Rituximab (RTX) induces more than 98% depletion of the CD20 +B cells in blood after a single injection, yet 35% to 50% of RA treated patients show a poor response to the therapy1. Despite the identification of many different biomarkers, mostly in the B cell compartment2, adequate prediction of response to RTX treatment is still quite challenging.
Objectives
To test the hypothesis that non-response to rituximab can be predicted by analysing B-cell receptor (BCR) repertoire characteristics before and shortly after rituximab therapy.
Methods
Paired peripheral blood (PB) samples and synovial tissue (ST) samples were available from a total of 21 patients before therapy with RTX, and at 4 and 16/24 weeks after treatment. Next-generation sequencing was used to analyse the BCR repertoire, and asses the frequency of high expanded clones (HECs:>0.5% of the sequenced reads)3 and load of somatic hypermutation (SHM). Clinical response was evaluated at 6 month following EULAR response criteria.
Results
In spite of the complete depletion of B cells (measured using CD19) with conventional flow cytometry, we detect a complete BCR repertoire at week 4 and 16/24 after RTX treatment. The post-treatment PB BCR repertoire is composed of fewer, but more expanded and more mutated clones compared to baseline (figure 1). Non-response associates with a higher number of HECs at week 4 (p<0.01) and with a higher overlap in the top-50 clones between the baseline and week 4 repertoire (p=0.03). In fact, in all non-responders some of the HECs detected at week 4 were already present at baseline. In these persisting clones the SHM load was higher than the median in the total repertoire. In the synovial tissue BCR repertoire the number of clones and HECs does not significantly change after RTX treatment. Like in PB, an increase in SHM load is observed after treatment but at the later time point (week 16). In ST the overlap within the top-50 clones with baseline is largely maintained at week 4, but then decreases at week 16. No baseline predictors of response to RTX treatment were identified.
Conclusions
Incomplete depletion of the baseline BCR clonal repertoire in peripheral blood within the first month of treatment predicts poor clinical response at 6 months, revealing the persistence of "rituximab-resistant" BCR clonal signatures associated with treatment failure. In all patients the PB BCR repertoire at 4 weeks after rituximab is dominated by few but highly expanded and highly mutated BCR clones, most likely CD20-negative plasmablasts, while less pronounced and delayed effects are observed in the ST BCR repertoire.
References
[1] Cohen SB, et al. Arthritis Rheum2006;54:2793–806. [2] Benucci M, et al. Autoimmun Rev2010;9:801–3. [3] Doorenspleet ME, et al. Ann Rheum Dis2014;73:756–628.
Patients with rheumatoid arthritis (RA) show modulated circadian rhythms of inflammatory cytokines and cortisol suggesting potential involvement of clock genes (CG) in the pathophysiology of RA. In our study we aimed to examine the expression and localization of clock genes in synovial tissues of patients with RA compared to the expression in samples of patients with osteoarthritis (OA). Additionally we investigated the expression of certain clock genes in the synovial tissue of RA or psoriatic arthritis (PsA) patients before and after administration of common TNF inhibitors.
Methods
Synovial fibroblasts (SFs) of patients with RA or OA were synchronised by serum shock or treated with TNFα and gene expression profiles of different CG were determined over 24 hours by real time PCR. The expression of CG in RA and PsA patients was determined before and after administration of TNF inhibitors by real time PCR. Additionally, the in situ expression of clock genes in synovial tissues was examined by immunohistochemistry.
Results
Immunohistochemical analysis revealed that CG are expressed in B-cells, macrophages and fibroblasts but not in T-cells of the synovial tissue but the percentage share of expressing cells is comparable between RA and OA patients. Furthermore we could show that cultured human SFs can be synchronised by TNF a and that this synchronization differs from the synchronization under standard conditions (serum shock). The inflammatory condition with TNFa leads to phase shifts in the expression profiles of some CG and therefore differences in the expressional intensity at different time points. However, the application of TNF blockers such as infliximab in RA patients or adalumimab in PsA patients had no effect on the expression patterns of CG in synovial tissue. Remarkably, a dramatic downregulation of most CG was observed 48 hrs after start of therapy, both in the inflimiximab and the placebo group which might be caused by the first biopsy (i.e. mechanical stress) performed immediately before start of therapy or, alternatively, be related to the fact that participation in the study may create a stressful situation for the hospitalised patient. This phenomenon would certainly deserve further investigation.
