Filgotinib

Clinical pharmacology of filgotinib in the treatment of rheumatoid arthritis: current insights

1. Introduction

Rheumatoid arthritis (RA) is a severe autoimmune inflammatory condition affecting around 1% of the population worldwide and causing progressive articular damage, high morbidity, and elevated socio–economical costs [1,2]. Synovitis, as inflamma- tion of the synovial tissue, represents the major manifestation of the disease, leading eventually to cartilage disruption, bone erosions, and severe disability [3–6]. Concurrently, the persis- tence of a chronic inflammatory response results in extra– articular manifestations and comorbidities in different domains (vascular, metabolic, bone and psychological) highlighting the systemic nature of the disease and its high mortality [1,7–12]. Up to date, the etiology of RA remains poorly understood [13–16]. Gene-polymorphisms together with environmental challenges create the substrate for the beginning of an auto- immune response. This results in a chronic synovitis character- ized by sustained influx of innate and adaptive immune cells, production of pro–inflammatory cytokines, activation of the resident stromal cells and failure in the spontaneous resolution of the inflammation [7,14,15,17–21]. Advances in the research field over the last decades provided us a better understand- ing of the pathophysiological mechanisms behind RA, with the identification of pivotal pro-inflammatory cytokines,such as tumor necrosis factor alpha (TNFa) and interleukin (IL)-6, as well as cell–associated molecules, such as CD20 and CD80/86, which have been successfully adopted as therapeutic targets [17,22–24]. Indeed, the treatment sce- nario of RA has been drastically improved with the intro- duction on the market of bioengineered drugs against key inflammatory mediators in the pathogenesis of the disease. Thanks to this change of prospective, the international scientific committee defined as primary goal both an early diagnosis of the disease and an early therapeutic interven- tion, adopting a treat-to–target approach, in which the treatment can be adjusted according to the individual ther- apeutic response, with the target to achieve remission or minimal disease activity [25–27].

The current clinical practice identifies methotrexate (MTX) as the most effective conventional synthetic disease- modifying antirheumatic drug (csDMARD), to be used in all new diagnosed RA patients as ‘anchor’ treatment [28]. The introduction of biologic disease-modifying antirheumatic drugs (bDMARDs) – including 5 TNF inhibitors, two IL-6 block- ers, one T-cell co-stimulation modulator, one IL-1 soluble receptor, and one B-cell depleting monoclonal antibody – significantly improved the possibility to achieve remission or low disease activity in MTX–insufficient responder (IR) patients [25,26,29–34]. However, the variety of bDMARDs available on the market seems to be still insufficient to guarantee remis- sion and/or minimal disease activity in a sustained number of patients. Real-data estimations suggested that about 50–70% of patients failed the therapeutic target, due to primary or secondary loss of response or discontinuation of therapy because of side effects or intolerance [33,35–38].

Therefore, the need of identifying new pathophysiological mechanisms and molecular candidates for the development of new therapeutic approaches came up. In the last few years, the research focus shifted from addressing the extracellular compartment and the mediators of cellular communication, to studying the intracellular area and the signaling cascades, which occur upon ligand-receptor binding. Here, a new class of drug was generated, consisting of low weight, orally avail- able ‘small molecules’ targeting intracellular kinases [34]. Up to date, the inhibitors of the Janus kinase (JAK) enzymes repre- sent the most successful class currently available [39,40]. Indeed, the JAK inhibitors (Jakinibs), classified as targeted synthetic disease-modifying antirheumatic drugs (tsDMARSDs), represent already a therapeutic strategy in alter- native to biologic treatments in MTX-IR patients [25,26,41].

The JAK family consists of four tyrosine kinases– JAK1, JAK2, JAK3 and TYK2-binding the intracellular domains of Type I and Type II cytokine receptors. Once a cytokine binds its own receptor, this goes through a multimerization step bringing different JAKs into proximity. This is a critical step for the activation and autophosphorylation of the different cytoplas- mic receptor-associated JAK pairs. As follows, the JAKs catalyze the phosphorylation of the receptor itself, which allow the recruitment and the further phosphorylation of the signal transducer and activator of transcription (STAT) family of DNA- binding proteins. Eventually, dimerized STATs enter to the nucleus and regulate the transcription of target genes [42].

