|Home | About | Journals | Submit | Contact Us | Français|
The development of new pharmacologic agents for the treatment of multiple sclerosis (MS) and advances in testing for exposure to the JC virus have led to changes in the treatment of MS. In addition several new agents are in late stage development for MS and their entry onto the market will provide additional treatment options. In 2012 and in early 2013, it is likely that both terifunomide and BG-12 will be approved by the United States Food and Drug Administration (FDA) for the treatment of relapsing forms of MS. The therapeutic environment has already changed and is likely to change rapidly over the next several years. Fingolimod was the first oral agent approved for the treatment of MS and this agent is now widely used in patients intolerant of injections and the side effects associated with the older platform therapies. In many settings it is also used a first-line agent. Owing to the risk of progressive multifocal leukoencephalopathy, natalizumab had previously been reserved for patients with active disease who were intolerant of first-line agents or patients who were worsening despite standard therapy. With the availability of JC virus antibody testing, natalizumab is now being used as a first-line agent in patients negative for JC virus antibodies. Teriflunomide and BG-12 will become available in the next year. Both agents have suitable efficacy and a favorable safety and tolerability profile. There are advantages and disadvantages associated with all of the oral agents. In this article we summarize the clinical trial results regarding the efficacy and safety of the oral agents and discuss the changes that are already taking place in the therapeutic landscape for MS.
The therapeutic landscape in the treatment of multiple sclerosis (MS) is now evolving quite rapidly and will continue to do so for the next several years. This has occurred after many years in which four first-line agents (glatiramer acetate, interferon [IFN] β-1b, IFN β-1a intramuscular [IM], and interferon β-1a subcutaneous [SQ]) constituted the principal treatment options. A variety of new agents potentially useful for the treatment of MS are now in phase II and III clinical trials and may soon be available in the United States and the European Union. In October 2010, the first oral agent was approved by the United States Food and Drug Administration (FDA). Fingolimod proved effective in decreasing relapse rates (RRs), new magnetic resonance imaging (MRI) lesions, disability progression, and brain volume loss in patients with relapsing–remitting MS (RRMS) [Kappos et al. 2010]. It was also proven superior to once weekly IFNβ-1a in decreasing RRs and new MRI lesions [Cohen et al. 2010]. While fingolimod was the first oral agent approved, several other oral agents are currently in phase III trials and are currently, or will soon be submitted to the FDA for approval. These agents include teriflunomide and oral fumarate (BG-12). In addition to the oral agents, several monoclonal antibodies are also in late stage development for MS. The purpose of this article is to provide a brief review of the changing treatment paradigm in relapsing forms of MS. The discussion is restricted to agents that will likely become available in the next year and to changes in the way currently available agents are being used.
Natalizumab (NTZB) is a highly effective therapy indicated to decrease the frequency of relapses and to reduce disability progression in patients with RRMS. It is a monoclonal antibody against VLA-4 and prevents the trafficking of lymphocytes from the periphery into the central nervous system (CNS) thereby preventing inflammation. In clinical trials with NTZB relapses were decreased by 68% and 3-month disability progression was reduced by 42%. Its effects on MRI measures of disease activity were robust. New T2 lesions were decreased 82% and gadolinium-enhanced lesions were decreased by 92% [Polman et al. 2006]. It was first approved by the FDA in November in 2004 but withdrawn from the market in February of 2005 after three cases of progressive multifocal leukoencephalopathy (PML) were reported in patients who had participated in clinical trials. PML is a lytic infection of oligodendrocytes caused by the John Cunningham (JC) virus [Major, 2010]. The JC virus is a ubiquitous virus that is not known to cause disease in nonimmunocompromised patients. It can be found in kidney, spleen, and tonsilar epithelium in normal individuals. In immunocompromised patients the virus can reactivate and cause PML. Patients with PML present with progressive neurological deficits involving neuropsychiatric manifestations, hemiparesis, cortical blindness, seizures, or myoclonus. It is associated with large, multifocal cavitary lesions often leading to death or severe disability. PML is usually encountered in severely immunocompromised patients such as those with HIV or in those with hematologic malignancies following immunosuppressive treatment. This infection had never been seen in MS patients regardless of exposure to multiple immunosuppressive agents such as cyclophosphamide, methotrexate, mitoxantrone, or azathioprine. The appearance of three cases in patients exposed to NTZB led to the withdrawal of NTZB form the market in both the US and Europe. A careful analysis of all clinical trial participants suggested the risk of PML was approximately 1.7/1000 after 18 months of therapy. As of June 2012, 258 cases of NTZB associated PML have been reported worldwide.
