Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Relapses in multiple sclerosis: Relationship to disability

Multiple Sclerosis and Related Disorders, Volume 6, March 2016, Pages 10–20


Multiple sclerosis (MS) is a recurrent inflammatory disease of the central nervous system, which ultimately causes substantial disability in many patients. A key clinical feature of this disease is the occurrence of relapses, consisting of episodes of neurological dysfunction followed by periods of remission. This review considers in detail the importance of the occurrence of relapses to the ultimate course of MS and the impact of relap setreatment (both acutely and prophylactically) on the long-term outcome for individuals. The ultimate goal of therapy in MS is the reduction of long-term disability. Clinical trials in MS, however, typically only extend for a very short time period compared to the time it takes for disability to evolve. Consequently, short-term outcome measures that are associated with, and predict, future disability need to be identified. In this regard, not only are relapses a characteristic feature of MS, they have also been proven to be associated with the occurrence of long-term disability. Moreover, treatments that reduce the number and severity of these attacks improve the long-term prognosis.


  • MS is a chronic, recurrent inflammatory disease of the brain and spinal cord.
  • Relapses are a key clinical feature of the MS disease process.
  • Relapses can cause disability both acutely and in the long-term.
  • Treatments that suppress the occurrence of relapses can improve long-term prognosis.

Keywords: Multiple sclerosis, Disability, Relapse, Treatment, MRI, Lesion.

1. Introduction

Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system (CNS) (Bar-Or et al, 1999, Compston et al, 2006, Conlon et al, 1999, and Hauser and Oksenberg, 2006), which ultimately causes substantial disability in many patients (Bar-Or et al, 1999, Compston et al, 2006, Conlon et al, 1999, and Hauser and Oksenberg, 2006). About 90% of MS cases are characterized by the occurrence of clinical attacks, which consist of episodes of neurologic dysfunction lasting for some period of time (usually defined as more than a day), and then followed by a remission of symptoms (Cook et al, 2012 and Lublin and Reingold, 1996). Indeed, the occurrence of such attacks has been an essential component of every diagnostic scheme for MS in the past 50 years – from the early, pre-magnetic resonance imaging (MRI) criteria of the Schumacker committee in 1965 (Schumacker et al., 1965) through the most recently revised International Panel criteria published in 2011 (Polman et al., 2011). Thus, clinical attacks are essential to the very definition of MS. Although the neurologic disability experienced during an attack can be quite marked, some neurological recovery from these attacks occurs in the majority of patients and is often seemingly complete. Therefore, the question naturally arises as to whether (or to what extent) these clinical attacks are responsible for, or contribute to, the ultimate disability experienced by individual MS patients.

2. Pathology of multiple sclerosis

In MS, there is pathological evidence of multi-focal injuries of varying ages to the myelin sheaths surrounding the axons, to the oligodendrocytes and, to a somewhat lesser extent, the nerve cells and their processes (Bar-Or et al, 1999, Compston et al, 2006, Conlon et al, 1999, and Hauser and Oksenberg, 2006). Axonal injury within active lesions and gray matter demyelination also both occur (Bo et al, 2003, Ferguson et al, 1997, Lucchinetti et al, 2011, Peterson et al, 2001, and Trapp et al, 1998). Within acute lesions, presumably under the influence of cellular adhesion molecules (CAMs) and pro-inflammatory cytokines, auto-reactive, cluster of differentiation (CD)4+, thymic-derived lymphocytes (T cells), CD8+ cytotoxic lymphocytes, CD20+ bone marrow-derived lymphocytes (B cells), and CD68+ macrophages cross the blood–brain barrier (BBB) to enter the CNS (Bar-Or et al, 1999, Compston et al, 2006, Conlon et al, 1999, and Hauser and Oksenberg, 2006). These activated cells are thought to contribute to the CNS tissue damage that is seen in acute MS lesions.

MRI lesions characterized by only T2-hyperintensity (i.e., T2-only lesions), are much more likely to have preserved myelin than MRI lesions characterized by persistent T1-hypointensity and a reduced magnetization transfer ratio (MTR), in addition to T2-hyperintensity (i.e., T2/T1/MTR lesions) – in fact, only 20–45% of T2-only lesions are associated with demyelination on histopathological examination (Fig. 1) compared to 80–83% of the T2/T1/MTR lesions (Fisher et al, 2007 and Moll et al, 2009). Nevertheless, regardless of the presence or absence of demyelination, most of the T2-only MRI lesions still contain activated microglia and evidence of BBB breakdown. Even in the so-called normal-appearing white matter, 30% of the regions sampled contain activated microglia. It remains to be determined whether activated microglia within myelinated T2-only lesions are causing new damage or are simply responding to a cytokine release associated with the breakdown of the BBB from Wallerian degeneration, or both. Indeed, it is possible that the microglia may actually be involved in cleaning up lesions (possibly even promoting repair), whereas the macrophages may be responsible for the actual tissue damage (Yamasaki et al., 2014).

Fig. 1

Fig. 1 Schematic representation of the pathological evolution of inflammatory-mediated demyelination of brain white matter. In normal appearing white matter (A) axons (red) are surrounded by myelin (green). (B) Acute multiple sclerosis (MS) lesion: An early event in white matter demyelination is the entry of immune cells (blue) from the blood. These cells participate in and are required for white matter demyelination. (C) Chronic active MS lesion: With time, the immune cells disappear from the center of the MS lesion, but remain at the border of the lesion where they slowly expand demyelination. (D) Chronic inactive MS lesion: Eventually, the immune cell composition of the lesion decreases and astrocytosis (yellow) is a prominent feature of the demyelinated area. (With permission from: Trapp BD). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

Thus, MS lesions show considerable histopathological diversity, ranging from chronic gliotic demyelinated scars, to highly inflamed demyelinating lesions, to less inflamed regions in which the myelin is preserved (Fig. 1).

3. The blood–brain barrier

The endothelial cells in the CNS are non-fenestrated and have extraordinarily tight junctions between them. For a long time, these tight junctions were thought to be primarily responsible for creating the BBB, but it is now known that this barrier is actually the result of a very complex interface between the vascular system and the CNS, which is called, collectively, the neurovascular unit (Engelhardt et al, 2014, Holman et al, 2011, and Muoio et al, 2014). This unit includes the endothelial cells, the extracellular matrix, the basement membrane, and also the cells surrounding the endothelial cells, notably pericytes and astrocytes. Together, this unit works to both provide mutual trophic support and to make the entry of hydrophilic molecules (by active transport or diffusion) and transcytosis into the CNS extremely selective. A focal breakdown of the BBB can be caused by any one of a variety of CNS insults including inflammation, toxic exposure, neoplasia, trauma, and ischemia.

In MS, the breakdown of the BBB is thought to represent a critical step in the development of a new MS lesion and the basis for an acute MS attack. Nevertheless, whether this BBB breakdown is the initial event in lesion formation is not entirely clear (Filippi et al, 1998 and Goodkin et al, 1998). Thus, using the magnetization transfer ratio (MTR), focal changes in the relative concentrations of free and bound water can be detected in those otherwise normal-appearing CNS white matter regions that, months later, are destined to become a gadolinium (Gd)-enhancing lesion on MRI. Presumably, these MTR changes reflect biochemical alterations, which are the initial events in lesion formation. It is nonetheless possible that these early events represent a selective breakdown in the BBB not detectable by conventional MRI and, in this view, the more general breakdown of the BBB, which is reflected by the Gd-enhancement, would be a secondary phenomenon.

4. The timing of MS attacks

Certain environmental factors may increase or decrease likelihood of an MS attack. For example, pregnancy, especially in the last trimester, reduces the risk of an attack compared to the pre-pregnancy and the post-partum state (Confavreux et al, 1998, Salemi et al, 2004, and Vukusic et al, 2004). Also, the periods surrounding those times when a patient is experiencing non-specific infectious syndromes (e.g., rhinorrhea, fever, cough, malaise, diarrhea, and gastrointestinal distress) are associated with increased risk of an MS attack compared to those times when a patient is free of infectious symptoms (Andersen et al, 1993, Edwards et al, 1998, Panitch, 1994, and Sibley et al, 1985). These attacks are often attributed to a triggering of the autoimmune response by these infections through molecular mimicry or other mechanisms (Bar-Or et al, 1999, Compston et al, 2006, Conlon et al, 1999, and Hauser and Oksenberg, 2006). Indeed, some believe that such molecular mimicry may represent the environmental mechanism by which MS is triggered in the first place.

