Ulcerative colitis disease severity affects the speed of symptom relief under filgotinib treatment: a post hoc analysis of the phase 2b/3 SELECTION study
Article information
Abstract
Background/Aims
Filgotinib is an oral, once-daily, Janus kinase 1 preferential inhibitor approved for the treatment of ulcerative colitis (UC). This study aimed to assess symptomatic response with filgotinib 200 mg (FIL200) according to disease severity using baseline partial Mayo Clinic Score (pMCS).
Methods
In the phase 2b/3 SELECTION study (NCT02914522), adults with moderate-to-severe UC were randomized to receive FIL200, filgotinib 100 mg, or placebo for 11 weeks in induction studies A (biologic-naive) and B (biologic-experienced). In this post hoc analysis, symptomatic remission (Mayo rectal bleeding subscore of 0 and stool frequency subscore ≤ 1) rates were assessed daily from baseline to day 15 and fortnightly from week 2 to week 10 by baseline pMCS (pMCS ≥ 7, pMCS < 7) in patients who received induction FIL200.
Results
Of those who received FIL200 in induction studies A and B, 90 and 148 patients had a pMCS ≥ 7, and 155 and 114 had a pMCS < 7, respectively. Symptomatic remission rates were generally significantly higher in the pMCS < 7 than ≥ 7 group from day 2–15 (day 2: 8.4% vs. 1.1%, P=0.009 [induction study A]; 8.8% vs. 0.7%, P=0.004 [induction study B]). However, by week 10, there was no longer a significant difference in the rates between the pMCS ≥ 7 and < 7 groups (43.3% vs. 54.8%, P=0.124 [induction study A]; 26.4% vs. 39.5%, P=0.099 [induction study B]).
Conclusions
Symptomatic response to FIL200 occurred more rapidly in the less severe disease groups than in the more severe disease groups; however, regardless of disease severity, both groups benefited from continued FIL200 treatment.
INTRODUCTION
Ulcerative colitis (UC) is a debilitating, chronic, relapsing disease, characterized by mucosal inflammation of the colon [1]. Symptoms include diarrhea, rectal bleeding, and frequent bowel movements, which can have a substantial negative impact on a patient’s quality of life [2,3].
When considering UC treatments, patients highly value rapid onset of action, as demonstrated by patient survey data and real-world evidence [4,5]. Treatment options that are available for patients with moderately to severely active UC for whom conventional therapies have failed include biologics targeting tumor necrosis factor (TNF); interleukin-12/23, interleukin-23, and α4β7-integrin antagonists; and small molecules such as Janus kinase (JAK) inhibitors and sphingosine-1 phosphate receptor modulating therapies [6].
By blocking the activity of one or more JAK enzymes, JAK inhibitors interfere with the JAK-signal transducer and activators of transcription signaling pathway to reduce cytokine signaling and to reduce the inflammatory response in UC [7]. JAK inhibitors are orally administered and non immunogenic, and have been shown to have a rapid onset of action in patients with UC [8-11].
Filgotinib is an oral, once-daily, JAK-1 preferential inhibitor approved in Europe, the UK, and Japan for the treatment of UC [12-14]. The efficacy and safety of filgotinib in patients with moderately to severely active UC were evaluated in the phase 2b/3 randomized, double-blind, placebo (PBO)-controlled SELECTION study (NCT02914522) [15].
A post hoc analysis of SELECTION study data showed rapid and sustained improvements in UC symptoms with filgotinib 200 mg (FIL200) treatment compared with PBO [11]. Further post hoc and exploratory analyses of SELECTION revealed sustained improvements in both health-related quality of life and work productivity with FIL200 treatment compared with PBO [16,17]. The objective of this study was to assess symptomatic response to FIL200 induction treatment according to disease severity using baseline partial Mayo Clinic Score (pMCS; a combination of the Mayo rectal bleeding, stool frequency, and Physician Global Assessment subscores). Symptomatic response was assessed daily from baseline to day 15 and fortnightly from week 2 to week 10.