Natalizumab was re-introduced in July 2006 after it became apparent that it was effective in patients with highly active disease. While head-to-head comparison trials with other agents have not been carried out NTZB is widely accepted as being superior in efficacy. In the AFFIRM trial, NTZB brought about a 68% decrease in RRs and was highly effective in decreasing MRI measures of disease activity [Polman et al. 2006]. New T2 lesions were decreased 82% and gadolinium-enhanced lesions were decreased by 92%. In a post hoc analysis looking highly at patients with highly active disease (defined as those with at least two relapses in the year prior to study entry and at least one gadolinium-enhanced [GD] lesion on the baseline MRI) NTZB decreased RR 84% [Hutchinson et al. 2009]. Subsequent post hoc analysis suggested that as many as 70% of NTZB treated patients had improvements of at least 0.5 Expanded Disability Status Scale (EDSS) points [Munschauer et al. 2008]. This substantiates the clinical impression that many patients improve while on NTZB treatment.
Following its re-introduction, NTZB became the drug of choice for patients with worsening MS despite standard therapy. Owing to the risk of PML, NTZB had been used only in patients failing standard therapy or in those with active disease who were intolerant or could not take any other first-line agent. With the recent availability of JC virus antibody testing NTZB is now being utilized as a first-line agent in patients who are negative for JC virus antibody [Bozic et al. 2011]. If patients test negative for JC virus antibodies, it suggests that they have not been exposed to the virus. In this population of patients the risk of PML is thought to be negligible because they do not harbor the virus.
The test for the JC virus antibody is a two-step assay that reliably determines prior exposure to the JC virus [Bozic et al. 2011]. If there has been no prior exposure to the JC virus, theoretically it should not be possible for those without JC virus exposure to develop PML. In those with prior exposure the JC virus resides in the kidney, tonsillar epithelium, and spleen [Major, 2010], it remains dormant and asymptomatic unless a series of events including immunosuppression lead to its reactivation and to PML. It is known that PML is caused by the JC virus and if patients have not been exposed to the JC virus the risk of PML should be zero. There a couple of issues deserving of consideration. First, the assay has a false-negative rate of 2.5% [Bozic et al. 2011]. Roughly 1 in 50 patients testing negative for the JC virus will have prior exposure (i.e. infection with the JC virus but will test negative). Testing for patients who are JC virus negative is carried out yearly. The chance of having two false-negative tests in a row (a year apart) will be 1/2500. NTZB-associated PML is a time-dependent event. In patients with no prior history of immunosuppressive exposure the risk of PML with treatment duration under 2 years is approximately 1/2000. Thus, if a patient has a false-negative test or new exposure to the JC virus they can remain on NTZB for 2 years before the risk of PML increases to approximately 1/250. If there is a prior history of immunosuppressive exposure the risk of PML is four times greater regardless of treatment duration or length of washout prior to starting NTZB [Fox and Rudick, 2012]. In patients with a prior history of immunosuppressive exposure the risk of PML with treatment duration over 2 years is 1/90. Therefore, in patients with prior exposure to immunosuppressive agents a false-negative test is of much greater importance. However, PML is still exceedingly rare in patients with less than 1 year of exposure to NTZB.
The result of having the availability of a test that determines whether a patient has been exposed to the JC virus is that NTZB may be used as a first-line agent in patients who test negative for the JC virus. Increasingly, this is the case in many MS centers. NTZB appears to have superior efficacy despite the lack of head-to-head trials proving superiority and in patients who are negative for JC virus it is an appropriate first-line agent.
Fingolimod is a sphingosine-1-phosphate analogue that acts as a functional antagonist of S1P receptors [Chun and Hartung, 2010]. These receptors are present on lymphocytes and follow a gradient of S1P that allows these cells to egress from lymphoid organs. Binding of phosphorylated fingolimod to the S1P receptor results in stimulation of the receptor and internalization of the receptor into the cell. The net result is a functional antagonism of the receptor. This effectively traps lymphocytes in lymphoid organs decreasing the availability of activated T lymphocytes so they cannot cross the blood–brain barrier and enter the CNS. Importantly, only naive T cells and memory T cells are sequestered [Chun et al. 2010]. Effector memory T cells that are CCR7 negative are not sequestered so that immune surveillance takes place normally. The result is decreased T-cell-mediated autoimmune CNS inflammation. While this mechanism is likely responsible for the increased risk of infection observed with fingolimod the actual mechanism of action probably takes place within the CNS. Sphingosine-1-phosphate receptors are widely distributed throughout many tissues including the CNS and are present on neurons, oligodendocytes, oligondendrocyte progenitor cells, astrocytes, and microglial cells [Gardell et al. 2007].
