Background/Purpose: Data suggest the lung may be a site of RA-related autoantibody (Ab) generation including findings of airways inflammation prior to onset of articular RA (Demoruelle 2012; Fischer 2012), and the presence of RA-related Abs in sputum but not blood in subjects at risk for future RA (Willis and Demoruelle 2013, in press). The mechanisms by which RA-related Abs may be generated in the lung are unknown; however, inducible bronchus-associated lymphoid tissue (iBALT) is known to form in the lung in response to inhaled antigens, can release Abs in the airways, and contains plasma cells generating RA-related Abs in RA (Rangel-Moreno 2006). Therefore, we hypothesize that iBALT formation is a mechanism by which RA-related Abs are generated in the lung, and aim to establish an association between iBALT and serum RA-related Abs.

Methods: We obtained stored blood and lung tissue from 48 subjects from the NIH Lung Tissue Research Consortium that includes subjects undergoing lung biopsy for clinical management of any underlying disease. Routine histologic staining of the lung was performed and interpreted in a blinded fashion by a trained pathologist. Lymphoid follicles consistent with iBALT and follicles with areas of inflammation organized into germinal centers (GCs) were quantified. Serum was tested for RA-related Abs: CCP2 (IgG, Axis-Shield), CCP3.1 (IgA/IgG, INOVA), and RF isotypes (IgM/A/G, INOVA). Differences in prevalence of positivity were compared using chi-squared/Fishers exact testing.

Results: Subjects with iBALT had similar rates of ever smoking (74% v. 64%, p=0.54), were younger (median age 60 v. 68, p=0.02), more often female (78% v. 44%, p=0.02), and more often had interstitial lung disease (ILD) (61% v. 8%, p<0.01) compared to subjects without iBALT. Nine subjects had a clinical diagnosis of RA; of these, 9 (100%) had airways disease or ILD, 9 (100%) were positive for ≥1 serum RA-related Ab, and 8 (89%) had iBALT. In the 39 subjects without RA, 35 (90%) had airways disease or ILD, 13 (33%) were positive for ≥1 serum RA-related Ab, and 15 (38%) had iBALT. In comparing all subjects, the prevalence of iBALT and GCs was associated with serum RA-related Ab positivity (Table).

Conclusion: iBALT is associated with serum RA-related Abs in subjects with and without RA. While these pilot results are driven by subjects with diagnosed RA, when combined with the published data discussed above, this association suggests that iBALT may be a mechanism of RA-related Ab generation. Notably, iBALT can develop in response to inflammation, and given our prior findings of airways inflammation associated with RA-related Abs in absence of synovitis, iBALT may be a mechanism of RA-related Ab generation prior to synovitis in RA. Going forward we will expand these studies to additional subjects to evaluate the generation of RA-related Abs within iBALT, and factors (e.g. microbial) that may trigger iBALT and Ab development in subjects with and without classified RA.