Background/Purpose: Interstitial lung disease (ILD) is an outcome with high morbidity and mortality in rheumatoid arthritis (RA).  Citrullinated proteins are observed in these lung tissues; however, the association of specific anti-citrullinated peptide antibodies (ACPA) with ILD in RA is unknown.

Methods: RA patients underwent multi-detector computed tomography (MDCT) of the chest with interpretation by a pulmonary radiologist for ILD features [ground glass opacification (GGO), reticulation (R), honeycombing (HC), and traction bronchiectasis (TB)].  A semi-quantitative ILD Score (ILDS; range 0-32) for ILD features was calculated.  Concurrent serum samples were assessed for anti-CCP (CCP2) and levels of a panel of antibodies against 4 non-citrullinated proteins [fibrinogen A, HSP60, apolipoprotein (Apo) A1, and Apo E] and 17 citrullinated full-length proteins or peptides within these proteins (see table for list) using a custom Bio-Plex bead array. Individual candidate citrullinated antigens were conjugated to spectrally-distinct fluorescently dyed beads. Peptide-conjugated beads were incubated with diluted patient sera, autoantibody binding detected with a phycoerythrin-conjugated secondary antibody, and levels of autoantibody binding quantitated on a Luminex 200 System.  High level ACPA was defined at ≥ the group 75^th percentile.  Pulmonary function testing (PFT) was performed in 156 patients a mean of 21±3 months later.

Results: Among 177 RA patients [60% female, 86% Caucasian, mean age 59±9 years, 11% current smokers], any ILD features (i.e. ILDS>0) was observed in 57 (32%).  Among those with any ILD, the median ILDS was 3 (range 1-10).  A predominant pattern of GGO was observed in 22 (39%) and R/HC/TB in 35 (61%).  PFT restriction or impaired diffusion was observed in 36 (23%).

                Levels of CCP2 and all specific ACPAs were 46-273% higher among RA patients with vs. those without ILD (all p-values<0.05), and higher levels correlated with higher ILDS.  In contrast, levels of non-citrullinated protein antibodies were not higher in those with ILD.  The median number of high level ACPA was significantly greater for those with an ILDS≥3 vs. those with an ILDS=0 (6 vs. 1; p=0.005, see Table).   Each ACPA was associated, on average, with a 0.10 unit increase in ILDS (p=0.001), an association that remained significant after adjusting for features associated with ILD [age, gender, current and former smoking, DAS28, current prednisone and leflunomide use].  More high level ACPA were observed in R/HC/TB compared to ILDS=0, but the difference was less robust for GGO vs. ILDS=0 (Table).  More high level ACPA were observed in those with PFT restriction or impaired diffusion compared to those without these PFT findings (Table).

Conclusion: Our findings of a broader ACPA repertoire in RA ILD suggest a role for ACPA in the pathogenesis of ILD and/or implicate inflamed lung parenchyma as a source of ACPA generation.