Introduction
T cells (especially CD4
+ T cells), monocytes and B cells are considered to be involved in the pathogenesis of rheumatoid arthritis (RA) [
1]. It is frequently considered that decreasing the number and/or activity of lymphocytes and other immune cells by RA treatment can reduce disease activity. The first evidence of this was seen in preliminary clinical trials in which targeting CD4
+ T cells with anti-CD4 monoclonal antibodies (mAbs) resulted in clinical improvement of RA, albeit only modestly [
2]. Abatacept, a cytotoxic T lymphocyte antigen 4 immunoglobulin recombinant fusion protein that inhibits CD4
+ T cell activation by blocking costimulation with antigen-presenting cells such as B cells and monocytes, showed clinical efficacy against RA and has been approved worldwide for the treatment of RA [
3]. Depletion of peripheral B cells by the anti-CD20 antibody rituximab also improves disease activity [
4].
The anti-interleukin-6 receptor (IL-6R) antibody tocilizumab (TCZ) is also clinically effective against RA [
5]. Several studies have shown that blocking the IL-6 signaling with TCZ can affect proportions of peripheral blood cells. Given that IL-6 was originally identified as a B cell differentiation factor [
6,
7], it is not surprising that TCZ affects proportions of B cell populations in patients with RA. IL-6 also influences differentiation of T cells into effector T cells (T
H1, T
H2 and T
H17 cells) or regulatory T cells (T
reg) [
8-
10]. In recent studies, researchers have shown that IL-6 blockade could favorably affect the T
H17/T
reg cell imbalance in patients with RA [
11,
12]. Moreover, IL-6 seems to affect the proliferation and activation of monocytes that express IL-6R [
13,
14]. However, because in previous studies the number of patients, the period of study and the examined cell populations were limited, it is not clear whether there is a key population of peripheral immune cells that attenuates RA clinical symptoms through anti-IL-6R therapies [
11,
12,
15]. If these relationships could be clarified, it would enable medical researchers to comprehend the pathogenesis of RA from the view of lymphocyte populations and to find surrogate markers in order to choose an optimal therapeutic strategy for RA.
The primary objective of this study was to evaluate multiple different types of peripheral blood cells by using flow cytometry analysis to identify populations modulated by anti-IL-6R therapy. The secondary objective was to determine whether any of these populations is strongly associated with various clinical measures in response to anti-IL-6R therapy.
Methods
Patients
Eligible patients were those who met the 1987 revised criteria of the American College of Rheumatology (ACR) for the classification of RA or the 2010 ACR/European League Against Rheumatism (EULAR) classification criteria [
16,
17]. Consecutive patients at our institute who commenced TCZ as their first biologic agent between March 2010 and April 2012 were enrolled. They all showed insufficient response to at least one conventional synthetic disease-modifying antirheumatic drug (csDMARD). The enrolled patients were administered 8 mg/kg TCZ every 4 weeks, either with or without other csDMARDs, including methotrexate (MTX). The study protocol was approved by the ethics committee at Keio University School of Medicine and was carried out in accordance with the Declaration of Helsinki and Good Clinical Practice. Written informed consent was obtained from all patients.
Clinical assessments and evaluation of effectiveness
Demographic and clinical characteristics including age, sex, disease duration, tender joint count (TJC), swollen joint count (SJC), patient global assessment (patient visual analogue scale (Pt-VAS)), physician global assessment (doctor’s visual analogue scale (D-VAS)), Health Assessment Questionnaire Disability Index (HAQ-DI) score, C-reactive protein (CRP) level, erythrocyte sedimentation rate (ESR), matrix metalloproteinase-3 (MMP-3) level, rheumatoid factor (RF) value and anticyclic citrullinated peptide (CCP) antibody value were obtained from the patients’ medical records.
Disease activity was assessed using the Clinical Disease Activity Index (CDAI) and Simplified Disease Activity Index (SDAI). The cutoff values for remission, low disease activity (LDA), moderate disease activity (MDA) and high disease activity (HDA) were as follows: for remission, CDAI ≤2.8, SDAI ≤3.3; for LDA, 2.8 < CDAI ≤ 10, 3.3 < SDAI ≤ 11; for MDA, 10 < CDAI ≤ 22, 11 < SDAI ≤ 26; and for HDA, CDAI > 22, SDAI > 26 [
18].
