1 Introduction
Siponimod (BAF312), a potent, oral, selective modulator of the sphingosine 1-phosphate (S1P) receptor subtypes 1 and 5 (S1P
1,5), limits the inflammatory effects mediated by B and T cells [
1]. It is currently under advanced phase of clinical development for the treatment of secondary progressive multiple sclerosis (SPMS) [
2]. The pharmacokinetic (PK) profile of siponimod in a single-ascending dose range study (0.1–75 mg) in 98 healthy subjects was measurable in the plasma as early as 0.25 h postdose, and the maximum plasma concentration (
Cmax) was reached within 3–8 h postdose (minimum–maximum 0.25–24 h). In a multiple-ascending dose range study (0.3–20 mg) in 48 healthy subjects over 28 days, the steady-state of plasma siponimod was reached after approximately 6 days; the mean accumulation ratio was 1.9–2.7 and the effective elimination half-life (
T½) was 22–38 h (mean 30 h) [
1]. The rate and extent of systemic exposure were increased in a dose-proportional manner after single (0.1–75 mg) and multiple (0.3–20 mg) doses of siponimod [
1].
Cytochrome P450 (CYP) 2C9 is the major enzyme responsible for the clearance of siponimod [
3,
4]. Siponimod is eliminated from the systemic circulation mainly due to metabolism and subsequent biliary/faecal excretion [
4‐
6]. Metabolite M3 is one of the main circulating metabolites of siponimod in humans; it is formed by glucuronidation of the hydroxylated M5 metabolite that results from metabolism primarily via CYP2C9 (79.2%), with a minor contribution from CYP3A4 (18.5%) [
3,
6,
7].
A drug–drug interaction (DDI) study was conducted with a CYP2C9 inhibitor to evaluate the effect of CYP2C9 inhibition on siponimod PK. In the absence of any strong CYP2C9 inhibitor, fluconazole was selected as it is one of the most potent CYP2C9 inhibitors used in clinical practice. Fluconazole is a moderate CYP2C9 and CYP3A inhibitor and is recommended in regulatory guidance as a prototype inhibitor to assess potential DDI by CYP2C9 inhibition [
8,
9]. This study will therefore support the development of clinical recommendations for siponimod coadministration with CYP2C9/CYP3A inhibitors.
CYP2C9 is a polymorphic enzyme. Of more than 50 single nucleotide polymorphisms (SNPs) described in the regulatory and coding regions of the CYP2C9 gene, only two coding SNPs, namely CYP2C9*2 and CYP2C9*3, have shown clinically relevant reductions in enzyme activity [
10]. These two SNPs result in six different genotypes that confer three functionally different phenotypes, namely extensive metabolisers (EMs; CYP2C9*1/*1), intermediate metabolisers (IMs; CYP2C9*1/*2, CYP2C9*1/*3 and CYP2C9*2/*2) and poor metabolisers (PMs; CYP2C9*2/*3 and CYP2C9*3/*3) [
10‐
12]. Results from in silico SimCYP simulations [a physiological-based PK (PBPK) model; SimCYP Ltd, Sheffield, UK] suggested an increased plasma exposure of siponimod in subjects with the CYP2C9*2 and CYP2C9*3 genotypes, due to a reduction in the enzyme activity [
3]. Exploratory PK/pharmacogenetic (PG) analyses indicated that heterozygous CYP2C9*3 carriers tended to have a higher area under the curve (AUC) of siponimod compared with subjects not carrying the *3 allele (unpublished data). A study was conducted to assess the impact of the reduced CYP2C9 enzyme activity on siponimod PK in CYP2C9 PMs (CYP2C9*2/*3 and CYP2C9*3/*3).
The present article reports on two studies: Study A, in vivo effects of the steady-state CYP2C9 enzyme inhibitor, fluconazole, on the PK and safety/tolerability of a single oral dose of siponimod 4 mg in healthy adult subjects; and Study B, the PK and safety/tolerability of a single dose and 3-day dosing of siponimod in healthy subjects with polymorphic variants of CYP2C9.
4 Discussion
The study in healthy subjects with the CYP2C9*1/*1 genotype found that coadministration with fluconazole led to a twofold increase in siponimod AUC, together with a 50% increase in T½, and a minor increase in Cmax compared with siponimod alone. In another study conducted in healthy subjects with polymorphic variants of CYP2C9, siponimod AUC was approximately twofold and fourfold greater in the CYP2C9*2/*3 and CYP2C9*3/*3 genotypes, respectively, and mean Cmax was increased by < 25% compared with the CYP2C9*1/*1 genotype.
For a drug that is a substrate of a polymorphic enzyme or transporter, the difference in drug exposures between EM and PM genotypes would generally represent the most extreme change that could be caused by a strong inhibitor of that pathway. An alternative to a genotype-specific PK study is to administer the investigational drug to EMs with and without concomitant administration of a known strong inhibitor of the metabolic pathway [
16]. This interaction approach is considered even more attractive when the prevalence of the PM genotype is very low, as for the CYP2C9*3/*3 genotype. However, in the absence of any strong CYP2C9 inhibitor in clinical use, this alternative approach could not be used to estimate the siponimod PK in CYP2C9 PMs. Two separate studies were therefore conducted to fully characterise the role of the CYP2C9 pathway on siponimod PK [
3].
