Teriflunomide: A Review of Its Use in Relapsing Multiple Sclerosis
Abstract
Teriflunomide (AubagioTM) is the main active metabolite of leflunomide, an established disease-modifying anti- rheumatic drug. Teriflunomide is an inhibitor of de novo pyrimidine synthesis, reducing lymphocyte proliferation, amongst other immunomodulatory effects; autoimmunity is believed to be one of the potential mechanisms of disease for multiple sclerosis. Teriflunomide is considered cytostatic but not cytotoxic: it does not affect resting or slowly dividing lympho- cytes. This article reviews the available pharmacological prop- erties of oral teriflunomide and its clinical efficacy and tolerability in patients with relapsing multiple sclerosis. While both the 7 and the 14 mg/day dosages are discussed, the 7 mg/day dosage is not approved in the EU. Both dosages are approved in the USA. In phase III trials, teriflunomide 7 or 14 mg/day was consistently demonstrated to be more effective than placebo and as effective as interferon beta-1a in the prevention of relapses in patients with relapsing forms of multiple sclerosis; moreover, teriflunomide 14 mg/day was also consistently shown to be more effective than placebo in prevention of disability progression. Teriflunomide was generally well tolerated in these patients. Long-term, extension data were generally similar to those observed in the shorter-term trials.
Teriflunomide is associated with increased liver enzyme levels, and is contraindicated in pregnant patients because of a potential risk of teratogenicity. As an oral treatment, it offers an alternative to the traditional, parenteral, disease- modifying therapies; however, further investigation into the efficacy and/or tolerability differences between teriflunomide and other available oral drugs would be of great use in the placement of this drug. At present, given the relatively limited long-term data, it is difficult to draw definite conclusions with regard to safety; however, as teriflunomide is the main active metabolite of leflunomide, long-term safety data can be extrapolated from the large amount of post-approval data available regarding its parent drug. Oral teriflunomide is a valuable addition to available treatment options for patients with relapsing multiple sclerosis, in particular those patients who prefer an oral drug.
1 Introduction
Until recently, the &2 million patients with multiple sclerosis worldwide (&85 % of whom have a relapsing- remitting disease course) were limited to treatment with parenteral disease-modifying therapy (DMT) [1]. DMTs (e.g. interferon beta and glatiramer acetate) have been available since the 1990s, and primarily act by impeding the activation, proliferation and migration of inflammatory cells across the blood-brain barrier [1, 2]. However, par- enteral application of drugs is linked to issues with adherence, which is of major concern with chronic dis- eases, and can also be associated with injection-site reac- tions [1–3].
As a result of these limitations, several new, oral, treatments have become available over the past few years: fingolimod, dimethyl fumarate (BG-12) and teriflunomide (AubagioTM) [1–3]. These new treatments have different mechanisms of action, potentially resulting in differing efficacy and tolerability profiles [1, 2], and provide a variety of potential treatments for patients, enabling indi- vidualized treatment based on patient preference, comorbid disease and other factors.
This article reviews the available pharmacological properties of oral teriflunomide and its clinical efficacy and tolerability in patients with relapsing multiple sclerosis. While both the 7 and the 14 mg/day dosages are discussed, the 7 mg/day dosage is not approved in the EU [4]. Both dosages are approved in the USA [5].
2 Pharmacodynamic Properties
Teriflunomide is the main active metabolite of leflunomide, a disease-modifying anti-rheumatic drug indicated for the treatment of active rheumatoid arthritis [5, 6]. The phar- macodynamic effects of leflunomide (focusing on its active metabolite teriflunomide [A77 1726]) have previously been extensively reviewed [7]; this section briefly summarizes the pharmacodynamic effects of teriflunomide, an immu- nomodulatory agent with anti-inflammatory properties. See Fig. 1 for the chemical structure of teriflunomide, and how it compares with that of leflunomide. Teriflunomide is an inhibitor of de novo pyrimidine synthesis (with no impact on the salvage pathway), achieved via the inhibition of dihydro-orotate dehydroge- nase (DHODH), a mitochondrial enzyme [4, 5]. While the exact mechanism of action for the therapeutic effect of teriflunomide in multiple sclerosis is unknown, evidence suggests that it may involve a reduction in the number of activated lymphocytes in the CNS [4, 5]. It is a selective immunomodulator, according to a preclinical study; teri- flunomide efficiently interferes with the expansion of T-cell clones with high-T-cell-receptor-affinity for anti- gens, while also permitting low-affinity-triggered T-cell responses [8].
Teriflunomide is a potent inhibitor of DHODH, although two in vitro studies gave differing values for the inhibition constant value (2.7 lmol/L [9] and 0.179 lmol/L [10]) and the half maximal efficacy concentration (1.08 lmol/L [11] and 0.3 lmol/L [10]). It is a competitive inhibitor of the ubiquinone binding site, and does not compete at the dihydro-orotate binding site, on DHODH [10].
Activation of lymphocytes leads to the induction of DHODH, required for the de novo synthesis of uridine monophosphate, which is a precursor for pyrimidine nucleotides, in actively dividing cells (Fig. 2) [7]. In vitro, inhibition of DHODH with teriflunomide or its precursor leflunomide is associated with inhibited uridine mono- phosphate de novo synthesis [9]; inhibited pyrimidine (uridine triphosphate [UTP] and cytidine triphosphate [CTP]) synthesis and (at higher concentrations) reduced purine (adenosine triphosphate [ATP] and guanosine tri- phosphate [GTP]) synthesis [12, 13]; inhibited DNA and RNA synthesis [14]; and, ultimately, inhibition of human T-lymphocyte [13, 15, 16] and B lymphocyte [16] prolif- eration and G1-phase cell cycle arrest in human T lym- phocytes [14] and murine B lymphocytes [17]. Additionally, in murine B lymphocytes, the change from S to G2/M phase is targeted [17]. Cell activation is also reduced, but to a lesser extent than cell proliferation (mean maximal inhibition was less than 50 %), in T lymphocytes, but not significantly affected in B lymphocytes [16]. Moreover, no significant effects on cell viability were observed [16].
Teriflunomide is considered cytostatic but not cytotoxic; it does not affect resting or slowly dividing lymphocytes, as they have low levels of DHODH activity, and utilise extracellular pyrimidines from the DHODH-independent salvage pathway for cell survival (Fig. 2) [6, 7]. Thus, the salvage pathway can sustain cells arrested in the G1 cell cycle phase (Fig. 2) [6, 7].
In vitro, teriflunomide has been shown to potentially have an inhibitory effect on cell adhesion, as well as inhibiting other functions, such as immunoglobulin syn- thesis, nuclear factor (NF) jB activation, NFjB-dependent reporter gene expression, cyclo-oxygenase activity induc- tion, protein tyrosine kinase activity, and oxygen radical production [6, 7]. Moreover, teriflunomide is associated with increased production of the immunosuppressive transforming growth factor-b1 protein, the inhibition of proinflammatory interleukin (IL)-2 levels, and the inhibi- tion of cell surface IL-2 and transferrin receptors in qui- escent peripheral blood lymphocytes, as well as inhibition of IL-1b expression and metalloproteinase-1 (a tissue destructive enzyme) production [7]. In an in vivo study, teriflunomide was associated with decreased chemokine (CXCL9 and CXCL10) levels in the CNS [18]. These may all contribute to its immunomodulatory effects; however, the clinical relevance and contribution to its effects in humans have not been confirmed, and the concentrations used in these studies were largely supraphysiological.
In the experimental autoimmune encephalomyelitis (EAE) rat model of multiple sclerosis, teriflunomide 3 [19] and 10 mg/kg/day [19, 20] were consistently associated with significantly delayed disease onset (p \ 0.01 vs. vehicle) [19] and reduced disease scores (p \ 0.01 vs. vehicle) [19, 20] when administered prophylactically, and significantly reduced disease scores when administered therapeutically (p \ 0.01 vs. vehicle) [19, 20]. At both administration time points, teriflunomide was also associ- ated with significantly reduced inflammation [19], demy- elination [19] and axonal loss [19, 20]. Moreover, teriflunomide 10 mg/kg/day treatment in the EAE rat model was associated with a reduction in neurological deficits (associated with reduced levels of spinal cord- infiltrating immunocytes) and attenuation of disease-asso- ciated changes in immunocyte numbers and distribution in whole blood and the spleen [21].
In placebo-controlled trials in patients with multiple sclerosis (see Sect. 4.1), teriflunomide was associated with a mean decrease in white blood cell count of &15 % (mainly neutrophils and lymphocytes); this decrease occurred during the initial 6 weeks of treatment, after which the count remained low during treatment [4, 5]. These reductions in neutrophil and lymphocyte counts from baseline occurred to a slightly greater extent with teriflunomide treatment than with placebo in the TEMSO study [22]; the relative difference from placebo in lym- phocyte count was -13 and -18 % for teriflunomide 7 and 14 mg, respectively, and the relative difference in neutro- phil count was -12 and -15 % [23]. For the most part, mean absolute counts remained within the normal range of 1 9 109 to 4 9 109/L for lymphocytes and 2.5 9 109 to 7.5 9 109/L for neutrophils [23].