5. Assessing outcome: disability

The 10-point Disability Status Scale (DSS), or its expanded version (EDSS), which is divided into half-step increments (except between 0 and 1), is used to quantify the degree of MS-related disability in individual patients (Kurtzke, 1955 and Kurtzke, 1983). When assessing short-term disability in randomized clinical trials (RCTs), investigators have relied on measures such as “confirmed” progression, which represents a change in EDSS score (often 1–2 points) that has been maintained for some period of time (typically 3 or 6 months). Even though this particular measure has generally been favored in the analysis of clinical trials, other disability measures exist, including the actual EDSS change (ΔEDSS) or the categorical change in EDSS score over the course of the RCT, the composite change in EDSS subscale scores, and the change in other outcome measures such as the timed 25-foot walk or the 9-hole peg test. Importantly, the clinical impact of any particular numerical change in EDSS score will be difficult to interpret due to the non-linear nature of the EDSS scale (Cohen et al., 1993). Similar concerns also pertain to the interpretation of changes in the other outcome measures. Consequently, any method that is chosen to assess disability during an RCT requires validation (Prentice, 1989).

The assessment of long-term disability has been even more variable among studies. Some authors have chosen to use so-called “hard” (i.e., unambiguous) outcomes such as the time to an unremitting EDSS of 6 or 7, conversion to secondary progressive multiple sclerosis (SPMS), or death (Fisher et al, 2002, Goodin et al, 2011, and Goodin et al, 2012a). Others have chosen to assess long-term disability either as the time to a “confirmed” EDSS change (Kappos et al, 2006, Pozzilli et al, 2005, Prosperini et al, 2009, and Tomassini et al, 2006) or as the magnitude of the ΔEDSS (Bermel et al., 2013) during the follow-up period. Each of these methods has potential drawbacks. For example, the ΔEDSS method confounds the disability outcome with the baseline state. Thus, by this method, a person who progresses from EDSS=0 to 4 is considered to be more disabled than a person who progresses from EDSS=4 to 7. Moreover, the use of this method assumes that the EDSS scale is linear, which, clearly, it is not (Cohen et al., 1993). When using the method of time to “hard” outcome, the rate of progression to the outcome for an individual will likely depend upon his or her disability level at trial entry and such a circumstance necessitates the inclusion of baseline variables in the regression equations. The same is true for the outcome of “confirmed” progression and, in addition, this method does not distinguish transient disability (which recovers after 3 or 6 months) from permanent disability. For example, when these “confirmed” outcomes were studied using the combined data from the placebo arms of two published clinical trials (Liu and Blumhardt, 2000), it was demonstrated that, even as early as the end of the RCT, half of those patients who experience sustained progression have already reverted to a non-progressed status regardless of the definition of the “confirmed” change that they employed (Table 1). Moreover, use of any fixed amount of “confirmed” EDSS change to measure disability again assumes that the EDSS scale is linear, which, clearly, it is not (Cohen et al., 1993).

Table 1 Relationship of “confirmed” EDSS change to RCT outcome (i.e., sustained EDSS change at the conclusion of the RCT) (Liu and Blumhardt, 2000).

Definition EDSS change (%) PPV
Confirmed Sustained
1 Point 3 Months 32.2 15.3 0.48
1 Point 6 Months 21.4 14.1 0.67
2 Points 3 Months 12.1 6.4 0.53
2 Points 6 Months 9.3 5.1 0.55

EDSS=Expanded Disability Status Score; RCT=randomized controlled trial.

Confirmed EDSS change=meets the definition of progression provided.

Sustained EDSS change=remains progressed at end of RCT.

PPV=positive predictive value=(sustained change/confirmed change).

Therefore, it is critical that any study on the long-term outcome in MS include, in the analysis, the patient’s pre-study or pre-therapy clinical and radiographic status. For example, those patients with higher EDSS scores, a greater burden of disease on MRI, or more active MS prior to starting therapy will often subsequently experience a more progressive or more active disease course. Failure to include these predictive baseline variables in the analysis may make it appear that certain “on-study” or “on-drug” events are important predictors of long-term outcome when, in fact, these events are merely a reflection of the patient’s baseline state. These concerns are especially important in the analysis of the long-term impact of a disease-modifying therapy (DMT). Thus, patients with a bad baseline state may appear to do worse while on therapy, even in circumstances where the drug is working equally well in all patients.

6. Impact of relapses on disability

6.1. Post-attack neurological status: Are there persistent deficits?

Many studies show that MS attacks can sometimes leave permanent residual neurological deficits. For example, the Mayo Clinic study (Keegan et al., 2002) on the use of plasma exchange (PE) in treatment of severe acute demyelinating episodes (median pre-attack EDSS=0; median post-attack EDSS=8; and with a poor recovery after pulse high-dose glucocorticoid treatment) seems to confirm this fact. Fifty-nine patients were followed for 1 year after their PE treatment. Many of the patients included in this study had an alternative (non-MS) final diagnosis such as neuromyelitis optica, Marburg variant, acute disseminated encephalomyelitis, or acute transverse myelitis (Keegan et al., 2002). Nevertheless, 22 patients had relapsing-remitting multiple sclerosis (RRMS) and, of these, 8 had marked improvement; 5 had mild or moderate improvement; and 9 had no improvement following PE therapy (Keegan et al., 2002). Thus, many of the RRMS patients who participated in this study suffered severe (and likely permanent) neurological consequences from their acute demyelinating event (Keegan et al., 2002). Similarly, in the Optic Neuritis Treatment Trial (ONTT), none of the treatment groups had 100% visual recovery following their clinical event (Beck et al., 1992). Although the initial study only reported outcomes at 6 months, at this time point, all of the curves seemed to be reaching plateaus that were well below full recovery (Fig. 2). Indeed, at the 15-year mark, 27% of the cohort still only had a partial recovery of visual function (The Optic Neuritis Study Group, 2008). More subtly, following a clinical episode of optic neuritis, there is frequently a fixed delay in the latency of the P100 component of the visual evoked potential (Compston et al., 2006). This delay indicates, quantitatively, that there is residual tissue damage (demyelination) in the anterior optic pathways. Such a delay can also be found in visually asymptomatic MS patients, presumably due to similar tissue damage from a sub-clinical event. In sum, there are multiple lines of conclusive evidence to indicate that MS attacks can be (or, perhaps, usually are) followed by detectable functional deficits and that sometimes these residual deficits, by themselves, can be quite debilitating.

Fig. 2

Fig. 2 Time course of recovery in visual fields following an episode of acute optic neuritis. The recovery is displayed in three treatment groups: (1) intravenous methylprednisolone (IVMP) followed by oral prednisone; (2) oral prednisone; and (3) placebo. Recovery in the glucocorticoid-treated groups is faster than in the placebo group, but in none is the recovery 100%. The difference in final outcome for IVMP and the other groups was not statistically significant (Beck et al., 1992).Copyright © 1992 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society.

Despite this conclusion, however, it is still unclear whether all such events are equally important and whether treatment of some events might have a greater impact on later disability status than others. As considered in subsequent sections, perhaps clinically evident events are more important than sub-clinical ones, or perhaps attacks occurring while on a DMT are more (or less) important than those occurring while off such therapy. Alternatively, perhaps MS events meeting stricter definitional standards may be more important than non-verified historical or subjective events. To assess this latter possibility, the authors of the recently completed CombiRx trial examined whether different attack definitions made any difference to the trial outcomes (Lublin et al., 2012). Possible attacks in this trial were divided into three categories. The first was protocol-defined exacerbations (PDE), in which patients were examined within one week of the attack onset, the clinical episode was of an appropriate duration, and a protocol-defined change in neurologic function was documented on examination. The second group was non-protocol-defined exacerbations (NPDE), in which the confirmatory examination was conducted more than a week after attack onset. The third group consisted of purely subjective exacerbations, in which patients reported new clinical symptoms of an appropriate duration but did not have changes in the EDSS and/or did not undergo any confirmatory examination. In comparing the treatment response in these three groups, it was clear that the attack definition made little difference to the trial outcomes (Fig. 3A and B). However, the event rates steadily decreased for each level of increased restriction on the definition (Fig. 3A). Thus, making these particular distinctions served only to reduce the number of events and likely, thereby, to reduce the power of the RCT to detect a treatment effect. At the extremes, however, the PDE may miss true relapses and the subjective exacerbations may include events that are not relapses. In such circumstances, the mixture of false negatives and false positives will weaken the predictive power of relapses to be associated with future outcomes.

Fig. 3

Fig. 3 The impact of attack definition on outcome in a clinical trial. Three definitions of an attack are employed.”Protocol-defined exacerbations (PDE)” and “non-protocol-defined exacerbations (NPDE)” both required confirmation by examination (see text for a discussion). For “subjective exacerbations,” no examination was performed and the attack was purely historical. Panel (A) shows the distribution of the different attack types, each of which represents about one-third of the total. Panel (B) shows the impact of treatment on each of the exacerbation type, which is the same regardless of attack definition. Treatments investigated in this study were weekly interferon beta-1a (IFN) and glatiramer acetate (GA) (Lublin et al., 2012).