As polymorphic variations in the CYP2C9 enzyme may lead to increased exposure of siponimod due to reduced metabolic clearance, Study A prescreened CYP2C9 genotypes and excluded subjects who were not homozygous for the CYP2C9*1 allele (wild-type) to reduce the intersubject variability in PK parameters and reduce sample size. Study B included subjects with the CYP2C9*2/*3 and CYP2C9*3/*3 genotypes because the largest impact on PK and drug metabolism through CYP2C9 polymorphism is expected in these populations [
17], as indicated by in vitro metabolism results and predicted with SimCYP PBPK modelling [
3]. In Caucasians, the prevalence of CYP2C9*1/*1 ranges from 62 to 65%; CYP2C9*1/*2 from 20 to 24%; CYP2C9*2/*2 from 1 to 2%; CYP2C9*1/*3 from 9 to 12%; CYP2C9*2/*3 from 1.4 to 1.7%; and CYP2C9*3/*3 from 0.4 to 0.5% [
18,
19].
The selection of dose in both studies was based on the PK data generated from the single (unpublished data) and multiple-ascending dose [
1] range studies in healthy subjects and was supported by the SimCYP simulations study (SimCYP Ltd) [
3]. Single doses of siponimod up to 25 mg were well tolerated and showed favourable safety profiles in healthy subjects in a previous study (unpublished data). In Study A, a dose of siponimod 4 mg was administered as this dose represented a relevant therapeutic dose level for siponimod in multiple sclerosis (MS) at the time this study was conducted. The selected dose provided a sufficient safety margin for an up to fivefold increased systemic exposure due to an interaction with fluconazole (similar exposure as with a single-dose maximum tolerated dose [MTD] of 25 mg). Based on the SimCYP simulation model, steady-state concentration of fluconazole was reached within 3 days following oral doses of 200 mg twice daily on day 1 and 200 mg administered once daily after day 2 [
3]. This was consistent with the fluconazole prescribing information indicating that administration of a loading dose (on day 1) of twice the usual daily dose results in plasma concentrations close to steady-state by the second day [
20]. To achieve a rapid steady-state plasma concentration of fluconazole, a loading dose of twice daily on the first day of therapy was recommended [
21]. A two to threefold increase in siponimod exposure (AUC
∞) with fluconazole treatment was expected based on the results of SimCYP simulations [
3]. In part 1 of Study B, the selected dose of 0.25 mg was 100-fold lower than the MTD defined in the phase I, single-ascending dose range study in healthy volunteers (unpublished data). In part 2, the target dose of 0.5 mg on day 3 was preceded by 0.25 mg once daily over 2 days in order to attenuate the bradyarrhythmic effects of siponimod, thereby mimicking the first three dosing days of the dose titration regimen used in clinical studies in patients with MS. The SimCYP simulations predicted that siponimod exposure was 4.5-fold higher (based on AUC
∞) in subjects with a CYP2C9*3/*3 versus CYP2C9*1/*1 genotype [
3]. The therapeutically relevant dose of siponimod in MS, as investigated in pivotal studies, is 2 mg [
22,
23]. In view of the dose-linear PK, the 0.5 mg dose in PM subjects was therefore expected to have a similar exposure compared with the therapeutic dose of 2 mg.
Coadministration of a single dose of siponimod and fluconazole at steady-state in healthy subjects (EMs: CYP2C9*1/*1 genotype) led to a twofold increase in plasma AUC of siponimod and an approximately 50% increase in terminal
T½. As expected, these results show that fluconazole did not affect the absorption phase of siponimod, and CYP2C9 inhibition only influenced the elimination phase. The results observed in this study were in line with, and confirmed, the predictions obtained through SimCYP simulations [
3]. Other enzymes and transporters might also play minor roles in siponimod disposition and hence inhibition of these pathways by fluconazole could have also partly contributed to the increase in siponimod exposure.
Although the expression and activity of the CYP2C9 enzyme is lower overall in the gut than in the liver, the surface area of the proximal small intestine is large, and intestinal CYP2C9 may well contribute to the first-pass metabolism of their substrate drugs [
24]. The first-pass effect of siponimod was estimated by SimCYP to be minor (in the range of 5% of the administered dose only (unpublished data). In the DDI study (Study A), the
Cmax of siponimod was slightly increased by 10% in the presence of fluconazole. It cannot be excluded that this slight increase could be due to the inhibition of the first-pass effect through fluconazole inhibition of the CYP2C9 enzyme expressed in the duodenum and small intestine.
