Background/Purpose:  Autoimmune rheumatic diseases, including rheumatoid arthritis (RA), have multifactorial pathogenesis associated with failure of immune tolerance. Advances in understanding the disease immunopathology have led to the development of biological drugs targeting components of the immune response that drive disease progression (including TNFα, IL-6, and B cells). However, these drugs are not effective in all individuals and many patients fail to respond for unknown reasons. The ability to predict patient responses to conventional and to biological treatment would provide a crucial step change improvement in the clinical management of patients with RA. We proposed a novel immunophenotyping approach to identify unique ‘signatures’ that could stratify patients for a more personalised approach to treatment of this complex disease. 

Methods:  We optimised a novel high through-put immunophenotyping platform, LEGENDScreen^TM, measuring the expression of 332 cell surface markers on B-cell and T-cell populations. We used this multi-dimensional approach to screen 50 RA patients, either treated with DMARDs or with anti-TNFα treatments, compared to patients with MS, SLE and healthy donors (HCs) as controls. The antigen-specific fluorescence intensity for each subset was examined. Differentially expressed markers (DEMs) between patient groups were determined using multiple unpaired t-test analysis (p<0.05). The results guided us to further stratify the data looking at B cell subsets, disease activity and immunogenicity (presence or absence ADA, measured by ELISA).

Results:  We identified phenotypic markers specific for anti-TNFα-treatment-naïve RA patients, developing an “immune-signature” that differentiates these patients from HCs, RA patients treated with anti-TNFα-treatment, and from patients with MS or SLE. We observed that the majority of DEMs, with significantly different expression between patient groups, were on CD19⁺ B-cells rather than CD4⁺T-cells, confirming the contribution of B-cells in the pathogenesis of RA. Comparing HCs to anti-TNFα-treatment-naïve patients we identified 31 DEMs that were significantly differentially expressed on B-cells but only 2 DEMs on T-cells. Further stratification of B-cells, looking at immature, mature, and memory subsets, identified 40 and 41 DEMs on immature and mature B-cells respectively, and only 20 DEMs on memory B-cells. We identified a unique immune-phenotype associated with the lack of response to anti-TNFa and the development of ADA.

Conclusion:  We have developed a new strategy, which clustered patients based on shared immune-phenotypes and distinguished between patients who developed anti-drug antibodies (ADA+) and those with active vs inactive disease. The predictive capacity of these immune-signatures is being validated in a prospective cohort of patients treated with anti-TNFα, which will also predict response to switching therapies (e.g. from anti-TNFα to rituximab). We proposed to use this novel immunophenotyping signature for a more personalised approach to treatment of this complex disease. Conducted as part of the ABIRISK consortium.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/novel-immunosignature-stratifies-patients-with-rheumatoid-arthritis-into-distinct-disease-sub-groups-and-predicts-response-to-anti-tnf%ce%b1-therapies/