The JAK-STAT pathway is involved in the pathogenesis of different autoimmune and inflammatory conditions. Indeed, pivotal cytokines for both innate and adaptive immune responses mediate their downstream effects through JAK- STAT signaling activation [43]. However, only specific groups of cytokines mediate their function through JAK-STAT signal- ing pathway, and these can be functionally distinguished according to their ability, upon receptor binding, to pair dif- ferent JAKs. Specifically, cytokines involved in lymphocyte proliferation, differentiation and homeostasis (such as IL-2, IL- 9, IL-21) cause JAK3 and JAK1 pairing, whereas cytokines and grow factors involved in differentiation and maintenance of hematopoietic stem cells (such as IL-3, granulocytes- macrophage colony-stimulating factor (GM-CSF), thrombo- poietin) convey their intracellular signal through JAK2 pairing with itself. Finally, cytokines mediating inflammatory responses lead to JAK1, JAK2 and TYK2 pairing (such as IL-6 and type-I interferons) [34,43]. JAK isoforms and STAT proteins are expressed in synovial tissue of patients with RA, suggest- ing their potential utility as therapeutic targets [39,44,45]. The first Jakinibs proposed and currently available on the market for treatment of RA are pan-Jakinibs [46]. These are competi- tive inhibitors of the active adenosine triphosphate (ATP) JH1 domain, which is a well-conserved structure among JAKs [47]. However, signal transmission of pro-inflammatory mediators involved in RA pathogenesis is mainly dependent on JAK1 signaling [48]. Hence, the recent interest is to develop small molecules with a selective JAK1 inhibitor activity, with the aim to reduce the dose-related toxicity due to the inhibition of other JAKs, without impairing the clinical efficacy [48]. High- resolution analysis of JAKs crystal structures was performed in order to identify specific amino-acid interactions within the ATP-binding domain, allowing the development of a second generation of Jakinibs with a more selective function on JAK1 [49].

2. Currently approved JAK inhibitors for the RA treatment

Jakinibs can be classified according to their ability to selec- tively inhibit JAK subtypes [47]. To date, three Jakinibs have been approved, as monotherapy or combination treatment with MTX, by both the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA), as alterna- tive of bDMARDs after failure and/or intolerance to one or more csDMARDs [50]. The first Jakinib available on the market were two pan-Jakinibs. Tofacitinib (approved by FDA in 2012 and by EMA in 2017) is a selective JAK1,3 inhibitor with a mild activity on JAK2 and TYK2, whereas baricitinib (approved by EMA in 2017 and FDA in 2018) is the result of tofacitinib structural modification and is a selective JAK1,2 inhibitor with moderate activity toward TYK2 and mild activity on JAK3 [51,52]. Randomized controlled trials (RCTs) demon- strated that both treatments are as efficacious as bDMARDs, showing a successful clinical efficacy in patients with RA [53]. Despite the reassuring safety data coming from RTCs and their extension, some concerns have been raised about the devel- opment of side effects following dose-dependent inhibition of multi-JAK, namely JAK2 and JAK3, such as increased rate of thromboembolic events or increased risk of herpes zoster (HZ) infections [54].

Thus, the development of a more selective generation of Jakinibs, specifically targeting JAK1 activity, with the aim to achieve the same clinical efficacy with a better safety profile. As mentioned above, JAK1 activity is important in mediating inflammatory cytokine signals relevant for RA pathogenesis, such as IL-6 (Figure 1) [55,56]. Upadacitinib is a selective and reversible JAKi engineered to have a stronger inhibitory capa- city on JAK1, with respect to JAK2, 3 and TYK2 [57]. To date, there are no head-to-head studies comparing the efficacy of the approved Jakinibs. However, a recently published network meta-analysis on RCTs reported that upadacitinib, with a dosage of 15 mg daily both in combination with MTX or in monotherapy, had a higher efficacy in terms of American College of Rheumatology (ACR) response criteria among the approved Jakinibs in patients with moderate-to-severe RA [50]. Furthermore, the specificity of the JAK1 inhibitors to selec- tively inhibit the kinase JAK1 without too much influence on the other subtypes remains still blurry, with data supporting a partial effect on JAK2 and JAK3 within therapeutic dosages on human cells [58,59]. Nevertheless, only robust clinical stu- dies involving head-to-head trials will ultimately determine whether there are clinically meaningful differences between these Jakinibs.

Figure 1. The role of JAK-STAT pathway in modulating the immune response.

3. Filgotinib as JAK1 inhibitor

Filgotinib (FIL) (Jyseleca ®) is an oral, ATP-competitive, rever- sible JAK1 inhibitor, developed by Galapagos NV and Gilead Science for treatment of inflammatory diseases, including RA [60]. The development of FIL is the result of a screening of BioFocus kinase-collection, tested in their capacity to inhibit the kinase domain of JAK1 in vitro through biochemical assays [61,62]. The molecules with an inhibitory capacity above 75% were then further evaluated through dose response tests measuring half maximal inhibitory concentration (IC50) values. FIL (GLPG0634) was then selected as selective inhibitor of the JAK1-dependent signaling [63]. In September 2020 FIL was then approved by EMA for the treatment of adult patients with moderate-to-severe active RA after csDMARDs failure with a dosage of 200 mg once daily. To date, FIL has not been approved by the FDA, whereas it has been licensed in Japan for RA patients with inadequate response to standard treatments [64,65].Clinical trials investigating the efficacy and safety of FIL in other autoimmune and inflammatory diseases are already ongoing, including inflammatory bowel disease and spondy- loarthritis, namely psoriatic arthritis, and ankylosing spondylitis.

