Evidence for the efficacy of interferon beta-1b in delaying the onset of clinically definite multiple sclerosis in individuals with clinically isolated syndrome

Abstract

The BEtaferon®/BEtaseron® in Newly Emerging MS For Initial Treatment (BENEFIT) trial assessed the efficacy of early versus delayed treatment with interferon beta-1b for patients with clinically isolated syndrome (CIS). Patients were randomly assigned to receive either interferon beta-1b 250 μg every other day (early treatment, n = 292) or placebo (delayed treatment, n = 176) for 2 years or until progression to clinically definite multiple sclerosis. Clinical and magnetic resonance imaging (MRI) outcomes were assessed after 2 years (at the end of the placebo-controlled phase) and then again at 3, 5, and 8 years post randomization. MRI assessments were made after 2, 3, and 5 years. The results showed a consistent advantage of early treatment across most clinical and MRI variables, although median Expanded Disability Status Scale scores remained consistently low, with no differences between groups. These findings suggest that early treatment with interferon beta-1b improves long-term outcomes for patients presenting with CIS.

Keywords

clinically isolated syndrome early treatment interferon beta-1b multiple sclerosis

Introduction

Multiple sclerosis (MS) is a chronic demyelinating disease of the central nervous system [Compston and Coles, 2008]. The majority of patients with MS (up to 85% by some estimates) present with a single episode of demyelination known as a clinically isolated syndrome (CIS) [Miller et al. 2012], which can include a wide variety of symptoms. In one retrospective analysis, the most common presentation was dysfunction of long tracts (52%), followed by combinations of symptoms (21%), isolated optic neuritis (18%), and isolated brain stem dysfunction (9%) [Confavreux et al. 2000]. Other studies have found optic neuritis to be the presenting symptom for approximately 50% of patients with CIS [Brex et al. 2002; Fisniku et al. 2008; O’Riordan et al. 1998]. CIS may also have a subacute presentation [Miller et al. 2012], after which patients may progress to MS without showing evidence of relapses.

The placebo-controlled, double-blind BEtaferon®/BEtaseron® in Newly Emerging MS For Initial Treatment (BENEFIT) trial was designed to assess the effects of early versus delayed treatment with interferon beta-1b (BEtaferon®/BEtaseron®; Bayer HealthCare Pharmaceuticals; Whippany, NJ, USA) for patients presenting with their first episodes suggestive of MS (Figure 1) [Kappos et al. 2006]. Patients with CIS were randomly assigned to receive either interferon beta-1b 250 μg every other day (early treatment, n = 292) or placebo (delayed treatment, n = 176) [Kappos et al. 2006]. They remained in these treatment groups for 2 years or until conversion to clinically definite MS (CDMS), after which they could take interferon beta-1b, a different disease-modifying therapy (DMT), or no medication for MS at all [Kappos et al. 2006]. Using this design, the minimum time on placebo was 1 month with a median of 23 months [Edan et al. 2013].

Figure 1.

Design of the BENEFIT trial.

All patients who completed the placebo phase were eligible to enter a preplanned long-term follow-up study lasting up to 5 years as well as the BENEFIT Extension study, which reassessed patients after 8 years [Edan et al. 2013; Kappos et al. 2006]. Analyses were conducted at the end of the placebo-controlled phase [Kappos et al. 2006] and then 3 [Kappos et al. 2007], 5 [Kappos et al. 2009], and 8 [Edan et al. 2013] years after randomization. The core clinical outcomes in BENEFIT were time to CDMS and time to McDonald MS (per 2001 criteria), annualized relapse rate (ARR), and assessment of disability. Additional analyses were conducted to assess magnetic resonance imaging (MRI) outcomes and other clinical variables.

Core clinical outcomes

Assessments of the core clinical outcomes showed that early treatment was consistently associated with lower risk of disease progression and lower relapse rates. Disability, however, did not show much difference between the two treatment arms. Importantly, the presence of neutralizing antibodies (NAbs) had little effect on these outcomes.

Rate of conversion to CDMS was lower in the early treatment group at 2 years, a difference that persisted through year 8 of the study (Figure 2) [Edan et al. 2013; Kappos et al. 2006, 2007, 2009]. There was a 32.2% reduction in the risk for CDMS in the early treatment group after 8 years [Edan et al. 2013]. Similar differences were observed when conversion to McDonald MS was analyzed, with reductions of 46%, 46%, and 45%, respectively, at 2, 3, and 5 years [Kappos et al. 2006, 2007, 2009].

