Improving the Management of Rheumatoid Arthritis-Associated Lung Disease in Veterans Using Real-World Data

Rheumatoid arthritis (RA) is a chronic autoimmune disease affecting 1.3 million individuals in the U.S., causing physical disability, reduced quality of life, premature mortality, and enormous health care costs. Veterans with RA die from respiratory diseases at a rate three times higher than the general population.

Much of this excess respiratory mortality affecting Veterans with RA is attributable to interstitial lung disease (ILD), which has a prognosis as poor as many cancers.

Despite advances in RA treatment over the past two decades with the adoption of novel therapies and more aggressive treatment strategies, the optimal management of RA-ILD is unknown.

Two critical limitations for effectively managing Veterans with RA-ILD are 1) the inability to identify Veterans with progressive RA-ILD?those most likely to benefit from anti-fibrotic or aggressive immunomodulatory therapies and 2) a lack of data on the comparative effectiveness and safety of disease- modifying RA therapies in this population.

Therefore, the overall objectives of this project are to leverage unique prospective Veteran RA-ILD cohorts and data linkages within the Veterans Health Administration (VHA) to 1) identify prognostic serum and genetic biomarkers for RA-ILD and 2) compare the effectiveness and safety of RA therapies in Veterans with RA-ILD.

Our central hypotheses are that serum and genetic biomarkers will be independently associated with, and accurately predict, RA-ILD progression, and select RA therapies will differentially slow ILD progression and improve related survival.

In Aim 1, we will utilize prospective RA-ILD cohorts to identify prognostic serum and genetic biomarkers and derive progressive RA-ILD predictive models.

We hypothesize that biomarkers from RA-ILD pathophysiologic domains?novel disease-related autoantibodies, genetic markers, pro-inflammatory cytokines, and matrix metalloproteinases?will be independently associated with ILD progression in Veterans with RA-ILD.

In Aim 2, we will link national VHA data sources and use advanced causal inference methodology to identify RA therapies that are associated with less ILD progression and improved survival in Veterans with RA-ILD.

We hypothesize that compared to tumor necrosis factor inhibitors (TNFi), non-TNFi biologic therapies (rituximab, abatacept, and tocilizumab) will be associated with less ILD progression and have a lower mortality risk in Veterans with RA-ILD.

Impact: The results from the proposed research will assist clinicians with personalized treatment selection in Veterans with RA-ILD, a high-risk population with little data to currently guide treatment selection.

The PI will complete this research plan under the mentorship of a multidisciplinary team of experts in clinical and pharmacoepidemiologic research within the VHA, building upon his early research productivity.

The accompanying mentored training program in pharmacoepidemiology and causal inference methodology obtained through advanced, immersive coursework and professional meetings will position the PI to conduct high-impact research in RA and RA-ILD within the VHA.

Upon completion of this award, the PI will be poised for an independent research career targeting improvements in the long-term outcomes for Veterans with RA.