Request a Rep Icon
Green Bush Image Green Bush Image Green Bush Image Green Bush Image Green Bush Image Green Bush Image Green Bush Image
6 phase 2-3 clinical trials across UC and MS, ~38,000 patients in clinical trials and post marketing setting, and ~40,000 patient years total exposure in clinical trials and most marketing setting 6 phase 2-3 clinical trials across UC and MS, ~38,000 patients in clinical trials and post marketing setting, and ~40,000 patient years total exposure in clinical trials and most marketing setting

Overall ozanimod exposure in parent and extension trials (phase 1-3 MS, phase 2-3 UC) was 17,321.31 patient years (PY) and estimated to be 22,652 PY in the post-marketing setting. The cumulative number of patients exposed to ozanimod in parent and extension trials (all indications) was 3789 and estimated to be 34,910 in the post-marketing setting. All trials had a data cutoff of May 19, 2023, and all post-marketing data had a cutoff date of April 30, 2023.10,11

ZEPOSIA has been studied across multiple indications in 4 clinical trials, including TRUE NORTH (Ph 3); TOUCHSTONE (Ph 2); and SUNBEAM (Ph 3) and RADIANCE (Ph 3). 496 patients receiving the 0.92 mg orally once daily dose of ZEPOSIA during induction in TRUE NORTH or TOUCHSTONE and 882 patients receiving the 0.92 mg orally once daily dose of ZEPOSIA in SUNBEAM or RADIANCE were assessed in the safety analysis.1-4,6

In UC, from the start of the TOUCHSTONE phase 2 clinical trial (December 26, 2012) through TRUE NORTH OLE study data cutoff (January 10, 2022). In MS, from the start of the RADIANCE phase 2 clinical trial (September 18, 2012) through the DAYBREAK OLE database lock (April 7, 2023). Only includes patients receiving the 0.92 mg oral daily dose of ZEPOSIA.1,2,5,12,13

ZEPOSIA has been studied across multiple indications in 6 phase 2-3 clinical trials.1,2,5,12,13

Moderate-to-severe UC: TRUE NORTH (NCT02435992), a phase 3, multicenter, randomized, double-blind, placebo-controlled clinical trial; TRUE NORTH OLE (NCT02531126), an ongoing phase 3, multicenter, open-label extension trial; TOUCHSTONE (NCT01647516), a phase 2, multicenter, randomized, double-blind, placebo-controlled trial.2,6-8
Relapsing MS: SUNBEAM (NCT02294058) and RADIANCE (NCT02047734), phase 3, multicenter, randomized, double-blind, double-dummy, active treatment-controlled studies; DAYBREAK (NCT02576717), a phase 3, multicenter, open-label extension trial.1,3,4,9,13

MS=multiple sclerosis; OLE=open-label extension; Ph=phase; UC=ulcerative colitis.

Safety Comparable to Avonex in Overall Incidence of AEs, and Generally Similar Safety in Long-Term Extension Study;
Nearly 10 Years* of Experience1,13-15

In Pivotal Trials: Incidence of Adverse Reactions1,14,15
SUNBEAM (1 YEAR) RADIANCE (2 YEARS)
Summary of Adverse Reactions Avonex n=445 ZEPOSIA n=448 Avonex n=440 ZEPOSIA n=434
Overall incidence of adverse reactions 75.5% 59.8% 83.0% 74.7%
Severe adverse reactions 2.2% 1.6% 4.3% 3.5%
Serious adverse reactions 2.5% 2.9% 6.4% 6.5%
SUNBEAM (1 YEAR)
Summary of Adverse Reactions Avonex n=445 ZEPOSIA n=448
Overall incidence of adverse reactions 75.5% 59.8%
Severe adverse reactions 2.2% 1.6%
Serious adverse reactions 2.5% 2.9%
RADIANCE (2 YEARS)
Summary of Adverse Reactions Avonex n=440 ZEPOSIA n=434
Overall incidence of adverse reactions 83.0% 74.7%
Severe adverse reactions 4.3% 3.5%
Serious adverse reactions 6.4% 6.5%

Adverse Reactions With an Incidence of at Least 2% in Patients Treated With ZEPOSIA and at Least 1% Greater Than Avonex1,14,15a

SUNBEAM AND RADIANCE: POOLED DATA

Adverse Reactions Avonex n=885 ZEPOSIA n=882
Upper respiratory infectionb 23% 26%
Hepatic transaminase elevationc 5% 10%
Orthostatic hypotension 3% 4%
Urinary tract infection 3% 4%
Back pain 3% 4%
Hypertensiond 2% 4%
Upper abdominal pain 1% 2%

Adverse reactions are sorted by decreasing incidence in patients treated with ZEPOSIA. For adverse reactions pertaining to liver function tests, increases were transient and generally resolved without discontinuation.
Elevations of 3-fold the ULN or greater occurred in 5.5% of patients taking ZEPOSIA and in 3.1% of patients taking Avonex. The majority (79%) continued treatment with ZEPOSIA with values returning to less than 3 times the ULN within approximately 2 to 4 weeks.1,14

From the first patient randomized (October 18, 2012) through the DAYBREAK database lock (April 7, 2023), the maximum continuous exposure was 117.2 months. The mean exposure to ZEPOSIA 0.92 mg oral daily dose in the parent trials and DAYBREAK was 74.8 months.1,13

Data are not an adequate basis for comparison of rates between ZEPOSIA and the active control.1

Includes the following terms: nasopharyngitis, upper respiratory tract infection, pharyngitis, respiratory tract infection, bronchitis, rhinitis, respiratory tract infection viral, viral upper respiratory tract infection, rhinorrhea, tracheitis, and laryngitis.1

Includes the following terms: alanine aminotransferase increased, gamma-glutamyl transferase increased, aspartate aminotransferase increased, hepatic enzyme increased, liver function test abnormal, and transaminases increased.1

Includes hypertension, essential hypertension, and orthostatic hypertension.1

AE=adverse event; ULN=upper limit of normal.

