AB0009 Genomic signatures may be associated with vascular pathology associated with rheumatoid arthritis

Abstract

BackgroundAccelerated atherosclerosis and cardiovascular (CV) disease have been associated with rheumatoid arthritis (RA). Many genes have been implicated in atherosclerosis, RA or both. However, most of these studies described SNPs in CD40, SMAD3, HLADR, CTLA4 and other alleles. Very few studies on genetic signatures have been performed that would link RA and CV pathology. We have previously associated some genomic profiles with pathological carotid atherosclerosis (ccIMT), arterial stiffness (PWV) and brachial artery flow-mediated vasodilation (FMD). In other studies we have also found 165 genes that separated anti-TNF responder patients from non-responders.ObjectivesHere we looked for associations between clinical and “vascular” response to biologics and vascular pathology in RA patients.MethodsIn this study, 19 RA patients were treated with either etanercept (ETN) or certolizumab pegol (CZP) for one year. We separated responders (R) and non-responders (NR) according to EULAR response criteria. Microarray gene expression study was performed (Affymetrix) followed by analysis using the GeneSpring software, hierarchy clustering and principal component analysis (PCA). “Vascular response” (VR) to biologics was defined as an at least 20% improvement in FMD, ccIMT or PWV. Good Vascular Response (GVR) was defined as an at least 20% improvement in two or three of these variables.ResultsAmong the 19 patients, 13 were R and 6 were NR. With respect to VR, FMD, ccIMT and PWV responded to anti-TNF treatment in 10, 9 and 8 patients, respectively. GVR was observed in 8 patients and 5 patients had VR in all 3 parameters. When comparing clinical response and VR, 7 out of 8 patients showing GVR also had good clinical response to biologics. Up-regulation of 99 and down-regulation of 67 genes separated clinical R and NR patients. Significant correlation was found between ccIMT improvement upon biological therapy and clinical response (R=0.418, p=0.04).ConclusionsGenomic signature analysis may be able to separate clinical responders and non-responders to biologics, as well as patients that show or do not show imporvement of vascular pathology.Disclosure of InterestNone declared