Even if different types of clinical events are difficult to distinguish by examination, it may still be the case that some MS attacks leave objective deficits whereas others do not. To address this question, Lublin et al. (2003) combined placebo data from several MS clinical RCTs to determine whether residual neurological deficits were present following clinical MS attacks. They identified a cohort of 224 patients who had undergone an EDSS assessment both before and at least 30 days after experiencing a clinical attack, and measured the ΔEDSS – the EDSS score after the attack minus the EDSS score before (Fig. 4A). Of these patients, 95 (42.4%) had experienced a detectable neurological deterioration in association with their clinical event (Fig. 4A). However, 43 patients (10.3%) had experienced an improvement over this same interval (Fig. 4A). Because attacks seem unlikely to cause improvement, these “improved” patients likely reflected either the inherent variability of the EDSS measurement or the well-documented fluctuating nature of a patient’s function. In either case, assuming that the distribution of these variable measurements are symmetrical around (ΔEDSS=0), one can subtract the number of observed improvements both from itself and from the number of observed deteriorations in each EDSS category in order to provide an estimate of the number and distribution of excessive deteriorations found (Fig. 4B). This analysis suggests that, in the cohort, there are 50 patients (22.3%) who have a residual disability of approximately 1 EDSS point when measured 30 days or more following an acute MS attack (Fig. 4B).

Fig. 4

Fig. 4 The measured change in EDSS from baseline (pre-attack) to more than 30 days after a clinical MS attack. Panel (A) shows the measured EDSS changes (ΔEDSS) for the entire cohort. Panel (B) shows the results of subtracting the improved scores (ΔEDSS range: −3.5 to 0) from both the same degrees of improvement and from themselves. The result displays the distribution of the excessive deteriorations found in this cohort (Lublin et al., 2003).Panel (A) reprinted with permission from Lublin et al. (2003).EDSS=expanded disability status scale; MS=multiple sclerosis.

6.2. Attack-treatment: Does it affect the post-attack neurological status?

Glucorticoids and melanocortin peptides have both anti-inflammatory and immunosuppressive properties (Arnason et al, 2013 and Catania et al, 2010) and have been widely used in the management of acute MS attacks. Several such agents (or agents that stimulate the release of glucocorticoids) have been used. These include the synthetic glucocorticoids such as prednisone, prednisolone, dexamethasone, and methylprednisolone (MP) as well as ACTH. The pivotal study of this therapeutic approach (Rose et al., 1970), followed 197 RRMS patients who had received either placebo or intramuscular ACTH and reported a benefit of active therapy on the speed of recovery over 4 weeks. The next most important study of glucocorticoid therapy was the ONTT, which compared intravenous (IV) MP (1000 mg/day for 3 days followed by 11 days of oral prednisone at 1 mg/kg/day and a 3-day taper), oral prednisone only, and placebo in the treatment of acute optic neuritis (The Optic Neuritis Study Group, 2008). This trial evaluated the effect of therapy on visual recovery in a cohort of 457 patients. In the IVMP group, visual recovery was faster than in the placebo group, although by 6 months, this recovery was not statistically different between groups (Fig. 2). On the basis of these trials (in addition to several corroborative smaller studies), glucocorticoid therapy or ACTH is considered effective in speeding functional recovery from acute MS attacks. The possibility that glucocorticoid therapy might have a more lasting benefit in some cases is not supported by these studies, but neither can it be excluded (Fig. 2).

It is also not possible at present to draw any conclusions about whether total dose, route of administration, or the nature of the agent has any importance for the clinical use of glucocorticoids in MS. It is noteworthy that, unlike the synthetic glucocorticoids, ACTH stimulates the adrenal cortex to produce both glucocorticoid and mineralocorticoid hormones and also has a direct, steroid-independent effect on immune cells because it acts at the melanocortin receptor. However, based on the experience using glucocorticoids in the management of infantile spasms (Hussain et al, 2014 and Snead et al, 1983) where there is evidence of superiority of ACTH over prednisone, the equivalence of treating MS attacks with either ACTH or the synthetic glucocorticoid agents should not be assumed without direct comparative evidence.

6.3. Relationship of clinical attacks to long-term disability

Several natural history studies suggest that more frequent attacks early in the disease course, a shorter interval between the first two attacks, residual neurological deficits after early attacks, and early involvement of motor or cerebellar pathways are associated with a greater likelihood for a patient to become disabled in the long run (Confavreux et al, 2003, Confavreux and Vukusic, 2006, Kremenchutzky et al, 2006, Novotna et al, 2015, Runmarker and Andersen, 1993, Scalfari et al, 2010, Tremlett et al, 2006, Tremlett et al, 2008, Weinshenker et al, 1989a, and Weinshenker et al, 1989b). More recently, the authors of a French natural history study (Confavreux et al, 2003 and Confavreux and Vukusic, 2006) have suggested that the emergence of progressive disability in MS—either SPMS or primary progressive multiple sclerosis (PPMS)—is just an effect of chronologic age and that the EDSS milestones of 4 and 6 are reached on a predefined schedule not obviously influenced by relapses. In this view, PPMS is just RRMS in which the relapsing-remitting phase has been skipped, and seems to suggest that the relapsing phase of MS has little bearing on the ultimate accumulation of disability (Fig. 5). Similarly, in a recent Canadian study (Scalfari et al., 2010), both a greater number of attacks during the first two years of disease and a shorter interval between the first and second attack were associated with a shorter time (from disease-onset) to reaching the “hard” disability outcomes of DSS=6, 8, and 10. By contrast, attacks occurring later in the disease course did not have perceptible influence on these outcomes. Thus, regardless of the correctness of the Confavreux hypothesis, these natural history studies both support the notion that high relapse rates, especially those that occur early in the disease course, are predictive of long-term disability.

Fig. 5.

Fig. 5 The natural history of MS. Panel (A) shows the time course of developing progressive disease in patients who either had a relapsing onset of their disease (SPMS) or had a progressive onset (PPMS). Based on the superimposition of these two curves, the authors concluded that the progressive course of MS was not influenced by attacks. Panel (B) shows the estimated age at the onset of the progressive phase of MS based on the estimated age of onset of the clinical relapses. Estimation is necessary because the publications did not include all of the necessary data. Estimated age was taken as the mid-point for the age ranges provided in the tables (for the youngest and oldest patients this range was taken as the preceding or following decade). Onset age was estimated by adding the estimated age plus the median time to progression provided for each age group (Confavreux et al., 2003).SPMS=secondary progressive multiple sclerosis; PPMS=primary progressive multiple sclerosis.

The idea that relapses make some difference to long-term disability is also supported by long-term follow-up (LTFU) studies of MS therapies (Bermel et al, 2010, Bermel et al, 2013, Fisher et al, 2002, Ford et al, 2010, Goodin et al, 2011, Goodin et al, 2012a, Goodin et al, 2012b, Kappos et al, 2006, Pozzilli et al, 2005, Prosperini et al, 2009, and Tomassini et al, 2006). These long-term studies are methodologically quite challenging (Goodin et al., 2012b). Two shorter-term, non-randomized, studies looked at the relationship of relapses in the first year to disability measured 4 or 5 years later (Prosperini et al, 2009 and Tomassini et al, 2006). Both studies assessed disability as the occurrence of a 6-month “confirmed” 1-point EDSS change at any time during the follow-up period. The larger of the two (Prosperini et al., 2009) reported a significant positive relationship between relapses in the first year and (potentially transient) disability.

Four other longer-term studies followed the pivotal RCTs, which led to DMT approval, but few focused specifically on the relationship of relapses to long-term disability. In two of these studies, the case-ascertainment rates for the original study cohort after 15 years were only around 40% (Bermel et al, 2010, Bermel et al, 2013, and Ford et al, 2010). After 15 years in the long-term follow-up (LTFU) of the glatiramer acetate pivotal trial, no attempt was made to relate relapses to the final disability status. By contrast, at the 8-year time point for the LTFU of the weekly interferon-beta (IFNβ)-1a trial (Kappos et al., 2006), the authors reported that the number of relapses during the RCT was a significant predictor of the final EDSS≥6 status, although baseline EDSS and multiple sclerosis functional composite (MSFC) were stronger predictors of this outcome than the number of relapses. Perplexingly, the direction of the impact that relapses had on disability was inconsistent – being either helpful or harmful depending upon the regression model used (Fisher et al., 2002). At the 15-year time-point of the same trial (Bermel et al., 2013), in those patients who were originally randomized to receive IFNβ-1a, relapses were associated with a significantly greater odds ratio (OR) for being in the worst ΔEDSS quartile at the long-term follow-up (OR=4.4; p=0.01). A similar association between relapses and the odds of this outcome was not found in the group randomized to receive placebo, although the confidence intervals for the observations in the two groups overlapped considerably (Bermel et al., 2013).

