For patients with Relapsing Remitting Multiple Scierosis Beta Interfaerons and Glatiramer Acetate were the first to be licensed for treatment. This review deals with one major question: when to initiate therapy? Through exploring the unique characteristics of the disease and treatement we suggest an approach that should be helpful in the process of decision-making.
Until the early nineties, the major challenge for Multiple Sclerosis (MS) neurologists was to establish the diagnosis when a patient presents clinical picture suggestive of MS. Treating MS patients consisted mainly of education, supportive measures, and symptomatic treatments. MS appeared as an unmodifiable illness with many stories of disappointment when using immunosuppressant therapy. When beta-Interferons (IFNs) and Glatiramer Acetate (GA) were introduced, and vigorously marketed, into the world of MS, patients’ and neurologists’ prospective about MS changed and focus in management shifted. Now, experts in MS around the globe strive to best answer the question that appears with every newly diagnosed patient: when to start treatment with Disease-Modifying Drugs (DMDs)? Which drug to use?
This review deals with the former question. We will address the second question in a later review. Despite the absence of class I recommendations, we support the worldwide trend towards early treatment. We explore the possible reasons behind this consensus. We also address the implications of starting treatment early, and how one could explain the reluctance of some MS patients to start DMDs despite these medications being partly or completely supported by health benefits in numerous countries. We review these treatments in Relapsing Remitting MS (RRMS), since as we discuss in this paper, IFNs and GA were licensed for this disease phenotype. Efficacy was not shown for Primary Progressive MS (PPMS). We conclude by detailing our approach to the question of when to start, and we emphasize the importance of individualizing this decision to the patient’s circumstances.
In this review we focus on IFNs and GA as they are first-line treatments and widely available. We will not review in detail when to start Mitoxantrone or Natalizumab. Suffice it to say that we exceptionally use them as first-line when Relapsing Remitting MS (RRMS) appears to be rapidly leading to disability. Fingolimod has not been commercialized long enough that its place in the armamentarium against MS can yet be determined.
In this section we discuss the features of MS that makes it a challenging disease. MS is mysterious for its true onset in individual patients, and thus talking about early versus late treatment remains to be a relative rather than an absolute argument. Here, we start by briefly exploring the difficulties that may face a neurologist diagnosing MS and prognosticating its path in individual patients.
Although patients may present with a highly suggestive clinical picture (typical age group and symptoms) supported by a brain MRI that fulfill the widely accepted Barkhof’s criteria [
One of the most puzzling aspects of MS is its great variability in terms of natural course. A considerable percentage of 15–20% of MS patients will continue to mobilize freely and live independently throughout their lives without treatment [
Most recent studies, however, suggest that the disease course during the first few years after clinical onset, and before the Expanded Disability Status Scale (EDSS) reaches a score of 3 to 4, is the most relevant in terms of predicting disability progression. In their natural history study of 806 essentially untreated MS patients, Scalfari et al. [
On the other hand, there remains no consensus on the definition of benign MS. While it is usually defined based on a long duration (longer than 15 years) and a low score of EDSS (no cutoff point has been agreed upon yet), it is important to keep in mind that EDSS scoring system does not parallel functionality, productivity, or quality of life. EDSS does not account for highly prevalent and disabling symptoms in MS like fatigue, depression, cognitive decline, and pain [
In an attempt to better understand benign MS and to reach a better definition for this disease phenotype, Rovaris et al. [
An important fact in MS is that the disease process starts several years prior to clinical presentation, if not very early in life. The epidemiological studies, particularly on the migrant populations, strongly suggest that a number of environmental factors may start their effect on genetically susceptible individuals very early in life, if not before birth [
This proposal is amplified by the finding of a high percentage of abnormal conventional brain MRIs in patients with CIS who convert to CDMS [
Advanced MRI studies reveal several changes at the very earliest stages of MS or in patients with CIS undetected by conventional MRI [
Sometimes one encounters patients years after the clinical onset of the disease, and the question of starting a DMD at this stage can be equally challenging. Two questions to consider here. The first is, would DMDs retain their efficacy even when started later during the relapsing remitting stage of MS? In other words, would the therapeutic window for DMDs within the relapsing remitting stage of the disease remain open as long as there are relapses and/or Gadolinium enhancing lesions, and probably closes when the progressive phase (without relapses) is reached? As clinical studies proved IFNs and GA to be much less effective in secondary progressive MS (SPMS) [
Most MS patients included in the pivotal IFNs and GA trials had a disease duration of one year or more; in one study disease duration of 6 months was allowed and in another, disease duration had to be of more than 3 years. Otherwise, disease duration was not an exclusion criterion except in one study when patients were excluded if they had MS for more than 15 years. Patients however were excluded if they had an EDSS score of more than 5.5 or 6, and in one study when EDSS was more than 3.5 [
A diagnosis of MS has to be made cautiously for its life-long implications. It is a stigma in the community and affects every aspect of life, including personal relationships, employment, and medical insurance. Tools for forecasting prognosis are lacking, and the natural course of illness is very variable. Disease behavior during the first two years and MRI changes early in the disease hold promise as ways of speculating the course of MS.