Figure 2.

Hazard ratios for conversion to CDMS.

ARR was also lower in the early treatment group at 3 [Kappos et al. 2007], 5 [Kappos et al. 2009], and 8 years [Edan et al. 2013] (Figure 3). After 8 years, there was a 22.9% reduction in risk for relapse in the early treatment group relative to the delayed treatment group [Edan et al. 2013]. Although the advantage of early treatment persisted throughout the study, it seems to have been largely driven by differences between the early and delayed arms that occurred in the first year of the study [Edan et al. 2013]. However, ARR was still significantly lower in the early treatment group when aggregate relapses in years 3–8 (i.e. after the completion of the placebo phase) were analyzed separately [Edan et al. 2013]. Importantly, some patients may have met the criteria used for disease progression without experiencing a relapse [i.e. through Expanded Disability Status Scale (EDSS) progression], although this is a rare occurrence in clinical practice, and therefore these patients might still have progressed to CDMS in BENEFIT without a corresponding increase in ARR.

Figure 3.

ARR across the 8 years of the BENEFIT study.

In contrast to these outcomes in which notable differences between early and delayed treatment were observed, relatively little difference in EDSS score was seen between the two arms at each analysis interval. Median EDSS changed little across the course of the study from the score of 1.5 that was reported for both groups at baseline [Kappos et al. 2006], with median values of 1.5 in both arms after 8 years [Edan et al. 2013]. However, despite the long-term stability in EDSS over the course of the study, significantly fewer patients in the early treatment arm had EDSS progression at 3 years (the first analysis point at which EDSS was reported) than in the delayed treatment arm [Kappos et al. 2007]. These differences at year 3 of BENEFIT likely reflected the higher relapse rates in the delayed treatment arm during the first 2 years of the study, given the lag between these events and the accumulation of disability in the early stages of MS [Compston and Coles, 2002]. However, by year 3, nearly all patients in the trial were undergoing active treatment with interferon beta-1b (and were therefore less likely to progress) and, as could be expected, the difference in EDSS progression between the two treatment arms did not persist through the later analyses, but the progression rate remained low and stable in both arms through the 5- and 8-year analyses [Edan et al. 2013; Kappos et al. 2009]. This lack of difference between the two groups may have resulted from the partial treatment received by patients in the delayed treatment arm or the natural tendency for the rate of progression in each arm to coalesce over time.

No significant effects of NAb status were seen on clinical outcomes after 2 or 3 years [Kappos et al. 2006, 2007]. In fact, there was actually a trend toward lower risk of progression to CDMS in patients with at least one NAb positive titer at 2 years (hazard ratio 0.63, 95% confidence interval 0.35–1.11, p = 0.11) [Kappos et al. 2006]. At the 5-year interval, NAb positivity was associated with increased brain MRI activity, but no effects on clinical outcomes were seen [Hartung et al. 2011]. NAb status was not assessed at the 8-year interval, but NAb titer at the end of the 5-year analysis did not significantly impact the outcomes after 8 years [Edan et al. 2013].

MRI outcomes

As with the core clinical outcomes, imaging outcomes in BENEFIT also supported early treatment with interferon beta-1b. Initial findings showed less lesion activity in the early treatment arm than in the delayed treatment arm, a finding that persisted up to the 5-year analysis. Patients in the early treatment arm also showed fewer persistent T1 hypointense lesions [persistent black holes (PBHs)].

As specified by the inclusion criteria, all patients in BENEFIT had at least two clinically silent T2 lesions with a size of at least 3 mm³ at study entry [Kappos et al. 2006]. At the end of the placebo-controlled phase, interferon beta-1b-treated patients had a lower number of newly active lesions, new T2 and gadolinium-enhanced (Gd+) lesions and a lower volume of Gd+ lesions [Kappos et al. 2006]. The decreased number of newly active lesions (either new or enlarging Gd+ or T2 lesions) in the early treatment group persisted through the 3- and 5-year analyses [Kappos et al. 2007, 2009]. Across the first 5 years of the study, T2 lesion volume decreased in both groups, with no significant differences between arms [Kappos et al. 2009]. MRI was not assessed at the observational 8-year interval, and therefore the persistence of the differences between early and delayed treatment beyond 5 years could not be determined [Edan et al. 2013]. Overall, results suggested a decrease in MRI activity in patients who had early treatment with interferon beta-1b relative to those who had delayed treatment.