3.1. Pharmacodynamics and pharmacokinetics

The inhibitory activity of FIL has been tested by different in vitro assays, showing that FIL preferentially inhibited JAK1/ JAK3-mediated IL-2, IL-4 and IL-15 signaling, JAK1/2-mediated IL-6 signaling and JAK1/TYK2-mediated type I interferon sig- naling. Moreover, in human whole blood assays, FIL (half maximal inhibitory concentration (IC50): 629 nM) showed a high selectivity on inhibition of JAK1-dependent IL-6-induced STAT1 phosphorylation, but almost no effect on JAK2-dependent GM-CSF induced STAT5 phosphorylation [63]. Moreover, filgotinib in vitro has been shown to dose- dependently inhibit Th1 and Th2 differentiation and to limit, albeit to a lower degree, the differentiation of Th17 cells [63]. Efficacy of FIL was also evaluated in in vivo mouse and rat models of collagen induced arthritis (CIA). Here, FIL demon- strated a comparable effect to etanercept in terms of clinical score, radiographic changes and histomorphological signs of inflammation. Pro-inflammatory cytokines including IL-6 were reduced in CIA animals treated with FIL in comparison to the disease-control group. Genes related to osteoclast activation were also reduced in FIL-treated mice [63]. Pharmacokinetic (PK) data have been collected from two early-phase trials performed in healthy volunteers with a dose up to 450 mg daily [66,67]. Consequently, a maximum daily dose of 200 mg has been chosen for the phase IIB trials with RA patients [62]. Oral FIL is absorbed rapidly, with the median peak plasma concentration (Cmax) of FIL reached 2–3 h post dose and that of its active metabolite reached 5 h post dose, with an apparent elimination half-life of 23 h. The stade–state plasma concentration levels of FIL and its active metabolite are reached after 2-3 days and 4 days, respectively, without further increase in the plasma or in the tissue of the metabolite [64].

In order to rule out possible genetic influence on drug metabolism, the PK and PD were assessed in a Japanese and a Caucasian cohort of healthy volunteers receiving FIL 200 mg daily or placebo for 10 days. The effect of FIL and its metabo- lite was comparable in the two ethnic groups [66].

