Sonia Dupont

Discovery and Optimization of an Azetidine Chemical Series As a Free Fatty Acid Receptor 2 (FFA2) Antagonist: From Hit to Clinic

FFA2, also called GPR43, is a G-protein coupled receptor for short chain fatty acids which is involved in the mediation of inflammatory responses. A class of azetidines was developed as potent FFA2 antagonists. Multiparametric optimization of early hits with moderate potency and suboptimal ADME properties led to the identification of several compounds with nanomolar potency on the receptor combined with excellent pharmacokinetic (PK) parameters. The most advanced compound, 4-[[(R)-1-(benzo[b]thiophene-3-carbonyl)-2-methyl-azetidine-2-carbonyl]-(3-chloro-benzyl)-amino]-butyric acid 99 (GLPG0974), is able to inhibit acetate-induced neutrophil migration strongly in vitro and demonstrated ability to inhibit a neutrophil-based pharmacodynamic (PD) marker, CD11b activation-specific epitope [AE], in a human whole blood assay. All together, these data supported the progression of 99 toward next phases, becoming the first FFA2 antagonist to reach the clinic.

AB0506 Identification of a Gene Signature and Response Biomarkers in Circulating Leukocytes of RA Patients After Treatment with the JAK1-Selective Inhibitor Filgotinib (GLPG0634)

Background The 4 Janus kinases (JAK1, JAK2, JAK3 and TYK2) are cytoplasmic tyrosine kinases that mediate intracellular signaling of cytokines (e.g. certain interleukins and interferons) and growth factors (e.g. erythropoietin). Filgotinib (formerly GLPG0634) is the first JAK inhibitor that displays a high selectivity for JAK1 over the 3 other JAK family members in biochemical and cellular assays. It showed a favorable safety and efficacy profile in two 4-week Phase 2a studies in rheumatoid arthritis (RA) patients. Objectives To assess the effect of RA disease effects, characterize the disease-relevant treatment responses to filgotinib and identify response biomarkers, we compared the gene expression profile of circulating leukocytes of RA patients, before and after up to 4 weeks of once-daily treatment with 200mg of filgotinib, with healthy volunteers. Methods RA patients participated in the Phase 2a Proof of concept, a randomized, double-blind, placebo-controlled study enrolling 24 patients with insufficient response to MTX. They were orally treated with placebo or 200 mg QD filgotinib for 4 weeks. Blood was sampled in PAXgene tubes at pre-dose and at the last day of treatment. Non-matched healthy volunteers were also sampled. mRNA was extracted, labeled and profiled using Affymetrix U219 micro-arrays. Data analysis was performed in R/BioConductor using linear regression models (limma). Results The leukocyte gene signature of 12 healthy subjects was first compared to the one obtained from the 24 RA patients prior to placebo or filgotinib treatment. Genes showing differential expression compared to healthy subjects allowed for definition of a disease signature, which includes novel players as well as genes already described in RA (TFPI or ITGA2B). Four weeks of treatment with filgotinib impacted the signal levels for 1005 probes (254 downregulated, 751 upregulated) in RA patient samples (p-value <0.05 and absolute log2-fold change >0.5 compared to pre-dose), while the signal levels of only 161 probes (86 down- and 75 upregulated) were impacted to the same extent in the placebo group. Several disease-relevant genes (CETP, CTSD, FSTL1, HTRA1, MYL9) were affected by filgotinib treatment. This was confirmed by qPCR. Response biomarkers could be identified by comparing strong responders and weaker responders based on percentage change in disease activity score 28 (DAS28). Functional classification of these biomarkers showed enrichment towards key RA pathways including regulation of immune cells, neutrophil recruitment, inositol phosphate metabolism and phosphatidylinositol signaling (PI3K). In addition, in the best responders, ECM-receptor interaction and focal adhesion pathways were enriched in the genes affected by filgotinib while in the weaker responders, regulation of actin cytoskeleton and TGF-beta signalling were affected by filgotinib. Conclusions Blood transcriptome analysis showed that 4-week filgotinib treatment partially reverses the disease effect and allows for the identification of treatment response biomarkers in circulating leukocytes of RA patients. Additional information to be obtained from the Darwin studies currently ongoing should provide more insights. Disclosure of Interest M. Ongenaert Grant/research support from: AbbVie, Employee of: Galapagos, S. Dupont Grant/research support from: AbbVie, Employee of: Galapagos, B. Vayssière Grant/research support from: AbbVie, Employee of: Galapagos, L. Meuleners Grant/research support from: AbbVie, Employee of: Galapagos, R. Brys Grant/research support from: AbbVie, Employee of: Galapagos, R. Galien Grant/research support from: AbbVie, Employee of: Galapagos

