Background Good evidence shows that treating the inflammation of rheumatoid arthritis (RA) early leads to less disability and improved patient outcomes . However, for patients receiving biologic therapies, only ∼30% will have a good response to the first drug tried ; currently there is no way to predict which drug will be effective in each individual patient. As RA is an autoimmune disease, we hypothesise that the immunophenotype can be used to inform selection of the treatment most likely to be effective in individual patients.
The first step is to identify immunophenotypic differences between RA patients with very active disease and controls in order to inform hypotheses as to which immune markers may be important in informing selection of effective biologic therapy.
Objectives 1) To compare the immunophenotype of RA patients about to start biologic therapy with controls.
2) To characterise the involvement of novel cell types such as invariant natural killer T (iNKT) cells and co-stimulatory pathways (CD200/CD200R) in RA compared to healthy controls.
Methods Peripheral blood mononuclear cells (PBMCs) were collected from patients about to be started on a biologic drug as part of the BRAGGSS study . PBMCs were stimulated under a number of different conditions and analysed by 16-parameter flow cytometry for:
a) Cell surface profiles (major T and B cell subsets, classical and non-classical monocytes, natural killer cells, dendritic cells, markers for iNKT cells and the expression of CD200 and CD200R), and
b) Intracellular cytokine quantification, following a stimulation assay (IFNγ, TNFα, IL-13, IL-10, IL-17A, GM-CSF, IL12-p40, IL-6).
Results 7 RA patients and 9 healthy control samples were examined in this pilot study.
There were large differences in the number of IFNγ-producing T cells in all samples, in both RA patients and healthy controls (range 8.19% to 65.1% in RA and 17.9% to 49.0% in controls,). There were also differences in the number of IL-17A-producing T cells, although absolute numbers were lower (range: 1.2% to 12%). Samples could be placed into four phenotypes: 1) “IFNγ”, 2) “IL-17A”, 3) “double-hi”, and 4) “low”.
Interestingly, there was also a large range in number of IFNγ-producing B cells in all samples (mean fluorescence intensities (MFI) ranged from 202 to 1197).
Follow up data was available for 7 patients. Five had been started on an anti-TNF, one on tocilizumab and one on rituximab. Four had a good EULAR response, had been started on an anti-TNF, and had either the “double-hi”, “IFNγ”, or “low” phenotypes. One non-responder had been started an anti-TNF but had the “IL-17” phenotype. The other non-responder had been started on rituximab and had both the “IL-17” phenotype and a low IFNγ B cell signature.
Conclusions RA patients may have very different B and T cell cytokine profiles on ex vivo stimulation of their PBMCs. This may be important in selecting the first biologic drug choice: For instance, a high IFNγ B cell phenotype might make rituximab a more attractive first choice, and so forth. These results require validation with larger sample numbers.
Stoffer M.A. et al. Ann Rheum Dis. 2016;75(1):16–22.
Rubbert-Roth A. and Finckh A. Arthritis Res Ther. 2009;11 Suppl 1:S1.
Potter C. et al. Ann Rheum Dis. 2009;68(1):69–74.
Disclosure of Interest None declared