Sphingosine 1-phosphate receptor modulators in multiple sclerosis treatment: A practical review
Abstract
Four sphingosine 1-phosphate (S1P) receptor modulators (fingolimod, ozanimod, ponesimod, and siponimod) are approved by the US Food and Drug Administration for the treatment of multiple sclerosis. This review summarizes efficacy and safety data on these S1P receptor modulators, with an emphasis on similarities and differences. Efficacy data from the pivotal clinical trials are generally similar for the four agents. However, because no head-to-head clinical studies were conducted, direct efficacy comparisons cannot be made. Based on the adverse event profile of S1P receptor modulators, continued and regular monitoring of patients during treatment will be instructive. Notably, the authors recommend paying attention to the cardiac monitoring guidelines for these drugs, and when indicated screening for macular edema and cutaneous malignancies before starting treatment. To obtain the best outcome, clinicians should choose the drug based on disease type, history, and concomitant medications for each patient. Real-world data should help to determine whether there are meaningful differences in efficacy or side effects between these agents.
Background
Sphingosine 1-phosphate (S1P) receptor modulators are G protein-coupled receptors found throughout the body. They mediate a broad range of functions via five distinct subtypes (S1P1–5).1, 2 Owing to the widespread expression of these receptor subtypes across multiple organ systems, they can influence the immune system, brain, lung, liver, heart, and vasculature (Fig. S1).1, 2 Considering the extensive presence of S1P receptors on cardiomyocytes and vascular endothelial cells, S1P receptor modulators are expected to have cardiovascular effects.3 These were observed in preclinical studies, particularly for S1P receptor modulators that target S1P1 and S1P3.4
Owing to their ability to bind with high affinity to one or more S1P receptor subtypes, S1P receptor modulators have a complex mechanism of action with both peripheral immunological and central nervous system (CNS) effects.1, 5 Consequently, S1P receptor modulators represent a therapeutic option for immune-mediated diseases, including multiple sclerosis (MS).1, 2 To date, four S1P receptor modulators, fingolimod, ozanimod, ponesimod, and siponimod, are approved by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) for the treatment of MS (Table 1).6-13 S1P receptor modulators function by blocking the capacity of lymphocytes to egress from lymph nodes, leading to a reduction in the number of lymphocytes in peripheral blood.14 While the full mechanistic pathway by which S1P receptor modulators exert therapeutic effects is unknown, lymphocyte sequestration and reduction of lymphocyte migration into the CNS were proposed as the primary mechanism of action in MS.6-9
| Fingolimod6 S1P1, S1P3, S1P4, and S1P5 | Ozanimod7 S1P1 and S1P5 | Ponesimod8 S1P1 | Siponimod9 S1P1 and S1P5 | |
|---|---|---|---|---|
| Indication(s) | Relapsing forms of MS, including CIS, RRMS, and active SPMS, in patients ≥10 years of age |
|
Relapsing forms of MS, including CIS, RRMS, and active SPMS, in adults | Relapsing forms of MS, including CIS, RRMS, and active SPMS, in adults |
| Recommended dosing |
|
|
|
|
| Pharmacokinetics | ||||
| Tmax | 12–16 h | 6–8 h CC112273 and CC1084037 (active metabolites): 10 and 16 h | 2–4 h | 3–8 h |
| Half-life | 6–9 days |
|
33 h | ~30 h |
| Time to reach steady state | 1–2 months |
|
3 days of maintenance dose | 6 days |
| Contraindications |
|
|
|
|
- AV, atrioventricular; CIS, clinically isolated syndrome; h, hour; MAO, monoamine oxidase; MS, multiple sclerosis; QD, once daily; RRMS, relapsing–remitting multiple sclerosis; S1P1, sphingosine 1-phosphate receptor subtype 1; S1P3, sphingosine 1-phosphate receptor subtype 3; S1P4, sphingosine 1-phosphate receptor subtype 4; S1P5, sphingosine 1-phosphate receptor subtype 5; SPMS, secondary progressive MS; Tmax time of peak plasma concentration; TIA; transient ischemic attack; UC, ulcerative colitis.
Fingolimod, a nonselective S1P receptor modulator targeting S1P1, S1P3, S1P4, and S1P5, was approved in the US in 2010 and the EU in 2011 as the first oral capsule for the treatment of MS.10, 15, 16 Fingolimod, the only S1P receptor modulator studied in children with MS, is indicated for patients at least 10 years of age with relapsing MS (RMS), including clinically isolated syndrome (CIS, US only), relapsing–remitting MS (RRMS), and active secondary progressive MS (SPMS).6 Fingolimod is a prodrug that requires phosphorylation to be converted to its active form, which is thought to contribute to its relatively long elimination half-life.2 Safety concerns, particularly the cardiac-related effects (e.g., bradycardia) associated with fingolimod led to the development of more selective, second-generation S1P receptor modulators: ozanimod, ponesimod, and siponimod.17 The FDA recently approved an orally disintegrating fingolimod tablet for the treatment of adult patients with RMS, which provides faster onset and improved bioavailability than the oral capsule formulation.18
The second generation of S1P receptor modulators do not require phosphorylation for activity and also have shorter half-lives, implying a shorter washout period.15 The exception is ozanimod, which has active metabolites with a longer mean elimination half-life than fingolimod (Table 1).19 Siponimod and ozanimod, which are selective for S1P1 and S1P5, received FDA approval in 2019 and 2020, respectively, for the treatment of adults with RMS, including CIS, RRMS, and active SPMS.7, 9 The EMA approved siponimod in 2020 for the treatment of SPMS and ozanimod in 2020 for the treatment of active RRMS.11, 12 Additionally, in 2021, ozanimod was the first in this class to receive FDA approval for moderately to severely active ulcerative colitis.7, 20 Ponesimod is selective for S1P1, and received FDA approval in 2021 for the treatment of adults with RMS, including CIS, RRMS, and active SPMS, and EMA approval in 2021 for the treatment of adults with active RMS.8, 13
S1P receptor modulators are of particular interest in the MS treatment landscape because they may have direct CNS effects and are able to cross the blood–brain barrier, whereas several other disease-modifying therapies (DMTs) (e.g., interferons, glatiramer acetate, and monoclonal antibodies) function only in the periphery.21, 22 A direct CNS effect, which is supported by preclinical evidence, may contribute to clinical benefits in MS, because S1P receptors are expressed by cells in the CNS.17 This direct mechanism of action potentially provides a benefit for S1P receptor modulators in the treatment of MS; however, clinical studies have not substantiated these findings. Notably, clinical studies clearly demonstrated reductions in brain volume loss, as is common with DMTs and particularly at early stages of disease,23 suggesting use of these agents early in the course of MS rather than waiting until progression is clinically evident.24
Considering the plethora of recent and ongoing research on S1P receptor modulators, this review aimed to summarize the data generated to date on the efficacy and safety of S1P receptor modulators for the treatment of MS, with an emphasis on detailing the similarities and differences of each agent.