THU0116 Biological Effects of the JAK1 Selective Inhibitor GLPG0634 on Inflammation Markers in Arthritic Mice

Background Non-selective Janus kinase (JAK) inhibitors have shown long-term efficacy in treating rheumatoid arthritis (RA). However, tolerated doses and thereby efficacy are limited due to JAK2-driven side effects. Selective inhibition of JAK1 may be associated with an improved safety profile while maintaining clinical efficacy. Currently available information for JAK1 inhibitors is limited. GLPG0634 is a JAK inhibitor which displays a high selectivity for JAK1 over JAK2 in human whole blood assays. It is currently being developed for RA treatment and displayed good efficacy and tolerability in a 4-week Phase 2a study. Objectives Characterize the mode of action of a JAK1-selective inhibitor by measuring the impact of GLPG0634 on JAK pathway activity and inflammation parameters in blood and paws in the mouse CIA model. Methods Mice with established collagen-induced arthritis (CIA) were treated with GLPG0634 (50 mg/kg po, bid) or etanercept (10 mg/kg ip, 3 times a week). Disease progression was evaluated using clinical score and histological parameters. Expression of genes related to JAK pathways and markers of inflammation were measured in hind paws as well as in circulating leukocytes (WBC) using qRT-PCR. Luminex technology was used to measure a panel of 35 cytokines and chemokines in mouse serum. Results Oral treatment with GLPG0634 decreased inflammation in CIA in mice as measured by clinical scores and paw histology, with the same efficacy as etanercept. In paws of arthritic mice, an increase of 20- to more than 200-fold in the expression of inflammation markers (IL-6, IL-1β, TNFα, CXCL1) was found. A 30% to 45% decrease of their expression was observed in GLPG0634-treated arthritic mice. A similar expression pattern was observed for the osteoclast differentiation factor RANKL and for proteases linked to inflammation (MMP3, MMP13). These findings corroborate the bone and cartilage protective effects of GLPG0634 observed by histological analysis of paws. In addition, decreases from 30 to 70% in serum levels of cytokines and chemokines (IL-6, IP-10/CXCL10, MCP-1) in GLPG0634-treated arthritic mice confirmed the immunomodulation and anti-inflammatory effects. Of interest, the increased expression of JAK1-induced genes Mx1 and Mx2 in arthritic mice both in WBC and hind paws was abolished after treatment of animals with GLPG0634, whereas expression of the JAK2-dependent gene HRH4 was not altered by GLPG0634 treatment. These findings show that the decrease in inflammation markers was associated with inhibition of JAK1 but not JAK2. Conclusions These data show that oral GLPG0634 administration reduces inflammation in the mouse CIA model to the same extent as parenteral etanercept. Effects on several inflammation markers in paws and serum demonstrate the anti-inflammatory activity of GLPG0634. These effects were achieved through selective JAK1 inhibition. These data establish that selective JAK1 inhibition by GLPG0634 is sufficient to mediate strong efficacy in an established arthritis model and provide some useful potential markers to follow pathology progression in arthritic mice and humans. Disclosure of Interest B. Vayssiere Employee of: Galapagos SASU, S. de Vos Employee of: Galapagos NV, D. Merciris Employee of: Galapagos SASU, M. Auberval Employee of: Galapagos SASU, S. Dupont Employee of: Galapagos SASU, N. Vandeghinste Employee of: Galapagos NV, L. Lepescheux Employee of: Galapagos SASU, P. Clement-Lacroix Employee of: Galapagos SASU, P. Delerive Employee of: Galapagos SASU, L. van Rompaey Employee of: Galapagos NV, R. Brys Employee of: Galapagos NV, R. Galien Employee of: Galapagos SASU

FRI0539 A safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) study with increasing oral doses of glpg1972 administered daily for 29 days shows a strong biomarker effect in patients with KNEE and/or HIP OA