In order to tell about the true onset of MS, brain MRIs may need to be done on at-risk populations at a very young age (e.g., offspring of affected parents in areas of high prevalence of MS), however may remain very difficult to know without a biological or genetic marker for MS.
Until recently, IFNs and GA were the only approved treatment for MS and remain the best studied to date. Despite the growing trend towards earlier treatment, DMDs will not be always instituted as soon as a diagnosis is made, or anticipated as in CIS. In order to understand why, we review the effects and limitations of the currently available DMDs, their safety, and the evidence supporting starting treatment after a CIS in the following section.
In randomised clinical trials (RCTs) efficacy of DMDs in RRMS was measured based on their effect on relapses frequency and severity, effect on MRI parameters (number of new T2 lesions, T2 lesions load, lesions enhancement, brain atrophy, etc.), and effect on confirmed changes in EDSS. In studies on patients with CIS the primary focus was the time to conversion to CDMS.
The pivotal studies that lead to licensing of IFNs and GA were heterogeneous to some extent and difficult to compare. Each had its own strengths and limitations. In the literature there is a considerable number of reviews and meta-analyses, interestingly with variable and sometimes conflicting conclusions, depending on the weight given by each author or reviewer to certain points of weakness or limitation, the time when the analysis was done (before or after other studies or open label extensions were completed), and probably personal experience.
In general, one can say that all currently available IFN preparations and GA proved superior to placebo in terms of effect on frequency of relapses in RRMS patients. In the pivotal trials [
On the other hand, open label studies for both agents, despite all the inherent limitations of this study design, suggest a sustained benefit for these DMDs on relapse frequency. It is however expected in MS patients to have fewer relapses as the disease duration increases, when degenerative changes override inflammatory events. It is also difficult to make absolute conclusions from these studies due to the fact that only a proportion of patients who were initially included continued to follow up. One may wonder if the patients who did not enter the follow up actually fared worse. Propensity to bias due to loss of randomization may render the trials less representative of the natural history of RRMS. Moreover, patients may have not been monitored, may have discontinued the drug for considerable amount of time, switched to another DMD, or added another one [
In addition to the effect on relapse rate, open label extensions of the original studies and nonrandomised retrospective studies showed delayed time to conversion to SPMS and to reaching high EDSS scores in the treated group [
The effects of the IFNs and GA on MRI parameters in the pivotal trials in MS was even more robust than the clinical effect, and further supported their role in managing MS. Studies, including those which followed the pivotal trials, suggest sustained effects on diseases burden as seen on MRI. A favourable effect was demonstrated on the number of enhancing lesions, rate of black holes formation, as well as on the rate of brain atrophy [
The relation between the rate of brain atrophy and the subsequent development of disability was well-demonstrated by Fisher et al. in an 8-year follow up study for the patients originally involved in a phase III IFN-1a trial [
In a small open-label, single-blinded study, a positive effect of IFN-1a on the gray matter fraction (GMF) but not white matter fraction (WMF) after a follow up period of 3 years was suggested. This finding may indicate different dynamics and mechanisms behind white matter and gray matter atrophy in MS, and that the favorable DMD effect on brain atrophy might be mainly through reducing GMF atrophy [
Although it is desirable to have less and milder relapses during the course of MS, the ultimate goal of DMDs therapy in MS remains to be preventing, or at least delaying, the progression of disability. A recent meta-analysis by Sormani et al. [ Another important aspect of measuring the efficacy of DMDs is through their effect on cognition and fatigue; two major sources of disability which are not captured by EDSS. An effect of DMDs on cognitive decline can be inferred from the fact that they may reduce brain atrophy, particularly their effect on gray matter, as discussed above. However it is certainly more convincing to look at the studies that compared DMDs-treated and untreated MS patients in relation to the degree of cognitive impairment. Pliskin et al. [ However, Montalban and Rio [ The only pivotal IFN study that incorporated neuropsychological testing was that for Avonex [ Same holds true for GA. More than 10 years ago, Weinstein and colleagues [ As we pointed out earlier in this review, fatigue is one of the most debilitating symptoms in MS, even in patients with low EDSS scores, and it affects quality of life and employment. A major problem with fatigue is