Efficacy and safety of filgotinib in the treatment of ulcerative colitis with a focus on rapid and sustained efficacy: a narrative review

1) Clinical Response and Remission Outcomes: Primary and Secondary Endpoints

Feagan et al. [5] was a pivotal study which reported on the preplanned primary and secondary endpoints of SELECTION. According to Feagan et al., FIL 200 mg significantly improved clinical remission rates (primary endpoint; defined in section above) at Week 10 versus placebo (biologic-naive patients: 26.1% vs. 15.3% [P=0.02]; biologic-experienced patients: 11.5% vs. 4.2% [P=0.01]), and at Week 58 versus placebo (37.2% vs. 11.2% [P<0.0001]). In the FIL 100 mg group, clinical remission was not significantly different compared with placebo at Week 10, but was significant by Week 58 (23.8% vs. 13.5%; P=0.04). The treatment effect of FIL 200 mg on clinical remission relative to placebo at Week 58 was consistent across prespecified subgroups (by biologic-naive/experienced, TNF antagonist failure, vedolizumab failure, TNF antagonist + vedolizumab failure statuses).

Pre-planned secondary endpoints, including MCS remission (total MCS score: ≤2, no single subscore of >1), endoscopic remission (Mayo endoscopic subscore: 0), and histologic remission (based on the Geboes score; no/mild increase in chronic inflammatory infiltrate in lamina propria, no neutrophils in the lamina propria/epithelium, and no erosion/ulceration/granulation tissue) were also met at Week 10 in biologic-naive patients treated with FIL 200 mg and Week 58 in FIL 200 mg-treated patients (i.e., significantly greater proportions achieving remission versus placebo), but these endpoints were not met in biologic-experienced patients at Week 10 and the FIL 100 mg group at Week 58 [5].

2) Rapid Symptom Relief and Achievement of pMCS Remission

Two post hoc analyses of the SELECTION trial reported on rapid symptom relief and achievement of partial MCS (pMCS) remission among FIL-treated patients [20,33]. In a post hoc analysis which assessed rapid symptom relief and pMCS remission (pMCS ≤ 2 and no rectal bleeding, stool frequency, or physician’s global assessment subscore >1) among FIL-treated patients with UC [20], FIL 200 mg significantly improved rectal bleeding and stool frequency within 7 days compared with placebo (P<0.05). Specifically, the treatment effect of FIL 200 mg versus placebo in achieving a combined improvement in rectal bleeding and stool frequency was observed by Day 9 in biologic-naive patients (18.8% vs. 9.5%; P=0.01) and Day 7 in biologic-experienced patients (10.7% vs. 4.2%; P=0.02). By Day 3, improvements in stool frequency alone were already observed in biologic-naive patients treated with FIL 200 mg versus placebo (29.0% vs. 19.0%; P=0.03).

By Week 2, greater proportions of FIL 200 mg-treated patients versus placebo achieved pMCS remission (biologic-naive: 15.1% vs. 8.0% [P=0.04]; biologic-experienced: 10.3% vs. 4.2% [P=0.03]), with similar treatment effects observed at Week 58 (P<0.0001) [20]. Day 7 rectal bleeding and stool frequency subscores were associated with MCS response at Weeks 10 and 58. Patients in pMCS remission at Weeks 10 and 58 also had greater improvements in IBD Questionnaire (IBDQ) scores than those not in pMCS remission, demonstrating that symptomatic relief were associated with both short-term and long-term improvements in health-related QoL (HRQoL).

As a supplement to the above study, Saruta et al. [33] assessed rapid symptom relief based on baseline disease severity (moderate: pMCS <7; severe: pMCS ≥7). The study found that Day 2 improvements in rectal bleeding and stool frequency with FIL 200 mg treatment were significantly higher in patients with baseline pMCS <7 versus pMCS ≥7 (biologic-naive: 8.4% vs. 1.1% [P=0.009]; biologic-experienced: 8.8% vs. 0.7% [P=0.004]). From Day 2–15, symptomatic remission rates increased regardless of prior exposure to biologics, and these rates generally continued to be significantly higher in those with baseline pMCS <7 versus pMCS ≥7. By Week 10, however, symptomatic remission rates with FIL 200 mg treatment were no longer significantly different between those with baseline pMCS <7 and pMCS ≥7. Hence, although symptomatic response to FIL 200 mg occurred more rapidly in patients with lower UC severity than those with higher severity, converging response rates over 10 weeks of treatment led to symptomatic remission regardless of baseline disease severity.