A subanalysis was conducted to examine the effects of early versus delayed treatment on the development of PBHs [Nagtegaal et al. 2014]. These lesions correlate more strongly with clinical disability than Gd+ or T2 lesions [Sahraian et al. 2010; Truyen et al. 1996; van Walderveen et al. 2001] and are also thought to reflect global brain atrophy [Barkhof et al. 1998; Van Waesberghe et al. 1999; van Walderveen et al. 1995, 1998]. In BENEFIT, T1 lesions with hypointensity that persisted through the last available scan were defined as PBHs [Nagtegaal et al. 2014].

At year 5, PBHs were observed in 150 patients [83 in the early treatment arm (31.1% of available patients), 67 in the delayed treatment arm (39.9% of available patients)] [Nagtegaal et al. 2014]. Both the mean number of PBHs and the total PBH volume per patient were lower in the early treatment arm than in the delayed treatment arm [Nagtegaal et al. 2014]. Overall, approximately 10% of the lesions of any type that were identified in the study evolved into PBHs, and the probability of development into a PBH was influenced by the presentation type, location, and size of the lesion [Nagtegaal et al. 2014]. Treatment did not affect the probability of a lesion developing into a PBH; however, patients in the early treatment group had fewer PBHs than those in the delayed treatment group because early treatment had a protective effect on the development of new lesions [Nagtegaal et al. 2014].

Additional analyses from the BENEFIT study

In addition to the clinical and MRI outcomes, several other analyses have been conducted on data from the BENEFIT trial to gain as much insight as possible into disease course and the effects of interferon beta-1b in MS. These subanalyses attempted to address issues of clinical importance, including effects on cognition, identification of biomarkers for treatment response, influence of changes in the diagnostic criteria, and identification of baseline and early treatment characteristics that predicted later disease course and response to treatment. While the search for treatment biomarkers continues, these analyses have identified improvements in cognition with early treatment, and indicated that the benefits of early treatment are consistent regardless of the criteria used to diagnose MS (Poser versus McDonald). Results from BENEFIT have also helped further define the role of vitamin D in MS.

Cognition was assessed in BENEFIT using the Paced Auditory Serial Addition Task (PASAT) component of the Multiple Sclerosis Functional Composite (MSFC) [Penner et al. 2012]. Baseline PASAT scores were relatively high in both the early and delayed treatment arms with no differences between the two groups (median 55.0 across both arms) [Penner et al. 2012]. By the end of the placebo phase, the mean increase in PASAT score was greater in the interferon beta-1b-treated group than in the placebo group (+2.3 versus +0.8, p = 0.018) [Penner et al. 2012]. After 5 years, mean change in PASAT score remained higher in the early treatment group (+3.4 versus +1.5, p = 0.005) [Penner et al. 2012]. The greater improvement in performance on PASAT in the early treatment group than in the delayed treatment group persisted up to the 8-year analysis time point [Edan et al. 2013].

Analyses for biomarkers of treatment response have yielded mixed results in BENEFIT. The immunoglobulin M antiglycan antibody gMS-Classifier I has been observed in patients with CIS who were more likely to have a rapid relapse [Freedman et al. 2012]. However, in BENEFIT this antibody did not significantly predict time to CDMS or McDonald MS (per 2005 criteria) after 5 years [Freedman et al. 2012]. However, gMS-Classifier I did significantly predict increased risk for confirmed EDSS progression [Freedman et al. 2012]. Several other attempts to identify prognostic biomarkers have been made, but results to date have not yielded much conclusive evidence.