Efficacy of S1P Receptor Modulators in Multiple Sclerosis
The efficacies of fingolimod, ozanimod, ponesimod, and siponimod were evaluated in phase 2 and phase 3 clinical trials. Efficacy data from the pivotal phase three trials for the key endpoints listed in the respective prescribing information for each drug are summarized in Table 2. The safety and efficacy of fingolimod was assessed in adults with RRMS (FREEDOMS,25 FREEDOMS II,30 and TRANSFORMS31) and in pediatric participants with RMS (PARADIGMS32); siponimod in adults with SPMS (EXPAND29); ozanimod in adults with RMS (SUNBEAM26 and RADIANCE27); and ponesimod in adults with RMS (OPTIMUM).28 The safety and efficacy of fingolimod has also been assessed in adults with primary progressive MS (INFORMS).33 However, in this phase 3 trial, fingolimod did not slow disease progression compared with placebo,33 suggesting that CNS penetration of fingolimod alone was inadequate to alter the course of progression in primary progressive MS. The results of this study are not further discussed in this review.
| Fingolimod6, 25 S1P1, S1P3, S1P4, and S1P5 | Ozanimod7, 26, 27 S1P1 and S1P5 | Ponesimod8, 28 S1P1 | Siponimod9, 29 S1P1 and S1P5 | |
|---|---|---|---|---|
| Comparator |
PBO (FREEDOMS, 2 years) IFN β-1a (TRANSFORMS, 1 year) |
IFN β-1a (SUNBEAM, ≥ 1 year; RADIANCE, 2 years) | Teriflunomide (OPTIMUM, 108 weeks) | PBO (EXPAND, 21-month median study duration) |
| Relapse | ||||
| ARR |
0.18 vs. 0.40 (PBO), p < 0.001 0.16 vs. 0.33 (IFN β-1a), p < 0.001 |
0.18 vs. 0.35 (IFN β-1a); relative reduction 48%, p < 0.0001 0.17 vs. 0.28 (IFN β-1a); relative reduction 38%, p < 0.0001 |
0.20 vs. 0.29 (teriflunomide); relative reduction 31%, p = 0.0003 | 0.07 vs. 0.16 (PBO); relative reduction 55%, p < 0.01 |
| Percentage of patients without relapse |
70% vs. 46% (PBO), p < 0.001 83% vs. 70% (IFN β-1a), p < 0.001 |
78% vs. 66% (IFN β-1a) 76% vs. 64% (IFN β-1a) |
71% vs. 61% (teriflunomide) | 89% vs. 81% (PBO) |
| Confirmed disability progression (CDP) | ||||
| 3-month CDP | 18% vs. 24% (PBO), p = 0.02 | 8% vs. 8% (IFN β-1a), NS (SUNBEAM and RADIANCE combined) | 11% vs. 13% (teriflunomide), NS | 26% vs. 32% (PBO), p = 0.013 |
| 3-month CDP hazard ratio |
0.70 (95% CI 0.52, 0.96), p = 0.02 0.71 (95% CI 0.42, 1.21), p = 0.21 |
0.95 (p = 0.77) (SUNBEAM and RADIANCE combined) | 0.83, p = 0.29 | 0.79 (95% CI 0.65, 0.95), p = 0.013 |
| MRI outcomes | ||||
| Mean (median) number of GdE lesions |
0.2 (0) vs. 1.1 (0) (PBO), p < 0.001 0.2 (0) vs. 0.5 (0) (IFN β-1a), p < 0.001 |
0.2 vs. 0.4 (IFN β-1a); relative reduction 63%, p < 0.0001 0.2 vs. 0.4 (IFN β-1a); relative reduction 53%, p = 0.0006 |
0.2 vs. 0.4 (teriflunomide); relative reduction 59%, p < 0.0001 | 0.1 vs. 0.6 (PBO); relative reduction 86%, p < 0.0001 |
| Mean number of new/enlarging T2 lesions |
2.5 (0) vs. 9.8 (5.0) (PBO), p < 0.001 1.6 (0) vs. 2.6 (1.0) (IFN β-1a), p = 0.002 |
1.5 vs. 2.8 (IFN β-1a); relative reduction 48%, p < 0.0001 1.8 vs. 3.2 (IFN β-1a); relative reduction 42%, p < 0.0001 |
1.4 vs. 3.2 (teriflunomide); relative reduction 56%, p < 0.0001 | 0.7 vs. 3.6 (PBO); relative reduction 81%, p < 0.0001 |
- ARR, annualized relapse rate; CDP, confirmed disability progression; CI, confidence interval; GdE, gadolinium-enhancing; IFN, interferon; MRI, magnetic resonance imaging; MS, multiple sclerosis; NS, not significant; PBO, placebo; S1P1, sphingosine 1-phosphate receptor subtype 1; S1P3, sphingosine 1-phosphate receptor subtype 3; S1P4, sphingosine 1-phosphate receptor subtype 4; S1P5, sphingosine 1-phosphate receptor subtype 5.