Fingolimod readily penetrates the CNS and it accumulates within the myelin fraction [Foster et al. 2007]. The onset of effectiveness as measured by monthly gadolinium-enhanced MRI begins about 5 weeks after daily administration is begun [Kappos et al. 2006]. Lymphocyte sequestration takes place with 72 hours. No effect on the frequency of enhancing lesion is seen for about 5 weeks. This is exactly the time frame in which the accumulation of fingolimod is seen within the CNS. Further, when administered intrathecally in animals with experimental allergic encephalomyelitis (EAE) it is active suggesting that the peripheral immune system is not required for an effect on CNS inflammation [Foster et al. 2007]. Within the CNS it binds to neurons, oligindendrocyte progenitor cells, astrocytes, and microglia [Foster et al. 2007; Jung et al. 2007; Coelho et al. 2007]. The efficacy of fingolimod in MS is probably due to its effect within the CNS and not on the peripheral immune system.
Fingolimod decreased relapses 54% versus placebo [Kappos et al. 2010]. It also showed efficacy in decreasing disability progression (30%), and in decreasing the accumulation of new T2-weighted lesions (74%), and gadolinium-enhanced lesions (82%). In an active comparator trial fingolimod decreased RRs 52% versus IFBβ-1a [Cohen et al. 2010]. Compared with IFNβ-1a fingolimod decreased new T2 lesions 65% (2.1 versus 1.5, p = 0.053). Gadolinium-enhanced lesions were decreased 55% compared with IFNβ-1a (0.51 versus 0.23, p < 0.0004). While there was no difference in disability progression between patients treated with IFNβ-1a versus fingolimod it should be pointed out that it was a 1-year trial in patients with relatively early RRMS and only a very small proportion of patients would be expected to show confirmed progression over a 1-year period.
In regard to safety and tolerability, fingolimod was generally well tolerated. Common side effects included diarrhea, back pain, alopecia, eczema, and asthenia. In the FREEDOMS trial 81% of the fingolimod 0.5 mg treated patients and 72% of placebo-treated patients completed the study on investigational product. Safety concerns were a greater issue. Fingolimod was associated with first-dose bradycardia, increased risk of bronchitis and influenza, lymphopenia, macular edema, and elevated liver function tests. There may also be a risk of fetal malformation in women of child-bearing potential. Since S1P receptors are involved in embryogenesis of the vascular system fingolimod could cause fetal malformations. In animal studies it was associated with ventricular septal defects and persistent truncus arteriosus.
The next agent most likely to receive FDA approval for the treatment of RRMS is teriflunomide. It is a dihydro-oratate dehydrogenase inhibitor that blocks pyrimidine synthesis [Fox et al. 1999]. It is the active metabolite of leflunomide (Arava®) used in the treatment of rheumatoid arthritis. Its mechanism of action may be related to its effect on pyrimidine synthesis but could also be more complex. It appears to bring about a shift from a Th1 cytokine profile to a Th2 profile [Merrill et al. 2009]. This is associated with an anti-inflammatory effect. It might also suppress the immune response by preventing clonal expansion of B and T cells and antibody production. In animal studies it has been shown to decrease inducible nitric oxide synthetase which produces damaging free radicals released by macrophages and astrocytes [Miljkovic et al. 2001]. These free radicals have been implicated as important mediators of neurodegeneration in MS lesions.
In phase II clinical trials it was administered once daily at 7 or 14 mg versus placebo [O’Connor et al. 2006]. The 7 mg dose decreased RRs 28% and the 14 mg dose decreased RRs 32%. Both doses brought about statistically significant reductions in the number of gadolinium-enhanced lesions. There were significant reductions at both the 7 mg dose and the 14 mg dose. It was well tolerated in phase II studies with few adverse events. Of note, there was some evidence of liver function test abnormalities. Approximately 10% of patients treated with teriflunomide in the phase II trial had elevations of liver function tests sometimes with concomitant elevations in bilirubin suggesting hepatocellular injury.
The results of a recently reported placebo-controlled phase III trial (TEMSO) confirmed and extended the results of the phase II trial. This trial recruited 1088 patients with RRMS and at least one relapse in the year prior to study entry [O’Connor et al. 2011]. Patients were randomized to placebo or treatment with teriflunomide at 7 or 14 mg. It was a 2-year trial and the primary outcome measure was the reduction in RR at 2 years. Key secondary outcome measures included the time to sustained disability progression, T2 burden of disease, number and volume of gadolinium-enhanced lesions, T1 hypointense lesions, and brain volume.
