Article Text

Download PDFPDF

Autoimmune necrotising myopathy and HMGCR antibodies
  1. Kushan Karunaratne1,
  2. Dimitri Amiras2,
  3. Matthew C Pickering3,
  4. Monika Hofer4,
  5. Stuart Viegas1
  1. 1 Department of Neurology, Imperial College Healthcare NHS Trust, London, UK
  2. 2 Department of Radiology, Imperial College Healthcare NHS Trust, London, UK
  3. 3 Department of Rheumatology, Imperial College Healthcare NHS Trust, London, UK
  4. 4 Department of Neuropathology, Oxford University Hospitals NHS Foundation Trust, Oxford, UK
  1. Correspondence to Dr Stuart Viegas, Neuromuscular Unit, Department of Neurology, Charing Cross Hospital, London W6 8RF, UK; stuart.viegas{at}nhs.net

Abstract

Statins lower serum cholesterol concentrations by inhibiting the enzyme 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR). Muscle side effects are relatively common and include asymptomatic elevation of serum creatine kinase (CK), myalgia, proximal muscle weakness and rhabdomyolysis. More recently, a subset of cases of immune-mediated necrotising myopathy has been found to have antibodies against HMGCR. It is often an aggressive and debilitating myopathy and has a complex pathogenesis characterised by fibre necrosis, usually with minimal associated inflammation. Not all such patients are taking statins. The general consensus is that best treatment involves withdrawing the statin and giving immunosuppressive and immunomodulatory treatment. We describe three cases of HMGCR-related immune-mediated necrotising myopathy, detailing their clinical course and subsequent management, illustrating the spectrum of this disorder.

  • HMG-CoA reductase
  • necrotizing
  • myopathy
  • statins
  • autoantibodies

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Case reports

Case 1

A 61-year-old Indian man was re-referred to the neuromuscular clinic after a recurrence of proximal muscle weakness and myalgia. He had been referred 6 years earlier with myalgia and proximal muscle weakness that had worsened over 18 months. He had taken simvastatin 20 mg for over 2 years. His serum creatine kinase (CK) was elevated at 5000 IU/L (24–195). A subsequent muscle biopsy showed non-specific myopathic changes. After stopping the simvastatin, the symptoms completely resolved and his CK normalised.

Four years later, his primary care physician started him on atorvastatin 20 mg for recurrence of hypercholesterolaemia. A few weeks later his muscle symptoms recurred, noting difficulties with stairs and getting up from the ground after praying at the temple. Examination confirmed proximal muscle weakness in upper and lower limbs and a myopathic gait, prompting re-referral to our service. We stopped his statin, but unlike with his previous presentation, the weakness continued to progress. Serum CK was again elevated at 8100 IU/L. He was found to have 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) antibodies but testing for other myositis antibodies was negative. An MR scan of the lower limb muscles showed widespread oedema that was more pronounced posteriorly (figure 1A,B). A muscle biopsy showed scattered necrotic and regenerating fibres with no significant inflammation. Given the clinical context and the presence of detectable HMGCR antibodies, the features were consistent with an immune-mediated necrotising myopathy. We gave intravenous methylprednisolone followed by oral prednisolone 50 mg. We later started methotrexate as a corticosteroid sparing agent and established a weaning plan for prednisolone. Screening for underlying malignancy was negative.

Figure 1

(A) Short tau inversion recovery axial MRI sequences at the level of the mid femur, showing muscle oedema affecting the hamstring muscles of semimembranosus, semitendinosus, biceps femoris and adductor magnus with subtle oedema in the quadriceps muscle group bilaterally. Note the relative sparing of sartorius and adductor longus. (B) Short tau inversion recovery axial MRI sequences at the level of the proximal tibia showing marked bilateral medial and lateral gastrocnemius muscle oedema.  AD, adductor magnus and longus; BFL, biceps femoris long head; GR, gracilis; LG, lateral gastrocnemius; MG, medial gastrocnemius; RF, rectus femoris; SA, sartorius; ST, semitendinosus; SM, semimembranosus; SOL, soleus; TA, tibialis anterior; TP, tibialis posterior; VL, vastus lateralis; VI, vastus intermedius; VM, vastus medialis.

Follow-up 20 months later

His CK normalised 3 months after starting treatment.  He still walks with the aid of one stick and manages to walk for 10 minutes before he needs to stop. He can climb stairs more easily. Objectively, he has mild weakness of deltoids and iliopsoas muscles. His balance remains poor. He still has several symptoms, many of which relate to his ongoing immunotherapy. He has mild breathlessness but no dysphagia. Lung function tests are satisfactory. He has developed corticosteroid-related diabetes mellitus and now takes antihyperglycaemic agents for glucose control. Oral corticosteroid weaning continues and he is tolerating weekly methotrexate 15 mg with satisfactory blood tests (see table 1 for summary for all cases).