Background Osteoarthritis (OA) is characterised by structural changes of the joint, of which degradation of articular cartilage is one of the major signs1. The main proteoglycan component of the extracellular matrix of articular cartilage is aggrecan. GLPG1972 as a potent and selective inhibitor of ADAMTS-5, a key aggrecan-cleaving enzyme involved in cartilage degradation, is being developed as a potential disease-modifying OA drug (DMOAD). Aggrecan cleavage by ADAMTS-5 results in release of N-terminal ARGS neo-epitope fragments of which serum levels significantly decreased in healthy subjects treated with GLPG1972 during 14 days in a previous study². Objectives To assess safety, tolerability, PK and PD (serum ARGS-aggrecan levels) during and following administration of GLPG1972 in patients with knee and/or hip OA. Methods This was a single centre, randomised, double-blind, placebo-controlled, age and gender stratified, ascending dose Phase Ib study, with three semi-sequential cohorts of 10 patients each, randomised to active drug or placebo in a 4:1 ratio. Doses tested were once daily 100, 200 and 300 mg. Treatment duration was 29 days. Patients had follow-up visits 14 and 21 days after last dosing for additional PD assessments. Methods for PD have been described previously³. Results Thirty patients were included. Of these, 24 patients (M/F rate 8/16, 14 aged 50–64 and 10 aged 65–75) received active medication. All adverse events (AE) were mild and transient. No serious AEs were reported during the study; one female patient in the 300 mg group was discontinued after 15 days of treatment due to drug-related elevated transaminase values which returned to normal 9 days later while her bilirubin levels remained normal. There were no overall trends in lab abnormalities over time or significant changes in vital signs, ECG and Holter parameters. Steady state in plasma exposure was reached after 3 days of dosing. Exposure increased dose-proportionally. Mean serum ARGS levels (SEM) decreased steadily over time in all patients receiving GLPG1972: −40% (2.9),%–46% (4.5) and −53% (2.8) at day 15 compared to baseline in the 100, 200 and 300 mg group respectively. These levels remained stable until last dose on day 29, then consistently returned to pre-dose levels for all groups 14 and 21 days after last dose. Placebo group levels remained unchanged. Conclusions When administered daily for 29 days in patients with knee and/or hip OA, GLPG1972 at oral doses of 100, 200 and 300 mg q.d. was generally well tolerated and safe. Serum ARGS levels, as a marker for target engagement and potential proxy of cartilage degradation, showed a dose-dependent decrease over time up to 53% below baseline in the 300 mg group. These PK-PD findings are consistent with what we observed in a previous study in healthy subjects² and reinforce the rationale for developing GLPG1972 as a DMOAD. References [1] Hunter DJ, et al. Curr. Opin. Rheumatol. 2009;21:110–117. [2] van der Aar E, et al. Arthritis Rheumatol. 2017;69(suppl 10). [3] Larsson, et al. Osteoarthritis Cartilage2014;22(2):242–9. Disclosure of Interest H. Deckx Employee of: Galapagos NV, Belgium, S. Hatch Consultant for: Galapagos NV, Belgium, M. Robberechts Employee of: Galapagos NV, Belgium, S. Dupont Employee of: Galapagos SASU, France, J. Desrivot Employee of: has been employee of Galapagos SASU, France, H. Coleman Paid instructor for: Covance has been contracted by Galapagos NV to conduct the study, S. Larsson: None declared, A. Struglics: None declared, E. van der Aar Employee of: Galapagos NV, Belgium, A. Fiew Employee of: Galapagos NV, Belgium

DOP070 The FFA2 antagonist GLPG0974: opportunity to treat neutrophil-driven inflammation

Background: KdPT is a tripeptide with broad anti-inflammatory activity. Pre-clinical data indicated that KdPT is safe and efficacy was shown in different murine models of intestinal inflammation. KdPT given orally proved to be safe and well tolerated in previous clinical studies. Methods: We performed a multi-centre, randomised, doubleblind, placebo-controlled prospective phase II trial (2011 002462 20) to evaluate efficacy and safety of KdPT in patients with mild to moderate ulcerative colitis (UC). As add-on therapy to aminosalicylates, azathioprine, and/or corticosteroids placebo or KdPT in three different doses (20, 50, 100mg BID) was administered. 168 patients were randomized (ITT: n = 162; PP: n = 116) in 6 countries. The primary objective was to determine time to response to KdPT, defined as time from Day 0 to earliest treatment visit at which sustained improvement in colitis activity index (CAI) of 50% was determined. Improvement in CAI was to be 50% at Week 8, irrespective of any interim decline, for improvement to be classified as sustained. Due to an unusually high placebo rate from week 6 on assumptions for the statistical evaluation of the primary endpoint were violated, thus appropriate statistical tests were run to compare treatment groups. Results: The primary endpoint for pooled KdPT (PP) approached significance using Log rank (Renyi family) test (p = 0.0525, onesided) and showed statistical significance for the Wilcoxon (Renyi family) test (p = 0.0368, one-sided). At Week 2 and Week 4, remission rates were approximately twice as high for pooled KdPT compared to placebo (2 weeks: p = 0.0361; 4 weeks: p = 0.0178). For 20mg KdPT remission occurred earlier compared to 50mg or 100mg K(D)PT. CAI response rates lost significance at later time points due to unexpected high placebo rates from week 6 on. Patients with active and more severe disease (CAI 9; baseline medication: aminosalicylates + corticosteroids and/or azathioprine) showed earlier and more pronounced response to KdPT compared to placebo (p = 0.0156). KdPT was safe and tolerated well at the investigated doses with no difference between placebo and KdPT groups.