that surrounding people with limited knowledge about MS may not appreciate the impact of this symptom on the ability to work. Fatigue is one major aspect of MS that could benefit from treatment with DMDs. A good number of studies looked at the correlation between brain atrophy and fatigue. The importance of this correlation relies in the fact that DMDs may actually prevent brain atrophy, as discussed above. From there one can infer that DMD may then reduce the incidence or severity of fatigue by reducing the rate or degree of brain atrophy. Studies, however, did not agree as whether or not fatigue can be correlated with MRI brain measures like lesion load or brain atrophy. Bakshi et al. [ On the other hand, a number of recently published studies did show an evidence for an association between brain atrophy in MRI and fatigue [ There are no placebo-controlled studies on the effect of DMDs on fatigue in MS patients. In his observational study, Ziemssen et al. [
There is a number of studies that dealt with the economic impact of multiple sclerosis by looking at the direct and indirect costs of the disease, and how do DMDs contribute to the total financial burden.
Directs costs of MS are the medical costs including payments made to healthcare providers for inpatients and outpatient care, physical therapy, rehabilitation, laboratory services, drugs, and so forth, Indirect costs are those related to loss of productivity and loss of Quality of Life (QOL).
It became obvious through the results of these studies that the more advanced the disease the higher are the costs, and that the indirect costs of the disease are the major contributor to the cost of illness in MS [
The evaluation of DMDs cost-effectiveness was for the major part done through cross-sectional study of direct costs, the so called Economic Modeling Studies. Cost-effectiveness was tested through measuring cost per quality adjusted year life (QALY). Studies were heterogeneous in many ways and difficult to compare. Parkin et al. [
Other economic studies reached different conclusions in regard to cost-effectiveness of treatment with IFNs, including for treatment after CIS [
The uncertainty around the cost-effectiveness of DMDs reflect the complexity of measuring this endpoint, and the high variability of study design prevented us from reaching a definite conclusion. Detournay [
IFNs and GA have been in use for about 20 years. They proved to be safe and well-tolerated.
Adverse events from the use of IFNs and GA may be divided in to three major categories. Common but non life-threatening side effects, mainly flu-like symptoms with IFNs injection, and subcutaneous adipose tissue necrosis with GA injections. Despite being referred to as “benign side effects”, flu-like symptoms and injection site reactions are recognized as major factors in impaired adherence to treatment, as we discuss below. Rare but serious side effects: fulminant liver failure and skin necrosis, with the use of IFNs and GA respectively. The emergence of NAB during the course of treatment with IFNs.
A recently published cross-sectional study, although insensitive to rare side effects, proved long-term safety of IFN-1b for RRMS after a 16-year follow up [
Two recent prospective studies on a small number of patients [
The current literature abounds in studies and analyses supporting the initiation of DMDs early in RRMS or even after the first demyelinating event suggestive of MS (CIS) [
This may represent a global trend [
However, this trend is justified by the available evidence. As discussed above, the characteristics of the disease (MS) and the characteristics of the treatment (IFNs and GA) support the use of DMDs after the CIS. Goodin and Bates [ Evidence of axonal injury early on in the disease [ Evidence of brain atrophy at the earliest stages of the disease; higher rate of atrophy occurred in RRMS, compared to SPMS [ In CIS patients, the number of lesions at baseline MRI, and the increase in lesion volume over the first 5 years, both correlated with EDSS at 14 years [ Treatment with IFNs or with GA delayed conversion to CDMS. In addition, larger treatment effects were obtained when treatment was started in the CIS stage than when started in the RRMS stage, and larger when started in the RRMS stage than when started in the SPMS [
Collectively, IFNs and GA are effective in modifying the natural course of MS through their partial, albeit significant, impact on relapse frequency, disease burden on MRI, and time to sustained disability. Their impact on other aspects on MS particularly disabling symptoms like fatigue and cognitive impairment remains to be confirmed. The inherently limited studies on cost-effectiveness support their use, and postmarketing surveillance proved their safety. The rationale behind starting treatment as early as after a CIS suggestive of MS and supported by abnormal MRI brain seems to be well-supported by a growing number of clinical and pathological studies, however other important considerations, as discussed below, argue against starting treatment after a CIS in every single patient, and thus starting early cannot be considered as a standard of care.