The two other post-RCT studies had more complete case-ascertainment (Goodin et al, 2011, Goodin et al, 2012a, and Kappos et al, 2006). Thus, in the 8-year LTFU study of patients from the RCT of thrice-weekly IFNβ-1a (Kappos et al., 2006), the case-ascertainment was 68.2%. Moreover, because the missing data in this trial was due to “center” drop-outs (which, unlike individual dropouts, are balanced by the randomized design), the likelihood of ascertainment bias is considerably lessened. Compared to placebo, the patients randomized to high-dose IFNβ had a significantly lower relapse rate during the RCT and also a significantly slower accumulation of disability after 8 years, as measured by a 1-point increase on the EDSS scale sustained for 3 months. Unfortunately, there was no attempt to correlate the two observations; thus any direct assessment of the relationship between relapses and long-term disability progression is not possible from this study.

The second study was the LTFU of the IFNβ-1b trial, which took place both 16 and 21 years later (Goodin et al, 2011, Goodin et al, 2012a, and Goodin et al, 2012b). At the 16-year mark, similar to the thrice-weekly IFNβ-1a LTFU, case ascertainment was 69.9% (Goodin et al, 2011 and Goodin et al, 2012b). At the 21-year mark, the case ascertainment was 98.4%, although the only outcome explored in this particular study was death (Goodin et al., 2012a). Nevertheless, the 16-year study compared the baseline and 5-year, placebo-controlled data for both the ascertained and the 30.1% non-ascertained patients and, in this case, the two groups seemed to be quite well-matched (Goodin et al, 2011 and Goodin et al, 2012b). To explore the validity of the various short-term outcomes used in MS clinical trials, this study examined the relationship of clinical variables (measured both at baseline and during the RCT) to the “hard” disability outcomes of an unremitting score of EDSS≥6, progression to SPMS, or death (Table 2). The establishment of such long-term relationships is an essential component to the validation of the short-term measures that we use to establish efficacy in short RCTs (Prentice, 1989). In this study, regardless of how long-term disability was defined, the only “on-RCT” variables that contributed to its prediction were the annualized relapse rate and clinical measures of short-term disability progression. By far, the best measure of short-term disability was the “actual” EDSS change (i.e., the score at the end of the RCT minus the score at the start). Indeed, it is notable, but maybe not surprising, that this particular measure accounted for more than 5 times the variance in long-term disability as did the 3-month “confirmed” 1-point EDSS progression during the RCT (Table 2). One possible reason for this finding is that a 1-point change lasting 3 months may not have the same reliability as an EDSS change which takes place over a longer time interval. Another may be the fact that a “confirmed” 1-point change often doesnot persist for more than 3–6 months (Table 1).

Table 2 Univariate and multivariate relationships with long-term outcome (Goodin et al., 2012b).

Univariate relationships Odds ratio R 2 p-value
Baseline variables
EDSS 2.07 0.22 <0.0001
T2 BOD (cm2) 1.03 0.07 0.001
3rd Ventricular width (mm) 1.18 0.04 0.011
MS duration (yr) 1.07 0.05 0.003
On-RCT variables
Annual relapse rate 1.82 0.12 <0.0001
Actual EDSS Δ 1.59 0.11 <0.0001
Categorical EDSS Δ 2.71 0.06 0.002
Confirmed EDSS Δ 1.84 0.02 0.05
Multivariate relationships Slope SE p-value
Baseline variables
EDSS 1.20 0.25 <0.0001
T2 BOD (cm2) 0.05 0.02 <0.0001
Gender 0.93 0.47 0.001
On-RCT variables
Actual EDSS Δ 0.86 0.21 <0.0001
Annual relapse rate 0.52 0.23 0.025

R2=Percent of the variance accounted for by the relationship;

On-RCT=outcomes occurring during the course of the RCT;

EDSS=Expanded Disability Status Score;

BOD=Burden of Disease;

Actual EDSS Δ=EDSS at the end minus EDSS at start of RCT;

Categorical EDSS Δ=≥1 point EDSS change between trial end and start;

Confirmed EDSS Δ=1 point EDSS change sustained for 3 months;

Slope=Regression coefficient; SE=Standard error; Gender: Men=1; Women=0;

Long-term outcome=EDSS≥6 or conversion to SPMS.

In summary, these LTFU studies provide evidence for an association between clinical relapses and long-term disability outcomes. Nevertheless, because pre-study relapses were not so associated (Goodin et al., 2012b), because the “on-RCT” relapse-rates will potentially interact with any impact of therapy on long-term outcome (Fisher et al, 2002, Goodin et al, 2011, Goodin et al, 2012a, and Kappos et al, 2006), and because the natural history studies suggest that only early relapses make a difference (Confavreux et al, 2003, Confavreux and Vukusic, 2006, Kremenchutzky et al, 2006, Runmarker and Andersen, 1993, Tremlett et al, 2006, Tremlett et al, 2008, Weinshenker et al, 1989a, and Weinshenker et al, 1989b), the strength of the relationship between the relapses and long-term disability outcome is considered to be weak. For example, in the only LTFUs that addressed this topic (Fisher et al, 2002 and Goodin et al, 2012b), the relationship with relapses accounted for only about 12% of the variance in final disability status (Table 2).

6.4. Relationship of MRI lesions to clinical relapses

Following its introduction in the 1980s, MRI revolutionized the study of MS because of its ability to provide in vivo images of its pathology (Barkhof, 2002, Compston et al, 2006, and Goodin, 2006). With the subsequent introduction of Gd-enhanced MRI, the ability to detect a breakdown in the BBB associated with an MS attack and, thus, the ability to identify currently active lesions became considerably improved. Despite this, it has proven difficult to correlate these Gd-enhanced lesions with clinically-evident attacks (Petkau et al., 2008). There are at least four possible reasons for this. First, many Gd-enhancing lesions occur in the absence of clinically evident attacks (Petkau et al., 2008). Second, attacks may occur in CNS locations (e.g., spinal cord) that are not imaged as well as the brain by current MRI techniques. Nevertheless, there are some indications that spinal lesions are quite important (Okuda et al., 2011) and, ultimately, spinal cord imaging may prove superior to brain imaging with regard to its correlation with the clinical events of relapse and disability progression. Third, the MRI may not be acquired within the short time period of enhancement. And fourth, single-dose gadolinium may miss some truly active lesions. To circumvent concerns about the exact timing of the MRI, Sormani and colleagues studied the utility of new T2 lesions as a surrogate measure for clinical relapses (Sormani et al, 2009 and Sormani et al, 2011). At both the trial level and at the individual level, new T2 lesions were useful surrogates for clinical relapses. Indeed, at the individual level, the number of new T2 lesions accounted for over 60% of the treatment effect on relapses and seemed to be mediated through drug effects on new T2 lesions (Sormani et al., 2011). In this study, however, the number of new T2 lesions was about 3-times more common than clinical episodes and no attempt was made to relate the location of the new lesions with the clinical manifestations of the attacks (Sormani et al., 2011). Nevertheless, it seems clear that new T2 lesions can reliably predict the occurrence of clinical relapses and can be used as a valid biomarker for this future outcome. Despite this, there may be certain agents (either at present or in the future) that, despite having comparable efficacy, do not impact the BBB to the same extent as others. In this setting, the exclusive use of new T2 lesions as a biomarker in preliminary trials may fail to identify some effective therapeutic strategies.

6.5. Relationship of brain MRI lesions to concurrent disability (EDSS)

The correlation between the number of lesions or lesion burden on brain MRI and concurrent EDSS disability, although statistically significant, has been relatively poor. A review of this literature (Goodin, 2006) found correlations of concurrent disability with brain MRI ranging between ρ=0.15 and 0.67, with larger series tending to report correlations at the low end of this interval. In considering these low correlations, it is important to note that asymptomatic MRI lesions occur commonly in MS patients. Of all MRI lesions seen in MS patients, 90–95% appear to be asymptomatic, both from the doctor’s and the patient’s perspectives (Goodin, 2006). With such a high rate of asymptomatic lesions, low correlations are only to be expected. In fact, this is the likely basis of the so-called “clinic-radiological paradox” (Barkhof, 2002 and Goodin, 2006). Using a simple mathematical model, it can be demonstrated that if 90% of lesions are relatively silent, then the maximum possible correlation between lesion number and disability is only about 0.2–0.3, even when brain MRI lesions are assumed to be the sole determinant of disability (Goodin, 2006).

Nevertheless, the reason that brain MRI lesions are commonly asymptomatic is not entirely clear. One suggestion has been that large areas of the brain are either “non-eloquent” or “redundant” and, thus, unlikely to cause noticeable symptoms when injured. Such an explanation may be plausible in the brain where certain areas (e.g., the periventricular white matter) are associated with few defined functions, compared to other areas (e.g., the white matter of the internal capsule), which have well-defined functional importance. Such an explanation, however, is less compelling when it comes to clinically silent lesions in the spinal cord (Silver et al., 2001) where most of the fiber pathways have defined functions and, thus, such lesions should produce symptoms.

