Multiple Sclerosis and Related Disorders 51 (2021) 102844 Contents lists available at ScienceDirect Multiple Sclerosis and Related Disorders journal homepage: www.elsevier.com/locate/msard Original article Ozanimod in relapsing multiple sclerosis: Pooled safety results from the clinical development program Krzysztof W. Selmaj a, *, Jeffrey A Cohen b, Giancarlo Comi c, Amit Bar-Or d, Douglas L Arnold e, Lawrence Steinman f, Hans Peter Hartung g, h, i, Xavier Montalban j, Eva Kubala Havrdova k, Bruce A.C. Cree l, Neil Minton m, James K Sheffield m, Ning Ding m, Ludwig Kappos n a Center for Neurology, Tylna 12, 90-324 Łódź, Poland and Collegium Medicum, Department of Neurology, University of Warmia and Mazury, Warszawska 30, 11-082 Olsztyn, Poland b Department of Neurology, Mellen Center for MS Treatment and Research, Neurological Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio, 44195, USA c Department of Neurology, San Raffaele Scientific Institute, Vita-Salute San Raffaele University, via Olgettina 48, 20132 Milan, Italy d Center for Neuroinflammation and Experimental Therapeutics, and Department of Neurology, Perelman School of Medicine, University of Pennsylvania, 3400 Spruce Street - 3 Dulles Building, Philadelphia, Pennsylvania, 19104, USA e NeuroRx Research and Montréal Neurological Institute, McGill University, 3801 University Street, Montreal, Quebec, H3A 2B4,Canada f Department of Neurology and Neurological Sciences, Beckman Center for Molecular Medicine, Stanford University Medical Center, 300 Pasteur Dr Stanford, Stanford, California 94305, USA g Department of Neurology, Medical Faculty, Heinrich-Heine University, University Hospital Dusseldorf, Moorenstr. 5 40225 Düsseldorf, Germany h Brain and Mind Centre, University of Sydney, Camperdown, NSW 2006 Sydney, Australia i Department of Neurology, Medical University Vienna, Spitalgasse 23, 1090 Vienna, Austria j Department of Neurology-Neuroimmunology, Centre d’Esclerosi Múltiple de Catalunya (Cemcat), Hospital Universitario Vall d’Hebron, Pg. Vall d’Hebron, 119-129, 08035 Barcelona, Spain k Department of Neurology and Center for Clinical Neuroscience, First Medical Faculty, Charles University, Katerinska 30, 120 00 Prague 2 Czech Republic l Weill Institute for Neurosciences, Department of Neurology, UCSF University of California San Francisco, 675 Nelson Rising Lane San Francisco, California 94158, USA m Bristol Myers Squibb, 3401 Princeton Pike, Lawrenceville, New Jersey 08648, USA n Research Center for Clinical Neuroimmunology and Neuroscience Basel (RC2NB), Departments of Medicine, Clinical Research, Biomedicine, and Biomedical Engineering, University Hospital and University of Basel, Petersgraben 4, CH-4031 Basel, Switzerland A R T I C L E I N F O A B S T R A C T Keywords: Background: Ozanimod, an oral sphingosine 1-phosphate receptor 1 and 5 modulator, is approved in multiple Multiple sclerosis countries for the treatment of relapsing multiple sclerosis (RMS). In phase 3 trials, ozanimod was well tolerated Ozanimod and superior to interferon beta-1a 30 µg once-weekly in reducing clinical and radiologic disease activity. The Safety objective of this integrated safety analysis was to evaluate the safety of extended ozanimod exposure in par­ Adverse events ticipants with RMS from all clinical trials and compare it with phase 3 trial data. Clinical trials Methods: We report pooled incidence and study duration‒adjusted incidence rates (IR) of treatment-emergent adverse events (TEAEs) from an interim data cut (January 31, 2019) of RMS participants treated with ozani­ mod. Data were pooled from a phase 1 pharmacokinetic/pharmacodynamic trial, a placebo-controlled phase 2 trial with dose-blinded extension, 2 large active-controlled phase 3 trials, and an open-label extension (OLE). Results were compared with pooled phase 3 trial data. Results: At the data cutoff, 2631 RMS participants had exposure to ozanimod 0.92 mg (mean 32.0 months) and 2787 had exposure to either ozanimod 0.46 or 0.92 mg (mean 37.1 months). The IRs per 1000 person-years (PY) for any TEAE (772.2) and serious TEAEs (33.2) in the overall population were similar to those in the phase 3 Abbreviations: ALC, absolute lymphocyte count; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CSR, central serous choroidopathy; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; IFN, interferon; IR, incidence rates; MAO-A, monoamine oxidase A; MAO-B, monoamine oxidase B; ME, macular edema; MERP, Macular Edema Review Panel; OCT, optical coherence tomography; OLE, open-label extension; PFT, pulmonary function test; PML, pro­ gressive multifocal leukoencephalopathy; PY, person-years; S1P, sphingosine 1-phosphate; ULN, upper limit of normal. * Corresponding author. E-mail addresses: kselmaj@gmail.com (K.W. Selmaj), COHENJ@ccf.org (J.A. Cohen), comi.giancarlo@hsr.it (G. Comi), amitbar@pennmedicine.upenn.edu (A. Bar-Or), douglas.arnold@mcgill.ca (D.L. Arnold), steiny@stanford.edu (L. Steinman), hans-peter.hartung@uni-duesseldorf.de (H.P. Hartung), xavier. montalban@cem-cat.org (X. Montalban), ehavr@lf1.cuni.cz (E.K. Havrdova), bruce.cree@ucsf.edu (B.A.C. Cree), Neil.Minton@bms.com (N. Minton), James. Sheffield@bms.com (J.K. Sheffield), Ning.Ding@bms.com (N. Ding), ludwig.kappos@usb.ch (L. Kappos). https://doi.org/10.1016/j.msard.2021.102844 Received 23 November 2020; Received in revised form 29 January 2021; Accepted 12 February 2021 Available online 15 February 2021 2211-0348/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 population (896.1 and 31.2, respectively). There were no serious opportunistic infections. There were no second- degree or higher atrioventricular blocks on electrocardiogram. Hepatic enzyme elevations declined over time. Malignancy rates remained low with longer exposure. Pulmonary function tests showed minimal reductions in lung function. Seven ozanimod-treated participants with comorbid risk factors had confirmed macular edema, including 3 in the ongoing OLE. Conclusions: Safety results in this larger RMS population with greater ozanimod exposure demonstrated no new safety concerns and were consistent with phase 3 trial results. 1. Introduction randomized trials: an unpublished phase 1 pharmacokinetic/pharma­ codynamic study, a phase 2 study (Cohen et al., 2016) that included a Ozanimod is a sphingosine 1-phosphate (S1P) receptor 1 and 5 dose-blinded extension (Cohen et al., 2019a), and 2 similarly designed modulator (Scott et al., 2016) approved in multiple countries for the phase 3 trials of differing durations (Cohen et al., 2019b; Comi et al., treatment of relapsing multiple sclerosis (RMS) (Zeposia package insert, 2019) (Fig. 1). Upon completion of any of those trials, participants were 2020; Zeposia summary of product characteristics 2020). Ozanimod eligible to enroll in DAYBREAK, an ongoing open-label extension (OLE) blocks the capacity of lymphocytes to egress from lymphoid tissue, study. Institutional review boards approved the use of human subjects in reducing the number of lymphocytes in peripheral blood (Scott et al., each study included in this analysis, and all participants provided 2016). While the exact mechanism of ozanimod in RMS is unknown, it written informed consent. may involve reduced migration of lymphocytes into the central nervous All parent trials enrolled men and women 18 to 55 years of age with system. MS diagnosed by 2010 McDonald criteria who exhibited a relapsing In the phase 