Bodyweight, gender, age, mild renal impairment [creatinine clearance (CLCR) ≥60 mL/min), and mild or moderate hepatic impairment (Child-Pugh A or B) did not influence the PK of FIL. In contrast, moderate or severe impairment of the renal function (defined as CLCR 15 to <60 mL/min) leads to accu- mulation of FIL and its metabolites and requires a dose reduc- tion to 100 mg once daily. No data are available on patients with end-stage renal failure (CLCR <15 mL/min) or severe hepatic disease (Child-Pugh C) [64]. 3.2. Clinical Efficacy 3.2.1. Phase IIA Trials After pivotal studies on healthy volunteers, clinical proof of concept for the efficacy and safety of FIL have been investi- gated in double-blinded, placebo-controlled phase IIA trials (Table 1). The first one, published in 2012, included 36 MTX- IR RA patients, who received 100 mg twice daily, 200 mg once daily or placebo for 4 weeks, respectively. A clinical improve- ment was observed in ACR20 and Disease Activity Score based on 28 joints and C-reactive protein (DAS28-CRP) comparing both dosages of FIL to the placebo group [68]. In the second trial, 91 MTX-IR RA patients have been enrolled for a dose- ranging study, where FIL at 30 mg, 75 mg, 150 mg, or 300 mg daily was compared to placebo for 4 weeks. Improvements in ACR20 and DAS28-CRP were observed in the group of patients receiving a dosage between 70 and 300 mg daily. Both trials showed similar results in the groups of patients treated 200 mg or 300 mg daily [68]. 3.2.2. Phase IIB Trials Two-phase IIb trials started in 2013 (Table 1): DARWIN 1 (NCT01888874) evaluating FIL as add-on therapy and DARWIN 2 (NCT01894516) evaluating FIL as monotherapy, both in MTX-IR RA patients [69,70]. DARWIN 1 trial was a 24-week, randomized, double-blinded, placebo-controlled, phase IIb, dose-finding study, in which 594 MTX-IR patients have been enrolled receiving either different doses and regimens of FIL (50, 100, or 200 mg once or twice daily) or placebo up to week 24 [69]. At week 12, the ACR20 rates (primary endpoint) were significantly higher for 100 mg once daily (64%), 200 mg once daily (69%), and 100 mg twice daily (79%) versus placebo (44%; p = 0.0435, p = 0.0068, p < 0.0001, respectively). Significant dose-dependent improvements were achieved in terms of ACR50, ACR-N index of improvement, DAS28-CRP, Clinical Disease Activity Index (CDAI), and Simple Disease Activity Index (SDAI) in FIL treated groups versus placebo already at week 12 and maintained up to week 24. Moreover, DAS28-CRP improvement has been already reported after one week of treatment with 100 or 200 mg/d FIL. Patient-reported outcomes (PROs) with FIL, including mea- surements of physical function through Health Assessment Questionnaire-Disability Index (HAQ-DI), pain based on visual analog scale (VAS), fatigue with (FACIT-F) and Short Form-36 physical component summary (SF-36 PCS) also improved from week 2 up to week 12 [71]. Finally, comparable data were observed between once or twice daily regimens [69]. In the DARWIN 2 study, a 24-week, randomized, double- blinded, phase 2b, dose-finding trial the clinical efficacy of FIL was investigated as monotherapy [70]. Here, 283 MTX-IR patients were randomized to receive FIL 50, 100, or 200 mg monotherapy once daily versus placebo, after a washout per- iod from MTX of more than 4 weeks. The ACR20 response rate (primary endpoint) was achieved in all the actively treated groups in comparison to the placebo group at week 12. Specifically, at week 12 the primary endpoint was met by 67% of patients in the 50 mg group, 66% in the 100 mg, and 73% in the 200 mg (p < 0.001) versus 29% in the placebo group. Similar improvements were observed in all FIL groups when considering ACR-N, DAS28-CRP, SDAI, and European League Against Rheumatism (EULAR) good response at week 12. The group receiving 200 mg showed a greater mean change of disease activity from baseline (measured by both DAS28-CRP and SDAI) and greater remission rates. Moreover, the onset of clinical response in terms of ACR20 was observed at week 1 in the FIL 200 mg group and in terms of DAS28-CRP and CDAI in all dose groups, at week 2 in terms of ACR50 in the FIL 200 mg and at week 4 for ACR70 in the same dose group. These clinical responses were maintained or even improved up to week 24. Genovese and colleagues published the effect of FIL based on PROs in RA patients selected from DARWIN 1 and 2 studies [71]. At week 12, all PROs, except for the SF-36 mental component in the DARWIN 1 study, were significantly improved in patients treated with FIL compared with placebo, with a very early onset of clinical response since the first week of therapy. FIL reduced HAQ-DI by 0.58–0.84 points, pain by 24.2–37.9 mm, FACIT-F by 6.9–11.4, and patient global assessment (PtGA) by 25.2–35.6. These results were maintained up to week 24. Patients in the placebo cohort, reassigned to receive FIL 100 mg at week 12, experienced similar improvements in PROs between weeks 12 and 24 [71]. Patients who had completed 24 weeks of treatment in DARWIN 1 or DARWIN 2 were eligible to enter the ongoing, open-label,long-termextension study DARWIN 3 (NCT02065700). In particular, 739 out of 877 patients of DARWIN1/2 were enrolled in DARWIN 3 (497 from DARWIN 1, 242 from DARWIN 2). Here, all patients received FIL 200 mg/day (apart from 15 males who received a 100 mg/day dosage) [72]. Preliminary analysis of DARWIN 3 showed that clinical efficacy of FIL is maintained up to week 156, as assessed by ACR20, ACR50 and ACR70 response rates (87%/72%/46% in patients in combination therapy and 90%/63%/40% in patients continuing the monotherapy), and DAS28-CRP low disease activity and remission rates (69%/53% in patients receiving FIL and MTX and 65%/46% in patients receiving only FIL) [72]. 