Cell surface staining and flow cytometry analysis
Peripheral blood mononuclear cells (PBMCs) were obtained at baseline and at weeks 24 and 52 of TCZ treatment. PBMCs were separated by density gradient with Ficoll-Paque Plus (GE Healthcare, Uppsala, Sweden) and cryopreserved in CELLBANKER 1 (Nippon Zenyaku Kogyo, Fukushima, Japan) until use. Thawed cells were stained for 30 minutes at room temperature under darkened conditions with the following fluorophore-labeled mAbs: anti-CD4-VioGreen (Miltenyi Biotec, Bergisch Gladbach, Germany); anti-CD3-Pacific Blue/fluorescein isothiocyanate (FITC), anti-CD8-Pacific Blue, anti-CD14-(APC)-Cy7, anti-CD20 allophycocyanin-cyanine 7 (APC-Cy7), anti-CD25 phycoerythrin (PE)-Cy5, anti-CD27-PE-Cy7, anti-CD38-PE-Cy5, anti-CD45RO-PE-Cy7, anti-CD56-PE/PE-Cy7, anti-CD69-APC/PE-Cy7, anti-CD80-FITC, anti-CD86-PE-Cy5, anti-CD127-FITC, anti-CD161-APC, anti-chemokine (C-X-C motif) receptor 3 (CXCR3)-PE and anti-HLA-DR-APC/APC-Cy7 (all from BD Biosciences, Franklin Lakes, NJ, USA); anti-CD16-Brilliant Violet 510 and anti-CCR6-Brilliant Violet 421 (both from BioLegend, San Diego, CA, USA); and anti-mouse immunoglobulin G isotype-matched controls (VioGreen from Miltenyi Biotec, the others from BD Biosciences).
Stained cells were washed twice with 2 ml of phosphate-buffered saline and analyzed on a MACSQuant analyzer (Miltenyi Biotec). Dead cells were confirmed with a propidium iodide fluorescence solution (Miltenyi Biotec) and excluded on the basis of scatter signals. The subsets analyzed were CD4 and CD8 T cells (including memory, effector and activation markers) and T
H1, T
H2, T
H17, T
reg, B cells, natural killer cells, and monocytes, including their subpopulations and activation markers. The peripheral cell subsets identified in this study were defined by using cell surface markers on the basis of peripheral cell subsets described in a previous report (Additional file
1: Table S1) [
19].
Statistical analysis
Continuous data are presented as median and interquartile range (IQR) or as a number with percentage value, as appropriate. The Wilcoxon test and Kruskal-Wallis test were used to examine the differences between continuous variables. Correlation of two continuous variables was analyzed using the Spearman rank correlation coefficient. Fisher’s exact test was used to compare the proportion of categorical data between groups. A P-value <0.05 was considered statistically significant. All statistical analyses were performed with JMP 10 (SAS Institute, Cary, NC, USA).
Discussion
In this article, we report the results of the first comprehensive study to show the effect of TCZ on various peripheral cell subsets. In this study, we have shown that the proportion of CD4+CD25+CD127low Treg cells among CD4+ cells and the proportion of HLA-DR+-activated Treg cells among Treg cells significantly increased from baseline over the course of treatment with TCZ, and also that the proportions of CD20+CD27+ memory B cells, HLA-DR+CD14+ activated monocytes, CD69+CD14+ activated monocytes, and CD16+CD14+ nonclassical monocytes significantly decreased from baseline. Among them, only the increase in Treg cells was significantly associated with achieving remission by TCZ treatment.
Related studies have reported the increase in T
reg cell levels 1 year after TCZ administration in a small population [
20], for 3 months [
15] and for 6 months [
12]. Though definitions of T
reg cells in our study, using only cell surface markers without stimulatory modification for cells
in vitro, were different from those in these previous studies, our results showed the same trends as previously reported. The mechanism is a question that is not yet fully understood. However, this is still an important biomarker, as no suitable biomarker that can help predict outcome of RA has not been identified.
