Background/Purpose:   Emerging data suggest that microorganisms and mucosal inflammation may play a role in the etiology of rheumatoid arthritis (RA). Furthermore, our published findings of inflammatory airways disease in a high proportion of RA-related autoantibody positive subjects without inflammatory arthritis (IA), some of whom later developed classifiable RA, suggest that the lung may play a role in early RA pathogenesis. Therefore, investigations of the lung microbiome in subjects at risk for future RA may provide insight into RA pathogenesis.   

Methods:   13 CCP positive cases without IA and 9 healthy seronegative controls were identified through community screening. Induced sputa was collected from all subjects. DNA extractions for microbiome analysis were performed using Qiagen EZ1 Advanced platform.  Established protocols with barcoded PCR primers were used to construct multiplexed amplicon pools and assign sequences to the appropriate subject sample. Microbial prevalence and median relative abundance in sputa were compared using Explicet. 

Results:   Cases were older and more frequently male than controls (Table), and more cases had been smokers, although they had quit smoking a median of 9 years prior to sputa collection. Bacteria were identified by rRNA pyrosequencing (>1,000 sequences per sample). Good’s coverage was >98.9% for all samples.  80 genera were identified across all samples with an average of 30 per sample.  Relative abundance of Haemophilus and Neisseria was elevated in cases compared to controls (p=0.04) (Table), and there was a trend toward increased relative abundance of Streptococcus in cases. There was no significant difference in Porphyromonas, Prevotella and Mycoplasma between groups.

Conclusion: These results show differences in sputa microbiota between RA-related autoantibody positive cases without IA and controls. This is particularly intriguing because our prior work has shown that 9 of these autoantibody positive cases had inflammatory airways disease on lung imaging, raising the possibility of a mechanistic link between the microbiome, mucosal inflammation and generation of RA-related autoimmunity in the lung. A caveat in this small pilot study is the unknown influence on the lung microbiome of differences in age and sex between groups. Also, differences in smoking history between groups may affect results as smoking is known to alter the lung microbiome. However, smoking is also associated with CCP positivity in established RA, albeit through unknown mechanisms. As such, it is possible that a mechanism that drives CCP generation is smoking (or other inhaled factor)-related changes in the lung microbiota. Additional study, including speciation of organisms and prospective follow-up of larger numbers of subjects and controls, is needed to determine the biologic relevance of the lung microbiome to the pathogenesis of early RA.