|Home | About | Journals | Submit | Contact Us | Français|
Multiple sclerosis is a progressive inflammatory disease of the central nervous system. With prevention or at least delay of disease progression as a key target in the management of multiple sclerosis, current opinion on treatment encourages early intervention with well-tolerated disease-modifying treatments in order to optimize long-term clinical outcomes. Patients presenting with a clinically isolated syndrome (CIS) may progress to clinically definite multiple sclerosis, and clinical trials have demonstrated that early treatment with interferon beta can reduce the conversion rate. Cognitive impairment may already be present in patients with CISs. Today there is evolving evidence that cognitive impairment may be relevant for prognosis and that early treatment with interferon beta may also have a protective effect on the cognitive function. As an accumulation of neuronal loss is now considered to underlie the development of persistent disability in multiple sclerosis, it is crucial that treatment can protect against neuronal damage. In addition to its anti-inflammatory activity, interferon beta may have direct and indirect neuroprotective effects, and several studies have explored the role of interferon beta in regulating neuroprotective factors. With over 15 years of clinical experience as evidence, the long-term safety and efficacy of interferon beta treatment is unquestionable. Results from the CIS studies have demonstrated the high percentage of patients converting to clinically definite multiple sclerosis without treatment and the short- and long-term benefits of an early use of disease-modifying treatments. These findings support starting disease-modifying treatment as soon as the diagnosis of MS is reasonably formulated.
Multiple sclerosis (MS) is characterized as a progressive inflammatory disease of the central nervous system in which axonal destruction occurs at early stages and is responsible for the accumulation of irreversible disability [Kuhlmann et al. 2002; De Stefano et al. 2001]. Current opinion on treatment encourages early intervention with well-tolerated disease-modifying treatments (DMTs) in order to optimize long-term clinical outcomes. MS affects predominantly young adults with approximately 500,000 patients in Europe and more than 2 million people worldwide [Flachenecker and Stuke, 2008]. The long disease duration has a high impact on both the individual patient and healthcare systems. Although the aetiology of MS remains unknown, the view has evolved that MS is a ‘whole-brain disease' driven by an autoreactive immune response against central nervous system (CNS) antigens, particularly myelin peptide antigens, that lead to axonal and neuronal loss [Hafler et al. 2005].
Four clinical types of MS have been described [Lublin and Reingold, 1996]. The majority of patients (80%) develop relapsing–remitting MS (RRMS), and may enter a progressive phase during the further course of disease [Noseworthy et al. 2000]. RRMS patients suffer from episodes of neurologic symptoms (relapse) followed by remission with either full recovery or residual deficits. Most RRMS patients initially present with a demyelinating event or a clinically isolated syndrome (CIS) such as optic neuritis or transverse myelitis [Miller et al. 2005]. Secondary progressive MS (SPMS) begins as RRMS and proceeds to progression of disability without discrete relapses. Approximately 90% of untreated RRMS patients will transition to SPMS after 20–25 years [Trojano et al. 2003]. Progressive–relapsing MS (PRMS) arises in patients with ongoing progression from the onset, but with one or more subsequent relapses superimposed on continuing progression [Lublin and Reingold, 1996]. This subtype is diagnosed rarely. Primary progressive MS (PPMS) represents around 10% of cases and is characterized by continuous disease progression from onset without relapse or remission [Andersson et al. 1999].
Key targets in the management of MS are reducing the rate of relapses and preventing or at least delaying disease progression [Comi, 2009], especially silent disease activity, which predates the appearance of initial symptoms and may also determine progressive disease courses in the long run. Agents such as interferon beta (IFN-β) and glatiramer acetate are recommended as first-line treatments for RRMS. They may be initiated once diagnosis is confirmed with reference to the McDonald criteria [Polman et al. 2005; McDonald et al. 2001]. However, clinical trials have demonstrated that early IFN-β treatment is also effective in preventing the conversion of the first isolated demyelinating episodes into clinically definite MS (CDMS) [Clerico et al. 2008], and recent data support early interferon treatment to significantly reduce the risk of progression [Trojano et al. 2009]. Possibly the most important aspect of MS therapy is to initiate treatment as early as possible to minimize further inflammation with axonal damage, and resulting irreversible disability, ongoing in patients who otherwise appear clinically stable. The US National MS Society (NMSS) recommends early therapy should be given without interruption, except in cases of lack of benefit, intolerable adverse effects, or availability of more effective therapy. In addition, other national MS societies have recommended treatment after the first relapse, particularly in patients at high risk for disease progression.
CIS is described as a status of patients with an acute or subacute first demyelinating event caused by either inflammation or demyelination confined to the CNS. It can be unifocal or multifocal, and the acute attack, which lasts for at least 24 hours, is classified according to its localization in the CNS as cerebral, optic nerve, brainstem or spinal cord [Miller et al. 2005; Brex et al. 2002]. CIS may or may not progress to CDMS, but when an episode occurs in conjunction with lesions on the initial MRI, CIS is highly predictive of developing further inflammation and future definite MS within a decade [Noseworthy et al. 2000]. In patients who first present with CIS suggestive of MS, the increases in the volume of the lesions, seen on MRI in the first 5 years, correlate with the degree of long-term disability [Brex et al. 2002]. The significantly increased risk of developing CDMS with a positive MRI and the need to consider treatment with DMTs has been supported by clinical observations. In a longitudinal study assessing a cohort of 71 CIS patients with serial MRI scans, 88% of subjects with abnormal MRI but only 19% with normal MRI at presentation had developed CDMS at reassessment after a mean of 14.1 years [Brex et al. 2002].