In the pivotal phase 3 trials, treatment with fingolimod25 or siponimod29 resulted in statistically significantly lower annualized relapse rates (ARR) compared with placebo, while fingolimod,31 ozanimod,26, 27 and ponesimod28 showed statistically significantly lower ARR compared with active comparators. The proportion of patients without relapse was higher with fingolimod compared with placebo and intramuscular interferon beta-1a (IFN β-1a),25, 31 ozanimod compared with IFN β-1a,26, 27 ponesimod compared with teriflunomide,28 and siponimod compared with placebo.29 Treatment with fingolimod25 and siponimod29 resulted in a significantly lower proportion of patients with 3-month confirmed disability progression (CDP) compared with placebo. The mean numbers of gadolinium-enhancing and new/enlarging T2 lesions were significantly reduced with fingolimod compared with placebo and IFN β-1a,25, 31 ozanimod compared with IFN β-1a,26, 27 ponesimod relative to teriflunomide,28 and siponimod compared with placebo.29
In terms of durability of efficacy, the response to S1P receptor modulator treatment appears to be sustained long term. In LONGTERMS, an open-label, phase 3b extension study, sustained efficacy was reported with fingolimod over 10 years in more than 4000 participants with MS who had participated in the core trials.34 Similarly, data from the open-label extension DAYBREAK study, ponesimod phase 2 extension study, and phase 3 EXPAND study confirmed maintenance of efficacy of ozanimod for up to 8 continuous years of exposure, ponesimod for up to 8 years, and siponimod for more than 5 years, respectively.35-37
Overall, while the efficacy data between the pivotal clinical trials of fingolimod, ozanimod, ponesimod, and siponimod are generally comparable, head-to-head clinical trials were not conducted; therefore, definitive conclusions comparing the efficacy of S1P receptor modulators cannot be made.2
Safety of S1P Receptor Modulators in Multiple Sclerosis
There are no direct comparative trials between S1P receptor modulators, and the currently available safety information for agents is from their individual trials; this should be considered when comparing event rates between agents.
Adverse events
The varying effects of S1P receptor modulators on different receptor subtypes (S1P1, S1P3, S1P4, and S1P5) may increase the risk of certain AEs, because these receptors are expressed on cells throughout the body.15 Therefore, it is important for clinicians to consider specific events known to be associated with a particular S1P receptor modulator when considering a treatment approach. Notably, pretreatment screening and continued follow-up can minimize risks, with the overall safety profile of this drug class remaining favorable.38 The most relevant AEs encountered with S1P receptor modulators include cardiovascular AEs, lymphopenia, serious infections, malignancy (including cutaneous malignancies), hepatotoxicity, seizures (mainly in children), and pulmonary effects (Table 3).42
| Fingolimod6, 25, 39 S1P1, S1P3, S1P4, and S1P5 | Ozanimod7, 26, 27, 40 S1P1 and S1P5 | Ponesimod8, 28 S1P1 | Siponimod9, 29 S1P1 and S1P5 | |
|---|---|---|---|---|
| Cardiac and vascular | ||||
| HR reduction from baseline | 8–10 bpm | 1.2 bpm | 6 bpm | 5–6 bpm |
| HR < 40 bpm | Rare | Not observed | Rare (3 patients) | Rare |
| Bradycardia | 0.6% vs. 0.1% (PBO) symptomatic following first dose (hypotension, dizziness, fatigue, palpitations, and/or chest pain that usually resolved within the first 24 h on treatment) |
First dose: 0.6% vs. 0.0% (IFN β-1a) After day 1: 0.8% vs. 0.7% (IFN β-1a) |
Treatment initiation: 5.8% vs. 1.6% (teriflunomide) | 4.4% vs. 2.9% (PBO); generally asymptomatic; symptomatic patients reported dizziness or fatigue that resolved within 24 h without treatment |
| AV blocks | Conduction abnormalities were generally transient, asymptomatic, and resolved within the first 24 h | There were no cases of Mobitz type 2 second- or third-degree AV blocks in patients who underwent dose titration | Conduction abnormalities were generally transient, asymptomatic, and resolved within 24 h | Conduction abnormalities were generally transient, asymptomatic, and resolved within 24 h |
| Patients with first-degree AV block | 4.7% vs. 1.6% (PBO) | 0.6% vs. 0.2% (IFN β-1a) | 3.4% vs. 1.2% (teriflunomide) | 5.1% vs. 1.9% (PBO) |
| Patients with second-degree AV block or higher |
Second-degree: 4% vs. 2% (PBO) Third-degree occurred in postmarketing |
No report | No report | <1.7% (second-degree) |
| QTc prolongation | 14.0 msec (1.25 and 2.5 mg doses) | No report | 11.8 msec (40 mg) | 7.8 msec (2 mg dose) |
| Hypertension | 8% vs. 4% (PBO); Compared with PBO, patients treated with fingolimod 0.5 mg had a mean increase in SBP of 3 mm Hg and in DBP of 2 mm Hg developing after 1 month of treatment and persisting with continued treatment | 3.9% vs. 2.1% (IFN β-1a); Compared with IFN β-1a, ozanimod treatment led to an increase of 1–2 mm Hg in SBP, with no change in DBP, developing after 3 months of treatment and persisting with continued treatment | 10.1% vs. 9.0% (teriflunomide); Ponesimod increased SBP by a mean of 2.9 mm Hg and DBP by 2.8 mm Hg (compared with 2.8 mm Hg and 3.1 mm Hg in teriflunomide, respectively). Increases were first seen after 1 month of treatment and persisting with continued treatment | 12.5% vs. 9.2% (PBO); Compared with PBO, patients treated with siponimod had an average increase in SBP of 3 mm Hg and DBP of 1.2 mm Hg developing after 1 month of treatment and persisting with continued treatment |
| Macular edema | 0.5% (0.5 mg) vs. 0.4% (PBO) | 0.3% vs. 0.3% (IFN β-1a) | 1.1% vs. 0% (teriflunomide) | 1.8% vs. 0.2% (PBO) |