METHODS
1. Study Design and Patients
The study design of the SELECTION trial (Supplementary Fig. 1) has been reported previously [15]. In brief, adults aged 18 to 75 years with moderately to severely active UC (i.e., Mayo endoscopic subscore ≥2, rectal bleeding subscore ≥1, stool frequency subscore ≥1, Physician Global Assessment subscore ≥2; total Mayo Clinic Score [MCS] 6–12) were randomized 2:2:1 to receive oral, once-daily FIL200, filgotinib 100 mg, or PBO treatment for 11 weeks in either induction study A (biologic-naive patients) or induction study B (biologic-experienced patients).
Patients in clinical remission or with an MCS response (definitions provided in Table 1) at week 10 (responders) could enter the 47-week maintenance study. Those who received induction filgotinib were rerandomized 2:1 to continue their induction filgotinib regimen or receive PBO in the maintenance study. Those who received induction PBO continued receiving PBO in the maintenance study. This post hoc analysis included patients who received induction FIL200 or PBO.
Definitions of Outcomes from the SELECTION Study and This post hoc Analysis
The SELECTION trial (NCT02914522) was carried out in accordance with the International Conference on Harmonization Good Clinical Practice Guidelines and the Declaration of Helsinki. The study protocol and amendments were reviewed and approved by the Independent Ethics Committee or Institutional Review Board at each study site (Jikei University Institutional Review Board and others). All patients provided written informed consent.
2. Outcomes
1) Symptomatic Remission
Using baseline pMCS, the proportions of patients in symptomatic remission (definition is provided in Table 1) were analyzed daily from baseline to day 15 and every 2 weeks from week 2 to week 10. Baseline pMCS cutoffs used were pMCS ≥7 and pMCS <7, which had been previously used for severe disease (pMCS 7–9) and moderate disease (pMCS 5–6), respectively [18]. The proportions of patients in symptomatic remission were also assessed daily from baseline to day 15 and every 2 weeks from week 2 to week 10 using baseline MCS (a combination of the Mayo rectal bleeding, stool frequency, endoscopic, and Physician Global Assessment subscores). Baseline MCS cutoffs used were MCS ≥ 9 and MCS ≤ 8.
2) Characteristics Associated with Early Symptomatic Remission
Baseline characteristics were assessed for an association with symptomatic remission at induction week 2 among patients who received FIL200.
3) Clinical and Health-Related Quality of Life Outcomes
The proportions of patients who experienced clinical remission, MCS remission, an MCS response, endoscopic remission, an endoscopic response, histologic remission, and Inflammatory Bowel Disease Questionnaire (IBDQ) remission (definitions are provided in Table 1) were assessed at week 10 by baseline pMCS (pMCS ≥7 and pMCS <7).
4) Work Productivity and Activity Impairment Questionnaire-Ulcerative Colitis
Changes in the scores for the Work Productivity and Activity Impairment Questionnaire-Ulcerative Colitis (WPAI:UC) items absenteeism, presenteeism, work productivity loss, and activity impairment from baseline to week 10 were assessed by baseline pMCS (pMCS ≥7 and pMCS <7).
3. Statistical Analyses
These post hoc analyses were conducted using the SELECTION induction full analysis set, which included all randomized patients who received at least 1 dose of the study drug. Patient demographics and baseline characteristics were analyzed using descriptive statistics. Odds ratios (ORs) and associated 95% confidence intervals (CIs) and nominal P-values used to assess baseline characteristics associated with symptomatic remission at week 2 were obtained by univariate logistic regression analysis (categorical variables) or univariate linear regression analysis (numerical variables).
Symptomatic remission rates were compared between the pMCS ≥7 and pMCS <7 groups, and the MCS ≥9 and MCS ≤8 groups, in the FIL200 and PBO arms using a Cochran–Mantel–Haenszel test that was adjusted by study randomization stratification factors. Symptomatic remission rates were analyzed daily from baseline to day 15 and every 2 weeks from week 2 to week 10. The proportions of patients who experienced clinical remission, MCS remission, an MCS response, endoscopic remission, an endoscopic response, histologic remission, and IBDQ remission were compared between the pMCS ≥7 and pMCS <7 groups in the FIL200 and PBO arms using a Cochran–Mantel–Haenszel test that was adjusted by study randomization stratification factors at week 10.
