Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients With Ulcerative Colitis

Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients With Ulcerative Colitis

Journal Pre-proof Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients with Ulcerative Colitis William J. Sandborn, Laurent Peyr...

4MB Sizes 0 Downloads 0 Views

Journal Pre-proof Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients with Ulcerative Colitis William J. Sandborn, Laurent Peyrin-Biroulet, Jinkun Zhang, Michael Chiorean, Séverine Vermeire, Scott D. Lee, Tanja Kühbacher, Bruce Yacyshyn, Christopher H. Cabell, Snehal U. Naik, Preston Klassen, Julián Panés PII: DOI: Reference:

S0016-5085(19)41526-6 https://doi.org/10.1053/j.gastro.2019.10.035 YGAST 62991

To appear in: Gastroenterology Accepted Date: 29 October 2019 Please cite this article as: Sandborn WJ, Peyrin-Biroulet L, Zhang J, Chiorean M, Vermeire S, Lee SD, Kühbacher T, Yacyshyn B, Cabell CH, Naik SU, Klassen P, Panés J, Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients with Ulcerative Colitis, Gastroenterology (2019), doi: https:// doi.org/10.1053/j.gastro.2019.10.035. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 by the AGA Institute

What you need to know: BACKGROUND AND CONTEXT: Etrasimod is an oral sphingosine 1-phosphate receptor modulator in development for immune-mediated inflammatory disorders that might be used to treat ulcerative colitis (UC). NEW FINDINGS: In a phase 2 randomized trial of patients with moderately to severely active UC, etrasimod (2 mg) was more effective than placebo in reducing clinical and endoscopic features of UC. LIMITATIONS: Larger studies of more patients are needed to determine the efficacy and safety of etrasimod in patients with UC. IMPACT: With further development, etrasimod might be used to treat patients with UC. Lay Summary: In a clinical trial, etrasimod reduced clinical and endoscopic features of UC.

Etrasimod, an oral, selective, S1P receptor modulator for patients with moderately-to-severely active ulcerative colitis Improvement in mMCS from baseline

Patients with endoscopic improvement, %

P = 0.009

P = 0.003

Patients with clinical remission, % P < 0.001

41.8

2.49

33.0

1.94 1.5

22.5

17.8

16.0 8.1

Etrasimod 2 mg (n = 50)

Etrasimod 1 mg (n = 52)

12-week treatment mMCS, modified Mayo Clinic Score (endoscopy findings, rectal bleeding, stool frequency), range 0–9 P values vs placebo

Placebo (n = 54)

Efficacy and Safety of Etrasimod in a Phase 2 Randomized Trial of Patients with Ulcerative Colitis Short Title: Etrasimod for Ulcerative Colitis William J. Sandborn,1 Laurent Peyrin-Biroulet,2 Jinkun Zhang,3 Michael Chiorean,4 Séverine Vermeire,5 Scott D. Lee,6 Tanja Kühbacher,7 Bruce Yacyshyn8 Christopher H. Cabell,3 Snehal U. Naik,3 Preston Klassen,3 Julián Panés9 1

University of California San Diego, La Jolla, California; 2Department of Gastroenterology, Nancy

University Hospital, Lorraine University, Vandoeuvre-lès-Nancy, France; 3Arena Pharmaceuticals, San Diego, California; 4Division of Gastroenterology, Virginia Mason Medical Center, Seattle, Washington; 5

Department of Gastroenterology & Hepatology, University Hospitals Leuven, Leuven, Belgium;

6

University of Washington Medical Center, Seattle, Washington; 7Asklepios Westklinikum Hamburg

and Christian Albrechts University, Hamburg, Germany; 8Department of Internal Medicine, University of Cincinnati, Cincinnati, Ohio; 9Hospital Clinic de Barcelona, IDIBAPS, CIBERehd, Barcelona, Spain

Funding This study was supported by Arena Pharmaceuticals, Inc. Medical writing assistance in development of the manuscript was funded by Arena Pharmaceuticals, Inc. Abbreviations used in this paper: AEs, adverse events; CI, confidence interval; LSM, least squares mean; MCS, Mayo Clinic score; S1P, sphingosine 1-phosphate; UC, ulcerative colitis; TEAEs, treatment-emergent adverse events; TNFα, tumor necrosis factor alpha. Correspondence:

William J. Sandborn, MD University of California San Diego 9500 Gilman Drive, MC 0956 La Jolla, CA 92093 Phone: 858-657-5331 Email: [email protected]

Disclosures WJS received research grants from AbbVie, Amgen, Atlantic Healthcare Limited, Celgene/Receptos, Genentech, Gilead Sciences, Janssen, Lilly, Pfizer, Prometheus Laboratories and Takeda; consulting fees from AbbVie, Allergan, Amgen, Arena Pharmaceuticals, Avexegen Therapeutics, BeiGene, Boehringer Ingelheim, Celgene, Celltrion, Conatus, Cosmo, Escalier Biosciences, Ferring, Forbion, Genentech, Gilead Sciences, Gossamer Bio, Incyte, Janssen, Kyowa Kirin Pharmaceutical Research, Landos Biopharma, Lilly, Oppilan Pharma, Otsuka, Pfizer, Precision IBD, Progenity, Prometheus Laboratories, Reistone, Ritter Pharmaceuticals, Robarts Clinical Trials (owned by Health Academic Research Trust, HART), Seres Therapeutics, Shire, Sienna Biopharmaceuticals, Sigmoid Biotechnologies, Sterna Biologicals, Sublimity Therapeutics, Takeda, Theravance Biopharma, Tigenix, Tillotts Pharma, UCB Pharma, Ventyx Biosciences, Vimalan Biosciences, Vivelix Pharmaceuticals; and stock or stock options from BeiGene, Escalier Biosciences, Gossamer Bio, Oppilan Pharma, Precision IBD, Progenity, Ritter Pharmaceuticals, Ventyx Biosciences, Vimalan Biosciences. WJS reports the following financial relationships for an immediate family member (spouse): consultant and stock or

stock options: Opthotech, and Progenity; employee and stock options: Escalier Biosciences, Oppilan Pharma, Precision IBD, Ventyx Biosciences, and Vimalan Biosciences. LP-B has received grant support from AbbVie, MSD, and Takeda; personal fees from AbbVie, Allergan, Alma, Amgen, Arena Pharmaceuticals, Biogen, Boehringer Ingelheim, Celgene, Celltrion, Enterome, Ferring, Genentech, Gilead, Hikma, Index Pharmaceuticals, Janssen, MSD, Nestlé, Pfizer, Pharmacosmos, Roche, Samsung, Sandoz, Sterna, Takeda, and Tillots; and holds stock options from CTMA-Clinical Trials Mobile Application. MC has received consulting and/or speaking fees from AbbVie, Arena Pharmaceuticals, Celgene, Janssen, Medtronic, Pfizer, Takeda, and UCB. SV has received grants from AbbVie, J&J, Pfizer, and Takeda, and has received consulting and/or speaking fees from AbbVie, Avaxia, Celgene, Dr. Falk Pharma, Ferring, Galapagos, Genentech-Roche, Gilead, Hospira, Janssen, Mundipharma, MSD, Pfizer, Prodigest, Prometheus, Robarts Clinical Trials, Second Genome, Shire, Takeda, and Tillots. SL has received financial support for research from AbbVie, Amgen, Genentech, Janssen, Pfizer Takeda and UCB, and consulting fees from Janssen, Takeda, and UCB. TK has received personal fees from Arena Pharmaceuticals. BY has received grant funding from Proctor and Gamble and Merck and served on an advisory board of Arena Pharmaceuticals and as a consultant to Gilead. JP has received research grants from Abbvie and MSD and received consulting and/or speaking fees from AbbVie, Abbott, Arena Pharmaceuticals, Boehringer Ingelheim, Celgene, Genentech-Roche, GoodGut, GSK, Janssen, MSD, Nestle, Oppilan, Pfizer, Progenity, Takeda, Theravance, and TiGenix. JZ, CHC, SUN, and PK are employed by Arena Pharmaceuticals.

Writing Assistance Medical writing support was provided by Steven Goodrick, PhD, and Elizabeth Strickland, PhD, inScience Communications, Springer Healthcare (Philadelphia, PA, USA), and funded by Arena Pharmaceuticals.

Author Contributions All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this manuscript and take responsibility for the integrity of the work as a whole. The first author and the sponsor (Arena Pharmaceuticals, Inc.) designed the study. Data were collected and analyzed by a contract research organization (Pharmaceutical Product Development). All authors had full access to the data. The authors and the sponsor jointly interpreted the data. The first author and medical writers from inScience Communications (funded by the sponsor) wrote the first draft of the manuscript. All authors participated in subsequent drafts and in the decision to submit the manuscript for publication, and vouch for the completeness and veracity of the data and analyses reported and for the adherence of the trial to the protocol.

Acknowledgments The authors would like to acknowledge the patients and staff of the participating study sites.

Abstract Background and Aims Etrasimod (APD334) is an oral, selective sphingosine 1-phosphate receptor modulator in development for immune-mediated inflammatory disorders. We assessed the efficacy and safety of etrasimod in patients with moderately to severely active ulcerative colitis (UC).

Methods In a phase 2, proof of concept, double-blind, parallel-group study, adult outpatients with modified Mayo Clinic scores (MCSs; stool frequency, rectal bleeding, and endoscopy findings) of 4–9, endoscopic subscores of 2 or more, and rectal bleeding subscores of 1 or more were randomly assigned to groups given once-daily etrasimod 1 mg (n=52), etrasimod 2 mg (n=50), or placebo (n=54) for 12 weeks. The study was performed from October 15, 2015 through February 14, 2018 at 87 centers in 17 countries. The primary endpoint was an increase in the mean improvement in modified MCS from baseline to week 12. Secondary endpoints included the proportion of patients with endoscopic improvement (subscores of 1 or less) from baseline to week 12. Exploratory endpoints, including clinical remission, are reported in the manuscript, although the study was only statistically powered to draw conclusions on the primary endpoint.

Results At week 12, the etrasimod 2 mg group met the primary and all secondary endpoints. Etrasimod 2 mg led to a significantly greater increase in mean improvement in modified MCS from baseline than placebo (difference from placebo, 0.99 points; 90% CI, 0.30–1.68; P=.009), and etrasimod 1 mg led to an

increase in mean improvement from baseline in modified MCS of 0.43 points more than placebo (90% CI, reduction of 0.24 to increase of 1.11; nominal P=.15). Endoscopic improvement occurred in 41.8% of patients receiving etrasimod 2 mg vs 17.8% receiving placebo (P=.003). Most adverse events were mild to moderate. Three patients had a transient, asymptomatic, low-grade atrioventricular block that resolved spontaneously and did not recur with further dosing; 2 of the 3 patients had evidence of atrioventricular block prior to etrasimod exposure.

Conclusions In patients with moderately to severely active UC, etrasimod 2 mg was more effective than placebo in producing clinical and endoscopic improvements. Further clinical development is warranted. Clinicaltrials.gov no: NCT02447302.

Keywords: Immunomodulatory Therapy; Inflammatory Bowel Disease; S1P Receptor Modulator; Ulcerative Colitis

Introduction Ulcerative colitis (UC) is a chronic, immune-mediated inflammatory disease of the colon that is characterized by diffuse mucosal inflammation.1,2 Current treatments for UC include mesalamine, corticosteroids, azathioprine, anti-tumor necrosis factor alpha (TNFα) agents, anti-α4β7 integrin agents, and Janus kinase inhibitors.3 Despite this growing treatment armamentarium, many patients still experience unacceptable side effects, inconvenience leading to poor adherence, or an inadequate or unsustained response, indicating an unmet need for additional therapies. Sphingosine 1-phosphate (S1P) is a membrane-derived lysophospholipid signaling molecule involved in regulation of multiple physiologic and pathophysiologic processes, mediated through the 5 isoforms of the G protein–coupled S1P receptor (S1P1 through S1P5).4,5 The function of each receptor is highly dependent on the cell type on which it is expressed. S1P1, S1P2, and S1P3 are widely expressed throughout the body, while S1P4 and S1P5 have more limited expression in immune and central nervous system cell subsets.6,7 S1P2 and S1P3 are involved in regulation of endothelial barrier function, fibrosis, and vasoconstriction.7,8 S1P4 may play a role in dendritic cell trafficking and function, and S1P5 has been implicated in regulation of natural killer cell trafficking.6,7 A key role of S1P in innate and adaptive immunity is regulation of lymphocyte trafficking, Th17 polarization, dendritic cell differentiation, and natural killer cell migration.5,9-12 S1P1 expression on lymphocytes plays a key role in lymphocyte trafficking by regulating lymphocyte egress from lymphoid organs.7,13 Upon binding to S1P1, synthetic modulators induce and sustain receptor internalization. Loss of cell surface–expressed S1P1 prevents cells from migrating along S1P gradients and results in retention of lymphocytes within lymphoid tissue and subsequent reduction of lymphocyte accumulation in peripheral tissues.9,14 S1P1 also has physiologic functions beyond its role in lymphocyte trafficking.