Conclusion
Although TNFα can modulate CG expression in vitro, expression of CG does not appear to be altered in synovial tissue of RA patients and CG expression is not affected by anti-TNF therapy.
The angiogenic factors angiopoietin-1 (Ang-1) and Ang-2 are differentially expressed in the serum and synovial fluid of patients with different forms of arthritis. Ang-1 and Ang-2 contribute to inflammation and joint destruction in rheumatoid arthritis (RA), but the underlying mechanisms for their differential expression and their consequences for synovial Tie2 activation are unknown. Here the authors examined relationships between synovial expression of Ang-1 and Ang-2, and Tie2 activation in patients with RA and psoriatic arthritis (PsA), and the expression of transcription factors (TF) regulating Ang expression.
Materials and methods
Ang-1, Ang-2, Tie2 and phosphorylated (p) active Tie2 expression was examined by immunohistochemistry combined with digital image analysis in synovial biopsies from 20 RA and 19 PsA patients. Relationships between Ang expression and Tie2 activation were compared. In silico analysis was conducted to identify possible TF candidates in the Ang-1 promoter, using four different TF binding site prediction programmes. RA synovial expression of TF and Ang-1 was examined in publicly available gene expression data sets. Candidate TF were characterised by immunohistochemistry.
Results
Synovial Ang-1 expression was elevated approximately fourfold in patients with RA (p<0.05), while Ang-2 expression was similar in RA and PsA. The ratio of Ang-1 expression to Ang-2 was increased in RA (integrated optical density±SEM, 4.87±1.02) versus PsA (0.57±0.20) (p<0.001), and Ang-1 expression inversely correlated with Ang-2 expression in RA synovial tissue (r=−0.784, p<0.005). Tie2 activation was enhanced in RA (0.62±0.10) compared to PsA (0.18±0.11) (p<0.05), and a strong positive correlation was observed between Tie2 engagement and the ratio of Ang-1/Ang-2 expression (r=0.786, p<0.01). Expression of three potential Ang-1 TF, USF-1, NF-YA and Elf-1 correlated strongly with Ang-1 expression in RA synovial tissue (p<0.05) and were readily detected by immunohistochemistry.
Conclusions
In RA, synovial Tie2 engagement is driven primarily by enhanced local Ang-1 production. Expression of Ang-1 is differentially regulated in RA and PsA, correlating with enhanced expression of putative Ang-1 TF in RA.
Recently, synoviolin, a novel E3 ubiquitin ligase, was identified in rheumatoid arthritis (RA) fibroblast-like synoviocytes (FLS) and synovial tissue, where it may contribute to the dysregulated proliferation and apoptosis seen in RA.1 2 Synoviolin expression is also associated with arthritis development in mice.1 The exact mechanism of synoviolin regulation and expression remains to be elucidated, but recently a role for tumour necrosis factor α (TNFα) and interleukin 1b was suggested as these cytokines increased the expression of synoviolin in RA FLS.3 Another interesting observation is that high level sustained expression of synoviolin in whole peripheral blood was related to decreased clinical response in RA patients treated with TNF blockade.4 Therefore, we investigated synoviolin expression in synovium and its relationship to later response to TNF blockade in patients with RA compared with psoriatic arthritis (PsA) and osteoarthritis (OA).
Synovial tissue samples were obtained …
Immunological and genetic studies implicate T cells in the initiation and progression of rheumatoid arthritis (RA) [1]. Additionally, the diversity of the T cell repertoire is known to be reduced in RA, suggesting an enrichment of disease related clonal signatures [2].
Objectives
To study whether changes in the T-cell receptor repertoire (TCR) in lymph node (LN) and peripheral blood (PB) precede the onset of arthritis in seropositive at-risk individuals.
Methods
Unique molecular identifier (UMI)-based RNA sequencing of the TCRβ repertoire in LN biopsies and PB was performed in 9 healthy controls and 26 IgM-RF and/or ACPA positive at-risk individuals. Six of these individuals developed arthritis (RA) during at least 3 years of follow-up. PB and LN biopsies were analyzed during the at-risk phase and at RA onset.