3 RADIANCE (Cohen et al., 2019b) and SUNBEAM clinical course, had a history of brain magnetic resonance imaging (MRI) (Comi et al., 2019) trials, oral ozanimod 0.92 mg (equivalent to ozani­ lesions consistent with MS, and an Expanded Disability Status Scale mod HCl 1 mg) or 0.46 mg (equivalent to ozanimod HCl 0.5 mg) once score of 0 to 5.0 (phase 2 and 3) or 0 to 6.0 (phase 1) (Cohen et al., 2016; daily was superior to intramuscular interferon (IFN) beta-1a 30 μg once Cohen et al., 2019b; Comi et al., 2019). This analysis pools weekly on measures of disease activity and in slowing brain volume loss. ozanimod-treated participants from all RMS trials, including the exten­ Ozanimod was generally well tolerated, with completion rates of sions, with a data cutoff of January 31, 2019. At data cutoff, only approximately 90% in the pivotal phase 3 trials. Both doses of ozanimod DAYBREAK was ongoing. demonstrated a favorable safety profile compared with IFN beta-1a, a well-established treatment for RMS. 2.2. Safety assessments and outcomes This integrated safety analysis of participants who received ozani­ mod in all RMS studies was undertaken to provide a comprehensive The safety assessment schedule is available in Table A.1. Treatment- perspective of the safety of ozanimod 0.92 mg and to characterize its emergent adverse events (TEAEs) were coded using MedDRA version safety profile in a large RMS population with a longer duration of 18.1. exposure than in the phase 3 trials. External experts in liver dysfunction examined overall hepatic safety and conducted in-depth reviews of specific cases of potential clinical 2. Methods significance (eg, concurrent elevations of aminotransferases and bili­ rubin). An independent expert Macular Edema Review Panel (MERP) 2.1. Ozanimod clinical trial study designs and RMS populations reviewed TEAEs related to macular edema (ME), cases of central foveal thickness >20% of baseline, and any relevant optical coherence to­ The ozanimod clinical development program in RMS included 4 mography (OCT) abnormalities. Fig. 1. RMS clinical development program for ozanimod: studies included in this analysis HCI: hydrochloride; IFN: interferon; IM: intramuscular; OLE: open-label extension; PO: per os (oral); QD: once daily. a In all trials, upon initiation of ozanimod, participants received 0.23 mg (equivalent to ozanimod HCl 0.25 mg) on days 1 to 4, 0.46 mg (equivalent to ozanimod HCl 0.5 mg) on days 5 to 7, and then their assigned dose of 0.46 mg or 0.92 mg (equivalent to ozanimod HCl 1 mg) on day 8 and thereafter. All participants entering the phase 2 dose-blinded extension period underwent dose escalation, even if treated with ozanimod in the parent trial, to maintain the blind. b In DAYBREAK, dose escalation was performed for all participants entering from one of the active-controlled phase 3 trials, irrespective of prior treatment assignment (to maintain the blinding in the parent trials); dose escalation was not performed for those entering from the phase 1 or 2 trials, unless the last dose of ozanimod was >14 days before entering DAYBREAK. 2 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Outcomes included incidence and study duration‒adjusted incidence 2.4. Data availability statement rates (IR) per 1000 person-years (PY) (100,000 PY for malignancies) of TEAEs, severe TEAEs, serious TEAEs, and TEAEs leading to discontinu­ Celgene, a Bristol Myers Squibb company, is committed to respon­ ation of study drug. We also report the incidence and study duration‒ sible and transparent sharing of clinical trial data with patients, adjusted IR of TEAEs of special interest, which include events potentially healthcare practitioners, and independent researchers for the purpose of associated with S1P receptor modulation as well as events potentially improving scientific and medical knowledge as well as fostering inno­ related to immune modulation. As first-dose cardiac effects are reported vative treatment approaches. Data requests may be submitted to Cel­ in the primary study publications, this analysis focused on longer-term gene, a Bristol Myers Squibb company, at https://vivli.org/ourmember/ cardiac safety outcomes only. Concurrent absolute lymphocyte count celgene/ and must include a description of the research proposal. (ALC) <0.2 × 109/L and infection was evaluated. Monoamine oxidase B (MAO-B) metabolizes one of the minor active metabolites of ozanimod, 3. Results RP101075, to form CC112273, the major active metabolite; further­ more, active metabolites of ozanimod (CC112273 and CC1084037) 3.1. Study population inhibit MAO-B with more than 1000-fold selectivity over MAO-A in vitro (Zeposia package insert, 2020; Zeposia summary of product character­ Overall, 2631 participants with RMS received ozanimod 0.92 mg istics, 2020). Therefore, we assessed the potential for drug interactions (Fig. 1). As of the data cutoff, mean (SD [standard deviation]) exposure in patients concurrently taking ozanimod and either serotonergic med­ to ozanimod 0.92 mg was 32.0 months (12.8) with 7058.5 PY on study, ications (which are metabolized primarily by MAO-A) or MAO in­ maximum exposure was about 75 months, and approximately 90% hibitors. We also report deaths and pregnancy outcomes among all (2256/2494) of DAYBREAK participants were still receiving ozanimod ozanimod-treated participants. 0.92 mg. For comparison, in the phase 3 trials, exposure to ozanimod 0.92 mg (n=882) was 18.1 months (6.0) with 1345.4 PY on study. 2.3. Statistical analyses In addition, 1033 participants had exposure to ozanimod 0.46 mg in one of the parent trials, 877 of whom subsequently received ozanimod Safety results among those exposed to at least one dose of ozanimod 0.92 mg in DAYBREAK. Participants who were exposed to any ozanimod 0.92 mg (the approved dose) are presented in the pooled population dose were included in select analyses (Section 2.3). At data cutoff, mean from all RMS trials as well as in the phase 3 trials as a point of reference. (SD) overall exposure to any ozanimod dose (n=2787) was 37.1 months Compared with the phase 3 trials, the pooled population had a longer (14.7) with 8688.3 PY in any RMS study. In the phase 3 trials, mean (SD) mean duration of exposure to ozanimod due to continued treatment in combined exposure to either ozanimod dose (n=1774) was 17.9 months DAYBREAK. For safety outcomes with low event rates and/or longer (5.97) with 2686.8 PY on study. Demographics and baseline disease time to onset (malignancies, ME, drug interactions, and pregnancies), characteristics are provided in Table 1. calculated incidences and IRs for all participants exposed to either dose of ozanimod (0.46 or 0.92 mg) are presented. TEAE rates are reported 3.2. TEAEs throughout the duration of ozanimod exposure, through the cutoff date; ALC and blood pressure (BP) over time are presented through month 42, The study duration‒adjusted IR of any TEAE was 772.2/1000 PY the last time point for which about 40% or more of participants have among participants exposed to ozanimod 0.92 mg in RMS trials; for data available. All outcomes were analyzed descriptively. comparison, the IR in the ozanimod 0.92 mg groups from the phase 3 Table 1 Demographic and baseline disease characteristics: safety population. Phase 3 Study Population Overall RMS Population Ozanimod 0.92 mg (N=882) Ozanimod 0.92 mg (N=2631) Any Ozanimod (N=2787)a Age, mean (SD) years 35.4 (9.1) 36.0 (9.2) 35.9 (9.1) Sex, n (%) Female 576 (65.3) 1765 (67.1) 1868 (67.0) Male 306 (34.7) 866 (32.9) 919 (33.0) Race, n (%) White 876 (99.3) 2608 (99.1) 2758 (99.0) Black 5 (0.6) 16 (0.6) 21 (0.8) Asian 1 (0.1) 4 (0.2) 4 (0.1) Other 0 3 (0.1) 4 (0.1) Hispanic ethnicity, n (%) 16 (1.8) 29 (1.1) 32 (1.1) Region, n (%) Eastern Europe 790 (89.6) 2365 (89.9) 2490 (89.3) Rest of world 92 (10.4) 266 (10.1) 297 (10.7) BMI, mean (SD), kg/m2 24.3 (4.8) 24.3 (4.8) 24.3 (4.8) Time since MS symptom onset, mean (SD), years 6.9 (6.3) 6.8 (6.2) 6.8 (6.1) Time since MS diagnosis, mean (SD), years 3.8 (4.7) 3.7 (4.6) 3.7 (4.6) EDSS score, mean (SD) 2.6 (1.2) 2.6 (1.2) 2.6 (1.2) Number of GdE lesions, mean (SD) 1.7 (3.6) 1.6 (3.3) 1.7 (3.3) Number of T2 lesions, mean (SD) 51.3 (36.3) 51.4 (36.2) 51.2 (36.1) Prior exposure to any MS disease-modifying therapy, n (%) 252 (28.6) 753 (28.6) 801 (28.7)b Abbreviations: BMI, body mass index; EDSS, Expanded Disability Status Scale; GdE, gadolinium-enhancing; MS, multiple sclerosis; OLE, open-label extension; RMS, relapsing multiple sclerosis; SD, standard deviation. a Includes participants who received only ozanimod 0.92 mg as well as those who received ozanimod 0.46 mg alone (if they did not enroll in the OLE study) or followed by ozanimod 0.92 mg in the OLE study. b Previous disease-modifying therapies used by the “any ozanimod” group included glatiramer acetate (n=317 [11.4%]), interferon beta-1a (n=288 [10.3%]), interferon beta-1b (n=229 [8.2%]), peginterferon beta-1a (n=35 [1.3%]), teriflunomide (n=33 [1.2%]), daclizumab (n=12 [0.4%]), dimethyl fumarate (n=9 [0.3%]), interferon (unspecified, n=8 [0.3%]), and mitoxantrone (n=2 [0.07%]). 3 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Table 2 TEAEs in participants with RMS who were treated with ozanimod 0.92 mg. Phase 3 Study Population (N=882) Overall RMS Population (N=2631) a Incidence, n (%) IR/1000 PY Incidence, n (%) IR/1000 PYa (95% CI) (95% CI) Any TEAE 592 (67.1) 896.1 (825.4‒971.3) 2106 (80.0) 772.2 (739.5‒805.9) Severe TEAEs 22 (2.5) 16.5 (10.4‒25.0) 129 (4.9) 18.7 (15.6‒22.2) Serious TEAEs 41 (4.6) 31.2 (22.4‒42.4) 224 (8.5) 33.2 (29.0‒37.9) Permanent discontinuation for TEAEs 26 (2.9) 19.4 (12.7‒28.5) 66 (2.5) 9.4 (7.3‒11.9) TEAEs in ≥5% of participants Nasopharyngitis 98 (11.1) 78.8 (64.0‒96.1) 457 (17.4) 72.9 (66.3‒79.9) Headache 78 (8.8) 61.7 (48.8‒77.0) 339 (12.9) 52.5 (47.0‒58.4) URTI 52 (5.9) 40.3 (30.1‒52.8) 249 (9.5) 37.6 (33.1‒42.6) Lymphopenia NAb NAb 222 (8.4) 33.1 (28.9‒37.8) ALC decreased NAb NAb 181 (6.9) 26.6 (22.9‒30.8) GGT increased 40 (4.5) 30.5 (21.8‒41.6) 174 (6.6) 25.8 (22.1‒30.0) Back pain 35 (4.0) 26.6 (18.5‒37.0) 162 (6.2) 23.9 (20.3‒27.8) Hypertension 30 (3.4) 22.8 (15.4‒32.5) 141 (5.4) 20.7 (17.4‒24.4) UTI 36 (4.1) 27.4 (19.2‒37.9) 138 (5.2) 20.2 (17.0‒23.9) ALT increased 47 (5.3) 36.2 (26.6‒48.1) 129 (4.9) 19.0 (15.8‒22.5) Influenza-like illness 44 (5.0) 34.4 (25.0‒46.2) 65 (2.5) 9.4 (7.3‒12.0) Serious TEAEs in ≥2 participants Appendicitis 3 (0.3) 2.2 (0.5‒6.5) 8 (0.3) 1.1 (0.5‒2.2) Uterine leiomyoma 1 (0.1) 0.7 (0.0‒4.1) 8 (0.3) 1.1 (0.5‒2.2) Pyelonephritis acute 1 (0.1) 0.7 (0.0‒4.1) 7 (0.3) 1.0 (0.4‒2.0) Intervertebral disc disorder 2 (0.2) 1.5 (0.2‒5.4) 6 (0.2) 0.9 (0.3‒1.9) Depression 0 0 (0.0‒2.7) 4 (0.2) 0.6 (0.2‒1.5) Intervertebral disc protrusion 1 (0.1) 0.7 (0.0‒4.1) 4 (0.2) 0.6 (0.2‒1.5) Pneumonia 0 0 (0.0‒2.7) 4 (0.2) 0.6 (0.2‒1.5) Uterine hemorrhage 0 0.0 (0.0‒2.7) 4 (0.2) 0.6 (0.2‒1.5) Abortion spontaneous 1 (0.1) 0.7 (0.0‒4.1) 3 (0.1) 0.4 (0.1‒1.2) Craniocerebral injury 1 (0.1) 0.7 (0.0‒4.1) 3 (0.1) 0.4 (0.1‒1.2) Epilepsy 1 (0.1) 0.7 (0.0‒4.1) 3 (0.1) 0.4 (0.1‒1.2) Headache 0 0 (0.0‒2.7) 3 (0.1) 0.4 (0.1‒1.2) ALC: absolute lymphocyte count; ALT: alanine aminotransferase; CI: confidence interval; GGT: gamma-glutamyl transferase; IR: incidence rate; NA: not applicable; PY: person-years; RMS: relapsing multiple sclerosis; TEAEs: treatment-emergent adverse events; URTI: upper respiratory tract infection; UTI: urinary tract infection. a IR/1000 PY, study duration‒adjusted incidence rate per 1000 person-years, calculated as number of participants with a TEAE of interest/PY x 1000, where PY was calculated as (date of first TEAE of interest – date of first dose of study drug + 1)/365.25; for participants without a TEAE of interest, time on study was the study duration (last date on study – date of first dose of study drug + 1)/365.25. b Investigators in the phase 3 RMS studies were blinded to lymphocyte count data (a key pharmacodynamic effect of ozanimod); therefore, TEAEs related to lymphocyte counts were not reported. trials was 896.1/1000 PY (Table 2). The most common TEAEs were 3.4. Hepatic effects nasopharyngitis, headache, upper respiratory tract infection, and lym­ phopenia. The study duration‒adjusted IR for serious TEAEs was 33.2/ Among participants treated with ozanimod 0.92 mg across the RMS 1000 PY among ozanimod 0.92 mg recipients across all RMS trials and trials, 3.9% had a maximum alanine aminotransferase (ALT) level ≥3 x 31.2/1000 PY in the phase 3 trials (Table 2). the upper limit of normal (ULN), 1.2% had a maximum aspartate aminotransferase (AST) level ≥3 x ULN, and 2.4% experienced maximum bilirubin >2 x ULN (Table 3); there was no clear temporal 3.3. Cardiovascular effects pattern to these abnormalities relative to start of treatment. Consecutive ALT ≥3 x ULN occurred in 1.1% and consecutive AST ≥3 x ULN Modest gradual increases in mean sitting/supine systolic and dia­ occurred in 0.2%. The majority of participants with ≥1 assessment after stolic BP continued over time during treatment with ozanimod 0.92 mg an abnormality was detected recovered to ≤1 x ULN while still on (Table A.2). Blood pressure increases were detectable approximately 3 treatment (ALT: 84/110 [76.4%]; AST: 24/25 [96.0%]; bilirubin: 27/52 months after treatment initiation. After 42 months of ozanimod 0.92 mg, [51.9%]). Except for bilirubin, the same was true for recovery in the systolic BP had increased from a mean (SD) of 116.5 (12.23) at baseline phase 3 trials (ALT: 25/41 [61.0%]; AST: 5/5 [100%]; bilirubin: 4/14 to 122.1 (12.49) mm Hg and diastolic BP increased from a mean (SD) of [28.6%]). Mean (SD) maximum change from baseline was 27.0 IU/L 74.3 (9.17) to 77.4 (9.14) mm Hg. In the phase 3 trials, the increase from (66.3) for ALT, 12.8 IU/L (44.0) for AST, and 6.05 µmol/L (5.4) for baseline in systolic BP was about 1 to 2 mm Hg greater with ozanimod bilirubin. There were no cases of severe drug-induced liver injury. than IFN beta-1a, whereas impact on diastolic BP was similar between The most common hepatobiliary TEAE was hyperbilirubinemia the two treatment groups. There was no increase in exposure-adjusted (n=24 [0.9%]; Table 3); six of these participants had bilirubin >1 × ULN IRs of hypertension-related TEAEs (22.8/1000 PY in phase 3 vs 20.7/ and one had bilirubin >2 × ULN at baseline. There were six serious 1000 PY in the overall RMS population) and no increase in the IRs of hepatobiliary TEAEs among all RMS participants treated with ozanimod bradycardias, cardiac conduction abnormalities, and ischemic heart 0.92 mg: one case each of cholelithiasis, biliary polyp, acute cholecys­ conditions with longer exposure (Table A.3). Two patients were hospi­ titis, chronic cholecystitis, cholestasis, and chronic hepatitis. talized for hypertension, which resolved without ozanimod interrup­ tion. There were no cases of second-degree Mobitz type 