In the TENERE study (see Sect. 4.2), the mean change from baseline at week 48 in neutrophils was -0.59 and -0.76 versus -0.72 Giga/L in recipients of teriflunomide 7 and 14 mg/day versus interferon beta-1a [24]. The respective change from baseline in lymphocytes was -0.15 and -0.29 versus -0.39 Giga/L. Further data regarding the tolerability aspect of white blood cell reduction in clinical trials are discussed in Sect. 5. According to in vivo data, teriflunomide may be asso- ciated with a delay in antiviral antibody response, but does not prevent the response altogether, implying that increased viral infection is unlikely in recipients of teriflunomide [25]. Teriflunomide 7 or 14 mg/day treatment (for C6 months) did not appear to affect the antibody response to influenza vaccine in patients with relapsing multiple sclerosis, when compared with that in patients receiving interferon beta-1 [26]. In another study in healthy volun- teers, teriflunomide 14 mg/day for 25 days was associated with a lower but effective antibody response to rabies vaccine, compared with placebo [27].
Teriflunomide was not associated with clinically sig- nificant corrected QT interval (QTc) prolongation in a placebo-controlled study in healthy volunteers [28]. The largest time-matched mean difference between terifluno- mide and placebo recipients in change from baseline in QTc (Fridericia) [QTcF] was 3.46 ms, with an upper bound of the 90 % confidence interval of \10 ms [28]. No QTcF values were [480 ms, and no subjects experienced a change from baseline of [60 ms [28].
3 Pharmacokinetic Properties
As a metabolite of leflunomide, teriflunomide treatment results in a similar range of teriflunomide plasma concen- trations to that with leflunomide, at the recommended dosages [5]. The pharmacokinetics of teriflunomide in patients with multiple sclerosis are similar to those in healthy volunteers [29].
Oral teriflunomide is absorbed rapidly; the maximum teriflunomide plasma concentration is reached between 1 and 4 hours post dose [4, 5]. The absolute oral bioavail- ability of oral teriflunomide is &100 % [4, 5], and the drug displays linear pharmacokinetics from 7 to 100 mg doses [29]. The estimated area under the concentration-time curve accumulation ratio is &34-fold, and steady state is reached after &100 days’ treatment [4]. Food has no clinically relevant effect on teriflunomide pharmacokinet- ics [4, 5, 29].
The volume of distribution of a single intravenous injection of teriflunomide is 11 L [4, 5, 29]; this may be an underestimation, as extensive organ distribution was observed in rats [4]. It is mainly distributed in plasma, and [99 % of the drug is bound to plasma protein, mainly albumin [4, 5, 29].
While unchanged teriflunomide is the major circulating moiety in plasma, following oral administration of 14C- teriflunomide in healthy volunteers, minor metabolites are produced via hydrolysis (primary pathway), oxidation, N-acetylation and sulphate conjugation (secondary path- ways) [4, 5, 29]. The main metabolite is 4-trifluoro- methylaniline oxanilic acid [30].
Teriflunomide is eliminated very slowly from plasma [5]. Without accelerated elimination, it takes an average of 8 months to reach plasma concentrations of \0.02 mg/L after treatment discontinuation; however, it may take as long as 2 years, as a result of individual variations in drug clearance [5].
Elimination of teriflunomide can be enhanced by the administration of cholestyramine 8 g (or 4 g if the higher dosage is not well tolerated) three times a day or activated charcoal 50 g twice daily [4, 5]; both decreased the con- centration of teriflunomide by [98 % after 11 days of administration (teriflunomide had been administered at 14 mg/day for 8–11 days, after a loading dose of teriflun- omide 70 mg/day for 3–4 days) [31]. Cholestyramine 8 g was more efficient than cholestyramine 4 g, which was in turn more efficient than activated charcoal [31].
Elimination of teriflunomide occurs mainly via direct biliary excretion of unchanged drug (most likely by direct secretion [4]), as well as the renal excretion of metabolites [4, 5]. A total of 60.1 % of the administered dose of oral 14C-teriflunomide in healthy volunteers was excreted over the 3 weeks following administration (37.5 % in faeces [mainly as unchanged drug [29]] and 22.6 % in urine [mainly as the 4-trifluoro-methylaniline oxanilic acid metabolite [29]]) [4, 5, 29]. Following an accelerated elimination procedure with cholestyramine, an additional 23.1 % was recovered, mostly in faeces [4, 5].
The total body clearance of teriflunomide following a single intravenous dose was 30.5 mL/h [4, 5]. The median elimination half-life of repeated doses of teriflunomide 7 and 14 mg/day was &18 and &19 days, respectively, based on a population analysis of healthy volunteers and patients with multiple sclerosis [4, 5].
As teriflunomide is a substrate of the Breast Cancer Resistant Protein (BCRP) [based on in vitro studies], BCRP inhibitors such as cyclosporine, eltrombopag or gefitinib may increase teriflunomide exposure [5]. Teri- flunomide is not metabolized by cytochrome P450 (CYP) or flavin monoamine oxidase enzymes [5].
Studies suggest that teriflunomide is an inhibitor of CYP2C8 (in vitro and in vivo evidence) [5], an inducer of CYP1A2 (in vivo evidence) [5], and an inhibitor of BCRP, hepatic uptake transporter OATP1B1/B3 and renal uptake transporter OAT3 (in vitro evidence) [5, 33].Teriflunomide did not affect the pharmacokinetics of bupropion (a CYP2B6 substrate), midazolam (a CYP3A4 substrate), S-warfarin (a CYP2C9 substrate), omeprazole (a CYP2C19 substrate) or metoprolol (a CYP2D6 substrate), when coadministered with these agents [5]. Table 1 shows the pharmacological effects observed and recommenda- tions for management when teriflunomide is coadminis- tered with certain drugs [4, 5].
The pharmacokinetics of teriflunomide were not sig- nificantly affected by mild or moderate hepatic impairment or severe renal impairment [4, 5]. No dosage adjustments are considered necessary in patients with mild or moderate hepatic impairment or mild, moderate or severe renal impairment (not undergoing dialysis [4]) [4, 5]. Teriflun- omide pharmacokinetics have not been investigated in patients with severe hepatic impairment, and teriflunomide is contraindicated in these patients [4, 5]. Teriflunomide is contraindicated in patients with severe renal impairment undergoing dialysis in the EU [4].
Female subjects had a 23 % decrease in teriflunomide clearance from that in male subjects, based on a population analysis [5]. In healthy volunteers and patients with mul- tiple sclerosis, the impact of age, body weight, race, albumin levels and bilirubin levels on teriflunomide phar- macokinetics was limited (B31 %), based on a populations analysis [4].
4 Therapeutic Efficacy
The efficacy of once-daily, oral teriflunomide 7 and 14 mg/ day in patients with remitting multiple sclerosis has been compared with that of placebo in three studies (one phase II study [34] [see Sect 4.1.1] and two phase III [TEMSO [22, 35–37] and TOWER [38–41]] studies [see Sect. 4.1.2]) and with that of interferon beta-1a in one phase III study (TENERE [24]) [see Sect 4.2]. Long-term data from the phase II study and TEMSO are available from their extension studies [42, 43] (see Sect 4.3). While both the 7 and the 14 mg/day dosages are discussed in this article, the 7 mg/day dosage is not approved in the EU [4].
4.1 Efficacy Versus Placebo
4.1.1 Phase II Study
In a double-blind phase II study, 179 patients with relapsing-remitting multiple sclerosis (88 % of patients) or secondary progressive multiple sclerosis with relapse (12 % of patients) were randomized to 36 weeks’ treatment with teriflunomide 7 (n = 61) or 14 mg/day (n = 57) or placebo (n = 61) [34]. Magnetic resonance imaging (MRI) scans were performed every 6 weeks; the primary endpoint was the number of combined unique active lesions (newly and persistently enhancing T1 lesions [pre- and postgado- linium-enhanced T1-weighted scan] and new and enlarging T2 active lesions [unenhanced proton-density T2-weighted scan]) per MRI scan. Following the 36-week double-blind treatment period, patients were permitted to enter a long- term extension study of teriflunomide 7 or 14 mg/day [43]; results from this study are presented in Sect. 4.3. At baseline, the mean number of combined unique active lesions per scan was 1.30, 2.48 and 2.16 in teriflunomide 7 and 14 mg/day and placebo recipients, respectively; the median number at baseline was 0.0, 0.5 and 0.5 [34].
After 36 weeks of treatment, teriflunomide 7 and 14 mg/ day recipients had significantly fewer (p \ 0.03 and p \ 0.01 respectively) combined unique active lesions per scan (adjusted for baseline activity, Expanded Disability Status Scale [EDSS] strata and study site) than placebo recipients; respectively, the mean numbers of combined unique active lesions per scan were 1.04 and 1.06 versus 2.68, leading to a relative reduction of 61.1 % with teri- flunomide 7 mg/day and 61.3 % with teriflunomide 14 mg/ day, versus placebo [34]. The median numbers of com- bined unique active lesions per scan were 0.2 and 0.3 versus 0.5.
The numbers of T1-enhancing lesions per scan and new or enlarging T2 lesions per scan were also significantly lower (p \ 0.05) in teriflunomide 7 and 14 mg/day than in placebo recipients, and recipients of teriflunomide 14 mg/ day had significantly reduced (p \ 0.02) T2 disease burden compared with placebo recipients at 36 weeks [34].
The mean annualized relapse rate was 0.58, 0.55 and 0.81 in teriflunomide 7 mg/day, teriflunomide 14 mg/day and placebo recipients, respectively [34]. Significantly fewer patients receiving teriflunomide 14 mg/day had increased disability (according to EDSS score) than pla- cebo recipients (7.4 vs. 21.3 %; p \ 0.04).
