The IL-17 axis plays an important role in the immunopathology of spondyloarthritis (SpA) as indicated by a wealth of genetic, immunopathological, experimental, and translational evidence (1). Randomized clinical trials have formally demonstrated the role of IL-17A, the prototypical IL-17 cytokine, in human SpA. Treatment with Secukinumab (anti-IL-17A mAb) reduced signs and symptoms of ankylosing spondylitis (AS) as well as of psoriatic arthritis (PsA), the major forms of axial and peripheral SpA, respectively (2, 3). Similarly, ixekizumab has proven clinical efficacy in PsA (4) and AS (4). Recent evidence, however, suggests that other cytokines of the IL-17A family, with as prime example IL-17F, can contribute to inflammation in SpA (5). Albeit not proven clinically, animal models have also suggested a potential role for IL-22 (6, 7). This raises the question whether targeting other molecules in the IL-17 axis could enhance clinical efficacy above and beyond selective IL-17A blockade.

One of the key molecules controlling the production of IL-17A and related cytokines is the transcription factor retinoic acid receptor-related orphan receptor RORγt (murine) or RORC (the human homologue). RORC is involved in the production and regulation of IL-17A by different cell types including Th17 cells (8), γδ T cells (9), innate lymphoid cells (10). In animal models a distinctive group of RORγt+ Tregs that is vital to maintain gastrointestinal homeostasis and avoid colitis (11–13). Moreover, it is crucial in orchestrating the differentiation of naïve CD4 T cells to Th17 cells (8). Enhanced gene transcription of RORC increases IL-17A production in a T cell line and human primary cells (14–16), whereas blockade of RORC in human CD4 T cells suppresses IL-17A and other inflammatory cytokines, including IL-17F and IL-22 (17, 18). Expression of IL-23R, CCR6, and IL-26 were also decreased upon RORC inhibition in Th17 cells, without affecting the gene signature of other T helper cell types (18, 19). Genetic lack of RORγt protected mice against experimental autoimmune encephalomyelitis (EAE), induced defects in Th17 differentiation and prevented T-cell-transfer-mediated colitis (8, 20). Pre-clinical studies in animal models, including imiquimod-induced psoriasis (21), spontaneous colonic inflammation (22), antigen induced arthritis (17), EAE (23), and experimental autoimmune uveitis (EAU) (24), confirmed that RORC inhibition markedly reduces local and systemic IL-17A levels and decreases tissue inflammation. Moreover, expression of IL-17F and IL-22 was also reduced upon in vivo RORC inhibition (21, 23). These findings indicate that RORC could be an interesting therapeutic target in IL-17A dependent pathology, blocking not only IL-17A but also related pro-inflammatory cytokines.

In this study, we aimed to assess RORC as potential therapeutic target in SpA by studying the effects of a small molecule RORC inhibitor in experimental SpA in HLA-B27 transgenic (tg) rats, a well validated model for human SpA (25–27). As we previously showed that both initiation and disease persistence in this model is partially but not completely inhibited by IL-17A blockade (28), we used this model to assess how RORC inhibition affects a panel of IL-17 related cytokines in vivo and if this biological effect translates into clinical efficacy.

The major findings of the current study are 1) that IL-17A is primarily produced upon restimulation of splenocytes from MTB immunized rats, with a peak response after 7 days of immunization. 2) IL-17A expression could be reduced upon in vitro as well as in vivo RORC inhibition, additionally in vivo RORC inhibition reduced IL-17F and IL-22 expression without affecting other T helper cell related genes. 3) Despite these molecular changes RORC inhibition did not inhibit, but rather stimulated disease development. These findings are surprising considering previous studies of RORC inhibition in a variety of inflammatory models including antigen-induced arthritis (17), imiquimod-induced psoriasis (21), IL-23 induced skin inflammation (30), intestinal inflammation (22, 31). Recently, Tan et al. (24) showed the reduced clinical severity of experimental autoimmune uveitis (EAU), and EAE using two different RORc antagonists (CQMU151 and CQMU152) (24).

All these models showed decreased levels of IL-17A and significant reduction of clinical symptoms upon pharmacological RORC inhibition or in RORC deficient animals. Also in 2018, Taurog et al. demonstrated the therapeutic efficicacy of the RORγt antagonist (A-1619758) in reducing inflammation, and suppression of both axial and peripheral skeletal bone changes in the HLA-B27 transgenic rat deficient for Dazl gene (21-3x283-2x17-9) (32).

A major question is whether the data of the current study could have been biased by a technical or biological artefact. Exposure measurements indicated that all rats in the treatment group have high levels of RORC inhibitor present in their serum shortly before and a few hours after treatment. The control group responded as expected, although disease incidence and severity were low. We could detect a biological effect, in terms of reduced IL-17A, IL-17F and IL-22 expression and finally clinical data are in line with histological data. Together these findings prove that groups were not switched during or after this experiment. The potential mechanism remains unknown.