Results
The TCRβ repertoire in LNs of autoantibody positive at-risk individuals harbors expanded TCRβ clones which can hardly be found in LNs of healthy controls. Regardless of whether the at-risk individuals developed arthritis or not, expanded TCRβ clones in LNs could be retrieved in peripheral blood. On RA onset, the LNs showed a marked decrease in the diversity of their TCRβ repertoire and, expanded TCRβ clones present during the at-risk phase could still be found in lymph nodes on arthritis onset.
Conclusion
Expanded TCRβ clones are present in lymph nodes and peripheral blood of individuals at risk of developing RA which can be retrieved in peripheral blood. Expanded TCRβ clones present in LNs during the at-risk phase persist on arthritis onset. Our analyses show a clear decrease in the diversity of the TCRβ repertoire towards onset of RA, implicating a key role for T-cell adaptive responses in steering onset of seropositive RA.
References
[1]Cope, A.P., T cells in rheumatoid arthritis. Arthritis Research & Therapy, 2008. 10(1): p. S1. [2]Jiang, X., et al., Comprehensive TCR repertoire analysis of CD4+ T-cell subsets in rheumatoid arthritis. Journal of Autoimmunity, 2020. 109: p. 102432.
Rheumatoid arthritis (RA) is a rheumatic disease mainly affecting women. Prolactin (PRL) is a sex hormone, which apart from inducing lactation, has also immunomodulatory properties. High prolactin levels are associated with an increased disease activity postpartum and some studies have shown that bromocriptin, decreasing prolactin levels, improves clinical activity of patients with RA. Furthermore, hyperprolactinemia is observed in 6% of RA patients compared to 3% in the normal population. Recently, the prolactin receptor (PRLR), belonging to the family of cytokine receptors, has been described in atherosclerotic plaques, mainly on macrophages. The objective of the study is to determine (1) the level of PRL in RA patients related to treatment effect (2) PRLR expression in synovial tissue of RA, psoriatic arthritis (PsA) and osteoarthritis (OA) patients (3) the phenotype of the PRLR expressing cell.
Material and methods
Serum prolactin levels were measured using immunofluorescent metric assay in patients with RA before and after tumour cecrosis factor α (TNFα) blockade (n=118). The expression of PRLR was determined in synovial tissue (ST), of RA (n=91), PsA (n=15) and OA (n=9) patients. Immunofluorescence (IF) was used to detect the PRLR expressing cell type.
Results
Hyperprolactinemia was observed in 4,2% of the patients with RA (PRL level 16–36 µg/l). Prolactin level is respectively highest in premenopausal, postmenopausal females compared to male. The level of prolactin was decreased in the group of responders compared to the non-responders to TNF treatment, respectively 7.0 (2.0–36) and 8.5 (4.0–19) (median (range) µg/l; p=0.048). Higher tertiles of prolactin levels, still within physiological range, were associated with reumafactor positivity (p=0.031), anticyclic citrullinated peptide (p=0.075) and erosive disease (p=0.095). The number of patients expressing PRLR in the synovium was comparable between RA and PsA (66% and 73%, respectively) versus 25% of the patients with OA (p=0.050). The levels of PRLR expressions were significantly higher in RA and PsA compared to OA, 0.055 (0.000–5.673), 0.182 (0.000–5.336) and 0.000 (0.000–0.908) (median (range) IOD/nuclei per mm2; p=0.024), see figure 1. There was no significant difference in PRLR expression between males and (pre/postmenopausal) females. Using IF, colocalisation was observed with markers of macrophages and endothelial wall.
Conclusion
Higher levels of prolactin were found in patients not responding to anti-TNF treatment. The expression of the prolactinreceptor in synovial tissue, mainly by macrophages, is higher in the inflammatory diseases (RA and PsA) compared to OA. Our data suggest a role of prolactin as sexhormone in rheumatoid arthritis.
Rheumatoid arthritis (RA) mainly affects women. Prolactin (PRL) is a sex hormone with immunomodulatory properties. High prolactin levels are associated with increased disease activity postpartum, and that the PRL-inhibitor bromocriptine improves disease activity of patients with RA. Hyperprolactinemia is observed in 6% of RA-patients, compared to 3% of healthy individuals. The prolactin receptor (PRLR), belonging to the family of cytokine receptors, has been described in atherosclerotic plaques, mainly on macrophages.