1 or higher atrioventricular block observed on electrocardiogram (ECG). Eleven 3.5. Macular edema participants (0.4%) exposed to ozanimod 0.92 mg experienced ischemic heart conditions (Table A.3). In the phase 3 trials, TEAEs of ME or cystoid ME were reported in 2/ 4 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Table 3 Hepatic laboratory abnormalities and TEAEs in participants with RMS treated with ozanimod 0.92 mg. Phase 3 Study Population (N=882) Overall RMS Population (N=2631) Based on laboratory testing n=878 n=2623 Maximum ALT ≥3 x ULN, n (%) 48 (5.5) 102 (3.9) ≥5 x ULN, n (%) 14 (1.6) 25 (1.0) ≥10 x ULN, n (%) 4 (0.5) 9 (0.3) Mean (SD) maximum change from baseline, IU/L 28.4 (61.0) 27.0 (66.3) Maximum AST ≥3 x ULN, n (%) 9 (1.0) 31 (1.2) ≥5 x ULN, n (%) 5 (0.6) 13 (0.5) ≥10 x ULN, n (%) 4 (0.5) 6 (0.2) Mean (SD) maximum change from baseline, IU/L 13.7 (34.4) 12.8 (44.0) Maximum Bilirubin >2 x ULN, n (%) 14 (1.6) 64 (2.4) >3 x ULN, n (%) 3 (0.3) 9 (0.3) Mean (SD) maximum change from baseline, µmol/L 4.6 (4.9) 6.1 (5.4) Hy’s Law casesa, n 0 0 N (%) IR/1000 PYb N (%) IR/1000 PYb (95% CI) (95% CI) Any Hepatobiliary TEAEs, n (%) 15 (1.7) 11.2 (6.3‒18.5) 73 (2.8) 10.5 (8.3‒13.2) Hepatobiliary TEAEs in ≥3 participants, n (%) Hyperbilirubinemia 3 (0.3) 2.2 (0.5‒6.5) 24 (0.9) 3.4 (2.2‒5.1) Cholecystitis chronic 1 (0.1) 0.7 (0.0‒4.1) 7 (0.3) 1.0 (0.4‒2.0) Biliary dyskinesia 0 0 (0.0‒2.7) 5 (0.2) 0.7 (0.2‒1.7) Hepatic cyst 0 0 (0.0‒2.7) 5 (0.2) 0.7 (0.2‒1.7) Cholelithiasis 1 (0.1) 0.7 (0.0‒4.1) 4 (0.2) 0.6 (0.2‒1.5) Chronic hepatitis 0 0 (0.0‒2.7) 3 (0.1) 0.4 (0.1‒1.2) Hepatitis 0 0 (0.0‒2.7) 3 (0.1) 0.4 (0.1‒1.2) Hepatitis toxic 2 (0.2) 1.5 (0.2‒5.4) 3 (0.1) 0.4 (0.1‒1.2) Hypertransaminasemia 1 (0.1) 0.7 (0.0‒4.1) 3 (0.1) 0.4 (0.1‒1.2) ALT: alanine aminotransferase; AST: aspartate aminotransferase; CI: confidence interval; IR: incidence rate; PY: person-years; RMS: relapsing multiple sclerosis; SD: standard deviation; TEAEs: treatment-emergent adverse events; ULN: upper limit of normal. a Hy’s Law, defined as ALT or AST ≥3 x ULN plus total bilirubin >2 x ULN without cholestasis and without alternative explanation, is used by the US Food and Drug Administration to identify drugs likely to cause severe drug-induced liver injury (Food and Drug Administration, 2009). An unblinded external panel of expert hepatologists reviewed all cases of concurrent ALT/AST elevations ≥3 x ULN and bilirubin >2 x ULN and concluded that none met Hy’s Law criteria due to alternate explanations and based on the pattern of abnormalities. b IR/1000 PY, study duration‒adjusted incidence rate per 1000 person-years, calculated as number of participants with a TEAE of interest/PY x 1000, where PY was calculated as (date of first TEAE of interest – date of first dose of study drug + 1)/365.25; for participants without a TEAE of interest, time on study was the study duration (last date on study – date of first dose of study drug + 1)/365.25. 882 (0.2%) participants treated with ozanimod 0.92 mg, 3/892 (0.3%) who discontinued ozanimod. As of March 2020, ME was ongoing in the treated with ozanimod 0.46 mg, and 3/885 (0.3%) treated with IFN patient who continued on ozanimod. beta-1a. Across the ozanimod clinical development program, there were 10 (0.4%) reported TEAEs of ME or cystoid ME among 2787 RMS par­ 3.6. ALC reductions ticipants treated with either ozanimod dose. The MERP confirmed seven cases (0.3%) of ME (IR 0.8/1000 PY; Among participants treated with ozanimod 0.92 mg, mean ALC was 95% confidence interval [CI] 0.3‒1.7), four in participants treated with reduced to 51% of baseline at the first assessment and 44% of baseline at ozanimod 0.92 mg and three in participants treated with ozanimod 0.46 month 3, after which mean ALC was maintained at approximately 40% mg. This included four cases (ozanimod 0.92 mg: n=1; 0.46 mg: n=3) of baseline (Fig. 2). Among the 2631 participants exposed to ozanimod that occurred during the phase 3 trials, affecting 0.2% of 1774 0.92 mg, ALC was less than the lower limit of normal, <0.8 × 109/L, ozanimod-treated participants (IR 1.5/1000 PY; 95% CI 0.4‒3.8); the <0.5 × 109/L, or <0.2 × 109/L on at least one measurement in 2488 other three cases occurred in DAYBREAK, in participants who had (94.9%), 2345 (89.5%), 1669 (63.4%), and 182 (6.9%) participants, switched to ozanimod from IFN beta-1a. Across all seven MERP- respectively. confirmed cases, onset of ME occurred 15 to 366 days after ozanimod was first initiated (Table A.4). There were no MERP-confirmed cases of ME in the phase 3 IFN beta-1a groups. 3.7. Infections All cases of confirmed ME had pre-existing risk factors or con­ founding conditions consisting of a history of ME, ME secondary to Across all RMS studies, the overall incidence of infection among ocular trauma, central serous choroidopathy (CSR), uveitis, prior unre­ participants treated with ozanimod 0.92 mg was 48.6% (IR 270.1/1000 ported uveitis (intraocular inflammation), pigment epithelial detach­ PY) (Table 4), which was similar to the incidence in the phase 3 trials ment with possible choroidal neovascularization, and a history of (35.1%; IR 300.5/1000 PY). The most common infections were naso­ retinopathy and optic neuritis (Table A.4). Ozanimod was permanently pharyngitis, upper respiratory tract infection, and urinary tract infection withdrawn in all but one person with trace ME due to CSR who was (Table 4). allowed to continue treatment because the choroidal process associated The incidence of serious infections was 1.7% in the overall RMS with CSR-related ME differs from intraretinal ME associated with S1P- population treated with ozanimod 0.92 mg (IR 6.3/1000 PY) and 1.0% related cases. As of the data cutoff, ME had resolved in all patients in phase 3 trial participants treated with the same dose (IR 6.7/1000 PY). Serious infections occurring in more than two participants are 5 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Fig. 2. Mean (SE) ALC by visit during treatment with ozanimod 0.92 mg: overall RMS population ALC: absolute lymphocyte count; LLN: lower limit of normal; RMS: relapsing multiple sclerosis; SE: standard error. shown in Table 2, and a complete list is provided in Table A.5. There was postbaseline visit in 171/2568 (6.7%) participants treated with ozani­ one infection-related death: a bed-ridden participant was hospitalized mod 0.92 mg across all RMS trials; the corresponding incidence for re­ and discontinued ozanimod due to community-acquired pneumonia ductions in FVC was 204/2568 (7.9%) at any visit and 146/2568 (5.7%) after >5.2 years of ozanimod treatment, and died of respiratory failure for two consecutive visits or on the last postbaseline visit. 