The Benefits and Risks of Switching from Fingolimod to Siponimod for the Treatment of Relapsing–Remitting and Secondary Progressive Multiple Sclerosis

Fingolimod (Gilenya) was the first sphingosine 1-phosphate (S1P) receptor 1 (S1P₁) modulator [13]. The primary activity of fingolimod associated with therapeutic outcomes is the functional inhibition S1P₁ receptor on T cells. [16]. Fingolimod prevents autoaggressive T cells from exiting lymph nodes and entering the CNS to attack the myelin sheath. Thereby, fingolimod reduces inflammatory damage in the CNS [17]. The sequestration of autoaggressive T cells is reversible upon discontinuation of fingolimod treatment.

In Europe, the use of fingolimod is indicated in patients with highly active RRMS, defined as two or more disabling relapses in 1 year, and with one or more gadolinium-enhancing lesions on brain MRI, or with a significant increase in lesion load on T2-weighted imaging [18]. However, in the USA and Australia, fingolimod can also be administered to SPMS patients with superimposed relapses to reduce relapse activity [19, 20].

The efficacy of fingolimod in SPMS has been investigated in some clinical trials, but none of them were randomized, and the results are not conclusive [21, 22]. In the INFORMS study [23], when compared with placebo, fingolimod was not able to delay sustained disability progression in patients with primary progressive MS (PPMS). A recent study analysing different treatments for SPMS noted that fingolimod is superior to low-efficacy therapy in reducing disease progression in patients with active relapse associated with SPMS but not in those with inactive (no more relapses) SPMS [24].

In line with human data, fingolimod did not prevent secondary progressive neurodegeneration in the MS animal model of experimental allergic encephalomyelitis (EAE) [25]. These data indicate that the therapeutic action of fingolimod in SPMS patients is limited, especially in patients who are not in an active phase of the disease, suggesting that, once RRMS patients progress to the SPMS disease phase, it would be advised to switch the treatment to siponimod.

Siponimod (Mayzent) is a second-generation S1PR modulator that binds with high affinity to S1PR1 and S1PR5. It was approved by the FDA in 2019 for adults with RRMS, SPMS or CIS [14, 29]. In Europe, based on the pivotal phase 3 trial, the authorized therapeutic indication for siponimod is for adults with SPMS with active disease, evidenced by relapses or imaging features of inflammatory activity [30]. Siponimod is currently the only approved therapy for SPMS [31]. In RRMS, it was shown that siponimod reduced the frequency of relapses and the number of brain lesions at MRI [32].

Similar to fingolimod, siponimod acts as a functional antagonist of S1P₁ on circulating T cells. Specifically, it induces receptor internalization and degradation (also called downmodulation) to inhibit cell egress from lymph nodes, and therefore, it prevents autoreactive lymphocyte infiltration into the CNS [33].

Siponimod may also exert therapeutic activity within the CNS [34]. In the EAE mouse model, siponimod prevented synaptic neurodegeneration, independently of its effects on circulating lymphocytes [35], and promoted remyelination in the Xenopus laevis model [36‐39].

There are currently no data on the effects of switching from fingolimod to siponimod for the treatment of RRMS [41]. It was shown that siponimod has therapeutic effects in RRMS. A phase 2 clinical trial of patients with RRMS in 2013 showed that siponimod reduced MRI lesion activity at 3 months in a dose-dependent manner (from 0.25 to 10 mg) [32]. Although no serious side effects were reported at 3 months, some of them (headache, bradycardia, dizziness and nasopharyngitis) were observed after 6 months in a small percentage of patients (from 5% to 19% in the dose groups studied). Similar efficacy and safety results were reported in the extension of this phase 2 clinical trial in 2016 [42].