| Lymphopenia | ||||
| Lymphocyte reduction from baseline | 70% | 57% | 60%–70% | 70%–80% |
| Approximate ALC nadir | 500 cells/μL | 759 cells/μL | 650 cells/μL | 560 cells/μL |
| Lymphocyte recovery to normal range | 1–2 months | 30 days | 1–2 weeks | 10 days |
| Infections | ||||
| Common infections | Bronchitis, herpes zoster, influenza, sinusitis, and pneumonia were more common with fingolimod | Ozanimod increased the risk of viral upper respiratory tract infections, urinary tract infections, and herpes infections | Ponesimod increased the risk of upper respiratory tract infections; Similar rates of herpes zoster were reported in ponesimod and teriflunomide (4.8%) | Herpes zoster, herpes infections, bronchitis, sinusitis, upper respiratory infection, and fungal skin infection were more common with siponimod |
| Overall infections | 72% (similar to PBO) | 35% vs. 34% (IFN β-1a) | 54.2% vs. 52.1% (teriflunomide) | 49.0% vs. 49.1% (PBO) |
| Herpetic infections | 9% (0.5 mg) vs. 7% (PBO) | 0.6% vs. 0.2% (IFN β-1a) (only includes herpes zoster) | 4.8% vs. 4.8% (teriflunomide) | 4.6% vs. 3.0% (PBO) |
| Serious infections | 2.3% (0.5 mg) vs. 1.6% (PBO) | 1% vs. 0.8% (IFN β-1a) | 1.6% vs. 0.9% (teriflunomide) | 2.9% vs. 2.5% (PBO) |
| Progressive multifocal leukoencephalopathy (PML)a |
61 PML cases Incidence rate: 5.88 per 100,000 PY (August 2022) |
1 PML case ~40,000 PY of exposure |
No cases reported |
3 PML cases ~30,000 PY of exposure (2022) |
| Cryptococcal meningitis (CM) infections | 74 CM cases (9 per 100,000 PY) | No cases reported | No cases reported | 1 CM case (>7236 PY) |
| Malignancies |
Basal cell carcinoma rate: 2% vs. 1% (PBO) (FREEDOMS trials) Postmarket reports of melanoma, squamous cell carcinoma, Merkel cell carcinoma, and cases of lymphoma (T cell, B cell, CNS) |
Melanoma, basal cell carcinoma, breast cancer, seminoma, cervical carcinoma, and adenocarcinomas, including rectal adenocarcinoma, were reported in controlled trials of MS and UC | Incidence of basal cell carcinoma was 0.4% (ponesimod) vs. 0.2% (teriflunomide); cases of other cutaneous malignancies, including melanoma, have been reported | Increased risk of basal cell (1.1%) and squamous cell carcinoma (0.2%) with siponimod treatment (EXPAND); cases of other cutaneous malignancies have been reported |
| Hepatic | ||||
| Hepatotoxicity |
Clinically significant liver injury has occurred in the postmarket setting; cases of acute liver failure requiring liver transplant have been reported Levels of serum transaminase returned to normal within ~2 months following discontinuation |
79% of patients with ALT elevations continued ozanimod treatment, with values decreasing within 2–4 weeks | The median time to 3x ULN of ALT was 3 months; the majority (89%) continued treatment and levels decreased within 2–4 weeks | Elevations in transaminases and bilirubin were reported in 10.1% (siponimod) vs. 3.7% (PBO). ALT returned to normal ~1 month after discontinuation |
| ALT ≥3 × ULN | 14% vs. 3% (PBO) | 5.5% vs. 3.1% (IFN β-1a) | 17.3% vs. 8.3% (teriflunomide) | 5.6% vs. 1.5% (PBO) |
| ALT ≥5 × ULN | 4.5% vs. 1% (PBO); most within 6–9 months of treatment initiation | 1.6% vs. 1.3% (IFN β-1a) | 4.6% vs. 2.5% (teriflunomide) | 1.4% vs. 0.5% (PBO); the majority occurred within 6 months of treatment initiation |
| Pulmonary |
Reductions in FEV1 and DLCO The decrease from baseline in the percent of predicted values for FEV1 was 2.8% with fingolimod 0.5 mg and 1.0% for PBO; for DLCO, 3.3% and 0.5%, respectively Changes in FEV1 appear to be reversible after treatment discontinuation |
Dose-dependent reductions in FEV1 3 months after start of treatment In a pooled analysis, decline from baseline with ozanimod compared to IFN β-1a was 60 mL (95% CI: −100, −20) at 12 months (mean difference 1.9%) Not enough evidence to determine reversibility of this effect |
Dose-dependent reductions in FEV1 and DLCO were observed mostly in the first month of treatment In OPTIMUM, the decrease from baseline in FEV1 at 2 years was 8.3% vs. 4.4% (teriflunomide) |
Dose-dependent reductions in absolute FEV1 after 3 months of treatment Decline in absolute FEV1 from baseline with siponimod vs. placebo was 88 mL at 2 years (mean difference, 2.8%) There is not enough evidence to determine the reversibility Changes in patients with MS who also had mild to moderate asthma and chronic obstructive pulmonary disease were similar to the overall population |
| Neurological | ||||
| Seizureb | 0.9% (0.5 mg) vs. 0.3% (PBO) | NA | 1.4% vs. 0.2% (teriflunomide) | 1.7% vs. 0.4% (PBO) |
| AEs leading to discontinuation | FREEDOMS: 7.5% (fingolimod 0.5 mg) vs. 7.7% (PBO) |
RADIANCE: 3.0% (1 mg) vs. 4.1% (IFN β-1a) SUNBEAM: 2.9% (1 mg) vs. 3.6% (IFN β-1a) |
8.7% vs. 6.0% (teriflunomide) | 8.5% vs. 5.1% (PBO) |
- AE, adverse event; ALC, absolute lymphocyte count; ALT, alanine transaminase; AV, atrioventricular; bpm, beats per minute; CM, cryptococcal meningitis; CNS, central nervous system; DPB, diastolic blood pressure; DLCO, diffusion lung capacity for carbon monoxide; FEV1, forced expiratory volume; h, hour; HR, heart rate; MS, multiple sclerosis; NA, not available; PBO, placebo; PML, progressive multifocal leukoencephalopathy; PY, person-years; UC, ulcerative colitis; VZV, varicella zoster virus; QTc, corrected QT interval; ULN, upper limit of normal; S1P1, sphingosine 1-phosphate receptor subtype 1; S1P3, sphingosine 1-phosphate receptor subtype 3; S1P4, sphingosine 1-phosphate receptor subtype 4; S1P5, sphingosine 1-phosphate receptor subtype 5; SBP, systolic blood pressure.