S1P1 expression on cardiac cells is involved in regulation of heart rate, and expression on endothelial cells maintains barrier integrity.6,7,15 Several S1P receptor modulators have been explored as therapies for immune-mediated diseases. Fingolimod and siponimod have been approved for treatment of relapsing forms of multiple sclerosis.16-20 Other S1P receptor modulators are in clinical development as potential therapies, including ozanimod in multiple sclerosis and UC,21-23 ponesimod in multiple sclerosis and psoriasis,24,25 and amiselimod in multiple sclerosis and Crohn’s disease.26-28 Fingolimod, a first-generation S1P receptor modulator, interacts nonselectively with S1P isoforms 1 to 5.9,16,29 This lack of selectivity has been suggested to contribute to some of the adverse events (AEs) observed in patients treated with fingolimod, including pulmonary fibrosis and elevation of liver enzymes.14 Fingolimod has also been associated with macular edema,30 which is likely partially due to an S1P1 effect that is exacerbated by S1P2 and S1P3,31 and with postmarketing cases of progressive multifocal leukoencephalopathy in patients with multiple sclerosis.32,33 Newer S1P receptor modulators with selective receptor isoform affinities are currently in development.5,21,34 Etrasimod is a selective S1P1, S1P4, and S1P5 receptor modulator that is being developed for treatment of immune-mediated inflammatory disorders.34,35 Following treatment with once-daily etrasimod 2 mg, an approximately 53% decrease from baseline in mean lymphocyte counts was observed in healthy volunteers at day 3, with a continued decline to 69% by day 21 and lymphocyte recovery to within 5% of baseline 7 days after drug discontinuation.36,37 The current study aimed to determine the safety and efficacy of etrasimod in patients with moderately to severely active UC.

Materials and Methods Study Design This phase 2, randomized, double-blind, parallel-group, placebo-controlled study of induction therapy with etrasimod was conducted from October 15, 2015, to February 14, 2018, at 87 centers in 17 countries (supplementary materials). The study was conducted in accordance with the International Conference on Harmonisation Guideline for Good Clinical Practice and was approved by the institutional review board at each center. All patients provided written informed consent. All authors had full access to the data, control of the content of the manuscript, and had final responsibility for the decision to submit for publication.

Patients Eligible patients were aged 18–80 years with UC and a modified Mayo Clinic score (MCS) of 4–9, including a centrally read endoscopic subscore of ≥2 and a rectal bleeding subscore of ≥1. Patients with disease limited to the rectum were excluded. Disease activity was measured with a modified MCS (range, 0–9; composed of 3 subscores [endoscopic findings, rectal bleeding, and stool frequency], each with range 0–3). Higher scores indicate more severe disease. Stable doses of oral mesalamine and corticosteroids (prednisone ≤10 mg/day, budesonide ≤9 mg/day, or equivalent steroid) were permitted except for rectally administered agents, which had to be discontinued 2 weeks before the first dose of study drug; biologics were discontinued 60 days or more before randomization; and immunosuppressants were discontinued at randomization. Complete inclusion and exclusion criteria are provided in the supplementary materials.

Randomization and Masking Patients were randomly assigned 1:1:1 to receive oral etrasimod 1 mg, etrasimod 2 mg, or placebo once daily for 12 weeks (Supplementary Figure S1). Randomization was performed centrally with block size of 6 and was stratified by current use of corticosteroids and previous exposure to anti-TNFα agents. Randomization codes were generated by a statistician not directly involved with the study. The sponsor, patients, and personnel involved with the conduct of the study, with the exception of the clinical supply staff, safety staff, and the statistician supporting the Data Safety Monitoring Board, were blinded to the identity of the study medication; placebo and study drug were supplied as capsules with the same appearance.

Procedures Patients were assessed for efficacy at baseline and week 12 with the MCS,38 including flexible sigmoidoscopy. Histological severity was evaluated using the Geboes Index.39 Endoscopic scores were assessed from a single read by a central, blinded reader; histological scores were assessed by a single blinded reader. Endoscopy and histology central readers were blinded to clinical information, treatment assignment, and visit sequence. Additional visits were performed for safety monitoring and collection of patient diary information. Blood samples were taken at each visit for clinical chemistry and hematologic analysis. Stool samples were taken during screening and at weeks 4, 8, and 12 for measurement of fecal calprotectin. Patients receiving corticosteroids at baseline continued treatment at stable doses; dose tapering was not permitted during the study period.

Outcomes The primary efficacy endpoint was improvement from baseline in the modified MCS at week 12. The primary efficacy endpoint of improvement from baseline in the modified MCS was prespecified prior to database lock, a change from the primary endpoint of clinical remission in the original protocol made to reduce the estimated required sample size. The proportion of patients with clinical remission was retained as a prespecified exploratory outcome. Secondary efficacy endpoints were the proportion of patients at week 12 who achieved endoscopic improvement (defined as an endoscopic subscore of ≤1 point), improvement in the 2-component MCS (range, 0–6, including rectal bleeding and endoscopy findings), and improvement in the total MCS (range, 0–12, composed of the modified MCS plus Physician Global Assessment). Exploratory outcomes assessed at week 12 included clinical remission (Mayo Clinic endoscopic subscore ≤1 [with absence of friability], rectal bleeding score ≤1, and stool frequency score ≤1 with a frequency decrease of ≥1 point from baseline), clinical response (met the criteria for clinical remission or had a decrease in modified MCS of ≥2 points and a decrease of ≥30%, with either a rectal bleeding score of ≤1 or a decrease in rectal bleeding of ≥1), histological improvement (Geboes score <3.1), histological remission (Geboes score <2.0), and peripheral blood lymphocyte counts. After the study commenced, the secondary endpoints of improvement in the 2-component MCS and improvement in the total MCS were added. The proportion of patients at week 12 with clinical response (originally a secondary endpoint) and clinical remission (originally the primary endpoint) were made exploratory endpoints. AEs were coded using the Medical Dictionary for Regulatory Activities, version 20.1. Anemia was based on investigator diagnosis. The complete schedule of procedures and monitoring for AEs is described in the supplementary materials.

Statistical Analysis Demographic and disease characteristics at baseline were summarized using descriptive statistics with no formal statistical testing comparing groups. Primary and secondary efficacy endpoints at week 12 were analyzed in the intention-to-treat population (all randomized patients who received 1 or more doses of study drug) using the multiple imputation method for missing data (including missing subscores and individual components of composite endpoints). Sensitivity analyses were conducted for the primary endpoint and for clinical remission to test the robustness of results. Exploratory endpoints were analyzed in either the intention-to-treat population using multiple imputation or in the modified intention-to-treat population (all randomized patients who received 1 or more doses of study drug and had a measurement at baseline and at least 1 post-randomization measurement) for all other variables. The safety population consisted of all randomized patients who received 1 or more doses of the study drug. No missing data were imputed for repeated-measures analyses or safety analyses. The primary efficacy endpoint was analyzed using an analysis of covariance model that incorporated treatment, current oral corticosteroid use, prior exposure to anti-TNFα agents, and baseline value as covariates. All statistical testing was 1-sided at the 0.05 level of significance, as appropriate for a proof of concept, phase 2 study.40 The primary comparison was between etrasimod 2 mg and placebo. Comparisons between all other groups were exploratory. Nominal P values were reported for all exploratory endpoints. A hierarchical testing strategy was pre-specified and applied to all secondary endpoints. Any endpoint subsequent to a comparison that did not meet statistical significance was considered exploratory in the hierarchical order of: improvement in modified MCS (the primary endpoint), proportion of patients achieving endoscopic improvement, improvement in 2-component MCS, and improvement in total MCS.

The secondary endpoints of improvement in 2-component MCS and total MCS were analyzed using the same model as the primary endpoint. The proportion of patients who achieved endoscopic improvement (a secondary endpoint) was analyzed using the Mantel–Haenszel method, adjusted for stratification factors of current corticosteroid therapy at baseline and previous exposure to anti-TNFα agents. For the exploratory endpoints, the Mantel–Haenszel method was used for proportion-based efficacy endpoints. A mixed-effects model for repeated measures that included current corticosteroid use, previous exposure to anti-TNFα agents, treatment, week, treatment-by-week interaction, and baseline value as covariates was used for longitudinal continuous efficacy endpoints. An unstructured covariance model was used for within-patient correlation for patient-reported outcomes. Dose-response analyses were performed using an analysis of covariance model for continuous variables or a logistic regression model for categorical variables (treatment group and appropriate baseline covariates were included in the analysis model). Subgroup analyses for improvement of modified MCS used an analysis of covariance with terms for treatment, subgroup, treatment-by-subgroup interaction, and baseline value as covariate. Prespecified analyses were for subgroups of sex, age, race, current oral corticosteroid use, prior exposure to antiTNFα agents, response to anti-TNFα agents, baseline fecal calprotectin, baseline C-reactive protein, and total MCS. Subgroup analyses by geographic region; exposure to anti-integrin agents, ≥1 biologic agent, ≥2 biologic agents, vedolizumab, and immunosuppressives; baseline lymphocyte count; duration of UC; and history of pancolitis, proctosigmoiditis, or pancolitis and proctosigmoiditis were post hoc. For comparison of the primary efficacy measure, a sample of ~39 eligible patients per group would provide ~80% power to detect a difference of 1.15 in improvement in the modified MCS at an overall level of significance of 0.05 (1-sided test) based on the estimated pooled standard deviation of 2.03. The

usage of a 90% confidence interval (CI) is appropriate and consistent with sample size consideration. The study was not powered to make formal statistical comparisons of the exploratory endpoints. For the exploratory endpoint of clinical remission, it was estimated that a sample of ~80 patients per group would be required to provide 80% power to detect a difference of 18% at an overall significance level of 5% (2-sided test).

Results Patient Characteristics A total of 156 patients were randomized to etrasimod 1 mg (n = 52), etrasimod 2 mg (n = 50), or placebo (n = 54) (Supplementary Figure S2). Demographics and disease characteristics of the patient groups are shown in Table 1. Of the randomized patients, 141 (90.4%) completed the 12-week treatment phase of the study. Fifteen patients discontinued treatment; the most common reason for discontinuation was the occurrence of treatment-emergent AEs (TEAEs). Three patients in the etrasimod 1-mg group and 4 patients in the etrasimod 2-mg group discontinued because of TEAEs (Table 2). The median duration of exposure to the study drug was 12.1 weeks in each treatment group (min–max, 0.3–17.7 across all groups).

Efficacy Primary Endpoint At week 12, the least squares mean (LSM) (standard error) improvement in the modified MCS was 1.94 (0.31), 2.49 (0.31), and 1.50 (0.30) points for the etrasimod 1 mg, etrasimod 2 mg, and placebo groups,

respectively (Figure 1A; Supplementary Table S1). A significant difference in the magnitude of improvement in the modified MCS was demonstrated at week 12 for the etrasimod 2-mg group compared with placebo (LSM difference, 0.99 [90% CI, 0.30–1.68]; P = .009). The LSM difference between etrasimod 1 mg and placebo was 0.43 (90% CI, −0.24 to 1.11; nominal P = .15) (Figure 1A). At final data analysis, among 156 patients in the intention-to-treat population, 14 patients (8.9%) had missing data for the primary endpoint of modified MCS; missing data were imputed using prespecified multiple imputation procedures and analysis methods. The results of sensitivity analyses support the findings for the primary efficacy endpoint of a statistically significant difference for change from baseline at week 12 in the modified MCS for the etrasimod 2-mg group (Supplementary Table S2). Secondary Endpoints At week 12, a significantly higher proportion of patients receiving etrasimod 2 mg achieved endoscopic improvement vs those receiving placebo (41.8% vs 17.8%; difference, 24.4% [90% CI, 9.8%–39.0%]; P = .003) (Figure 1B; Supplementary Table S1); 22.5% of patients receiving etrasimod 1 mg achieved endoscopic improvement (difference vs placebo, 4.1% [90% CI, −9.1% to 17.2%]; nominal P = .31). The etrasimod 2-mg group had significant improvement in the 2-component MCS vs placebo (LSM difference, 0.84 [90% CI, 0.36–1.32]; P = .002); the etrasimod 1-mg group had a difference vs placebo of 0.39 (90% CI, −0.08 to 0.85; nominal P = .09) (Figure 1C; Supplementary Table S1). Total MCS improved significantly in the etrasimod 2-mg group vs placebo (LSM difference, 1.27 [90% CI, 0.37– 2.17]; P = .010). The difference in the etrasimod 1-mg group vs placebo was 0.60 (90% CI, −0.27 to 1.48; nominal P = .13) (Figure 1D; Supplementary Table S1). Exploratory Endpoints