Studies examining the impact of 25-hydroxyvitamin D (25[OH]D) on outcomes in the BENEFIT trial have yielded more consistent results. 25(OH)D levels in the first 12 months after CIS were found to be predictive of disease activity and rate of progression up to 5 years after the start of the trial [Ascherio et al. 2014]. Patients with serum 25(OH)D levels at least 50 nmol/liter had a fourfold lower change in T2 lesion volume, a twofold lower rate of brain atrophy, and lower EDSS score than patients with 25(OH)D less than 50 nmol/liter [Ascherio et al. 2014]. These data suggest that 25(OH)D insufficiency had an impact on the disease course and that the patients with CIS might benefit from 25(OH)D supplementation, but more research is needed to identify the optimal levels of 25(OH)D for this patient population [Ascherio et al. 2014]. The possibility also exists that disease activity can be affected in an additive manner by 25(OH)D and interferon beta-1b.

Additional analyses have also been conducted to assess the effects that changes in the diagnostic criteria for MS would have had on the outcomes of BENEFIT. Revisions to the McDonald diagnostic criteria that were intended to improve sensitivity for detecting early MS without compromising specificity became available in 2005 [Polman et al. 2005], well after the BENEFIT trial was designed. These guidelines were further refined in 2010 [Polman et al. 2011]. When retrospectively applied, 34.7% of the BENEFIT population met the criteria for McDonald 2010 MS at the start of the trial [Montalbán et al. 2012]. Application of these diagnostic criteria to the BENEFIT population also permitted earlier diagnosis of MS over the first 5 years of the study and, importantly, the benefits of early treatment were still evident when assessing time to McDonald MS in that there was a 41% reduction in the risk of MS over 5 years in patients initially randomized to interferon beta-1b relative to those in the delayed treatment arm [Montalbán et al. 2012]. The sensitivity for diagnosis of MS was further increased by adding cerebrospinal fluid (CSF) data to the analysis [Montalbán et al. 2013].

Several subanalyses of the BENEFIT trial have attempted to identify patient characteristics that predicted disease course and response to treatment. Identification of these prognostic factors can help to optimize long-term disease management for patients with MS, particularly for patients in the early stages of the disease. The effect of key baseline demographic, clinical, and MRI parameters on time to CDMS was assessed at the conclusion of the placebo phase [Polman et al. 2008]. In this analysis of patient subgroups, risk for CDMS was higher in younger patients, patients who were treated with steroids at CIS (most likely precipitated by more severe disease symptoms), and patients with oligoclonal bands in CSF [Polman et al. 2008]. Importantly, treatment with interferon beta-1b significantly predicted longer time to CDMS across all subgroups except for patients without oligoclonal bands in CSF [Polman et al. 2008]. Treatment effects were also stronger for patients with monofocal disease who had at least nine T2 or Gd+ lesions [Polman et al. 2008]. To further the understanding of how baseline characteristics might help to predict future disease course, an additional analysis looked at the 12 patients (7.2%) in the placebo arm of BENEFIT who had no evidence of clinical or MRI disease activity after 2 years [Edan et al. 2007]. These patients tended to have older age, no oligoclonal bands in CSF, and up to nine T2 lesions at screening [Edan et al. 2007].

A separate analysis examined the prognostic value of baseline MRI characteristics for conversion to CDMS after 3 years [Moraal et al. 2009]. The presence of at least nine T2 lesions and the presence of at least three periventricular lesions were the MRI characteristics with the strongest prognostic value for negative disease outcomes [Moraal et al. 2009]. Treatment with interferon beta-1b did not affect the predictive capacity of MRI imaging criteria [Moraal et al. 2009].

A regression analysis of clinical, MRI, and laboratory parameters at baseline and after 1 year of treatment that predicted disease course at 5 years was conducted in only the patients in the early treatment arm (N = 268) [Freedman et al. 2011]. The results showed that age, number of Gd+ lesions, and number of T2 lesions at baseline predicted both ARR and annualized MRI activity after year 1 [Freedman et al. 2011]. During the first year of treatment, relapses and new MRI lesions during year 1 predicted ARR. In addition, new MRI lesions during year 1 and NAb titer at year 1 predicted annualized MRI activity [Freedman et al. 2011]. No significant predictors of confirmed or sustained EDSS progression were identified, perhaps due to the low rate of EDSS progression in the trial [Freedman et al. 2011].

