Phenome-Wide Association Study of Novel Anti-Citrullinated Peptide Antibodies in Rheumatoid Arthritis

Katherine Liao¹, Jeffrey A. Sparks¹, Boris Hejblum², I-Hsin Kuo¹, Jing Cui¹, Lauren J. Lahey³, Andrew Cagan⁴, Vivian Gainer⁴, Weidong Liu⁵, Tony Cai⁶, Jeremy Sokolove⁷ and Tianxi Cai², ¹Division of Rheumatology, Immunology, and Allergy, Brigham & Women's Hospital, Boston, MA, ²Biostatistics, Harvard School of Public Health, Boston, MA, ³Medicine, VA Palo Alto HealthCare System and Stanford University, Palo Alto, CA, ⁴Research Computing, Partners HealthCare, Charlestown, MA, ⁵Institute of Natural Sciences, Shanghai Jiao Tong University, Shanghai, China, ⁶Statistics, University of Pennsylvania, Philadelphia, PA, ⁷Medicine, VA Palo Alto Health Care System and Stanford University, Palo Alto, CA

Meeting: 2015 ACR/ARHP Annual Meeting

Date of first publication: September 29, 2015

Keywords: anti-citrullinated protein/peptide antibodies (ACPA), Big data, phenotypes and rheumatoid arthritis (RA)

Session Information

Date: Monday, November 9, 2015

Title: Epidemiology and Public Health Poster II: Pathogenesis and Treatment of Systemic Inflammatory Diseases

Session Type: ACR Poster Session B

Session Time: 9:00AM-11:00AM

Background/Purpose:

RA patients develop autoantibodies against a spectrum of antigens. However, the clinical significance of these autoantibodies after RA diagnosis is unclear. The Phenome-Wide Association Study (PheWAS) is one approach to studying the clinical significance of novel biomarkers. The PheWAS was designed to analyze biomarker data linked with electronic medical records (EMR), allowing for high-throughput screening of potential associations between biomarkers and phenotypes. Using the PheWAS approach, we examined the association between variations in autoantibodies and subphenotypes of RA.

Methods:

We conducted this study in an electronic medical record (EMR) based RA cohort with patients from 2 large academic institutions. Using a validated multiplex bead assay, we measured 36 autoantibodies targeting epitopes from 10 antigens (citrullinated & non-citrullinated) implicated in RA (Figure). We extracted all unique ICD9 codes for each subject. ICD9 codes were then grouped using a published method into disease categories or PheWAS codes. For example, Felty’s syndrome (ICD9 714.1) was included in the ‘RA’ PheWAS code along with RA (ICD9 714.0), while juvenile idiopathic arthritis (ICD9 714.3) was not. Autoantibody data were grouped by antigen target, e.g. fibrinogen. We studied all PheWAS codes with ≥3% prevalence in the cohort. Association analyses were performed between each autoantibody group and all PheWAS codes, adjusting for age, gender, race and correlation between individual autoantibody tests. A false discovery rate (FDR) <0.1 was considered significant. For significant associations, we reviewed the medical records of 50 subjects with each PheWAS code to determine the positive predictive value (PPV) of the PheWAS code.

Results:

We studied 1006 RA subjects, mean age 61 years (SD 12.9) and 79% female. The RA cohort had 3,568 unique ICD9 codes grouped into 625 PheWAS codes; 206 PheWAS codes had a prevalence of ≥3%. The PheWAS grouped by autoantibody target identified 13 associations with FDR<0.1 (Figure). Among the associations, the PheWAS code with the highest PPV (96%) on chart review was ‘Other alveolar and parietoalveolar pneumopathy’, which included diagnoses of cryptogenic organizing pneumonia, obliterative bronchiolitis, lipoid pneumonia, and alveolitis. These data suggest a link between autoantibodies targeting fibrinogen, both citrullinated & non-citrullinated, with chronic inflammatory lung diseases (p=2.6×10^-4) in RA.

Conclusion:

We demonstrated the application of a bioinformatics method, the PheWAS to screen for the clinical significance of novel autoantibodies in RA. The PheWAS identified a potential link between autoantibodies targeted at fibrinogen and inflammatory lung conditions in RA. These findings are of particular interest since pulmonary disease is recognized as a major cause of mortality in RA.

Disclosure: K. Liao, None; J. A. Sparks, None; B. Hejblum, None; I. H. Kuo, None; J. Cui, None; L. J. Lahey, None; A. Cagan, None; V. Gainer, None; W. Liu, None; T. Cai, None; J. Sokolove, Novimmune, 5; T. Cai, None.

To cite this abstract in AMA style:

Liao K, Sparks JA, Hejblum B, Kuo IH, Cui J, Lahey LJ, Cagan A, Gainer V, Liu W, Cai T, Sokolove J, Cai T. Phenome-Wide Association Study of Novel Anti-Citrullinated Peptide Antibodies in Rheumatoid Arthritis [abstract]. Arthritis Rheumatol. 2015; 67 (suppl 10). https://acrabstracts.org/abstract/phenome-wide-association-study-of-novel-anti-citrullinated-peptide-antibodies-in-rheumatoid-arthritis/. Accessed .

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/phenome-wide-association-study-of-novel-anti-citrullinated-peptide-antibodies-in-rheumatoid-arthritis/