Diagnostic criteria for MS have evolved over time to include MRI findings as an integral part of diagnosis to disseminate lesions in time and space [McDonald et al. 2001; Poser et al. 1983]. In 2001 the International Panel on the Diagnosis of Multiple Sclerosis updated the criteria to include specific guidelines for using MRI, visual evoked potentials (VEP) and cerebrospinal fluid (CSF) analysis [McDonald et al. 2001]. The high specificity, positive predictive value and accuracy of the new criteria more than doubled the rate of diagnosis of MS within a year of presentation with CIS [Dalton et al. 2002], highlighting the prognostic significance of the pathological changes occurring on MRI during these early years. The integration of MRI parameters into the diagnostic criteria thus enables a more rapid and accurate diagnosis, permitting earlier initiation of treatment for patients with CIS [Polman et al. 2005].
The impact of applying MRI-supported diagnostic criteria for MS has been intensely monitored and discussed over recent years. The increasing practical experience and study of different diagnostic approaches as potential ways to simplify MRI diagnostic prerequisites led to a recently published new algorithm [Montalban et al. 2010]. It is hoped that the new criteria will be easier to implement both in the routine clinical evaluation and research studies of patients with typical CIS while having diagnostic accuracy for MS similar to the original version of the McDonald criteria. Therefore, further testing in new prospectively followed up CIS cohorts is recommended.
MRI was described as a relevant tool to assist with early diagnosis and prognosis without a need of correlation with clinical symptoms after a retrospective study showed rapid early conversion to CDMS in a subset of high-risk patients with incidentally detected asymptomatic lesions [Lebrun et al. 2008]. In contrast, about one third of patients with an otherwise typical CIS suggestive of demyelination have entirely normal MRI apart from symptomatic lesion(s) [Tintore et al. 2006; Barkhof et al. 1997; Morrissey et al. 1993]. However, in a 10–14-year follow up, one fifth of such patients emerged with CDMS [Optic Neuritis Study Group, 2004; Ormerod et al. 1987]. MRI-negative CIS patients can have a positive result in CSF analysis; inconsistently, others do not [Soderstrom et al. 1998]. In a recent prospective study, 572 patients with CIS underwent brain MRI and analysis of CSF within 3 months of first attack [Tintore et al. 2008]. Follow up for a mean of 50 months revealed that the presence of oligoclonal bands doubles the risk for having a second relapse, independently of MRI, but does not seem to have an influence on the development of disability. For patients with this strictly isolated syndrome in time and space, CIS may be the only possible diagnosis [Weinshenker et al. 1990].
The significance of the CIS has been emphasized in four large-scale clinical trials to determine whether early treatment following a CIS can delay the second clinical event and therefore a diagnosis of CDMS: CHAMPS1 [Jacobs et al. 2000], ETOMS2 [Comi et al. 2001a], BENEFIT3 [Kappos et al. 2006], and PreCISe4 [Comi et al. 2009] (Table 1). Importantly with IFN-β and with glatiramer acetate a similar range of efficacy could be reached supporting the clinical relevance of both treatment principles; the major difference between these studies was rather based on study design and preplanned follow-up extensions. Based on findings from these studies it is important to identify CIS patients with little or no disability, because the earlier these treatments are administered, the more effective they are, producing better long-term, as well as short-term, outcome.
Some of the disabilities of CIS are not just physical. Cognitive deficits can occur independently of physical disability, which complicates their identification and recognition [Rao et al. 1991a]. In addition, MS patients with cognitive impairment experience a greater disturbance in activities of daily living than cognitively intact patients [Rao et al. 1991b]. This underscores the need for timely and accurate assessment of cognitive deficits in MS patients. Preserving cognitive function with early use of DMTs is quite important to maintain patients’ abilities such as everyday life activities, employment status, or social participation [Chiaravalloti and DeLuca, 2008].
The underlying mechanism for the irreversible neurodegeneration in MS is assumed to be a dysregulation of the innate immune system [Lassmann et al. 2007]. Active inflammation determines the initial event, but disease progression occurs independently of acute inflammation. The inflammatory phase is associated with a potent cellular and humoral immune response against potential CNS antigen(s) leading to neuronal loss and brain atrophy [Frohman et al. 2006; Peterson and Trapp, 2005; Trapp et al. 1998]. These complex cascades include immune-mediated cytotoxicity, demyelination, reduced neurotrophic support, metabolic impairment and altered intracellular processes [Frohman et al. 2006; Trapp et al. 1998], and result in an interindividual and stage-dependent heterogeneity of lesions [Lassmann et al. 2007].
It is presumed that a coincidence of genetic and environmental factors is involved in the activation of self-reactive lymphocytes against CNS antigen(s), such as myelin basic protein and proteolipid protein [Munz et al. 2009]. Disruption of the blood–brain barrier integrity affords the transendothelial migration of activated lymphocytes, as well as chemokines and cytokines, to the CNS [Minagar and Alexander, 2003]. This initiates a complex immunological cascade leading to epitope spreading, a phenomenon thought to contribute to the chronicity of MS [McMahon et al. 2005]. Disease progression may be caused by an abnormal activation of dendritic cells, resident CNS microglia, macrophages, natural killer cells and astrocytes. The production of pro-inflammatory cytokines, excitotoxic and cytotoxic molecules, e.g. tumour necrosis factor-α (TNF-α), interferon-γ (IFN-γ), glutamate, matrix metalloproteases, reactive oxygen species (ROS) and reactive nitrogenous species (RNS) is upregulated. This triggers new attacks and is implicated in acute axonal transection, neuronal loss and contraction of glial tissue [Frohman et al. 2006]. Axonal injury starts at disease onset and its progression correlates with the degree of inflammation. Demyelinated axons may further degenerate due to lack of trophic support following the loss of myelin [Trapp and Stys, 2009]. Long-term disability may be due to irreversible neurodegeneration that begins early not only within active focal lesions, but also in chronic silent plaques, and normal-appearing white and grey matter [Peterson et al. 2001; Trapp et al. 1998].