There is, however, another possible explanation than location alone, which is frequently posited in the literature regarding this paradox (Barkhof, 2002). In this view, part of the reason for the poor correlation is that the T2 lesions, which demonstrate a wide range of hyperintensity, are non-specific with respect to their histopathological correlates. By contrast, the brightest T2 lesions, which persist beyond the acute stage, can often also be detected as dark regions on the less sensitive, spin-echo T1-weighted images (i.e., the so-called “black holes”). These black holes are thought to indicate both matrix destruction and axonal loss and consequently to correlate better with clinical disability (Barkhof, 2002). Such a possibility is underscored by the marked histopathological diversity of MS lesions discussed earlier (Fig. 1). Although variations in the eloquence or redundancy of different CNS regions may play some role in producing this diversity, they cannot be the only (or maybe even the dominant) explanation.

6.6. Relationship of brain MRI lesions to future disability (EDSS)

Several longitudinal studies have demonstrated that both the total disease burden and the number of brain MRI lesions at baseline are significantly correlated with future EDSS disability many years later (Fisniku et al, 2008, Popescu et al, 2013, and Tintore et al, 2006). Thus, in the long-term follow-up of patients with clinically isolated syndromes (CIS) from Queen Square, the authors reported a strong relationship between an increasing baseline number and volume of lesions and an increased likelihood of reaching the endpoints of both EDSS>3 and EDSS≥6 after 20 years (Fisniku et al., 2008). Patients who were ultimately classified as having SPMS showed an increasing accumulation of excessive disease burden over time, in comparison to patients classified as RRMS or as CIS (Fig. 6B).

Fig. 6.

Fig. 6 Relationship of baseline lesion number and volume to long-term disability outcome after 20 years in a cohort of CIS patients. Panel (A) shows the increasing likelihood of reaching (EDSS>3) and (EDSS≥6) based on an increased number of lesions at the clinical onset of disease. Panel (B) shows the increasing baseline burden of disease in those patients who were ultimately (at long term follow-up) classified as SPMS, RRMS, and CIS. At every time point, the order of disease burden was the same (SPMS>RRMS>CIS), the difference between groups was increasing, and the volume was significantly correlated with the EDSS score at 20 years (Fisniku et al., 2008).Panel (B) reprinted with permission from Fisniku et al. (2008).CIS=clinically isolated syndromes; EDSS=expanded disability status scale; SPMS=secondary progressive multiple sclerosis; RRMS=relapsing-remitting multiple sclerosis.

Similarly, in the LTFU of the IFNβ-1b trial (Goodin et al, 2011, Goodin et al, 2012a, and Goodin et al, 2012b), the baseline MRI burden of disease strongly correlated both with long-term disability and with mortality, which is, itself, a reflection of death due to advanced MS (Goodin et al., 2012a). Despite this strong association with baseline disease burden, however, the number of new lesions occurring during the course of the RCT was not significantly correlated with long-term disability status (Table 2). Perhaps, this dissociation is a reflection of the fact that new lesions over a short time interval have a wide variation in both pathological characteristics and functional consequences and the fact that the “on-RCT” time period is typically very short relative to the long-term outcomes of interest. The baseline MRI, by contrast, estimates the total amount of tissue injury that has accumulated over the entire disease course up to that point – a time-span that, in most cases, includes a period of many years before the clinical onset (Okuda et al., 2009).

Somewhat counter to this observation is the study of Bermel et al. (2013) on the predictive value of MRI in the LTFU of the weekly IFNβ-1a trial. Among patients on active treatment during the RCT, the enhancing MRI lesions and clinical relapses were associated with an increased disability (as measured by the ΔEDSS). By contrast, in the placebo group, none of these odds ratios were significantly different from unity. Nevertheless, the confidence intervals were considerably overlapped between the placebo and treated groups, so that interpretation of the findings is uncertain. Regardless, this study raises (but does not answer) the important question of whether those clinical or MRI events that occur while a person is on a DMT, might be more concerning (as a potential early indicator of suboptimal response to therapy), compared to those same events that occur while a person is untreated.

Other studies (not following RCTs) have also examined the relationship between MRI lesions and future outcome. For example, Tomassini and coworkers followed 68 IFNβ-treated patients for a mean of 5.9 years (Tomassini et al., 2006). Disease “worsening” in this study was defined as having a 6-month “confirmed” 1-point EDSS change at any time between the first and sixth year of follow-up. The authors reported that the MRI disease burden (on both T1- and T2-weighted images), measured after 1 year of therapy, was associated with a greater likelihood of having a “worsened” status, although this finding was fully accounted for by differences in disease burden at baseline (Tomassini et al., 2006). In another study, Prosperini et al. (2009) followed 394 IFNβ-treated patients for a mean of 4.8 years. Disease progression in this study was defined as having a 6-month “confirmed” 1-point EDSS change (compared to baseline) at any time during the follow-up interval. This study reported a significant relationship between MRI activity (either Gd-enhancement or one or more new T2 lesions) during the first year of therapy and the likelihood of disease progression (Prosperini et al., 2009). Nevertheless, it is noteworthy that both of these studies used a measure of long-term disability that has been previously demonstrated to be unreliable, even in the short-term (Table 1 and Table 2).

In summary, the available studies provide convincing evidence for an association of both brain lesion number (at clinical onset) and intracranial disease burden (either at disease onset or later in the disease course) with long-term disability outcome in MS (Fig. 6A and B). By contrast, there is not convincing evidence for a relationship between the occurrence of new brain MRI activity (measured over the short-term) and the long-term consequences of the disease. Whether the inclusion of spinal cord imaging will improve matters in the future remains to be determined.

7. Summary

There are several reasons to conclude that relapses play some role in the final disability status that many (most) MS patients experience. First, conclusive evidence indicates that clinical MS attacks can cause irreparable neurological damage in some MS patients. In some, this damage can be quite disabling (Keegan et al., 2002). Second, clinical MS attacks, especially those occurring early in the course of the disease and those occurring over the short 2- to 3-year interval of an RCT, are significantly and consistently correlated with long-term outcome (Fisher et al, 2002, Goodin et al, 2011, Goodin et al, 2012a, Kappos et al, 2006, and Prosperini et al, 2009). And, third, final disability outcome is unequivocally associated with disease burden present at baseline (Bermel et al, 2013, Fisniku et al, 2008, Goodin et al, 2011, Goodin et al, 2012b, Popescu et al, 2013, and Tintore et al, 2006), which likely reflects the accumulated amount of tissue damage from relapses (both clinical and subclinical) over the entire course of the disease up to that point.

The data are less convincing, however, regarding the extent to which relapses contribute to disability, especially later in the disease course. Thus, the contribution of relapses to the variance of the long-term disability during an RCT, while statistically significant, is unimpressive in terms of its magnitude (i.e., 12%). In addition, clinical relapses seem to have greater predictive value for long-term disability when they occur early in the relapsing phase of the illness compared to when they occur later (Confavreux et al, 2003, Confavreux and Vukusic, 2006, Kremenchutzky et al, 2006, Runmarker and Andersen, 1993, Tremlett et al, 2006, Tremlett et al, 2008, Weinshenker et al, 1989a, and Weinshenker et al, 1989b).

Nevertheless, this still leaves open the question of whether individual MS attacks should be treated. Certainly, if a treatment can sometimes improve the final recovery from an episode of MS (which plasma exchange seems to do for very severe attacks) and if therapy is well tolerated (which it generally is for short-term courses), then, at least, some and possibly most MS attacks should be treated. Even in the circumstance where treatment is thought merely to speed the recovery but not to alter either the attack’s ultimate resolution or the person’s long-term outcome, a safe treatment would still seem to be warranted in many cases. Perhaps it makes sense to be more aggressive in the treatment of relapses early in the disease course or in the treatment of severe relapses. However, the relationship between treatment of such attacks and long-term outcome has never actually been assessed.

Consequently, we conclude that most patients who experience MS attacks should be offered treatment with glucocorticoid or glucocorticoid-releasing agents unless there are reasons why such therapy is either unnecessary or might not be well tolerated. Nevertheless, for individual patients, the choice of which agent (ACTH or one of the synthetic glucocorticoids), which dose and regimen, and which route to use (oral, IV, subcutaneous, or intramuscular) will depend upon a combination of factors including comparative efficacy (proven or perceived), individual tolerability, patient preferences, and cost. In the event that a patient experiences a severe MS attack that has not responded to initial glucocorticoid or ACTH therapy, PE therapy should be offered/considered if it is available.

Conflict of Interest

Douglas S. Goodin, M.D. has no known conflicts of interest associated with this publication.