More patients receiving etrasimod 2 mg than those receiving placebo achieved clinical remission (33.0% vs 8.1%; nominal P < .001) (Figure 2A). Clinical response was achieved by 50.6% of patients treated with etrasimod 2 mg and 32.5% treated with placebo (nominal P = .03) (Figure 2B). Histological improvement was reported in 31.7% of patients receiving etrasimod 2 mg vs 10.2% in those receiving placebo (nominal P = .006) (Figure 2C). Histological remission occurred in 19.5% of patients treated with etrasimod 2 mg compared with 6.1% of patients receiving placebo (nominal P = .03) (Figure 2D). Absolute lymphocyte counts decreased in a dose-proportional manner starting at week 1 (Figure 3). At week 12, etrasimod 2 mg and etrasimod 1 mg resulted in decreases from baseline of 39.9% and 19.8%, respectively (nominal P < .001 for difference from placebo, for both doses). The results of all prespecified exploratory analyses are listed in Supplementary Table S3. Subgroup Analyses Overall, the results of subgroup analyses for improvement in modified MCS were similar to those for the primary analysis, including for patients with prior exposure to anti-TNFα therapy. The exceptions were prior vedolizumab use and a history of pancolitis (Supplementary Figure S3). Safety The incidences of study drug–related TEAEs and TEAEs leading to discontinuation were higher in the etrasimod groups than the placebo group (Table 2). Overall, 55.1% of patients reported 1 or more TEAEs during the study, and TEAEs that were deemed to be related to study drug were reported in 7.7% of patients. Ten serious TEAEs in 9 patients (5.8% of overall patients, 3 patients receiving etrasimod 1 mg and 6 patients receiving placebo) and 9 TEAEs leading to discontinuation of the study drug in 7 patients (4.5%; of overall patients, 3 patients receiving etrasimod 1 mg and 4 patients receiving etrasimod 2 mg) were reported. Across all treatment groups, most TEAEs (75%) were mild to moderate

(grades 1 or 2); no drug-related TEAEs were life-threatening, and there were no patient deaths during the study. The most commonly reported TEAEs in all groups included UC worsening, upper respiratory tract infection, nasopharyngitis, and anemia. No patients treated with etrasimod reported a grade 3 or above TEAE of infection; disorders of eye, blood (except anemia), hepatobiliary system, or lymphatic system; or lymphocyte count <0.2 x 109/L. Three patients reported TEAEs of special interest (all were asymptomatic, occurred and resolved on day 1, and did not recur with further dosing): heart rate lowering and second-degree atrioventricular block type 1 in 1 patient receiving etrasimod 2 mg that resolved by hour 8, and first-degree atrioventricular block in 2 patients receiving etrasimod 2 mg (1 reported as a TEAE of first-degree atrioventricular block and the other assessed as not clinically significant by the site Investigator). In 2 of the 3 patients, review of pre-dose electrocardiogram and/or Holter monitoring results found evidence of atrioventricular block prior to etrasimod exposure. Details of cardiac events are provided in the supplementary materials and all TEAEs in Supplementary Table S4.

Discussion In this phase 2 study, etrasimod 2 mg was effective as induction therapy in patients with moderately to severely active UC. Once-daily etrasimod 2 mg led to a significantly greater improvement in modified MCS, which includes stool frequency, rectal bleeding, and endoscopy findings, compared with placebo. Treatment also resulted in significantly greater improvements in 2-component and total MCS and a significantly higher proportion of patients achieving endoscopic improvement compared with placebo. Although this proof of concept phase 2 study was not powered to make formal statistical comparisons for exploratory endpoints, more patients treated with etrasimod 2 mg achieved clinical remission and clinical response compared with placebo, and etrasimod 2 mg met most exploratory efficacy endpoints,

including histological improvement and histological remission. Results with etrasimod 1 mg were greater than placebo for most outcome measures but had a nominal P > .05.

The primary endpoint was changed to improvement in modified MCS from the proportion of patients who achieved clinical remission to reduce the sample size needed per treatment group (~39 patients for modified MCS versus ~80 patients for clinical remission) to assess a treatment effect by etrasimod in this proof of concept phase 2 trial. The modified MCS is a continuous-variable outcome across 3 domains that was believed to be more sensitive to differences between etrasimod and placebo than the dichotomous outcome of clinical remission. The modified MCS includes the clinically relevant measures of endoscopic findings, rectal bleeding, and stool frequency, while omitting the subjective Physician Global Assessment, which has been shown to have only fair agreement with endoscopic findings.41 Etrasimod 2 mg met the primary endpoint of a significant difference in the magnitude of improvement in modified MCS from baseline compared with placebo (Figure 1A) as well as the secondary endpoint of endoscopic improvement (Figure 1B), a clinically important component of the modified MCS. The clinically important endpoint of the proportion of patients with clinical remission was included as an exploratory outcome in the current study, with 33% of patients receiving etrasimod 2 mg experiencing clinical remission at week 12, compared with 8.1% of patients receiving placebo (nominal P < .001; Figure 2).

Treatment with etrasimod appeared to be generally safe and well tolerated in the study population. Fewer than 10% of patients discontinued in both treatment arms, mostly because of TEAEs. Although TEAEs leading to drug discontinuation were higher with active treatment than placebo, most TEAEs were mild to moderate in severity and were similar to those reported for other next-generation S1P

receptor modulators.18,26 Cardiac events, including bradycardia, conduction abnormalities, or both, have been reported with S1P receptor modulators,16,17,21,25,42 usually occurring within hours of the first dose administration. S1P1, S1P2, and S1P3 receptor subtypes are expressed in cardiac tissue.7 In humans, activation of S1P1 on the cell surface of cardiomyocytes results in the activation of the G protein– coupled inwardly rectifying potassium channel, which has a role in cardiac conduction and heart rate.7,43 Hyperpolarization of myocytes and a temporary reduction in excitability caused by S1P receptor activation of G protein–coupled inwardly rectifying potassium channels before S1P antagonism or S1P receptor internalization may explain the transient bradycardia observed with members of the S1P receptor modulator class of therapeutic agents.7,43,44 For some S1P receptor modulators, the magnitude of first-dose cardiac effect necessitates titration of the study drug.20,21,24 Because etrasimod had a relatively modest, transient, and asymptomatic first-dose effect in phase 1 studies,37 no dose titration was used in the current study. Transient, asymptomatic, and low-grade (first-degree or second-degree type 1) atrioventricular block was seen in 3 patients receiving etrasimod 2 mg in the current study. In each instance, the change occurred on day 1, resolved spontaneously, and did not recur with further dosing. In 2 of the 3 patients, there was evidence of atrioventricular block prior to etrasimod exposure.

Targeted small-molecule S1P receptor modulators, such as etrasimod, represent a new class for the treatment of immune-mediated inflammatory diseases with several advantages over conventional treatments, particularly over treatment with biologic therapies, including oral delivery and a once-daily dosing regimen. Because etrasimod is a small molecule, no immunogenicity is expected. Etrasimod is a selective S1P1, S1P4, and S1P5 receptor modulator,34 and its specificity for these receptor subtypes may contribute to its safety profile. S1P1, S1P4, and S1P5 are primarily associated with immune system regulation, while functional abrogation of S1P2 and S1P3 may have adverse cardiovascular and

pulmonary effects.14,34 In addition to its receptor subtype selectivity, etrasimod has a rapid onset and offset of action. In the current study in patients with moderately to severely active UC, a mean 39% reduction in lymphocytes was observed 1 week after treatment with etrasimod 2 mg. This confirmed the results of a previous study with healthy adults that demonstrated the rapid onset and offset of action of etrasimod, with rapid depletion of circulating lymphocytes (~53% depletion at day 3) and rapid recovery to within 5% of baseline within a week of treatment cessation.36,37 In contrast, ozanimod, an S1P receptor agonist with an elimination half-life of 17–21 hours42 and long-acting active metabolites with half-lives of approximately 10 days,45 showed only ~15% reduction at day 3 and lymphocyte levels ~40% below baseline 2 weeks after the last dose in healthy adults.42 However, caution must be taken with cross-trial comparisons, and the true clinical significance of receptor specificity can only be determined by head-to-head comparisons and larger trials. The current study has some limitations. First, this induction study was only 12 weeks in duration, and the safety and efficacy during longer-term maintenance therapy remain to be elucidated. Second, a larger sample size is needed to fully determine the safety profile of etrasimod. In conclusion, etrasimod 2 mg was more effective than placebo for improving the modified MCS at week 12 in patients with moderately to severely active UC in this phase 2 trial. The safety and efficacy of etrasimod should be further characterized in phase 3 clinical studies.

Data Sharing Statement

The study protocol and data collected for the study, including individual patient data, will not be made available to others.

References 1.

Kornbluth A, Sachar DB. Ulcerative colitis practice guidelines in adults: American College of Gastroenterology, Practice Parameters Committee. Am J Gastroenterol 2010;105:501–523.

2.

Ungaro R, Mehandru S, Allen PB, et al. Ulcerative colitis. Lancet 2017;389:1756–1770.

3.

Olivera P, Danese S, Peyrin-Biroulet L. Next generation of small molecules in inflammatory bowel disease. Gut 2017;66:199–209.

4.

Kihara Y, Maceyka M, Spiegel S, et al. Lysophospholipid receptor nomenclature review: IUPHAR Review 8. Br J Pharmacol 2014;171:3575–3594.

5.

Perez-Jeldres T, Tyler CJ, Boyer JD, et al. Cell trafficking interference in inflammatory bowel disease: therapeutic interventions based on basic pathogenesis concepts. Inflamm Bowel Dis 2019;25:270–282.

6.

Blaho VA, Hla T. An update on the biology of sphingosine 1-phosphate receptors. J Lipid Res 2014;55:1596–1608.

7.

Camm J, Hla T, Bakshi R, et al. Cardiac and vascular effects of fingolimod: mechanistic basis and clinical implications. Am Heart J 2014;168:632–644.

8.

Blankenbach KV, Schwalm S, Pfeilschifter J, et al. Sphingosine-1-Phosphate Receptor-2 Antagonists: Therapeutic Potential and Potential Risks. Front Pharmacol 2016;7:167.

9.

Brinkmann V, Billich A, Baumruker T, et al. Fingolimod (FTY720): discovery and development of an oral drug to treat multiple sclerosis. Nat Rev Drug Discov 2010;9:883–897.

10.

Garris CS, Blaho VA, Hla T, et al. Sphingosine-1-phosphate receptor 1 signalling in T cells: trafficking and beyond. Immunology 2014;142:347–353.

11.

Arlt O, Schwiebs A, Japtok L, et al. Sphingosine-1-phosphate modulates dendritic cell function: focus on non-migratory effects in vitro and in vivo. Cell Physiol Biochem 2014;34:27–44.

12.

Walzer T, Chiossone L, Chaix J, et al. Natural killer cell trafficking in vivo requires a dedicated sphingosine 1-phosphate receptor. Nat Immunol 2007;8:1337–1344.

13.

Matloubian M, Lo CG, Cinamon G, et al. Lymphocyte egress from thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature 2004;427:355–360.

14.

Peyrin-Biroulet L, Christopher R, Behan D, et al. Modulation of sphingosine-1-phosphate in inflammatory bowel disease. Autoimmun Rev 2017;16:495–503.

15.

Xiong Y, Hla T. S1P control of endothelial integrity. In: Oldstone MBA, Rosen H, eds. Sphingosine-1-phosphate signaling in immunology and infectious diseases. Cham: Springer International Publishing, 2014:85–105.

16.

Cohen JA, Barkhof F, Comi G, et al. Oral fingolimod or intramuscular interferon for relapsing multiple sclerosis. N Engl J Med 2010;362:402–415.

17.

Kappos L, Radue EW, O'Connor P, et al. A placebo-controlled trial of oral fingolimod in relapsing multiple sclerosis. N Engl J Med 2010;362:387–401.

18.

Kappos L, Bar-Or A, Cree BAC, et al. Siponimod versus placebo in secondary progressive multiple sclerosis (EXPAND): a double-blind, randomised, phase 3 study. Lancet 2018;391:1263–1273.

19.

Gilenya [package insert]. East Hanover, NJ: Novartis Pharmaceuticals Corp; 2016.

20.

Mayzent [package insert]. East Hanover, NJ: Novartis Pharmaceuticals Corp; 2019.

21.

Sandborn WJ, Feagan BG, Wolf DC, et al. Ozanimod induction and maintenance treatment for ulcerative colitis. N Engl J Med 2016;374:1754–1762.

22.

Cohen JA, Comi G, Selmaj KW, et al. Safety and efficacy of ozanimod versus interferon beta-1a in relapsing multiple sclerosis (RADIANCE): a multicentre, randomised, 24-month, phase 3 trial. Lancet Neurol September 3, 2019 [Epub ahead of print].

23.

Comi G, Kappos L, Selmaj KW, et al. Safety and efficacy of ozanimod versus interferon beta-1a in relapsing multiple sclerosis (SUNBEAM): a multicentre, randomised, minimum 12-month, phase 3 trial. Lancet Neurol September 3, 2019 [Epub ahead of print].