3) Durable and Long-term Efficacy

Three analyses reported on the durable and long-term efficacy of FIL (Feagan et al. [4-year interim data], Laharie et al., and Feagan et al. [5-year interim data]) [6,23,27].

When evaluating the durable action of FIL, an interim analysis of SELECTIONLTE [6] reported that among FIL-treated patients who achieved clinical remission at Week 10, 79.3% (FIL 200 mg) and 63.0% (FIL 100 mg) were in clinical remission or had an MCS response at Week 58; these rates were numerically greater than patients re-randomized at Week 11 to placebo (FIL 200 mg-placebo: 44.4%; FIL 100 mg-placebo: 41.7%).

When the long-term efficacy of FIL was evaluated in Feagan et al. [6], the proportions of FIL 200 mg-treated patients achieving pMCS (score: ≤ 1), IBDQ (score: ≥ 170), and fecal calprotectin (FCP; ≤250 μg/g) remission were maintained over time from long-term extension baseline to Week 144 among SELECTION Week 10 responders (80.0%, 86.4%, and 86.0%, respectively); among Week 10 non-responders, these proportions increased over time from long-term extension baseline to Week 192 (62.1%, 76.7%, and 59.3%, respectively). These responses were similarly maintained up to 5 years, where 85.0% and 84.8% of Week 10 responders achieved pMCS and IBDQ remission, respectively; among Week 10 non-responders, 66.7% and 77.8% achieved pMCS and IBDQ remission, respectively [23]. Generally, differences between patients treated with FIL 200 mg and 100 mg were minimal, and the FIL induction dose did not influence efficacy outcomes among FIL 200 mg-treated patients in SELECTIONLTE [6]. Even among initial non-responders, gradual improvements and maintenance of efficacy measures were observed upon continued FIL treatment in SELECTIONLTE. Furthermore, improvements up to 4 years with FIL treatment in both Week 10 responders and non-responders were accompanied by low corticosteroid use, and high proportions of patients in corticosteroid-free pMCS remission [6].

Laharie et al. [27], a post hoc analysis of SELECTION assessing the prolonged benefit (i.e., sustained response) of FIL, demonstrated that continuous FIL 200 mg treatment significantly reduced the risk of PSDW compared with patients who were re-randomized from FIL 200 mg to placebo (FIL 200 mg-placebo: hazard ratio [HR], 0.26; P<0.0001). Among patients who were in clinical remission or had an MCS response at Week 10, 66.8% treated with continuous FIL 200 mg had a prolonged benefit at Week 58, compared with 32.7% in the FIL 200 mg-placebo arm. Additionally, in the continuous FIL 200 mg arm, 82.0% of patients with prolonged benefit at Week 58 had IBDQ remission at Week 58, demonstrating that prolonged benefit was associated with improvements in HRQoL.

4) Predictive Markers of Efficacy

Seven analyses included in this review identified patient factors associated with greater efficacy responses to FIL treatment or beneficial long-term response trajectories [21,22,29-31,34,36].

In a pre-planned exploratory analysis which identified circulating biomarkers of UC [30], FIL 100 mg and 200 mg significantly lowered levels of almost all circulating biomarkers associated with UC pathology at Weeks 4 and 10 versus placebo, especially at the 200 mg dose. Notably, there was a consistent decrease in almost all systemic inflammatory biomarkers, neutrophil activation biomarkers, T helper (Th)17- and Th2-related factors, and immunoglobulin M levels, with similar effects observed in biologic-naive and biologic-experienced patients. Therefore, changes in certain biomarker levels at Week 4 may be early mediators of FIL treatment effects at Week 10.

Across 6 post hoc analyses [21,22,29,31,34,36], factors such as biologic-naive status, smoking, female sex (vs. male), recent UC diagnosis (<1 year), FIL 200 mg treatment (vs.100 mg), and Week 10 assessments including stool frequency improvements (subscore: ≤1), FCP concentration (<250 μg/g), low endosopic inflammatory activity (subscore: ≤ 1), Geboes histologic remission, and HRQoL improvements (IBDQ total score: >170) were identified as predictive markers of sustained clinical remission to approximately 1 year. Aside from biologic-naive status, female sex, and lower baseline C-reactive protein (CRP) levels, Feagan et al. [22], also found that rapid symptom relief (i.e., improved rectal bleeding and stool frequency) on Day 7 predicted MCS response/remission at Week 10.