Anthony T. Reder, M.D. has received consulting and travel reimbursements from Questcor and from other companies developing therapies for MS, including Acorda, Bayer, Biogen Idec, Genentech, Genzyme, Merck Serono, Novartis, and Teva.

Robert A. Bermel, M.D. has received personal compensation for consulting work with Biogen Idec, Novartis, Genzyme, Teva, and Questcor and receives or has received research support (through his institution) from Biogen Idec and Novartis.

Gary R. Cutter, Ph.D. is a member of a data and safety monitoring board for Apotek, Biogen Idec, Cleveland Clinic (Vivus), GlaxoSmithKlein Pharmaceuticals, Gilead Pharmaceuticals, Modigenetech/Prolor, Merck/Ono Pharmaceuticals, Merck, Merck/Pfizer, Neuren, Sanofi-Aventis, Teva, Washington University, NHLBI (Protocol Review Committee), NINDS, and NICHD (OPRU oversight committee). He is a consultant for or attended advisory boards for Consortium of MS Centers (grant), D3 (Drug Discovery and Development), Genzyme, Jannsen Pharmaceuticals, Klein-Buendel Incorporated, Medimmune, Novartis, Opexa Therapeutics, Receptos, Roche, EMD Serono, Teva Pharmaceuticals, and Transparency Life Sciences. Dr. Cutter is employed by the University of Alabama at Birmingham and is President of Pythagoras, Inc. a private consulting company located in Birmingham, AL.

Robert J. Fox, M.D., F.A.A.N. has received personal consulting fees from Biogen Idec, GlaxoSmithKline, MedDay, Novartis, Questcor, Teva, and XenoPort; has served on advisory committees for Biogen Idec and Novartis; received research grant funding from Novartis; and serves on the editorial boards for Neurology and Multiple Sclerosis Journal.

Gareth R. John, Ph.D. has no known conflicts of interest associated with this publication.

Fred D. Lublin, M.D., F.A.A.N., F.A.N.A. has received research funding from Acorda Therapeutics, Inc., Biogen Idec, Novartis Pharmaceuticals Corp., Teva Neuroscience, Inc., Genzyme, Sanofi, Celgene, Transparency Life Sciences, NIH, and NMSS Consulting. He has agreements with, attended advisory boards for, or is a member of a data safety monitoring board for Bayer HealthCare Pharmaceuticals, Biogen Idec, EMD Serono, Inc., Novartis, Teva Neuroscience, Actelion, Sanofi-Aventis, Acorda, Questcor, Roche, Genentech, Celgene, Johnson & Johnson, Revalesio, Coronado Bioscience, Genzyme, MedImmune, Bristol-Myers Squibb, Xenoport, Receptos, Forward Pharma, and to-BBB technologies. Dr. Lublin is a speaker (non-promotional) for Genzyme; the Co-Chief Editor for Multiple Sclerosis and Related Diseases; and has financial interests/stock ownership in Cognition Pharmaceuticals, Inc.

Claudia F. Lucchinetti, M.D. has received grant support from Novartis, Alexion, and Biogen.

Aaron E. Miller, M.D. has received research support from Acorda, Novartis, Genentech, Genzyme, Sanofi-Aventis, Biogen Idec, Roche, and Questcor. He is a consultant for Genzyme/Sanofi-Aventis, Biogen Idec, GlaxoSmithKline, EMD Serono (Merck Serono); Novartis, Acorda, Accordant Health Services, Teva, and Roche.

Daniel Pelletier, M.D. has received research grants from Biogen Idec, Genzyme, Hoffman-LaRoche, the National Multiple Sclerosis Society, and the NIH/NINDS. He has received consulting honoraria from Genzyme, Biogen Idec, Genentech, Novartis, Acorda, Vaccinex, and Questcor.

Michael K. Racke, M.D. is a consultant for Diogenix, Biogen Idec, Teva Neuroscience, Roche/Genentech, and Novartis.

Bruce D. Trapp, Ph.D. has no known conflicts of interest associated with this publication.

Timothy Vartanian, M.D., Ph.D. is a member of a speakers bureau for Questcor and Teva and is a consultant for EMD-Serono and Genzyme.

Emmanuelle Waubant, M.D., Ph.D. has no known conflicts of interest associated with this publication.

Role of the funding source

Editorial assistance was provided by MedVal Scientific Information Services, LLC, which was supported by Autoimmune and Rare Diseases Business (formerly Questcor), Mallinckrodt Pharmaceuticals.


The MS Summit 2013 was held in June 2013 in Fort Lauderdale, Florida and was supported by Autoimmune and Rare Diseases Business (formerly Questcor), Mallinckrodt Pharmaceuticals.

Members of the expert committee included Robert A. Bermel, M.D., Cleveland, OH; Gary R. Cutter, Ph.D., Birmingham, AL; Robert J. Fox, M.D., F.A.A.N., Cleveland, OH; Douglas S. Goodin, M.D., San Francisco, CA; David A. Hafler, M.D., M.Sc., New Haven, CT; Gareth R. John, Ph.D., New York, NY; Fred D. Lublin, M.D., F.A.A.N., F.A.N.A., New York, NY; Claudia F. Lucchinetti, M.D., Rochester, MN; Aaron Miller, M.D., New York, NY; Daniel Pelletier, M.D., New Haven, CT; Michael K. Racke, M.D., Columbus, OH; Anthony T. Reder, M.D., Chicago, IL; Bruce D. Trapp, Ph.D., Cleveland, OH; Timothy Vartanian, M.D., Ph.D., New York, NY; Emmanuelle Waubant, M.D., Ph.D., San Francisco, CA


