Pulmonary Damage Biomarkers for Risk of Rheumatoid Arthritis–Associated Interstitial Lung Disease: Enabling Early Detection and Risk Stratification

Abstract

Rheumatoid arthritis-associated interstitial lung disease (RA-ILD) has garnered increasing attention because of reports of higher-than-expected prevalence with persistent excess mortality, despite the availability of antifibrotics as novel therapeutic options. [1][2][3][4] Although antifibrotics have only modest efficacy, emerging therapeutic options have bolstered efforts for risk stratification, early detection, and identifying risk factors for progression of RA-ILD. Chest high-resolution computed tomography (HRCT) scan is the gold standard for ILD diagnosis. Because of ionizing radiation, cost, and logistical barriers associated with chest HRCT scan, a risk-stratified screening approach to RA-ILD has been proposed. 5 Previous studies identified RA-ILD risk factors already available in daily practice, eg, male sex, older age, smoking, high articular RA disease activity, and elevated RA-related autoantibodies. [6][7][8] However, performance characteristics of these factors are not currently sufficient to risk stratify who should or should not receive a chest HRCT, and pulmonary function tests (PFTs) are not sensitive enough to detect early ILDs, so biomarkers may improve risk stratification in combination with clinical variables. 7,8 In addition to the potential for clinical risk stratification, biomarkers may also provide biologic insight into the pathogenesis of RA-ILD and, perhaps, other fibrotic diseases. The MUC5B promoter variant is the strongest genetic risk factor for usual interstitial pneumonia (UIP), the prototypic fibrotic RA-ILD subtype. 9,10 Peripheral blood biomarkers have been shown to predict RA-ILD. [11][12][13] Establishing biomarkers of pulmonary damage may be particularly helpful in identifying preclinical RA-ILD, even before changes in PFTs. Matrix metalloproteinases (MMPs) are responsible for regulation and degradation of the extracellular matrix and have been associated with RA-ILD previously, 14 so they may be a