24.

Olsson T, Boster A, Fernandez O, et al. Oral ponesimod in relapsing-remitting multiple sclerosis: a randomised phase II trial. J Neurol Neurosurg Psychiatry 2014;85:1198–1208.

25.

Vaclavkova A, Chimenti S, Arenberger P, et al. Oral ponesimod in patients with chronic plaque psoriasis: a randomised, double-blind, placebo-controlled phase 2 trial. Lancet 2014;384:2036– 2045.

26.

Kappos L, Arnold DL, Bar-Or A, et al. Safety and efficacy of amiselimod in relapsing multiple sclerosis (MOMENTUM): a randomised, double-blind, placebo-controlled phase 2 trial. Lancet Neurol 2016;15:1148–1159.

27.

Kappos L, Arnold DL, Bar-Or A, et al. Two-year results from a phase 2 extension study of oral amiselimod in relapsing multiple sclerosis. Mult Scler 2018;24:1605–1616.

28.

D'Haens GR, Danese S, Hibi T, et al. A controlled trial of amiselimod, a selective S1P receptor modulator in Crohn's disease. Gastroenterology 2019;156:S-217.

29.

Calabresi PA, Radue EW, Goodin D, et al. Safety and efficacy of fingolimod in patients with relapsing-remitting multiple sclerosis (FREEDOMS II): a double-blind, randomised, placebocontrolled, phase 3 trial. Lancet Neurol 2014;13:545–556.

30.

Jain N, Bhatti MT. Fingolimod-associated macular edema: incidence, detection, and management. Neurology 2012;78:672–680.

31.

Pul R, Osmanovic A, Schmalstieg H, et al. Fingolimod associated bilateral cystoid macular edema-wait and see? Int J Mol Sci 2016;17:E2106.

32.

Berger JR, Cree BA, Greenberg B, et al. Progressive multifocal leukoencephalopathy after

fingolimod treatment. Neurology 2018;90:e1815–e1821. 33.

Berger JR. Classifying PML risk with disease modifying therapies. Mult Scler Relat Disord 2017;12:59–63.

34.

Al-Shamma H, Lehmann-Bruinsma K, Carroll C, et al. The selective sphingosine 1-phosphate receptor modulator etrasimod regulates lymphocyte trafficking and alleviates experimental colitis. J Pharmacol Exp Ther 2019;369:311–317.

35.

Buzard DJ, Kim SH, Lopez L, et al. Discovery of APD334: design of a clinical stage functional antagonist of the sphingosine-1-phosphate-1 receptor. ACS Med Chem Lett 2014;5:1313–1317.

36.

Schreiber S, Morgan M, Christopher R, et al. Etrasimod (APD334), a potent, selective, oral S1P receptor modulator with preclinical autoimmune disease-modifying activity exhibits favorable PK/PD properties in healthy volunteers. Advances in Inflammatory Bowel Diseases (AIBD) 2016; December 8–10, 2016; Orlando, FL.

37.

Peyrin-Biroulet L, Adams J, Turner S, et al. Safety and immune modulatory properties of etrasimod (APD334), a next-generation oral, selective sphingosine 1-phosphate receptor (S1PR) modulator, in healthy volunteers. 13th Congress of the European Crohn’s and Colitis Organisation (ECCO); February 14–17, 2018; Vienna, Austria.

38.

Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. N Engl J Med 1987;317:1625–1629.

39.

Geboes K, Riddell R, Ost A, et al. A reproducible grading scale for histological assessment of inflammation in ulcerative colitis. Gut 2000;47:404–409.

40.

Fleming TR, Richardson BA. Some design issues in trials of microbicides for the prevention of HIV infection. J Infect Dis 2004;190:666–674.

41.

Regueiro M, Rodemann J, Kip KE, et al. Physician assessment of ulcerative colitis activity

correlates poorly with endoscopic disease activity. Inflamm Bowel Dis 2011;17:1008–1014. 42.

Tran JQ, Hartung JP, Peach RJ, et al. Results from the first-in-human study with ozanimod, a novel, selective sphingosine-1-phosphate receptor modulator. J Clin Pharmacol 2017;57:988– 996.

43.

Cannavo A, Liccardo D, Komici K, et al. Sphingosine kinases and sphingosine 1-phosphate receptors: signaling and actions in the cardiovascular system. Front Pharmacol 2017;8:556.

44.

Gergely P, Nuesslein-Hildesheim B, Guerini D, et al. The selective sphingosine 1-phosphate receptor modulator BAF312 redirects lymphocyte distribution and has species-specific effects on heart rate. Br J Pharmacol 2012;167:1035–1047.

45.

Tran JQ, Zhang P, Surapaneni S, Selkirk J, Yan G, Pamlisano M. Absorption, metabolism, and excretion, in vitro pharmacology, and clinical pharmacokinetics of ozanimod, a novel sphingosine 1-phosphate receptor agonist. 35th Congress of the European Committee for Treatment and Research in Multiple Sclerosis; September 11–13, 2019; Stockholm, Sweden.

Figure Legends Figure 1. Primary and secondary efficacy endpoints at week 12 (intention-to-treat population). (A) LSM improvement from baseline in the modified MCS, which includes stool frequency, rectal bleeding, and endoscopy findings (primary endpoint). (B) The proportion of patients who achieved endoscopic improvement, defined as a score of ≤1 point on the Mayo Clinic endoscopic subscore (secondary outcome). (C) LSM improvement from baseline in the 2-component MCS, including rectal bleeding and endoscopy findings (secondary outcome). (D) LSM improvement from baseline in total MCS (secondary outcome). Values in parentheses indicate 90% confidence intervals. LSM, least squares mean; MCS, Mayo Clinic score.

Figure 2. Selected exploratory efficacy outcomes at week 12. (A) The proportion of patients who achieved clinical remission, defined as having an endoscopic subscore ≤1 (with absence of friability), a rectal bleeding score ≤1, and a stool frequency score ≤1 with a frequency decrease of ≥1 point from baseline (intention-to-treat population). (B) The proportion of patients who achieved clinical response, defined as having met the criteria for clinical remission or having a decrease in the modified MCS of ≥2 points and a decrease of ≥30% with either a decrease in rectal bleeding of ≥1 or a rectal bleeding score of ≤1 (intention-to-treat population). (C) The proportion of patients with histologic improvement, defined as having a Geboes score <3.1 (modified intention-to-treat population). (D) The proportion of patients with histologic remission, defined as having a Geboes score <2.0 (modified intention-to-treat population). Values in parentheses indicate 90% confidence intervals. MCS, Mayo Clinic score.

Figure 3. Lymphocyte count over time (modified intention-to-treat population). Decrease in absolute circulating lymphocyte count from baseline to week 12. Percent change from baseline (∆) is based on a

mixed-effects model for repeated measures with current oral corticosteroid use, prior exposure to antitumor necrosis factor alpha agents, treatment, week, treatment-by-week interaction as factors, and baseline value as covariates. P values are for the comparison with placebo in mean change from baseline. The error bars indicate standard error of the mean. The asterisks indicate nominal P < .001 vs placebo. LLN, lower limit of normal.

Tables Table 1. Patient Demographics and Baseline Characteristics (Safety Population) Placebo

Etrasimod 1 mg

Etrasimod 2 mg

(n = 54)

(n = 52)

(n = 50)

44.8 (14.85)

43.2 (12.22)

40.4 (12.39)

32 (59.3)

30 (57.7)

27 (54.0)

51 (94.4)

48 (92.3)a

49 (98.0)

25.8 (4.6)

24.9 (3.6)

24.0 (5.5)

25.2 (18.2–41.1)

24.2 (17.9–32.2)

23.2 (17.7–46.6)

Mean (SD)

8.7 (1.72)b

8.8 (1.43)

8.9 (1.47)

Median (min–max)

9.0 (4–12)

9.0 (6–12)

9.0 (6–11)

Mean (SD)

6.5 (1.51)b

6.5 (1.23)

6.6 (1.17)

Median (min–max)

7.0 (2–9)c

7.0 (4–9)

7.0 (4–8)

Mean (SD)

8.6 (7.16)

7.0 (6.11)

6.2 (4.69)

Median (min–max)

6.0 (1–32)

5.3 (1–22)

4.8 (1–18)

Proctosigmoiditis

34 (63.0)

34 (65.4)

30 (60.0)

Pancolitis

23 (42.6)

20 (38.5)

14 (28.0)

Age, years Mean (SD) Sex, n (%) Male Race, n (%) White BMI, kg/m2 Mean (SD) Median (min–max) Baseline total MCS

Baseline modified MCS

Duration of UC, years

History of disease extent, n (%)d

Baseline fecal calprotectin, µg/g

Mean (SD) Median (min–max)

2350.8 (3121.69)

3051.7 (4977.91)e

2491.8 (3679.85)f

1429.0 (67–16,226)

1210.0 (30–24,190)

1449.0 (71–21,559)

79.30 (120.66)b

142.98 (345.97)

92.39 (139.10)d

34.57 (2.95–562.87)

46.19 (2.00–2333.38)

46.86 (2.76–732.40)

125.5 (33.47)b

122.0 (33.75)

116.7 (32.78)g

127.0 (53–185)

121.0 (56–187)

116.5 (58–208)

16 (29.6)

13 (25.0)

18 (36.0)

31.4 (1.1–1304.4)

26.1 (1.6–260.9)

25.7 (2.1–432.4)

18 (33.3)

15 (28.8)

17 (34.0)

41.4 (8.7–524.4)

99.1 (28.3–321.9)

54.4 (5.9–228.9)

33 (61.1)

17 (32.7)

26 (52.0)

57.3 (1.1–652.0)

104.4 (2.6–355.6)

52.6 (4.0–417.4)

12 (22.2)

4 (7.7)

7 (14.0)

26.3 (2.1–240.0)

54.1 (31.6–75.9)

43.9 (16.1–80.7)

53 (98.1)

49 (94.2)

46 (92.0)

146.3 (19.3–1464.6)

118.9 (25.3–1081.0)

112.9 (17.3–792.3)

Baseline CRP, nmol/L Mean (SD) Median (min–max) Baseline IBDQ total score Mean (SD) Median (min–max)

Prior and concomitant treatments

h

Current oral corticosteroids, n (%) Duration, median (min–max), weeks Previous anti-TNFα agents, n (%) Duration, median (min–max), weeks Previous immunosuppressants, n (%) Duration, median (min–max), weeks Previous anti-integrin agent, n (%) Duration, median (min–max), weeks Previous oral 5-aminosalicylates, n (%) Duration, median (min–max), weeks

BMI, body mass index; CRP, C-reactive protein; IBDQ, Inflammatory Bowel Disease Questionnaire; max, maximum; MCS, Mayo Clinic score; min, minimum; SD, standard deviation; TNFα, tumor necrosis factor alpha; UC, ulcerative colitis. a

One patient reported race of both white and Asian–Japanese.

b

n = 53.

c

One patient was inadvertently enrolled with baseline modified MCS of 2 (endoscopic score of 2, stool

frequency subscore of 0, rectal bleeding subscore of 0). This patient was randomized and received at least 1 dose of study medication before the discrepancy was noted, and the patient was withdrawn from the study at day 15. d

For history of proctosigmoiditis and pancolitis, the responses are not mutually exclusive. Some patients

reported a history of both proctosigmoiditis and pancolitis: etrasimod 1 mg, n = 11; etrasimod 2 mg, n = 9; placebo, n = 8. e

n = 49.

f

n = 45.

g

n = 48.

h

Treatment duration is the total number of weeks on treatment for each individual using imputed partial

dates. Duration of treatment summary includes prior and concomitant treatments.