In addition to analyzing predictive factors, Schreiber et al. [34] identified distinct patient subgroups of response to FIL based on pMCS trajectories over time. In this study, beneficial trajectory groups generally had higher proportions of patients who were recently diagnosed (<1 year; 4%–9% vs. 4%–5%), receiving FIL 200 mg (43%–65% vs. 36%–46%), and biologic-naive (54%–70% vs. 35%–58%) compared with relapsing trajectory groups. Furthermore, 55.4% of patients had sustained beneficial trajectories, with ≥43% having low baseline endoscopic subscores (subscore: 2) and ≥61% having strong Week 10 FCP responses (>50% decrease in FCP from baseline). Sustained beneficial trajectory groups also had a higher probability of achieving comprehensive disease control (CDC) at Week 58 than other groups (31%–32% vs. 0%–7%).

When the efficacy of FIL based on prior biologic failure was analyzed post hoc, Dotan et al. [21] showed that FIL 200 mg induced and maintained benefits relative to placebo in terms of achieving Week 10 clinical remission (biologic-naive, OR: 1.98; had prior biologic failure, OR: 3.91) and reducing the risk of PSDW at Week 58 (biologic-naive, HR: 0.22; had prior biologic failure, HR: 0.22), with the estimated therapeutic benefit being higher in biologic-naive patients than patients with prior biologic failure. Furthermore, among patients with prior biologic failure, the likelihood of PSDW at Week 58 generally increased with increasing numbers of prior biologics.

5) Health-Related Quality of Life

Schreiber et al. [35] reported on pre-planned exploratory analyses of SELECTION data, evaluating the effect of FIL on generic HRQoL (measured by 36-Item Short-Form Survey [SF-36] and EuroQol 5-dimension [EQ-5D] scores) and disease-specific HRQoL (measured by IBDQ score). At Weeks 10 and 58, patients who received FIL 200 mg had greater least-squares mean increases from induction and maintenance baseline, respectively, in SF-36, EQ-5D, and IBDQ scores than those who had received placebo (P<0.05); generally, these treatment differences were numerically larger among biologic-experienced than biologic-naive patients. Overall, treatment with FIL 200 mg improved generic and disease-specific HRQoL in the short term (10 weeks) among both biologic-naive and biologic-experienced patients, with improvements maintained during long-term treatment (58 weeks).

6) CDC: A Multi-Component Endpoint

Two studies assessed a multi-component endpoint, CDC, defined as achievement of pMCS remission, endoscopic improvement (subscore: ≤1), inflammatory biomarker remission (FCP <150 μg/g), and IBDQ remission (score: ≥ 170) assessed at individual patient level [6,35].

In a post hoc SELECTION analysis [35], a greater proportion of patients treated with FIL 200 mg achieved CDC than placebo at Week 10 (10.9% vs. 2.9%; P<0.001) and Week 58 (22.1% vs. 7.1%; P=0.002), and this trend was more pronounced in biologic-naive than biologic-experienced patients. CDC was also associated with clinically important improvements in HRQoL (e.g., SF-36, EQ-5D scores) and histological remission over both periods (all P<0.05). This suggests that CDC achievement may be associated with HRQoL restoration and controlled disease at the cellular level.

Similarly, according to an interim analysis of SELECTIONLTE [6], achieving CDC with FIL 200 mg at Week 58 of SELECTION was associated with improved efficacy, improved HRQoL, and reduced corticosteroid use during 3 years of follow-up, as shown by the significantly higher proportions of Week 58 CDC achievers than non-achievers who reached pMCS remission (pMCS ≤1), IBDQ remission (IBDQ score: ≥170), and corticosteroid-free pMCS remission (pMCS ≤ 1; not taking corticosteroids on the day of study visit) up to long-term extension Week 96 (all P<0.05). Generally, higher proportions of CDC achievers at Week 58 reached remission outcomes in SELECTIONLTE than those achieving individual components of CDC, suggesting that this multi-component endpoint may provide more rigorous assessment of treatment efficacy compared with individual endpoints.

7) Corticosteroid-Sparing Effects

Three studies included in this review investigated the corticosteroid-sparing effects of FIL, based on data from the SELECTION/SELECTIONLTE trials [5,26,28].

Briefly, in the pivotal SELECTION study [5], a significantly greater proportion of patients treated with FIL 200 mg versus placebo had 6-month corticosteroid-free remission at Week 58 (27.2% vs. 6.4%; P=0.01).