Baseline characteristics were also analyzed for predictive significance after 8 years in a regression analysis that included both the early and delayed treatment arms [Freedman et al. 2013]. Assignment to the early treatment group significantly predicted both a lower rate of conversion to CDMS and lower ARR [Freedman et al. 2013]. Patients were more likely to have evidence of clinical disease (i.e. confirmed one-point EDSS progression, occurrence of relapses, or conversion to CDMS) if they had at least one Gd+ lesion, at least nine hyperintense lesions, at least nine hyperintense lesions not enhancing on T1, or a median T2 hyperintense lesion volume of at least 1.883 cm³, or if they met the criteria for McDonald 2010 MS or dissemination in time per McDonald 2010 at baseline [Freedman et al. 2013].

While these additional analyses may inform clinical practice, it is important to note that some were not part of the preplanned analyses of the BENEFIT trial. As such, they may lack sufficient statistical power to draw definitive conclusions and some patient selection bias may exist. That said, these post hoc findings are still useful for demonstrating the potential benefits of early treatment and can guide future research in larger trials prospectively designed to examine the effects.

Safety and tolerability

Throughout the study the safety profile of interferon beta-1b has been consistent with the known profile of this agent. In the placebo-controlled phase of BENEFIT, the most common adverse events (AEs) in the interferon beta-1b arm were injection site reactions and flu-like symptoms [Kappos et al. 2006]. Elevated liver enzymes were also noted in the interferon beta-1b group, but increases were transient and mostly occurred only in the first 3 months of treatment [Kappos et al. 2006]. The frequency of AEs was largely the same in the delayed treatment group once patients were given the option of active treatment [Kappos et al. 2007]. There was, however, some evidence of a higher rate of leucopenia and elevated alanine aminotransferase in the early treatment group relative to the delayed treatment group after 3 years [Kappos et al. 2007]. The AE profile remained largely consistent with these findings for the remainder of the study [Edan et al. 2013; Kappos et al. 2009]. Overall, incidence of AEs tended to decline with time in the study [Reder et al. 2013]. Consistent with this AE profile, quality of life (QoL) was initially better in the early treatment arm, but this difference disappeared as patients in the delayed treatment arm accumulated more time on therapy [Edan et al. 2013; Kappos et al. 2007, 2009].

Summary and conclusions

The BENEFIT trial is one of only a handful of trials that have examined the importance of early treatment for patients who have a first demyelinating episode suggestive of MS. The findings from this trial continue to be relevant to the MS community as long-term follow up provides more evidence to support early intervention for patients with CIS. Early results showed improvements in clinical (relapses, conversion to MS) and MRI outcomes (number of new lesions, lesion volume) up to 5 years after randomization. Later analyses in the BENEFIT Extension study showed that most of these outcomes remained consistently better in the patients who received early treatment with interferon beta-1b relative to those initially assigned to placebo, although disability outcomes remained generally similar across groups. Advantages of early treatment were also seen in several other parameters, including cognition and QoL. Overall, these findings suggest that early treatment with interferon beta-1b had a positive effect on long-term outcomes for patients with CIS, with an acceptable balance between side effects and the benefits of treatment.

Other DMTs have been investigated in the treatment of patients with CIS with some similar results to those that been found in the BENEFIT trial. In the early glatiramer acetate treatment in delaying conversion to CDMS in subjects Presenting with a Clinically Isolated Syndrome (PreCISe) study, patients who had early treatment with glatiramer acetate (Copaxone; TEVA Neuroscience, Inc.; Overland Park, KS, USA) had a reduced risk of developing CDMS and improved MRI outcomes relative to patients who had delayed treatment at 3 and 5 years post randomization [Comi et al. 2009, 2013]. ARR was also significantly lower in the early treatment arm at the last follow up point [Comi et al. 2013]. Similarly, treatment with either of two doses of subcutaneous interferon beta-1a (Rebif; Merk Serono, SA, USA; Geneva, Switzerland) at CIS reduced the probability of conversion to CDMS and improved MRI outcomes in the Rebif Flexible dosing in early MS (REFLEX) study after 2 years [Comi et al. 2012].

