Sage Journals: Discover world-class research

Abstract

Magnetic resonance imaging (MRI) of the brain and spinal cord plays a crucial role in the diagnosis and monitoring of multiple sclerosis (MS). There is conclusive evidence that brain and spinal cord MRI findings in early disease stages also provide relevant insight into individual prognosis. This includes prediction of disease activity and disease progression, the accumulation of long-term disability and the conversion to secondary progressive MS. The extent to which these MRI findings should influence treatment decisions remains a subject of ongoing discussion. The aim of this review is to present and discuss the current knowledge and scientific evidence regarding the utility of MRI at early MS disease stages for prognostic classification of individual patients. In addition, we discuss the current evidence regarding the use of MRI in order to predict treatment response. Finally, we propose a potential approach as to how MRI data may be categorized and integrated into early clinical decision making.

Plain language summary

Can MRI help select appropriate therapy for recently diagnosed multiple sclerosis?

MS is a chronic autoimmune disease of the brain and spinal cord that causes physical and cognitive disability. Initially, most people with MS (pwMS) experience attacks of new symptoms and periods of partial recovery; this is called relapsing-remitting MS (RRMS). RRMS transitions to secondary progressive MS (SPMS), where there is a gradual worsening of disability. MS medications dampen parts of the immune system. They reduce the risk of relapses and delay transition to SPMS if started early. Once a person has SPMS, treatment can slow but not stop further deterioration. MS medications vary in their effects on the immune system, level of efficacy, and treatment risks. The course of MS is highly individual. When starting therapy, it can therefore be difficult to decide whether a drug with lower or higher efficacy is required. Some of the acute and chronic inflammatory changes in MS are shown as focal lesions (‘spots’) on MRI of the brain and spinal cord. They are very useful for diagnosing MS and determining disease activity. Even if there are no relapses, new lesions indicate that a MS medication is not fully effective. In addition, MRI provides a snapshot of tissue damage that has accumulated up to the examination. At the time of diagnosis, MRI reflects the natural history of MS in the individual, even before the first attack, and contains prognostic information. We review studies that investigate an association between certain MRI findings obtained early after the initial attack and the later course of MS. We propose that these metrics can be applied to a concept of grading and staging of MS as well as estimating functional reserve. We review thresholds that identify pwMS at risk of disability progression and transition to SPMS, who should be recommended highly effective therapy first line. Leveraging the prognostic capabilities of MRI may support initial treatment decisions.

Keywords

brain MRI multiple sclerosis prognosis spinal cord

Introduction

Multiple sclerosis (MS) is the most frequent chronic inflammatory demyelinating disease of the central nervous systems in young adults and can lead to long-term disability.¹ In addition to the clinical presentation and liquid biomarkers, magnetic resonance imaging (MRI) of the brain and spinal cord plays an important role in evaluating MS patients. This includes the establishment of an early and accurate diagnosis according to the 2017 revisions of the McDonald criteria as well as monitoring of treatment effectiveness and safety.^2–4 Expert panel guidelines such as the MAGNIMS-CMSC-NAIMS recommendations clearly established the role of MRI for these purposes and stressed the importance of standardized image acquisition as well as the role of gadolinium use.^5–9 In addition to diagnostic and treatment monitoring purposes, there is conclusive evidence that brain and spinal cord imaging findings contain important information regarding the prognosis of individual patients, including disease progression and the prediction of treatment response.^10,11 It has been demonstrated that certain imaging findings in the brain and spinal cord obtained early in the course of the disease, particularly after the first clinical demyelinating attack, provide important prognostic information regarding the conversion to clinically definite MS (CDMS), long-term disability, disease progression, disability progression and the conversion from relapsing-remitting MS (RRMS) to secondary progressive MS (SPMS).^10,12 This suggests that imaging findings may be of important clinical relevance in terms of treatment decision making at the time of diagnosis. However, whether and to what extent MRI findings during the first clinical presentation might aid in terms of treatment choice in clinical practice is still under debate.