3.2.3. Phase III Trials The phase III development program of FIL in RA is called FINCH and includes four RCTs conducted in different disease subsets. FINCH 2 is the only phase III trial, whose data have been published so far. However, information from the FINCH 1, FINCH 3 and from the extension study FINCH 4 have been presented as abstracts during international meetings (Table 1). FINCH 1 (NCT02889796) is a 52-week randomized, double- blinded, placebo- and active-controlled study testing the clinical efficacy of FIL and evaluating the radiographic progression in MTX-IR RA patients [73]. In the study, 1755 patients were enrolled and randomized to receive either FIL 100 mg or 200 mg once daily (n = 480 and 475, respectively), adalimumab (ADA) 40 mg every 2 weeks (n = 325) or placebo (n = 475) in addition to csDMARD (MTX). After 24 weeks, patients in the placebo group were re-randomized to receive either FIL 100 mg or 200 mg once daily up to week 52. The ACR20 clinical response at the week 12 was defined as primary endpoint [73]. Specifically, 70% of the patients in the FIL 100 mg and 77% in the 200 mg groups reached the primary endpoint in comparison to the 50% in the placebo group (p ≤ 0.001). The ACR50 and ACR70 responses were also higher in the FIL 100 mg and 200 mg groups than in the placebo group (36% and 47% versus 20%; 19% and 26% versus 7%; nominal p ≤ 0.001). Accordingly, the proportions of patients in FIL 100 mg and 200 mg groups achieving DAS28-CRP low disease activity (score ≤3.2) at week 12 was higher than in the placebo group [39% and 50% versus 23% (p < 0.001)], as well as for the DAS28-CRP remission (score <2.6) [34% and 24% versus 9% (nominal p < 0.001)]. Moreover, this trial reported the non- inferiority of FIL 200 mg in comparison to ADA in terms of DAS28-CRP low disease activity at 12 weeks (50% vs. 43%; p ≤ 0.001 for noninferiority). The extension up to 52 weeks confirmed the efficacy of FIL in terms of ACR20, ACR50, ACR70 and DAS28-CRP low disease activity [74]. The FINCH 1 study addressed also the effect of FIL on the radiographic progression of active patients with RA. The data showed a significantly lower radiographic progression at week 24 (FIL 100 mg or 200 mg), assessed by the modified Total Sharp Score (mTSS) in compar- ison to the placebo recipients (mean change from baseline 0.17 and 0.13 vs. 0.37; p ≤ 0.001). Moreover, significant improvements in PROs (HAQ-DI, SF-36 PCS, and FACIT-Fatigue) were reported in the FIL groups in contrast to placebo recipients [73]. The FINCH 2 (NCT02873936) study investigated the clinical efficacy of FIL in patients with active RA, who had an inade- quate response or intolerance to ≥1 bDMARDs [75]. This 24- week randomized, double-blinded, placebo-controlled phase III trial included 448 patients, divided in different treatment groups: FIL 100 mg or 200 mg once daily (n = 153 and 147) or placebo (n = 148). All patients additionally received csDMARDs. The ACR20 response (primary endpoint) at week 12 was observed in a significant higher percentage of patients receiv- ing FIL 100 mg or 200 mg than in those receiving placebo: 58% and 66% versus 31% (p < 0.001). Similar data were observed for ACR50 (32% and 43% versus 15%; nominal p ≤ 0.001) and ACR70 (14% and 22% versus 7%; nominal p ≤ 0.05) suggesting the superiority of FIL in comparison to placebo. Accordingly, clinical response in terms of DAS28-CRP low disease activity (37% and 41% versus 16%; p < 0.001) and DAS28-CRP remission (26% and 22% versus 8.1%; nominal p < 0.001) at week 12 was higher in FIL 100 mg or 200 mg group compared with placebo. These clinical improvements were sustained through the fol- low-up period up to week 24 (nominal p ≤ 0.05). Moreover, HAQ-DI, SF36 physical component score and the fatigue (FACIT- Fatigue) assessment were significantly improved in FIL 200 and 100 mg versus placebo at both weeks 12 and 24. In particular, HAQ-DI showed a reduction of ≥0.22 points (66% and 67% vs. 44%; nominal p < 0.001) [75]. These data were further confirmed in a post hoc analysis of the PROs in both FINCH 1 and FINCH 2 studies. This analysis showed rapid and sustained improvements across multiple aspects of patients: Health-related quality of life (HRQoL) mea- sured by SF-36 and HAQ-DI when FIL was used in combination with csDMARD for treatment of MTX-IR or bDMARD-IR RA [76]. The FINCH 3 (NCT02886728) trial is a 52-week randomized, double-blinded, active-controlled phase III study investigating the efficacy and safety of FIL in MTX naïve patients with moderate-to-severe disease activity. The data reported so far suggested that FIL in combination with MTX significantly improved signs and symptoms of disease and HRQoL and slowed radiographic progression in active RA patients [77]. Of 1249 enrolled patients, 207 and 416 received FIL 100 mg or 200 mg once daily plus MTX, respectively, whereas 210 were treated with FIL 200 mg as monotherapy and the last 416 with MTX alone. The primary endpoint was the proportion of patients who achieved an ACR20 response at week 24. The frequency of patients reaching the primary endpoint was significantly higher in the group receiving combination ther- apy of FIL 100 mg or 200 mg plus MTX, than in the group receiving MTX as monotherapy: 80% and 81% versus 71% (p < 0.05). Conversely, there was no significant difference between FIL 200 mg monotherapy and MTX monotherapy in terms of ACR20 response rate (78% versus 71%). Interestingly, 24-week evaluation of patients receiving FIL (100 mg or 200 mg) plus MTX or FIL 200 mg in monotherapy in comparison to MTX monotherapy control group, showed better clinical response: ACR50 (57%, 62%, and 58% vs. 46%; p < 0.01) and ACR70 (40%, 44%, and 40% versus 26%; p ≤ 0.001), DAS28-CRP low disease activity (63%, 69%, and 60% versus 46%; p ≤ 0.001) and DAS28-CRP remission (43%, 54%, and 42% versus 29%; p ≤ 0.001). The onset of clinical response was rapid, already present after 2 weeks, and main- tained up to week 24. Patients receiving FIL 200 mg mono- therapy had less radiographic progression than patients receiving MTX (mean change from baseline in mTSS −0.04 versus 0.52; p < 0.01). However, there were no significant differences between patients receiving combination treat- ment of FIL (100 mg or 200 mg) plus MTX and those receiv- ing only MTX (mTSS 0.22 and 0.21 versus 0.52). Regarding the PROs assessment, the mean change from baseline in HAQ-DI at week 24 was significantly (p ≤ 0.01) higher in the groups receiving the combinatory treatment (FIL 100 mg or 200 mg plus MTX) as compared to MTX alone (−0.90 and −0.94 versus −0.79) [77]. The long-term extension up to 52 weeks of the study showed a sustained clinical efficacy of FIL, with higher ACR20 (75%, 73.4%, 74.8% versus 61.8%; p < 0.01), ACR50 (62.3%, 59.4%, 61.4% versus 48.3%; p < 0.01) and ACR70 (47.8%, 40.1%, 45.2% versus 29.8%; p < 0.05) in the FIL 100 mg and 200 mg combination therapy and FIL 200 mg monotherapy groups than in the MTX group. Accordingly, an increased proportion of patients in the FIL- treated groups achieved DAS28-CRP low disease activity and remission. Overall, a consistently higher numerical efficacy for FIL 200 vs 100 mg was described. Patients receiving combi- nation therapy or FIL 200 mg monotherapy had less radio- graphic progression than patients receiving MTX (0.27, 0.21 and 0.23 vs. 0.74; nominal p < 0.05) at week 52. The mean change from baseline in HAQ-DI scores was greater with FIL 200 mg combination therapy and FIL 200 mg monotherapy than with MTX (nominal p < 0.05), with no differences between the FIL 100 mg combination therapy and metho- trexate groups [78]. Recently, a post hoc analysis of all three FINCH studies was conducted to assess the impact of FIL on pain. FIL 200 and 100 mg provided rapid, clinically meaningful pain relief as assessed by VAS-pain in patients with active RA with ≥40% of patients in all studies having a ≥ 50% reduction in pain [79]. 3.2.4. Safety JAK-STAT signaling is involved in immune responses against pathogens as well as in cellular homeostasis, differentiation and proliferation [40]. These facts raised safety concerns on the routine use of this therapeutic approach. In particular, JAK3 deficiency results in immunodeficiency diseases, whereas JAK2 loss of function affects erythropoiesis as well as GM-CSF function. Consequently, JAK1 inhibitors could reduce the potential adverse events (AEs) due to a pan-JAK blockade. So far, the majority of safety data available on Jakinibs comes from RTCs and their extensions, except for tofacitinib, whose safety profile has been already reported in post-marketing real-life studies [80,81]. Overall, current data show that Jakinibs safety profile is comparable to bDMARDs and that, despite the theoretic difference between JAK specific inhibi- tion, safety data are overlapping between Jakinibs. Data obtained from studies on healthy volunteers showed a good safety profile of FIL [82]. Comparable results were reported in the phase IIA trials, where mild AEs and only one serious AE (SAE) were reported. Regarding lab parameters, there was no evidence of neutropenia; however, a limited decrease in absolute neutrophilic cell count was observed. No effect on lymphocytic cell count was described, whereas a dose-related improvement in hemoglobin value was observed. Finally, no significant change in the value of LDL or transaminases was observed [68]. A recent integrated safety analysis of FIL