DAYBREAK OLE: Incidence of Adverse Events13

Summary of TEAEs (PRIMARY ENDPOINT) ZEPOSIA (n=2494) IR/1000 PY
Any TEAE 89.0% 650.8
Severe TEAEs 9.6% 19.8
Serious TEAEs 15.3% 32.4
TEAEs leading to permanent treatment discontinuation 3.9% 7.7

TEAEs in ≥5% of Patients Treated With ZEPOSIA13

TEAEs ZEPOSIA (N=2494) IR/1000 PY
Nasopharyngitis 21.3% 49.6
Headache 17.1% 38.3
COVID-19 16.5% 33.5
Upper respiratory tract infection 12.4% 26.7
Lymphopeniae 10.3% 22.2
Back pain 9.6% 20.1
ALC decreasede 9.4% 20.0
Hypertensionf 9.2% 19.3
GGT increased 8.0% 16.7
Urinary tract infection 6.8% 13.9
Respiratory tract infection 6.6% 13.6
Arthralgia 6.5% 13.3
Bronchitis 6.3% 12.9
Depression-related TEAEsg 5.9% 12.0
Viral respiratory tract infection 5.8% 11.9
ALT increased 5.1% 10.2

Similar safety patterns were seen in the continuous once-daily oral ZEPOSIA 0.92 mg (n=881) population1,13

ALC reductions are an expected pharmacodynamic effect related to the mechanism of ozanimod; although investigators were not required to report ALC reductions as TEAEs, lymphopenia and ALC decreases were reported as TEAEs according to investigator determination.13

Includes preferred terms of hypertension, essential hypertension, labile hypertension, and systolic hypertension.13

Includes preferred terms of depression, depressed mood, and depressive symptoms.13

ALC=absolute lymphocyte count; ALT=alanine aminotransferase; GGT=gamma-glutamyl transferase; TEAE=treatment emergent adverse event.

Demonstrated Tolerability Profile With Low Discontinuation Rates Due to Adverse Events1,13-15

≥90% of Patients Stayed on Therapy Through Completion
of Pivotal Trials

SUNBEAM (1 Year)

94%

Remained on ZEPOSIA at 1 Year and 92% remained on Avonex1

 

RADIANCE (2 Years)

90%

Remained on ZEPOSIA at 2 Years and 85% remained on Avonex1

LOW DISCONTINUATION RATES
DUE TO AEs13-15

SUNBEAM (1 Year)

2.9%

For ZEPOSIA and 3.6% for Avonex14

 

RADIANCE (2 Years)

3.0%

For ZEPOSIA and 4.1% for Avonex15

DAYBREAK (Up to 7 Years) n=2494

3.9%

For ZEPOSIA in Long-Term Extension Study13

Rates of Serious Infections
and Malignancies Consistent
vs Avonex

The rate of serious infections at 1 year for ZEPOSIA was 1.1% vs 0.7% for Avonex, and the rate at 2 years for ZEPOSIA
was 0.9% vs 0.9% for Avonex. The rate of malignancies at 1 year for ZEPOSIA was 0.2% vs 0% for Avonex, and the rate at 2 years for ZEPOSIA was 0.9% vs 0.5% for Avonex.14,15

► Overall Infections

In SUNBEAM and RADIANCE, the overall rate of infections with ZEPOSIA (35%) was similar to Avonex (34%). ZEPOSIA causes a reduction in peripheral blood lymphocyte count and may increase the risk of infection.1

Controlled Lymphocyte Reductions

ALC was consistently maintained near the lower limit of normal across both pivotal trials, and the mean ALC for both SUNBEAM and RADIANCE was ≈0.8 × 109/L.14-16

VIEW ALC DATA

Herpetic Infections: In active-controlled MS trials, herpes zoster was reported as an adverse reaction in 0.6% of patients treated with once-daily oral ZEPOSIA 0.92 mg and in 0.2% of patients taking Avonex.1

ZEPOSIA Consistently Maintained ALC Near the Lower Limit of Normal14-16

ZEPOSIA® consistently maintained ALC near the lower limit of normal

Lymphocyte numbers can be restored to normal values by discontinuing therapy1,14,15,17

  • After discontinuing daily oral ZEPOSIA 0.92 mg, the median time for peripheral blood lymphocytes to return to the normal range was 30 days, with approximately 90% of patients in the normal range within 3 months
  • Mean ALC was approximately 0.8 x 109 cells/L for both SUNBEAM and RADIANCE (at 1 year and 2 years, respectively)
  • ZEPOSIA causes a mean reduction in peripheral blood lymphocyte count to approximately 45% of baseline values because of reversible sequestration of lymphocytes in lymphoid tissues; ZEPOSIA may therefore increase the susceptibility to infections, some serious in nature

BL=baseline; LLN=lower limit of normal.

Herpetic Infections: In active-controlled MS trials, herpes zoster was reported as an adverse reaction in 0.6% of patients treated with once-daily oral ZEPOSIA 0.92 mg and in 0.2% of patients taking Avonex.1

Get your patients
started on ZEPOSIA
Patient
Start Form
Enroll your patients in ZEPOSIA 360 Support
and help them get started on treatment.
BMS eSign
Did your patient forget to sign the ZEPOSIA Start Form? Send them a link to the form below to sign electronically via CoverMyMeds.
Explore the efficacy data
ZEPOSIA dosing & initiation

This website is best viewed
using the horizontal display on
your tablet device.

This website is best viewed
using the vertical display on
your mobile device.