Principles of a New Treatment Algorithm in Multiple Sclerosis

Hans-Peter Hartung; Xavier Montalban; Per Soelberg Sorensen; Patrick Vermersch; Tomas Olsson


Expert Rev Neurother. 2011;11(3):351-362. 

In This Article

First Key Decision Point: Initiating Therapy

Early Treatment: Rationale, Evidence & Prediction of Long-term Response

Widespread and irreversible axonal pathology has been shown to occur at the earliest stages of the multiple sclerosis (MS) disease process,[1] including clinically isolated syndrome (CIS)[2] (e.g., optic neuritis, partial transverse myelitis or acute brainstem syndrome). Consequently, there is a strong argument for early therapy, and several randomized controlled clinical trials have been conducted in patients with a CIS to ascertain the benefits of early treatment with disease-modifying therapies (DMTs). These trials, which are discussed in further detail later, have consistently demonstrated the benefits of early intervention with IFNβ formulations and glatiramer acetate in delaying conversion to clinically definite MS (CDMS) (Table 1).[3–7] In addition, immediate treatment following a CIS has been demonstrated to provide consistent long-term reductions in the risk of CDMS and disability progression, relative to delayed treatment.[8,9] As the definition of a CIS is not consistent across these trials, for the purpose of this publication a CIS is considered to be a single episode with neurological symptoms suggestive of a demyelinating disease, in a patient having lesions of a shape and location typical for MS, and for whom a thorough differential diagnosis has been completed.

Overall, approximately 50% of patients who develop a CIS will experience a second event within 2 years (CDMS). To ensure that the most vulnerable patients are protected while avoiding overtreatment, it is important to identify which patients with a CIS are at the highest risk of progressing to CDMS. Of the prognostic factors that have been studied, the presence of demyelinating lesions on MRI scans (in recent CIS trials, typically, at least two) at the time of CIS presentation appears to be the most robust predictor of subsequent conversion to CDMS.[10–14] Therefore, it would appear prudent to consider initiating early treatment in a patient with a CIS who also presents with an MRI typical of MS (e.g., showing more than eight T2 gadolinium-enhancing [Gd+] lesions), and even more important if there are oligoclonal bands in their cerebrospinal fluid (i.e., four or more T2 lesions plus the presence of oligoclonal bands).

Selecting a Treatment for First-line Use: Considerations for Constructing a Treatment Algorithm Prior to the Era of Oral Therapies

To date, clinicians and their patients have selected first-line treatment from among the three IFNβ formulations, glatiramer acetate or, rarely, in patients with high disease activity, natalizumab. In addition, mitoxantrone has been used (without indication) if natalizumab is unavailable, although it is avoided when possible owing to its association with serious adverse events (SAEs), as described later. Ideally, the initial treatment choice would be the option offering the highest efficacy and the lowest burden (including factors such as convenience, tolerability and safety). However, in practice, clinicians and patients have accepted some degree of trade-off between efficacy and the burden of therapy. As discussed below, therapies offering higher efficacy may also present a greater degree of burden, so the initial treatment decision largely comes down to informed patient preference. However, some patients, such as those with higher-activity disease, may willingly select more effective drugs, even when those drugs carry an increased burden. Of course, clinicians play a role in the determination of initial therapy, and their propensity for risk avoidance (for example) may also be a factor in patients' treatment decisions.

As already described, IFNβ formulations and glatiramer acetate have demonstrated consistent efficacy benefits compared with placebo in patients with early disease (i.e., patients presenting with a CIS) in randomized controlled trials (Class 1 evidence).[3–8] There are no head-to-head trials of these agents in CIS, and comparing across trials with different designs and baseline characteristics is associated with inherent limitations. Nevertheless, it is reassuring that the observed benefit reported with these agents compared with placebo in CIS is generally comparable, with an adjusted 35–55% reduction in the risk of progression to CDMS over 2–3 years (Table 1). These studies have also demonstrated that these agents limit other measures of disease activity, including relapse rates and new MRI lesion activity. Consequently, there appears to be little to distinguish among IFNβ and glatiramer acetate therapies from an efficacy perspective in CIS.

The results from the Early Treatment Of MS (ETOMS) study, which used a lower dose of IFNβ than that approved for use in standard clinical practice, demonstrated that, in contrast to data obtained in relapsing–remitting MS (RRMS) patients, even a low dose of IFNβ seems to be effective when used in the early phase of the disease.[5] In addition, it has recently been reported, though not yet published, that IFNβ-1a significantly delayed conversion to CDMS in CIS patients in the 2-year Phase III Rebif Flexible Dosing in Early MS (REFLEX) study.[101] However, this IFN therapy has not, to date, been approved for use in CIS patients.