Effective MS treatment targets multiple disease pathways by blocking inflammatory processes, reducing oxidative stress and enhancing cell growth [Lopez-Diego and Weiner, 2008]. Therapies targeting the immune system, especially during the inflammatory phase, could potentially slow the progression of MS. As these therapies need to be given early, accurate early diagnosis of MS has become paramount.
Across different MS stages patients have measurable amounts of whole brain atrophy, both in the white and grey matter, and brain atrophy reflects the net result of irreversible and destructive pathological processes. It worsens over time, often without clinical manifestations as assessed by the Expanded Disability Status Scale (EDSS) score [Tao et al. 2009; Rudick et al. 1999]. Paraclinical assessments such as serial MRI are able to detect subclinical lesions and accurately monitor the pathologic changes in MS [Zivadinov et al. 2008b]. Using standard MRI acquisitions, the rate of atrophy can already be detected and accurately quantified in patients with CIS, who will only later develop CDMS [Anderson et al. 2007]. As measures of global brain atrophy lack adequate specificity for the determination of the degree of inflammation and underlying neurodegenerative changes, such as loss of myelin or axons and increase in glial content, only newer, nonconventional imaging techniques provide the potential to detect early neuroprotective effects of treatment [Giacomini and Arnold, 2008]. Improved specificity is achieved by diffusion tensor imaging (DTI). The evolution of new lesions into persistent black holes can be characterized with T1 and T2 relaxometry and magnetization transfer ratio (MTR), and their biochemical composition probed with magnetic resonance spectroscopic imaging [Barkhof et al. 2009; Hasan et al. 2005; van Waesberghe et al. 1999].
Despite the continued improvement of MRI techniques and the introduction of more sensitive imaging modalities, MRI findings in isolation are insufficient to confirm a diagnosis of MS. A combination of clinical and paraclinical measures remains indispensable for diagnosis; these include a careful history, clinical examination, panel of evoked responses and CSF analysis in addition to MRI.
Recent research demonstrating that cognitive impairment occurs early in the disease and can have a negative impact on patients’ lives has important implications for physicians. Findings from two observational studies, CogniCIS and CogniMS, showed early occurrence of depression and fatigue alongside cognitive deficits [Langdon et al. 2009a, 2009b]. Patients’ relatives were able to detect even minor changes in cognitive performance, in earlier stages of the disease than the MS patients themselves. Early cognitive decline in MS patients is insufficiently recognized, but could provide an important target for a better disease management.
An increasing understanding of the immunopathology of MS has led to the approval and widespread early use of DMT. As early initiation appears to have an impact on the effectiveness of treatment, disease-modifying agents should be started as soon as possible following a diagnosis of MS and may also be considered in patients with CIS who are at high risk based on initial MRI lesion load [National Clinical Advisory Board, 2008]. Numerous studies confirmed that the level of inflammation, as indicated by gadolinium enhancement activity on MRI, is highest in the early relapsing disease, and that axonal damage starts before symptoms are evident [Comi, 2009]. Ongoing development of brain lesions and atrophy is evident even in individuals who are clinically in remission. DMTs decrease the mean relapse rate by approximately 30% and positively affect MRI in RRMS [Ebers and PRISMS (Prevention of Relapses and Disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis) Study Group, 1998]. In addition, IFN-β may slow disease progression [Panitch et al. 2002; Li et al. 2001].
The anti-inflammatory activity of interferons is well understood and considered an important contributor to its overall effect [Kovarik et al. 2007]. Recent investigations with the relevant animal model of experimental autoimmune encephalomyelitis (EAE) showed that the effect of type I interferons on myeloid cells within the CNS was crucial to the protective effect of the drug [Prinz et al. 2008]. In this model the effect of IFN-β on myeloid cells in the CNS suppressed inflammation and the processing of antigenic peptides, which otherwise exacerbate the disease by activating infiltrating T cells and causing further inflammation. With glatiramer acetate a broader range of putative therapeutic mechanisms can be assumed, involving a Th2 shift of pathogenic T cells [Neuhaus et al. 2000], but also the secretion of brain-derived neurotrophic factor (BDNF) from invading T cells whose relevance has recently been shown in conditional mouse mutants suffering from EAE [Linker et al. 2010].
Additional effects of interferons to mitigate brain damage in the early stages of the disease to be discussed here include the enhancement of growth factor production in the brain and the reduction of oxidative stress.
Experimental studies have shown, first with glatiramer acetate, that glatiramer-reactive T cells can secrete BDNF and also penetrate the blood–brain barrier during EAE, thus serving as a possible transporter vehicle. BDNF is a potent neuroprotective agent that has been found in active and inactive MS lesions [Stadelmann et al. 2002]. Therefore, the term neuroprotective autoimmunity was coined: T cells and macrophages/microglia in MS lesions seem to release BDNF, and nerve cells in the immediate vicinity are responsive to BDNF, which helps to preserve axons. In RRMS patients under treatment with IFN-β, BDNF secretion from peripheral blood mononuclear cells (PBMCs) is upregulated significantly [Azoulay et al. 2009], indicating a potential direct neuroprotective mechanism of IFN-β. In addition, it has been shown that IFN-β promotes the production of nerve growth factor (NGF) by human brain microvascular endothelial cells [Biernacki et al. 2005]. An effect on NGF production was triggered by interaction with T lymphocytes pretreated with IFN-β in vitro or derived from MS patients treated with IFN-β in vivo. These data are consistent with an effect of IFN-β on brain atrophy and axonal injury detected on MRI measures when given early in the course of MS [Zivadinov et al. 2008a].