- a PML is an opportunistic viral infection of the brain caused by the JC virus that typically occurs in patients who are immunocompromised and that usually leads to death or severe disability.
- b Mainly in children.41
Each S1P receptor modulator is associated with a transient reduction in heart rate upon initiation, that is highest with fingolimod (decrease of 8.1 beats per minute [bpm]), followed by ponesimod (decrease of 6 bpm), siponimod (decrease of 5–6 bpm), and ozanimod (decrease of 1.2 bpm).6-9, 43 Meta-analysis of S1P receptor modulator clinical trials showed that cardiovascular AEs occurred in 10.9% in patients treated with S1P receptor modulators compared with 4.8% in those receiving the control treatment, with a significantly increased risk of arrhythmia (primarily bradyarrhythmia) and hypertension in patients treated with S1P receptor modulators compared with control treatment.3 Fingolimod and siponimod are associated with greater rates of first-degree AV block than placebo,6, 9 whereas ozanimod and ponesimod are associated with greater rates of bradycardia and first-degree AV block than IFN β-1a or teriflunomide, respectively.7, 8 Second- or third-degree AV block was reported with fingolimod and siponimod6, 9 but not with ozanimod and ponesimod.7, 8 Fingolimod, ponesimod, and siponimod are associated with prolongation of QTc intervals,6, 8, 9 whereas ozanimod is not.7
While S1P receptor modulators were associated with an increased risk of macular edema,42 there was no increase in these events with ozanimod compared with IFN β-1a in pivotal trials.26, 27 Macular edema was reported in 0.3% of patients treated with ozanimod and 0.3% of patients treated with IFN β-1a.7 Fingolimod increases the risk of macular edema in a dose-dependent manner,6 and in clinical trials, macular edema generally occurred in the first 3–4 months of treatment and partially or completely resolved after fingolimod discontinuation.42 Macular edema was reported in 1.1% and 1.8% of patients treated with ponesimod and siponimod, respectively, and in 0% and 0.2% of patients receiving teriflunomide or placebo, respectively.8, 9 Notably, patients with a history of uveitis, diabetes mellitus, and cataract surgery are at increased risk of macular edema during treatment with an S1P receptor modulator.42, 44
S1P receptor modulators reduce the capacity of lymphocytes to egress from lymph nodes, thereby reducing the number of lymphocytes in peripheral blood, which in turn can lower the ability of the immune system to fight infections.2, 45 Reductions in lymphocyte counts from baseline ranged from 57% with ozanimod to ~70% with fingolimod, siponimod, and ponesimod, with a nadir of ~760 cells/μL with ozanimod, 500 cells/μL with fingolimod, 560 cells/μL with siponimod, and 650 cells/μL with ponesimod (Table 3).6, 8, 9, 26, 27 Fastest lymphocyte recovery to normal range was 10 days with siponimod and 1–2 weeks with ponesimod, while it was 1 month with ozanimod and 1–2 months with fingolimod (Table 3).6-9 All four S1P receptor modulators are associated with a risk of serious infections. In fingolimod trials, two patients died from herpetic infections, one due to disseminated primary herpes zoster and one to herpes simplex encephalitis. Both patients received high-dose corticosteroids to treat suspected MS relapses.6 Other serious infections that occurred during S1P receptor modulator treatment include progressive multifocal leukoencephalopathy and cryptococcal meningitis (Table 3).
Basal cell carcinoma and other skin malignancies were reported as AEs for all four S1P receptor modulators.6-9 Lymphoma, particularly non-Hodgkin's lymphoma, has also been associated with fingolimod use, and a recent analysis noted a small increase in the overall risk of invasive cancer with fingolimod.6, 46 With respect to liver function, alanine and aspartate transaminase elevations are common across all four S1P receptor modulators.6-9 For ozanimod and ponesimod, these elevations typically resolved within 2–4 weeks of continued treatment,7, 8 while for siponimod and fingolimod, these elevations returned to normal 1–2 months after treatment discontinuation.6, 9
Pulmonary effects were observed in the first 1–3 months of treatment with all S1P receptor modulators, with dose-dependent reductions in forced expiratory volume.6-9 Routine monitoring with pulmonary function tests is not recommended for any S1P receptor modulator. Spirometry and diffusion lung capacity of carbon monoxide evaluation should be performed when clinically indicated.
No significant new safety concerns were reported in long-term studies, including for over 14 years for fingolimod, up to 8 years of continuous ozanimod exposure, 8 years for ponesimod, and >5 years for siponimod.34-37 AEs most commonly reported with the four S1P receptor modulators in long-term follow-up studies were viral upper respiratory tract infection, headache, hypertension, lymphopenia, and nasopharyngitis.34-37 However, instances of disease rebound were reported following discontinuation of fingolimod.6, 47 In most of these cases, patients did not return to the functional status they had before stopping fingolimod treatment.6 Symptoms generally occurred within 12 weeks but were reported for up to 24 weeks after discontinuation.6 One such condition, relapse with tumefactive demyelinating lesions, was observed within the initial 9 months following treatment initiation as well as 6 months after treatment discontinuation.6, 48 Due to the association of fingolimod with tumefactive MS, the condition should be considered when a severe MS relapse occurs during fingolimod treatment, especially during initiation, or after discontinuation of treatment, prompting imaging evaluation and subsequent appropriate treatment initiation.6, 48 To our knowledge, no cases of severe disease rebound were reported following the discontinuation of ozanimod, ponesimod, or siponimod.8, 9, 49 Mild to moderate relapses occurred in 2.3% of patients with RMS following ozanimod discontinuation, and 70% of those experienced a complete recovery within 30 days of onset.50
Safety and tolerability during pregnancy
Animal studies demonstrated the potential for fetal harm with S1P receptor modulators in the absence of maternal toxicity.6-9 S1P receptor modulators and/or their metabolites were also detected in the milk of treated lactating rats.51 Dedicated human pregnancy studies were not conducted.