Currently, there are no data available with natalizumab or mitoxantrone in the treatment of patients with a CIS. However, there is emerging evidence that use of these types of immunosuppressant treatments as a short-term 'induction therapy' followed by maintenance therapy with an immunomodulatory treatment may be an effective strategy in patients who present with aggressive disease.[15,16]

For patients who have relapsing CDMS, the IFNβ formulations, glatiramer acetate, natalizumab and mitoxantrone have all shown statistically significant benefits versus placebo in terms of decreasing annualized relapse rates (ARRs) and reducing the risk of disability progression in Phase III, multicenter, randomized studies (Class 1 evidence; Table 2).[17–22] However, it should be pointed out that compared with the other trials shown in Table 2, the mitoxantrone pivotal trial (Mitoxantrone In Multiple Sclerosis [MIMS]) was a more limited trial conducted in patients with more advanced MS,[19] although extensive clinical experience has subsequently supported its results.[23]

In addition to the pivotal trials abovementioned, six randomized head-to-head clinical trials have been conducted with the IFNβ formulations and glatiramer acetate (Class 1 evidence; Table 3).[24–29] Overall, no differences in efficacy were seen between subcutaneous (SC) IFNβ and glatiramer acetate, or between the SC formulations of IFNβ in these trials. Nevertheless, results from two of the studies (Independent Comparison of Interferon [INCOMIN] and Evidence for Interferon Dose Response: European–North American Comparative Efficacy [EVIDENCE]) suggested that the more frequently administered SC formulations of IFNβ may have benefits with regard to relapse frequency and MRI activity over intramuscular (IM) IFNβ-1a given once weekly.[27,28] However, INCOMIN and EVIDENCE were criticized for their short treatment duration (in EVIDENCE, advantages were not sustained for a full year) and/or methodological limitations (in INCOMIN, there were some differences in baseline characteristics).[30] There are no completed randomized head-to-head trials to date comparing IM IFNβ-1a with glatiramer acetate, nor any such trials for natalizumab and mitoxantrone, although some trials are in progress (e.g., natalizumab vs mitoxantrone[102]). Despite the absence of data from comparative trials for natalizumab and mitoxantrone, the numerical differences in the findings seen in Natalizumab Safety and Efficacy in Relapsing–Remitting Multiple Sclerosis [AFFIRM] and MIMS suggest that these agents may be more effective than the IFNβ formulations and glatiramer acetate in terms of reducing ARR (Table 2),[20] a conclusion that is also supported by data from clinical practice (Level IV evidence).

As mentioned previously, clinicians and patients also consider the potential burden when choosing a therapy. Natalizumab is associated with a risk of progressive multifocal leukoencephalopathy (PML), a rare (1:1000) opportunistic viral infection of the brain that can lead to severe disability or death (Table 4)[31–34] and, in addition, may also be associated with hepatotoxicity (unconfirmed), hypersensitivity reactions and an increased risk of certain infections.[103] The risk of PML increases after 24 months of treatment and in patients who have received prior immunosuppressant therapy.[35] Therefore, this agent is only recommended as an initial treatment for MS under restricted conditions, and patients must provide new informed consent if they wish to continue receiving treatment with natalizumab beyond 2 years. As discontinuation of natalizumab therapy has been associated with a rapid resumption of disease activity,[36,37] strategies to minimize these effects are currently being evaluated.

The current indication in Europe for natalizumab is primarily for patients with high disease activity despite treatment with IFNβ formulations or glatiramer acetate. However, natalizumab can also be used in patients with rapidly evolving, severe RRMS. The US FDA indication currently recommends natalizumab for patients who have had an inadequate response to, or are unable to tolerate, an alternate MS therapy.

As mentioned above, mitoxantrone is usually avoided for first-line use since it is associated with SAEs (Table 4), including profound myelosuppression, increased risk of infections, potentially fatal cardiotoxicity and, most notably, treatment-induced acute leukemia.[23,38,39,104] Because of their safety profiles, both natalizumab and mitoxantrone require close patient monitoring.

Consequently, the IFNβ formulations and glatiramer acetate have been the foundation of initial treatment for MS patients. Given that these agents appear to have similar efficacy, treatment decisions have been based mainly on differences in the agents' tolerability profiles (Table 4). In addition, the European Federation of Neurological Societies has indicated that the risk of developing IFNβ-neutralizing antibodies should be integrated into treatment decisions,[40] and this proposal has been adapted into a consensus article issued by an expert panel with participants from both sides of the Atlantic Ocean.[41]

Convenience is also an important consideration for patients initiating first-line therapy – particularly since patients are expected to remain on therapy for prolonged time periods. Adherence is negatively affected by higher injection frequency, as well as mode of application (see Table 4 for dosing routes and schedules of the IFNβ formulations and glatiramer acetate), as has been demonstrated in other therapy areas.[42,43]

Selecting a Treatment for First-line Use in the Era of Oral Therapies: What has Changed & Why?