Table 2. Treatment-emergent Adverse Events (Safety Population)a Placebo

Etrasimod 1 mg Etrasimod 2 mg

(n = 54)

(n = 52)

(n = 50)

27 (50.0)

31 (59.6)

28 (56.0)

Number of TEAEs, n

64

66

78

Patients with TEAEs leading to death, n

0

0

0

Patients discontinued due to ≥1 TEAE, n (%) [no. of events]

0

3 (5.8) [5]

4 (8.0) [4]

Ulcerative colitis—worsening

0

2 (3.8)

2 (4.0)

Anal abscess

0

1 (1.9)

0

Electrocardiogram T-wave abnormal

0

0

1 (2.0)

Dehydration

0

1 (1.9)

0

Hypokalemia

0

1 (1.9)

0

Neurological symptom

0

0

1 (2.0)

6 (11.1) [7]

3 (5.8) [3]

0

5 (9.3)

2 (3.8)

0

Abdominal pain

1 (1.9)

0

0

Ulcerative colitis—worsening

3 (5.6)

2 (3.8)

0

Duodenal ulcer perforation

1 (1.9)

0

0

Hepatobiliary disorders, n (%)

1 (1.9)b

0

0

Infections and infestations, n (%)

0

1 (1.9) c

0

Renal and urinary disorders, n (%)

1 (1.9)d

0

0

Grade 1—mild

19 (35.2) [32]

21 (40.4) [41]

23 (46.0) [53]

Grade 2—moderate

14 (25.9) [26]

18 (34.6) [23]

14 (28.0) [23]

5 (9.3) [5]

2 (3.8) [2]

2 (4.0) [2]

Patients with any TEAE, n (%)

Patients with serious TEAEs, n (%) [no. of events] Gastrointestinal disorders, n (%)

Severity (all TEAEs), n (%) [no. of events]

Grade 3—severe

Grade 4—life-threatening

1 (1.9) [1]

0

0

0

0

0

Grade 1—mild

2 (3.7) [3]

2 (3.8) [3]

2 (4.0) [2]

Grade 2—moderate

3 (5.6) [3]

2 (3.8) [2]

4 (8.0) [5]

Grade 3—severe

0

0

0

Grade 4—life-threatening

0

0

0

Grade 5—death related to TEAE

0

0

0

27 (50.0) [57]

30 (57.7) [61]

26 (52.0) [71]

3 (5.6) [6]

4 (7.7) [5]

5 (10.0) [7]

Atrioventricular block second degree (type 1)

0

0

1 (2.0)f

Heart rate lowering

0

0

1 (2.0)f

Ulcerative colitis—worsening

4 (7.4)

5 (9.6)

2 (4.0)

Upper respiratory tract infection

2 (3.7)

4 (7.7)

2 (4.0)

Nasopharyngitis

4 (7.4)

2 (3.8)

1 (2.0)

Anemiag

2 (3.7)

2 (3.8)

3 (6.0)

0

2 (3.8)

2 (4.0)

Abdominal pain

2 (3.7)

2 (3.8)

0

Headache

1 (1.9)

0

3 (6.0)

Nausea

2 (3.7)

1 (1.9)

1 (2.0)

Fecal calprotectin increased

2 (3.7)

1 (1.9)

1 (2.0)

Grade 5—death related to TEAE Severity (treatment-related TEAEs), n (%) [no. of events]

TEAE relation to study drug, n (%) [no. of events]¶ Not related Related Treatment-related TEAEs of special interest, n (%)e

TEAEs reported by ≥2 patients in any treatment group, n (%)

Urinary tract infection

Arthralgia

2 (3.7)

1 (1.9)

1 (2.0)

Blood creatine phosphokinase increased

0

1 (1.9)

2 (4.0)

Gamma-glutamyltransferase increased

0

1 (1.9)

2 (4.0)

Sinusitis

0

1 (1.9)

2 (4.0)

Fever

1 (1.9)

0

2 (4.0)

Anal fissure

2 (3.7)

0

0

Hyperlipasemia

0

0

2 (4.0)

Neutrophil count increasedh

0

2 (3.8)

0

AEs, adverse events; n, number of patients with observations; TEAE, treatment-emergent adverse event. a

TEAEs were defined as AEs with an onset date on or after the first dose of study medication. Events

were coded using the Medical Dictionary for Regulatory Activities, version 20.1. b

Jaundice.

c

Anal abscess.

d

e

Hydronephrosis.

Treatment-related TEAEs included events categorized by the investigator as related. If relatedness was

missing, the event was assumed to be related. f

One patient experienced 1 event each of atrioventricular block second degree (type 1) and heart rate

lowering that were considered treatment related. One additional patient experienced 1 event each of atrioventricular block first degree and sinus bradycardia that were not considered treatment related. See Supplementary Materials for additional details. g

Anemia was based on investigator diagnosis.

h

Neutrophil count increase was based on investigator assessment of central laboratory values.

Disposition of patients. The number (%) of patients in the MITT and completers populations is for the primary endpoint. AE, adverse event; ITT, intention to treat; MITT, modified intention to treat.

Supplemental Material This supplemental material has been provided to give readers additional information about the study. Table of Contents 1. List of Principal Investigators Participating in the Study 2. Supplementary Methods a. Inclusion Criteria b. Exclusion Criteria c. Selection of Doses d. Schedule of Procedures and Visits 3. Adverse Events of Special Interest 4. Supplementary Tables and Figures Table S1. Summary Results of the Primary and Sensitivity Analyses Results of the Primary Efficacy Endpoint (Etrasimod 2 mg vs Placebo) Table S2. Results of Exploratory Efficacy Endpoints at Week 12 Table S3. All Treatment-Emergent Adverse Events Figure S1. Trial Design Figure S2. Disposition of Patients Figure S3. Subgroup Analyses 5. References

1

1.

Principal Investigators Participating in the Study

Investigator

City

Austria Harald Vogelsang

Vienna

Belgium Elisabeth Macken Severine Vermeire

Edegem Leuven

Bulgaria Asen Petrov Plamen Petrov Simeon Stoinov Galina Vasileva Borislav Vladimirov

Sofia Sofia Sofia Ruse Sofia

Canada Charles Bernstein Hughie Fraser A. Hillary Steinhart

Winnipeg Bridgewater Toronto

Germany Tanja Kühbacher Ingolf Schiefke Stefan Schreiber

Hamburg Leipzig Kiel

Hungary László Bene Tibor Gyökeres Gyula G. Kiss Árpád Patai Márta Varga

Budapest Budapest Debrecen Szombathely Békéscsaba

Isreal Doron Schwartz

Beer Sheva

Latvia Aldis Pukitis

Riga

Lithuania Edita Kazenaite

Vilnius

Poland Marcin Hańczewski Marek Horyński Malgorzata Kondusz-Szklarz Krzysztof Niezgoda Karolina Radwan Janusz Rudzinski Dariusz Solowiej Piotr Walczak

Poznan Sopot Wroclaw Elblag Rzeszów Bydgoszcz Kielce Kraków

 

2

Republic of Korea Byung Ik Jang Hyun-Soo Kim Yoon Jae Kim

Daegu Wonju-si Incheon

Romania Eugen Adrian Goldis Radu Bogdan Mateescu Lucian Negreanu Carol Stanciu

Timisoara Bucharest Bucharest Iasi

Russian Federation Olga Alexeeva Alexander Gordienko Elena Malova Alexey Nizov Andrey Obrezan Marina Osipenko Marina Pesegova Vasiliy Trofimov Aleksey Yakovlev

Nizhny Novgorod Saint Petersburg Samara Ryazan Saint Petersburg Novosibirsk Krasnoyarsk Saint Petersburg Rostov-on-Don

Spain Federico Argüelles Arias Julian Panes Estela Fernandez Salgado Miquel Sans

Sevilla Barcelona Pontevedra Barcelona

Ukraine Oleksii Datsenko Andriy Dorofeyev Galyna Fadieienko Oleksandr Fediv Oleksandr Golovchenko Olena Levchenko Inna Markevych Sergiy Prokopchuk Yaroslava Rishko Mykola Stanislavchuk Roman Yatsyshyn

Kharkiv Kiev Kharkiv Chernivtsi Vinnytsia Odessa Kiev Kiev Uzhhorod Vinnytsia Ivano-Frankivsk

United Kingdom Matthew Brookes Richard Johnston James Lindsay

Wolverhampton Torquay London

United States Asad Sheikh Aziz Leonard Baidoo Haresh Boghara Michael Chiorean Mehmet Emin Donat Ana M. Elosegui Alister George Eugene Greenberg

Hoffman Estates, IL Chicago, IL DeSoto, TX Seattle, WA Troy, MI Sweetwater, FL Thousand Oaks, CA Urbana, IL

3

Ammar Hemaidan Samuel Idarraga Christopher Michael Johnson John Francis Kuemmerle, Jr. Mark Lamet Scott D. Lee John Earl Lowe Talha Malik Travis James Rutland Lawrence Saubermann Javier Sobrado Dario Sorrentino George Whitfield Bruce R. Yacyshyn Ziad Younes

Port Orange, FL Wauwatosa, WI Temple, TX Richmond, VA Hollywood, FL Seattle, WA Ogden, UT Birmingham, AL Dothan, AL Rochester, NY Miami, FL Roanoke, VA Hermitage, TN Cincinnati, OH Germantown, TN

4

2.

Supplementary Methods

a. Inclusion Criteria Patients were required to meet all of the following inclusion criteria to be enrolled in the study: 1. Males or females aged 18 to 80 years, inclusive 2. Had to be able to give signed informed consent 3. Had to be willing and able to comply with the study requirements 4. Had to be considered in stable health in the opinion of the investigator, as determined by:  A prestudy physical examination with no clinically significant abnormalities unrelated to ulcerative colitis (UC)  Vital signs at screening: pulse rate ≥55 beats per minute, systolic blood pressure ≥90 mm Hg, and diastolic blood pressure ≥ 55 mm Hg  Liver function tests (alanine aminotransferase/aspartate aminotransferase, bilirubin, and alkaline phosphatase) <2× upper limit of normal  All other prestudy clinical laboratory findings within normal range, or if outside of the normal range were not deemed clinically significant in the opinion of the investigator  12-lead electrocardiogram (ECG) that showed no clinically significant abnormalities (exclusion criterion #24)  A chest X-ray showed no evidence of active pulmonary disease. Note: A chest X-ray taken within the previous 12 months from the screening visit could also have been used  Ophthalmology evaluation (by an ophthalmologist) without evidence of macular edema, supported with optical coherence tomography where available (dependent on-site capability) 5. Had a diagnosis of UC established ≥6 months before screening by clinical and endoscopic evidence and corroborated by histopathology report 6. Had moderately to severely active UC defined as a three-component Mayo Clinic score (MCS) of 4 to 9 that included an endoscopic subscore of ≥ 2 and a rectal bleeding score of ≥ 1 (using 3 of the 4 components of the total MCS [endoscopic findings, rectal bleeding, and stool frequency]). These values were to be obtained from patient diary entries of rectal bleeding and stool frequency within 10 days before randomization and flexible proctosigmoidoscopy results as determined by a blinded central reader within 10 days before randomization 7. Had evidence of colonic UC activity on endoscopy (i.e., UC extending ≥ 15 cm proximal to the rectum) 8. For patients having a history of extensive colitis or pancolitis of > 8 years’ duration or left sided colitis of >12 years’ duration, documented evidence that a surveillance colonoscopy was performed within 12 months of the initial screening visit (if not, the patient had to undergo a colonoscopy in lieu of a flexible proctosigmoidoscopy during screening) 9. Had demonstrated over the previous 5-year period an inadequate response to, loss of response to, or intolerance of ≥ 1 of the following agents as defined below:  5-aminosalicylic acids (5-ASAs) (e.g., mesalamine)  Corticosteroids – Signs and symptoms of persistently active disease despite a history of at least one 4-week induction regimen that included a dose equivalent to prednisone 30 mg daily, or – Two failed attempts to taper corticosteroids to below a dose equivalent to prednisone 10 mg daily, or – History of intolerance of corticosteroids (including but not limited to Cushing syndrome, osteopenia/osteoporosis, hyperglycemia, insomnia, and infection).  Immunosuppressives – Signs and symptoms of persistently active disease despite a history of at least one 8-week regimen of oral azathioprine (≥1.5 mg/kg) or 6 mercaptopurine (≥0.75 mg/kg), or – History of intolerance to ≥1 of these immunosuppressives (including but not limited to nausea/vomiting, abdominal pain, pancreatitis, liver function test abnormalities, lymphopenia, TPMT genetic mutation, or infection)  Anti-tumor necrosis factor alpha (TNFα) agents – Signs and symptoms of persistently active disease despite a history of completing an induction regimen with ≥1 of the following: infliximab, adalimumab, or golimumab at doses per the current labeling and/or institutional standard of care, or – Recurrence of symptoms during maintenance dosing with infliximab, adalimumab, or golimumab following prior clinical benefit (discontinuation despite clinical benefit did not qualify), or