Pooled data from clinical trials of fingolimod reported 66 pregnancies with in utero exposure (the total number of pregnancies including exposure to fingolimod, placebo, or IFN β-1a, was 89).52 Among these, there were 28 live births, 9 spontaneous abortions, 24 elective abortions, and 5 that were ongoing or lost to follow up at the time.52 In the postmarketing fingolimod pregnancy registry, 6.0% of reported pregnancies resulted in major congenital malformations in live births; 6.6% in major congenital malformations in live births, stillbirths, and pregnancies that were terminated due to fetal anomalies.53 In the general population, the incidence of congenital malformations in live births is around 3% and the incidence of spontaneous abortion is estimated to range from 12% to 20%.54-56 The ozanimod trials reported 78 pregnancies, including 42 live births, 1 report of duplex kidney, 12 spontaneous abortions, and 15 elective abortions.57 Over the course of the ponesimod clinical program (which included the 30-day period following treatment discontinuation), a total of 19 pregnancies were reported.58 Among these, six resulted in live, normal births; three were spontaneous abortions, and there were eight elective abortions. Of the 12 abortion cases, one reported benign hydatidiform mole, while fetal abnormalities were not known/reported for the other cases.59 As of March 2022, 31 pregnancies were reported in women treated with siponimod, with no events of congenital malformation or fetal/maternal complications.60
Safety outcomes with concomitant SARS-CoV-2 infection
Considering the reduction in lymphocyte count observed with S1P receptor modulator treatment, it was initially thought that patients with MS undergoing treatment may be more susceptible to poor outcomes from COVID-19.61 However, severe disease and mortality in patients treated with all four S1P receptor modulators were comparable to the general population.35, 62, 63
Some studies demonstrated that patients with MS treated with S1P receptor modulators, particularly fingolimod, may have limited ability to develop a humoral immune response after COVID-19 vaccination.64-68 Fingolimod-, ponesimod-, or siponimod-treated patients were seropositive after second and third mRNA vaccinations, but with lower immunoglobulin G levels compared with untreated patients.68 Following a second mRNA vaccine, 25% of fingolimod patients had a T-cell response. Nearly all vaccinated participants with serological data in the ozanimod DAYBREAK open-label extension study mounted a serologic response after full COVID-19 vaccination.69 A literature review found that fingolimod treatment was associated with blunted antibody vaccine responses (seroconversion rate of 28%–77%) compared with untreated individuals, while seroconversion rates seen in the majority of people treated with siponimod, ozanimod, and ponesimod are higher (71%–100%).63 Differences in (COVID-19) vaccine responses between fingolimod and newer S1P receptor modulators may be due to differences in S1P receptor subtype modulation.63 Fingolimod targets S1P1, S1P3, S1P4, and S1P5, while ponesimod, siponimod, and ozanimod target S1P1, and siponimod and ozanimod also target S1P5.63 The stronger antibody vaccine responses of ponesimod, siponimod, and ozanimod may be due to their limited impact on S1P3 and S1P4 and their effect on the formation and function of the germinal center, important in generating neoantigen antibody responses.63 While, reduced T-cell responses have been reported with fingolimod, data on the peripheral blood T-cell responses in patients treated with second generation S1P modulators have been inconsistent.63
Clinical considerations when prescribing S1P receptor modulators to patients with multiple sclerosis
Clinical considerations for patients with MS who are candidates for S1P receptor modulator therapy are listed in Table 4.
| Fingolimod6 S1P1, S1P3, S1P4, and S1P5 | Ozanimod7 S1P1 and S1P5 | Ponesimod8 S1P1 | Siponimod9 S1P1 and S1P5 | |
|---|---|---|---|---|
| Lymphocyte dynamics | ||||
| Maximum lymphocyte reduction from baseline | 70%–80% | 55% | 60%–70% | 70%–80% |
| Time to lymphocyte restoration after discontinuation | 4–8 weeks | Median of 30 days with 90% in normal range within 3 months | Within 1 week for 90% of patients | 10 days for most patients (90%) but can take up to 3–4 weeks |
| Pretreatment assessments | ||||
| Genetic | NA | NA | NA | Genotype for CYP2C9 |
| Cardiac | ECG | ECG | ECG | ECG |
| Hepatic | Transaminase and bilirubin from the past 6 months | Transaminase and bilirubin from the past 6 months | Transaminase and bilirubin from the past 6 months | Transaminase and bilirubin from the past 6 months |
| Blood | Recenta CBC | Recenta CBC | Recenta CBC | Recenta CBC |
| Antibodies |
Immunity to VZV HPV vaccination and pap test |
Antibodies to VZV or VZV vaccination | Antibodies to VZV or VZV vaccination | Antibodies to VZV or VZV vaccination |
| Ophthalmic examinations | All patients | At-risk patientsb | All patients | All patients |
| Skin examinations | Periodic skin examination is recommended for all patients, particularly those with risk factors for skin cancer | NA | NA | All patients |
| First-dose monitoring | ||||
| Requires dose titration | No | Yes, 8 days | Yes, 15 days | Yes, 5 days |
| Requires first-dose monitoring |
Yes, 6 h for all patients Overnight for those with bradycardia or other cardiovascular conditions |
No | Recommended for 4 h for patients with bradycardia, first- or second-degree (Mobitz type I) AV block, or a history of myocardial infarction or heart failure | Recommended for 6 h for patients with bradycardia, first- or second-degree (Mobitz type I) AV block, or a history of myocardial infarction or heart failure |
| Continued monitoring during treatment | ||||
| Blood pressure | Monitor and manage appropriately | Monitor and manage appropriately | Monitor and manage appropriately | Monitor and manage appropriately |
| Ocular examinations | Regular examinations of the fundus and macula for patients with a history of uveitis or diabetes mellitus, and for all patients whenever a change in vision is noted | Regular examinations of the fundus and macula for patients with a history of uveitis or diabetes mellitus, and for all patients whenever a change in vision is noted | Regular examinations of the fundus and macula for patients with a history of uveitis or diabetes mellitus, and for all patients whenever a change in vision is noted | Regular examinations of the fundus and macula for patients with a history of uveitis or diabetes mellitus, and for all patients whenever a change in vision is noted |
| Serious infections | Monitor during treatment and during the time of lymphocyte restoration after discontinuation | Monitor during treatment and during the time of lymphocyte restoration after discontinuation | Monitor during treatment and during the time of lymphocyte restoration after discontinuation | Monitor during treatment and during the time of lymphocyte restoration after discontinuation |
| Skin examinations | Examine annually by a dermatologist; evaluate any suspicious skin lesions | – | Examine annually by a dermatologist; evaluate any suspicious skin lesions | Examine- annually by a dermatologist; evaluate any suspicious skin lesions |
| Spirometric evaluations | If clinically indicated | If clinically indicated | If clinically indicated | If clinically indicated |
| Complete blood count | Monitor during treatment | Monitor during treatment | Monitor during treatment | Monitor during treatment |
| Liver aminotransferase levels | Checked periodically | Checked if clinically indicated | Checked if clinically indicated | Checked if clinically indicated |
- AV, atrioventricular; BP, blood pressure; bpm, beats per minute; CBC, complete blood count; CYP2C9, cytochrome P450 family 2 subfamily C member 9; ECG, electrocardiogram; HPV, human papillomavirus; HR, heart rate; MRI, magnetic resonance imaging; MS, multiple sclerosis; QTc, corrected QT interval; S1P1, sphingosine 1-phosphate receptor subtype 1; S1P3, sphingosine 1-phosphate receptor subtype 3; S1P4, sphingosine 1-phosphate receptor subtype 4; S1P5, sphingosine 1-phosphate receptor subtype 5; VZV, varicella-zoster virus.