Least-squares (LS) mean changes from baseline to week 10 for WPAI items, and associated 95% CIs and P-values, were assessed using an analysis of covariance model, which included baseline pMCS subgroup (pMCS ≥7 or pMCS <7), concomitant use of oral systemic corticosteroids (yes or no) and immunomodulators (yes or no) at day 1, and baseline score as covariates. Given the post hoc nature of these analyses, all P-values are nominal. Nominal P-values of <0.05 were considered to be statistically significant.
RESULTS
1. Patient Disposition and Baseline Characteristics
Overall, 382 biologic-naive and 404 biologic-experienced patients in induction studies A and B, respectively, were included in these analyses (Table 2). Of the patients who received FIL200 in SELECTION induction study A, 90 had a baseline pMCS of ≥7 and 155 had a baseline pMCS of <7. Of those who received FIL200 in induction study B, 148 had a baseline pMCS of ≥7 and 114 had a baseline pMCS of <7. Baseline characteristics of patients by pMCS (≥ 7 and <7) are shown in Table 2.
Baseline Characteristics of Patients by pMCS (≥7 and <7)
2. Proportions of Patients in Symptomatic Remission Over Time
1) Symptomatic Remission by Baseline pMCS
At day 2, symptomatic remission rates with FIL200 treatment were significantly higher in patients with a baseline pMCS of <7 than in those with a baseline pMCS of ≥ 7 (8.4% vs. 1.1%, P=0.009 [induction study A]; 8.8% vs. 0.7%, P=0.004 [induction study B]) (Fig. 1A and B). From baseline to day 15, symptomatic remission rates increased in both groups and, except for day 7 for induction study A and day 9 for induction study B, continued to be significantly higher in patients with a baseline pMCS of <7 than in those with a baseline pMCS of ≥ 7.
Symptomatic remission rates by baseline pMCS (≥7 and <7) from baseline to day 15 in induction study A (A) and B (B) and from baseline to week 10 in induction study A (C) and B (D). Blue asterisks denote statistically significant nominal P-values for comparisons between the pMCS ≥7 and <7 groups in the FIL200 arm. Red asterisks denote statistically significant nominal P-values for comparisons between the pMCS ≥7 and <7 groups in the PBO arm. Nominal P-values were obtained using a Cochran–Mantel–Haenszel test that was adjusted by study randomization stratification factors. Symptomatic remission was defined as a Mayo rectal bleeding subscore of 0 and a stool frequency subscore of ≤1. *P<0.05, **P<0.01, ***P<0.001. BL, baseline; CI, confidence interval; FIL200, filgotinib 200 mg; PBO, placebo; pMCS, partial Mayo Clinic Score.
From week 2, symptomatic remission rates with FIL200 treatment generally continued to increase in both the pMCS ≥7 and pMCS <7 groups (Fig. 1C and D). By week 10, symptomatic remission rates with FIL200 treatment were numerically different but were no longer significantly different between patients with a baseline pMCS of ≥ 7 and those with a baseline pMCS of <7 (43.3% vs. 54.8%, P=0.124 [induction study A]; 26.4% vs. 39.5%, P=0.099 [induction study B]).
In induction study A and induction study B, there was a similar trend in the proportions of patients in symptomatic remission in the pMCS ≥7 and pMCS <7 groups; however, these proportions were generally lower in induction study B than in induction study A.
2) Symptomatic Remission by Baseline MCS
Significant differences in symptomatic remission rates with FIL200 treatment between the MCS ≤8 and ≥9 groups were also seen at early timepoints, with significantly higher rates observed in the MCS ≤8 than ≥9 group at day 3 in induction study A (12.1% vs. 5.4%, P=0.033) and day 2 in induction study B (9.9% vs. 2.1%, P=0.036) (Supplementary Fig. 2A and B). From baseline to day 15, symptomatic remission rates generally increased in both groups and were numerically higher at all timepoints in the MCS ≤8 than ≥9 group. Compared with the analysis by pMCS groups, there were fewer timepoints in the first 2 weeks at which significant differences were observed between the groups, in particular in induction study B.