An inflammatory, but also a degenerative cascade that includes neuronal oxidative stress and excitotoxicity is initiated in MS by microglia activation [Gonsette, 2008]. Reducing free radicals and oxidative stress can protect against tissue damage and neural death, and preliminary data suggests IFN-β can upregulate genes that are able to inhibit oxidative stress [Croze et al. 2009a]: In RRMS patients 2 months of treatment with IFN-β-1b consistently induced increased metallothionein (MT) mRNA expression in lymphocytes. MTs have anti-inflammatory and anti-apoptotic properties [Hwang et al. 2008], protecting against oxidative stress and acting as neuroprotectants preventing demyelination, axonal damage, and also increasing oligodendrocyte precursors. The expression of the nuclear receptor co-activator NCOA7 is also rapidly induced in MS patients treated with IFN-β-1b [Croze et al. 2009b; Reder et al. 2008]. As demonstrated in previous studies this protein belongs to a conserved family of eukaryotic oxidative resistance proteins, which are highly expressed in human foetal brain cells and astrocytes [Durand et al. 2007; Shao et al. 2002]. It is assumed that proteins of this class protect brain cells from oxidative stress-related DNA damage. Expression of NCOA7 was significantly elevated in mononuclear cells isolated from RRMS patients 4 hours after a single dose of IFN-β1-b [Croze et al. 2009b].
In patients with MS, ‘black holes’, i.e. hypointense lesions on a T1-weighted MRI scan, are thought to represent irreversible axonal damage. Limiting the evolution of newly formed lesions to chronic black holes has thus been suggested as a surrogate marker for neuroprotective properties of therapeutic interventions [Miller, 2004] and has first been described in therapeutic trials with glatiramer acetate [Comi et al. 2001b]. In SPMS patients IFN-β-1b treatment significantly reduced the development of hypointense T1-weighted lesions, suggesting a reduction in axonal damage [Barkhof et al. 2001]. In the recent BECOME study5 that analysed the formation of chronic black holes in patients treated either with glatiramer acetate or IFN-β-1b, the majority of new brain lesions that occurred in patients with RRMS did not result in chronic black holes [Cadavid et al. 2009]. Compared with treatment with glatiramer acetate, significantly fewer lesions developed into chronic black holes with IFN-β-1b treatment, suggesting an anti-oxidative effect of IFN-β-1b.
Brain-volume increases on serial MRI provide a sensitive overall measure of neuroprotection in MS trials [Barkhof et al. 2009]. Data for this parameter are available from the recent BEYOND6 study that compared clinical and MRI outcomes in treatment-naïve RRMS patients randomized to two dosages of IFN-β-1b or glatiramer acetate [O'Connor et al. 2009]. As reflected by the mean changes in brain volume, there was no additional atrophy observed in years 2 and 3, and the overall median increases in brain volume from baseline to year 3 were similar in all treatment groups. Furthermore, early IFN-β-1b therapy may not only slow the accumulation of axonal damage but also reverse some of the early axonal dysfunction and delay the accumulation of permanent axonal loss [Narayanan et al. 2001], as suggested by an increase in the N–acetylaspartate/creatine ratio, a neuronal measure of axonal injury, following treatment of RRMS patients with IFN-β-1b.
The best available clinical evidence, based on a high retention rate of patients during the follow up, for clinically relevant neuroprotection of IFN-β-1b therapy is provided by the recently released 5-year data from the BENEFIT trial that confirmed and extended what had already been seen at the 3-year time point [Kappos et al. 2009, 2007]. In this study, investigators measured progression of patient disability using EDSS. In the 3-year analysis a delay of IFN-β-1b treatment for up to 2 years was associated with a significantly higher risk of developing confirmed EDSS progression [Kappos et al. 2007]. After 5 years, the absolute number of patients with confirmed EDSS progression remained lower in the early treatment group, although the difference was no longer statistically significant since at this time point also the placebo group had been switched to IFN-β-1b for at least 2 years [Kappos et al. 2009]. Analyses of cognitive function as measured by the Paced Auditory Serial Addition Test (PASAT), a standardized measure of working memory and attention, also suggest that treatment with IFN-β-1b at the first signs of MS results in better preservation of cognitive functioning compared with results in patients whose therapy was delayed [Kappos et al. 2009].
Cognitive deficits are a confirmed concomitant of MS, with any form or stage of the disease, and a high prevalence dependent on the research setting. Clinic-based cross-sectional studies on MS-related cognitive impairment provide estimates ranging between 54% and 65%, whereas large-scale, community-based surveys show prevalence estimates of 43% and 46%, respectively [Amato et al. 2006b; McIntosh-Michaelis et al. 1991; Rao et al. 1991a]. Cognitive dysfunction can vary widely between individuals regarding severity and symptoms including attention, information processing efficiency, executive functioning and complex attention [Benedict et al. 2006]. Processing speed, visual learning and episodic and working memory seem to be most consistently affected and deficits are evident in approximately 40–65% of patients [Rao et al. 1993], whereas dementia and language deficits are uncommon [Chiaravalloti and DeLuca, 2008].
Cognitive deficits in MS per se can have a negative impact on several aspects of the subject’s everyday life, including behaviour, socialization, relationships and family, competence in legal and financial matters, driving skills, adjustment to disability, adherence to treatment regimens and ability to benefit from rehabilitation. Thus, they result in a considerable disruption of patients’ social and private lives thereby detrimentally affecting their overall quality of life (QoL). Although physical ability is important for the everyday performance, it cannot account for the extent of difficulties that individuals with MS encounter predominantly in those activities that require a substantial cognitive load. Onset of MS accompanied with cognitive disabilities typically occurs between the ages of 20 and 40 and leads to the loss of gainful employment status for a large number of patients. Between 50% and 80% of patients are unemployed 10 years after the onset of the disease, with only 15% due to physical restrictions [Rao et al. 1991b]. In addition to walking ability and age, the cognitive parameters of verbal fluency and two memory measures were identified among the five variables that account for 49% of the change in employment status in MS [Beatty et al. 1995a]. Moreover, measures of cognitive performance such as processing speed, verbal memory and executive functioning could predict the vocational status of MS patients even after adjustment for age, education, sex, depression and disease course [Benedict et al. 2006]. Cognitive impairment may thus be counted among the most important factors predicting employment status in MS patients.
