Network analysis of synovial RNA sequencing identifies gene-gene interactions predictive of response in rheumatoid arthritis

Abstract

BackgroundTo determine whether gene-gene interaction network analysis of RNA sequencing (RNA-Seq) of synovial biopsies in early rheumatoid arthritis (RA) can inform our understanding of RA pathogenesis and yield improved treatment response prediction models.MethodsWe utilized four well curated pathway repositories obtaining 10,537 experimentally evaluated gene-gene interactions. We extracted specific gene-gene interaction networks in synovial RNA-Seq to characterize histologically defined pathotypes in early RA and leverage these synovial specific gene-gene networks to predict response to methotrexate-based disease-modifying anti-rheumatic drug (DMARD) therapy in the Pathobiology of Early Arthritis Cohort (PEAC). Differential interactions identified within each network were statistically evaluated through robust linear regression models. Ability to predict response to DMARD treatment was evaluated by receiver operating characteristic (ROC) curve analysis.ResultsAnalysis comparing different histological pathotypes showed a coherent molecular signature matching the histological changes and highlighting novel pathotype-specific gene interactions and mechanisms. Analysis of responders vs non-responders revealed higher expression of apoptosis regulating gene-gene interactions in patients with good response to conventional synthetic DMARD. Detailed analysis of interactions between pairs of network-linked genes identified the SOCS2/STAT2 ratio as predictive of treatment success, improving ROC area under curve (AUC) from 0.62 to 0.78. We identified a key role for angiogenesis, observing significant statistical interactions between NOS3 (eNOS) and both CAMK1 and eNOS activator AKT3 when comparing responders and non-responders. The ratio of CAMKD2/NOS3 enhanced a prediction model of response improving ROC AUC from 0.63 to 0.73.ConclusionsWe demonstrate a novel, powerful method which harnesses gene interaction networks for leveraging biologically relevant gene-gene interactions leading to improved models for predicting treatment response.