11 days later. Pulmonary adverse events occurred in 39/2631 (1.5%) participants Opportunistic infections occurred in 113/2631 (4.3%) participants with RMS who received ozanimod 0.92 mg, including one report treated with ozanimod 0.92 mg (IR 16.4/1000 PY) across all trials, (<0.1%) of serious chronic obstructive pulmonary disease. Dyspnea including 16/882 (1.8%) cases during phase 3 (IR 12.0/1000 PY). occurred in 10/2631 (0.4%) participants exposed to ozanimod 0.92 mg Herpes virus infections were the most frequent opportunistic infection across all RMS trials; one participant discontinued ozanimod because of and included oral herpes and herpes zoster/varicella zoster virus dyspnea. infection (Table 4). None of the herpes zoster infections were serious, and the affected participants continued treatment with ozanimod 3.10. Potential drug interactions without adverse clinical consequences. There were no serious opportu­ nistic infections, including progressive multifocal leukoencephalopathy During the RMS trials, 573/2787 (20.6%) participants concurrently (PML). received ozanimod 0.92 or 0.46 mg and serotonergic medications, One participant (0.5%) had an ALC <0.2 × 109/L around the time of including psychoanaleptic or psycholeptic drugs (12.2%), analgesics onset of a serious infection (pyelonephritis) and one (0.5%) had an ALC (7.8%), nasal preparations (2.0%), cough/cold preparations (1.5%), and <0.2 × 109/L around the onset of a nonserious opportunistic infection anesthetics (0.7%). No participants developed serotonin syndrome (oral herpes) (Table 4). Both infections occurred during DAYBREAK. (preferred terms serotonin syndrome, neuroleptic malignant syndrome, Thus, there was no apparent association between ALC <0.2 × 109/L and hyperthermia malignant). Three (0.1%) participants in the RMS trials serious or opportunistic infections. concurrently took ozanimod 0.92 or 0.46 mg and an MAO inhibitor (moclobemide in all three cases); none experienced hypertensive crisis. 3.8. Malignancies 3.11. Pregnancy outcomes Among 2787 participants with any ozanimod exposure in any RMS study, there have been 11 nonmelanoma skin cancers, 1 melanoma, and Participants with reproductive potential were required to use effec­ 13 noncutaneous malignancies that were treatment emergent (Table 5). tive contraception during ozanimod trials. Nonetheless, 36 of 1868 fe­ The overall IR of malignancy (289.3/100,000 PY; 95% CI 187.2‒427.1) male participants became pregnant while taking either dose of was similar to the rate in participants with ≤24 months’ ozanimod ozanimod as of the data cutoff; one was initially carrying twins. Out­ exposure in phase 3 trials (298.2/100,000 PY; 95% CI 128.7–587.6) comes include 24 live births: 18 normal and 3 premature but normal (Table 5). For comparison, in the phase 3 trials, there were two infants (12.5% of live births), and 1 report each of neonatal icterus, late treatment-emergent malignancies (basal cell carcinoma and chronic intrauterine growth retardation with subsequent normal progress over lymphocytic leukemia) reported in 885 participants treated with IFN the first year, and duplex kidney. There were five early spontaneous beta-1a (IR 150.8/100,000 PY; 95% CI 18.3‒544.8). Prior to data cutoff, abortions (13.5%) (of which one was loss of a twin) and seven elective three deaths occurred in patients with malignancies, including meta­ terminations. One participant refused consent to follow-up. static malignant neoplasm of unknown primary, metastatic pancreatic cancer, and glioblastoma. 3.12. Mortality 3.9. Pulmonary function There were eight deaths, of which three were from malignancies and one from pneumonia, as described above. Two were accidents: a Pulmonary function test (PFT) abnormalities (forced expiratory drowning reported in phase 3 RADIANCE (Cohen et al., 2019b) and a volume in 1 second [FEV1] or forced vital capacity [FVC] decrease to pedestrian-train accident. One participant died from a pulmonary em­ <80% of baseline) occurred in 311/2631 (11.8%) RMS participants bolism after a 38-day hospitalization for surgical repair of a lower limb treated with ozanimod 0.92 mg across all RMS clinical trials, which fracture. Another died as a result of chronic kidney failure approxi­ included 90/882 (10.2%) participants treated with ozanimod 0.92 mg in mately 10 months after prematurely discontinuing ozanimod 0.92 mg the phase 3 studies. FEV1 was <80% of baseline at any visit in 242/2568 during phase 3 RADIANCE due to Guillain-Barré syndrome and posterior (9.4%) and for two consecutive postbaseline visits or on the last reversible encephalopathy syndrome. 6 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Table 4 Infections in participants with RMS who were treated with ozanimod 0.92 mg. Phase 3 Study Population (N=882) Overall RMS Population (N=2631) a Incidence, n (%) IR/1000 PY Incidence, n (%) IR/1000 PYa (95% CI) (95% CI) Any infection 310 (35.1) 300.5 (268.0‒335.9) 1278 (48.6) 270.1 (255.5‒285.3) Any serious infectionb 9 (1.0) 6.7 (3.1‒12.8) 44 (1.7) 6.3 (4.6‒8.4) Any opportunistic infection 16 (1.8) 12.0 (6.9‒19.5) 113 (4.3) 16.4 (13.5‒19.7) Infections in ≥2% of participants Nasopharyngitis 98 (11.1) 78.8 (64.0‒96.1) 457 (17.4) 72.9 (66.3‒79.9) URTI 52 (5.9) 40.3 (30.1‒52.8) 249 (9.5) 37.6 (33.1‒42.6) UTI 36 (4.1) 27.4 (19.2‒37.9) 138 (5.2) 20.2 (17.0‒23.9) Bronchitis 23 (2.6) 17.3 (11.0‒26.0) 118 (4.5) 17.2 (14.2‒20.6) Pharyngitis 28 (3.2) 21.2 (14.1‒30.6) 91 (3.5) 13.2 (10.6‒16.2) Respiratory tract infection 18 (2.0) 13.5 (8.0‒21.4) 110 (4.2) 16.0 (13.1‒19.2) Respiratory tract infection viral 21 (2.4) 15.8 (9.8‒24.2) 99 (3.8) 14.3 (11.7‒17.5) Influenza 9 (1.0) 6.7 (3.1‒12.8) 73 (2.8) 10.5 (8.2‒13.2) Rhinitis 19 (2.2) 14.3 (8.6‒22.4) 77 (2.9) 11.1 (8.8‒13.9) Sinusitis 13 (1.5) 9.8 (5.2‒16.7) 76 (2.9) 10.9 (8.6‒13.7) Opportunistic infection in ≥2 participantsc Oral herpes 6 (0.7) 4.5 (1.6‒9.7) 40 (1.5) 5.7 (4.1‒7.8) Herpes zoster (including VZV) 5 (0.6) 3.7 (1.2‒8.7) 37 (1.4) 5.3 (3.7‒7.3) Herpes simplex 1 (0.1) 0.7 (0.0‒4.1) 12 (0.5) 1.7 (0.9‒3.0) Genital herpes 0 0 (0.0‒2.7) 5 (0.2) 0.7 (0.2‒1.7) Fungal infection 0 0 (0.0‒2.7) 4 (0.2) 0.6 (0.2‒1.5) Candida infection 0 0 (0.0‒2.7) 3 (0.1) 0.4 (0.1‒1.2) Oral fungal infection 0 0 (0.0‒2.7) 3 (0.1) 0.4 (0.1‒1.2) Herpes dermatitis 0 0 (0.0‒2.7) 2 (0.07) 0.3 (0.0‒1.0) Genital fungal infection 1 (0.1) 0.7 (0.0‒4.1) 2 (0.07) 0.3 (0.0‒1.0) Minimal postbaseline ALC <0.5 × 109/L 480 (54.4) 542.5 (495.1‒593.3) 1669 (63.4) 450.6 (429.3‒472.8) Minimal postbaseline ALC <0.2 × 109/L 29 (3.3) 21.9 (14.7‒31.4) 182 (6.9) 26.7 (23.0‒30.9) ALC <0.2 × 109/L around onset of any infection 2 (6.9) 1.5 (0.2‒5.4) 18 (9.9) 2.6 (1.5‒4.0) ALC <0.2 × 109/L around onset of serious infectiond 0 0 (0.0‒2.7) 1/182 (0.5) (pyelonephritis) 0.1 (0.0‒0.8) ALC <0.2 × 109/L around onset of opportunistic infectiond 0 0 (0.0‒2.7) 1/182 (0.5) (oral herpes) 0.1 (0.0‒0.8) ALC: absolute lymphocyte count; CI: confidence interval; IR: incidence rate; PY: person-years. RMS, relapsing multiple sclerosis; TEAEs, treatment-emergent adverse events; VZV, varicella-zoster virus. a IR/1000 PY, study duration‒adjusted incidence rate per 1000 person-years, calculated as number of participants with a TEAE of interest/PY x 1000, where PY was calculated as (date of first TEAE of interest – date of first dose of study drug + 1)/365.25; for participants without a TEAE of interest, time on study was the study duration (last date on study – date of first dose of study drug + 1)/365.25. b Serious infections occurring in >2 participants can be found in Table 2; a complete list of serious infections in participants with RMS who were treated with ozanimod 0.92 mg is available in table A-5. c Additional opportunistic infections that occurred in a single participant each (IR <0.1/1000 PY) across all participants exposed to ozanimod 0.92 mg in any of the RMS trials included anal fungal infection, gastrointestinal candidiasis, esophageal candidiasis, oral candidiasis, ophthalmic herpes simplex, herpes virus infection, nasal herpes, and varicella; 4 of these 8 infections occurred during the phase 3 trials. d Participants who experienced an initial serious infection or opportunistic infection and had an ALC <0.2 × 109/L at the laboratory visit prior to the event to either the time of the event or the assessment just after onset of the event. 