Others showed, using the exact same RORC inhibitor, that while IL-17A levels were reduced, the production of IL-22 continued (33, 34) and treatment with the RORC inhibitor abrogated experimental colitis (34). Specific subsets of iNKT and γδ T cells showed to be among the IL-22 producing cells upon RORC inhibition (33). Similarly another RORC inhibiting compound could selectively impact specific cell types, including Th17 cells, while local IL-17A or IL-22 production by ILC3s was not reduced in a mouse model for intestinal inflammation (22). IL-22 has been shown to be a key player in intestinal host defence and mucosal homeostasis (35), and as we did observe reduced IL-22 expression, it could by hypothesized that RORgt inhibition could result in gut inflammation by a decrease in IL-22. This gut inflammation could then induce the SpA phenotype in the rats, similar to the induction of disease by orchitis, which is a consistent finding in this animal model. However, the rats showed no weight loss (Figures 5A, B) and gut inflammation was absent after RORgt inhibition, which makes this hypothesis very unlikely.

In contrast, in another model for intestinal inflammation the RORgt inverse agonist (TAK-828F) suppressed intestinal inflammation. In this model the compound significantly reduced Th17 and Th17/Th1 cell population in mesenteric lymph nodes (MLN) which was accompanied by suppressed/decreased gene expression of not only IL17A, IL17F but also IL-22. Local effects on ILC3 function in the intestine was not studied. However, in normal mice this RORgt inverse agonist did reduce numbers of Th17 cells and IL3s in the laminal propria (31).

In order to test the responsiveness of cells in the MTB immunized HLA-B27 tg rat CD4⁺ T cells and γδ T cells were isolated from spleen and draining lymph nodes and restimulated in the presence of RORC inhibitor (n=9). IL-17A and IL-22 expression were similarly decreased in both cell subsets upon RORC inhibition.

It is important to note, however, that this is not the first time that clinical or preclinical data demonstrate that the IL-17 axis does not work in a linear manner. It was reported by Chong et al., that loss of IL-17A produced by Th17 cells didn’t reduce the pathogenicity of these cells as expected and instead increased the expression of other Th17 cytokines (i.e. GM-CSF and IL-17F) (36). This was attributed to a Th17 cell-intrinsic autocrine loop induced by IL-17A binding to its receptor resulting in IL-24 induction, which in turn repressed the Th17 cytokine program. They showed that in vivo IL-24 treatment ameliorated Th17-induced EAU, whereas silencing of IL-24 in Th17 cells enhanced disease. However, these findings are unlikely to explain the paradoxical effects of RORC inhibition in our model as we observed the down regulation of other TH17 cytokines (e.g. IL-17F).

Clinical studies in psoriasis patients indicated that both blockade of IL-23 as well as IL-17A are effective (37, 38). However in Crohns disease blockade of IL-23 was effective while blockade of IL-17A or IL-17RA was not (39–41). Finally in SpA, blockade of IL-23p19 seems to be effective in psoriatic arthritis (42) but not in ankylosing spondylitis (43). Similarly in a previous study with our HLA-B27 tg rats, we showed that blockade of the IL-23 receptor (IL-23R) significantly reduced levels of IL-17A and IL-22 (but not IL-17F) in vivo. While prophylactic blockade of the IL-23R completely prevented spondylitis and arthritis development, therapeutic blockade did not impact clinical or histological spondyloarthritis in the HLA-B27 tg rat (44). Moreover prophylactic as well as therapeutic blockade of IL-17A was effective in this model (28).

The RORγt+ Tregs subset might be involved in these unanticipated effects. The function of these regulatory cells was thought to be controlled partly by RORγt (45). Kleinewietfeld et al. showed that a low molecular weight RORγt inhibitor could influence the frequencies of regulatory T-cells (Tregs) beside the Th17 response. The equilibrium between RORγt and FoxP3 expression level would equally control Treg and Th17 cells, depending on the cytokine environment. RORγt inhibition would twist the Th17/Treg cell ratio towards the Treg pathway (46). Despite the apparent beneficial effect for inducing immune tolerance, it seems that RORγt modulation would alter both the downstream pro- and anti-inflammatory pathways and induce undesirable effects (45). Affecting the regulatory function of these RORγt Tregs would predispose to inflammation (even when the expression of Th17 signature genes and cytokines is decreased). The question remains to what extend these findings are relevant for other models or for human disease. Based on our previous findings regarding IL-23R and IL-17A blockade in the HLA-B27 tg rats we conclude that it is uncertain whether RORC inhibition might be a good therapeutic option for all IL-17A driven diseases.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

All animal experiments were approved by the AMC Animal Care and Use Committee, in line with national and international regulations and guidelines.

MT and LD, study design, experimental procedures, analyzing data, writing manuscript. MM, analyzing data, writing manuscript. JW, experimental procedures. ML, study design, experimental procedures, analyzing data. MS, analyzing data, writing manuscript. GN and DB, study design, analyzing data, writing manuscript. All authors contributed to the article and approved the submitted version.

This work was sponsored by Boehringer-Ingelheim Pharmaceuticals Inc., Ridgefield, CT, USA.

ML and GN are employees of Boehringer-Ingelheim Pharmaceuticals Inc. MS received consultancy fees from Novartis and Abbvie and research grants from Janssen, Novartis and Eli Lilly. DB is an employee of UCB.

The authors declare that this study received funding from Boehringer-Ingelheim Pharmaceuticals Inc. The funder shared in study design, data analysis and data collection.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.