1746: Inflammatory- and Fibrosis-Related Biomarkers and Rheumatoid Arthritis-Associated Interstitial Lung Disease Risk

Background/Purpose: There exists a need to identify informative peripheral blood biomarkers for RA-interstitial lung disease (RA-ILD) and elucidate the pathophysiological processes driving RA-ILD onset. Several inflammatory- and fibrosis-related mediators have been identified in diseases characterized by tissue fibrosis such as idiopathic pulmonary fibrosis (IPF) and heart failure, but few have been evaluated specifically in RA-ILD. We investigated the association of inflammatory- and fibrosis-related biomarkers with RA-ILD risk in a large, prospective RA cohort.

Methods: Using a multicenter, prospective cohort of U.S. Veterans with RA, we performed a cross-sectional study of prevalent RA-ILD and cohort study of incident RA-ILD. Inflammatory- and fibrosis-related biomarkers discovered in IPF and heart failure: growth differentiation factor-15 (GDF-15), interleukin-18, interferon-gamma inducible protein-10, interferon-inducible T-cell alpha chemoattractant, monokine induced by gamma interferon, pentraxin-3, and soluble vascular cell adhesion molecule, were measured from serum at enrollment using the MesoScale Discovery platform. Biomarker concentrations were log-transformed and standardized, as well as categorized into quartiles to assess non-linear associations. RA-ILD was identified through informatics-based screening and validation by a board-certified rheumatologist requiring a clinical diagnosis plus imaging or biopsy findings. The associations of candidate biomarkers with RA-ILD were assessed separately in multivariable logistic (prevalent RA-ILD) and Cox (incident RA-ILD) regression models adjusted for age, sex, race, smoking status, body mass index category, anti-CCP antibody positivity, DAS28, and comorbidity burden.

Results: Cross-sectional analyses were conducted among 2,737 RA participants (87% male, mean age 64 years), with 124 having prevalent RA-ILD. Higher concentrations of GDF-15 (per 1 SD aOR 1.42 [1.08, 1.87]) and pentraxin-3 (aOR 1.30 [1.07, 1.58]) were significantly associated with prevalent RA-ILD (Table 1). Referent to the lowest quartile, higher quartiles of GDF-15 (Q3 aOR 3.76 [1.68, 8.41]; Q4 4.10 [1.80, 9.31]) and the highest quartile of Pentraxin-3 (aOR 2.18 [1.30, 3.65]) were associated with prevalent ILD (Figure 1). Concentrations of other biomarkers were not associated with prevalent RA-ILD. After excluding those with prevalent RA-ILD, 174 participants developed incident RA-ILD over 20,957 patient-years of follow-up. None of the analyte concentrations were significantly associated with incident RA-ILD risk (Table 1). However, the highest quartile of GDF-15 was associated with a >3-fold higher risk of incident RA-ILD (adjusted HR 3.21 [1.94, 5.32]) (Figure 2).

Conclusion: GDF-15, a member of the transforming growth factor-β family, was independently associated with both prevalent and incident RA-ILD in a large RA cohort. Other inflammatory- and fibrosis-related mediators seen in other fibrotic diseases, such as IPF and heart failure, were not associated with RA-ILD risk, suggesting there may be unique disease specific inflammatory and fibrotic pathophysiologic pathways in RA-ILD.