From week 2, symptomatic remission rates with FIL200 treatment generally continued to increase in both the MCS ≤ 8 and ≥ 9 groups (Supplementary Fig. 2C and D). In induction study A, symptomatic remission rates with FIL200 treatment at week 10 were numerically different but were no longer significantly different between the MCS ≤ 8 and ≥ 9 groups (56.0% vs. 45.7%, P=0.154). In induction study B, symptomatic remission rates with FIL200 treatment were significantly higher from week 4 through to week 10 in the MCS ≤8 than ≥9 group (at week 10: 45.1% vs. 27.2%, P=0.048).
3. Characteristics Associated with Early Symptomatic Remission
Given the significant differences in symptomatic remission rates at week 2 among the pMCS ≥7 and pMCS <7 groups, other baseline characteristics were assessed for an association with symptomatic remission at week 2. In induction study A, baseline characteristics that had a significant association with increased odds of symptomatic remission at week 2 were a baseline stool frequency subscore of ≤2 (vs. >2: OR =2.58, P=0.008) and a pMCS of <7 (vs. ≥ 7: OR=2.27, P=0.019) (Fig. 2A). Baseline C-reactive protein (CRP) levels of >4.415 and ≤ 11.550 mg/L (vs. ≤ 1.690 mg/L) were significantly associated with reduced odds of symptomatic remission at week 2 (OR=0.38, P=0.030).
Associations between baseline characteristics and symptomatic remission at week 2 among patients who received FIL200. ORs and associated 95% CIs and nominal P-values were obtained by univariate logistic regression analysis (categorical variables) or univariate linear regression analysis (numerical variables). aThe baseline cutoffs for CRP were determined based on data set quartile values for patients in induction study A and B. FIL200, filgotinib 200 mg; OR, odds ratio; CI, confidence interval; CRP, C-reactive protein; BMI, body mass index; MCS, Mayo Clinic Score; MES, Mayo endoscopic subscore; RB, rectal bleeding; SF, stool frequency; pMCS, partial Mayo Clinic Score; ref, reference; TNF, tumor necrosis factor.
In induction study B, baseline variables that had a significant association with increased odds of symptomatic remission at week 2 were an MCS of ≤ 8 (vs. ≥ 9: OR= 2.28, P=0.013), a stool frequency subscore of ≤ 2 (vs. >2: OR=3.92, P<0.001), and a pMCS of <7 (vs. ≥ 7: OR=2.88, P=0.001) (Fig. 2B). Other baseline variables that had a significant association with reduced odds of symptomatic remission at week 2 were CRP levels (OR =0.96, P=0.016), CRP levels >11.550 mg/L (vs. ≤1.690 mg/L: OR =0.22, P=0.003), prior use of vedolizumab (OR=0.39, P=0.003), prior failure of vedolizumab (OR=0.48, P=0.024), and prior use of a TNF antagonist and vedolizumab (OR=0.35, P=0.002).
4. Clinical, Health-Related Quality of Life, and Work Productivity Outcomes at Week 10
In induction study A, similar proportions of patients who received FIL200 in the pMCS ≥7 group and in the pMCS <7 group experienced clinical remission (25.6% vs. 26.5%; P=0.710) (Fig. 3A), MCS remission (24.4% vs. 24.5%; P=0.752) (Fig. 3B), an MCS response (66.7% vs. 66.5%; P=0.958) (Fig. 3C), endoscopic remission (12.2% vs. 12.3%; P=0.859) (Fig. 3D), an endoscopic response (36.7% vs. 32.3%, P=0.724) (Fig. 3E), histologic remission (33.3% vs. 36.1%, P=0.584) (Fig. 3F), and IBDQ remission (52.2% vs. 58.1%, P=0.735) (Fig. 3G) at week 10.