4.1.2 Phase III Studies
A total of 1,088 [22] and 1,169 [38, 39] patients with relapsing multiple sclerosis took part in the TEMSO [22] and TOWER [38, 39] studies, respectively. Patients were randomized to receive teriflunomide 7 or 14 mg/day or placebo for 108 weeks [22] or for a minimum of 48 weeks with a variable (mean of 18 months) follow-up period (study treatment was administered for 48–152 weeks; the core treatment period lasted until 48 weeks after the ran- domization of the last patient) [38, 39]. Randomization was stratified according to baseline EDSS score (B3.5 or [3.5) and trial site [22, 39]. Study details and patient character- istics are shown in Table 2. No significant inter-group differences in baseline characteristics were found in either study [22, 39].
Patients in TEMSO who completed the study were permitted to enter an ongoing, long-term, blinded extension study of teriflunomide 7 or 14 mg/day [42]; results from this study are presented in Section 4.3. Patients in TOWER who completed the study were permitted to enter an open- label extension study of teriflunomide 14 mg/day [38, 39]; results are as yet unavailable for this study. In TEMSO, suspected and confirmed relapses could be treated with intravenous glucocorticoids at the discretion of the treating neurologist [22].
The primary endpoint was the annualized relapse rate in both studies [22, 38, 39]. For definitions of annualized relapse rate and other terms used in this review, please refer to Table 3. Teriflunomide 7 and 14 mg/day were significantly more effective than placebo in patients with relapsing multiple sclerosis [22, 38]. The adjusted annualized relapse rate was significantly lower with both dosages of teriflunomide than with placebo (see Table 4), with a relative reduction versus placebo of 31 % (TEMSO) [22] and 22 % (TOWER) [38, 39] for teriflunomide 7 mg/day and 32 % [22] and 36 %
[38, 39] for teriflunomide 14 mg/day (primary endpoint; see Fig. 3).
A significantly lower proportion of teriflunomide 14 mg/ day recipients experienced confirmed sustained disability progression than placebo recipients (p \ 0.05); terifluno- mide 7 mg/day recipients did not significantly differ from placebo recipients in this outcome (see Table 4 and Fig. 3; hazard ratios versus placebo 0.76 [95 % CI 0.56–1.05] and 0.70 [95 % CI 0.51–0.97] for teriflunomide 7 and 14 mg/ day, respectively, in TEMSO [22], and 0.955 [95 % CI 0.677–1.347 [44]] and 0.685 [95 % CI 0.467–1.004 [44]] in TOWER [38, 39]).
At 48 weeks, the estimated sustained disability pro- gression rate in TOWER was 12.1 and 7.8 versus 14.2 % in teriflunomide 7 and 14 mg/day versus placebo groups, respectively [39]. The treatment effect on annualized relapse rate and time to 12-week sustained disability progression was investi- gated in a pooled analysis (TEMSO plus TOWER) of two subgroups of patients with high disease activity (defined as either: a C2 relapses in the year before study entry; or b use of DMTs in the past 2 years plus either C1 relapse in the year before study entry or C1 Gadolinium-enhancing lesion at baseline, or no DMT use in the past 2 years, EDSS score C1.5, and either C2 relapses in the year before study entry or one relapse with Cgadolinium-enhancing lesion at baseline) (n = 710 and 1,260 for the respective subgroups) [45]. Efficacy results in these subgroups were consistent with those in the main TEMSO and TOWER study populations.The proportions of patients who were free from relapse over the treatment period were significantly (p B 0.01).
Post-hoc analyses of TEMSO and TOWER found that teriflunomide significantly reduced annualized rates of relapses with (investigator-defined) sequelae and relapses leading to hospitalization [41, 47]. In TEMSO, the annu- alized rate of relapses with sequelae (investigator assessed) was reduced by 25 % with teriflunomide 7 mg/day and 53 % with teriflunomide 14 mg/day compared with pla- cebo (p \ 0.0001 for teriflunomide 14 mg/day vs. pla- cebo), and the risk of sequelae per relapse was reduced by 31 % with teriflunomide 14 mg/day compared with pla- cebo (p \ 0.01) [47]. The annualized rate of relapse lead- ing to hospitalization was reduced by 36 % with teriflunomide 7 mg/day (p \ 0.05) and 59 % with teri- flunomide 14 mg/day (p \ 0.0001) compared with pla- cebo, and the risk of hospitalization per relapse with teriflunomide 14 mg/day was reduced by 43 % (p \ 0.001) compared with placebo [47]. In TOWER, the annualized rate of relapse with sequelae with teriflunomide 14 mg/day was reduced by 54 % (p = 0.0004) compared with pla- cebo, and the rate of relapse leading to hospitalization was reduced by 34 % (p = 0.0155) [41]; moreover, terifluno- mide 14 mg/day recipients spent significantly fewer nights in hospital than placebo recipients (1.7 vs. 3.4 nights; p = 0.0285) [41].
4.2 Efficacy Versus Interferon beta-1a
The comparative efficacy of 48–115 (median 63.6) weeks’ past 1 and 2 years, multiple sclerosis subtype, sex or previous use of multiple sclerosis drugs) was found on teriflunomide’s effect on annualized relapse rate; however, sex (p = 0.017) and number of relapses in the past 1 (p = 0.018) and 2 years (p = 0.045) were found to influence the effect of terifluno- mide 14 mg/day on disability progression [40].
Fig. 4 Efficacy of 108 weeks’ treatment with once-daily, oral teriflunomide 7 or 14 mg/day versus placebo in patients with relapsing multiple sclerosis. Magnetic resonance imaging outcomes from the TEMSO study [22, 35]. a Mean change from baseline in lesion volume. b Estimated number of lesions at end of treatment period. The volume of hyperintense lesion components on T2- weighted images was calculated as the portion of the total lesion volume that appeared hyperintense on T2-weighted images but did not appear hypointense on T1-weighted images obtained after the administration of gadolinium. The 7 mg/day dosage is not approved in the EU [4]. Refer to main text and Table 2 for study details. *p \ 0.05, **p \ 0.001 vs. placebo.
Teriflunomide 7 or 14 mg/day does not affect health- related quality of life, as assessed using the Short-Form 36 treatment with oral teriflunomide 7 and 14 mg/day versus subcutaneous interferon beta-1a 44 lg (three doses per week) was investigated in a randomized, single-blind, multicentre, phase III study (TENERE; n = 324) in patients with relapsing multiple sclerosis [24]. Randomi- zation was stratified by country and baseline disability (EDSS score ≤3.5 or [3.5) [24]. The core treatment period lasted until 48 weeks after randomization of the last patient [24]. Teriflunomide recipients were blinded as to dosage; interferon beta-1a treatment was open-label [24]. Interferon beta-1a dosage was titrated from an initial dosage of 8.8 lg three times weekly to the final dosage of 44 lg three times weekly; a dose reduction to 22 lg three times weekly was permitted if patients had tolerability issues [24]. Patients in TENERE who completed the study were permitted to enter an ongoing, long-term, open-label extension study of teri- flunomide 14 mg/day [46]; results are as yet unavailable for this study. Study details and patient characteristics are shown in Table 2.
Fig. 5 Comparative efficacy of 48–115 weeks’ treatment with once- daily, oral teriflunomide 7 or 14 mg/day versus subcutaneous interferon beta-1a 44 lg three times weekly in patients with relapsing multiple sclerosis in the TENERE trial [24]. Cumulative percentage of estimated treatment failures (primary endpoint) at 96 weeks using Kaplan-Meier estimates. The 7 mg/day dosage is not approved in the EU [4]. Refer to main text and Table 2 for study details.
The primary endpoint was the time to treatment failure in the intent-to-treat population during the core treatment period (48–115 weeks) [24]. For the definition of treatment failure, please refer to Table 3. Patients with relapsing multiple sclerosis receiving ter- iflunomide 7 or 14 mg/day did not significantly differ from those receiving interferon beta-1a 44 lg (three doses per week) with regard to treatment failure (primary endpoint) after 48–115 weeks’ treatment (see Fig. 5) [24]. Treatment failure consisted of confirmed relapse (42.2 and 23.4 vs. 15.4 %, respectively), permanent treatment discontinuation (6.4 and 13.5 vs. 24.0 %, respectively) or other reasons (patients who were never treated or received the wrong treatment; 0 and 0.9 vs. 2.9 %, respectively) at 96 weeks [24].
At 48 weeks, the cumulative percentage of estimated treatment failures was 36 and 33 versus 37 % in teriflun- omide 7 and 14 mg/day versus interferon beta-1a recipi- ents, respectively [24]. Annualized relapse rate during the core treatment per- iod, mean change from baseline to 48 weeks in FIS score, and mean Treatment Satisfaction Questionnaire for Medi- cation (TSQM) scores at 48 weeks are presented in Table 5 [24]. No significant difference in annualized relapse rate was observed between recipients of teriflunomide 14 mg/ day and recipients of interferon; however, teriflunomide 7 mg/day was associated with a higher relapse risk than interferon beta-1a (p = 0.03). Teriflunomide 7 mg/day, but not teriflunomide 14 mg/day, was associated with a sig- nificantly (p = 0.03) smaller change in score on the FIS than interferon beta-1a. Mean scores on the TSQM were significantly higher with teriflunomide treatment at either dosage than with interferon beta-1a treatment, aside from teriflunomide 14 mg/day on the effectiveness subscale. A total of 6 % of teriflunomide 7 mg/day, 4 % of teri- flunomide 14 mg/day and 2 % of interferon beta-1a recipients discontinued treatment as a result of lack of efficacy [24].