Table 1

Summary of clinical, laboratory, imaging, neurophysiology, histopathology and treatment

Case 2

A 52-year-old Indian man was referred from rheumatology to the neuromuscular clinic with a 2-month history of dysphagia, weight loss, myalgia, as well as difficulties getting out of a chair and managing stairs. He had taken simvastatin 20 mg for over 10 years and the dose had increased to 40 mg 12 months before. Examination confirmed symmetrical proximal weakness in his upper and lower limbs. Serum CK was elevated at 11 000 IU/L (24–195). CT scan of chest, abdomen and pelvis, upper gastrointestinal endoscopy and colonoscopy were normal, although a barium swallow did identify oesophageal dysmotility. MR scan of his lower limb muscles showed widespread oedema while electromyography showed myopathic changes with evidence of irritability. He had detectable HMGCR antibodies but no other myositis-specific antibodies. The muscle biopsy findings were consistent with an immune-mediated necrotising myopathy (figure 2A,B). After stopping the statin, his symptoms markedly improved. His swallowing improved and weight started to increase. He could walk for half a mile on a flat surface and could manage two or three flights of stairs. Two months later, there was only mild weakness in his deltoid muscles and he could perform a squat and get up from the floor without difficulty. His serum CK was 2700 IU/L but he continued to improve without treatment and felt almost 100% back to normal.

Figure 2

(A) H&E stained section showing scattered fibres at various stages of necrosis and regeneration (pale, purple, sometimes vacuolated fibres). Secondary changes (increase in variability of fibre diameters and internal nuclei are also seen). Note the relative lack of inflammatory cells, which are mostly restricted to necrotic fibres. (B) Acid phosphatase highlighting various stages of myophagocytosis in necrotic fibres (red staining).

Follow-up 20 months later

He did not require immunosuppressive treatment. He can now walk a few miles at a time and can also run. Examination shows no objective weakness. His most recent serum CK was 52 IU/L.

Case 3

A 46-year-old woman presented with a 1-year history of difficulty climbing stairs, getting up from the floor while at work, and weakness in her arms. She had been previously well and had never taken a statin. She was referred to a rheumatologist. Her serum CK was raised at 3800 IU/L (24–170) and electromyography showed myopathic changes with evidence of irritability. MR scan of the lower limb muscles showed oedema, suggesting bilateral myositis. She was started on prednisolone 40 mg once a day. A muscle biopsy showed features consistent with an immune-mediated necrotising myopathy. Testing for myositis antibodies was negative although HMGCR antibodies were subsequently identified. Although her serum CK fell to 1000 IU/L over a few weeks, her limb weakness continued to progress despite a course of intravenous methylprednisolone. She was subsequently referred to us for a second opinion. Gradually, the oral prednisolone was reduced and methotrexate introduced as a corticosteroid sparing agent. She later developed sensory symptoms and was found to have an underlying small-fibre neuropathy that has been treated symptomatically.

Follow-up 26 months later

Despite corticosteroid and methotrexate therapy, her serum CK fluctuated between 500 and 1000 IU/L and subjectively she felt weaker when prednisolone was weaned to below 10 mg. She showed no objective improvement following intravenous immunoglobulin. Repeat muscle imaging showed new signal change in proximal musculature. We recently started her on rituximab as well as ongoing prednisolone and methotrexate. Her CK has fallen to 319 IU/L, the lowest since presentation, whilst her muscle strength has also improved.  Extensive repeated investigations including a PET CT scan, mammography, upper gastrointestinal endoscopy and colonoscopy have not found any underlying malignancy.

Discussion

Immune-mediated necrotising myopathy is a recently identified subgroup of idiopathic inflammatory myopathies. It is considered a rare disease, but accounts for approximately 20% of idiopathic inflammatory myopathies.1 According to the European Neuromuscular centre, signal recognition particle and HMGCR antibodies are strongly associated with immune-mediated necrotising myopathy.2 HMGCR antibody-related immune-mediated necrotising myopathy has a different pathogenic mechanism to the other statin-associated muscle complaints (myalgia, weakness and rhabdomyolysis).3 This discussion will focus on HMGCR immune-mediated necrotising myopathy and is aimed at the general neurologist.

HMGCR antibodies were first described in 2010 by Christopher-Stine and colleagues as anti-200/100, a novel antibody associated with immune-mediated necrotising myopathy.4 Further work by this group identified the autoantigenic target in the cholesterol synthesis cascade and the autoantibody was later defined as anti-HMGCR.