5

History of intolerance to infliximab, adalimumab, or golimumab (including but not limited to infusion- or injection-related reaction, demyelination, congestive heart failure, or infection)  Anti-integrin agents – Recurrence of symptoms during maintenance dosing with vedolizumab following prior clinical benefit (discontinuation despite clinical benefit did qualify), or – History of intolerance to vedolizumab (including but not limited to infusion-related reaction) 10. Could have been receiving a therapeutic dose of the following drugs:  Oral 5-ASA compounds, provided that the dose had been stable for 2 weeks immediately before randomization  Oral corticosteroid therapy (prednisone at a stable dose of ≤10 mg/day, budesonide at a stable dose of ≤9 mg/day, or equivalent steroid), provided that the dose had been stable for 4 weeks immediately before screening endoscopy assessment, if corticosteroids had recently been initiated  Azathioprine or 6-mercaptopurine, provided that the dose had been stable for 8 weeks immediately before screening. Note: These immunosuppressive agents had to be discontinued at the time of randomization  Probiotics (e.g., Culturelle®, Saccharomyces boulardii), provided that the dose had been stable for 2 weeks immediately before randomization  Antidiarrheals (e.g., loperamide, diphenoxylate with atropine) for control of chronic diarrhea 11. Eligible female patients had to be:  Nonpregnant, evidenced by a negative serum human chorionic gonadotropin pregnancy test at screening and a urine dipstick pregnancy test at day 1  Nonlactating  Sexually abstinent, surgically sterile, postmenopausal, or agreed to continue to use an accepted method of birth control during and for ≥30 days after the last study drug administration. Acceptable methods of birth control were the following: – Hormonal contraceptives (patients had to be consistently taking the hormonal contraceptives for ≥3 months [90 days] before screening) – Double-barrier method (condom or occlusive cap [diaphragm or cervical cap] with spermicide) – An intrauterine device – Surgical sterility for at least 6 months before screening for tubal ligation performed laparoscopically, hysterectomy, and/or bilateral oophorectomy, and/or postmenopausal (defined as ≥2 years without menses) Note: Contraceptive measures such as Plan B (used after unprotected sex) were not considered to be acceptable methods of contraception for this study 12. Eligible male patients had to be either:  Surgically sterile (i.e., vasectomy), for ≥3 months (90 days) before screening  Agreed to use a condom with spermicide when sexually active with a female partner who was not using an acceptable method of birth control during the study and for 30 days after last study drug administration 13. Eligible male and female subjects had to agree not to participate in a conception process (i.e., active attempt to become pregnant or to impregnate, sperm donation, in vitro fertilization) for 30 days after the last dose of study drug –

b. Exclusion Criteria Patients were not enrolled in the study if they met any of the following exclusion criteria: 1. Had evidence of abdominal abscess or toxic megacolon at the initial screening visit 2. Had previous extensive colonic resection (subtotal or total colectomy) 3. Had underwent ileostomy, colostomy, or had known fixed symptomatic stenosis of the intestine 4. Had received any of the following therapies within 30 days before randomization for the treatment of underlying disease:  Nonbiologic therapies (e.g., cyclosporine, tacrolimus, tofacitinib, thalidomide) other than those specifically accepted (inclusion criterion #10).  A nonbiologic investigational therapy.  An approved nonbiologic therapy in an investigational protocol. 5. Had received any of the following agents within 60 days before randomization:  Infliximab, adalimumab, golimumab, certolizumab, or vedolizumab.  Any other investigational or approved biologic agent.

6

6. Had any prior exposure to natalizumab, efalizumab, or rituximab 7. Had received previous treatment with more than three biologic agents 8. Had received topical (rectal) treatment with 5-ASA or corticosteroid enemas/suppositories within 2 weeks of administration of the first dose of study drug 9. Had evidence of or had received treatment for Clostridium difficile infection within 60 days or other intestinal pathogen within 30 days before randomization 10. Had currently required or anticipated requiring surgical intervention for UC during the study 11. Had current evidence of adenomatous colonic polyps that were not removed 12. Had current evidence of colonic mucosal dysplasia 13. Had a diagnosis of Crohn’s colitis or indeterminate colitis 14. Had an infection with hepatitis B or C virus 15. Had active or latent tuberculosis (TB), regardless of treatment history, as evidenced by any of the following:  History of TB (that had not been acceptably treated)  A positive diagnostic TB test within 1 month of randomization defined as: – A positive QuantiFERON® (QFT) test or two successive indeterminate QFT tests, or – A tuberculin skin test reaction ≥10 mm (≥5 mm in patients receiving the equivalent of prednisone >15 mg/day)  Chest X-ray within 12 months of randomization in which active or latent pulmonary TB could not be excluded 16. Had any known history of congenital or acquired immunodeficiency (e.g., common variable immunodeficiency, human immunodeficiency virus infection [enzyme linked immunosorbent assay and western blot] test result, organ transplantation) 17. Had clinically significant extraintestinal infection (e.g., pneumonia, pyelonephritis) within 30 days before randomization 18. Had previously participated in any study of etrasimod 19. Had a history of any clinically significant medical condition that, in the investigator's opinion, would have precluded participation in the study 20. Had a recent history (within 6 months of screening visit) of cardiovascular or cerebrovascular disease, acute coronary syndrome, myocardial infarction, unstable angina, cerebrovascular accident, or transient ischemic attack at screening 21. Had underwent any surgical procedure that required general anaesthesia within 30 days before randomization or planned to undergo major surgery during the study period 22. Had a history of retinal macular edema 23. Had a history of or signs and symptoms of progressive multifocal leukoencephalopathy (PML) as assessed by a PML checklist 24. Had a history of cardiac arrhythmia, conduction system disease (including atrioventricular [AV] node dysfunction, second- or third-degree heart block, and sick sinus syndrome), or use of Class Ia and Class III anti arrhythmic agents, or baseline-corrected QT interval ≥500 ms 25. Had a forced expiratory volume in the first second or forced vital capacity of <80% of predicted values (i.e., abnormal) 26. Had an infection that required hospitalization or intravenous antimicrobial therapy, or opportunistic infection within 4 weeks of screening 27. Had a history of ≥1 episode of herpes zoster or any episode of disseminated zoster 28. Were without documented positive varicella zoster virus (VZV) immunoglobulin G antibody status or had completed VZV vaccination within 30 days before randomization 29. Had received a live vaccine within 4 weeks before screening 30. Had a history of lymphoproliferative disorder, lymphoma, leukemia, myeloproliferative disorder, or multiple myeloma 31. Had a history of malignancy, except for adequately treated basal cell skin cancer 32. Had received any investigational therapy, excluded medications (as defined by exclusion criteria #6 and #7), or any approved therapy in an investigational protocol within 30 days before screening 33. Had a history of severe allergic or anaphylactic reactions that required medical attention 34. Had current or recent history (within 1 year prior to randomization) of alcohol dependence or illicit drug use 35. Had a history of clinically significant leukopenia or lymphopenia at screening 36. Had active psychiatric problems that, in the investigator’s opinion, could have interfered with compliance with the

7

study procedures 37. Had been using moderate to strong inhibitors of cytochrome P450 (CYP)2C9 38. Had a history of severe renal impairment 39. Had a history of severe hepatic impairment 40. Could not have attended all the study visits or comply with study procedures 41. Had a history of primary nonresponse to a treatment regimen of vedolizumab per the current labeling and/or institutional standard of care. This exclusion did not apply to patients who were intolerant of vedolizumab or who developed a secondary loss of response c.

Selection of Doses

The etrasimod doses (1 mg and 2 mg) used in this study were selected on the basis of results of a single ascending dose study and a multiple ascending dose study conducted in healthy adults. In the single ascending dose study, etrasimod exposure was dose proportional for doses ranging from 0.1 to 5 mg. Single etrasimod doses of 3 mg and 5 mg induced a dose-responsive decline in the absolute number of peripheral blood B cells, T cells, natural killer cells, and T-cell subsets, except effector memory T cells. The maximum tolerated dose in the study was etrasimod 3 mg. In the multiple ascending dose study, a dose-dependent effect on lymphocyte lowering was observed, plateauing at 2 mg with daily dosing despite increased exposure with 3 mg. Additionally, the heart rate effects for initiating at doses ≤2 mg without titration occurred within the first week, with much of the effect occurring with the first dose on day 1, and attenuation from day 4 through the end of the study, whereas titration resulted in further heart rate lowering at the time of dose increase. On the basis of these results from the single and multiple ascending dose studies, etrasimod doses of 1 and 2 mg without titration were selected for this study.

8

d.

Schedule of Procedures and Visits

Screening Period Evaluation

Informed consent Inclusion/exclusion criteria Medical and social historyc UC history Physical and neurologic examinatione Serology screen (HIV, HBsAg, HCV, VZV IgG antibody) Chest X-rayg PFT PML monitoring TB screening Urine drug screeni Pregnancy testj Genomic DNAk CBC with differential and plateletsl CRP Ophthalmoscopy with OCTn Laboratory testsp Stool sampleq Fecal calprotectin Flexible proctosigmoidoscopyr Biopsyt TMSu PRO2v Diary instructionw Diary reviewx IBDQz

Days −28 to −1 X X X X Xf

Days −10 to −1b

Day −1

2-week Follow-up Visita

Induction Treatment Week 0 (Day 1)

Week 1

Week 2

Week 4

Week 8

Week 12

Xd Xd

X

X

X X X X X X X

X

X

X

X

Xh X

X

X

X X X X X X X Xs X X X

X X X X X

X X

X X

X X

X X

X

X

X

X X X

X X X

X X X X

X

X

X

X

X

X

Xy

X

X

X

X

X X X X X X

X X

9

X X

Xh X

Xm X Xo X

PK assessmentsaa X X X X X X Randomization X Vital signsbb X X X X X X X X Holter monitor application X Holter monitoringcc X 12-lead ECGdd X X X X X X X Study drug administrationee X X X X X X Drug dispensation/accountability X X X X X X AE assessment X X X X X X X X X Concomitant medications X X X X X X X X X Concomitant procedures X X X X X X X X X AE, adverse event; CBC, complete blood count; CRP, C-reactive protein; DNA, deoxyribonucleic acid; ECG, electrocardiogram; HBsAg, hepatitis B virus surface antigen; HCV, hepatitis C virus; HIV, human immunodeficiency virus; IBDQ, Inflammatory Bowel Disease Questionnaire; IgG, immunoglobin G; MCS, Mayo Clinic score; OCT, optical coherence tomography; PFT, pulmonary function test; PK, pharmacokinetic(s); PML, progressive multifocal leukoencephalopathy; PRO2, two item patient reported outcome; TB, tuberculosis; TMS, total Mayo Clinic score; UC, ulcerative colitis; VZV, varicella zoster virus. a. For subjects not enrolling in the extension Study APD334-005 at week 14 b. Within screening period, but had to occur within 10 days prior to randomization c. Included prior therapies and tobacco, caffeine, and alcohol use d. Partial examination, to update findings from the examination performed at screening e. Complete physical examination (including assessments of the skin, head, eyes, ears, nose, throat, neck, thyroid, lungs, heart, abdomen, back, lymph nodes, extremities, and body weight) and neurologic examination (including assessments of the cranial nerves, motor and sensory function, coordination, and mental status) f. Height was measured at screening only g. A chest X-ray taken within the previous 12 months from the screening visit could also be used h. Because forced vital capacity and forced expiratory volume in the first second are highly effort-dependent, abnormal values or significant changes from baseline or the previous measurement had to be verified with a repeat assessment i. Included amphetamines, barbiturates, cocaine metabolites, opiates, benzodiazepines, and cannabinoids j. Only for female subjects who were not diagnosed as postmenopausal. Serum human chorionic gonadotropin test was performed at screening and week 12 visit; urine pregnancy test on day 1, and at weeks 4 and 8 k. Optional blood sample (4 mL) for DNA testing l. Samples for CBC with differential, platelet count, and lymphocyte counts were obtained prior to dosing m. Required if CBC with differential, platelet count and lymphocyte counts were abnormal at week 12 visit n. Optical coherence tomography was to be performed by the study sites that had the capabilities. Retinal photos were to be taken at each ophthalmoscopy o. To be conducted only if abnormal results were assessed at week 12 p. Clinical laboratory tests included haematology (including coagulation), serum chemistry, urinalysis, and CRP and had to be completed predose at the scheduled study visits q. Stool sample was for both fecal calprotectin (screening, Weeks 4, 8, and 12) and culture, ova and parasite evaluation for C. difficile assay at screening and at

10

any point in the study when a subject became symptomatic, including worsening or return of disease activity r. Read by a blinded central reader who determined subject eligibility to participate in current study s. Endoscopy results were read by blinded central reader. However, the local read by investigator at the study site determined subject eligibility for participation in the extension study t. Biopsy was assessed by Geboes Index for assessment of UC histologic disease activity u. TMS (Physician Global Assessment, endoscopy findings, rectal bleeding, and stool frequency) was calculated at screening and Week 12 visits; entry criteria were based on the 3-component MCS (endoscopy findings, rectal bleeding, and stool frequency) as defined in inclusion criteria. The MCS was evaluated during screening using subject diary entries and flexible proctosigmoidoscopy results within 10 days prior to randomization. The MCS at week 12 was calculated using the week 12 proctosigmoidoscopy, and stool frequency and rectal bleeding scores had to be completed by the subject 7 days prior to visit v. PRO2 (based on stool frequency and rectal bleeding patient-reported outcomes collected daily using electronic patient diaries) excluded the findings on endoscopy and the Physician Global Assessment. The PRO2 was derived at screening, week 0 (Day 1), weeks 1, 2, 4, 8, and 12 w. Subject began diary entries (stool frequency and rectal bleeding) beginning the first day of screening after diary training was completed x. To capture daily stool frequency and rectal bleeding; the 2 patient-reported outcome measures of the MCS y. Had to be completed prior to dosing z. Inflammatory Bowel Disease Questionnaire (32 questions) aa. Pharmacokinetic blood samples were collected predose, and at 2- and 6-hour postdose on week 0 (Day 1), and predose at weeks 1, 2, 4, 8, and 12 bb. Included blood pressure, pulse, respiratory rate, and oral temperature captured predose. On Week 0 (Day 1), heart rate and blood pressure were captured at baseline (predose) and hourly through 6 hours postdose cc. Twenty-four hours predose through 24 hours postdose of day 1 dd. Safety ECG measurements were to be completed at screening and weeks 0, 1, 2, 4, 8 and 12. On week 0 (Day 1), ECG at baseline (predose) and hourly through 6 hours postdose; predose only at weeks 1, 2, 4, 8, 12 ee. Study drug had to be taken on an empty stomach (after ~8-hour overnight fast) and food had to be avoided for ~1-hour postdose. On days with scheduled study visits, subjects were not to take their dose of study drug at home in order to complete predose study procedures

11

3.