Proportions of patients by baseline pMCS (<7 vs. ≥7) experiencing clinical remission (A), MCS remission (B), an MCS response (C), endoscopic remission (D), an endoscopic response (E), histologic remission (F), and IBDQ remission (G) at week 10. Differences in response rates and associated 95% CIs (shown in parentheses) and nominal P-values were obtained using a Cochran–Mantel–Haenszel test that was adjusted by study randomization stratification factors. Error bars indicate 95% CIs for the response rates. aClinical remission was defined as an MES of 0 or 1, a Mayo rectal bleeding subscore of 0, and a decrease in Mayo stool frequency subscore of ≥1 point from induction baseline for a subscore of 0 or 1; bMCS remission was defined as an MCS of ≤2 and no single subscore of >1; cMCS response was defined as a decrease in MCS of ≥3 points and a ≥30% decrease from baseline, with a decrease in rectal bleeding subscore of ≥1 point, or an absolute rectal bleeding subscore of 0 or 1; dEndoscopic remission was defined as an MES of 0. eEndoscopic response was defined as an MES of 0 or 1; fHistologic remission was based on the Geboes Scale and required no or a mild increase in chronic inflammatory infiltrate in the lamina propria, no neutrophils in the lamina propria or epithelium, and no erosion, ulceration, or granulation tissue (grade 0 of ≤0.3, grade 1 of ≤1.1, grade 2a of ≤2A.3, grade 2b of 2B.0, grade 3 of 3.0, grade 4 of 4.0, and grade 5 of 5.0); gIBDQ remission was defined as an IBDQ total score of ≥170. pMCS, partial Mayo Clinic Score; MCS, Mayo Clinic Score; IBDQ, Inflammatory Bowel Disease Questionnaire; MES, Mayo endoscopic subscore; CI, confidence interval; FIL200, filgotinib 200 mg; PBO, placebo.
In induction study B, rates of outcomes at week 10 were generally non significantly lower in the pMCS ≥ 7 group than in the pMCS <7 group: clinical remission (8.1% vs. 15.8%, P=0.228) (Fig. 3A), MCS response (48.6% vs. 58.8%, P=0.385) (Fig. 3C), endoscopic remission (1.4% vs. 6.1%, P=0.083) (Fig. 3D), endoscopic response (14.2% vs. 21.1%, P=0.585) (Fig. 3E), histologic remission (15.5% vs. 25.4%, P=0.208) (Fig. 3F), and IBDQ remission (39.2% vs. 48.2%, P=0.460) (Fig. 3G). An exception was the rates of MCS remission, which were significantly lower in the pMCS ≥ 7 group than in the <7 group in induction study B (4.7% vs. 15.8%, P=0.026).
From baseline to week 10, there were no significant differences between the pMCS ≥7 and pMCS <7 groups with FIL200 treatment for LS-mean changes in absenteeism (−8.2% vs. −10.8%, P=0.458), presenteeism (−18.3% vs. −19.5%, P=0.769), work productivity loss (−21.8% vs. −22.9%, P=0.813), or activity impairment (−23.3% vs. −23.3%, P=0.999) in induction study A (Fig. 4). Similarly, in induction study B, there were no significant differences between the pMCS ≥7 and pMCS <7 groups with FIL200 treatment for absenteeism (−12.1% vs. −8.8%, P=0.366), presenteeism (−17.4% vs. −14.9%, P=0.521), work productivity loss (−24.4% vs. −19.8%, P=0.326), or activity impairment (−26.0% vs. −23.3%, P=0.443) (Fig. 4).
Change from baseline to week 10 by baseline pMCS (<7 vs. ≥7) in the WPAI:UC items absenteeism (A), presenteeism (B), work productivity loss (C), and activity impairment (D). The LS means, 95% CIs, and nominal P-values were obtained using an ANCOVA model, which included baseline pMCS subgroup, concomitant use of oral systemic corticosteroids (yes or no) and immunomodulators (yes or no) at day 1, and baseline score as covariates. pMCS, partial Mayo Clinic Score; WPAI:UC, Work Productivity and Activity Impairment Questionnaire-Ulcerative Colitis; LS, least-squares; CI, confidence interval; ANCOVA, analysis of covariance; FIL200, filgotinib 200 mg; PBO, placebo.