3.13. Clinical rebound with the phase 3 population. Slowing of heart rate can occur after first-dose exposure to S1P modulators; however, as reported previously The number of participants with posttreatment follow-up data is (Cohen et al., 2016; Cohen et al., 2019b; Comi et al., 2019; Tran et al., limited, as most were continuing in DAYBREAK at data cutoff. However, 2017), effects on heart rate and atrioventricular conduction are atten­ as of this analysis, there have been no posttreatment AEs indicative of uated by dose escalation when initiating ozanimod and such effects are rebound (MS flare). uncommon with continued use. There were no second-degree Mobitz type 1 or higher atrioventricular blocks based on ECG, and few (0.4%) 4. Discussion participants exposed to ozanimod 0.92 mg in any of the RMS trials experienced ischemic heart conditions. This analysis provides a comprehensive perspective of the safety of Hepatic enzyme elevations and hepatic-related TEAEs did not in­ ozanimod 0.92 mg. The 2631 participants from all ozanimod RMS crease with continued ozanimod exposure. Hepatic enzyme elevations clinical trials had a mean duration of exposure to ozanimod 0.92 mg that were generally asymptomatic, transient, and did not lead to severe drug- was approximately 14 months longer than ozanimod exposure in the 12- induced liver injury. to 24-month phase 3 studies (Cohen et al., 2019b; Comi et al., 2019). ME cases independent of coexisting risk factors have been reported in The safety profile in this population was generally consistent with that studies of fingolimod (0.4% with 0.5 mg/d), siponimod (2% with 2 mg/ reported in the phase 3 trials (Cohen et al., 2019b; Comi et al., 2019) and d), and ponesimod (0.9% with 10, 20, or 40 mg/d) (Kappos et al., 2018; does not point to any new safety concerns with continued ozanimod use. Kappos et al., 2014; Olsson et al., 2014). ME rates with ozanimod were Long-term cardiovascular safety was evaluated because S1P re­ 0.3% to 0.4%, and all confirmed cases had predisposing comorbid ceptors are expressed by cardiomyocytes (Li and Zhang, 2016), and conditions. cardiovascular abnormalities have been reported with other S1P mod­ ALC reductions are an expected part of the mechanism of action of ulators (Kappos et al., 2018; Kappos et al., 2014; Olsson et al., 2014). S1P modulators. Ozanimod 0.92 mg reduced ALC to approximately 40% Mean systolic and diastolic BP increased with continued use of ozani­ of baseline, and about 7% of participants who received ozanimod 0.92 mod 0.92 mg; however, there was no increase in the IR of mg developed ALC counts <0.2 × 109/L on at least one laboratory hypertension-related TEAEs in the overall RMS population compared assessment. For comparison, ALC counts <0.2 × 109/L were reported in 7 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 Table 5 Treatment-emergent malignancies by preferred term in participants with RMS who were exposed to any dose of ozanimod (0.46 and/or 0.92 mg). Phase 3 Study Population (N=1774) Overall RMS Population (N=2787) a Incidence, n (%) IR/100,000 PY (95% CI) Incidence, n (%) IR/100,000 PYa (95% CI) Treatment-emergent malignancies 8 (0.5) 298.2 (128.7‒587.6) 25 (0.9) 289.3 (187.2‒427.1) Cutaneous 4 (0.2) 149.0 (40.6‒381.6) 12 (0.4) 138.8 (71.7‒ 242.4) Basal cell carcinoma 3 (0.2) 111.8 (23.0‒326.6) 9 (0.3) 104.0 (47.6‒197.5) Keratoacanthoma 1 (0.06) 37.2 (0.9‒207.4) 1 (0.04) 11.5 (0.3‒64.3) Squamous cell carcinoma 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Malignant melanoma 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Nonmelanoma skin cancer 4 (0.2) 149.0 (40.6–381.6) 11 (0.4) 127.2 (63.5‒227.6) Noncutaneous 4 (0.2) 148.9 (40.6‒381.4) 13 (0.5) 150.1 (79.9‒256.7) Breast cancer (women only)b 3/1174 (0.3) 168.7 (34.8‒493.0) 5/1868 (0.3) 86.4 (28.0‒201.5) Cervix carcinoma 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3-64.3) Testicular seminoma (pure) stage Ic 1 (0.06) 37.2 (0.9‒207.4) 1 (0.04) 11.5 (0.3‒64.3) Bile duct cancerd 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Clear cell renal carcinoma 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Glioblastoma 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Malignant neoplasm 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Pancreatic carcinoma metastatic 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Papillary thyroid cancer 0 0 (0.0‒2.7) 1 (0.04) 11.5 (0.3‒64.3) Treatment-emergent malignancies, excluding NMSC 4 (0.2) 148.9 (40.6‒381.4) 14 (0.5) 161.7 (88.4‒271.2) CI: confidence interval (based on the Poisson distribution); IR: incidence rate; NMSC: nonmelanoma skin cancer; PY: person-years; RMS: relapsing multiple sclerosis. a IR/100,000 PY, study duration‒adjusted incidence rate per 100,000 person-years, calculated as number of persons having the malignancy of interest/person-years x 100,000, where person-years = (date first malignancy of interest was documented – date of first dose of study drug + 1)/365.25; for participants not having the malignancy of interest, the time on study is the study duration (last date on study – first dose date of study drug +1)/365.25. b Breast cancer includes cases using preferred terms of invasive breast cancer, breast cancer, and breast neoplasm. c Diagnosed on study day 51. d Diagnosis changed and confirmed after the data cutoff to hydatid cysts. 18% of FREEDOMS participants treated with fingolimod 0.5 mg/ Mild but persistent declines in FEV1 and FVC were noted in d (Kappos et al., 2014). ozanimod-treated RMS patients beginning as early as 3 months of In the overall RMS population that had longer ozanimod 0.92 mg treatment. These changes were generally not symptomatic or associated exposure, overall rates of infections and serious infections were similar with reported TEAEs and were not progressive. Dose-dependent re­ to those with ≤24 months’ exposure in phase 3 trials. The incidence and ductions in FEV1 have been reported with other S1P modulators (May­ IR for opportunistic infections, particularly herpes infections, were zent package insert 2020; Calabresi et al., 2014; Olsson et al., 2014). higher in the overall RMS population than in the phase 3 subgroup; About 21% of ozanimod-treated participants were concurrently however, the 95% CIs for the IRs in the overall population overlapped using serotonergic drugs, and 0.1% were using MAO inhibitors. To date, completely with the wide 95% CIs from the phase 3 trials. As of the data there have been no cases suggestive of serotonin toxicity or serotonin cutoff, no serious opportunistic infections, including PML, have syndrome among concurrent users of serotonergic medications, and no occurred in ozanimod-treated participants. An association between ALC cases of hypertensive crisis among the three participants who concom­ <0.2 × 109/L and serious or opportunistic infections was not detected. itantly used ozanimod and an MAO inhibitor during the RMS trials. The overall malignancy rate observed among all ozanimod-treated Although the mean effective half-life of ozanimod active metabolites RMS study participants (289.3/100,000 PY) was similar to those with is approximately 11 days, elimination may take up to 3 months (Zeposia less exposure in phase 3 trials (298.2/100,000 PY). It was also consistent package insert, 