Physicians and patients have desired orally administered treatment in MS to lower the burden of therapy associated with the parenteral mode of application (especially IM and SC injections) (Table 4). In September 2010, the first oral agent for MS, fingolimod, a sphingosine-1-phosphate receptor modulator,[44,45] gained FDA approval. In January 2011, the European Medicines Agency's Committee for Medicinal Products for Human Use (CHMP) recommended approval of fingolimod for RRMS in patients with high disease activity despite treatment with IFNβ and those with rapidly evolving RRMS. As with the existing DMTs, fingolimod has shown efficacy versus placebo (in terms of reducing ARR and delaying disability progression) in a pivotal, randomized clinical trial of treatment-naive patients with CDMS (FTY720 Research Evaluating Effects of Daily Oral Therapy in Multiple Sclerosis [FREEDOMS]; Class 1 evidence; Table 2).[46] Data are not currently available for fingolimod in CIS.

In addition to the pivotal study mentioned above, a head-to-head Phase III study has been conducted comparing daily oral fingolimod and IM IFNβ-1a in patients with MS (Trial Assessing Injectable Interferon versus FTY720 Oral in Relapsing–Remitting Multiple Sclerosis [TRANSFORMS]; Class 1 evidence; Table 3).[47] There was a significantly greater reduction in ARR with fingolimod than with IFNβ, although the risk of disability progression was similar over 12 months. Conclusions from TRANSFORMS may be limited by the fact that almost 50% of patients in the study had previously been treated with an IFNβ formulation. Given that the study entry criteria specified baseline disease activity in the previous year, this might indicate that a notable proportion of the study population was already responding suboptimally to IFNβ.

Serious adverse events and events leading to discontinuations were more frequent with oral fingolimod than with IM IFNβ-1a in TRANSFORMS.[47] The most frequent SAEs with oral fingolimod were bradycardia and atrioventricular block. Oral fingolimod may also increase the risk of some infections, with two deaths resulting from viral infections in patients receiving fingolimod 1.25 mg. One of these deaths was caused by disseminated primary varicella zoster infection, while the second death was caused by herpes simplex encephalitis. Bronchitis and pneumonia were also more frequent with fingolimod.[46,47] Fingolimod-related leukopenia and lymphopenia,[46,47] with a 73–76% reduction in peripheral-blood lymphocyte counts, were reported in TRANSFORMS and FREEDOMS, and these reductions may have led to the increased risk of infection with fingolimod treatment. In addition, fingolimod may be associated with an increased risk of macular edema, hepatic effects and fetal risk, as noted in the fingolimod prescribing information.[105] Patients randomized to receive the lower fingolimod dose (0.5 mg) in FREEDOMS did not show an increased number of SAEs compared with placebo.[46] It is conceivable that lower doses of fingolimod may have an improved risk–benefit profile, and part of the FDA approval involved a commitment to study this treatment at doses lower than the approved 0.5 mg dose.

Because of these safety concerns, the FDA has implemented a risk evaluation and mitigation strategy (REMS) for fingolimod, which requires the manufacturer to communicate comprehensive safety information to healthcare providers through direct mailings for 5 years post-launch. This safety information states that healthcare providers must closely monitor patients to avoid bradyarrhythmia and atrioventricular block (which are known to occur on first dosing), infections, macular edema, respiratory effects, hepatic effects and fetal risk.[106] This will involve an additional time commitment, not only for patients, but also for healthcare providers.