4.3 Long-term Efficacy
A total of 742 patients with relapsing multiple sclerosis entered the long-term, double-blind, extension study fol- lowing completion of the phase III TEMSO study (see Sect. 4.1.2 for TEMSO study details [22]); data are avail- able for up to 9 years’ exposure to teriflunomide [42]. Long-term data are also available after up to 8.5 years (mean 5.6 years, median 7.1 years) from 147 patients who entered the long-term extension study following
completion of the phase II study (see Sect. 4.1.1) [43]. In both extension studies, patients who initially received ter- iflunomide 7 or 14 mg/day continued on the same dosage; patients who initially received placebo were re-randomized to treatment with teriflunomide 7 or 14 mg/day [42, 43].
Teriflunomide 7 or 14 mg/day appeared to continue to be effective in patients with relapsing multiple sclerosis in the long term, in both the phase III TEMSO study [42] and the phase II study [43].
After up to 9 years in TEMSO, the adjusted annualized relapse rates were 0.225 for recipients of placebo then teriflunomide 7 mg/day, 0.198 for recipients of terifluno- mide 7 mg/day during both periods, 0.177 for recipients of placebo then teriflunomide 14 mg/day, and 0.171 for recipients of teriflunomide 14 mg/day during both study periods [42]. There were no significant between-group differences in the risk of disability progression (sustained for 12 weeks) [remained low] or mean EDSS (remained relatively stable) throughout the extension study. At week 240, the respective changes in total lesion volume were +3.946, +2.521, +3.979 and +3.557; the mean number of gadolinium-enhancing T1 lesions per MRI scan were 0.214, 0.366, 0.444 and 0.294, respectively (lower than baseline for all groups). At the time of analysis, the cumulative duration of exposure to teriflunomide 7 and 14 mg/day was 1,375.60 and 1,342.36 patient-years, respectively, and 468 of the original 1,088 patients remained in the study.
Over the 372-week evaluation period in the phase II extension study, annualized relapse rates decreased in all groups [43]. At week 372, the annualized relapse rate during the extension study treatment period was 0.224 for recipients of placebo then teriflunomide 7 mg/day, 0.296 for recipients of teriflunomide 7 mg/day during both peri- ods, 0.213 for recipients of placebo then teriflunomide 14 mg/day, and 0.181 for recipients of teriflunomide 14 mg/day during both study periods [43]. Respectively, the mean change in EDSS score from baseline was +0.97, +0.23, -0.17 and +0.61, and the mean change from baseline in T2-weighted MRI lesion volume (burden of disease) +5,714.79, +3,846.58, +2,039.98 and +1,969.68 mm3 [43].
5 Tolerability
Oral teriflunomide is generally well tolerated in patients with relapsing multiple sclerosis [22, 24, 38, 39, 42, 43, 48]. The overall incidence of treatment-emergent adverse events ranged from 84.1 to 93.6 % in teriflunomide 7 and 14 mg/day recipients across the three main studies [22, 24, 38, 39]; this was similar to the incidence observed in pla- cebo [22, 38, 39] and interferon beta-1a [24] recipients (Table 6). Moreover, the incidences of serious treatment- emergent adverse events and adverse events leading to death appeared to be generally similar across treatment groups [22, 24, 38, 39] (Table 6). The proportion of patients discontinuing treatment as a result of adverse events was numerically higher in the interferon beta-1a group than in teriflunomide groups [24] (Table 6). Safety data from TEMSO, TOWER and the phase II trial were pooled, showing a safety profile that was consistent with that in the individual trials, and providing data from [2,500 patient-years of teriflunomide exposure [49]. No unexpected safety outcomes were found in this analysis, and both teriflunomide dosages were well tolerated, with generally similar safety profiles [49].
The most common (incidence of ≥10 % in any treat- ment group) treatment-emergent adverse events observed in the TEMSO study are presented in Fig. 6; the most common of these (incidence of ≥15 % in any treatment group) include nasopharyngitis, headache and diarrhoea [22]. Diarrhoea, elevated alanine aminotransferase (ALT) levels (investigator-reported), nausea and thinning hair/ decreased hair density (alopecia) appeared to occur at a greater rate among teriflunomide recipients than among placebo recipients (statistical analyses not reported), and had a dose effect; however, these events rarely led to treatment discontinuation (the discontinuation rate associ- ated with these events was ≤0.5 % for all) [22].
The most common (occurring in ≥8 % of teriflunomide [either dosage] recipients and ≥1 % more frequently than placebo) treatment-emergent adverse events observed in TOWER included headache (14.7 and 12.4 vs. 10.9 % of teriflunomide 7 and 14 mg/day vs. placebo recipients), increased ALT levels (11.2 and 14.0 vs. 8.3 %), thinning hair (10.3 and 13.5 vs. 4.4 %), diarrhoea (12.0 and 11.1 vs. 7.3 %) and nausea (8.3 and 10.2 vs. 8.8 %) [38, 39].
Teriflunomide appeared to be as well tolerated as interferon beta-1a; of the most common adverse events in TENERE (incidence of ≥10 % in any treatment group), nasopharyngitis (25.5, 20.0 and 17.8 %, in teriflunomide 7 and 14 mg/day and interferon beta-1a, respectively), diar- rhoea (22.7, 20.9 and 7.9 %), hair thinning (5.5, 20.0 and 1.0 %), paraesthesia (12.7, 10.0 and 7.9 %) and back pain (9.1, 10.0 and 6.9 %) had a numerically higher incidence in teriflunomide recipients than in interferon beta-1a recipi- ents, and influenza-like illness (3.6, 2.7 and 53.5 %), increased ALT levels (10.9, 10.0 and 30.7 %) and head- ache (20.9, 15.5 and 25.7 %) had a numerically higher incidence with interferon beta-1a than with teriflunomide treatment [24].
The most common serious adverse event in TENERE was increased ALT levels, which occurred in 2.7 % of teriflunomide 7 mg/day, 0.9 % of teriflunomide 14 mg/day and 1.0 % of interferon beta-1a recipients (all serious ALT level elevations were asymptomatic and reversible); all other serious adverse events occurred in ≤1 % of patients in each study group [24]. Increased ALT levels was also the most common adverse event leading to treatment dis- continuation, with 1.8, 3.6 and 8.9 % of patients, respectively.
Most cases of alopecia were diffuse or generalized over the scalp, and mainly occurred during the initial 6 months of treatment; 90 % of patients with alopecia receiving teriflunomide 14 mg/day experienced resolution [4]. Blood pressure elevation may occur with teriflunomide treatment [4]. Increased blood pressure-related adverse events occurred in 5.4 % of teriflunomide 7 mg/day, 5.0 % of teriflunomide 14 mg/day and 3.1 % of placebo recipi- ents in TEMSO; no patients discontinued treatment as a result of increased blood pressure [22]. One patient in TOWER (a teriflunomide 7 mg/day recipient) discontinued treatment as a result of increased blood pressure [38]. The mean change from baseline in supine systolic blood pres- sure was +2.87, +2.72 and -0.99 mmHg, respectively, in TEMSO [22] and +2.55, +2.69 and -0.22 mmHg, respectively, in TOWER [38]. The mean change in supine diastolic blood pressure was +1.38, +1.31 and -0.87 mmHg, respectively, in TEMSO [22], and +1.74, +2.20 and -0.09 mmHg, respectively, in TOWER [38]. In TENERE, the mean increase in systolic blood pressure from baseline to week 48 was 1.49 and 4.70 versus 0.04 mmHg in teriflunomide 7 and 14 mg/day versus interferon beta-1a recipients; the mean increase in diastolic blood pressure was 0.99 and 4.39 versus 0.29 mmHg, respectively [24].
There is a boxed warning regarding hepatotoxicity in the teriflunomide US prescribing information [5]. Severe liver injury (including fatal liver failure and dysfunction) has been reported with leflunomide treatment; a similar risk may be expected with teriflunomide treatment [5]. In TEMSO, the incidence of ALT levels ≥1 times the upper limit of normal (ULN) was higher with teriflunomide 7 and 14 mg/day than with placebo (54.0 and 57.3 vs. 35.9 %); however, the incidence of ALT levels ≥3 times the ULN was similar across treatment groups in both TEMSO (6.3 and 6.7 vs. 6.7 %, respectively) [22] and TOWER (7.6 and 7.8 vs. 5.7 %, respectively) [38, 39]. In TEMSO, one patient in each group had ALT levels ≥3 times the ULN as well as total bilirubin levels ≥2 times the ULN [22]; in TOWER, two teriflunomide 7 mg/day and two placebo recipients had ALT levels ≥3 times the ULN and total bilirubin levels ≥2 times the ULN [38, 39]. Alternative explanations for these elevations included hepatitis C, incidental bile-duct stenosis and cytomegalovirus infection in TEMSO [22] and concomitant cortisone-pulse therapy, hepatitis C, Gilbert’s syndrome and alcoholic liver disease in TOWER [38, 39]. Most elevations occurred within the first year of treatment, and half of the cases returned to normal without treatment discontinuation [5]. In TENERE, a total of 6 % of teriflunomide 7 mg/day, 10 % of teri- flunomide 14 mg/day and 16 % of interferon beta-1a recipients had ALT levels [3 times the ULN [24].