Cognitive and neurological deficits are not necessarily associated in MS [Rao et al. 1991a]. Owing to the individual variability in natural history and lesion location, the degree of cognitive impairment is only mildly associated with the physical disability status measured by EDSS, and the same applies to disease duration [Lynch et al. 2005; Beatty et al. 1990]. However, a number of concomitant disease factors have been identified which may influence the degree of cognitive dysfunction. Cognitive deficits have been found in patients with early RRMS [Amato et al. 1995; Beatty et al. 1995b; Rao et al. 1991a], probable MS [Achiron, 2004] and in patients with CIS [Potagas et al. 2008; Glanz et al. 2007] with the frequency of cognitive dysfunction declining progressively from SPMS to PPMS, RRMS and finally to CIS [Potagas et al. 2008]. The natural history of cognitive impairment in 50 patients, followed from the first diagnosis of MS and compared with 70 healthy control subjects, revealed that cognitive dysfunction can be detected in some patients even in the initial phase of MS [Amato et al. 1995]. The number of patients with cognitive defects tended to increase with disease progression, and a number of subjects who had remained unchanged after 4 years were found to show deteriorated cognitive function after 10 years [Amato et al. 2001].
Even though disease duration appears to be a poor predictor of the degree of cognitive impairment, the course of the disease has an influence on its pattern. Comparing MS subtypes, patients with PPMS or SPMS course have typically demonstrated a greater severity of cognitive impairment than patients diagnosed with a RRMS course, and SPMS generally results in more severe involvement than PPMS [Chiaravalloti and DeLuca, 2008]. The different subtypes of MS have also been associated with different profiles of cognitive impairment. PPMS and SPMS patients may be more likely than RRMS patients to suffer from attention, speed of processing, executive and abstraction deficits, while RRMS patients may be more likely than healthy controls to suffer from memory deficits. Owing to the potential influence of other factors such as age, gender and disability level, however, using the disease course as a predictor of cognitive impairment has to be questioned.
Cognitive dysfunction may also affect a considerable number of patients with so-called benign MS (BMS), a relapsing–remitting form of MS with patients experiencing only few relapses with slight or no residual damage or disability. The definition of BMS has been heavily weighted towards physical disability; however, BMS patients with normal physical abilities may be heavily disabled by nonmotor symptoms such as fatigue, pain, depression and cognitive dysfunction. In a clinical cohort of patients that fulfilled currently accepted criteria for benign MS (disease duration >15 years and EDSS <3.0), significant fatigue and depression were found in 49% and 54% of patients, respectively, and cognitive performance was significantly lower compared with healthy subjects [Amato et al. 2006a]. In 38% of cases, cognitively impaired patients had reduced their social and work activities measured on the Environmental Status Scale (ESS), despite substantial preservation of motor abilities.
In BMS patients cognitive impairment has been related to higher cortical and subcortical brain damage as assessed by quantitative MRI measures [Rovaris et al. 2008; Amato et al. 2004]. However, functional cortical changes may limit the clinical impact of tissue injury, as cortical reorganization occurs during high cognitive processes in patients with CIS [Audoin et al. 2003]: using functional magnetic resonance imaging (fMRI) while performing the conventional PASAT, CIS patients showed significantly greater activation in the right frontopolar cortex, bilateral lateral prefrontal cortices and right cerebellum than education-, age-, and sex-matched healthy controls. There was no difference in the PASAT score between patients and controls, indicating the existence of compensatory cortical activations, mainly located in regions involved in executive processing. Furthermore, the results suggest that fMRI could become a useful method to characterize early compensatory cortical reorganization in MS and could provide objective indices of brain function in the evaluation of the disease evolution.
Both cognitive assessment and paraclinical MRI-based markers are relevant in the clinical assessment of BMS patients, as they may predict short-term disease evolution and correctly identify ‘truly benign’ and ‘pseudo-benign’ (cognitively impaired) MS patients. In comparison with cognitively impaired patients, cognitively preserved ‘truly benign’ MS patients exhibited lower T2-weighted and T1-weighted white matter lesion load, as well as higher cortical volumes and magnetization transfer ratio values [Amato et al. 2008]. In a prospective study, 29% of patients with ‘pseudo-benign’ MS significantly worsened over a 5-year follow-up period, whereas ‘truly benign’ patients had a significantly higher probability of remaining stable over 5 years [Portaccio et al. 2009]. Worsening at follow up was closely related to specific aspects such as male gender and cognitive dysfunction at baseline as well as high white matter T1-weighted lesion load. Therefore, in the model of ‘benign MS’, cognitive impairment represents a sensitive marker of brain tissue integrity or pathology and may play a prognostic role in predicting subsequent disease evolution.
Recent findings in CIS subjects demonstrated that the presence of cognitive impairment may predict the conversion to CDMS [Zipoli et al. 2010]. Fifty six patients with either monofocal or multifocal CIS enrolled within no more than months since onset of symptoms were evaluated at baseline and after a mean follow up of 3.5 years. At baseline 14% of CIS patients met the conditions for cognitive impairment of failing at least three neuropsychological tests, and 57% of patients fulfilled McDonalds’s criteria for dissemination in space. While 46% of all patients converted to CDMS, the percentage in the cognitively impaired subgroup was 88%. Using a Cox regression model, both cognitive impairment and the presence of McDonalds’s dissemination in space were shown to be predictors for conversion from CIS to CDMS.