With longer follow up, the open-label Controlled High-Risk Subjects Avonex Multiple Sclerosis Prevention Study in Ongoing Neurologic Surveillance (CHAMPIONS) study, an extension of the CHAMPS trial, also found reduced risk of conversion to CDMS in patients who had early treatment with intramuscular interferon beta-1a (Avonex; Biogen Idec, Inc.; Cambridge, MA, USA) relative to those who had delayed treatment 5 years after randomization, but no improvements in MRI outcomes were observed [Kinkel et al. 2006]; however, few of the original CHAMPS patients remained in this study. Later follow up at 10 years after randomization still showed no differences in MRI outcomes, but probability of conversion to CDMS and ARR were both lower in the early than delayed treatment group [Kinkel et al. 2012]. When examining the results from these studies, it is important to consider that PreCISe and CHAMPS enrolled primarily patients with monofocal disease presentation whereas patients in BENEFIT and REFLEX could have had monofocal or multifocal disease onset at the start of the trial. In addition, these studies did not report findings on cognition and tended to have lower ascertainment with long-term follow up relative to BENEFIT.

Long-term follow up from extension trials like these have the advantage over observational studies in that both the patient population and the treatment plan are relatively homogenous. While BENEFIT did allow patients to take any DMT recommended by their physicians after the placebo-controlled phase was completed, each patient’s initial treatment was limited to interferon beta-1b or placebo, which makes medication effects easier to identify. Observational studies, however, may benefit from larger numbers of patients but they lack prospectively defined treatment strategies, thereby making identification of the effects of individual DMTs on patient outcomes more difficult than in extension studies

It is important to note that the inclusion criteria of BENEFIT selected for patients with relatively active disease and the study protocol conferred only a relatively small ‘head start’ for the early treatment arm. While all patients were given the option of treatment with interferon beta-1b by the end of the second year of the study, the mean time for patients to switch off of placebo was just a little over 1 year. Despite this small difference in the time to start interferon beta-1b, the early treatment arm still showed significant benefits over the long term in this population of patients who would be expected to progress relatively quickly. This finding emphasizes the impact that events in the first year after the first event suggestive of MS can have for patients, potentially leading to profound changes in the patient’s clinical course. By effectively treating patients early, physicians can lessen the overall morbidity of MS for years to come.

The BENEFIT study will continue to generate long-term data on the effects of early versus delayed treatment with interferon beta-1b in patients with CIS through the 11-year follow up, which recently concluded enrollment [ClinicalTrials.gov identifier: NCT01795872]. This study re-enrolled as many patients from the original cohort as possible to provide a thorough analysis of long-term outcomes for patients with CIS who received early or delayed treatment. Outcomes to be assessed include relapses, conversion to CDMS/McDonald MS, EDSS, Multiple Sclerosis Functional Composite, cognition, fatigue, employment, resource use, MRI measures, and optical coherence tomography.

Footnotes

Acknowledgements

I am grateful to the patients and the investigators in the BENEFIT trial for their continuing contributions to the study. Additionally, I would like to thank Gustavo Suarez (Bayer Health-Care Pharmaceuticals) and Dirk Pleimes (Myelo Therapeutics GmbH, formerly of Bayer HealthCare Pharmaceuticals) for assistance with preparation of the manuscript. Robert C. Ristuccia, PhD (Precept Medical Communications) provided medical writing assistance (funded by Bayer HealthCare Pharmaceuticals).

Funding

This work was supported by Bayer HealthCare Pharmaceuticals.

Conflict of interest

M.S. Freedman has received compensation from Actelion, Bayer HealthCare, Biogen Idec, Celgene, EMD Canada, Genzyme, Glycominds, Hoffmann-La Roche, Merck Serono, Novartis, Opexa, Sanofi Aventis, and Teva Canada Innovation for consulting services, and has received research/educational grants from Bayer HealthCare and Genzyme. He also participates in a Genzyme-sponsored speaker’s bureau.

References

Ascherio

Munger

White

Kochert

Simon

Polman

. (2014) Vitamin D as an early predictor of multiple sclerosis activity and progression. JAMA Neurol 71: 306–314.

Barkhof

McGowan

Van Waesberghe

Grossman

(1998) Hypointense multiple sclerosis lesions on T1-weighted spin echo magnetic resonance images: their contribution in understanding multiple sclerosis evolution. J Neurol Neurosurg Psychiatry 64(Suppl. 1): S77–S79.

Brex

Ciccarelli

O’Riordan

Sailer

Thompson

Miller

(2002) A longitudinal study of abnormalities on MRI and disability from multiple sclerosis. N Engl J Med 346: 158–164.