Three teriflunomide recipients developed moderate neu- tropenia (neutrophil count of\0.9 9 109/L) in TEMSO; this resolved spontaneously (after treatment discontinuation in one patient) [22]. In TEMSO, Common Terminology Criteria for Adverse Events (CTCAE) grade 4 reductions in lympho- cyte counts occurred in one patient in each of the teriflunomide groups; CTCAE grade 4 reductions in neutrophil counts occurred in two teriflunomide 7 mg/day recipients [23]. Neutropenia (not further defined) occurred in 7.1 % of teri- flunomide 7 mg/day, 9.4 % of teriflunomide 14 mg/day and
2.9 % of placebo recipients in TOWER [38]. In TENERE, grade 3 neutropenia occurred in 0 % of teriflunomide 7 mg/ day, 1 % of teriflunomide 14 mg/day and 3 % of interferon beta-1a recipients; no grade 4 neutropenia was observed [24]. Grade 3 reductions in lymphocyte counts occurred in 0, 2 and 3 %, respectively, in TENERE, and no grade 4 reductions were observed [24].
Further data regarding the white blood cell counts in clinical trials are discussed in Sect. 2.
Rare cases of pancytopenia, agranulocytosis and thrombocytopenia have been reported with leflunomide in the postmarketing setting; it is logical to assume a similar risk with teriflunomide treatment [5]. No cases of serious pancytopenia have been reported in premarketing clinical trials of teriflunomide [5].
As a drug with immunosuppressant potential, terifluno- mide may increase the risk of infection, including oppor- tunistic infections [4, 5]. However, no increased risk of infection was evident in placebo-controlled trials [4, 5, 22, 38]. Serious infections occurred in 1.6, 2.5 and 2.2 % of teriflunomide 7 and 14 mg/day and placebo recipients, respectively, in TEMSO, and no serious opportunistic infections occurred [22]. Corresponding rates of serious treatment-emergent infections in TOWER were 3.4, 3.0 and 2.9 %, and 0.2, 0.8 and 0.3 % experienced suspected opportunistic infections (hepatitis C concomitant with cytomegalovirus [placebo], bacteraemia due to Gram- negative bacteria [teriflunomide 7 mg/day], and bacterial bilateral pneumonia, bacterial sepsis and intestinal tuber- culosis [all teriflunomide 14 mg/day]) [38, 39]. Two deaths as a result of infection occurred in TOWER (placebo and teriflunomide 14 mg/day) (Table 6) [38, 39]. There was no increased risk of serious infection with teriflunomide 7 or 14 mg/day than with interferon beta-1a in TENERE (1.8 and 1.8 vs. 1.0 %) [24] Fatal infections have been reported in leflunomide recipients, in the postmarketing setting; most cases also involved concomitant immunosuppressant therapy and/or comorbid disease [5].
Some immunosuppressant drugs are associated with an increased risk of malignancy; despite having immunosup- pressant potential, teriflunomide did not appear to increase the risk of malignancy compared with placebo [4, 5, 22]. In TEMSO, four patients had malignant neoplasms (three placebo recipients [one each of breast cancer, thyroid cancer and cervical cancer] and one recipient of teriflun- omide 14 mg/day [cervical cancer in situ, stage 0, after
1.5 years of treatment]) [22]. No patients experienced increased amylase or lipase levels [22]. One placebo reci- pient (and no teriflunomide recipients) had pancreatitis [22].
In patients with rheumatoid arthritis receiving lefluno- mide, rare cases of Steven-Johnson syndrome and toxic epidermal necrolysis have been reported, and a similar risk may be expected with teriflunomide treatment [4, 5]. A total of 10.3 % of teriflunomide 7 mg/day, 11.2 % of ter- iflunomide 14 mg/day and 7.2 % of placebo recipients had hypersensitivity or skin disorders in TEMSO; no cases of anaphylactic shock or serious hypersensitivity reactions occurred [22].
Peripheral neuropathy (including polyneuropathy and mononeuropathy) occurred more frequently among teri- flunomide 7 and 14 mg/day recipients than among placebo recipients (1.2 and 1.9 vs. 0 %, respectively, confirmed by nerve conduction studies) in TEMSO [5]. Treatment was discontinued in two patients with peripheral neuropathy, one of whom recovered following treatment discontinua- tion [5]. In a pooled analysis of TEMSO and TOWER, the incidence of peripheral neuropathy was 2.2 % in recipients of teriflunomide 14 mg/day and 0.6 % in recipients of placebo [4].
A total of 1.2 % of 844 teriflunomide (either dosage) recipients and 0 % of 421 placebo recipients had transient acute renal failure with a creatinine measurement increased by ≥100 % from baseline serum values, in a pooled ana- lysis of TEMSO and phase II data [5]. However, no asso- ciated symptoms were documented, and the creatinine level was normal on the next reported measurement (6–48 days later), despite continued teriflunomide use. Creatinine level increases occurred between 12 weeks and 2 years of initiating teriflunomide treatment [5]. A possible explanation for these cases is acute uric acid nephropathy; teriflunomide is associated with increases in uric acid clearance, leading to mean decreases of 20–30 % in serum uric acid [4, 5].
In a pooled analysis of TEMSO and phase II data, treatment-emergent hyperkalaemia (defined as serum potassium levels [7.0 mmol/L) occurred in 1.0 % of 829 teriflunomide recipients (two patients also had acute renal failure) and 0.2 % of placebo recipients [5]. Mild hypophosphatemia (phosphorus levels of ≥0.6 mmol/L and below the lower limit of normal) occurred in 18 % of teriflunomide and 9 % of placebo recipients in clinical trials; moderate hypophosphatemia (phosphorus levels of ≥0.3 and \0.6 mmol/L) occurred in 5 % and 1 %, respectively [5]. No patients had serum phosphorus levels of \0.3 mmol/L.
In the premarketing database of &2,600 patients exposed to teriflunomide, there have been four reports of cardiovascular deaths, including three sudden deaths and one myocardial infarction (in a patient with a history of hyperlipidaemia and hypertension) [5]. All occurred during uncontrolled extension studies, 1–9 years following treat- ment initiation, and no relationship between teriflunomide and cardiovascular death has been established [5].
As interstitial lung disease and worsening of pre-exist- ing interstitial lung disease has been reported with leflun- omide treatment, a similar risk can be expected with teriflunomide [5].
5.1 In Pregnancy
There is a boxed warning regarding a risk of teratogenicity in the teriflunomide US prescribing information [5]. Teri- flunomide was selectively teratogenic and embryolethal in animal studies, at doses lower than those used clinically, and, based on these studies, may cause major birth defects when administered to a pregnant woman [4, 5]. Pregnancy was reported as an adverse event in TEMSO [22]. Of a total of 11 pregnancies in the trial, four were spontaneous abortions (three in the teriflunomide 14 mg/day group and one in the placebo group) and six were induced abortions (five in the teriflunomide 7 mg/day group and one in the teriflunomide 14 mg/day group). The remaining pregnancy went to term, resulting in a healthy baby, with no health concerns after 2 years. All pregnant patients discontinued treatment and underwent an 11-day elimination period.
Overall, 81 pregnancies have been identified in a ret- rospective analysis of a clinical trial database of teriflun- omide study participants; 69 in patients receiving teriflunomide treatment, the remaining 12 in patients receiving placebo, interferon beta-1a or blinded therapy [50]. Of these pregnancies, 25 resulted in live births (21 in patients receiving teriflunomide; all underwent drug elim- ination in the first trimester [fetal exposure was ≤11 weeks]), 35 in induced abortions (28 in teriflunomide recipients), 14 in spontaneous abortions (13 teriflunomide recipients), and 7 were ongoing pregnancies at the time of analysis (all teriflunomide recipients). In the 21 newborns exposed to teriflunomide, no structural defects or func- tional deficits had been reported at the time of analysis, and mean birth weight and gestational age were consistent with those in the general population.
Teriflunomide has been detected in human semen; the US prescribing information recommends the use of con- traception in men receiving teriflunomide, or, in men wishing to father children, the discontinuation of treatment [5]. In a retrospective analysis of clinical trial data, 20 pregnancies in female partners of 17 male patients (receiving teriflunomide [n = 16] or placebo [n = 4] at the time of pregnancy) were reported, with the following outcomes: 15 live births (12 in partners of teriflunomide recipients), 2 induced abortions (1 in a partner of a teri- flunomide recipient), 1 spontaneous abortion (in a partner of a teriflunomide recipient) and 2 ongoing pregnancies (both in partners of teriflunomide recipients) [50]. All newborns were healthy and free from structural and func- tional abnormalities.
5.2 Long-Term Tolerability
Long-term safety and tolerability results from TEMSO were consistent with those observed in the core study [22, 42]. After a median treatment duration of 722 days (teri- flunomide 7 mg/day) and 762 days (teriflunomide 14 mg/ day), 83.6 % of 383 teriflunomide 7 mg/day recipients and 84.6 % of 357 teriflunomide 14 mg/day recipients had experienced at least one treatment-emergent adverse event, 15.4 and 11.5 % had experienced at least one serious treatment-emergent adverse event, 0.3 and 0.3 % had died as a result of treatment-emergent adverse events, and 9.1 and 6.2 % had discontinued treatment as a result of treat- ment-emergent adverse events [22].