The aim of this review is to present and discuss the current knowledge and scientific evidence regarding the role of MRI during the early stages of MS for prognostic classification of patients, its potential in predicting treatment response and its application in initial treatment decisions.

Current concept of MS pathology and MRI correlates

One of the main histopathological hallmarks of MS is focal inflammatory demyelination. Characterized by perivenular inflammation, it leads to the characteristic perivascular distribution of MS lesions in the periventricular and deep white matter. Blood-brain barrier disruption is evident by contrast enhancement on MRI.¹³ In addition to white matter involvement, MS is characterized by leucocortical inflammatory demyelination. MRI correlates include juxtacortical and various types of cortical grey matter lesions.^14,15 Another clinically relevant feature of MS with strong impact on disease course and disability progression is focal inflammatory demyelination in the spinal cord. This is frequently already present at early disease stages and accelerates over time, particularly in later progressive disease stages.^16,17 (Sub)acute focal inflammatory demyelinating lesions may show signs of partial or complete tissue repair, such as remyelination, during the first 6 months.^13,18,19 New or enlarging T2 lesions and contrast-enhancing lesions, which indicate ongoing focal inflammation, are the most relevant MRI marker of disease activity. These are central to treatment monitoring in clinical practice, particularly during non-progressive MS stages.⁸

Certain MS lesions demonstrate a slow expansion that may be visible on sequential MRI studies. Histopathologically, slowly expanding lesions (SEL) are characterized by the combination of a fully demyelinated, inactive centre without any signs of remyelination or prominent inflammation, and an active border including iron-laden phagocytes. On MRI, some of these T2-hyperintense lesions show a low signal intensity rim on susceptibility-based imaging, and are thus called paramagnetic rim lesions (PRL); these show an association with chronic T1-hypointense lesions (‘black holes’).^20,21 Also called ‘smouldering lesions’, SEL are clinically associated with progressive MS disease stages and course as well as with disability progression.²²

Another presentation of chronic inflammation in MS is characterized by clusters of B- and T-cells in the leptomeninges, which are frequently associated with demyelination of the adjacent subpial cortical grey matter.²³ It has been suggested that focal leptomeningeal inflammation corresponds to small hyperintense areas on late contrast-enhanced 3D fluid-attenuated inversion recovery (FLAIR) images.²⁴ Whether the accumulation of leptomeningeal immune cells directly causes cortical demyelination remains unclear. In other diseases, leptomeningeal inflammation is not typically associated with cortical demyelination.²⁵ A number of studies have reported an association of leptomeningeal contrast enhancement (LMCE) with focal atrophy of the adjacent cortex, but not focal demyelination,^26–28 suggesting a link to diffuse rather than focal neuroinflammation. Interestingly, LMCE also correlates with focal demyelinating lesions in the thalamus and hippocampus^28,29 as well as cortical and total grey matter volume (GMV).^30,31 Areas of LMCE usually remain stable during follow-up, and the in vivo response to disease-modifying treatment (DMT) remains unclear.²⁴ Resolution of LCME has been reported in two patients treated with high-dose corticosteroids³² and in 9 out of 20 patients treated with anti-CD20 therapies.³³ In a rat model of proteolipid protein-induced experimental autoimmune encephalitis, treatment with evobrutinib, an inhibitor of Bruton’s tyrosine kinase (BTK), promoted the resolution of focal meningeal inflammation on serial MRI.³⁴

Apart from focal inflammation, MS, in particular in progressive stages, is also characterized by diffuse inflammation that can appear on conventional MRI as areas of slightly increased T2-hyperintensity.^35–37 Various quantitative MRI methods to detect diffuse neuroinflammation have been described. On positron emission tomography, several radioligands binding to the 18-kDa translocator protein and ¹¹C-acetate label activated microglia and astrocytes, respectively.^38,39 Besides detecting diffuse neuroinflammation, these ligands may also bind in a focal pattern, associated both with the acute formation and chronic activity of lesions.³⁹