Several mechanisms may contribute to the potentiation of T
reg cells by TCZ therapy. It is conceivable that IL-6 has an effect on T
reg differentiation via inhibiting the expression of a specific transcriptional factor, FoxP3 [
21,
22]. This may suggest a mechanism by which neutralizing of IL-6 signaling in patients with RA can induce an increase in the number and function of T
reg cells. The quantitative increase in T
reg cells expressing elevated levels of human leukocyte antigen (HLA)-DR in patients with RA treated with TCZ could also be a result of activation and proliferation of preexisting T
reg cells or of their differentiation by conversion from Foxp3
− precursors.
One question arising from the result is whether the increase of T
reg proportion after TCZ therapy is a result of disease remission or is attributable to use of TCZ. To answer the question, we also analyzed peripheral T
reg cells in 12 patients with RA longitudinally treated with MTX alone (Additional file
1: Table S4). All the patients achieved LDA or remission at week 52 after starting MTX therapy (Additional file
1: Figure S3). The proportion of T
reg cells did not show a certain tendency or change over 52 weeks after administration of MTX (
P = 0.729) (Additional file
1: Figure S4A). In addition, the change in T
reg cells at week 52 was not associated with the change in CDAI and SDAI scores at week 52 (Additional file
1: Figure S4B). These results suggest that, albeit the comparison in a small number of patients with MTX, the increase in T
reg proportion after TCZ therapy is not a result of disease remission but caused by TCZ therapy itself.
Another question is whether other biologic agents for RA treatment also induce T
reg cells. It has been reported that neither adalimumab, an anti-tumor necrosis factor (anti-TNF) mAb, nor etanercept, a soluble TNF receptor, modified percentages or absolute numbers of circulating CD4
+CD25
high T
reg cells or other T
reg phenotypes after being administered for 6 and 12 weeks to patients with RA, regardless of their response [
23]. However, there is one report that high concentrations of TNFα can block the immunosuppressive functions of T
reg cells
in vitro and that the treatment of patients with RA with infliximab, an anti-TNF mAb, bolsters T
reg suppression of the proliferation of effector cells [
24]. It is assumed that these conflicting results might be ascribed to a small number of subjects (
N = 10 to 30) and a short observational period (12 to 24 weeks). Regarding abatacept, the therapy diminishes the absolute numbers of T
reg cells but enhances their function in patients with RA [
25]. It is conceivable that the different targets of these therapies influence the different performance of T
reg cells and that the results of the present study indicate a part of the unique mechanism of TCZ.
Although much still remains to be clarified about how Treg defects might contribute to the pathogenesis of RA, approaches that specifically boost Treg activity could be useful in the treatment of RA. In this study, the change in Treg was correlated with disease activities after TCZ therapy. This is the first time that the effect of increasing Treg has been shown to be strictly associated with clinical efficacy.
The effects of IL-6 on the late stages of B cell differentiation
in vitro are well documented [
26].
In vivo IL-6 overexpression is associated with B cell hyperactivity, autoantibody production and immunopathology [
27,
28]. In patients with RA, chronic activation of B cells and an accumulation of memory B cells in the peripheral blood and synovial membranes have been described [
29,
30]. Within this context, B cell–targeted therapies utilizing rituximab have been widely explored in RA. Because IL-6 has been described as an important B cell–stimulating factor with effects on memory B cell survival and on plasma cell differentiation and survival in the bone marrow, it is easy to comprehend the effect of TCZ on peripheral B cells, especially the ratio of naïve B cells to memory B cells [
31]. Although the proportion of memory B cells significantly decreased over 52 weeks of TCZ therapy, it did not correlate with any component of activity status, SJC, TJC, Pt-VAS, D-VAS, CRP and ESR. Therefore, the decrease can be attributed to the effect of TCZ therapy rather than to disease activity. When we compared the proportion of B cell subsets in the same 12 patients that were effectively treated with MTX alone during 52 weeks as mentioned above (Additional file
1: Table S5), we observed that the proportion of memory B cells tended to decrease in patients with MTX therapy, as in the case with TCZ, suggesting that the trend was not specific to TCZ therapy. However, the proportions of CD80
+ and CD86
+ B cells among all B cells did not change in patients who received MTX therapy. Therefore, the decrease in the proportion after TCZ therapy may be characteristic of TCZ.