  • Andersen et al., 1993 O. Andersen, P.E. Lygner, T. Bergstrom, et al. Viral infections trigger multiple sclerosis relapses: a prospective seroepidemiological study. J. Neurol.. 1993;240:417-422 Crossref
  • Arnason et al., 2013 B.G. Arnason, R. Berkovich, A. Catania, et al. Mechanisms of action of adrenocorticotropic hormone and other melanocortins relevant to the clinical management of patients with multiple sclerosis. Mult. Scler. J.. 2013;19:130-136 Crossref
  • Bar-Or et al., 1999 A. Bar-Or, E.M. Oliveira, D.E. Anderson, et al. Molecular pathogenesis of multiple sclerosis. J. Neuroimmunol.. 1999;100:252-259 Crossref
  • Barkhof, 2002 F. Barkhof. The clinico-radiological paradox in multiple sclerosis revisited. Curr. Opin. Neurol.. 2002;15:239-245 Crossref
  • Beck et al., 1992 R.W. Beck, P.A. Cleary, M.M. Anderson Jr., et al. A randomized, controlled trial of corticosteroids in the treatment of acute optic neuritis. N. Engl. J. Med.. 1992;326:581-588 Crossref
  • Bermel et al., 2010 R.A. Bermel, B. Weinstock-Guttman, D. Bourdette, et al. Intramuscular interferon beta-1a therapy in patients with relapsing-remitting multiple sclerosis: a 15-year follow-up study. Mult. Scler.. 2010;16:588-596 Crossref
  • Bermel et al., 2013 R.A. Bermel, X. You, P. Foulds, et al. Predictors of long-term outcome in multiple sclerosis patients treated with interferon beta. Ann. Neurol.. 2013;73:95-103 Crossref
  • Bo et al., 2003 L. Bo, C.A. Vedeler, H.I. Nyland, et al. Subpial demyelination in the cerebral cortex of multiple sclerosis patients. J. Neuropathol. Exp. Neurol.. 2003;62:723-732
  • Catania et al., 2010 A. Catania, C. Lonati, A. Sordi, et al. The melanocortin system in control of inflammation. Sci. World J.. 2010;10:1840-1853 Crossref
  • Cohen et al., 1993 R.A. Cohen, H.R. Kessler, M. Fischer. The Extended Disability Status Scale (EDSS) as a predictor of impairments of functional activities of daily living in multiple sclerosis. J. Neurol. Sci.. 1993;115:132-135 Crossref
  • Compston et al., 2006 A. Compston, C. Confavreux, H. Lassmann. McAlpine’s Multiple Sclerosis. (Churchill Livingston, London, 2006)
  • Confavreux et al., 1998 C. Confavreux, M. Hutchinson, M.M. Hours, et al. Rate of pregnancy-related relapse in multiple sclerosis. Pregnancy in Multiple Sclerosis Group. N. Engl. J. Med.. 1998;339:285-291 Crossref
  • Confavreux and Vukusic, 2006 C. Confavreux, S. Vukusic. Natural history of multiple sclerosis: a unifying concept. Brain. 2006;129(Part 3):606-616 Crossref
  • Confavreux et al., 2003 C. Confavreux, S. Vukusic, P. Adeleine. Early clinical predictors and progression of irreversible disability in multiple sclerosis: an amnesic process. Brain. 2003;126:770-782 Crossref
  • Conlon et al., 1999 P. Conlon, J.R. Oksenberg, J. Zhang, et al. The immunobiology of multiple sclerosis: an autoimmune disease of the central nervous system. Neurobiol. Dis.. 1999;6:149-166 Crossref
  • Cook et al., 2012 S.D. Cook, S. Dhib-Jalbut, P. Dowling, et al. Use of magnetic resonance imaging as well as clinical disease activity in the clinical classification of multiple sclerosis and assessment of its course: a report from an International CMSC Consensus Conference. Int. J. MS Care.. 2012;14:105-114 Crossref
  • Edwards et al., 1998 S. Edwards, M. Zvartau, H. Clarke, et al. Clinical relapses and disease activity on magnetic resonance imaging associated with viral upper respiratory tract infections in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry. 1998;64:736-741 Crossref
  • Engelhardt et al., 2014 S. Engelhardt, S. Patkar, O.O. Ogunshola. Cell-specific blood–brain barrier regulation in health and disease: a focus on hypoxia. Br. J. Pharmacol.. 2014;171:1210-1230 Crossref
  • Ferguson et al., 1997 B. Ferguson, M.K. Matyszak, M.M. Esiri, et al. Axonal damage in acute multiple sclerosis lesions. Brain. 1997;120(Part 3):393-399 Crossref
  • Filippi et al., 1998 M. Filippi, M.A. Rocca, G. Martino, et al. Magnetization transfer changes in the normal appearing white matter precede the appearance of enhancing lesions in patients with multiple sclerosis. Ann. Neurol.. 1998;43:809-814 Crossref
  • Fisher et al., 2007 E. Fisher, A. Chang, R.J. Fox, et al. Imaging correlates of axonal swelling in chronic multiple sclerosis brains. Ann. Neurol.. 2007;62:219-228 Crossref
  • Fisher et al., 2002 E. Fisher, R.A. Rudick, J.H. Simon, et al. Eight-year follow-up study of brain atrophy in patients with MS. Neurology. 2002;59:1412-1420 Crossref
  • Fisniku et al., 2008 L.K. Fisniku, P.A. Brex, D.R. Altmann, et al. Disability and T2 MRI lesions: a 20-year follow-up of patients with relapse onset of multiple sclerosis. Brain. 2008;131(Part 3):808-817 Crossref
  • Ford et al., 2010 C. Ford, A.D. Goodman, K. Johnson, et al. Continuous long-term immunomodulatory therapy in relapsing multiple sclerosis: results from the 15-year analysis of the US prospective open-label study of glatiramer acetate. Mult. Scler.. 2010;16:342-350 Crossref
  • Goodin, 2006 D.S. Goodin. Magnetic resonance imaging as a surrogate outcome measure of disability in multiple sclerosis: have we been overly harsh in our assessment?. Ann. Neurol.. 2006;59:597-605 Crossref
  • Goodin et al., 2011 D.S. Goodin, J. Jones, D. Li, et al. Establishing long-term efficacy in chronic disease: use of recursive partitioning and propensity score adjustment to estimate outcome in MS. PLoS One. 2011;6:e22444
  • Goodin et al., 2012a D.S. Goodin, A.T. Reder, G.C. Ebers, et al. Survival in MS: a randomized cohort study 21 years after the start of the pivotal IFNβ-1b trial. Neurology. 2012;78:1315-1322 Crossref
  • Goodin et al., 2012b D.S. Goodin, A. Traboulsee, V. Knappertz, et al. Relationship between early clinical characteristics and long term disability outcomes: 16 year cohort study (follow-up) of the pivotal interferon β-1b trial in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry. 2012;83:282-287 Crossref
  • Goodkin et al., 1998 D.E. Goodkin, W.D. Rooney, R. Sloan, et al. A serial study of new MS lesions and the white matter from which they arise. Neurology. 1998;51:1689-1697 Crossref
  • Hauser and Oksenberg, 2006 S.L. Hauser, J.R. Oksenberg. The neurobiology of multiple sclerosis: genes, inflammation, and neurodegeneration. Neuron. 2006;52:61-76 Crossref
  • Holman et al., 2011 D.W. Holman, R.S. Klein, R.M. Ransohoff. The blood–brain barrier, chemokines and multiple sclerosis. Biochim. Biophys. Acta. 2011;1812:220-230 Crossref
  • Hussain et al., 2014 S.A. Hussain, S. Shinnar, G. Kwong, et al. Treatment of infantile spasms with very high dose prednisolone before high dose adrenocorticotropic hormone. Epilepsia. 2014;55:103-107 Crossref
  • Kappos et al., 2006 L. Kappos, A. Traboulsee, C. Constantinescu, et al. Long-term subcutaneous interferon beta-1a therapy in patients with relapsing-remitting MS. Neurology. 2006;67:944-953 Crossref
  • Keegan et al., 2002 M. Keegan, A.A. Pineda, R.L. McClelland, et al. Plasma exchange for severe attacks of CNS demyelination: predictors of response. Neurology. 2002;58:143-146 Crossref
  • Kremenchutzky et al., 2006 M. Kremenchutzky, G.P. Rice, J. Baskerville, et al. The natural history of multiple sclerosis: a geographically based study 9: observations on the progressive phase of the disease. Brain. 2006;129:584-594 Crossref
  • Kurtzke, 1955 J.F. Kurtzke. A new scale for evaluating disability in multiple sclerosis. Neurology. 1955;5:580-583
  • Kurtzke, 1983 J.F. Kurtzke. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983;33:1444-1452
  • Liu and Blumhardt, 2000 C. Liu, L.D. Blumhardt. Disability outcome measures in therapeutic trials of relapsing-remitting multiple sclerosis: effects of heterogeneity of disease course in placebo cohorts. J. Neurol., Neurosurg. Psychiatry. 2000;68:450-457 Crossref
  • Lublin et al., 2012 LublinF., Cofield, S., Gustafson, T., et al., 2012. Assessment of relapse activity in the CombiRx randomised clinical trial. Presented at: 28th Congress of the European Committee for Treatment and Research in Multiple Sclerosis, October 10–13, 2012, Lyon, France.
  • Lublin et al., 2003 F.D. Lublin, M. Baier, G. Cutter. Effect of relapses on development of residual deficit in multiple sclerosis. Neurology. 2003;61:1528-1532 Crossref
  • Lublin and Reingold, 1996 F.D. Lublin, S.C. Reingold. Defining the clinical course of multiple sclerosis: results of an international survey. National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology. 1996;46:907-911 Crossref
  • Lucchinetti et al., 2011 C.F. Lucchinetti, B.F. Popescu, R.F. Bunyan, et al. Inflammatory cortical demyelination in early multiple sclerosis. N. Engl. J. Med. 2011;365:2188-2197 Crossref
  • Moll et al., 2009 N.M. Moll, M.B. Cossoy, E. Fisher, et al. Imaging correlates of leukocyte accumulation and CXCR4/CXCL12 in multiple sclerosis. Arch. Neurol.. 2009;66:44-53
  • Muoio et al., 2014 V. Muoio, P.B. Persson, M.M. Sendeski. The neurovascular unit-concept review. Acta Physiol.. 2014;210:790-798 Crossref
  • Novotna et al., 2015 M. Novotna, M.M. Paz Soldan, Z.N. Abou, et al. Poor early relapse recovery affects onset of progressive disease course in multiple sclerosis. Neurology. 2015; Jul 24 [Epub ahead of print]
  • Okuda et al., 2009 D.T. Okuda, E.M. Mowry, A. Beheshtian, et al. Incidental MRI anomalies suggestive of multiple sclerosis: the radiologically isolated syndrome. Neurology. 2009;72:800-805 Crossref
  • Okuda et al., 2011 D.T. Okuda, E.M. Mowry, B.A. Cree, et al. Asymptomatic spinal cord lesions predict disease progression in radiologically isolated syndrome. Neurology. 2011;76:686-692 Crossref
  • Panitch, 1994 H.S. Panitch. Influence of infection on exacerbations of multiple sclerosis. Ann. Neurol.. 1994;36(Suppl.):S25-S28 Crossref
  • Peterson et al., 2001 J.W. Peterson, L. Bo, S. Mork, et al. Transected neurites, apoptotic neurons, and reduced inflammation in cortical multiple sclerosis lesions. Ann. Neurol.. 2001;50:389-400 Crossref
  • Petkau et al., 2008 J. Petkau, S.C. Reingold, U. Held, et al. Magnetic resonance imaging as a surrogate outcome for multiple sclerosis relapses. Mult. Scler.. 2008;14:770-778 Crossref
  • Polman et al., 2011 C.H. Polman, S.C. Reingold, B. Banwell, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald Criteria. Ann. Neurol.. 2011;69:292-302 Crossref
  • Popescu et al., 2013 V. Popescu, F. Agosta, H.E. Hulst, et al. Brain atrophy and lesion load predict long term disability in multiple sclerosis. J. Neurol. Neurosurg. Psychiatry. 2013;84:1082-1091 Crossref
  • Pozzilli et al., 2005 C. Pozzilli, L. Prosperini, E. Sbardella, et al. Post-marketing survey on clinical response to interferon beta in relapsing multiple sclerosis: the Roman experience. Neurol. Sci.. 2005;26(Suppl. 4):S174-S178 Crossref
  • Prentice, 1989 R.L. Prentice. Surrogate endpoints in clinical trials: definition and operational criteria. Stat. Med.. 1989;8:431-440 Crossref
  • Prosperini et al., 2009 L. Prosperini, V. Gallo, N. Petsas, et al. One-year MRI scan predicts clinical response to interferon beta in multiple sclerosis. Eur. J. Neurol.. 2009;16:1202-1209 Crossref
  • Rose et al., 1970 A.S. Rose, J.W. Kuzma, J.F. Kurtzke, et al. Cooperative study in the evaluation of therapy in multiple sclerosis. ACTH vs. placebo—final report. Neurology. 1970;20:1-59
  • Runmarker and Andersen, 1993 B. Runmarker, O. Andersen. Prognostic factors in a multiple sclerosis incidence cohort with twenty-five years of follow-up. Brain. 1993;116(Part 1):117-134
  • Salemi et al., 2004 G. Salemi, G. Callari, M. Gammino, et al. The relapse rate of multiple sclerosis changes during pregnancy: a cohort study. Acta Neurol. Scand.. 2004;110:23-26 Crossref
  • Scalfari et al., 2010 A. Scalfari, A. Neuhaus, A. Degenhardt, et al. The natural history of multiple sclerosis: a geographically based study 10: relapses and long-term disability. Brain. 2010;133:1914-1929 Crossref
  • Schumacker et al., 1965 G.A. Schumacker, G. Beebe, R.F. Kibler, et al. Problems of experimental trials of therapy in multiple sclerosis: report by the panel on the evaluation of experimental trials of therapy in multiple sclerosis. Ann. N. Y. Acad. Sci.. 1965;122:552-568
  • Sibley et al., 1985 W.A. Sibley, C.R. Bamford, K. Clark. Clinical viral infections and multiple sclerosis. Lancet. 1985;325:1313-1315 Crossref
  • Silver et al., 2001 N.C. Silver, C.D. Good, M.P. Sormani, et al. A modified protocol to improve the detection of enhancing brain and spinal cord lesions in multiple sclerosis. J. Neurol.. 2001;248:215-224 Crossref
  • Snead et al., 1983 O.C. Snead III, J.W. Benton, G.J. Myers. ACTH and prednisone in childhood seizure disorders. Neurology. 1983;33:966-970
  • Sormani et al., 2009 M.P. Sormani, L. Bonzano, L. Roccatagliata, et al. Magnetic resonance imaging as a potential surrogate for relapses in multiple sclerosis: a meta-analytic approach. Ann. Neurol.. 2009;65:268-275 Crossref
  • Sormani et al., 2011 M.P. Sormani, B. Stubinski, P. Cornelisse, et al. Magnetic resonance active lesions as individual-level surrogate for relapses in multiple sclerosis. Mult. Scler.. 2011;17:541-549 Crossref
  • The Optic Neuritis Study Group, 2008 The Optic Neuritis Study Group. Visual function 15 years after optic neuritis: a final follow-up report from the Optic Neuritis Treatment Trial. Ophthalmology. 2008;115:1079-1082
  • Tintore et al., 2006 M. Tintore, A. Rovira, J. Rio, et al. Baseline MRI predicts future attacks and disability in clinically isolated syndromes. Neurology. 2006;67:968-972 Crossref
  • Tomassini et al., 2006 V. Tomassini, A. Paolillo, P. Russo, et al. Predictors of long-term clinical response to interferon beta therapy in relapsing multiple sclerosis. J. Neurol.. 2006;253:287-293 Crossref
  • Trapp et al., 1998 B.D. Trapp, J. Peterson, R.M. Ransohoff, et al. Axonal transection in the lesions of multiple sclerosis. N. Engl. J. Med.. 1998;338:278-285 Crossref
  • Tremlett et al., 2006 H. Tremlett, D. Paty, V. Devonshire. Disability progression in multiple sclerosis is slower than previously reported. Neurology. 2006;66:172-177 Crossref
  • Tremlett et al., 2008 H. Tremlett, Z. Yinshan, V. Devonshire. Natural history of secondary-progressive multiple sclerosis. Mult. Scler.. 2008;14:314-324 Crossref
  • Vukusic et al., 2004 S. Vukusic, M. Hutchinson, M. Hours, et al. Pregnancy and multiple sclerosis (the PRIMS study): clinical predictors of post-partum relapse. Brain. 2004;127(Part 6):1353-1360 Crossref
  • Weinshenker et al., 1989a B.G. Weinshenker, B. Bass, G.P. Rice, et al. The natural history of multiple sclerosis: a geographically based study. 2. Predictive value of the early clinical course. Brain. 1989;112:1419-1428
  • Weinshenker et al., 1989b B.G. Weinshenker, B. Bass, G.P. Rice, et al. The natural history of multiple sclerosis: a geographically based study. I. Clinical course and disability. Brain. 1989;112:133-146
  • Yamasaki et al., 2014 R. Yamasaki, H. Lu, O. Butovsky, et al. Differential roles of microglia and monocytes in the inflamed central nervous system. J. Exp. Med.. 2014;211:1533-1549 Crossref