Adverse Events of Special Interest

One patient receiving etrasimod 2 mg had heart rate lowering and second-degree atrioventricular (AV) block type 1. The patient, who had a predose electrocardiogram (ECG) showing sinus rhythm, a baseline PR interval of 200 ms, and poor precordial R-wave progression, had evidence of first-degree AV block at 1 hour post dose (heart rate 53 beats per minute). At 2 hours post dose, the patient had evidence of second-degree AV block type 1 (ventricular rate 36 beats per minute), which continued at hour 6 (ventricular rate of 49 beats per minute). At hour 3, the patient received a single dose of atropine 0.5 mg, and by hour 8 the second-degree AV block had resolved with evidence of first-degree AV block (59 beats per minute). Drug intake was interrupted after the first dose but was restarted on day 4, and the patient completed the study without a further episode of AV block. One patient receiving etrasimod 2 mg had sinus bradycardia and first-degree AV block. The patient had a PR interval of 215 ms on the screening ECG, and the predose ECG showed first-degree AV block with PR intervals >220 ms. The predose heart rate (58–61 beats per minute) slowed to 48 beats per minute at 2 and 4 hours postdose and to 45 beats per minute at hour 5. At hour 6, heart rate increased to 53 beats per minute, and the sinus bradycardia was considered to be resolved. Another patient in the etrasimod group had first-degree AV block at hour 2 that was not considered to be clinically significant and had resolved by hour 4.

12

Supplementary Tables and Figures

Table S1. Results of Primary and Secondary Efficacy Endpoints at Week 12 (ITT Population, MI Method)

Modified MCSa,b LSM (SE) improvement from baseline 90% CI One-sided P value LSM (SE) difference from placebo 90% CI One-sided P value Endoscopic improvementc,d Patients with endoscopic improvement, % MH estimate for difference from placebo, % (SE) 90% CI for difference from placebo One-sided P value 2-component MCSa,e LSM (SE) improvement from baseline LSM (SE) difference from placebo 90% CI One-sided P value Total MCSa,f

Placebo n = 54

Etrasimod 1 mg n = 52

Etrasimod 2 mg n = 50

1.50 (0.30) 1.01 to 2.00 <0.0001 — — —

1.94 (0.31) 1.42 to 2.45 <0.0001 0.43 (0.41) −0.24 to +1.11 0.146

2.49 (0.31) 1.98 to 3.01 <0.0001 0.99 (0.42) 0.30 to 1.68 0.009

17.8 — — —

22.5 4.1 (7.98) –9.1 to +17.2 0.306

41.8 24.4 (8.87) 9.8 to 39.0 0.003

0.92 (0.21) — — —

1.30 (0.22) 0.39 (0.28) –0.08 to +0.85 0.086

1.75 (0.22) 0.84 (0.29) 0.36 to 1.32 0.002

LSM (SE) improvement from baseline 2.08 (0.39) 2.69 (0.41) 3.35 (0.41) LSM (SE) difference from placebo — 0.60 (0.53) 1.27 (0.55) 90% CI — –0.27 to +1.48 0.37 to 2.17 One-sided P value — 0.128 0.010 CI, confidence interval; LSM, least squares mean; ITT, intention to treat; MCS, Mayo Clinic score; MH, Mantel–Haenszel; MI, multiple imputation; SE, standard error. a Analysis of covariance model with terms for treatment, current oral corticosteroid use, prior exposure to anti-tumor necrosis factor alpha agents, and baseline value as covariates. b Modified MCS ranged from 0 to 9, with larger numbers indicating greater disease severity and included subscores for stool frequency, rectal bleeding, and endoscopy findings. The modified MCS (primary outcome) ranged from 0 to 9, where higher scores indicated greater disease severity. c Mantel–Haenszel method, adjusted for stratification factors of current corticosteroid therapy at baseline and previous exposure to anti-TNFα agents. d Endoscopic improvement was defined as a score of ≤1 point on the Mayo Clinic endoscopic subscore (secondary outcome). e The 2-component MCS, which ranged from 0 to 6 (higher scores indicated greater disease severity), included rectal bleeding and endoscopy findings (secondary outcome). f Total MCS ranged from 0 to 12 (higher scores indicated greater disease severity) and included stool frequency, rectal bleeding, endoscopy findings, and Physician Global Assessment (secondary outcome).

13

Table S2. Summary of the Primary and Sensitivity Analyses Results of the Primary Efficacy Endpoint (Etrasimod 2 mg vs Placebo)a

Modified MCS LSM (SE) improvement at week 12 Etrasimod 2 mg Placebo Difference from placebo, etrasimod 2 mg

Primary Analysis ITT Population (Multiple Imputation)

Sensitivity Analysis MITT Population

Completers Population

ITT Population (Worst Case Imputation)

(etrasimod 2 mg: n = 50; placebo: n 54)

(etrasimod 2 mg: n = 44; placebo: n = 48)

(etrasimod 2 mg: n = 44; placebo: n = 46)

(etrasimod 2 mg: n = 50; placebo: n = 53)

2.49 (0.31) 1.50 (0.30)

2.54 (0.33) 1.51 (0.31)

2.52 (0.33) 1.56 (0.32)

2.10 (0.35) 1.10 (0.34)

LSM (SE) 0.99 (0.42) 1.04 (0.44) 0.96 (0.44) 1.00 (0.47) 90% CI 0.30–1.68 0.31–1.76 0.24–1.69 0.23–1.78 One-sided P value .009 .010 .015 .017 CI, confidence interval; LSM, least squares mean; ITT, intention to treat; MCS, Mayo Clinic score; MITT, modified intention to treat; SE, standard error. a Analysis of covariance model with terms for treatment, current oral corticosteroid use, prior exposure to anti-tumor necrosis factor alpha agents, and baseline value as covariates. Modified MCS ranged from 0 to 9, with larger numbers indicating greater disease severity and included subscores for stool frequency, rectal bleeding, and endoscopy findings.

14

Table S3. Results of Exploratory Efficacy Endpoints at Week 12a

Clinical remission (ITT, MI method) N Patients with clinical remission, % MH estimate for difference, % (SE) 90% CI for difference from placebo One-sided P value Clinical response (ITT, MI method) N Patients with clinical response, % MH estimate for difference, % (SE) 90% CI for difference from placebo One-sided P value MCS stool frequency subscore (MITT, MMRM method) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LSM difference from placebo One-sided P value MCS rectal bleeding subscore (MITT, MMRM method) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LSM difference from placebo One-sided P value MCS endoscopic subscore (ITT, MI method) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LSM difference from placebo One-sided P value Physician Global Assessment (ITT, MI method) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LSM difference from placebo One-sided P value Partial MCS (ITT, MI method) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LSM difference from placebo One-sided P value 2-component MCS, including stool frequency and rectal bleeding (MITT, MMRM method) N LSM (SE) change from baseline LSM (SE) difference from placebo

Placebo

Etrasimod 1 mg

Etrasimod 2 mg

54 8.1 — — —

52 16.0 7.1 (6.46) −3.5 to +17.7 .14

50 33.0 25.8 (7.47) 13.5 to 38.1 <.001

54 32.5 — — —

52 43.7 11.4 (10.14) −5.3 to +28.1 .13

50 50.6 18.9 (9.92) 2.6 to 35.3 .03

49 −0.61 (0.13) — — —

49 −0.66 (0.13) −0.05 (0.18) −0.36 to +0.26 .39

45 −0.74 (0.14) −0.13 (0.19) −0.45 to +0.18 .24

49 −0.63 (0.12) — — —

49 −0.89 (0.12) −0.25 (0.16) −0.53 to +0.02 .06

45 −0.97 (0.12) −0.33 (0.17) −0.61 to −0.05 .03

54 −0.32 (0.13) — — —

52 −0.43 (0.14) −0.11 (0.18) −0.41 to +0.18 .26

50 −0.78 (0.14) −0.46 (0.18) −0.76 to −0.16 .006

54 −0.56 (0.12) — — —

52 −0.72 (0.13) −0.15 (0.17) −0.43 to −0.12 .18

50 −0.83 (0.13) −0.26 (0.17) −0.55 to +0.02 .07

54 −1.77 (0.31) — — —

52 −2.25 (0.32) −0.47 (0.42) −1.16 to +0.21 .13

50 −2.57 (0.32) −0.79 (0.43) −1.50 to −0.09 .03

49 −1.25 (0.21) —

49 −1.56 (0.22) −0.31 (0.30)

45 −1.70 (0.22) −0.44 (0.30)

15

90% CI for LSM difference from placebo One-sided P value Improvement in rectal bleeding (MITT) N Patients with improvement in rectal bleeding, % MH estimate for difference, % 90% CI for difference from placebo One-sided P value Histologic improvement (MITT) N Patients with histologic improvement, % MH estimate for difference, % 90% CI for difference from placebo One-sided P value Histologic remission (MITT) N Patients with histologic remission, % MH estimate for difference, % 90% CI for difference from placebo One-sided P value Endoscopic remission (ITT, MI method) N Mean (SE) responder rate at week 12 Log odds (SE) Odds ratio (90% CI) vs placebo One-sided P value Improvement in absolute stool frequency (MITT) N Patients with improvement in absolute stool frequency, % MH estimate for difference, % 90% CI for difference from placebo One-sided P value Improvement in MCS stool frequency sub-score (MITT) N Patients with improvement in MCS stool frequency subscore, % MH estimate for difference, % 90% CI for difference from placebo One-sided P value IBDQ total score (MITT) N LSM (SE) change from baseline LSM (SE) difference from placebo 90% CI for LS mean difference from placebo One-sided P value Trichotomous composite clinical remission/response score (ITT, MI method) N Mean (SE) composite score at week 12 Clinical response or remission odds (score of 1 or