Early identification of MS-related cognition impairment may be beneficial, as earlier intervention may improve overall prognosis of the disease. Five-year data from the BENEFIT trial have shown that early treatment with IFN-β-1b may have a beneficial effect on cognition that becomes even more pronounced over time [Kappos et al. 2009]. Mean scores for Multiple Sclerosis Functional Composite (MSFC) subscales or the overall MSFC improved over the 5 years in most patients, with no significant differences between early and delayed IFN treatment. However, improvement of the overall MSFC score was mainly due to an improvement in the cognitive function subtest PASAT, and patients with early IFN treatment had better mean PASAT scores than patients with delayed treatment (p=0.0045). Additional analyses showed a highly significant association between categorized EDSS scores and PASAT scores, as well as a significant association between Kurtzke cerebral function score and PASAT scores at the beginning of the study. No significant or consistent correlates of baseline MRI variables such as T1 enhanced and T2 lesion number and volume, black hole lesion volume, brain volume on the PASAT-3” score at baseline and year 5 could be observed.
The observations in BENEFIT with repeated measurements of the PASAT seem to suggest that most control patients improved but reached a plateau. The fact that cognitive improvement over the long time observation was more pronounced in the early treatment group implies additional capacity for recovery or compensation that could be better preserved with early treatment.
Cognitive impairment, a major cause of disability in MS patients, can be detected at the earliest stages of the disease and may be relevant for prognostic predictions. Therefore, regular screening from the early phases of MS is important in clinical practice.
In 1993, IFN-β-1b was the first immunomodulatory therapy to be approved for the treatment of MS by the US Food and Drug Administration, and it is the only IFN-β to receive a license for treatment of SPMS. With over 15 years of clinical experience as evidence, the long-term safety and efficacy of IFN-β treatment is unquestionable. The concept of initiating the treatment of MS in its earliest phase dates from the early years of this century, when IFN-β was first demonstrated to delay conversion to CDMS in patients with a CIS. However, arguments against early treatment were also discussed.
While CIS is the first clinically detectable event characterizing the onset of RRMS, MS may not always be suspected at the time of the attack. Not all patients with CIS will develop RRMS: a diagnosis of MS should be made when there is clinical and/or paraclinical evidence of temporal and spatial dissemination of the lesions [Polman et al. 2005; McDonald et al. 2001] and only after accurate diagnostic procedures have eliminated alternative diagnoses. Is the concern that ‘not every CIS is MS’ still a hurdle to treating MS early and what have we learned from findings for placebo patients in the CIS studies CHAMPS, ETOMS, BENEFIT and PreCISe [Comi et al. 2009; Kappos et al. 2009, 2007; Comi et al. 2001a; Jacobs et al. 2000]?
In the CHAMPS study, 383 patients with a first acute clinical demyelinating event (optic neuritis, incomplete transverse myelitis or brainstem or cerebellar syndrome), and evidence of demyelination on MRI of the brain, were randomly assigned to receive weekly intramuscular injections of IFN-β-1a 30µg or placebo [Jacobs et al. 2000]. The primary endpoint was the development of CDMS, defined as the occurrence of either a new visual or neurologic event or progressive neurologic deterioration. The trial was stopped early after the preplanned interim analysis revealed that the cumulative probability of the development of CDMS during the 3-year follow up was significantly reduced in the IFN-β-treated patients (35%) compared with the placebo group (50%). Patients in the IFN-β group also had a significant relative reduction in the volume of brain lesions, significantly fewer new or enlarging lesions, and significantly fewer gadolinium-enhanced lesions at 18 months. With 82% of placebo-treated CIS patients showing MRI conversion at 18 months, CHAMPS demonstrated that the large majority of CIS patients evolved to MS according to the McDonald criteria [Polman et al. 2005] and that initiating treatment with IFN-β at the time of a first demyelinating event was beneficial for patients with brain lesions on MRI that indicated a high risk of CDMS.
Similar findings were achieved in the 2-year ETOMS trial, which randomized patients with initial findings suggestive of MS within the previous 3 months to once-weekly treatment with either 22µg IFN-β-1a subcutaneously or placebo [Comi et al. 2001a]. MRI conversion, defined by the appearance of at least one new MRI T2 lesion of diameter more than 10mm or three lesions of diameter less than 10mm diameter at 24 months, was 94% for placebo-treated patients. Compared with placebo, IFN-β showed significant positive effects on clinical outcomes, with a reduced rate of conversion to CDMS (34% versus 45%) and delayed time to the occurrence of a second exacerbation (569 days versus 252 days; hazard ratio [HR] 0.65). Importantly, in this study the therapeutic benefit on relapses was supported by MRI findings showing that both lesion activity and the accumulation of lesion burden were reduced compared with placebo.
The PRECISE trial randomized 481 unifocal CIS patients with a first clinical event and at least two T2-weighted brain lesions at least 6mm in size on MRI to glatiramer acetate 20mg/day given subcutaneously or placebo [Comi et al. 2009]. Conversion to CDMS according to the McDonald criteria at 3 years defined by the time to new T2 lesion of CDMS was seen in approximately 90% of placebo-treated patients. Conversion of CIS patients to CDMS was significantly reduced in the glatiramer acetate group compared with the placebo group (25% versus 43%), and the time to conversion was significantly prolonged from 336 days to 722 days (115%). Risk reduction of developing CDMS was 45% compared with placebo. MRI activity, including the number of enhancing lesions and the number of new T2 lesions, was significantly lower in the glatiramer acetate group. Actively treated patients showed a 61% decrease in new T2 lesions and a 60% decrease in new contrast lesions compared with placebo.