After up to 9 years’ follow-up of TEMSO, the most common adverse events were nasopharyngitis, headache and increased ALT levels, and the most common serious adverse events ([4 patients) were increased ALT levels and intervertebral disc protrusion [42]. Three deaths had occurred at this time; none were considered related to study treatment. Adverse events leading to treatment discontinuation occurred in 82 patients; the most common of these was increased ALT levels. The incidence of first adverse events was highest in the first year of active treatment.
In the 8.5-year follow-up to the phase II study extension, 98.8 % of 81 teriflunomide 7 mg/day and 100 % of 66 teriflunomide 14 mg/day recipients had experienced treat- ment-emergent adverse events [43]. These included infec- tions/infestations (88.9 and 84.8 %, respectively), gastrointestinal disorders (80.2 and 75.8 %), nervous sys- tem disorders (95.1 and 92.4 %), psychiatric disorders (55.6 and 51.5 %), skin/subcutaneous tissue disorders (63.0 and 63.6 %), musculoskeletal/connective tissue disorders (86.4 and 80.3 %), general disorders (75.3 and 78.8 %) and renal/urinary disorders (40.7 and 34.8 %). A total of 29.6 % of teriflunomide 7 mg/day and 28.8 % of teriflun- omide 14 mg/day recipients had increased ALT levels [43]. Serious treatment-emergent adverse events occurred in 35.8 % of teriflunomide 7 mg/day and 28.8 % terifluno- mide 14 mg/day recipients; the most common of these was increased hepatic enzyme levels (6.2 and 7.6 %) [43]. No serious opportunistic infections or hypersensitivity reac- tions were reported [43]. Tolerability results were similar after 9 years of follow- up in the phase II trial (n = 70), with 97.5 and 100 % of teriflunomide 7 and 14 mg/day recipients reporting at least one adverse event, 39.5 and 31.8 % reporting at least one serious adverse event, and 21.0 and 18.2 % experiencing adverse events leading to treatment discontinuation [48].
6 Dosage and Administration
The recommended dosage in the USA for patients with relapsing forms of multiple sclerosis is oral teriflunomide 7 or 14 mg once daily, with or without food [5]. Recom- mended safety monitoring before and during treatment in the USA includes transaminase and bilirubin levels, ALT levels, complete blood cell counts, screening for latent tuberculosis infection, and blood pressure [5].For adult patients with relapsing-remitting multiple sclerosis in the EU, the recommended dosage is oral teri- flunomide 14 mg/day, with or without food [4]. Recom- mended safety monitoring before and during treatment in
the EU includes blood pressure, ALT levels and complete blood cell counts [4].
The US prescribing information has a boxed warning regarding hepatotoxicity and risk of teratogenicity with teriflunomide use [5]. Severe liver injury including fatal liver failure has been reported in leflunomide recipients, and may occur in teriflunomide recipients (see Sect. 5). Patients with pre-existing acute or chronic liver disease or with serum ALT levels [29 the ULN should not normally be treated with teriflunomide. Teriflunomide was selec- tively teratogenic and embryolethal in animal studies, and may cause major birth defects when administered to a pregnant woman (see Sect. 5.1) [5].
Under certain circumstances (e.g. pregnancy), acceler- ated elimination of teriflunomide with cholestyramine or activated charcoal powder is recommended [4, 5]. Vaccination with live vaccines is not recommended while undergoing teriflunomide treatment, and coadminis- tration of teriflunomide and antineoplastic or immunosup- pressive therapies used for the treatment of multiple sclerosis has not been evaluated [4, 5]. The safety and efficacy of teriflunomide in paediatric patients and patients aged [65 years have not been established [4, 5].Local prescribing information should be consulted for further, detailed information, including contraindications, precautions, drug interactions, use in special patient pop- ulations, and accelerated elimination instructions.
7 Place of Teriflunomide in the Management of Relapsing Multiple Sclerosis
The most recent guidelines on the treatment of relapsing multiple sclerosis were published before any oral therapies were approved, and thus emphasize the use of such disease- modifying treatments as interferon beta-1a and -1b and glatiramer acetate, under certain criteria, for relapse pre- vention [51, 52]. Other treatments discussed include aza- thioprine, mitoxantrone, intravenous immunoglobulin, plasma exchange and methylprednisolone [51, 52]. New guidelines from NICE are expected in 2014, which are likely to cover the oral therapies [53].
There are difficulties in the treatment of multiple scle- rosis, not least of which is that the range of available treatments spans from more effective but potentially harmful to less effective but much safer; a highly effective treatment with a generally safe tolerability profile is yet to be discovered [54]. Multiple sclerosis treatment is currently preventive and not curative; adherence is an important consideration for the best possible outcome for the patient, along with individualized safety concerns.
Until recently, treatment for multiple sclerosis has been limited to parenteral therapy. However, several oral therapies are now available, offering a preferred route of administration (allowing, for example, potentially improved adherence, and a lack of injection-site reactions). These therapies have differing mechanisms of action, and thus each provides its own individual efficacy and safety profile, with its own positive and negative aspects [54]. Table 7 presents a summary of the main safety consider- ations associated with the three approved oral DMTs for multiple sclerosis.
The most commonly accepted theory for the disease mechanism of multiple sclerosis is that it is caused by an autoimmune-mediated attack on CNS myelin antigens, lead- ing to inflammation and damage to myelin and axons and the formation of plaques [54]. Other theories include infection and degenerative causes [54]; however, most current treat- ments are based around the autoimmune theory.
Teriflunomide is a DHODH inhibitor, exerting its pro- posed action on patients with multiple sclerosis by inhib- iting T-cell proliferation via the inhibition of the pyrimidine synthesis pathway in rapidly dividing cells, thereby reducing their infiltration into the CNS; additional mechanisms have also been proposed (see Sect. 2). In phase III clinical trials, teriflunomide was more effective than placebo and as effective as interferon beta-1a in the prevention of relapse in patients with relapsing multiple sclerosis (see Sect. 4). The data presented in TOWER were generally similar to those in TEMSO in terms of magnitude of effect on the key outcome measures of annualized relapse rate and sustained disability pro- gression versus placebo. Moreover, teriflunomide was associated with better MRI outcomes than placebo, according to results from one phase III trial (see Sect. 4.1.2). However, quality of life and fatigue impact scores were not significantly affected by teriflunomide treatment. Based on long-term extension trials with smaller patient numbers, teriflunomide continued to be effective for up to 8.5 years (see Sect. 4.3).
Given that teriflunomide and interferon beta-1a had similar efficacy in the primary analysis of TENERE (see Sect. 4.2), patient satisfaction plays a large part in choosing between the drugs. Patient-rated TSQM scores were sig- nificantly higher in teriflunomide 7 mg/day recipients than interferon beta-1a recipients in all four categories (effec- tiveness, side effects, convenience and global satisfaction), and teriflunomide 14 mg/day was associated with signifi- cantly higher scores in three of the four categories (side effects, convenience and global satisfaction) than inter- feron beta-1a. The scores on convenience, in particular, are of interest, as this is a major perceived drawback of the injected drugs.
Overall, teriflunomide was generally well tolerated in clinical trials, with a similar incidence of treatment- emergent adverse events to that observed with placebo and interferon beta-1a treatment (see Sect. 5). Teriflunomide appeared to be as well tolerated as interferon beta-1a; diarrhoea and thinning hair appeared to occur more fre- quently in teriflunomide recipients than interferon beta-1a recipients, and influenza-like illness and increased ALT levels appeared to occur more commonly with interferon beta-1a than with teriflunomide treatment.
Long-term tolerability results were consistent with those observed in the shorter studies (see Sect. 5.2). As teri- flunomide is the main metabolite of leflunomide, some extrapolation can be made from the large amount of data built up regarding leflunomide safety. Approximately 2.5 million patient-years of exposure have been catalogued since the approval of leflunomide in 1998 (for the treatment of active rheumatoid arthritis); during this time, five cases of progressive multifocal leukoencephalopathy (PML) have been reported in leflunomide recipients [44]. Longer observation periods are required to see if the risk of this rare adverse event extends to teriflunomide treatment (no cases of PML have been observed in the teriflunomide clinical program or extension studies to date), both in the general multiple sclerosis population (now that terifluno- mide is approved) and in further, large trials over longer periods of treatment.
Teriflunomide is associated with increased ALT levels, and there is a boxed warning regarding hepatotoxicity in the US prescribing information. There is also a boxed warning regarding a risk of teratogenicity, and terifluno- mide is contraindicated in pregnant patients, based on data from animal studies; as ethical reasons prohibit pregnancy trials, data will slowly become available regarding teri- flunomide safety in pregnant humans from individual, likely mostly accidental, cases (see Sect. 5.1). A pregnancy registry is currently underway in the US to collect infor- mation from pregnancy cases; a similar registry is planned in the EU.
As a drug with immunosuppressant potential, terifluno- mide may increase the risk of infection, including oppor- tunistic infections; however, no increased risk of infection was evident in placebo-controlled trials (see Sect. 5). While teriflunomide has a long elimination half-life, and thus has the potential to remain in the system for a long time (Sect. 3), it can be rapidly eliminated under certain circumstances (e.g. pregnancy, possible overdose or emerging serious adverse effects), with cholestyramine or activated charcoal powder [4, 5].The use of teriflunomide as a single agent in 618 patients with a first clinical episode consistent with multi- ple sclerosis was also investigated (the TOPIC trial) [58]. Results from this trial (available as an abstract) demon- strate that teriflunomide 7 and 14 mg/day both significantly reduce the risk of conversion from clinically isolated
syndrome to clinically definite multiple sclerosis, as com- pared with placebo (by 37 % [p \ 0.05] and 43 % [p \ 0.01], respectively), over a 2-year study period [59]. Teriflunomide as concomitant treatment with interferon beta therapy in patients with relapsing multiple sclerosis was also being investigated; however, this study (TERA- CLES) has been terminated, for reasons not linked to safety concerns [60]. Two studies have been completed investi- gating the efficacy of teriflunomide as adjunctive therapy with glatiramer acetate [61] and interferon beta [62]; a long-term extension study of these has also been completed [63]. None of these treatment options are currently approved, and these studies are not discussed further in this review.