— —

−0.80 to −0.19 .15

−0.95 to +0.06 .07

49 49.0 — — —

49 65.3 15.89 −0.33 to +32.11 .05

45 73.3 24.91 8.44 to 41.39 .008

49 10.2 — — —

49 20.4 9.94 −1.83 to +21.71 .09

41 31.7 21.24 7.53 to 34.95 .006

49 6.1 — — —

49 10.2 3.42 −5.68 to +12.52 .27

41 19.5 13.34 1.93 to 24.76 .03

54 0.053 (0.034) −2.88 (0.68) — —

52 0.137 (0.049) −1.80 (0.43) 2.95 (0.80–10.85) .09

50 0.153 (0.054) −1.69 (0.42) 3.27 (0.90–11.94) .07

49 75.5

49 71.4

45 84.4

— — —

−6.37 −20.77 to +8.02 .77

8.74 −4.68 to +22.16 .15

49 49.0

49 49.0

45 55.6

— — —

−1.8 −17.79 to +14.20 .57

5.61 −10.79 to +22.01 .29

50 21.11 (4.59) — — —

51 25.03 (4.71) 3.92 (6.24) −6.42 to +14.25 .27

47 24.53 (4.70) 3.42 (6.37) −7.13 to +13.97 .30

54 0.41 (0.09) 0.39

52 0.60 (0.11) 0.63

50 0.84 (0.13) 1.08

16

2) Odds ratio (90% CI) vs placebo



1.61 (0.83–3.14) .12

2.78 (1.40–5.51) .007

One-sided P value — Fecal calprotectin (µg/g) (MITT, MMRM method) N 47 42 37 LSM (SE) change from baseline at week 12 156.57 (505.76) 311.67 (542.51) −1301.91 (565.26) LSM (SE) difference from placebo — 155.09 (728.76) −1458.48 (752.99) 90% CI for LSM difference from placebo — −1051.95 to +1362.14 −2705.65 to −211.30 One-sided P value .58 .03 C-reactive protein (nmol/L) (MITT, MMRM method) N 47 46 42 LSM (SE) change from baseline at week 12 19.84 (17.66) −20.89 (18.16) −24.93 (18.38) LSM (SE) difference from placebo — −40.72 (24.65) −44.78 (25.08) 90% CI for LSM difference from placebo — −81.53 to +0.08 −86.29 to −3.27 One-sided P value — .05 .04 Lymphocyte counts (x109/L) (MITT, MMRM method) N 43 40 40 LSM (SE) change from baseline at week 12 0.129 (0.075) −0.556 (0.078) −0.775 (0.078) LSM (SE) difference from placebo — −0.685 (0.106) −0.904 (0.106) 90% CI for LSM difference from placebo — −0.860 to −0.510 −1.079 to −0.728 One-sided P value — <.001 <.001 Lymphocyte counts percentage (MITT, MMRM method) N 43 40 40 LSM (SE) change from baseline at week 12 17.347 (5.304) −19.836 (5.523) −39.927 (5.481) LSM (SE) difference from placebo — −37.183 (7.475) −57.275 (7.516) 90% CI for LSM difference from placebo — −49.557 to −24.810 −69.715 to −44.835 One-sided P value — < .001 < .001 CI, confidence interval; IBDQ, Inflammatory Bowel Disease Questionnaire; ITT, intention to treat; LSM, least squares mean; MCS, Mayo Clinic score; MH, Mantel–Haenszel; MI, multiple imputation; MITT, modified intention to treat; MMRM, mixed-effects model for repeated measures; SE, standard error. a The difference from placebo for score measures was estimated using an analysis of covariance model, proportion measures were estimated using an MH model, and longitudinal results of biomarkers were estimated using an MMRM model. All methods were adjusted for current oral corticosteroid therapy at baseline and prior exposure to anti-tumor necrosis factor alpha agents. All P values reported are nominal. Endpoints were based on Mayo Clinic subscores1 (each ranging from 0–3), histology scoring using the Geboes Index,2 or the 32-item inflammatory bowel disease questionnaire3 with higher numbers indicating greater disease severity for all measures, and are defined as follows: clinical remission (an endoscopic score ≤1 [with absence of friability], a rectal bleeding score ≤1, and a stool frequency score ≤1 with a decrease ≥1 point from baseline); clinical response (met criteria for clinical remission or had a decrease in MCS score of ≥2 points and a decrease of ≥30% with either a decrease of rectal bleeding ≥1 or rectal bleeding score ≤1); partial MCS (a 3-component MCS including rectal bleeding, stool frequency, and Physician Global Assessment subscores); histologic improvement (Geboes score <3.1); histologic remission (Geboes score <2.0); endoscopic remission (Mayo Clinic endoscopy score of 0); improvement in stool frequency (change from baseline <0); and trichotomous composite score of clinical remission/response (score ranging from 0 to 2 [score 2 for achieving both clinical remission and clinical response, 1 for achieving only clinical response, and 0 for achieving neither]).

17

Table S4. Treatment-emergent Adverse Events (Safety Population)a n (%) [no. of events]b Total TEAEs Patients with ≥1 TEAE Infections and Infestations Anal abscess Bronchitis Enterobiasis Gastroenteritis viral Hordeolum Laryngitis Nasopharyngitis Pneumonia Respiratory tract infection Sinusitis Tooth abscess Upper respiratory tract infection Urinary tract infection Gastrointestinal disorders Abdominal pain Anal fissure Aphthous ulcer Colitis ulcerative Diarrhea Dry mouth Duodenal ulcer perforation Duodenitis Flatulence Gastroesophageal reflux disease Hemorrhoids Large intestine polyp Nausea Toothache Vomiting Investigations Basophil count increased Blood alkaline phosphatase increased Blood creatine phosphokinase-MB increased Blood creatine phosphokinase increased Blood glucose increased Blood pressure increased Chest x-ray abnormal Clostridium test positive Electrocardiogram T wave abnormal Fecal calprotectin increased Gamma glutamyl transferase increased Hemoglobin decreased

Placebo (n = 54) 64 27 (50.0) 6 (11.1) 0 0 0 0 0 0 4 (7.4) 0 0 0 0 2 (3.7) 0 14 (25.9) [18] 2 (3.7) 2 (3.7) 1 (1.9) 4 (7.4) 0 0 1 (1.9) 1 (1.9) 1 (1.9) 1 (1.9) 1 (1.9) 1 (1.9) 2 (3.7) 1 (1.9) 0 5 (9.3) 0 1 (1.9)

Etrasimod 1 mg (n = 52) 66 31 (59.6) 12 (23.1) [13] 1 (1.9) 0 0 1 (1.9) 1 (1.9) 0 2 (3.8) 0 0 1 (1.9) 1 (1.9) 4 (7.7) 2 (3.8) 8 (15.4) [10] 2 (3.8) [3] 0 0 5 (9.6) 0 1 (1.9) 0 0 0 0 0 0 1 (1.9) 0 0 8 (15.4) [15] 1 (1.9) 0

Etrasimod 2 mg (n = 50) 78 28 (56.0) 11 (22.0) [15] 0 1 (2.0) 1 (2.0) 1 (2.0) 0 1 (2.0) 1 (2.0) [2] 1 (2.0) 1 (2.0) 2 (4.0) 0 2 (4.0) 2 (4.0) [3] 5 (10) [7] 0 0 0 2 (4.0) 1 (2.0) 0 0 0 1 (2.0) 0 0 0 1 (2.0) 1 (2.0) 1 (2.0) 8 (16.0) [10] 0 0

1 (1.9)

1 (1.9) [2]

0

0

1 (1.9)

2 (4.0)

0 0 0 0 0 2 (3.7) 0

1 (1.9) 1 (1.9) 0 1 (1.9) 0 1 (1.9) [2] 1 (1.9)

0 0 1 (2.0) 0 1 (2.0) 1 (2.0) 2 (4.0)

1 (1.9)

1 (1.9)

1 (2.0)

18

Neutrophil count increasedc Platelet count increased Protein urine present Weight decreased Musculoskeletal and connective tissue disorders Arthralgia Back pain Flank pain Musculoskeletal pain Myalgia Osteochondrosis Pain in extremity Spinal osteoarthritis Spinal pain Tendon pain Nervous system disorders Autonomic nervous system imbalance Dizziness Dysgeusia Headache Hypoesthesia Migraine Neurological symptom Somnolence Tremor General disorders and administration site conditions Axillary pain Chest discomfort Fatigue Influenza-like illness Edema peripheral Pyrexia Metabolism and nutrition disorders Decreased appetite Dehydration Diabetes mellitus Hyperamylasemia Hyperglycemia Hyperlipasemia Hypokalemia Iron deficiency Vitamin D deficiency Blood and lymphatic system disorders Anemiad Thrombocytosis Vascular disorders Hot flush Hypertension Peripheral venous disease Phlebitis

0 0 0 0 5 (9.3) [6]

2 (3.8) 1 (1.9) 0 1 (1.9) 4 (7.7) [8]

0 1 (2.0) 1 (2.0) 0 3 (6.0)

2 (3.7) [3] 0 0 0 1 (1.9) 0 0 0 1 (1.9) 1 (1.9) 2 (3.7) [3] 0

1 (1.9) 1 (1.9) 1 (1.9) 0 0 1 (1.9) 1 (1.9) 1 (1.9) [3] 0 0 4 (7.7) 0

1 (2.0) 0 0 1 (2.0) 0 0 1 (2.0) 0 0 0 6 (12.0) [7] 1 (2.0)

1 (1.9) 0 1 (1.9) [2] 0 0 0 0 0 3 (5.6) [5]

1 (1.9) 0 0 1 (1.9) 1 (1.9) 0 0 1 (1.9) 3 (5.8)

0 1 (2.0) 3 (6.0) 0 0 1 (2.0) 1 (2.0) 0 3 (6.0) [4]

1 (1.9) 0 1 (1.9) 1 (1.9) 1 (1.9) 1 (1.9) 2 (3.7) 2 (3.7) 0 0 0 0 0 0 1 (1.9) 1 (1.9) 2 (3.7) [4] 2 (3.7) [3] 1 (1.9) 3 (5.6) 0 1 (1.9) 1 (1.9) 1 (1.9)

0 0 1 (1.9) 1 (1.9) 1 (1.9) 0 3 (5.8) [4] 1 (1.9) 1 (1.9) 1 (1.9) 0 0 0 1 (1.9) 0 0 2 (3.8) [3] 2 (3.8) [3] 0 1 (1.9) 0 1 (1.9) 0 0

0 1 (2.0) 1 (2.0) 0 0 2 (4.0) 3 (6.0) [5] 0 0 0 1 (2.0) 1 (2.0) [2] 2 (4.0) 0 0 0 3 (6.0) 3 (6.0) 0 2 (4.0) 1 (2.0) 1 (2.0) 0 0

19

Eye disorders Blepharitis Diplopia Pigment dispersion syndrome Vision blurred Vitreous detachment Psychiatric disorders Anxiety Depression Insomnia Mental disorder Cardiac disorders Atrioventricular block second degree Bradycardia Palpitations Sinus bradycardia Ventricular extrasystoles Renal and urinary disorders Dysuria Hydronephrosis Oliguria Proteinuria Renal colic Ureteric obstruction Skin and subcutaneous tissue disorders Acne Dermatitis Rash Urticaria Reproductive system and breast disorders Dysmenorrhea Vaginal hemorrhage Respiratory, thoracic, and mediastinal disorders Bronchiectasis Dyspnea Oropharyngeal pain Sinus congestion Upper-airway cough syndrome Wheezing Ear and labyrinth disorders Tinnitus Hepatobiliary disorders Jaundice Injury, poisoning, and procedural complications Procedural pain

1 (1.9) 1 (1.9) 0 0 0 0 1 (1.9) 0 0 1 (1.9) 0 0 0

1 (1.9) 0 0 0 0 1 (1.9) 0 0 0 0 0 1 (1.9) 0

3 (6.0) 0 1 (2.0) 1 (2.0) 1 (2.0) 0 4 (8.0) 1 (2.0) 1 (2.0) 1 (2.0) 1 (2.0) 3 (6.0) [4] 1 (2.0) [2]

0 0 0 0 1 (1.9) [2] 0 1 (1.9) 0 0 0 1 (1.9) 2 (3.7) [4] 1 (1.9) [3] 1 (1.9) 0 0 0

0 1 (1.9) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 (3.8)

1 (2.0) 0 1 (2.0) 1 (2.0) 3 (6.0) [4] 1 (2.0) 0 1 (2.0) 1 (2.0) 1 (2.0) 0 2 (4.0) 0 0 1 (2.0) 1 (2.0) 0

0 0 1 (1.9)

1 (1.9) 1 (1.9) 0

0 0 1 (2.0) [5]

0 0 0 1 (1.9) 0 0 0 0 1 (1.9) [2] 1 (1.9) [2] 1 (1.9)

0 0 0 0 0 0 1 (1.9) 1 (1.9) 0 0 0

1 (2.0) 1 (2.0) 1 (2.0) 0 1 (2.0) 1 (2.0) 0 0 0 0 0

1 (1.9)

0

0

AEs, adverse events; TEAEs, treatment-emergent adverse events. Events were coded using the Medical Dictionary for Regulatory Activities, version 20.1. a TEAEs were defined as AEs with an onset date on or after the first dose of study medication.

20

b

At each level of patient summarization, a patient was counted once if the patient reported one or more events. Unless otherwise indicated, the number of events = the number of patients. c Neutrophil count increase was based on investigator assessment of central laboratory values. d Anemia was based on investigator diagnosis.

21

Supplementary Figure Legends Figure S1. Trial design. Patients were randomly assigned 1:1:1 to receive oral etrasimod 1 mg, etrasimod 2 mg, or placebo once daily. Following a 28-day screening period, patients entered the 12-week treatment period, after which they could be eligible to continue directly into a 34-week extension study. Patients who were not eligible or did not join the extension study had a further safety follow-up visit at week 14 or 2 weeks after last study visit. QD, once daily. Figure S2. Disposition of patients. The number (%) of patients in the MITT and completers populations is for the primary endpoint. AE, adverse event; ITT, intention to treat; MITT, modified intention to treat. Figure S3. Subgroup analyses (intention-to-treat population). Subgroup analyses of placebo-adjusted change (90% confidence interval) at week 12 in the etrasimod 2 mg group. The improvement in the modified MCS (A and C) and the proportion of patients achieving clinical remission (B and D) are shown for baseline disease characteristics (A and B) and prior or concurrent therapies (C and D). Missing data were imputed using the multiple imputation method. Least squares mean differences were estimated using an analysis of covariance model that included treatment group, subgroup factor, and interaction term of treatment group and subgroup factor. Values in parentheses are the sample sizes for etrasimod 2 mg, placebo. CRP, C-reactive protein; CS, corticosteroid; IS, immunosuppressant; MCS, Mayo Clinic score; TNFα, tumor necrosis factor alpha; UC, ulcerative colitis.

22

4.

References

1. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. N Engl J Med 1987;317:1625–1629. 2. Geboes K, Riddell R, Ost A, Jensfelt B, Persson T, Lofberg R. A reproducible grading scale for histological assessment of inflammation in ulcerative colitis. Gut 2000;47:404–409. 3. Guyatt G, Mitchell A, Irvine EJ, et al. A new measure of health status for clinical trials in inflammatory bowel disease. Gastroenterology 1989;96:804–810.

23