was conducted evaluating the three phase III trials (FINCH 1–3), the two phase II (DARWIN 1-2) and the two long-term extensions still ongoing (FINCH 4 and DARWIN 3) up to 5.5 years [83]. Here, exposure-adjusted incidence rates (EAIR) per 100 patient-years exposure (PYE) of treatment- emergent adverse events (TEAEs) have been reported. The percentage of patients receiving FIL 200 mg or 100 mg once daily was 96% after 12 weeks, 74.2% after 52 weeks and 8.7% after 156 weeks. Up to 12 weeks, there were no differences in terms of rates of TEAEs (EAIR 195.4, 176.3 versus 175.9), TEAEs Grade ≥3 (EAIR 12, 11.5 versus 10.6), SAEs (EAIR 10.9, 12.8 versus 8.9), and TEAEs leading to discontinuation (EAIR 8.7, 6.3 versus 8.8) or death (EAIR 0.6, 0.6 versus 0.6) between FIL 200 mg or 100 mg and placebo. In the long-term treated set, TEAEs were lower in the FIL 200 mg versus 100 mg (40.4 versus 64.2), whereas the EAIR were lower but comparable between doses for TEAEs Grade ≥3, SAEs, and TEAEs leading to discontinuation or death. In the placebo set, the rates of infections and serious infections was higher in the FIL-treated groups as compared to placebo (76.9, 67.3 versus 58 and 3.9, 3.3 versus 2.4, respectively). However, no opportunistic infec- tions, including tuberculosis, were observed across the groups. With respect to the HZ infections, there was no difference between FIL 200 mg or 100 mg and placebo (0.6, 1.1 versus 1.1). As for the long-term treatment exposure, EAIR of infec- tions were higher in FIL 100 mg versus 200 mg group (34.4 versus 24.8), as well as the rate of serious infections (3.1 versus 1.6). In contrast, HZ infections were more frequent in the FIL 200 mg group (1.8 versus 1.1). Major cardiovascular events (MACE) and venous thromboembolisms (VTE) were also eval- uated. In the placebo set, no cases of MACE were observed in FIL 200 mg, whereas the EAIR for FIL 100 mg and placebo were 1.7 and 1.1/100 PYE, respectively. No VTE were observed. In the treated set, MACE EAIR were comparable for FIL doses and VTE EAIR were 0.2 and 0.0/100 PYE for FIL 200 and 100 mg, respectively. The rate of malignancies including and excluding skin melanoma were not common and numerically comparable between the groups [83]. Another integrated safety analysis of the seven trials on FIL also reported comparisons of AEs rates between FIL, ADA, MTX and placebo groups. In this integrated analysis, TEAEs were comparable between FIL, placebo, MTX and ADA groups (ser- ious TEAEs 6.5, 7.7, 7.6, 7.9 and 9.3; TEAEs leading to death 0.4, 0.4, 0.3, 0 and 0.3). Moreover, there was no dose-dependent effect of FIL on serious TEAEs and TEAEs leading to death. In FIL 200 mg and 100 mg, ADA, MTX and placebo groups, the EAIRs/100 PYE for infections were 28.9, 39.0, 44.5, 44.1 and 52.7, and the serious infections were 1.7, 3.3, 3.4, 2.2 and 2.3, respectively. In particular, HZ virus infections were uncommon overall, but numerically slightly higher for FIL relative to pla- cebo, ADA, and similar to MTX (EAIRs 1.7 and 1.1 versus 1.0, 0.7 and 1.1). Opportunistic infections, including tuberculosis, were also low across the groups with comparable number between FIL and placebo. Moreover, in the FIL groups, a low rate of MACE as well as VTE, malignancies (other than non- melanoma skin cancer) and nonmelanoma skin cancer (≤1.1) were observed. Similar data were described across all treat- ment groups [84]. Common and uncommon adverse events as detailed by the EMA have been summarized in Table 2 (https://www.ema.europa.eu/en/documents/rmp-summary/jys eleca-epar-risk-management-plansummary_en.pdf). The safety profile of different Jakinibs, including FIL, has also been evaluated in vitro at clinically relevant doses to calculate the ability of these drugs to suppress erythroid progenitor cell expansion and maturation, NK cell proliferation and liver X receptor (LXR) agonist-induced cholesteryl ester transfer protein (CETP) expression. Indeed, Di Paolo et al. per- formed a dose-response in vitro assay comparing Jakinibs with different selectivity with regard to their respective inhibition (FIL, tofacitinib, baricitinib and upadacitinib). They showed that FIL (100 mg and 200 mg) resulted in lower calculated cellular inhibition than the other Jakinibs for NK cell prolifera- tion and CETP expression. However, no difference was observed on erythroid progenitor cell differentiation or maturation across the Jakinibs evaluated [85]. More recently, safety concern has been raised regarding high dosage of FIL and male fertility. Animal studies showed decreased fertility, impaired spermatogenesis, and histopatholo- gical effects on male reproductive organs. These effects were dose-dependent and not fully reversible (https://www.ema. europa.eu/en/documents/product-information/jyseleca-epar- product-information_en.pdf). However, the potential effect of FIL on sperm production and male fertility in humans is currently unknown. Currently, two different RCTs are addressing this topic: the Study to Evaluate the Testicular Safety of FIL in Adult Males with Moderately to