MS is a chronic disorder, and it is not yet known when it is ideal to stop DMDs once started. Patients knowing this about their treatment may reduce their tolerance to what we, as physicians, identify as minor side effects, and may impair long-term adherence. Starting DMDs after a CIS could mean starting during early to mid-twenties, when a patient is busiest studying or working, with many plans and dreams on board. For women of child-bearing age, the decision to start a DMDs might well be affected by or affect their plans to conceive, and that should be taken in consideration too.
Treatment discontinuation ranges from 14 to 44% in long-term postmarketing studies of IFNs [
There is yet no global consensus on the clinical significance of the development of NABs in MS patients treated with IFNs. However a panel of European experts believe that NABs affect the clinical efficacy of IFNs, and they recommended a set of actions to be taken depending on how well the patient is doing clinically [
Many factors contribute to the emergence of NABs including the formula of the IFN used, route of administration, dose and frequency [
Developing NABs early during the disease course, as in the case of starting treatment after a CIS, may have different effects on treatment success than when NABs develop later. Indeed this has not being studied before, but is worth considering when deciding about initiating treatment early.
Being on one of the current DMDs may deprive the patient from the chance to participate in clinical trials testing new treatments in MS, or may delay his or her entry to the study. It also means that it will be more difficult to form a control group to which treated patients can be compared prospectively or retrospectively.
In a patient with a single demyelinating event suggestive of MS, the decision of whether to treat or not and when to start a DMD has to be made by mutual agreement between the treating neurologist and his/her patient. In order to optimize adherence and achieve maximum benefit from treatment, a patient must be well-informed, contribute to decision making, be comfortable with the decisions made, and be willing to commit.
Ross [
The patients should be made aware of the limitations of the current treatment options. Realistic goals need to be set, and the patients should learn that this is not a “symptomatic” treatment. Having relapses does not necessarily indicate treatment failure. Side effects need to be addressed and strategies to decrease their frequency, intensity, and impact on their lives should be taught. Patients’ personal circumstances may prevent them from starting treatment at a certain point of time, when they cannot handle extra stress in their lives. For the sake of optimizing adherence, it could be wise therefore to wait till after patients are more capable of handling all the difficulties around starting a new treatment. If the different companies have setup support services for MS patients starting on DMDs, it is our deep-felt conviction that these companies cannot do as good a job as the MS clinic nurse who already knows the patient, his/her personal circumtances, character and approach to life. This nurse should be the person in direct contact with the patient to help him/her to make the appropriate decision as to when and what to use as a DMD.
Patients will ask about newly emerging therapies. The effects on being on one of the current first-line DMD prior to starting treatment with one of the newer agent are not yet known. The possible impact on the response to treatment with the newer agent or on development of side effects with the newer treatment is not known as well.
Finally, in an ideal world, financial limitations should not play a major role when it comes to making decisions about treatment. However an ideal world does not exist. A neurologist’s decision will be frequently influenced by the health care system regulations. Free medical care is not available everywhere. In certain areas, certain treatments are not available to every patient, depending on a number of factors including the type of insurance they hold, the rules and regulation of the country or province where they live, and the type of disease they have. The Helsinki accord which has not been signed by all DMD companies, call for trials not to take place in the countries where the drug will not be available for patients post trial. Unfortunately the number of these countries has been increasing rather than decreasing. We hope this will convince big Pharma to make the drugs more available rather than continue the race for the most expensive products.
After considering all the different factors discussed in this review, our answer to the question