Background/Purpose: There are many clinical similarities between polymyalgia rheumatica (PMR) and rheumatoid arthritis (RA). It has been reported that the specificity for discriminating PMR from RA was up to 70% with using 2012 EULAR/ACR provisional classification criteria. The recent technology of microarrays had made it possible to reveal the expression profiles of non-coding RNAs together with protein-coding RNAs (pcRNAs). So far, the differences of these expression profiles between PMR and RA have not been elucidated. In this study, we tried to identify the gene expression signature that may distinguish PMR from RA.

Methods: The study included 71 RA, 17 PMR and 11 osteoarthritis (OA) patients. Peripheral blood was drawn from the patients who were newly diagnosed. The samples were prepared and subjected to RNA extraction. Messenger RNA levels were then measured with using Agilent whole human genome 60K, and the log-transformed raw intensity data were normalized with a quantile algorithm. Based on the differences in gene expression among RA, PMR and OA by ANOVA, differentially expressed genes (DEGs) were selected, and then subjected to a hierarchical clustering with assessment of the statistical robustness. In order to discriminate RA from PMR, we also performed discriminant analysis with using DEGs selected by t-test.

Results: The hierarchical clustering showed major 3 clusters with using expression profiles of 556 pcRNAs selected from top1000 DEGs. OA samples were aggregated in the edge of 1^st cluster, while PMR samples were distributed among RA samples. In the top 100 DEGs between RA vs. OA, OA vs. PMR and PMR vs. RA, the long intergenic non-coding RNAs (lincRNAs) accounted for 7%, 9% and 26% respectively. In the comparison of PMR with RA, 86% of the 26 lincRNAs were upregulated in PMR. With using top 49 DEGs containing both pcRNAs and lincRNAs, PMR is differentiated from RA with retrospective accuracy of 98.9% by discriminant analysis. The accuracy was also calculated as 94.3% with leave-one-out (LOO) cross-validation. Meanwhile, with using only 14 lincRNAs, it was calculated as 92.0% for retrospective accuracy and as 86.4% for LOO cross-validation. In comparison with pcRNAs, performance of the discriminant analysis with using lincRNAs was better, if the number of applied genes was same. When the 26 lincRNAs were selected, expression profile-based hierarchical clustering of all samples showed major 3 clusters. Although PMR and OA samples were distributed in RA samples, most of PMR and OA samples were segregated into 1^st and 2^nd clusters respectively; while 3^rd cluster was consist only of RA samples. Analysis of clinical characteristics suggests that the averages of RF/CCP titers of patients in the 3^rd cluster were significantly lower as compare with RA patients in other clusters.

Conclusion: The similarity of gene expression profile between PMR and RA would give us the explanation of the difficulty in the differential diagnosis. However, if genes are selected properly, evaluation of expression profiles might be a promising tool for the diagnosis. Although we are only beginning to understand the nature and extent of the involvement of lincRNAs in diseases, our results inspire us to investigate about lincRNAs in rheumatic diseases.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/distinctive-expression-profiles-of-long-intergenic-non-coding-rna-between-polymyalgia-rheumatica-and-rheumatoid-arthritis-might-contains-helpful-suggestions-for-understanding-and-discriminating-these/