As siponimod displays dose-proportional and time-independent PK exposure, the observed magnitude increase on exposure in single-dose studies can therefore be extrapolated to steady-state conditions (repeat dose) [
25]. SimCYP PBPK modelling [
3] was carried out under single-dose and steady-state conditions. Both simulation conditions predicted similar exposure ratios for siponimod under fluconazole inhibition (unpublished data). Therefore, the impact of coadministration of fluconazole on the repeat-dose PK of siponimod (steady state) is predicted to result in an approximately twofold increase in mean AUC
tau,ss, similar to the single-dose situation (based on AUC
∞).
In a multiple-ascending dose study, siponimod doses up to 20 mg administered daily for 28 days were well tolerated, demonstrating a favourable safety profile at all doses [
1]. Siponimod exposure was increased by twofold in the presence of fluconazole, which provided an adequate safety margin of approximately fivefold considering a siponimod therapeutic dose of 2 mg once daily. The increased exposure of siponimod in the presence of moderate CYP2C9 inhibitors is therefore considered not to be associated with safety and tolerability risks for short-term combination treatment. However, for longer coadministration periods, the reduced CYP2C9 metabolic activity may be compensated by reducing the dose of siponimod when administered in the presence of a moderate CYP2C9/CYP3A inhibitor (e.g. fluconazole).
Some key drug-metabolising enzymes, such as CYPs, are known to be modulated by systemic proinflammatory cytokines released during infection or inflammation, resulting in alteration in biotransformation and elimination of small-molecule substrates of the affected CYPs [
26]. Systemic levels of interleukin (IL)-6, a potent proinflammatory cytokine, have been found to be elevated in patients with various systemic inflammatory diseases, including MS [
27]. A cocktail clinical DDI study conducted in patients with rheumatoid arthritis reported that the administration of an anti-IL-6 receptor monoclonal antibody reversed IL-6-induced suppression of CYP enzyme activity, including CYP2C9 [
28]. The influence of disease state on siponimod exposure was investigated in two clinical studies in MS patients and in two population PK evaluations. No significant differences in the PK of MS patients and healthy subjects were observed (unpublished data). Based on these results, systemic IL-6 release is unlikely to significantly affect siponimod PK in patients with MS.
The genetic variation in the CYP2C9 enzyme affects the metabolism of siponimod. After a single oral dose of siponimod 0.25 mg, the AUC
∞ and AUC
last of siponimod were approximately two and fourfold higher in PMs with the CYP2C9*2/*3 and CYP2C9*3/*3 genotypes, respectively, compared with CYP2C9 EMs (CYP2C9*1/*1). The results observed in the present study were in line with the predictions obtained through the SimCYP simulation method [
3]. Due to the reduced CYP2C9 enzymatic activity in PMs, the siponimod plasma
T½ was prolonged in subjects with the CYP2C9*2/*3 and CYP2C9*3/*3 genotypes (50.9 and 126 h, respectively) compared with EMs (CYP2C9*1/*1: 28.1 h). The metabolite M3 accounts for approximately 28% of the exposure to siponimod, whereas M5 represents only approximately 2.5% of the parent plasma exposure [
6]. Both M3 and M5 have weak pharmacological activity on the S1P
1 receptor only (M5 is approximately 470-fold and M3 is approximately 10,000-fold less potent than siponimod on the S1P
1 receptor). The metabolite M5 results from metabolism primarily via CYP2C9, with a minor contribution from CYP3A4. A decrease in M3 and M5 exposure is therefore expected when the metabolite activity of CYP2C9 is decreased in CYP2C9 PMs (unpublished data). As expected in M3 and M5 metabolites,
Cmax and AUC were markedly lower in the CYP2C9*2/*3 and CYP2C9*3/*3 genotypes compared with the CYP2C9*1/*1 genotype, with a maximum 95% reduction in AUC for the M5 CYP2C9*3/*3 genotype compared with the CYP2C9*1/*1 genotype, confirming the major role of CYP2C9 in siponimod metabolism.
The magnitude of the effect observed in both studies is consistent with the major contribution of the CYP2C9 pathway to total metabolic clearance (fm; 79%) determined in vitro [
3].
For drugs where PG is important for PK variability, pharmaceutical companies are encouraged by regulators to assess if enzyme polymorphism may lead to a different benefit–risk in certain genetic subpopulations [
29]. These investigations are aimed at evaluating whether exposure in genetic subpopulations is different to such an extent that this would require a change in the posology of the drug for the specific subpopulation to achieve an exposure that is shown to be effective and safe.
Based on the results of Study B on the CYP2C9 genetic polymorphism, a siponimod genotype-based dosing may be considered for individuals with certain genetic polymorphisms, to adjust for the reduced CYP2C9 metabolic activity and avoid potential long-term safety risks of chronic higher exposure.
Final recommendations on the concomitant use of siponimod and CYP2C9/3A4 inhibitors, and on the use of siponimod in SPMS patients with different CYP2C9 genetic polymorphisms, will be made considering all pertinent clinical and in silico data.