Comi

Freedman

Barkhof

Polman

Uitdehaag

. (2012) Comparison of two dosing frequencies of subcutaneous interferon β-1a in patients with a first clinical demyelinating event suggestive of multiple sclerosis (REFLEX): a phase 3 randomised controlled trial. Lancet Neurol 11: 33–41.

Comi

Martinelli

Rodegher

Moiola

Bajenaru

Carra

. (2009) Effect of glatiramer acetate on conversion to clinically definite multiple sclerosis in patients with clinically isolated syndrome (PreCISe study): a randomised, double-blind, placebo-controlled trial. Lancet 374: 1503–1511.

Comi

Martinelli

Rodegher

Moiola

Leocani

Bajenaru

. (2013) Effects of early treatment with glatiramer acetate in patients with clinically isolated syndrome. Mult Scler 19: 1074–1083.

Compston

Coles

(2002) Multiple sclerosis. Lancet 359: 1221–1231.

Compston

Coles

(2008) Multiple sclerosis. Lancet 372: 1502–1517.

Confavreux

Vukusic

Moreau

Adeleine

(2000) Relapses and progression of disability in multiple sclerosis. N Engl J Med 343: 1430–1438.

10.

Edan

Barkhof

Freedman

Hartung

Kappos

Miller

. (2007) The BENEFIT studies: characteristics of placebo-treated patients with a clinically isolated syndrome who do not show any further disease activity over 2 years. Presented at the Annual Meeting of the American Academy of Neurology. Boston, MA, 28 April–5 May.

11.

Edan

Kappos

Montalban

Polman

Freedman

Hartung

. (2013) Long-term impact of interferon β-1b in patients with CIS: 8-year follow-up of BENEFIT. J Neurol Neurosurg Psychiatry. 11 November 2013. [Epub ahead of print]

12.

Fisniku

Brex

Altmann

Miszkiel

Benton

Lanyon

. (2008) Disability and T2 MRI lesions: a 20-year follow-up of patients with relapse onset of multiple sclerosis. Brain 131: 808–817.

13.

Freedman

Hartung

Miller

Montalbán

Kappos

Polman

. (2011) Predictors of disease activity in CIS patients treated with IFNβ-1b: Lessons from the BENEFIT study. Presented at the Annual Meeting of the European Committee for Research and Treatment in Multiple Sclerosis. Amsterdam, The Netherlands, 19–22 October.

14.

Freedman

Kappos

Edan

Montalbán

Polman

Hartung

. (2013) Predictors of disease activity in patients with CIS treated with interferon β-1b in the BENEFIT trial. Presented at the Annual Meeting of the European Committee for Research and Treatment in Multiple Sclerosis. Copenhagen, Denmark, 2–5 October.

15.

Freedman

Metzig

Kappos

Polman

Edan

Hartung

. (2012) Predictive nature of IgM anti-alpha-glucose serum biomarker for relapse activity and EDSS progression in CIS patients: a BENEFIT study analysis. Mult Scler 18: 966–973.

16.

Hartung

Freedman

Polman

Edan

Kappos

Miller

. (2011) Interferon β-1b-neutralizing antibodies 5 years after clinically isolated syndrome. Neurology 77: 835–843.

17.

Kappos

Freedman

Polman

Edan

Hartung

Miller

. (2007) Effect of early versus delayed interferon β-1b treatment on disability after a first clinical event suggestive of multiple sclerosis: a 3-year follow-up analysis of the BENEFIT study. Lancet 370: 389–397.

18.

Kappos

Freedman

Polman

Edan

Hartung

Miller

. (2009) Long-term effect of early treatment with interferon β-1b after a first clinical event suggestive of multiple sclerosis: 5-year active treatment extension of the phase 3 BENEFIT trial. Lancet Neurol 8: 987–997.

19.

Kappos

Polman

Freedman

Edan

Hartung

Miller

. (2006) Treatment with interferon β-1b delays conversion to clinically definite and McDonald MS in patients with clinically isolated syndromes. Neurology 67: 1242–1249.

20.

Kinkel

Dontchev

Kollman

Skaramagas

O’Connor

Simon

(2012) Association between immediate initiation of intramuscular interferon β-1a at the time of a clinically isolated syndrome and long-term outcomes: a 10-year follow-up of the Controlled High-Risk Avonex Multiple Sclerosis Prevention Study in Ongoing Neurological Surveillance. Arch Neurol 69: 183–190.