Head-to-head comparisons of teriflunomide with other approved oral multiple sclerosis drugs (i.e. fingolimod and dimethyl fumarate) would be of great interest, as would comparisons with the other approved DMTs: interferon beta-1b, glatiramer acetate, mitoxantrone and natalizumab. A study investigating the actual treatment adherence dif- ferences between oral and parenteral DMTs would also be of obvious benefit, to confirm or conflict with the theory that oral drugs improve patient adherence to treatment, thus improving patient outcomes longterm. Data from TENERE indicate that only one patient discontinued treatment as a result of poor compliance; that patient was in the interferon beta-1a treatment group [24]. Further analysis from TE- NERE on treatment compliance would be of interest.
No formal economic analysis has been conducted as yet to determine the cost effectiveness of teriflunomide in comparison with other multiple sclerosis drugs; this would also be of interest, once more comparative efficacy trials have been conducted.While further data are required to accurately place ter- iflunomide in the management of multiple sclerosis, it has been speculated that, based on the proposed mechanism of action, and using the immune model of multiple sclerosis pathogenesis, teriflunomide may be at its most effective during the early stages of the disease up to the relapsing- remitting stage, when autoimmune responses and inflam- matory disease can potentially be controlled by the inhi- bition of peripheral activation of T cells [1]. It would also be a viable option for those patients who may prefer to use oral drugs (e.g. patients with needle phobia or those with injection-related adverse events).
In conclusion, teriflunomide is effective and generally well tolerated in patients with relapsing multiple sclerosis, and is as effective and well tolerated as interferon beta-1a in these patients. As an oral treatment, it offers an alter- native to the traditional, parenteral DMTs; however, further investigation into the efficacy and/or tolerability differ- ences between teriflunomide and other available oral drugs would be of great use in the placement of this drug. At present, given the relatively limited long-term data, it is difficult to draw definite conclusions with regard to safety; however, as teriflunomide is the main active metabolite of leflunomide, long-term safety data can be extrapolated from the large amount of post-approval data available regarding its parent drug. Oral teriflunomide is a valuable addition to available treatment options for patients with relapsing multiple sclerosis, in particular those patients who prefer an oral drug.
Disclosure The preparation of this review was not supported by any external funding. During the peer review process,the manufacturer of the agent under review was offered an opportunity to comment on this article. Changes resultingfrom comments received were made by the authors on the basis of scientific and editorial merit.
References
1. Gold R, Wolinsky JS. Pathophysiology of multiple sclerosis and the place of teriflunomide. Acta Neurol Scand. 2011;124(2): 75–84.
2. Gold R. Oral therapies for multiple sclerosis: a review of agents in phase III development or recently approved. CNS Drugs. 2011;25(1):37–52.
3. Girouard N, Soucy N. Patient considerations in the management of multiple sclerosis: development and clinical utility of oral agents. Patient Prefer Adherence. 2011;5:101–8.
4. Sanofi-Aventis. Aubagio® (teriflunomide film-coated tablets) EU
summary of product characteristics. 2013. http://www.ema.europa. eu/docs/en_GB/document_library/EPAR_-_Product_Information/ human/002514/WC500148682.pdf. Accessed 10 Sept 2013.
5. Sanofi. Aubagio® (teriflunomide oral tablets): US prescribing
information. 2012. http://products.sanofi.us/aubagio/aubagio.pdf.
Accessed 8 Aug 2013.
6. Claussen MC, Korn T. Immune mechanisms of new therapeutic strategies in MS: teriflunomide. Clin Immunol. 2012;142(1): 49–56.
7. Prakash A, Jarvis B. Leflunomide: a review of its use in active rheumatoid arthritis. Drugs. 1999;58(6):1137–64.
8. Posevitz V, Chudyka D, Hucke S, et al. The anti-proliferative effect of teriflunomide correlates with antigen affinity [abstract no. P974 plus poster]. In: 29th Congress of the European Com- mittee for Treatment and Research in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
9. Greene S, Watanabe K, Braatz-Trulson J, et al. Inhibition of dihydroorotate dehydrogenase by the immunosuppressive agent leflunomide. Biochem Pharmacol. 1995;50(6):861–7.
10. Davis JP, Cain GA, Pitts WJ, et al. The immunosuppressive metabolite of leflunomide is a potent inhibitor of human dihy- droorotate dehydrogenase. Biochemistry. 1996;35(4):1270–3.
11. Knecht W, Loffler M. Species-related inhibition of human and rat dihydroorotate dehydrogenase by immunosuppressive isoxazol and cinchoninic acid derivatives. Biochem Pharmacol. 1998;56(9):1259–64.
12. Ru¨ckemann K, Fairbanks LD, Carrey EA, et al. Leflunomide inhibits pyrimidine de novo synthesis in mitogen-stimulated T-lymphocytes from healthy humans. J Biol Chem. 1998;273(34):21682–91.
13. Cherwinski HM, Byars N, Ballaron SJ, et al. Leflunomide interferes with pyrimidine nucleotide biosynthesis. Inflamm Res. 1995;44(8):317–22.
14. Cherwinski HM, Cohn RG, Cheung P, et al. The immunosup- pressant leflunomide inhibits lymphocyte proliferation by inhib- iting pyrimidine biosynthesis. J Pharmacol Exp Ther. 1995;275(2):1043–9.
15. Cao WW, Kao PN, Chao AC, et al. Mechanism of the antipro- liferative action of leflunomide. A77 1726, the active metabolite of leflunomide, does not block T-cell receptor-mediated signal transduction but its antiproliferative effects are antagonized by pyrimidine nucleosides. J Heart Lung Transplant. 1995;14(6 Pt 1):1016–30.
16. Li L, Liu J, Ringheim G, et al. The effects of teriflunomide on lymphocyte subpopulations in human peripheral blood mononu- clear cells [abstract no. P938]. Mult Scler. 2011;10(Suppl. 1):S422–3.
17. Siemasko KF, Chong AS, Williams JW, et al. Regulation of B cell function by the immunosuppressive agent leflunomide. Transplantation. 1996;61(4):635–42.
18. Pachner A, Keyes L, Gilli F. Teriflunomide treatment decreases chemokine expression in the CNS in the Theiler’s virus-induced demyelinating disease model of multiple sclerosis [abstract no. P369 plus poster]. In: 29th Congress of the European Committee for Research and Treatment in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
19. Merrill JE, Hanak S, Pu SF, et al. Teriflunomide reduces behavioral, electrophysiological, and histopathological deficits in the Dark Agouti rat model of experimental autoimmune encephalomyelitis. J Neurol. 2009;256(1):89–103.
20. Iglesias-Bregna D, Hanak S, Ji Z, et al. Effects of prophylactic and therapeutic teriflunomide in transcranial magnetic stimula- tion-induced motor-evoked potentials in the dark agouti rat model of experimental autoimmune encephalomyelitis. J Pharmacol Exp Ther. 2013;347(1):203–11.
21. Ringheim G, Lee L, Laws-Ricker L, et al. Teriflunomide atten- uates immunopathological changes in the Dark Agouti rat model of experimental autoimmune encephalomyelitis [abstract no. P448 plus poster]. In: 27th Congress of the European Committee for Treatment and Research in Multiple Sclerosis; 19–22 Oct 2011; Amsterdam.
22. O’Connor P, Wolinsky JS, Confavreux C, et al. Randomized trial of oral teriflunomide for relapsing multiple sclerosis. N Engl J Med. 2011;365(14):1293–303.
23. Comi G, Benzerdjeb H, Wang L. Effect of teriflunomide on lymphocyte and neutrophil levels in patients with relapsing multiple sclerosis: results from the TEMSO study [abstract no. P995]. In: 28th Congress of the European Committee for Treat- ment and Research in Multiple Sclerosis; 10–13 Oct 2012; Lyon.
24. Vermersch P, Czlonkowska A, Grimaldi LM, et al. Teriflunomide versus subcutaneous interferon-beta 1a in patients with relapsing multiple sclerosis: a randomised, controlled phase 3 trial. Mult Scler (in press).
25. Pachner A, Li L. Effect of teriflunomide on the viral load and anti-viral antibody responses in the Theiler’s virus model of MS [abstract no. P02.143]. Neurology. 2012;78(1 Meeting Abstracts).
26. Bar-Or A, Freedman MS, Kremenchutzky M, et al. Teriflunomide effect on immune response to influenza vaccine in patients with multiple sclerosis. Neurology. 2013;81(6):552–8.
27. Bar-Or A, Larouche R, Legrand B, et al. Immune respose to neoantigen and recall antigens in healthy subjects receiving ter- iflunomide [abstract no. P622]. In: 29th Congress of the European Committee for Research and Treatment in Multiple Sclerosis; 2-5 Oct 2013; Copenhagen.