Severely Active Inflammatory Bowel Disease (MANTA) and the Study to Evaluate the Effect of Filgotinib on Semen Parameters in Adult Males With Active Rheumatoid Arthritis, Psoriatic Arthritis, Ankylosing Spondylitis, or Non- radiographic Axial Spondyloarthritis (MANTA-RAy). The first is a placebo-controlled double-blind phase II trial with two arms (FIL 200 mg and placebo) with the aim to evaluate the testicular safety of FIL in adult males with moderately to severely active inflammatory bowel disease (https://clinicaltrials.gov/ct2/show/ NCT03201445). The second is a double-blind phase II study with two arms (FIL 200 mg versus placebo), an observation time up to 13 weeks and extension up to 156 weeks. The primary aim of this study is to evaluate the effect of FIL on semen parameters in adult males with active disease (RA, psoriatic arthritis, ankylosing spondylitis, or non-radiographic axial spondyloarthritis) (https:// clinicaltrials.gov/ct2/show/NCT03926195). Data from both stu- dies have not been published yet. 4. Conclusion FIL has been recently approved by EMA as new JAK1 selective inhibitor for the treatment of active RA patients after csDMARDs failure or intolerance. The selective capability of FIL to inhibit JAK1 has been tested with in vitro assays and the relatively longer elimination of its metabolite contributes to the overall PD of FIL. To date, FIL has been tested in phase II and phase III trials, with long-term extensions still ongoing. Based on these data, FIL showed a long-lasting efficacy compared to placebo and non- inferior data compared to ADA. RCTs reported also a good toler- ability and an overall acceptable safety profile of FIL. Moreover, clinical efficacy and safety of FIL are currently under investigation with RCTs for the treatment of spondylarthritis and inflammatory bowel diseases. Indeed, if the selective inhibition of JAK1 proved to be efficacious in RA patients, this approach needs still to be tested in the context of other rheumatic diseases with a different cytokine profile and different intracellular signaling pathways. 5. Expert opinion Filgotinib will be marketed for the treatment of RA as the fourth product in the JAKi class. The presence of such a large number of treatment options in an already busy mar- ket may make it unnecessary to add another arrow to the rheumatologist’s bow. However, some key aspects need to be considered. First of all, the strategy of managing RA by a treat- to-target approach has certainly improved the long-term out- come of the disease, but it has also led to the need for increasingly ambitious treatment targets. This process has been facilitated by the introduction of targeted drugs, which have enabled clinical response rates unimaginable before the era of biologics and the introduction of Jakinibs. Nevertheless, a significant proportion of patients still suffer from what is now called ‘difficult-to-treat RA,’ defined as a pattern of RA refractory to multiple treatments with alternative mechanisms of action [86]. This aspect justifies the need to further expand the range of choices available with a product such as FIL, which was developed in a program conducted on different subsets of the disease, including cases of multi-refractory RA. The second crucial aspect is related to the selectivity demon- strated in vitro by FIL against JAK1. It is a fact that the market penetration of the first two Jakinibs (tofacitinib and baricitinib) was very rapid and impactful, at least until some concerns about their safety profile partially dampened their growth curve. The limitations imposed by FDA on the use of bariciti- nib only at a dose of 2 mg and only in insufficient responders to anti-TNF, and by EMA on the use of tofacitinib only in subjects under 65 years of age, are certainly unclear aspects of this drug class. In this scenario, selective anti-JAK1 drugs represent, at least in theory, an opportunity to mitigate the occurrence of adverse events while maintaining a similar clin- ical efficacy. The data available so far on FIL seem to point in this direction. Even with the limitation of the lack of head-to- head studies comparing Jakinibs, indirect comparisons with other compounds in the same class seem to suggest substan- tially overlapping clinical efficacy in all RA subsets in which the drug has been tested. In addition, for some of the currently most controversial aspects of anti-JAK safety, such as HZ infections and especially venous thromboembolism, FIL seems to show a potentially more favorable profile compared to tofacitinib and baricitinib. However, the real weight of the possible FIL toxicity on male fertility remains to be clarified through the ad hoc clinical trials currently ongoing, and could certainly represent a limitation to the extensive use of the drug, even in a predominantly female disease as RA. Finally, it is important to stress that, beyond RCTs data, only post- marketing observational studies will really shed light on the actual safety profile of FIL, allowing a comparison with other Jakinibs when used in a real-life setting.