A post hoc analysis of SELECTION assessing the corticosteroid-sparing effects of FIL 200 mg at Week 58 by Loftus et al. [28] also found that among patients receiving corticosteroids at maintenance baseline, 30.4%, 29.3%, 27.2%, and 21.7% had been in corticosteroid-free remission for ≥1, ≥3, ≥6, or ≥8 months, respectively, versus 6.4% receiving placebo across all thresholds (P<0.05).

In addition, 6-month corticosteroid-free remission based on baseline characteristics were assessed by Kobayashi et al. [26], which observed that a lower baseline MCS endoscopic subscore (2 vs. 3) was associated with increased odds of corticosteroid-free remission in patients treated with FIL 200 mg and 100 mg, while female sex, smoking, and biologic-naive status was associated with increased odds of corticosteroid-free remission in those treated with FIL 200 mg versus 100 mg. These results indicated that although steroid tapering can be achieved in patients with UC receiving FIL 200 mg independently of baseline characteristics (e.g., clinical activity, illness duration), the likelihood of achieving corticosteroid-free remission may be higher among patients who are biologic-naive, smokers, had low endoscopic inflammatory burden, and females.

Overall, based on the above findings, FIL 200 mg demonstrated corticosteroid-sparing effects and maintained corticosteroid-free clinical remission in patients with UC.

8) RWE on Effectiveness

Although randomized clinical trials represent the optimal approach for evaluating efficacy and safety, RWE studies serve as a valuable complement to the SELECTION trials, by providing insights into FIL’s effect among broader, more diverse populations. Moreover, RWE studies can capture data on treatment outcomes of emerging importance (e.g., fecal urgency) that were not evaluated in the SELECTION trials, along with key treatment parameters such as drug persistence. The ongoing GALOCEAN study, for example, assesses both urgency and persistence [39]. Here, we report the results from several RWE studies [18,19,24,25,32,38].

Gros et al. [24] was the first cohort study to report data on real-world outcomes of FIL in patients with UC (n=91), which comprised a substantial proportion (67%) of biologic/small molecule-naive patients. Based on this retrospective analysis conducted in Scotland, pMCS (<2), biochemical (CRP ≤5 mg/L), and fecal biomarker (FCP <250 μg/g) remission outcomes were achieved in over 70% of patients at Week 12, and these improvements were maintained through Week 24. At Week 12, there were significant reductions in median pMCS, CRP levels, and FCP levels (all P<0.0001), whereas no significant changes were observed in total cholesterol and alanine transaminase levels. An exploratory analysis was performed to identify factors associated with clinical remission at Week 12, but no factors were identified. At the end of median follow-up of 42 weeks, 82.4% of patients remained on FIL and median time to FIL discontinuation was 13 weeks. The primary reason for drug discontinuation was primary non-response (9.9%), followed by secondary loss of response (2.2%). It was found that decreased albumin levels (<36 g/L) at baseline were associated with worse drug persistence (P=0.011). Consistent with analyses of SELECTION data, these real-world findings demonstrated that FIL effectively induced and maintained remission in patients with UC, with favorable outcomes persisting up to 6 months.

Similarly, Nogami et al. [18] was the first multicenter cohort study to assess outcomes in FIL-treated patients with UC (n=98). This retrospective observational analysis conducted in Japan demonstrated Week 8 clinical response and remission rates of 55.1% and 46.9%, respectively, with similar rates being observed regardless of previous exposure to tofacitinib (tofacitinib-naive: 54.9% and 48.8%; tofacitinib-exposed: 56.3% and 37.5%). Over a median follow-up of 191 days, 52.0% of patients treated with FIL discontinued treatment. Plachta-Danielzik et al. [32] reported the results from a prospective, multicenter observational FilgoColitis study, which investigated the clinical effectiveness of newly introduced FIL therapy in patients with active UC. At the 6-month interim analysis (n=164), treatment persistence was 87.2%, and clinical response and remission rates were 59.2% and 41.7%, respectively. Notably, no significant difference in clinical remission rates was observed between biologic-naive (41.7%) and biologic-experienced (41.4%) patients. For those who responded by Week 10, the effectiveness was even more pronounced at 6 months, with response and remission rates of 85.0% and 58.8% respectively. Additionally, patients experienced significant improvements in HRQoL, as evidenced by an increase in mean Short IBDQ (sIBDQ) scores, from 42.0 to 56.0 (P<0.001) [32].