DISCUSSION
In this post hoc analysis of the phase 2b/3 SELECTION study, patients in the less severe disease group (pMCS <7) initially showed significantly higher symptomatic remission rates with FIL200 treatment (as early as day 2 and generally up to and including day 15), compared with those in the more severe disease group (pMCS ≥7). However, by week 10, there were no longer significant differences in symptomatic remission rates between these groups. Similar trends in symptomatic remission rates over time were observed when patients were stratified by MCS, which includes an endoscopic assessment not present in the pMCS. These findings are consistent with the post hoc analysis of SELECTION study data, which showed rapid and sustained improvements in UC symptoms with FIL200 treatment compared with PBO [11]. These findings indicate that the speed with which patients respond to filgotinib may be dependent on patients’ baseline disease severity.
Baseline characteristics associated with less severe disease, such as a stool frequency subscore of ≤2, an MCS of ≤8 (for biologic-experienced patients), and a pMCS of <7, were associated with a significantly increased likelihood of early symptomatic remission (i.e., at week 2) among patients who received FIL200 treatment. These findings support a previous study, which demonstrated high positive predictive values for symptomatic improvement indicators (i.e., a rectal bleeding subscore of 0 and/or a stool frequency subscore of ≤1) at day 7 for long-term MCS response in patients receiving filgotinib [11]. Conversely, elevated CRP levels, prior experience with vedolizumab, prior failure of vedolizumab, and prior use of a TNF antagonist and vedolizumab were associated with a decreased likelihood of early symptomatic remission. It is important to acknowledge that most (90%) of the FIL200-treated biologicexperienced patients who were exposed to vedolizumab also had prior use of a TNF antagonist. Therefore, it is possible that the lower likelihood of symptomatic remission among these patients was due to refractory disease rather than vedolizumab use. Moreover, while for biologic-naive patients baseline CRP levels >4.415 and ≤11.550 mg/L (vs. ≤1.690 mg/L) were significantly associated with reduced odds of symptomatic remission at week 2, CRP levels >11.550 mg/L (vs. ≤1.690 mg/L) were not. Further investigation is needed to understand the extent to which CRP levels are associated with early symptomatic remission. Nevertheless, along with the symptomatic remission results over time, these data suggest that patients with less severe UC may be more likely to experience rapid symptomatic relief with FIL200 treatment than patients with more severe UC. The results from this study also suggest that patients who do not experience rapid symptomatic relief owing to their level of disease severity may benefit from continued FIL200 treatment, because there were no significant differences between the rates of symptomatic remission and the rates of most of the clinical, health-related quality of life, and work productivity outcomes at week 10 among the pMCS disease severity groups. Together, these findings indicate that the disease severity of patients should be an important consideration when assessing the efficacy of filgotinib in the management of UC, for example when selecting drugs before starting treatment and when assessing the efficacy of filgotinib after starting treatment.
In induction studies A and B, there was a similar trend in the proportions of patients achieving symptomatic remission and clinical, health-related quality of life, and work productivity outcomes at week 10. However, the proportions of patients achieving these outcomes were generally lower and there were greater numerical differences (and a significant difference for MCS remission) between the pMCS disease severity groups in induction study B than induction study A. Treatment benefits were therefore less pronounced in patients who were biologic-experienced than in those who were biologicnaive, suggesting that treatment history should also be considered when assessing filgotinib efficacy.
The limitations of this study include the inherent constraints associated with post hoc analyses, which may limit the generalizability and inferential power of the findings. There are also some limitations associated with the use of the pMCS to categorize disease severity groups: first, the inclusion of a Physician’s Global Assessment introduces subjectivity; second, the pMCS, despite being commonly used in clinical trials, may not reflect the tools and criteria used in routine clinical practice to identify patients with differing therapeutic needs; and third, the lack of endoscopic assessment could affect the accurate stratification of disease severity. Notably, similar trends in symptomatic remission rates over time were observed when patients were stratified by MCS, suggesting that the absence of endoscopic assessment in the pMCS did not significantly impact the accurate stratification of disease severity.
