Investigation of genetically regulated gene expression and response to treatment in rheumatoid arthritis highlights an association between IL18RAP expression and treatment response

Abstract

Objectives: In this study, we sought to investigate whether there was any association between genetically regulated gene expression (as predicted using various reference panels) and anti-tumour necrosis factor (anti-TNF) treatment response (change in erythrocyte sedimentation rate (ESR)) using 3158 European ancestry patients with rheumatoid arthritis.

Methods: The genetically regulated portion of gene expression was estimated in the full cohort of 3158 subjects (as well as within a subcohort consisting of 1575 UK patients) using the PrediXcan software package with three different reference panels. Estimated expression was tested for association with anti-TNF treatment response. As a replication/validation experiment, we also investigated the correlation between change in ESR with measured gene expression at the Interleukin 18 Receptor Accessory Protein (IL18RAP) gene in whole blood and synovial tissue, using an independent replication data set of patients receiving conventional synthetic disease modifying anti-rheumatic drugs, with directly measured (via RNA sequencing) gene expression.

Results: We found that predicted expression of IL18RAP showed a consistent signal of association with treatment response across the reference panels. In our independent replication data set, IL18RAP expression in whole blood showed correlation with the change in ESR between baseline and follow-up (r=-0.35, p=0.0091). Change in ESR was also correlated with the expression of IL18RAP in synovial tissue (r=-0.28, p=0.02).

Conclusion: Our results suggest that IL18RAP expression is worthy of further investigation as a potential predictor of treatment response in rheumatoid arthritis that is not specific to a particular drug type.

Keywords: pharmacogenetics; rheumatoid arthritis; treatment.

Figure 1

Manhattan plots of p values from tests of association between genetically regulated gene expression and the change in erythrocyte sedimentation rate for the 1575-person UK data set. The genetically regulated gene expression was estimated with (A) the MATURA reference panel, (B) the GTEx reference panel and (C) the DGN reference panel. On each panel, the red dashed line represents the experiment-wide significance level computed using a Bonferroni correction for the number of tests performed. The black diamond represents the IL18RAP gene. The white diamond represents the ARV1 gene.

Figure 2

Manhattan plots of p values from tests of association between genetically regulated gene expression and the change in erythrocyte sedimentation rate for the 3158-person expanded European ancestry data set. The genetically regulated gene expression was estimated with (A) the MATURA reference panel, (B) the GTEx reference panel and (C) the DGN reference panel. On each panel, the red dashed line represents the experiment-wide significance level computed using a Bonferroni correction for the number of tests performed. The black diamond represents the IL18RAP gene. The white diamond represents the ARV1 gene.

Figure 3

Confirmation of the IL18RAP expression quantitative trait locus and clinical consequences in rheumatoid arthritis. (A) Manhattan plot showing expression quantitative trait locus analysis comparing influence of SNPs at the IL18RAP locus on IL18RAP expression in blood measured by RNA-seq. (B) Scatter plot of SNP rs10439410 in the 5’ upstream region of IL18RAP and IL18RAP expression in whole blood. (C and D) Correlation between the change in erythrocyte sedimentation rate between baseline and 6 months of combination disease modifying anti-rheumatic drug therapy and IL18RAP expression measured by RNA-seq in whole blood (C) and synovial tissue (D).