At present, there are not phase 3 studies on the effect of siponimod in RRMS. Thus, a comparison with phase 3 studies of fingolimod in RRMS is not feasible. The term ‘relapsing multiple sclerosis’ (RMS) includes relapsing–remitting multiple sclerosis (RRMS) and SPMS with relapses. The latter is termed ‘active’ SPMS, for which RMS- or active-SPMS-approved disease modifying therapies (DMTs) can be used. However, the level of evidence supporting efficacy and safety in SPMS differs between DMTs approved for RMS and SPMS [43]. In the aforementioned phase 2 clinical trials of patients with RRMS, it is concluded that evaluation of therapeutic effects and tolerability in RRMS warrant investigation in phase 3 trials [32, 42].

Although S1PR modulators have better efficacy and safety profiles than other neuroimmunomodulatory drugs for the treatment of RRMS [44], there is currently insufficient evidence indicating that switching from fingolimod to siponimod for the treatment of RRMS has potential benefit in terms of therapeutic efficacy and side effects. Thus, further preclinical and clinical studies are necessary to evaluate the advantages and disadvantages of this treatment switch in patients with RRMS.

Recently reported cost-effectiveness analyses of siponimod use in SPMS concluded that, in patients with active SPMS, switching to siponimod is a cost-effective and clinically beneficial treatment approach as compared with the continuation with other DMDs [45], including fingolimod [46]. Therefore, in patients with active SPMS, switching from fingolimod to siponimod may have both clinical and economic benefits, especially when this switching is supported by appropriate analyses to diagnose active SPMS.

Despite these indications on the potential benefit of switching from fingolimod to siponimod, this translation is likely not without risk. Some recent case reports have evidenced possible initial complications that need to be closely monitored. A SPMS patient who switched from fingolimod to siponimod experienced disease reactivation in the absence of medication during a gap between the two treatments [47]. Another patient developed acute optic neuritis after a few weeks of treatment with siponimod, which persisted after treatment discontinuation [48]. Furthermore, another SPMS patient developed severe lymphopenia 1 month after switching from fingolimod to siponimod [49].

In a recent case series with 13 SPMS patients switching from fingolimod to siponimod, it was reported that three patients experienced disease reactivation in the first 3 months after switching [50]. Based on pharmacokinetics of these two drugs, the authors suggest that disease reactivation after switching from fingolimod to siponimod may be due to upregulation of S1P₃, which triggers a pro-inflammatory cytokine cascade via activation of NFκB [51].

These case reports highlight the potential risks of switching from fingolimod to siponimod for the treatment of SPMS. Careful monitoring of disease activity is therefore recommended before and after switching these treatment regimens.

The current knowledge on the mechanisms of action of fingolimod and siponimod mostly comes from studies using animal and cell culture models [26, 33, 34] (Table 1). These studies have shown that, although these two drugs have a similar mechanism of action, they can activate different receptors in the sphingolipid-regulated signalling pathway. Sphingolipids constitute a highly complex class of molecules that serve as structural components of cellular membranes and as signalling molecules capable of eliciting apoptosis, differentiation, chemotaxis and other responses in mammalian cells [52].

One important sphingolipid is S1P, a physiological signalling molecule that acts as a ligand for a group of cell surface receptors, the S1P receptors (S1PRs). These receptors are expressed in various tissues and regulate diverse physiological and pathological cellular responses involved in innate and adaptive immunity as well as in cardiovascular and neurological functions [53]. In particular, S1PR1 is expressed on the cell surface of lymphocytes, which are well known for their major role in the pathogenesis of MS. Moreover, S1PR1 plays an important regulatory role in the egress of lymphocytes from lymphoid organs into lymphatic circulation [33].

Both fingolimod and siponimod target S1PR1 and thus modulate immune cell trafficking through the sequestration of autoreactive lymphocytes in lymphoid organs. This sequestration reduces their recirculation and subsequent infiltration into the CNS [33]. Therefore, these two drugs reduce MS immuno-inflammation.

Upon entry into the CNS, fingolimod and siponimod also reduce inflammatory signalling in glial cells and astrocytes expressing S1P1 [54]. In addition, both drugs can bind S1P₅, which is predominantly expressed in oligodendrocytes and brain endothelial cells [55]. S1P₅ activation protects adult oligodendrocytes from apoptosis and contributes to the maintenance of blood–brain barrier (BBB) integrity [56]. Moreover, S1P₅ activation may support reparative action by promoting oligodendrocyte maturation and survival and myelination.