Further real-world insights emerge from the interim analysis of the prospective, observational, multicenter GALOCEAN study [25], which investigated FIL’s effectiveness with and without concomitant UC therapies (CT) in patients with UC. At the time of analysis, 223 patients had available baseline data, of which 62.3% and 37.2% completed 10 and 24 weeks of treatment, respectively; the remainder of the patients may have discontinued before these timepoints or may not have reached these timepoints at data cutoff. At Week 24, MCS remission was achieved by 30% of patients (FIL-CT: 29%; FIL+CT: 31%), while pMCS remission rates were 47% across both groups. Minimal clinically important differences (MCIDs) in sIBDQ scores were achieved by 66% of the overall cohort (FIL-CT: 75%; FIL+CT: 58%), and 43% (FIL-CT: 50%; FIL+CT: 36%) for the Urgency Numerical Rating Scale (NRS) score. This analysis underscores FIL’s overall effectiveness in a real-world cohort over 24 weeks, demonstrating comparable benefits whether used as monotherapy or with concomitant treatments [25].

A post hoc analysis of the early real-world outcomes from the same GALOCEAN study specifically examined FIL’s effectiveness in patients stratified by baseline disease activity (moderately vs. severely active UC). At the time of analysis, 219 patients had available baseline data, of which 62.6% and 37.4% completed 10 and 24 weeks of treatment, respectively; the remainder of the patients may have discontinued before these timepoints or may not have reached these timepoints at data cutoff. In this interim analysis, MCS remission rates at Week 24 were 33% in patients with moderately active UC compared to 25% in those with severely active UC. Similarly, pMCS remission rates were 56% and 43% for moderately and severely active patients, respectively. Clinically meaningful improvements in patient-reported outcomes were also observed, with MCID in the sIBDQ score achieved by 75% of moderately active UC patients and 68% of severely active UC patients, and MCID in the Urgency NRS score achieved by 35% and 42%, respectively. These findings indicate that FIL generally leads to improved UC symptoms and reduced disease activity over 24 weeks in real-world settings, with numerically greater proportions of patients with moderately active UC achieving remission and clinically meaningful sIBDQ score improvements [38].

Akiyama et al. [19], a real-world multicenter retrospective study, identified FIL-treated patients with UC (n=238), across 12 institutions in Japan. In this study, clinical remission rates based on per-protocol analysis at 10, 26, and 58 weeks were 47%, 55.8%, and 64.6%, respectively. At a median follow-up of 28 weeks, the rates of clinical remission, clinical response, corticosteroid-free remission, and endoscopic improvement were 39.9%, 54.7%, and 36.5%, and 43.5%, respectively. Notably, these rates were comparable between biologic/JAK inhibitor-naive and -experienced patients. At this median follow-up, the discontinuation rate of FIL was 39%. The majority of patients discontinued due to lack of efficacy (83%), followed by intolerance (14%). Most discontinuations occurred within 10 weeks of induction. The discontinuation rate was significantly higher among biologic/JAK inhibitor-experienced patients compared to biologic/JAK inhibitor-naive patients (47% vs. 30%; P=0.011). Multivariate analysis from this study showed clinical remission with FIL at the most recent follow-up was inversely related to baseline platelet count and number of prior biologics/JAK inhibitors (odds ratio [OR]=0.953; P=0.0102 and OR=0.768; P=0.0433, respectively). Furthermore, baseline platelet count and partial Mayo score were significantly associated with the risk of FIL discontinuation (OR=1.045; P=0.0178 and OR=1.196; P=0.0255, respectively), suggesting that patients with higher disease activity are more likely to discontinue FIL. Meanwhile, discontinuation was inversely related to concomitant use of immunomodulators (OR=0.485; P=0.0499). These findings suggest that platelet count could serve as a unique biomarker for both predicting efficacy and monitoring disease activity in FIL-treated patients, although further investigations are warranted to clarify the molecular links between platelet counts and JAK1 signaling inhibition in UC [19].

Future RWE studies involving longer follow-up durations and identifying predictive factors of effectiveness could help strengthen the evidence supporting FIL treatment in UC therapy.

