Background: Rheumatoid arthritis is an immune-mediated chronic disease characterized by the inflammation of synovial joints, which in many cases ends up in the complete destruction of the joint. Prevalence in global population is estimated between the 0.5 and the 1% and is higher in women. Patients’ best treatment option used to be the disease-modifying anti-rheumatic drugs (DMARDs), being the most common used the anti-TNF (tumour necrosis factor) agents. However, about the 30% of the patients experience an unfavourable response to this treatment, developing side effects that, in many cases, are more harmful than the disease itself. Therefore, is crucial the elucidation of biomarkers that could accurately predict the response to anti-TNF therapies in rheumatoid arthritis patients.

Objectives: We hypothesized that transcriptomic data could be used as a source of biomarkers in rheumatoid arthritis patients. The objective is reporting a transcriptomic signature capacity to predict the response to anti-TNF treatment in patients with rheumatoid arthritis, prior to its initiation.

Methods: Initially, we obtained data from the Gene Expression Omnibus (GEO) database GSE138746, which contains 80 samples with gene expression profiling on PBMC, including 38 and 42 samples from RA patients treated with adalimumab and etanercept. We continued by performing an analysis of differential expressed genes to establish a transcriptomic signature. Then, we ranked these genes based on the discriminatory power of each of them and we predicted the accuracy of the most discriminatory genes. Our collaborators validated in this sample set the 6 gene small-scale signature, consisting in KCNK17, GLS2, GTPBP2, DNTTIP1, IL18R1 and COMTD1.

Results: A total of 53 differentially expressed genes were identified between responders and non-responders. However, after performing functional enrichment analysis, any significant relation was found between the DEGs and the response. We evaluated the discriminatory power of these genes to create a transcriptomic signature that could accurately distinguish between these patient group using ROC analysis. A predictive model of 6 genes was selected to effectively discriminate between responders and non-responders: COMTD1, DNTTIP1, GLS2, GTPBP2, IL18R1 and KCNK17. Finally, we validated these results using another GEO dataset, obtaining an elevated AUC that confirms the predictive power of our small-scale signature.

Conclusion: Although TNF inhibitors have revolutionized rheumatoid arthritis treatment, a significant proportion of patients still exhibit partial response or nonresponse, leading to considerable clinical and economic burdens. Therefore, there’s a urge of predictive biomarkers to guide individualized treatment strategies. This gene signature will help to predict response to anti-TNF in rheumatoid arthritis patients before the initiation of the therapy, reducing the duration a patient may spend on ineffective treatment, as well as prevent the potential development of negative side effects.

REFERENCES: NIL.

Acknowledgements: The authors would like to acknowledge to every participant in the development of this study, including all financial supporters, technicians and patients.

Disclosure of Interests: None declared.

© The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.