However, siponimod seems to have a more specific and selective activity than fingolimod in the CNS, at least in cellular cultures and animal models [34, 57]. Experiments in the Xenopus tadpole model showed that spontaneous remyelination does not occur in S1P₅-knockout tadpoles and that remyelination increases in response to highly selective S1P₅ agonists, but not to S1P₁ agonists [39]. In EAE mice, siponimod administration induced partial restoration of cortical neuronal circuit function, lowering inflammatory cell infiltration in cortical grey matter lesions and restoring cortical network functionality [57]. Although a neuroprotective action of both fingolimod and siponimod has been observed in cell culture and preclinical models [26], clinical data for fingolimod have not conclusively shown a consistent effect on disease progression or other clinically relevant outcome measures of neurodegeneration [41]. The effect of siponimod on MS progression is more robust than that of fingolimod, resulting in greater efficacy for the treatment of SPMS. Siponimod appears to be more selective than fingolimod in targeting S1P₅ [34]. Therefore, its action at S1P₅ may be more potent in the CNS [34, 56]. In in vitro functional assays with astrocytes generated from human fibroblasts, it was shown that both siponimod and fingolimod have a comparable neuroprotective action on S1P₁ receptor in glial cells [37]. Nonetheless, the specific target selectivity (S1P_1,5) of siponimod and its CNS/blood pharmacokinetic (PK)/pharmacodynamic (PD) profiles allow optimal realization of a dual mode of action: inhibiting peripheral/central inflammation and promoting remyelination. This could also explain the different effects of siponimod and fingolimod in SPMS patients [33, 41].

When discussing the pharmacokinetic/pharmacodynamic (and side effects) of fingolimod and siponimod, it should be noted that fingolimod is a nonselective functional antagonist of four of the five S1PRs (1, 3, 4, and 5) [26]. In contrast, siponimod exhibits high-affinity binding to only S1PR1 and S1PR5, resulting in a favourable side effect profile compared with that of fingolimod.

At the S1P₁-receptor level, both drugs act as functional antagonists and cause the internalization and degradation of this receptor [22]. The S1P₃ receptor is instead only internalized by fingolimod, but not by siponimod [22]. These pharmacokinetic differences explain the more manageable cardiac effects of siponimod compared with fingolimod. In fact, the S1P₃ receptor appears to be a key direct regulator of cardiac rhythm and is involved in the risk of bradycardia and type I and type II atrioventricular block, as demonstrated in mouse models [58].

The interaction of fingolimod with S1P₃ may be responsible for serious side effects [59], such as atrioventricular nodal block [60] and macular oedema [61]. Nonetheless, similar side effects were also reported with siponimod [34, 62], bringing into question the effective role of the S1P₃ receptor in such complications. It should be noted that the lack of action of siponimod on the S1P₃ receptor, while bringing benefits in terms of cardiac complications, could also be the reason for disease reactivation after switching from fingolimod to siponimod, as previously stated. When fingolimod is withdrawn, there may be an upregulation of the S1P₃ receptor which is present in astrocytes. This upregulation is able to trigger a pro-inflammatory response through the ability to activate RhoA [51].

In addition to these pharmacokinetic differences, there are also pharmacodynamic differences between fingolimod and siponimod [22, 63]. It was shown that fingolimod is able to penetrate the CNS to a greater extent than siponimod [64]. Siponimod, however, has a much faster elimination half-life (2–5 h) than fingolimod (12–16 h) [22, 63]. However, the bioavailability of the drug in the CNS may not correspond to the levels of free drug required to act on the receptors. Thus, therapeutic efficacy depends on a combination of factors including molecular composition, half-life, ability to cross nervous tissue membranes, and CNS transport and metabolism mechanisms (phosphorylation, oxidation) [65].

A representative scheme of the interactions of these two drugs with S1PRs is shown in Fig. 1.