In addition to neuroinflammation, MS is also a neurodegenerative disease. Signs of axonal damage are present already in the early stages, and pronounced axonal loss occurs as the disease progresses. Associated focal and global atrophy of white and grey matter structures in the brain and spinal cord are established outcome measures of neurodegeneration in clinical trials. Various DMTs can slow down grey and white matter atrophy. For the routine clinical assessment of individual disease progression, imaging of neurodegeneration and atrophy is however not recommended at this time.^8,40

Prognostic value of MRI from the historical perspective

The first evidence suggesting the prognostic value of MRI measures was derived in patients presenting with a first clinical demyelinating event [later described as a clinically isolated syndrome (CIS)]. A four-parameter dichotomized MRI model including the number of lesions (using a cutoff of nine lesions), contrast enhancement and localization of lesions (juxtacortical, periventricular, infratentorial) showed a predictive value regarding the conversion to CDMS (Figure 1).⁴¹ These so-called Barkhof criteria were later validated in a second cohort, providing a sensitivity of 73%, a specificity of 73%, an accuracy of 73%, a positive predictive value of 55% and a negative predictive value of 85%.⁴² Further prediction models using a combination of these MRI criteria, cerebrospinal fluid analyses and neurophysiological findings were developed and validated.^43,44 Barkhof MRI criteria – modified by including spinal cord lesions – were later incorporated into the 2001 and 2005 revised McDonald criteria for the demonstration of dissemination in space.^45,46 These criteria were further simplified and continued to be included in the subsequent revisions of the McDonald criteria.^2,47 In addition, the Barkhof criteria, were incorporated into the Okuda criteria and more recent diagnostic criteria for patients with radiologically isolated syndromes (RIS).^48,49 The modifications of the McDonald criteria in 2001, 2005, 2010 and 2017, including adjustments to MRI and cerebrospinal fluid criteria, have progressively increased the sensitivity of diagnosing MS at the first clinical event. As a result, nearly 70% of patients experiencing their first clinical event now receive a diagnosis of MS.⁵⁰ Instead of a detailed discussion of the evolving definitions of CIS, we refer to patients with a first demyelinating attack, and interpret the risk of conversion to CDMS as risk of relapse.

Figure 1.

MRI images of the brain (a–e) and spinal cord (e, f) obtained in MS patients presenting with the first clinical event demonstrate MRI features that provide prognostic information regarding the clinical outcome. (a) Axial FLAIR image showing a high T2 lesion load in the periventricular, deep and juxtacortical white matter; (b) axial contrast-enhanced T1-weighted image showing two contrast-enhancing lesions in the deep white matter of the left hemisphere; (c) axial contrast-enhanced T1-weighted image showing several T1-hypointense lesions (‘black holes’); (d) axial DIR image showing several leucocortical lesions; (e) axial DIR image showing multiple infratentorial lesions including a brain stem lesion; (f, g) sagittal STIR and contrast-enhanced T1-weighted image showing asymptomatic lesions in the cervical spinal cord.

Prognostic value of brain and spinal cord MRI at disease onset

Diagnostic work-up at the time of the first clinical event suggestive of MS should include a contrast-enhanced MRI of the brain and spinal cord.⁸ In the following paragraphs and in Table 1, we describe and discuss various MRI measures at disease onset and/or in early RRMS that contain relevant prognostic information regarding further inflammatory disease activity, disability progression and the conversion to SPMS (Figure 1). These outcomes are better predicted by MRI findings than by clinical findings in early MS.^51–53

Table 1.

Stratification of risk level for MRI predictors and associated outcomes.