This SELECTION post hoc analysis found that, among patients receiving FIL200 in the SELECTION study, symptomatic remission rates were higher at day 2 and were generally higher up to and including day 15 in patients in the less severe disease group (pMCS <7) than in those in the more severe disease group (pMCS ≥ 7). However, 10 weeks of FIL200 treatment led to similar rates of symptomatic remission and other clinical, health-related quality of life, and work productivity outcomes regardless of patients’ disease severity.
Notes
Funding Source
The SELECTION trial was sponsored by Gilead Sciences, Inc. (Foster City, California, USA). Galapagos NV (Mechelen, Belgium) collaborated in the SELECTION study. This post hoc analysis was funded by Gilead Sciences K.K. (Tokyo, Japan), Eisai Co., Ltd. (Tokyo, Japan), and EA Pharma Co., Ltd. (Tokyo, Japan).
Conflict of Interest
Saruta M has received scholarship grants from AbbVie GK, CMIC CMO Co., Ltd., Kissei Pharmaceutical Co., Ltd., Mochida Pharmaceutical Co., Ltd., PPD-SNBL K.K., and Zeria Pharmaceutical Co., Ltd., and lecture fees from AbbVie GK, EA Pharma Co., Ltd., Gilead Sciences K.K., Janssen Pharmaceutical K.K., Kissei Pharmaceutical Co., Ltd., Mitsubishi Tanabe Pharma Corporation, Mochida Pharmaceutical Co., Ltd, Nobelpharma Co., Ltd., Takeda Pharmaceutical Co., Ltd., and Viatris Pharmaceutical Co., Ltd. outside the submitted work. Danese S reports consultancy fees from AbbVie, Alimentiv, Allergan, Amgen, AstraZeneca, Athos Therapeutics, Biogen, Boehringer Ingelheim, Celgene, Celltrion, Dr Falk Pharma, Eli Lilly, Enthera Pharmaceuticals, Ferring Pharmaceuticals, Gilead, Hospira, Inotrem, Janssen, Johnson & Johnson, MSD, Mundipharma, Mylan, Pfizer, Roche, Sandoz, Sublimity Therapeutics, Takeda, TiGenix, UCB, and Vifor; and lecture fees from AbbVie, Amgen, Ferring Pharmaceuticals, Gilead, Janssen, Mylan, Pfizer, and Takeda. Takatori Y and Kaise T are employees of Gilead Sciences K.K. and are shareholders of Gilead Sciences, Inc. Rudolph R is an employee of Alfasigma S.p.A. Ferrante M has received financial support for research from AbbVie, Biogen, EG Pharmaceuticals, Janssen Pharmaceuticals, Pfizer, Takeda, and Viatris; speaker fees from AbbVie, Biogen, Boehringer Ingelheim, Dr Falk Pharma, Ferring Pharmaceuticals, Janssen-Cilag, MSD, Pfizer, Takeda, Truvion Healthcare, and Viatris; and is a consultant for AbbVie, Agomab Therapeutics, Boehringer Ingelheim, Celgene, Celltrion, Eli Lilly, Janssen-Cilag, MRM Health, MSD, Pfizer, Takeda, and Thermo Fisher Scientific. Hibi T has received lecture fees from AbbVie GK, EA Pharma Co., Ltd., Janssen Pharmaceutical K.K., JIMRO Co., Ltd., Mitsubishi Tanabe Pharma Corporation, Mochida Pharmaceutical Co., Ltd., Pfizer Japan Inc., Sandoz K.K., Takeda Pharmaceutical Co., Ltd., and Zeria Pharmaceutical Co., Ltd.; advisory/consultancy fees from AbbVie GK, Celltrion Healthcare Japan K.K., EA Pharma Co., Ltd., Eli Lilly Japan K.K., Gilead Sciences K.K., Janssen Pharmaceutical K.K., Mitsubishi Tanabe Pharma Corporation, Nichi-Iko Pharmaceutical Co., Ltd., Nippon Kayaku Co., Ltd., Takeda Pharmaceutical Co., Ltd., and Zeria Pharmaceutical Co., Ltd.; and research grants from AbbVie GK, Activaid, Alfresa Pharma Corporation, Bristol Myers Squibb K.K., Eli Lilly Japan K.K., Ferring Pharmaceuticals Co., Ltd., Gilead Sciences K.K., Janssen Pharmaceutical K.K., JIMRO Co., Ltd., JMDC Inc., Kyorin Pharmaceutical Co., Ltd., Miyarisan Pharmaceutical Co., Ltd., Mochida Pharmaceutical Co., Ltd., Nippon Kayaku Co., Ltd., Pfizer Japan Inc., and Zeria Pharmaceutical Co., Ltd. Hibi T is an editorial board member of the journal but was not involved in the peer reviewer selection, evaluation, or decision process of this article.