- a Within 6 months or after discontinuation of prior MS therapy.
- b Ophthalmic evaluation is recommended for patients with diabetes mellitus or uveitis, prior to treatment initiation with follow-up.
Contraindications
Certain contraindications are unique to individual agents. For example, fingolimod is contraindicated in patients with a baseline QTc interval ≥ 500 msec and cardiac arrhythmias requiring anti-arrhythmic treatment with Class Ia or Class III anti-arrhythmic drugs6; ozanimod in patients with sino-atrial block, severe untreated sleep apnea, and those taking an MAO inhibitor7; ponesimod in patients with sino-atrial block8; and siponimod in patients with CYP2C9*3/*3 genotype.9 Contraindications for all four S1P receptor modulators are listed in Table 1.
Additional considerations for specific drug–drug interactions
Live, attenuated vaccines should be avoided during treatment with all four S1P receptor modulators and for 4 weeks following discontinuation.6-9 In addition, clinicians are advised to use caution when concomitantly prescribing beta-blockers, digoxin, calcium-channel blockers, or QT-prolonging therapies, considering the cardiovascular effects of S1P receptor modulators. Antineoplastic, immune-modulating, or corticosteroid or non-corticosteroid immunosuppressive therapies should also be used with caution owing to additive effects with S1P receptor modulators.
Some clinical considerations apply only to a particular medication in this class, owing to specific interactions. Coadministration of ozanimod with opioids, selective serotonin reuptake inhibitors (SSRI), serotonin-norepinephrine reuptake inhibitors (SNRI), tricyclic antidepressants, or tyramine is not recommended.7 Although a post hoc analysis of the DAYBREAK study did not find an excess number of adverse events, including serotonin syndrome–related TEAEs or hypertension in patients taking an SSRI or SNRI in combination with ozanimod.70 Concomitant use of beta-blockers and nondihydropyridine calcium channel blockers with ponesimod could lead to severe bradycardia and heart block; therefore, advice from a cardiologist should be sought before treatment initiation to determine the most appropriate monitoring for patients prescribed these medications.8 Administration of ponesimod is also not recommended with strong CYP3A4 and UGT1A1 inducers (e.g., rifampin, phenytoin, carbamazepine).8 Concomitant use of siponimod with modafinil plus all CYP2C9 and CYP3A4 inhibitors or inducers is not recommended due to a significant increase or decrease in siponimod exposure, respectively.9
Pretreatment assessments
Patients initiating treatment with siponimod should be tested to determine CYP2C9 genotype.9 Considering the cardiovascular effects of S1P receptor modulators, an electrocardiogram should be obtained to determine whether preexisting conduction abnormalities are present.6-9 A cardiologist should be consulted before initiating treatment in patients with significant QT prolongation, arrhythmias requiring treatment, ischemic heart disease, heart failure, history of cardiac arrest or myocardial infarction, cerebrovascular disease, uncontrolled hypertension, second-degree Mobitz type II or higher AV block, sick-sinus syndrome, or sinoatrial heart block.6-9 Transaminase and bilirubin levels and a complete blood count should be obtained before initiating S1P receptor modulators.6-9 Patients without a confirmed history of chickenpox or without documentation of a full course of vaccination against varicella zoster virus (VZV) should be tested for antibodies to VZV before initiating S1P receptor modulators6-9; VZV vaccination of antibody-negative patients is recommended before starting treatment with ozanimod, ponesimod, fingolimod, and siponimod.6-9 An ophthalmologic evaluation should be performed by an ophthalmologist for all patients starting fingolimod, ponesimod, and siponimod, and at-risk patients starting ozanimod.6-9 The ophthalmologic evaluation should be conducted throughout the treatment period of the S1P receptor modulator in patients with a history of uveitis, macular edema, and diabetes mellitus.71 A skin examination should be performed by a dermatologist in all patients before starting siponimod.9
First dose monitoring and dose escalation protocols
When treating patients with S1P receptor modulators, the most important clinical consideration is the effect of initial dosing on heart rate and subsequent hypertension.72 Clinicians should follow the recommended guidelines for initial titration of ozanimod until day 7, ponesimod until day 14, and siponimod until day 4 and day 5 to reach the 1 and 2 mg maintenance dose, respectively.7-9 First-dose monitoring is recommended depending on the specific agent and comorbidities in a setting where resources are available to appropriately manage symptomatic bradycardia as described in Table 4 for fingolimod, ponesimod, and siponimod.6, 8, 9 First-dose monitoring is not noted as required in the label for ozanimod. In patients with mild or moderate hepatic impairment (Child-Pugh class A or B) beginning ozanimod, the prescribing information recommends an every-other-day dosing schedule following the 7-day titration period7; however, there is a lack of published data for this recommendation.