In the BENEFIT study 468 patients with a first clinical demyelinating event suggestive of MS and at least two clinically silent brain MRI lesions were randomized to receive subcutaneous IFN-β-1b 250µg or placebo every other day for up to 2 years or until CDMS was reached [Kappos et al. 2007, 2006]. After 2 years 45% of patients in the placebo group had converted to CDMS and 85% fulfilled the McDonald criteria. Treatment with IFN-β significantly decreased the conversion to CDMS from 45% in the placebo group to 28%. Moreover the risk of conversion to MS according to McDonald criteria was also significantly reduced, resulting in a risk reduction of 50% for CDMS and 46% for McDonald MS [Kappos et al. 2006]. The design of the study included an optional 3-year follow up with open-label IFN-β-1b for all patients up to a maximum of 5 years after randomization [Kappos et al. 2009]. The results of this active treatment extension confirmed that the risk for conversion to CDMS was significantly lower in the early treatment group (46%) than in the delayed treatment group (57%), and time to MS diagnosed with McDonald criteria was delayed by early treatment with IFN-β.
Taken together these studies demonstrate both the high proportion of CIS patients converting to CDMS without treatment and the benefit of early treatment with IFN-β. An epidemiological 25-year model-based cost–utility analysis further supports the economic benefit of early treatment with IFN-β. Based on the results of the BENEFIT study, 40,420 of 50,000 patients treated with IFN-β-1b from CIS diagnosis are expected to convert to CDMS by the end of a 25-year follow-up period compared with 43,700 in a 50,000 patient cohort of untreated individuals [Lazzaro et al. 2009]. The estimated cumulative probability of converting to CDMS during the first 3 years is significantly different: 72.90% or 84.94%. As all four placebo-controlled phase III trials in CIS patients described above have demonstrated a conversion of placebo treated CIS patients to CDMS in more than 80% of cases [Comi et al. 2009, 2001a; Kappos et al. 2006; Jacobs et al. 2000], early treatment with IFN-β-1b is considered highly cost effective and dominant from the societal viewpoint.
Another issue in the discussion on ‘to treat or to wait’ was that some patients may be destined to have benign MS, so waiting might be appropriate. As discussed before, the current definitions of benign MS are mainly weighted for the patients’ motor abilities and fail to capture relevant disease-related cognitive, psychological and social problems such as fatigue, depression and cognitive impairment resulting in a reduction of social and work activities [Amato et al. 2006a]. In addition, BENEFIT has demonstrated favourable effects on cognitive performance over a 5-year observation period [Kappos et al. 2009]. Both studies suggest that even in patients with a longstanding history of mild or ‘benign’ disease course, the therapeutic decision process should be based on a comprehensive assessment of different disease-related dimensions.
Ten years ago, as well as today, concerns regarding the size of clinical benefit and long-term efficacy in terms of disability prevention and QoL remain [Pittock, 2009]. The studies described in the previous section may answer these questions.
CHAMPS [Jacobs et al. 2000], ETOMS [Comi et al. 2001a] and BENEFIT [Kappos et al. 2006] have consistently demonstrated a reduction in the cumulative probability of developing CDMS in CIS patients receiving early treatment with IFN-β compared with placebo-treated patients. A Cochrane meta-analysis of these three trials with IFN-β demonstrated a significantly lower risk of developing CDMS both after 1 year (pooled odds ratio [OR] 0.53) and after 2 years of follow up (pooled OR 0.52) [Clerico et al. 2008]. A little later, results of the PreCISe study on treatment of CIS patients with glatiramer acetate or placebo for up to 36 months were published [Comi et al. 2009], showing a statistically significant reduction in the risk of developing CDMS of 45% (hazard ratio 0.55).
The long-term benefit of early treatment with IFN-β was demonstrated in extension studies and long-term follow-up data for these trials in CIS patients. In the CHAMPIONS7 study [Kinkel et al. 2006], an open-label extension study of the CHAMPS study, the cumulative probability of development of CDMS was significantly lower in the patients who received continuous treatment for 5 years compared with the delayed treatment group of placebo-treated patients in CHAMPS who only received IFN-β during the extension study (36% versus 49%). In the CHAMPIONS 10-year follow up, patients treated with IFN-β from the beginning maintained a significantly reduced risk of conversion to CDMS (58%) compared with a failure rate of 69% for patients with delayed treatment [Kinkel et al. 2009]. Similar findings were reported after 4 years of follow up in an extension of the ETOMS study with CDMS conversion rates of 46.1% versus 57.8% for IFN-β treatment compared with placebo [Comi, 2009; Comi et al. 2002]. The 5-year results of the BENEFIT integrated data set also confirmed that the course of MS can be changed over time: at the end of the 5-year observational period, the risk for confirmed EDSS progression was 25% in the early treatment group and 29% in the delayed treatment group displaying a risk reduction of 24% over 5 years (HR 0.76). Time to confirmed EDSS progression was shifted 549 days in favour of patients from the early treatment group.
Taken together these findings support greater efficacy of early versus delayed IFN treatment in patients with a first clinical event suggestive of MS. However, long-term follow-up data are also available for RRMS: in an Italian observational study, 1504 RRMS patients were followed for up to 7 years. Patients received IFN-β-1a intramuscularly or subcutaneously, IFN-β-1b subcutaneously or no IFN treatment [Trojano et al. 2007]. IFN-β slowed progression in RRMS with a highly significant reduction in the incidence of secondary progression (HR 0.38), EDSS score of 4 (HR 0.70) and EDSS score of 6 (HR 0.60) when compared with untreated patients. About 28% of untreated patients compared with 20.5% of the IFN-treated group had reached an EDSS score of 4. This 20.5% threshold was reached with a delay of 1.7 years (7 years for IFN-treated versus 5.3 years for untreated patients).