Data Availability Statement
Gilead Sciences shares anonymized individual patient data upon request or as required by law or regulation with qualified external researchers based on submitted curricula vitae and reflecting non conflict of interest. The request proposal must also include a statistician. Approval of such requests is at Gilead Sciences’ discretion and is dependent on the nature of the request, the merit of the research proposed, the availability of the data, and the intended use of the data. Data requests should be sent to datarequest@gilead.com.
Author Contributions
Conceptualization: Takatori Y, Kaise T. Formal analysis: Saruta M. Investigation: Saruta M, Danese S, Ferrante M, Hibi T. Methodology: Takatori Y, Kaise T, Rudolph C. Project administration: Saruta M, Takatori Y. Writing - review & editing: all authors. Approval of final manuscript: all authors.
Additional Contributions
The authors thank Ruby Oberin PhD and Tamsyn Stanborough PhD of Oxford PharmaGenesis, Melbourne, Australia, for providing medical writing support, which has been funded by Gilead Sciences K.K., Tokyo, Japan, Eisai Co., Ltd., Tokyo, Japan, and EA Pharma Co., Ltd., Tokyo, Japan, in accordance with Good Publication Practice 2022 (GPP 2022) guidelines (https://www.ismpp.org/gpp-2022).
Supplementary Material
Supplementary materials are available at the Intestinal Research website (https://www.irjournal.org).
Supplementary Fig. 1.
SELECTION study design. Treatment arms in the maintenance study are greyed out as they are not relevant to this post hoc analysis. aPatients were eligible to enter the maintenance study if they experienced clinical remission (defined as an MES of 0 or 1, a rectal bleeding subscore of 0, and a ≥1-point decrease in stool frequency from induction baseline to achieve a subscore of 0 or 1) or an MCS response (defined as a reduction of ≥3 points in MCS and a ≥30% decrease from induction baseline, with an accompanying decrease in rectal bleeding subscore of ≥1 point, or an absolute rectal bleeding subscore of 0 or 1) at week 10. FIL100, filgotinib 100 mg; FIL200, filgotinib 200 mg; MCS, Mayo Clinic Score; MES, Mayo endoscopic subscore; PBO, placebo; PTM, placebo to match.
Supplementary Fig. 2.
Symptomatic remissiona rates by baseline MCS (≥9 and ≤8) from baseline to day 15 in induction study A (A) and B (B) and from baseline to week 10 in induction study A (C) and B (D). *P<0.05, **P<0.01. Blue asterisks denote statistically significant nominal P-values for comparisons between the pMCS ≥9 and ≤8 groups in the FIL200 arm. Red asterisks denote statistically significant nominal P-values for comparisons between the pMCS ≥9 and ≤8 groups in the PBO arm. Nominal P-values were obtained using a Cochran–Mantel–Haenszel test that was adjusted by study randomization stratification factors. aSymptomatic remission was defined as a Mayo rectal bleeding subscore of 0 and a stool frequency subscore of ≤1. BL, baseline; CI, confidence interval; FIL200, filgotinib 200 mg; PBO, placebo; MCS, Mayo Clinic Score.