Continued monitoring during treatment
Continued and regular monitoring of patients during treatment should be considered based on the AE profile of the S1P receptor modulator chosen for therapy.6-9 Monitoring should focus on blood pressure, ocular examinations, possible serious infections, skin examinations, white blood cell and lymphocyte counts, and liver aminotransferase levels (Table 4). Skin examinations and screening for cutaneous malignancies should be performed annually by a dermatologist in people receiving treatment when indicated. Clinicians should not only monitor patients for serious infections during treatment but also continue monitoring after discontinuation of therapy, until lymphocyte levels are restored.6-9 Patients should be counseled about the risk for progressive multifocal leukoencephalopathy (PML). If PML is suspected, based on symptomatic and/or magnetic resonance imaging findings, treatment with S1P receptor modulators should be discontinued.6-9 PML is a rare infection that predominantly occurs in cases of prolonged moderate-to-severe lymphopenia, and in older patients.73 So far, in clinical studies, PML occurred in 61 patients treated with fingolimod,74 three patients treated with siponimod,75 one patient treated with ozanimod,35 and no patients treated with ponesimod.8 These data indicate that there is a risk of PML in fingolimod-treated patients, particularly in those of older age. Immune reconstitution inflammatory syndrome (IRIS) has also been reported in MS patients who developed PML, generally within a few months of S1P receptor modulator discontinuation. Patients who discontinue treatment due to PML should be monitored for the development of IRIS.7 A small number of cases of posterior reversible encephalopathy syndrome (PRES) have occurred with fingolimod use. Symptoms include sudden onset severe headache, visual disturbances, altered mental status, and seizure. If PRES is not promptly treated, it can lead to ischemic stroke or cerebral hemorrhage. Treatment should be discontinued if PRES is suspected.6 Clinicians should educate patients of childbearing potential of the requirement for effective contraception during treatment and for the prescribed duration following treatment discontinuation. Concomitant use of S1P receptor modulators is not expected to decrease the efficacy of hormonal contraceptives based on a lack of relevant pharmacokinetic interaction between these medications. This included the coadministration of fingolimod 0.5 mg daily or ozanimod with an oral contraceptive containing ethinyl estradiol and norethindrone, coadministration of ponesimod with an oral hormonal contraceptive containing 1 mg norethisterone/norethindrone and 35 μg ethinyl estradiol, and coadministration of siponimod with an oral contraceptive containing 30 μg ethinyl estradiol and 150 μg levonorgestrel.6-9 All patients receiving S1P receptor modulators should be made aware of the risk of relapse causing severe disability following treatment discontinuation, particularly those treated with fingolimod. If this occurs, it is typically within 3 months of treatment cessation and may be associated with IRIS. Thus far, there is no evidence of rebound secondary to discontinuation when using the newer S1P receptor modulators. Importantly, a clear-cut plan to monitor and treat exacerbations should be in place.
Conclusions
S1P receptor modulators have emerged as an oral disease-modifying treatment option for MS that has a unique mechanism of action compared with other therapies. While efficacy data across the pivotal clinical trials of fingolimod, ozanimod, ponesimod, and siponimod appear to be relatively similar in patients with MS, each agent is unique and may be better suited for specific patients. For example, siponimod was the first S1P receptor modulator to be approved for patients with active SPMS; although the other agents could potentially be effective in this disease state, their efficacy in SPMS has not been proven. Clarification is needed whether all S1P receptor modulators have rebound effects, and more data are needed regarding their impact on vaccine response and the risk for infections, including PML.
Author Contributions
All authors have contributed to the writing of the manuscript and have seen and approved the submitted version.
Acknowledgements
Writing and editorial assistance was provided by Noud van Helmond, MD, PhD, of Peloton Advantage, LLC, an OPEN Health company, and was funded by Bristol Myers Squibb.
Funding Information
This review was supported by Bristol Myers Squibb, Princeton, NJ.
Conflict of Interest
PKC has consulted for Accordant, Biogen, Bristol Myers Squibb, Eli Lilly and Company, Genentech, GlaxoSmithKline, Horizon Therapeutics, LabCorp, Mylan, Novartis, Sanofi/Genzyme, and TG Therapeutics; and has received research support from Actelion, Alkermes, Celgene, CorEvitas, Genentech/Roche, Janssen Pharmaceuticals, MedDay, NINDS, Novartis, and Sanofi/Genzyme. MSF has received research or educational grants from Sanofi/Genzyme Canada; received honoraria or consultation fees from Alexion/Astra Zeneca, Biogen Idec, EMD Inc/EMD Serono/Merck Serono, F. Hoffmann-La Roche, Find Therapeutics, Novartis, Quanterix, Sanofi/Genzyme, and Teva Canada Innovation; is a member of a company advisory board, board of directors or other similar group of Actelion/Janssen (J&J), Alexion/Astra Zeneca, Atara Biotherapeutics, Bayer Healthcare, Celestra Health, EMD Inc./Merck Serono, F. Hoffmann-La Roche, Find Therapeutics, Novartis, Sanofi/Genzyme, and Setpoint Medical; has participated in a company sponsored speaker's bureau for Sanofi/Genzyme and EMD Serono. BAC has received personal compensation from Mylan for consulting, and from Applied Clinical Intelligence for serving as an external adjudicator for a clinical trial conducted by Sanofi. BACC reports personal compensation for consulting for Alexion, Atara, Autobahn, Avotres, Biogen, Boston Pharma, EMD Serono, Gossamer Bio, Hexal/Sandoz, Horizon, Immunic AG, Kyverna, Neuron23, Novartis, Sanofi, and TG Therapeutics and research support from Genentech. CEM has consulted for Alexion, ANI, Banner Life Sciences, Biogen, Bristol Myers Squibb, EMD Serono, Genentech/Roche, Sanofi/Genzyme, Horizon Therapeutics, Immunic AG, Janssen/Actelion, Novartis, and TG Therapeutics.