28. Menguy-Vacheron F, Msihid J, Poitiers F, et al. Lack of effect on cardiac repolarisation with teriflunomide compared with placebo: a thorough ECG study [abstract no. P458]. Mult Scler. 2011;17(10 Suppl. 1):S192.
29. Limsakun T, Menguy-Vacheron F. Pharmacokinetics of oral teriflunomide, a novel oral disease-modifying agent under investigation for the treatment of multiple sclerosis [abstract no. P05.032]. In: 62nd Annual Meeting of the American Academy of Neurology; 10–17 Apr 2010; Toronto.
30. Wang L, Turpault S, Zeng Z. In vivo comparative metabolism of [14C]-labelled teriflunomide in mice, rats, rabbits, dogs and humans [abstract no. P1533]. Eur J Neurol. 2011;18(Suppl. s2):262.
31. Miller A, Turpault S, Menguy-Vacheron F. Rapid elimination procedure of teriflunomide with cholestyramine or activated charcoal [abstract no. P10]. In: Fourth Cooperative Meeting of CMSC and ACTRIMS; 30 May–2 Jun 2012; San Diego.
32. Turpault S, Miller B, Menguy-Vacheron F. No adverse impact of repeated oral doses of teriflunomide on the pharmacokinetics of oral contraceptive steroids (ethinylestradiol and levonorgestrel) in young healthy female subjects [abstract no. PI-60]. Clin Phar- macol Ther. 2012;91(Suppl. 1):S29–30.
33. Weitz D, Schmider W, Menguy-Vacheron F, et al. Teriflunomide: potential for transporter mediated drug-drug interactions [abstract no. PIII-72]. Clin Pharmacol Ther. 2013;93(Suppl. 1):S112–3.
34. O’Connor PW, Li D, Freedman MS, et al. A Phase II study of the safety and efficacy of teriflunomide in multiple sclerosis with relapses. Neurology. 2006;66(6):894–900.
35. Wolinsky JS, Narayana PA, Nelson F, et al. Magnetic resonance imaging outcomes from a phase III trial of teriflunomide. Mult Scler. 2013. doi:10.1177/1352458513475723.
36. Miller AE, O’Connor P, Wolinsky JS, et al. Pre-specified sub- group analyses of a placebo-controlled phase III trial (TEMSO) of oral teriflunomide in relapsing multiple sclerosis. Mult Scler. 2012;18(11):1625–32.
37. Miller A, Lublin F, O’Connor P, et al. Effect of teriflunomide on relapses with sequelae and relapse leading to hospitalization in a population with relapsing forms of multiple sclerosis: results from the TEMSO study [abstract no. S30.003]. Neurology. 2012;78(1_MeetingAbstracts).
38. Kappos L, Comi G, Confavreux C, et al. The efficacy and safety of teriflunomide in patients with relapsing MS: results from TOWER, a phase III, placebo-controlled study [abstract no. 153]. In: 28th Congress of the European Committee for Treatment and Research in Multiple Sclerosis; 10–13 Oct 2012; Lyon.
39. Miller AE, Kappos L, Comi G, et al. Teriflunomide efficacy and safety in patients with relapsing multiple sclerosis: results from TOWER, a second, pivotal, phase 3 placebo-controlled study [abstract no. S01.004 plus oral presentation]. In: 65th annual meeting of the American Academy of Neurology; 16–23 Mar 2013; San Diego.
40. Moses H, Freedman M, Kappos L, et al. Pre-defined subgroups analyses of TOWER, a placebo-controlled phase 3 trial of
teriflunomide in patients with relapsing multiple sclerosis [abstract no. S41.006]. Neurology. 2013;80(1_MeetingAbstracts).
41. O’Connor P, Lublin F, Wolinsky J, et al. Teriflunomide reduces relapse-related sequelae, hospitalizations and corticosteroid use: a post-hoc analysis of the phase 3 TOWER study [abstract no. P07.109]. Neurology. 2013;80(1_MeetingAbstracts).
42. Freedman M, Wolinsky J, Comi G, et al. Long-term safety and efficacy of teriflunomide in patients with relapsing forms of multiple sclerosis in the TEMSO extension trial [abstract no. P544 plus poster]. In: 29th Congress of the European Committee for Research and Treatment in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
43. Confavreux C, Li DK, Freedman MS, et al. Long-term follow-up of a phase 2 study of oral teriflunomide in relapsing multiple sclerosis: safety and efficacy results up to 8.5 years. Mult Scler. 2012;18(9):1278–89.
44. Data on file, Genzyme, 2013.
45. Kappos L, Comi G, Freedman M, et al. Pooled efficacy data from two phase 3 placebo-controlled trials of oral, once-daily teri- flunomide [abstract no. P618 plus poster]. In: 29th congress of the European Committee for Treatment and Research in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
46. Sanofi. A study comparing the effectiveness and safety of teri- flunomide and interferon beta-1a in patients with relapsing mul- tiple sclerosis (TENERE) [CliniclTrials.gov identifier NCT00883337]. US National Institutes of Health, Clinical Tri- als.gov. 2013. http://clinicaltrials.gov/. Accessed 1 Mar 2013.
47. O’Connor PW, Lublin FD, Wolinsky JS, et al. Teriflunomide reduces relapse-related neurological sequelae, hospitalizations and steroid use. J Neurol. 2013;260(10):2472–2480.
48. Confavreux C, O’Connor P, Freedman MS, et al. Long-term safety and tolerability of teriflunomide in multiple sclerosis: 9-year follow-up of a phase II study [abstract no. P914]. Mult Scler. 2011;17(10 Suppl. 1):S409–10.
49. Leist T, Freedman M, Kappos L, et al. Pooled safety data from three placebo-controlled teriflunomide studies [abstract no. P633 plus poster]. In: 29th congress of the European Committee for Treatment and Research in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
50. Kieseier B, Benamor M, Delhay J-L, et al. Updated pregnancy outcomes from the teriflunomide clinical development pro- gramme: retrospective analysis of the teriflunomide clinical trial database [abstract no. P541 plus poster]. In: 29th congress of the European Committee for Treatment and Research in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
51. National Institute for Health and Care Excellence. Multiple sclerosis: management of multiple sclerosis in primary and sec- ondary care. 2003. http://guidance.nice.org.uk/CG8/NICEGuidance/ pdf/English. Accessed 8 Aug 2013.
52. Goodin DS, Frohman EM, Garmany GP Jr, et al. Disease mod- ifying therapies in multiple sclerosis: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and the MS Council for Clinical Practice Guidelines. Neurology. 2002;58(2):169–78.
53. National Institute for Health and Care Excellence. Guidance in development: multiple sclerosis (2014). 2013. http://www.nice.
org.uk/guidance/index.jsp?action=byID&o=13595. Accessed 8
Aug 2013.
54. Fontoura P, Garren H. Multiple sclerosis therapies: molecular mechanisms and future. Results Probl Cell Differ. 2010;51: 259–85.
55. Biogen Idec Inc. TecfideraTM (dimethylfumarate oral delayed- release capsules): US prescribing information. 2013. http://www. accessdata.fda.gov/drugsatfda_docs/label/2013/204063lbl.pdf. Accessed 8 Aug 2013.
56. Novartis Pharmaceuticals Corporation. GilenyaTM (fingolimod capsules): US prescribing information. 2012. http://www.pharma. us.novartis.com/cs/www.pharma.us.novartis.com/product/pi/pdf/ gilenya.pdf. Accessed 8 Aug 2013.
57. Novartis Pharmaceuticals Corporation. GilenyaTM (fingolimod capsules):EU summary of product characteristics. 2012. http:// www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_ Product_Information/human/002202/WC500104528.pdf. Accessed 18 Sept 2013.
58. Sanofi. Phase III study with teriflunomide versus placebo in patients with first clinical symptom of multiple sclerosis (TOPIC) [ClinicalTrials.gov identifier NCT00622700]. US National Insti- tutes of Health, ClinicalTrials.gov. 2013. http://clinicaltrials.gov/. Accessed 8 Aug 2013.
59. Miller A, Wolinsky J, Kappos L, et al. TOPIC main outcomes: efficacy and safety of once-daily oral teriflunomide in patients with clinically isolated syndrome [abstract no. 99 plus oral pre- sentation]. In: 29th Congress of the European Committee for Research and Treatment in Multiple Sclerosis; 2–5 Oct 2013; Copenhagen.
60. Sanofi. Efficacy and safety of teriflunomide in patients with relapsing multiple sclerosis and treated with interferon-beta (TERACLES) [ClinicalTrials.gov identifier NCT01252355]. US National Institutes of Health, ClinicalTrials.gov. 2013. http:// clinicaltrials.gov/. Accessed 8 Aug 2013.
61. Sanofi. Phase II study of teriflunomide as adjunctive therapy to glatiramer acetate in subjects with multiple sclerosis [Clinical- Trials.gov identifier NCT00475865]. US National Institutes of Health, ClinicalTrials.gov. 2012. http://www.clinicaltrials.gov. Accessed 8 Aug 2013.
62. Freedman MS, Wolinsky JS, Wamil B, et al. Teriflunomide added to interferon-beta in relapsing multiple sclerosis: a randomized phase II trial. Neurology. 2012;78(23):1877–85.
63. Sanofi. Long term safety of teriflunomide when added to interferon- beta or glatiramer acetate in patients with multiple sclerosis [ClinicalTrials.gov identifier NCT00811395]. US National Insti- tutes of Health, ClinicalTrials.gov. 2012. http://www.clinicaltrials.gov. Accessed 8 Aug 2013.