9) Overall Efficacy Findings

In summary, the publications included in this review demonstrated the rapid and durable efficacy of FIL in treating UC, in terms of achieving clinical remission, improving HRQoL, reducing biomarker levels, symptomatic relief, reducing corticosteroid dependency, and multidimensional disease control. The SELECTION trial and its post hoc analyses demonstrated significant efficacy improvements with FIL as early as 7 days (e.g., improved rectal bleeding and stool frequency) [5,20,33]. Long-term data from the SELECTIONLTE extension studies (4-year and 5-year interim analyses), along with a post hoc analysis of SELECTION assessing sustained response, showed responses extending up to approximately 5 years (e.g., clinical remission, MCS response, improvements in IBDQ), even in patients who did not initially respond after 10 weeks of treatment [6,23,27]. RWE consistently supports FIL’s effectiveness in improving UC symptoms, reducing disease activity, and improving HRQoL. The overall effectiveness of FIL in real-world cohorts appears comparable whether used as monotherapy or with concomitant treatments, and effectiveness is similar across various patient populations, including those with prior biologic experience [18,19,24,25,32,38].

Achieving these efficacy outcomes is vital for enhancing patients’ QoL and alleviating disease burden. Specifically, rapid symptom relief leads to swift improvements in discomfort and pain, thereby mitigating the emotional and physical stress associated with undesirable symptoms [35]. Such rapid benefits can enhance treatment adherence, increase patient satisfaction, and improve long-term health outcomes. Additionally, durable efficacy outcomes in clinical remission and HRQoL improvements facilitate sustained symptom relief and long-term disease management, minimizing the need for frequent therapy adjustments. Prolonged periods of UC remission also decrease patients’ anxiety concerning disease flares, and reduce reliance on additional medications (e.g., corticosteroids) which may have harmful side effects [20,40]. This stability conferred by durable efficacy outcomes enables patients to maintain their daily activities and productivity, while lowering the long-term health risks associated with untreated or poorly managed UC.

When considering efficacy outcomes by patient subgroups, these findings suggest the potential of FIL as a robust long-term treatment option in patients with UC, including those with prior biologic experience. However, those with lower (vs. higher) disease severity may experience more rapid symptomatic remission, and biologic-naive patients (vs. patients with prior biologic experience/failure) and those who use corticosteroids sparingly (vs. high corticosteroid use) may experience greater sustained clinical improvements on FIL treatment [20,33].

Furthermore, these findings support the utility of FIL in reducing corticosteroid dependency, which is crucial for patient safety and long-term disease management. The results also indicated a potential dose-response relationship, with the 200 mg dose generally eliciting greater efficacy responses than the 100 mg dose, and FIL 100 mg demonstrating greater efficacy than placebo in terms of clinical remission only in the maintenance study [5,26,28]. It is important to note that for moderate to severe UC, Japan’s Pharmaceuticals and Medical Devices Agency only indicates FIL 100 mg for maintenance therapy or for patients with severe or moderate kidney impairment. For all other instances where existing, conventional treatments are insufficient in effectiveness, FIL 200 mg is indicated [11]. Conversely, FIL 200 mg is indicated for both induction and maintenance therapy in the EU, UK, Singapore, and South Korea. FIL 100 mg is only indicated for maintenance therapy in elderly patients (i.e., ≥ 65 years of age) or adults at higher risk of venous thromboembolic event (VTE), major adverse cardiovascular events (MACE), and malignancy, or patients with severe or moderate kidney impairment [7-10].

As evidenced from various post hoc analyses, it is also important to consider factors including baseline assessment of serum biomarkers, clinical symptoms, patient demographics, and rigorous multi-component endpoints for optimizing UC management, as these predictive markers may facilitate the development of personalized and holistic management strategies in treating UC. Lastly, these findings support the possibility of using biomarkers and symptom monitoring as proxies for invasive mucosal assessments (e.g., endoscopy) at early timepoints of UC management [21,22,29-31,34,36].

1) Safety Outcomes in SELECTION and SELECTIONLTE

Table 4 summarizes the safety outcomes of 3 papers reporting on SELECTION and/or SELECTIONLTE data, at Week 58 and up to Week 202 [5,6,37]. Across the 3 safety analyses that report on overlapping patient groups enrolled in SELECTION and SELECTIONLTE, there were low/moderate occurrences of serious treatment-emergent AEs (TEAEs; 0.0%–18.1%) and TEAEs leading to study discontinuation (2.0%–25.6%), and very low occurrences of death (0%–0.8%) among FIL-treated patients, with comparable rates observed in the placebo group (Table 4) [5,6,37]. Regarding AEs of interests, infections were slightly higher in patients treated with FIL 200 mg (25.3%–52.7%) versus 100 mg (25.7%–33.8%), while serious infections (0%–5.8%) and herpes zoster infections remained low (0%–3.6%) across all treatment arms and studies [5,6,37]. Similar rates of AE occurrences were observed in FIL-treated patients compared with the placebo group [5,37].