The study met its primary outcome measure. Annualized RR was reduced 31.2% (p = 0.0002) in the 7 mg group and 31.5% (p = 0.0005) in the 14 mg group. Time to sustained progression of disability was reduced by 30% in the 14 mg group (hazard ratio [HR] 29.8%, p = 0.0279) but did not reach statistical significance in the 7 mg group (HR = 23.7%, p = 0.0835). There were significant delays in time to first relapse for both dose groups and change in T2 lesion volume was reduced 39.4% compared with placebo (p = 0.0317) in the 7 mg group and 67.4% (p = 0.0003) in the 14 mg group.
Teriflunomide was generally well tolerated but was associated with troublesome side effects in a minority of patients. A total of 74% of patients randomized to active treatment completed the trial on study drug. The most common side effects were diarrhea, elevation of alanine aminotransferase (ALT), nausea, and alopecia. Elevation of liver function tests occurred more commonly in the treated group but were not associated with elevations of bilirubin suggestive of hepatocellular injury. However, the parent compound leflunomide carries a block box warning for the risk of fatal hepatotoxicity. Mild-to-moderate reductions in neutrophil counts were noted in the first 3 months of teriflunomide treatment but treatment interruption was not required and counts stabilized. This did not appear to be dose related. Four malignancies were reported during the trial. Three occurred in the placebo group and one case of cervical carcinoma in situ was observed in the treated group.
Among the most significant potential adverse events of concern with teriflunomide is the potential for fetal malformation. The parent compound has been associated with fetal malformation and carries a black box warning for fetal malformation. In the TEMSO trial there were 11 pregnancies. Four ended in spontaneous abortion (one in the placebo group and three in the 14 mg group), six ended in elective abortion, and one baby was born full term with no apparent malformations. This patient underwent cholesteramine washout to remove circulating teriflunomide. In the absence of cholesteramine or activated charcoal washout it may take many months for levels to decrease to 0.02 mg/l which is a level associated with a minimal teratogenetic risk.
In summary, teriflunomide as a monotherapy appears effective but unimpressive. However, phase II data in combination with interferon or glatiramer acetate suggest much more promising results. When added to IFN or glatiramer acetate MRI activity is markedly reduced. As a result, teriflunomide might prove highly effective as add-on therapy. This use has significant potential. If the combination of IFN or glatiramer acetate and teriflunomide proves as effective as NTZB the benefits might outweigh the risks associated with this agent.
Dimethylfumarate (BG-12) is a fumaric acid ester that is metabolized to monomethyl fumarate which is the active compound. Fumaric acid has been used in the European Union as a treatment for severe systemic psoriasis. Owing to severe gastrointestinal side effects treatment with monomethylfumarate is typically limited to several weeks. The oral formulation of enteric coated dimethylfumarate has a much better tolerability profile. The administration of fumaric acid esters is typically associated with a decrease in the white cell count which is primarily due to a decrease in circulating lymphocytes. This is associated with a shift from a Th1 to a Th2 cytokine profile. BG-12 also increased secretion of IL-4, IL-10, and IL-5 which help foster a Th2 response. In EAE there is a corresponding decrease in macrophage infiltration resulting in preservation of myelin and axonal integrity [Schilling et al. 2006; de Jong et al. 1996]. Fumaric acid esters also bring about apoptosis of activated T cells in the periphery and downregulate intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule (VCAM) resulting in a decrease in the migration of activated T cells across the blood–brain barrier.
BG-12 may have effects which are potentially neuro protective but this has yet to be demonstrated in humans. It activates nuclear factor E2-related factor 2 (Nrf2) transcriptional pathway. Activation of this pathway upregulates NAD(P)H: quinone reductase and increases the cellular content of glutathione [Scannevin et al. 2012]. This is an important antioxidant that may reduce cellular damage in an inflammatory milieu. The Nrf2 pathway may have other neuroprotective roles such as the inhibition of neuronal damage caused by excitotoxicity and regulation of myelin maintenance.