THU0138 Glpg0634m1, A Major Metabolite of the Jak1-Selective Inhibitor Glpg0634, is Also Jak1-Selective and Efficient in the Rat CIA Model

Background GLPG0634 is a JAK inhibitor displaying a high selectivity for JAK1 over JAK2 in human whole blood assays. It is currently being developed as an oral treatment for rheumatoid arthritis (RA) and showed good efficacy and tolerability in two 4-week Phase 2a RA studies. GLPG0634 has a major metabolite, GLPG0634m1, with a half-life which might contribute to its clinical efficacy in RA patients. Objectives Characterize the pharmacological properties of the GLPG0634m1, its activity on JAK-driven pathways and its efficacy in the rat CIA (collagen-induced arthritis) model. Methods Selectivity of the metabolite GLPG0634m1 was measured in vitro in radiometric recombinant kinase assays and ex vivo in whole blood assays (human, dog and monkey) by monitoring STAT phosphorylation by flow cytometry. Rats with established arthritis were treated with the GLPG0634m1 molecule (60 mg/kg po, QD) or etanercept (10 mg/kg ip, 3 times a week). The clinical score as well as histological parameters were used to monitor/quantify disease progression. Plasma concentration of the molecule was quantified by LC-MS/MS. Results Biochemical analysis of potency of GLPG0634m1 on recombinant JAK kinases showed that this compound is 10-fold less active against JAK1 and JAK2 than GLPG0634 with IC50s of 546 nM and 624 nM, respectively. Potency on JAK3 and TYK2 over 3 μM indicates that this molecule is more selective for JAK1 and JAK2 compared to JAK3 and TYK2. In human whole blood assays (WBA), GLPG0634m1 inhibited a JAK1-dependent event (IL-6-induced STAT1 phosphorylation) with an IC50 of 11.9 μM and a JAK2-dependent event (GMCSF-induced STAT5 phosphorylation) with an IC50 exceeding 100 μM, revealing the >10-fold selectivity of this molecule for JAK1 over JAK2. The JAK1 potency was confirmed with assays using IL2-induced STAT5 and IFNα-induced STAT1 phosphorylation, respectively triggering the JAK1/JAK3 and JAK1/TYK2 pathways. When given orally to rats with established arthritis, GLPG0634m1 displayed a pronounced efficacy comparable to etanercept, strongly reducing the impact of disease on e.g. paw swelling and Larsen score. These effects were confirmed at histological level with decreased pannus severity, bone and cartilage lesion as well as cell infiltration indexes. Plasma levels of GLPG0634m1 exceeded its WBA-derived JAK1 IC50 but were far below its WBA-derived JAK2 IC50, implying that the metabolite efficacy in the CIA model is driven by JAK1 inhibition, as observed previously with the parent molecule. Conclusions While displaying a lower potency compared to its parent molecule, GLPG0634m1 has a similar JAK1 selectivity. Oral administration of this metabolite, albeit at a 10-times higher dose than GLPG0634, reduces inflammation in the rat CIA model to the same extent as parenteral etanercept, an effect that is supported by JAK1 inhibition but appears independent of JAK2 inhibition. These findings, together with the high exposure and long half-life of this metabolite observed in phase 1 and phase 2 clinical studies in humans, strongly suggest that it may contribute to the clinical efficacy of the parent compound GLPG0634 in RA. Disclosure of Interest : C. Belleville-Da-Costa Grant/research support: Abbvie, Employee of: Galapagos SASU, D. Merciris Grant/research support: Abbvie, Employee of: Galapagos SASU, B. Vayssière Grant/research support: Abbvie, Employee of: Galapagos SASU, N. Houvenaghel Grant/research support: Abbvie, Employee of: Galapagos NV, A. Monjardet Grant/research support: Abbvie, Employee of: Galapagos SASU, L. Lepescheux Grant/research support: Abbvie, Employee of: Galapagos SASU, S. Dupont Grant/research support: Abbvie, Employee of: Galapagos SASU, T. Christophe Grant/research support: Abbvie, Employee of: Galapagos NV, M. Borgonovi Grant/research support: Abbvie, Employee of: Galapagos SASU, P. Clément-Lacroix Grant/research support: Abbvie, Employee of: Galapagos SASU, C. Menet Grant/research support: Abbvie, Employee of: Galapagos NV, L. Van Rompaey Grant/research support: Abbvie, Employee of: Galapagos NV, R. Brys Grant/research support: Abbvie, Employee of: Galapagos NV, R. Galien Grant/research support: Abbvie, Employee of: Galapagos SASU DOI 10.1136/annrheumdis-2014-eular.4291