2020; Zeposia summary of product characteristics, with reported malignancy rates among MS patients treated with other 2020); therefore, the 36 women who became pregnant during ozanimod disease-modifying therapies (IR 290 to 1190/100,000 PY) (Cook et al., treatment and continued their pregnancies after stopping ozanimod 2019) and the rate of invasive cancers in a general Swedish age-matched likely had some level of drug exposure during the majority of their first non-MS population (310/100,000 PY) (Alping et al., 2020). Although trimester. The rate of spontaneous abortion (13.5%) was within the there were more malignancies reported with ozanimod than with IFN expected rate (about 20%) in the general population (Rossen et al., beta-1a in the phase 3 trials, (Cohen et al., 2019b; Comi et al., 2019) the 2018), and the rate of preterm birth (8.3% of pregnancies; 12.5% of live low number of malignancies, relatively short duration of follow-up, and births) was similar to the European estimate of 8.7% (Chawanpaiboon IRs with wide, overlapping CIs do not suggest an increase in the overall et al., 2019). The only congenital abnormality, duplex kidney, occurs in risk of malignancies with ozanimod. There were no lymphomas or other the general population at a frequency of about 1.8% (Privett et al., lymphoproliferative diseases. Five cases of breast cancer have been re­ 1976). Nonclinical data indicate potential teratogenic effects; it is rec­ ported in ozanimod-treated women with RMS (IR 86.4/100,000 PY); this ommended that women use effective contraception during ozanimod rate remains within the expected incidence of breast cancer of 5.35 treatment and for at least 3 months thereafter (Zeposia package insert events over the treatment period, derived by applying the breast cancer 2020; Zeposia summary of product characteristics 2020). Labels for IR (92.4/100,000 PY) in an age-matched female population from the other S1P receptor modulators contain similar statements regarding Surveillance, Epidemiology, and End Results to the female RMS partici­ fetal risk (Gilenya package insert 2019; Mayzent package insert 2020). pants with exposure to ozanimod (5789.5 PY). The IR (95% CI) per 100, This analysis included a large pooled RMS population comprising 000 PY for nonmelanoma skin cancers was comparable in the overall participants with varying durations of disease, baseline levels of clinical RMS population (127.2 [63.5‒227.6]) and the phase 3 study population and MRI disease activity, disability, and history of disease-modifying (149.0 [40.6–381.6]), suggesting that with longer exposure, the inci­ therapy use. These data allow a comparison between the phase 3 dence did not increase. Furthermore, the IR (95% CI) in the overall RMS experience and longer-term exposure in a larger number of patients due population was not markedly different from the IR (95% CI) of non­ to inclusion of the OLE. One limitation is the lack of a control group in melanoma skin cancer in a UK population of MS patients aged ≤59 years the OLE, preventing direct comparisons to other disease-modifying (101.8 [69.66–143.78]/100,000 PY) (adapted from Persson et al., 2020). therapies. Given that >99% of the participants were white, and nearly 8 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 90% hailed from Eastern Europe, generalizability of the findings to other Genzyme, Merck, Novartis, Roche, Sanofi, and Teva. populations is unknown. This is the longest duration of ozanimod B.A.C. Cree reports personal compensation for consulting for Akili, exposure reported to date (mean 32 months); however, since disease- Alexion, Biogen, EMD Serono, Novartis, Sanofi, and TG Therapeutics. modifying therapy for RMS in clinical practice is of indefinite dura­ N. Ding is an employee of Bristol Myers Squibb. tion, it is important to continue to collect safety data from longer term H.-P. Hartung reports personal fees for consulting, serving on steer­ follow-up, which is ongoing in DAYBREAK. ing committees, and speaking from Bayer Healthcare, Biogen, Celgene, GeNeuro, Genzyme, Merck, MedImmune, Novartis, Octapharma, Roche, 5. Conclusions Sanofi, and Teva. E.K. Havrdova reports personal compensation for consulting and Safety results in this larger RMS population were generally consistent speaking for Actelion, Biogen, Celgene, Merck, Novartis, Roche, Sanofi, with those of the phase 3 trials, providing reassurance of the safety of and Teva, and is supported by Czech Ministry of Education, project continued use of ozanimod relative to previous shorter periods of PROGRES Q27/LF1. observation. Planned future analyses from DAYBREAK, the ongoing OLE L. Kappos’s institution (University Hospital Basel) has received in the study, will provide further information about longer-term safety and last 3 years the following, which was used exclusively for research tolerability of ozanimod. support: steering committee, advisory board, consultancy fees, and support of educational activities from Actelion, Allergan, Almirall, CRediT authorship contribution statement Baxalta, Bayer, Biogen, Celgene, CSL Behring, Desitin, EXCEMED, Eisai, F. Hoffmann-La Roche, Genzyme, Japan Tobacco, Merck, Minoryx, Krzysztof W. Selmaj: Conceptualization, Formal analysis, Method­ Novartis, Pfizer, Sanofi Aventis, Santhera, and Teva; license fees for ology, Writing - original draft, Writing - review & editing. Jeffrey A Neurostatus-UHB products; and the research from the MS Center in Basel Cohen: Conceptualization, Formal analysis, Methodology, Writing - has been supported by grants from Bayer, Biogen, European Union, original draft, Writing - review & editing. Giancarlo Comi: Conceptu­ Innosuisse, Novartis, Roche Research Foundations, Swiss MS Society, alization, Formal analysis, Methodology, Writing - original draft, and Swiss National Research Foundation. Writing - review & editing. Amit Bar-Or: Conceptualization, Formal N. Minton is an employee of Bristol Myers Squibb. analysis, Methodology, Writing - original draft, Writing - review & X. Montalban has received speaking honoraria and travel expenses editing. Douglas L Arnold: Conceptualization, Formal analysis, Meth­ for scientific meetings, has been a steering committee member of clinical odology, Writing - original draft, Writing - review & editing. Lawrence trials or participated in advisory boards of clinical trials in the past 3 Steinman: Conceptualization, Formal analysis, Methodology, Writing - years with Actelion, Alexion, Bayer, Biogen, Celgene, EMD Serono, original draft, Writing - review & editing. Hans Peter Hartung: EXCEMED, Genzyme, Immunic, MedDay, Merck, MS International Conceptualization, Formal analysis, Methodology, Writing - original Federation, Nervgen, National Multiple Sclerosis Society, Novartis, draft, Writing - review & editing. Xavier Montalban: Conceptualiza­ Roche, Sanofi Genzyme, Teva Pharmaceuticals, and TG Therapeutics. tion, Formal analysis, Methodology, Writing - original draft, Writing - K.W. Selmaj reports consulting for Biogen, Celgene, Genzyme, review & editing. Eva Kubala Havrdova: Conceptualization, Formal Merck, Novartis, Ono Pharma, Roche, Synthon, and Teva. analysis, Methodology, Writing - original draft, Writing - review & J.K. Sheffield is an employee of Bristol Myers Squibb. editing. Bruce A.C. Cree: Conceptualization, Formal analysis, Meth­ L. Steinman reports consulting for AbbVie, Atreca, Celgene, Novartis, odology, Writing - original draft, Writing - review & editing. Neil Teva, Tolerion, and EMD Serono, and research support from Atara, Minton: Conceptualization, Formal analysis, Methodology, Writing - Biogen, and Celgene. original draft, Writing - review & editing. James K Sheffield: Concep­ tualization, Formal analysis, Methodology, Writing - original draft, Supplementary materials Writing - review & editing. Ning Ding: Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Ludwig Kappos: Supplementary material associated with this article can be found, in Conceptualization, Formal analysis, Methodology, Writing - original the online version, at doi:10.1016/j.msard.2021.102844. draft, Writing - review & editing. References Acknowledgments Alping, P., Askling, J., Burman, J., Fink, K., Fogdell-Hahn, A., Gunnarsson, M., Hillert, J., The ozanimod RMS trials were sponsored by Celgene International II. Langer-Gould, A., Lycke, J., Nilsson, P., Salzer, J., Svenningsson, A., Vrethem, M., The sponsor was involved in data analysis and interpretation, and Olsson, T., Piehl, F., Frisell, T., 2020. Cancer risk for fingolimod, natalizumab, and rituximab in MS patients. Ann. Neurol. 87 (5), 688–699. manuscript preparation, review, and approval. All authors vouch for Calabresi, P.A., Radue, E.W., Goodin, D., Jeffery, D., Rammohan, K.W., Reder, A.T., data accuracy, reviewed all drafts, and approved the final manuscript. Vollmer, T., Agius, M.A., Kappos, L., Stites, T., Li, B., Cappiello, L., von Rosenstiel, P., Support for third-party writing assistance for this manuscript was pro­ Lublin, F.D., 2014. Safety and efficacy of fingolimod in patients with relapsing- remitting multiple sclerosis (FREEDOMS II): a double-blind, randomised, placebo- vided by Peloton Advantage, LLC, an OPEN Health company, and was controlled, phase 3 trial. Lancet Neurol. 13 (6), 545–556. funded by Bristol Myers Squibb. Chawanpaiboon, S., Vogel, J.P., Moller, A.B., Lumbiganon, P., Petzold, M., Hogan, D., D.L. Arnold reports personal fees for consulting and/or grants from Landoulsi, S., Jampathong, N., Kongwattanakul, K., Laopaiboon, M., Lewis, C., Rattanakanokchai, S., Teng, D.N., Thinkhamrop, J., Watananirun, K., Zhang, J., Albert Charitable Trust, Biogen, Celgene, F. Hoffmann-La Roche, Fre­ Zhou, W., Gulmezoglu, A.M., 2019. Global, regional, and national estimates of levels quency Therapeutics, MedDay, Merck Serono, Novartis, and Sanofi of preterm birth in 2014: a systematic review and modelling analysis. Lancet Glob Aventis, and an equity interest in NeuroRx Research. Health 7 (1), e37–e46. Cohen, J.A., Arnold, D.L., Comi, G., Bar-Or, A., Gujrathi, S., Hartung, J.P., Cravets, M., A. Bar-Or participated as a speaker in meetings sponsored by and Olson, A., Frohna, P.A., Selmaj, K.W., 2016. Safety and efficacy of the selective received consulting fees and/or grant support from Atara Bio­ sphingosine 1-phosphate receptor modulator ozanimod in relapsing multiple therapeutics, Biogen, BMS-Celgene, EMD Serono, Sanofi Genzyme, sclerosis (RADIANCE): a randomised, placebo-controlled, phase 2 trial. Lancet Neurol. 15 (4), 373–381. Novartis, and Roche-Genentech. Cohen, J.A., Comi, G., Arnold, D.L., Bar-Or, A., Selmaj, K.W., Steinman, L., Havrdova, E. J.A. Cohen reports personal compensation for consulting for Ada­ K., Cree, B.A., Montalban, X., Hartung, H.P., Huang, V., Frohna, P., Skolnick, B.E., mas, Atara, Bristol Myers Squibb, Convelo, MedDay, and Mylan; and Kappos, L., 2019a. Efficacy and safety of ozanimod in multiple sclerosis: dose- serving as an Editor of Multiple Sclerosis Journal. blinded extension of a randomized phase II study. Mult. Scler. 25 (9), 1255–1262. Cohen, J.A., Comi, G., Selmaj, K.W., Bar-Or, A., Arnold, D.L., Steinman, L., Hartung, H.P., G. Comi reports compensation for consulting and/or speaking ac­ Montalban, X., Havrdova, E.K., Cree, B.A.C., Sheffield, J.K., Minton, N., tivities from Almirall, Biogen, Celgene, EXCEMED, Forward Pharma, Raghupathi, K., Huang, V., Kappos, L., 2019b. Safety and efficacy of ozanimod 9 K.W. Selmaj et al. Multiple Sclerosis and Related Disorders 51 (2021) 102844 versus interferon beta-1a in relapsing multiple sclerosis (RADIANCE): a multicentre, Mayzent [package insert]. 2020. Novartis Pharmaceuticals, East Hanover, NJ. randomised, 24-month, phase 3 trial. Lancet Neurol. 18 (11), 1021–1033. Olsson, T., Boster, A., Fernandez, O., Freedman, M.S., Pozzilli, C., Bach, D., Berkani, O., Comi, G., Kappos, L., Selmaj, K.W., Bar-Or, A., Arnold, D.L., Steinman, L., Hartung, H.P., Mueller, M.S., Sidorenko, T., Radue, E.W., Melanson, M., 2014. Oral ponesimod in Montalban, X., Havrdova, E.K., Cree, B.A.C., Sheffield, J.K., Minton, N., relapsing-remitting multiple sclerosis: a randomised phase II trial. J. Neurol. Raghupathi, K., Ding, N., Cohen, J.A., 2019. Safety and efficacy of ozanimod versus Neurosurg. Psychiatry 85 (11), 1198–1208. interferon beta-1a in relapsing multiple sclerosis (SUNBEAM): a multicentre, Persson, R., Lee, S., Minton, N., Niemcryk, S., Lindholm, A., Jick, S., 2020. Rate of randomised, minimum 12-month, phase 3 trial. Lancet Neurol. 18 (11), 1009–1020. nonmelanoma skin cancer in patients diagnosed with multiple sclerosis [abstract]. Cook, S., Leist, T., Comi, G., Montalban, X., Giovannoni, G., Nolting, A., Hicking, C., Eur. J. Neurol. 27 (Suppl 1), 944. Galazka, A., Sylvester, E., 2019. Safety of cladribine tablets in the treatment of Privett, J.T., Jeans, W.D., Roylance, J., 1976. The incidence and importance of renal patients with multiple sclerosis: an integrated analysis. Mult. Scler. Relat. Disord. 29, duplication. Clin. Radiol. 27 (4), 521–530. 157–167. Rossen, L.M., Ahrens, K.A., Branum, A.M., 2018. Trends in risk of pregnancy loss among Food and Drug Administration Guidance for Industry: Drug-Induced Liver Injury: US women, 1990-2011. Paediatr. Perinat. Epidemiol. 32 (1), 19–29. Premarketing Clinical Evaluation. 2009. (accessed June 22, 2020). G., Timony, G.A., Martinborough, E., Rosen, H., Roberts, E., Boehm, M.F., Peach, R. Gilenya [package insert], 2019. Novartis Pharmaceuticals Corporation. East Hanover, J., 2016. Ozanimod (RPC1063) is a potent sphingosine-1-phosphate receptor-1 NJ. (S1P1) and receptor-5 (S1P5) agonist with autoimmune disease-modifying activity. Kappos, L., Bar-Or, A., Cree, B.A.C., Fox, R.J., Giovannoni, G., Gold, R., Vermersch, P., Br. J. Pharmacol. 173 (11), 1778–1792. Arnold, D.L., Arnould, S., Scherz, T., Wolf, C., Wallstrom, E., Dahlke, F., 2018. Tran, J.Q., Hartung, J.P., Peach, R.J., Boehm, M.F., Rosen, H., Smith, H., Brooks, J.L., Siponimod versus placebo in secondary progressive multiple sclerosis (EXPAND): a Timony, G.A., Olson, A.D., Gujrathi, S., Frohna, P.A., 2017. Results from the first-in- double-blind, randomised, phase 3 study. Lancet 391 (10127), 1263–1273. human study with ozanimod, a novel, selective sphingosine-1-phosphate receptor Kappos, L., Cohen, J., Collins, W., de Vera, A., Zhang-Auberson, L., Ritter, S., von modulator. J. Clin. Pharmacol. 57 (8), 988–996. Rosenstiel, P., Francis, G., 2014. Fingolimod in relapsing multiple sclerosis: an Zeposia [package insert] 2020. Bristol Myers Squibb, Princeton, NJ, USA. integrated analysis of safety findings. Mult. Scler. Relat. Disord. 3 (4), 494–504. Zeposia [summary of product characteristics] 2020. Celgene Distribution B.V., Utrecht, Li, N., Zhang, F., 2016. Implication of sphingosin-1-phosphate in cardiovascular Netherlands. regulation. Front. Biosci. (Landmark Ed) 21, 1296–1313. 10