The initial study that evaluated the effects of dimethylfumarate in MS was an open-label prospective study of 18 weeks duration and this revealed a significant reduction in gadolinium-enhanced lesions [Schimrigk et al. 2006]. A second, larger, phase II study using BG-12 instead of monomethylfumarate evaluated the effects of 120 mg daily, 120 mg three times daily, and 240 mg three times daily compared with placebo [Kappos et al. 2008]. The study was 24 weeks in duration and the primary outcome measure was the cumulative number of enhanced lesions at week 12, 16, 20, and 24. While the two lower doses appeared to decrease the frequency of enhanced lesions the difference did not reach statistical significance. In contrast, the higher dose brought about a 69% decrease in the frequency of enhanced lesions (p < 0.0001) and 32% reduction in RR that did not reach statistical significance. Gastrointestinal irritation and flushing were the two most common adverse events but appeared to decrease with time becoming minimal after 4–8 weeks of treatment. Other common side effects included headache, pruritus, diarrhea, abdominal pain, and nausea. There was no increase in the risk of infection.
While the study was positive, the reduction in inflammatory measures of disease activity was modest (enhanced lesions and RR reduction). The perception of poor tolerability in conjunction with modest decreases in disease activity led to little excitement about this agent. Nevertheless, two phase III trials were designed and carried out on the basis of these results.
The first phase III trial was BG-12 was completed recently. This was a prospective, randomized, double-blind trail comparing BG-12 with placebo. The two doses were tested were 240 mg twice daily and 240 mg three times daily. The trial was 2 years in duration and patients had to have one relapse in the year prior to study entry to be included. The primary outcome measure was the proportion of patients who relapsed on trial. The formal results have only been reported in abstract form. Surprisingly, the data are far more robust than that reported in the phase II studies.
Patients treated with BG-12 in the twice-daily dose group had a 53% reduction in RR, a 90% reduction in gadolinium-enhanced lesion number, and an 85% reduction in new T2 lesions. The patients treated with 240 mg three times daily had 75% decrease in GD-enhancing lesions and a 73% decrease in new or enlarging T2 lesions. These results are better than expected on the basis of the phase II data and have brought about a reappraisal of the efficacy of this agent. While it is difficult and hazardous to compare results across clinical trials the results of the DEFINE study suggest that BG-12 could be more effective than anticipated previously. The findings reported in the DEFINE trial have been reproduced in the CONFIRM trial but have not yet been formally reported. BG-12 brought about a decrease in RRs of approximately 50% and decreased new T2 lesions by approximately 70 to 75%. This is not as robust as reported in the initial press release. BG-12 did not achieve superiority over glatiramer in disability progression. Nevertheless, the results of both trials suggest BG-12 could be an effective first-line agent in the treatment of relapsing forms of MS.
The treatment of RRMS is likely to undergo considerable change in the next few years. One factor already having an effect is the use of NTZB as a first-line agent in patients who test negative for JC virus. This has already changed the therapeutic landscape in that many more patients are starting therapy and staying on NTZB. In the past this agent had been used as a rescue therapy and was given only in cases in which patients were failing standard therapies. Another factor now having an effect on the treatment strategy and the market is the introduction of fingolimod. In the US, it is approved as a first-line agent. The lack of long-term safety data has led some to be very cautious in its use but the absence of any unexpected safety signals since its introduction has led to a wider acceptance. Many patients are reluctant to continue taking injectable medications when oral agents are available and are now switching to fingolimod. Newly diagnosed patients are also reluctant to take injectable medication when oral agents are available. The result thus far has been that a gradual shift to fingolimod and to NTZB in JC-virus-negative patients.
In the next year, BG-12 and teriflunomide are likely to enter the market. The BG-12 phase III trial data are superior to the efficacy shown in phase II trials. This has generated considerable excitement regarding BG-12. Unfortunately, this agent does not appear to be as effective as it initially appeared based on the data reported in the press release. This agent has a good safety profile without evidence of immunosuppression. There have been some tolerability issues in the form of nausea, vomiting, flushing, and pruritus that are generally short lived and resolved within about 8 weeks of starting therapy. Aside from this there is little to impede its widespread acceptance as an effective first-line therapy.
As a monotherapy, teriflunomide is unlikely to be widely adopted as a first-line agent. It is clearly effective but its efficacy is marginal in light of the superior data available for NTZB, fingolimod, and BG-12. Importantly, there have been no head-to-head comparative trails. Cross-trial comparisons are always difficult and often wrong. That being said, at the present time we have no better way to compare these agents. The factor that could substantially affect the use of teriflunomide is combination therapy. As pointed out above, the combination of IFN with teriflunomide could prove highly effective with a very favorable safety profile. The phase III combination trials are ongoing and this could have a profound effect on the treatment strategy for RRMS. The future of the MS therapeutics is changing rapidly because of the availability of oral agents with good efficacy and safety profiles and because of tests that allow for the stratification of risks with highly effective therapies such as NTZB. As more options become available the long-term outcomes for patients with MS will be vastly improved.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement: The author declares no conflict of interest in preparing this article.