Objectives
The objective of the study is to determine 1) the level of PRL in RA-patients related to treatment effect 2) PRLR expression in synovial tissue of RA, psoriatic arthritis (PsA) and osteoarthritis (OA) patients 3) the phenotype of the PRLR expressing cell.
Methods
Serum prolactin levels were measured using immunofluorescent metric assay in patients with RA before TNF-α blockade (n=98). The expression of PRLR was determined in synovial tissue (ST) of RA (n=91), PsA (n=15) and OA (n=9) patients using digital image analysis. Immunofluorescence (IF) was used to detect the PRLR expressing cell type.
Results
Hyperprolactinemia (PRL-level: 16-24 μg/L) was found in 3.8% of the patients with RA. Prolactin levels were highest in premenopausal compared to postmenopausal females and males. Baseline PRL-levels were significantly lower in responders (median (range): 7.0 (2.0-24) μg/L) than in non-responders (9.3 (4.0-19) μg/L)) on TNF treatment (P=0.009). Higher tertiles of PRL (but within the physiological range) were associated with RF-positivity (P=0.005), aCCP-positivity (P=0.06) and erosive disease (P=0.024). After adjustment for these potential confounders, and for baseline-DAS28, baseline-PRL appeared to be a predictor of non-response to anti-TNF treatment (OR: 4.5; P=0.018; table 1). RF and aCCP did not independently contribute. The proportion of patients expressing PRLR in the synovium was similar in RA (66%) and PsA (73%) patients, and lower in OA patients (25%; P=0.05). PRLR expression was higher in RA (median (range): 0.055 (0.000-5.673) IOD/nuclei/mm2) and PsA (0.182 (0.000-5.336)) compared to OA (0.000 (0.000-0.908); P=0.024). Males and (pre-/postmenopausal) females had similar PRLR expression. Using IF, co-localisation was observed with macrophages and endothelial cells.
Conclusions
Higher levels of PRL independently predicts a non-response to anti-TNF treatment. The expression of the PRLR in synovial tissue, mainly by macrophages, is higher in the inflammatory diseases (RA and PsA) than in OA. Our combined data suggest an important role of prolactin and its receptor in RA.
T-cells are thought to be key players in the initiation and progression of rheumatoid arthritis (RA). Earlier we showed that already at the seropositive "at risk" stage uninflamed synovial tissue contains T-cell infiltrates1. In another study we showed that inflamed synovium selectively harbours expanded T-cell clones that are hardly present in paired blood samples2.
Objectives
Following up on these observations, we longitudinally investigated whether the same expanded T-cell clones found in the inflamed synovial tissue at onset of RA are already present in the synovium in the seropositive "at risk" stage.
Methods
Fifty-five individuals without arthritis but seropositive for IgM rheumatoid factor and/or anti-citrullinated protein antibody (ACPA) were prospectively followed. In five aCCP+ individuals synovial biopsies and paired blood samples at inclusion ("at risk" stage) and after development of RA (ACR2010 criteria; mean time to arthritis 27 months (range 11.7–47.3)) were available for analysis. T-cell clones were identified by their unique TCRβ sequence using RNA-based next generation sequencing3. For each sample, 3570 TCRβ sequences were analysed. Clones with a frequency of ≥0.5% were arbitrarily considered as highly expanded clones (HECs). ANOVA and t-test were used for statistical analysis.
Results
T-cell repertoires in "at risk" and RA synovium were similar (mean (± SD) number of clones 488±70 vs 567±204 respectively, p=0.46), number of HECs (37±7 vs 31±18, p=0.41) and the impact of HECs collectively on the TCR repertoire (mean 48% ± 13% vs 50% ± 20%, p=0.84). Interestingly, of the HECs present in the synovium at onset of arthritis 23% (±9%) were already present as HECs in the synovium at the seropositive "at risk" stage. This overlap was significantly higher than that with paired blood samples taken at the arthritis (3% ± 3%; p=0.01) or at the seropositive "at risk" stage (5% ± 7%; p=0.01; Figure 1a-c patient example; Figure 1d summary of results). Further characterization of the synovial CDR3 sequences (length, total charge, polar, aromatic and aliphatic side chains) showed no significant differences between RA HECs that were and those that were not expanded in the seropositive "at risk" stage.
