Open Access Open Badges Research article

Anti-MDA5 autoantibodies in juvenile dermatomyositis identify a distinct clinical phenotype: a prospective cohort study

Sarah L Tansley1, Zoe E Betteridge2, Harsha Gunawardena3, Thomas S Jacques4, Catherine M Owens5, Clarissa Pilkington6, Katie Arnold7, Shireena Yasin7, Elena Moraitis6, Lucy R Wedderburn8, Neil J McHugh9* and on behalf of UK Juvenile Dermatomyositis Research Group

Author Affiliations

1 Royal National Hospital for Rheumatic Diseases, Bath, UK

2 Department of Pharmacy and Pharmacology, University of Bath, Bath, UK

3 Department of Rheumatology, North Bristol NHS Trust, Bristol, UK

4 Department of Histopathology, Great Ormond Street Hospital NHS Foundation Trust, London and Neural Development Unit, University College London, London, UK

5 Department of Cardiothoracic Radiology, Great Ormond Street Hospital NHS Foundation Trust, London, UK

6 Rheumatology Unit, Great Ormond Street Hospital NHS Foundation Trust, London, UK

7 Rheumatology Unit UCL Institute of Child Health, London, UK

8 Rheumatology Unit UCL Institute of Child Health and Arthritis Research UK Centre for Adolescent Rheumatology at UCL, UCLH and GOSH l, London, UK

9 Royal National Hospital for Rheumatic Diseases, Bath, BA1 1RL, Bath, UK

For all author emails, please log on.

Arthritis Research & Therapy 2014, 16:R138  doi:10.1186/ar4600

Published: 2 July 2014



The aim of this study was to define the frequency and associated clinical phenotype of anti-MDA5 autoantibodies in a large UK based, predominantly Caucasian, cohort of patients with juvenile dermatomyositis (JDM).


Serum samples and clinical data were obtained from 285 patients with JDM recruited to the UK Juvenile Dermatomyositis Cohort and Biomarker Study. The presence of anti-MDA5 antibodies was determined by immunoprecipitation and confirmed by ELISA using recombinant MDA5 protein. Results were compared with matched clinical data, muscle biopsies (scored by an experienced paediatric neuropathologist) and chest imaging (reviewed by an experienced paediatric radiologist).


Anti-MDA5 antibodies were identified in 7.4% of JDM patients and were associated with a distinct clinical phenotype including skin ulceration (P = 0.03) oral ulceration (P = 0.01), arthritis (P <0.01) and milder muscle disease both clinically (as determined by Childhood Myositis Assessment Score (P = 0.03)) and histologically (as determined by a lower JDM muscle biopsy score (P <0.01)) than patients who did not have anti-MDA5 antibodies. A greater proportion of children with anti-MDA5 autoantibodies achieved disease inactivity at two years post-diagnosis according to PRINTO criteria (P = 0.02). A total of 4 out of 21 children with anti-MDA5 had interstitial lung disease; none had rapidly progressive interstitial lung disease.


Anti-MDA5 antibodies can be identified in a small but significant proportion of patients with JDM and identify a distinctive clinical sub-group. Screening for anti-MDA5 autoantibodies at diagnosis would be useful to guide further investigation for lung disease, inform on prognosis and potentially confirm the diagnosis, as subtle biopsy changes could otherwise be missed.