Another line of reasoning was based on safety aspects: while patients might benefit from early treatment, starting IFN-β after a first event might not be appropriate due to the side effects.
Results from the 5-year active treatment extension of the BENEFIT study in CIS patients [Kappos et al. 2009] demonstrated a safety and tolerability profile of IFN-β-1b consistent with the profile reported for the placebo-controlled phase. Injection-site reaction and flu-like syndrome were the most common adverse events in the IFN-β group, and occurred at an incidence of 21% in the early and 24% in the delayed treatment group receiving IFN-β-1b 250µg subcutaneously every other day. Most side effects were reported less often in the BENEFIT trial than in the pivotal trials in RRMS patients, which may be attributed to several management improvements such as stepwise dose escalation at therapy start, use of injection devices and prophylactic use of paracetamol or ibuprofen. Therefore, patient support such as training and injection supervision is essential, particularly over the first 6–12 months of treatment.
Treatment support programmes are designed to help patients become familiar with treatment management. Elements may include the use of autoinjectors, support by specialist nurses or the MSGateway website. The 1-year interim results of an observational study evaluating the influence of support elements on RRMS or SPMS patients’ QoL suggest that the use of the specialist nurse programme or the specific, neutral patient-oriented websites is an effective support element in the treatment of patients with MS. Data from this study showed a mean difference in Functional Assessment in MS total score (FAMS-TS) for patients choosing a specialist nurse compared with patients not choosing a specialist nurse [Pozzilli et al. 2009].
Comprehensive MS patient support should include a proven and reliable product performance, the latest application system and a range of customized therapy aids. It should understand the day-to-day needs of MS patients and provide personalized solutions, experience and innovation in the field of MS. Last, but not least, it should provide face-to-face guidance and support.
Some opponents to early treatment of CIS claim that because MS DMTs have a partial effect and MS is a long-lasting disease, there is no reason to begin treatment prior to the clinical onset of the disease. The evidence that axonal damage begins in the early stages of MS, before symptoms are evident, provides a strong support for early intervention with immunomodulatory agents because what is lost is never regained. Early DMT therapy is therefore a logical strategy to counteract the inflammatory processes particularly relevant in the first years of the disease. The results of all four described pivotal clinical trials have demonstrated benefits for an immediate use of DMT to effectively reduce the risk of developing CDMS in CIS patients with a first demyelinating event and positive brain MRI [Comi, 2009]. To conclude, treatment for suspected MS patients should begin at the earliest stage possible to prevent initial axonal damage during the inflammatory stage of the disease. The cost of delaying treatment in MS could be stated as ‘What is lost is not regained’ [Schwid and Bever, 2001].
At present, the only option available to reduce the adverse effects of the progressive inflammatory disease MS on physical ability and cognition is to prevent nerve damage inflicted by inflammation at the earliest stages. The DMT should be started as soon as the diagnosis of MS is reasonably formulated, by a combination of clinical and paraclinical evaluations such as imaging findings, EDSS, MSFC, cognition, QoL, employment status and eligibility.
The findings of the DMT studies, and the established safety profiles, make it an appropriate choice for starting immunomodulatory treatment at the time of the first event suggestive of MS. In particular, patients at risk for acquiring high disability should be considered, and if they do not respond to DMT further escalating therapies may be selected. Comprehensive patient support programmes are useful in supporting and helping patients to start and stay on therapy.
Future studies will focus on earlier interventions to precisely determine the best approach to preventing initial neuronal damage with the proven DMT and whether early therapy can prevent, in addition to delay, transition to progressive MS.
1Controlled High-Risk Subjects Avonex Multiple Sclerosis Prevention Study.
2Early Treatment Of Multiple Sclerosis.
3Betaferon in Newly Emerging Multiple Sclerosis for Initial Treatment.
4Early glatiramer acetate treatment in delaying conversion to clinically definite multiple sclerosis in subjects Presenting with a Clinically Isolated Syndrome.
5Betaseron versus Copaxone in MS with Triple-Dose Gadolinium and 3-T MRI Endpoints.
6Betaferon Efficacy Yielding Outcomes of a New Dose.
7Controlled High risk Avonex Multiple sclerosis Prevention In Ongoing Neurological Surveillance.
The authors kindly acknowledge the contribution of Physicians World Europe for medical writing assistance.
This work was supported by Bayer Schering Pharma AG.
RG has received honoraria for speaking, consulting and board activities from Teva, BiogenIdec, Novartis, Bayer Schering, Merck Serono and Sanofi Aventis. JSW has received honoraria (and travel expenses where applicable) for services on steering committees for Novartis Pharmaceuticals and Sanofi Aventis, a scientific advisory board for Teva pharmaceuticals, and data safety monitoring boards for Eli Lilly and UCB; for consultancy (Genentech, Novartis Pharmaceuticals, Sanofi Aventis, Teva Neuroscience, Teva Pharmaceuticals, Acorda, Actelion, Bayer Health Care, Facet Biotech); for lectures (Consortium MS Centers, Sanofi Aventis New Zealand, Sterling Meeting Services, USF Health Professionals, Texas Neurological Society, Teva Pharmaceuticals, Lone Star Chapter NMSS, ICHE, Pfizer, EMD Serono, SUNY Stony Brook Fdn, UTMB, Medscape CME, University of Buffalo, Serono Symposia Intl Fnd). JSW also declares receiving royalties for out-licensed monoclonal antibodies through UTHSCH from Millipore (Chemicon Intl) Corp. MPA declares receiving money for board memberships, honoraria, and grants/grants pending from Bayer Schering, Merck Serono, and Biogen Dompé, and honoraria from Sanofi Aventis. GC has received honoraria (and travel expenses where applicable) for consultancy from Bayer Schering and grants/grants pending from Novartis.