Cladribine could potentially be the second oral agent granted FDA approval for the treatment of MS. However, the CHMP recently issued a negative opinion of the drug, stating that based on current data, the benefits do not outweigh the risks, with the number needed to treat (NNT) to show benefit being too high. Applications for approval of oral cladribine in the USA and Europe were based on the findings from a single Phase III, randomized, placebo-controlled clinical trial in previously untreated patients with CDMS (Cladribine Tablets Treating MS Orally [CLARITY]; Class 1 evidence; Table 2).[48] There are currently no head-to-head data available for cladribine versus other MS therapies; a trial with cladribine in CIS is currently ongoing (Oral Cladribine in Early MS [ORACLE]). In the Phase III CLARITY trial, oral cladribine significantly reduced ARR and decreased the risk of sustained progression of disability versus placebo. It should be noted that patients in the placebo group of this study seemed to have uncharacteristically low disease activity; the significant results seen with cladribine may indicate that this treatment could be a promising therapeutic option in patients with higher disease activity. Lymphocytopenia was the most commonly reported treatment-emergent event, occurring more frequently with cladribine (27%) than with placebo (2%). Herpes zoster was also more commonly seen with cladribine (20 patients, 2%) than with placebo (no patients), and a correlation was found between patients with the lowest lymphocyte counts and risk of infections. One concern for treatments that carry immunosuppressive properties is an increased risk of secondary malignancies.[49] Malignant (cervical, ovarian and pancreatic carcinoma and melanoma: three patients in total) and benign or unspecified neoplasms were more frequent with cladribine (ten patients, 1.1%) than with placebo (no patients). These included five benign uterine leiomyomas that required inpatient hospital visits for treatment. Overall, there was no apparent pattern in any specific neoplasms.

The Phase III Teriflunomide in Reducing the Frequency of Relapses and Accumulation of Disability in Patients With Multiple Sclerosis (TEMSO) study of teriflunomide, an inhibitor of dehydro-orotate dehydrogenase and DNA synthesis,[50] in relapsing MS was recently completed, and initial results were reported at the 2010 European Committee for Treatment and Research in Multiple Sclerosis meeting in Sweden.[51,52] Compared with placebo, teriflunomide had significant positive effects on MRI outcomes, ARR and disability progression. There were no reported differences in SAEs, including serious infections or infestations, between the two treatments.[51,52] However, the full, peer-reviewed publication of the study will offer a better understanding of the safety and efficacy profile of teriflunomide. A comparative Phase III study of teriflunomide and IFNβ-1a is currently ongoing.[107]

A key question is where these oral therapies fit into the existing treatment algorithm. Cladribine and fingolimod have the convenience of oral administration and evidence of substantial efficacy.[53] For the lowest dose of fingolimod, severe side effects were rare and overall safety concerns may be fewer. The postmarketing REMS plan that has been put in place for fingolimod should help to better determine the safety–tolerability profile of this agent. As seen with many other drugs across a wide range of therapy areas, serious side effects not seen in Phase III trials can emerge once a drug starts being routinely prescribed to patients, which underlines the need for an adequate REMS. The advantage of current first-line therapies (IFNβ and glatiramer acetate) is the extensive knowledge of their long-term safety. This is offset by the potential inconvenience of their invasive mode of application. The use of fingolimod or cladribine appears to be associated with a rare risk of severe side effects, so convenience/safety trade-offs need to be addressed when discussing treatment options. This is especially true in patients with early MS[53] or in women of childbearing age, as fetal effects cannot be excluded, and the evolution of the disease in cases of drug discontinuation has yet to be evaluated extensively. Some patients may conclude that the inconvenience (and cost) of the monitoring requirements when using oral treatments with immunosuppressive properties outweighs the convenience of oral administration.

Until longer-term safety data are available with these oral therapies and clinicians feel more comfortable with their use, it is likely that the current status quo will remain, and IFNβ formulations and glatiramer acetate will continue to provide the mainstay of first-line therapy. As a consequence of the oral agents' safety profile and the surveillance required, first-line use of the first new oral agents will likely be restricted to those patients with more active disease who are nonresponsive to treatment with IFNβ or glatiramer acetate, and those who reject invasive therapies and are willing to accept certain risks. High-quality postmarketing studies will be required to establish the risk–benefit profile of all new agents. If the efficacy and safety of teriflunomide are in line with initial reports of the Phase III data, it may have wider application in first-line therapy.

For the reasons mentioned above, natalizumab use in treatment-naive patients will continue to be restricted over the short-to-medium term for those patients with more active MS. In the near term, the choice between oral fingolimod and natalizumab in these patients will come down to physician and patient preference, taking into account the benefit–risk ratios of both agents. This may change over the longer term, and further attempts to tailor treatments to individuals need to be made. An ELISA that can detect JC virus (JCV) antibodies in serum and plasma is currently under investigation.[54] Infection with JCV is a prerequisite for developing PML; therefore, the ability to identify JCV-seronegative patients using a sensitive and specific assay would enable clinicians to identify patients who can be treated with natalizumab with a very low risk of developing PML. Such approaches will help to reduce the risks for individual patients and allow better stratification of treatment decisions, especially in early MS. Finally, the use of mitoxantrone in treatment-naive patients will likely be further limited due to the side-effect profile of this chemotherapeutic agent. A summary of the treatment considerations for first-line therapy, based on the information that is currently available, is outlined in Table 5.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.