MRI metric	Risk levels	Domain	Associated outcomes	References
T2 lesion count	Moderate – ⩾3 High – ⩾10 Very high – ⩾20	Staging Change on follow-up: Grading	Risk of relapse EDSS after 5–20 years Time to SPMS	52–58
T2 lesion volume	Moderate – ⩾1 ml High – ⩾2.5 ml	Staging	EDSS after 10, 14 and 20 years SPMS after 20 years	55, 59, 60
T1-hypointense lesion volume	Moderate – ⩾2.5 ml	Staging	EDSS progression after 10 years MSFC after 6 years	61, 62
Gd⁺ lesion count	Moderate – 1 High – ⩾2	Grading	Risk of relapse EDSS after 6 years EDSS 6.0 after 10 years EDSS and SPMS after 15 years	51, 57, 62–64
Spinal lesions	High – ⩾1	Reserve	Risk of relapse EDSS and EDSS progression after 5 years EDSS 3.0 after 2 years	51, 63, 65–70
Brainstem lesions	Moderate – 1 High – ⩾2	Reserve	Risk of relapse EDSS 3.0 after 7–9 years SPMS after 30 years	51, 71–73
Cortical lesions	Moderate – ⩾4 High – ⩾7	Staging	EDSS progression after 1.5 and 5 years Cognitive performance after 5 years Time to SPMS SPMS after 6.5 years	74–76
Atrophy of whole brain, grey matter, deep grey matter	Moderate – if present	Reserve	Risk of relapse EDSS progression after 2 years EDSS and MSSS after 10 years EDSS 4.0 after 10.6 years EDSS progression after 13 years SPMS after 10 and 13 years	60, 77–80
Atrophy of cervical spinal cord	Moderate	Reserve	EDSS after 5 years	68
Paramagnetic rim lesions	High – ⩾4	Staging	Progressive disease course Disability (EDSS, MSSS, PASAT, SDMT) EDSS progression after 3.2 years	22, 81, 82
Slowly expanding lesions	High – ⩾4	Staging	EDSS progression after 3.2 years	81

EDSS, Expanded Disability Status Scale; MRI, magnetic resonance imaging; MSFC, Multiple Sclerosis Functional Composite; MSSS, Multiple Sclerosis Severity Scale; PASAT, Paced Auditory Serial Addition Test; SDMT, Symbol Digit Modalities Test; SPMS, secondary progressive multiple sclerosis.

T2-hyperintense white matter lesions

After the first clinical event, the presence and number of T2 white matter lesion (WML) has consistently been identified as the strongest independent predictor of further relapses.^{52,53–56,83,84} Even a single T2 WML confers a substantial risk of a second attack.^52,85,86 This risk is highest in patients with numerous lesions, for example, ⩾10 or more.^56,57 Risk, and time to relapse are modified by lesion topology in the brain^71,87–90 and presence of infratentorial or spinal cord lesions.^65–67,71

While the extent or development of T2 lesions are not closely associated with present clinical findings,⁹¹ the number and volume of T2 WML are moderate predictors of disability as measured by the Expanded Disability Status Scale (EDSS) after 5, 10, 14 or 20 years as well as the time to reach disability milestones.^55,57,59,60 Presence of 10 or more T2 WML at baseline was associated with accelerated development of disability after a first demyelinating attack and an 11-fold increased risk (95% CI: 6.7–19.3) of reaching EDSS 3.0 after a median follow-up of 81 months,⁵³ whereas 20 T2 WML or at least 2 Gd⁺ (gadolinium-enhancing) lesions at baseline predicted aggressive MS, defined by an EDSS of 6 after 10 years.⁵⁷ Patients who had progressed to SPMS at 20 years had higher T2 lesion volumes at baseline (median, 2.5 ml; range, 0–55 ml) compared to those who had not (0.7 ml; range, 0–13.7 ml).⁵⁵ In a study evaluating several brain MRI metrics, presence of two or more infratentorial lesions best predicted EDSS 3.0 after a median follow-up of 8.7 years.⁷² In patients with at least 9 T2 lesions at onset, infratentorial lesions increased the risk of relapse and of reaching EDSS 3.0 after a median follow-up of 7.7 years.⁷¹ Identification of at least one infratentorial lesion at baseline indicates an increased risk of EDSS >3.5 at 30 years [odds ratio (OR) 16.8; 95% CI: 2.0–139.7] and of having SPMS at 30 years disease duration (OR 26.0; 95% CI: 3.1–215.0).⁷³

