Background The pro-inflammatory cytokine IL-17A is a central driver of pathology in human spondyloarthritis (SpA). IL-17A production was originally proposed to be dependent on the upstream cytokine, IL-23. Emerging preclinical and clinical evidence from SpA-related diseases suggest, however, that IL-17A and IL-23 have a partially overlapping but distinct biology.
Objectives Here, we aimed to assess to what extend pathogenic IL-17A is dependent on IL-23 in SpA by selectively targeting the IL-23R in the HLA-B27/Huβ2m transgenic rat model of SpA, which we showed previously to be IL-17A-dependent.
Methods HLA-B27/Huβ2m tg rats were immunized with low dose heat-inactivated M. tuberculosis/IFA. Rats were treated with a depleting anti-mouse/rat chimeric IL-23R antibody or PBS in a prophylactic (treatment initiation after immunization, before disease onset) or therapeutic (treatment initiation after disease onset) experiment. Clinical measurements included spondylitis and arthritis scores and hind paw swelling (plethysmometry). At the end of the study spleen and lymph nodes were used to assess cytokine expression, serum samples were analyzed for exposure to anti-IL23R.
Results In the prophylactic treatment strategy, 58% and 67% of the rats in the control group developed spondylitis and arthritis, respectively. The average arthritis score at the end of the study was 3.9±1.1 and the average hind paw swelling was 0.35±0.09 cm3. Prophylactic treatment with anti-IL-23R completely protected the rats against the development of spondylitis as well as arthritis. In the therapeutic treatment strategy, however, anti-IL23R treatment failed to reduce the incidence or decrease the severity of experimental SpA (fig. 1). With an average increase in arthritis score after the start of treatment of 1.6±2.8 versus 2.1±2.5 and an increase in paw swelling of 0.6±0.7 versus 0.3±0.6 cm3in anti-IL23R treated versus control animals. The differential effect of IL-23R targeting in the initiation phase versus established disease could not be explained by pharmacokinetic differences as serum analyses revealed similar exposure levels. Mechanistically, the expression of presumably downstream effector cytokines such as IL-17A (p<0.05) and IL-22 (p<0.01) was reduced in the popliteal lymph nodes of rats treated prophylactically with anti-IL23R versus controls, with a similar trend in spleen. Accordingly, IL-17A production upon ex vivo re-stimulation was reduced in samples from prophylactically treated rats. In contrast, similar popliteal lymph node expression data in samples from the therapeutic experiment indicate a twofold increase in IL-17A expression and no difference in IL-22 expression in the anti-IL23R treated rats compared to controls.
Conclusions IL-17A expression and production is dependent on the IL-23 axis in the initiation phase of experimental SpA but not in established disease. Accordingly, targeting of this axis with an anti-IL23R antibody completely prevented the onset of arthritis and spondylitis in HLA-B27/Huβ2m transgenic rats, but failed to reduce axial and peripheral joint inflammation in established disease. The cellular origin of IL-23-independent IL-17A production in established disease and the relevance to human SpA remains to be further investigated.
Disclosure of Interest M. Van Tok: None declared, S. Na Employee of: Eli Lilly and Co, J. Taurog Grant/research support from: AbbVie, Anges, Inc, and Celgene, D. Baeten Grant/research support from: AbbVie, Pfizer, UCB, MSD, Roche, BMS, Novartis, Eli Lilly, Janssen, Glenmark, Boehringer-Ingelheim, Employee of: part-time employee of UCB, L. van Duivenvoorde Grant/research support from: Abbvie, Boehringer-Ingelheim