21.

Kinkel

Kollman

O’Connor

Murray

Simon

Arnold

. (2006) IM interferon β-1a delays definite multiple sclerosis 5 years after a first demyelinating event. Neurology 66: 678–684.

22.

Miller

Chard

Ciccarelli

(2012) Clinically isolated syndromes. Lancet Neurol 11: 157–169.

23.

Montalbán

Kappos

Freedman

Hartung

Edan

Miller

. (2012) Application of the revised versions of the McDonald multiple sclerosis 2010 criteria: retrospective comparison using the BENEFIT clinical study dataset. Poster presented at the Annual Meeting of the European Committee for Treatment and Research in Multiple Sclerosis. Lyon, France, 10–13 October.

24.

Montalbán

Kappos

Freedman

Hartung

Edan

Miller

. (2013) Contribution of CSF to the diagnosis of McDonald MS: retrospective analysis in the BENEFIT study. Poster presented at the Annual Meeting of the American Academy of Neurology. San Diego, CA, 16–23 March.

25.

Moraal

Pohl

Uitdehaag

Polman

Edan

Freedman

. (2009) Magnetic resonance imaging predictors of conversion to multiple sclerosis in the BENEFIT study. Arch Neurol 66: 1345–1352.

26.

Nagtegaal

Pohl

Wattjes

Hulst

Freedman

Hartung

. (2014) Interferon β-1b reduces black holes in a randomised trial of clinically isolated syndrome. Mult Scler 20: 234–242.

27.

O’Riordan

Thompson

Kingsley

MacManus

Kendall

Rudge

. (1998) The prognostic value of brain MRI in clinically isolated syndromes of the CNS. A 10-year follow-up. Brain 121: 495–503.

28.

Penner

Stemper

Calabrese

Freedman

Polman

Edan

. (2012) Effects of interferon β-1b on cognitive performance in patients with a first event suggestive of multiple sclerosis. Mult Scler 18: 1466–1471.

29.

Polman

Kappos

Freedman

Edan

Hartung

Miller

. (2008) Subgroups of the BENEFIT study: risk of developing MS and treatment effect of interferon β-1b. J Neurol 255: 480–487.

30.

Polman

Reingold

Banwell

Clanet

Cohen

Filippi

. (2011) Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 69: 292–302.

31.

Polman

Reingold

Edan

Filippi

Hartung

Kappos

. (2005) Diagnostic criteria for multiple sclerosis: 2005 revisions to the ‘McDonald Criteria’. Ann Neurol 58: 840–846.

32.

Reder

Oger

Kappos

O’Connor

Rametta

(2013) Short-term and long-term safety and tolerability of interferon β-1b in multiple sclerosis. Mult Scler Related Dis 3: 294–302.

33.

Sahraian

Radue

Haller

Kappos

(2010) Black holes in multiple sclerosis: definition, evolution, and clinical correlations. Acta Neurol Scand 122: 1–8.

34.

Truyen

Van Waesberghe

van Walderveen

van Oosten

Polman

Hommes

. (1996) Accumulation of hypointense lesions (‘black holes’) on T1 spin-echo MRI correlates with disease progression in multiple sclerosis. Neurology 47: 1469–1476.

35.

Van Waesberghe

Kamphorst

De Groot

van Walderveen

Castelijns

Ravid

. (1999) Axonal loss in multiple sclerosis lesions: magnetic resonance imaging insights into substrates of disability. Ann Neurol 46: 747–754.

36.

van Walderveen

Barkhof

Hommes

Polman

Tobi

Frequin

. (1995) Correlating MRI and clinical disease activity in multiple sclerosis: relevance of hypointense lesions on short-TR/short-TE (T1-weighted) spin-echo images. Neurology 45: 1684–1690.

37.

van Walderveen

Kamphorst

Scheltens

Van Waesberghe

Ravid

Valk

. (1998) Histopathologic correlate of hypointense lesions on T1-weighted spin-echo MRI in multiple sclerosis. Neurology 50: 1282–1288.

38.

van Walderveen

Lycklama

Ader

Jongen

Polman

Castelijns

. (2001) Hypointense lesions on T1-weighted spin-echo magnetic resonance imaging: relation to clinical characteristics in subgroups of patients with multiple sclerosis. Arch Neurol 58: 76–81.