Conclusions
Many T-cell clones found in early RA synovial tissue are already present in the pre-clinical "at risk" phase. The resemblance in TCR repertoires indicates that the process leading to disease – at least at the T-cell level – constitutes a smooth development. These clones, being already present in the very early stage of this disease and persisting as dominant clones during contraction of active arthritis, form attractive candidates for further characterization.
References
de Hair MJ et al. Arthritis Rheum. 2014. Klarenbeek PL et al. Ann Rheum Dis. 2012. Klarenbeek PL et al. Immunol Lett. 2010.
Macrophage migration inhibitory factor (MIF) has recently emerged as an important cytokine possibly linking rheumatoid arthritis (RA) and atherogenesis. Because atherogenesis is accelerated in RA this study was conducted to investigate whether anti-tumour necrosis factor (TNF) therapy could lead to sustained downregulation of systemic MIF levels and improvement in lipid profiles.Fifty RA patients with active disease (disease activity score in 28 joints (DAS28) >or=3.2), who started adalimumab therapy at 40 mg every other week, were included. At baseline, weeks 16 and 52 serum levels of MIF and lipids were assessed. In addition, the DAS28 and serum C-reactive protein (CRP) levels and erythrocyte sedimentation rate (ESR) were determined.After 16 weeks of adalimumab therapy, both DAS28 and MIF levels were significantly decreased (p<0.001 and p = 0.020, respectively). This was sustained up to week 52 (p<0.001 and p = 0.012, respectively). CRP levels and ESR were significantly reduced after 16 and 52 weeks of adalimumab therapy (p<0.001). High-density lipoprotein cholesterol levels increased at week 16 (p<0.001), but returned to baseline at week 52. Apolipoprotein (apo) A-I levels increased at week 16 (p<0.001) and remained stable (p = 0.005). This resulted in an improved apo B/A-I ratio.The results underline the sustained downregulation of MIF as a potential new mechanism by which anti-TNF therapy might reduce vascular inflammation, and as such perhaps cardiovascular morbidity in RA patients. This hypothesis is supported by an improved apo B/A-I ratio as well as reduced CRP levels in these patients.
Psoriatic arthritis (PsA) is an inflammatory joint disease associated with psoriasis. Alefacept, a lymphocyte function-associated antigen (LFA)-3 Ig fusion protein that binds to CD2 and functions as an antagonist to T cell activation, has been shown to result in improvement in psoriasis and has limited effectiveness in PsA. Interleukin-20 (IL-20) is a key pro-inflammatory cytokine involved in the pathogenesis of psoriasis. The effects of alefacept treatment on IL-20 expression in the synovium of patients with psoriasis and PsA are currently unknown.
Material and methods
Eleven patients with active PsA and chronic plaque psoriasis were treated with alefacept (7.5 mg per week for 12 weeks) in an open study. Skin biopsies were taken before and after 1 and 6 weeks while synovial biopsies were obtained before and after 4 and 12 weeks of treatment. Synovial biopsies from patients with rheumatoid arthritis (RA) (n=10) were used as disease controls. Immunohistochemical analysis was performed to detect IL-20 expression, and stained synovial tissue sections were evaluated by digital image analysis. Double staining with IL-20 and CD68 (macrophages) and CD55 (fibroblast-like synoviocytes, FLS) was performed to determine the phenotype of IL-20 positive cells in PsA synovium. IL-20 expression in skin sections (n=6) was analysed semi-quantitatively.
Results
IL-20 was abundantly expressed in both RA and PsA synovial tissues. In inflamed PsA synovium CD68+ macrophages and CD55+ FLS co-expressed IL-20 and its expression correlated with the numbers of FLS. IL-20 expression in lesional skin of PsA patients correlated positively with the Psoriasis Area and Severity Index (PASI). While IL-20 expression in PsA synovium was not affected by alefacept treatment, IL-20 expression in lesional skin decreased significantly (p=0.04) after 6 weeks of treatment.
Conclusions
Conceivably, the relatively limited effectiveness of alefacept in PsA patients might be explained in part by persistent FLS-derived IL-20 expression.