In peripheral blood, two monocyte subpopulations with distinct functional properties have been defined by their expression of CD14 and CD16 molecules. Compared with classical CD14
+CD16
− monocytes, CD16
+ nonclassical monocytes have been shown to possess several features of inflammatory tissue macrophages, notably, higher expression of major histocompatibility complex class II antigens and several adhesion molecules and lower expression of IL-10, transforming growth factor β, macrophage colony-stimulating factor, IL-1β and TNFα [
32]. The pathophysiologic significance of the CD16
+ nonclassical monocyte subset has been demonstrated by its expansion under various inflammatory conditions, such as RA, sepsis, asthma and solid tumors. We demonstrated that TCZ reduced the peripheral level of CD16
+ nonclassical monocytes. Although it remains to be clarified how monocytes differentiate into CD16
+ nonclassical monocytes, this study revealed that IL-6 appears to be involved in the proliferation of CD16
+ nonclassical monocytes or in shifting the balance of monocytes to CD16
+ nonclassical monocytes. In the same 12 patients with RA described above who were treated with MTX alone, the proportions of HLA-DR
+CD14
+ monocytes and CD69
+CD14
+ monocytes among all monocytes did not change (Additional file
1: Table S5), which was different from the results for TCZ. The decrease of those activated monocytes after TCZ therapy did not seem to be the cause of disease remission, but rather the effect of TCZ. The proportion of CD16
+CD14
+ nonclassical monocytes tended to decrease during 52 weeks of MTX therapy compared with that of TCZ therapy. This may have been a result of the improvement of RA disease activity, or there may some actions on immune cells in common between TCZ and MTX.
Competing interests
JK, KY, NN and YK declare that they have no competing interest. MH is an employee of Chugai Pharmaceutical Co, Ltd. TT has received grants from Astellas Pharma Inc, Bristol-Myers KK, Chugai Pharmaceutical Co, Ltd, Daiichi Sankyo Co, Ltd, Eisai Co, Ltd, Mitsubishi Tanabe Pharma Corp, Pfizer Japan, Santen Pharmaceutical Co, Ltd, Takeda Pharmaceutical Co, Ltd, Teijin Pharma Ltd, AbbVie GK, Asahi Kasei Pharma Corp, Taisho Toyama Pharmaceutical Co, Ltd, and SymBio Pharmaceuticals Ltd; speaking fees from AbbVie GK, Bristol-Myers KK, Chugai Pharmaceutical Co, Ltd, Eisai Co, Ltd, Janssen Pharmaceutical KK, Mitsubishi Tanabe Pharma Corp, Pfizer Japan Inc, Takeda Pharmaceutical Co, Ltd, Astellas Pharma Inc, and Daiichi Sankyo Co, Ltd; and consultant fees from Astra Zeneca KK, Eli Lilly Japan KK, Novartis Pharma KK, Mitsubishi Tanabe Pharma Corp, Asahi Kasei Medical Co, Ltd, AbbVie GK, Daiichi Sankyo Co, Ltd, and Bristol-Myers KK.
Authors’ contributions
JK carried out flow cytometry, participated in the design of the study, analyzed the data, performed the statistical analysis, evaluated the results and drafted the manuscript. MH carried out flow cytometry, participated in the design of the study, analyzed the data, evaluated the results and drafted the manuscript. YK participated in the design of the study, evaluated the results and helped to draft the manuscript. KY participated in flow cytometry, evaluated the results and helped to draft the manuscript. NN analyzed the data, evaluated the results and helped to draft the manuscript. TT conceived of the study, participated in its design and coordination, evaluated the results and helped to draft the manuscript. All authors read and approved the final manuscript.