Background/Purpose: The idiopathic inflammatory myopathies (IIM) are a heterogeneous group of rare autoimmune diseases characterised by muscle weakness and extramuscular manifestations. The most common autoantibody in IIM is anti-Jo-1, present in ~20% of patients and associated with anti-synthetase syndrome; a distinct clinical entity characterised by the presence of myositis, interstitial lung disease, Raynaud’s and mechanics’ hands. The development of anti-Jo-1 antibodies is also associated with the presence of HLA-DRB1*03:01. Smoking is a risk factor for ACPA+ rheumatoid arthritis in patients possessing the shared epitope. A previous study has suggested an interaction between smoking and DRB1*03 for the development of anti-Jo-1 antibodies in IIM. We sought to replicate this in a large single-country cohort using high resolution HLA imputation data.

Methods: Eight hundred and seventy two UK patients of Caucasian descent were recruited through the UK Myositis Network (UKMYONET) and Juvenile Dermatomyositis Research Group (JDRG) comprising predominantly of adult and juvenile dermatomyositis and polymyositis patients. Juvenile cases were included as a group of patients not exposed to smoking. Smoking was defined as ‘having ever smoked at least one cigarette a day for as long as a year’ at the time of recruitment. Antibody testing was conducted using immunoprecipitation. DRB1*03:01 was imputed from SNP genotyping information using SNP2HLA. Five hundred and ninety one patients had complete data for smoking, DRB1*03:01 and anti-Jo-1 status.

Results: A strong effect was seen with DRB1*03:01 and the development of anti-Jo-1 antibodies (see table 1). In DRB1*03:01 negative cases, an association was found between smoking and the development of anti-Jo-1 antibodies (p=0.04). In DRB1*03:01 positive cases, an association between smoking and the development of anti-Jo-1 antibodies almost reached statistical significance (p=0.052). No departure from a multiplicative effect between smoking and DRB1*03:01 was observed using anti-Jo-1 as the outcome measure, suggesting that there is no interaction between smoking and DRB1*03:01 status. When repeating the analysis in adult cases only, the effect of smoking did not reach statistical significance for the development of anti-Jo-1 antibodies.

Conclusion: Smoking may be associated with an increased risk of developing anti-Jo-1 antibodies, independent of DRB1*03:01 status. The exposure to smoking in adulthood may contribute to the increased frequency of anti-Jo-1 antibodies in adult compared to juvenile IIM cases.