease, interstitial lung disease, uveitis, lupus-like syndromes, sarcoidosis, vasculitis, autoimmune hepatitis and antiphospholipid syndrome. 3
Polymyositis or dermatomyositis usually presents with progressive symmetrical proximal myopathy and elevated muscle enzymes, with little pain (with an exception in necrotizing myositis), and may be associated with systemic symptoms.
In contrast, focal myositis typically presents with localised muscle pain with or without swelling or erythema and without systemic symptoms or muscle weakness. 4 There is localised inflammatory swelling within a skeletal muscle or a single group of muscles predominantly in the lower limbs (i.e adductor muscle, vastus lateralis, groin and gastrocnemius), without systemic manifestations. 5 Focal myositis has also been reported in the neck, ocular and upper limb musculature.5,6 Common causes include autoimmune such as SLE, Behcet’s, Crohn’s disease, systemic sclerosis, ANCA vasculitis, sarcoidosis, tumours, infection, trauma, radiculopathy and idiopathic.4,5 However, TNFi-induced focal myositis has not been reported.
Diagnosis of focal myositis requires a combination of MRI showing evidence of muscle inflammation, myopathic changes on EMG and muscle biopsy for histological analysis. Histological findings include focal myopathic and neurogenic changes, fibrosis, inflammation and occasional presence of eosinophils. 5 Due to the focal nature of the disease, blood investigations including muscle enzymes are usually normal. The clinical course can be self-limiting, chronic or relapsing. 4 Treatments including methotrexate, azathioprine, glucocorticoids and intravenous immunoglobulin have been used. 4
Brunasso et al. reported a series of 20 patients with TNFi-induced dermatomyositis or polymyositis and anti-synthetase syndrome in patients with underlying rheumatoid arthritis, Crohn’s disease and ankylosing spondylitis. 1 The main causative agents reported include adalimumab, infliximab and etanercept. The onset of myositis ranged from 2 weeks to 34 months (mean 12.7 months) with TNFi use. Interstitial lung disease was also observed in patients with anti-synthetase antibodies. Withdrawal of TNFi led to clinical improvement in all patients. A subset of patients required additional immunosuppressive agents including corticosteroids, methotrexate and azathioprine. 1 Myositis with severe manifestations has also been treated with cyclophosphamide and rituximab. 7
It is also notable that our patient had a TNFi-induced anti-Mi-2 antibody without clinical features of dermatomyositis. This became undetectable 8 months after the cessation of anti-TNF. Given the anti-Mi-2 positivity, it is unclear whether the focal myositis observed in our patient would have progressed to dermatomyositis if infliximab was further continued. However, Tariq et al. reported a case of TNFi-induced polymyositis who had resolution of symptoms with glucocorticoids despite the continuation of etanercept. 3 The disappearance of anti-ds-DNA upon infliximab discontinuation further supports anti-TNF-driven autoimmunity in this patient. In general, anti-TNF-induced antibodies can persist for months to years despite cessation.
Although focal myositis in this patient could also be associated with ulcerative colitis, her bowel disease was quiescent on infliximab during the time of onset of her muscle symptoms. Orbital myositis and isolated gastrocnemius myositis are variants focal myositis that have been rarely reported in patients with IBD, especially in Crohn’s disease. Interestingly, TNFi therapy has been used to successfully treat focal myositis in these cases.8,9
Hence, the emergence of new autoantibodies (i.e ANA, anti-ds-DNA and anti-Mi-2), the temporal relationship with infliximab therapy and the resolution of focal myositis upon cessation, favours TNFi as the inciting agent. To the best of the authors’ knowledge, this case represents the first description of focal myositis attributed to TNFi therapy.
Conclusion
Tumour necrosis factor inhibitors can induce various autoantibodies and be associated with autoimmune diseases including focal myositis. A differential diagnosis of focal myositis should be considered in patients with unexplained local muscle pain who are being treated with a TNFi. Diagnosis of focal myositis is based on a constellation of symptoms, MRI changes and muscle histology. Treatment includes cessation of TNFi and if required, commencement of immunosuppressive therapy.
Footnotes
Author contributions
AM was the lead author involved in drafting of the manuscript , literature review and patient management. WL was involved in the article concept and the consultant leading the patient management . PW and NM were involved in the concept, design, editing of the manuscript. All authors reviewed and approved the final version of the manuscript.
Availability of data and material
The datasets generated and/or analysed during the current study are available from the corresponding author.
Ethical approval
Westmead Hospital does not require ethical approval for reporting individual cases or case series because patient information has been adequately anonymized.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Informed consent
Written informed consent was obtained from the patient(s) for their anonymised information to be published in this article