Table 4.

Summary of Safety Outcomes Reported for SELECTION and/or SELECTIONLTE

Feagan et al. [5] was the first publication to assess safety outcomes at induction (Week 10) and maintenance (Week 58) of SELECTION. Across the induction and maintenance studies, the incidences of serious AEs, AEs of interest (including infections), and treatment discontinuation due to AEs were similar between FIL and placebo groups; FIL was generally well tolerated at both 100 mg and 200 mg, and most AEs were mild or moderate in severity. Similarly, occurrences of serious infections and herpes zoster infections (n=6; none were serious/complicated or resulted in study discontinuation) were low in all treatment groups. Frequently observed AEs included nasopharyngitis, headache, worsening of UC, and arthralgia. One VTE (pulmonary embolism) was reported among FIL 200 mg-treated patients at induction, and 2 VTEs occurred among placebo-treated patients at maintenance. Malignancies and non-melanoma skin cancers each occurred in 3 FIL-treated patients. Two deaths were reported in the FIL 200 mg group during the maintenance study; neither were related to treatment.

Feagan et al. [6], reported the long-term safety outcomes (up to 3.9 years) of FIL in SELECTIONLTE as a primary endpoint. In this study, FIL’s long-term safety profile was generally consistent with the results of SELECTION trial and the previous data cutoff for SELECTIONLTE [5,37]. Occurrence of AEs was similar between patients treated with FIL 200 mg and 100 mg. Generally, the occurrence of TEAEs were low, with FIL 200 mg and 100 mg groups reporting overall TEAE incidences of 111.5 exposure-adjusted incidence rate per 100 censored patient-years of exposure (EAIR/100 cPYE) and 103.4 EAIR/100 cPYE, respectively. Serious TEAEs and TEAEs of grade 3 or higher were also uncommon (<10.0 EAIR/100 cPYE). Six deaths occurred during SELECTIONLTE in the FIL 200 mg group, deemed unrelated to the study drug, and mostly related to coronavirus disease 2019 (COVID-19) infections and/or related complications. The most frequent AEs were worsening of UC (FIL 200 mg: 16.5%; FIL 100 mg: 23.1%), COVID-19 infection (FIL 200 mg: 4.4%; FIL 100 mg: 1.3%), and anemia (FIL 200 mg: 3.8%; FIL 100 mg: 1.3%). Herpes zoster infections were rare (n=34; FIL 200 mg: 1.5 EAIR/100 cPYE; FIL 100 mg: 1.0 EAIR/100 cPYE) and mostly mild (all grade 2 or lower except for one grade 3 case), but incidence rates were somewhat higher in the Asian population compared with other ethnicities. There were also no increased risks of opportunistic infections, malignancies, MACE, or VTE associated with continuous FIL treatment.

In an integrated interim analysis of SELECTION and SELECTIONLTE representing a total of 3,326.2 patient-years of exposure (PYE) [37], the safety of FIL was compared in different patient subpopulations (e.g., by age, previous biologic exposure). Consistent with Feagan et al. [5], FIL was well tolerated with no unexpected safety signals in this integrated interim analysis; FIL also had an acceptable safety profile pertaining to infection risk, regardless of previous biologic exposure or age. Incidences of serious infection, thromboembolic events, and MACE were consistently low across treatment groups, with the majority of patients with MACE having cardiovascular risk factors. The incidence (measured by EAIR/cPYE) of herpes zoster infections was numerically higher for FIL 200 mg (1.44) and FIL 100 mg (0.69) than for placebo (0.26), and incidence of all infections was numerically higher in biologic-experienced (FIL 200 mg: 45.3; placebo: 72.9) than biologic-naive patients (FIL 200 mg: 28.9; placebo: 25.3). In line with previous reports of age as a risk factor for AEs such as malignancy, incidences of AEs including infections (e.g., herpes zoster), MACE, and malignancies were higher in patients aged ≥ 65 years than those < 65 years. Four deaths occurred, including 3 cardiovascular deaths, none of which were considered related to FIL.

An integrated post hoc analysis, pooling safety data from all phase 2 and 3 trials of FIL 100 mg and 200 mg in both rheumatoid arthritis and UC, aimed to evaluate the long-term safety of FIL. Particularly, the risks for MACE, VTE, and malignancies were assessed. Specifically for UC, the analysis included data from the SELECTION and SELECTIONLTE trials, representing over 2,800 PYE [41].