a Multiple Sclerosis Center, University of California, San Francisco Medical Center, San Francisco, CA, United States

b Department of Neurology, University of California, San Francisco School of Medicine, San Francisco, CA, United States

c Department of Neurology, The University of Chicago, Chicago, IL, United States

d Mellen Center for Multiple Sclerosis, Cleveland Clinic, Cleveland, OH, United States

e Department of Biostatistics, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, United States

f Mellen Center for Multiple Sclerosis, Neurological Institute, Cleveland Clinic, Cleveland, OH, United States

g Multiple Sclerosis Research Laboratory, Corinne Goldsmith Dickinson Center for Multiple Sclerosis, Friedman Brain Institute, New York, NY, United States

h Department of Neurology, Mount Sinai School of Medicine, New York, NY, United States

i Corinne Goldsmith Dickinson Center for Multiple Sclerosis, Icahn School of Medicine at Mount Sinai, New York, NY, United States

j Mayo Clinic College of Medicine, Rochester, MN, United States

k Neuro-Immunology Division and Yale Multiple Sclerosis Center, Advanced Imaging in Multiple Sclerosis (AIMS) Laboratory, Yale University School of Medicine, New Haven, CT, United States

l Department of Neurology, Wexner Medical Center at The Ohio State University, Columbus, OH, United States

m Department of Neurosciences, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States

n Judith Jaffe Multiple Sclerosis Center, New York-Presbyterian Hospital/Weill Cornell Medical Center, Weill Cornell Medical College, United States

o UCSF Regional Pediatric MS Center, Race to Erase MS, San Francisco, CA, United States

Correspondence to: UCSF MS Center, Department of Neurology, University of California, 675 Nelson Rising Lane, Suite #221D, San Francisco, CA 94158, United States. Fax: +1 415 514 2470.

Search this site

Stay up-to-date with our monthly e-alert

If you want to regularly receive information on what is happening in MS research sign up to our e-alert.

Subscribe »

About the Editors

  • Prof Timothy Vartanian

    Timothy Vartanian, Professor at the Brain and Mind Research Institute and the Department of Neurology, Weill Cornell Medical College, Cornell...
  • Dr Claire S. Riley

    Claire S. Riley, MD is an assistant attending neurologist and assistant professor of neurology in the Neurological Institute, Columbia University,...
  • Dr Rebecca Farber

    Rebecca Farber, MD is an attending neurologist and assistant professor of neurology at the Neurological Institute, Columbia University, in New...

This online Resource Centre has been made possible by a donation from EMD Serono, Inc., a business of Merck KGaA, Darmstadt, Germany.

Note that EMD Serono, Inc., has no editorial control or influence over the content of this Resource Centre. The Resource Centre and all content therein are subject to an independent editorial review.

The Grant for Multiple Sclerosis Innovation
supports promising translational research projects by academic researchers to improve understanding of multiple sclerosis (MS) for the ultimate benefit of patients.  For full information and application details, please click here

Journal Editor's choice

Recommended by Prof. Brenda Banwell

Causes of death among persons with multiple sclerosis

Gary R. Cutter, Jeffrey Zimmerman, Amber R. Salter, et al.

Multiple Sclerosis and Related Disorders, September 2015, Vol 4 Issue 5