Statin exposure

Several studies have shown a strong association with statin exposure, ranging from 37.5% to 94%.1 Typical cases usually are younger statin-naïve patients or older statin-exposed patients. Statins upregulate HMGCR expression in cells and it is possible that enzyme overexpression can lead to developing autoimmunity against HMGCR in susceptible people.5 However, to date, the role of statins triggering the disease remains unclear. With statin-naïve patients, it is important to explore the dietary history, as some dietary supplements and foods such as oyster mushrooms contain statins.2 Statin unexposed patients may also be more resistant to treatment.2

Clinical features

Characteristically, patients present subacutely with severe symmetrical, proximal muscle weakness leading to muscle atrophy if untreated. Myalgia is common although extramuscular manifestations with skin and lung involvement are relatively rare.2

Risk of malignancy

Patients with this condition appear to have a higher risk of malignancies and require screening.1 6 7 A recent study of 115 patients identified malignancy in 17.3% in the anti-HMGCR cohort, with lung, breast, renal, oesophagus, gastric, ovarian and endometrial cancers detected.6 We typically perform whole-body fluorodeoxyglucose positron emission tomography-CT imaging, with or without gastrointestinal endoscopy, with testicular ultrasound in men and mammograms in women.

Laboratory investigations

Serum CK is often markedly elevated with a median concentration of 3500 IU/L.2 As with other inflammatory myopathies, serum CK is a good marker for disease activity and can be used for disease follow-up.1 4 At present, testing for HMGCR antibodies relies on a research assay. Once the assay is validated for routine clinical use, it should be included as part of the myositis-specific antibody panel that is offered by most established immunology laboratories.

Electromyogram

Electromyogram findings typically show an irritable myopathy. The findings include low amplitude and short duration of motor unit potentials with spontaneous fibrillation potentials, positive sharp waves and myotonic or pseudomyotonic discharges.

Imaging

MR imaging of the lower limb muscles is an important part of the diagnostic algorithm. Typically, patients have symmetrical and diffuse intramuscular and intrafascial oedema in affected muscles, in anterior, medial and posterior compartments.8 MRI is also useful for localising affected muscles, to guide muscle biopsy and to improve the diagnostic yield.

Muscle biopsy

Muscle biopsies in these patients typically show scattered necrotic and regenerating fibres with minimal inflammatory infiltrate (figure 2A,B). Other features that can help distinguish this condition include upregulation of major histocompatability complex class 1 antigen in necrotic and non-necrotic fibres and microvascular deposition of membrane attack complex.1 9

Treatment

There is as yet no clinical trial evidence to determine the optimal treatment strategy in HMGCR immune-mediated necrotising myopathy. Some reports indicate that simply stopping the statin is sufficient (as in patient 2). The vast majority of patients, however, require immunosuppressive/immunomodulatory therapy. The first-line treatment is usually prednisolone (typically 1 mg/kg/day). However, in most patients, prednisolone monotherapy is insufficient to control the disease and patients need corticosteroid sparing agents (methotrexate and azathioprine).

In early reports of the disease, two oral agents and intravenous immunoglobulin were used to achieve disease remission, indicating the need for aggressive therapy.2 Recently, a case series reported three statin-exposed HMGCR immune-mediated necrotising myopathy patients who were successfully treated with intravenous immunoglobulin as monotherapy.10 In more resistant cases, both cyclophosphamide and rituximab have been used.2

Conclusion

These three patients highlight a spectrum in the clinical phenotype and treatment response observed with HMGCR immune-mediated necrotising myopathy. As always, the differential diagnosis for neuromuscular disorders remains wide and as with any disorder, the workup must start with basic investigations. Certain features in the patient with myalgia and weakness such as serum CK above 2000 IU/L, lack of extramuscular manifestations and symptoms persisting despite stopping a statin should raise suspicion of immune-mediated necrotising myopathy and prompt additional investigations. We would always suggest MR imaging (provided no contraindications) and a guided muscle biopsy in addition to antibody testing. With the risk of associated malignancy, we also recommend thorough screening. Awareness of this condition is important as treatment often requires aggressive immunosuppressive treatment.

Key points

  • 3-Hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) immune-mediated necrotising myopathy can occur in both statin-exposed and statin-naïve patients.

  • HMGCR antibodies are not part of current routine myositis antibody panels and need to be requested separately.

  • There is a risk of malignancy associated with this condition and patients require appropriate screening.

  • Symptoms usually persist despite stopping the statin; the treatment usually involves immunosuppressive/immunomodulatory treatment.

Acknowledgments

We thank Dr Peter Kelleher (Consultant Immunologist, Department of Infection and Immunity, Imperial College Healthcare NHS Trust) for helpful input regarding HMGCR antibody testing (undertaken by Immunology Laboratory, Oxford University Hospitals NHS Foundation Trust). We would also like to thank Dr Maxine Hogarth (Consultant Rheumatologist, North West London Healthcare NHS Trust) for referring patient 2.

References

Footnotes

  • Contributors SV is the consultant responsible for care of all three patients. MP responsible for care of patient 3. DA and MH have been involved in management of all three patients. All authors have been involved in revising the manuscript.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed. This paper was reviewed by David Hilton-Jones, Oxford, UK.

Linked Articles

  • Review
    David Hilton-Jones

Other content recommended for you