Among individuals with UC, MACE incidence was slightly higher in older (≥65 years) versus younger (<65 years) patients (EAIR: 1.7/100 PYE vs. 0.2/100 PYE), indicating an age-related increase in risk. No cases of VTE were reported in UC patients aged ≥65 years. Across both MACE and VTE, the EAIRs in patients with UC were generally similar between FIL 200 mg and FIL 100 mg, suggesting no consistent dose-related effect for these events. Malignancy rates generally followed a similar age-related trend, with a slightly higher incidence among older patients receiving FIL 200 mg compared to younger patients (EAIR: 2.3 PYE vs. 0.4/100 PYE, respectively) [41].

Overall, the safety data of FIL are consistent with previous overall safety analyses, demonstrating overall low rates of AEs of interest (AEI). Numerically higher rates of AEI occurred in patients aged ≥ 65 year, and there was no consistent FIL dose effect on AEI, suggesting that the occurrence of these AEI is predominantly influenced by the patients’ advancing age and underlying inflammatory disease, rather than a direct drug effect, thereby supporting dose adjustments for older patients [41].

Beyond overall incidence rates, the integrated post hoc analysis also explored baseline characteristics associated with the occurrence of specific AEI in patients with UC. While no formal statistical analysis was performed, certain patterns of association were observed irrespective of FIL dose. For MACE events in UC, there was a higher proportion of male patients, patients aged ≥65 years, and those who had a history of hypertension or dyslipidemia. A higher mean body mass index (BMI) was also noted in UC patients who experienced MACE compared to those who did not. UC patients who had malignancy events (excluding non-melanoma skin cancer [NMSC]), were more likely to be aged ≥ 65 years or have a history of hypertension, irrespective of FIL dose. Additionally, patients with UC were more likely to have a higher mean FCP level or a history of diabetes mellitus, though the interpretation of these associations is limited by the small number of events. Similarly, UC patients who developed NMSC were more likely to be aged ≥ 65 years or have a history of hypertension or dyslipidemia at baseline. These exploratory findings collectively suggest that older age and pre-existing cardiovascular risk factors (e.g., hypertension, dyslipidemia, higher BMI), along with indicators of higher disease activity or comorbidities, may be associated with an increased likelihood of MACE and malignancies in UC patients treated with FIL, irrespective of the specific FIL dose [41].

2) RWE on Safety

To complement the results of clinical trials, a RWE study of patients with UC in Scotland reported a 16.5% occurrence of AEs during the overall study period (July 2022–May 2023) [24]. These included mild AEs (5.5%; e.g., headache, fatigue, mild nausea, joint pain) and moderate AEs leading to temporary discontinuation of FIL (8.8%; mostly respiratory tract infections). Serious AEs leading to definitive drug discontinuation or death occurred in 2 patients.

Similarly, in Nogami et al. [18], the occurrence of AEs was reported in 24.5% of FIL-treated patients with UC in Japan during the overall study period (March 2022–February 2024). Serious AEs occurred in only 1 patient and resolved after drug discontinuation, while AEs leading to drug discontinuation occurred in 6.1% of patients. The occurrence of infections was 11.2%, with 3.1% developing herpes zoster infection. None of the patients experienced MACE, thrombosis, or malignancies.

3) Overall Safety Findings

FIL 100 mg and 200 mg were generally well tolerated with no unexpected safety signals in SELECTION and SELECTIONLTE, and had an acceptable short-term (Week 10) and long-term (up to 4 years) benefit-risk profile; a similar safety profile of FIL was observed in real-world studies. Incidence of AEs were comparable between FIL-treated patients and the placebo group, as well as between patients treated with FIL 100 mg and 200 mg. Across all safety analyses, the occurrences of serious AEs, serious infections, and herpes zoster infections were low. Additionally, the occurrence of infections was numerically higher in biologic-experienced (versus biologic-naive) patients and older (versus younger) patients. MACE and malignancy rates, although low overall, increased with advancing age in UC patients. Exploratory analysis of baseline characteristics further indicated that older age and pre-existing comorbidities (including hypertension and dyslipidemia) were associated with a higher likelihood of these AEs, irrespective of FIL dose [41]. Nevertheless, FIL was shown to have an acceptable safety profile regardless of the FIL dose received, previous biologic exposure, age, or corticosteroid use.