The presence of at least one spinal cord lesion at onset predicted a second attack and increased the odds of EDSS progression after 5 years.^51,66 Especially when asymptomatic, spinal cord lesions increased the risk of reaching EDSS 3.0 [hazard ratio (HR) 29.8; 95% CI: 1.1–786.5].⁶⁷ Moreover, the number of baseline spinal lesions is independently associated with EDSS level and EDSS worsening after a median of 5.2 or 5.1 years.^68,69 Within 2 years of disease onset, spinal cord lesions independently predicted EDSS ⩾4.0 after 7 years.⁷⁰ At least one spinal cord lesion at baseline predicted EDSS at 15 years and indicated strongly higher odds of having SPMS after 15 years (OR 4.71; 95% CI: 1.72–12.92).⁶³ Asymptomatic spinal cord lesions were found in 43 of 143 patients (30.1%) with a non-spinal first demyelinating attack; unlike brain lesions, they were identified as an independent risk factor for short-term disability.⁶⁷

In RIS patients, infratentorial and spinal cord lesions are independent predictors of a first clinical attack or conversion to primary progressive MS.^92,93

T1-hypointense lesions

Lesions with persistent, marked T1-hypointensity in the absence of Gd enhancement on T1-weighted (turbo/fast) spin echo sequences are historically addressed as chronic/persistent black holes (CBH) (Figure 1). They are not frequently observed in early MS stages. CBH signify permanent tissue destruction, including neuroaxonal loss, and are correlated with disability in cross-sectional studies.^94,95 The relation between baseline T1 lesion volume and long-term disability outcomes is not well understood. One study reported a significant association between a composite of baseline CBH volume and increase over time with disability after 10 years.⁶¹ At the first demyelinating attack, T1 lesion load predicted MSFC after 6 years, whereas baseline volume did not predict disability in another study.^62,96 T1 spin echo sequences are not commonly performed in the routine examination of MS patients, limiting the availability of this measure.

Gd⁺ lesions

Gd⁺ lesions indicate acute inflammation leading to blood-brain barrier disruption, that is, active disease. At onset, they predict the occurrence of a second attack.^51,64 If detected at the first demyelinating attack or within 1–3 years after diagnosis of MS, indicating persistent inflammatory disease activity, Gd⁺ lesions are linked to mid-to-long term disability and higher odds of having SPMS after 15 years; this association is even more pronounced in patients with spinal lesions (Figure 1).^51,62,63 Demonstration of two or more Gd⁺ lesions 3–5 months after a first demyelinating attack identified patients with aggressive MS (EDSS ⩾6.0 after 10 years) in a large cohort.⁵⁷

Cortical grey matter lesions

Imaging of cortical grey matter lesions (CL) requires appropriate acquisition parameters including sufficient spatial resolution, magnetic field strength (preferably 3T) and pulse sequences such as 3D double inversion recovery (DIR) and/or phase-sensitive inversion recovery (PSIR) (Figure 1). The clinical relevance of CL particularly regarding cognition has been conclusively demonstrated.⁷⁴ The number of CL at disease onset is an independent predictor of disability and of the risk and time until conversion to SPMS.^75,76 In patients with seven or more CL present at baseline, half developed SPMS within 6.5 years.⁷⁶ This demonstrates the importance of cortical lesions when detected in the diagnostic work-up of patients with suspected MS. Accordingly, CL were included in the 2017 revised McDonald criteria for the demonstration of dissemination in time.² Presently, inclusion of DIR or PSIR sequences is not recommended for routine monitoring purposes due to limited availability and concerns of reproducibility.^8,97

Volumetric data

Brain atrophy is a marker of irreversible neuroaxonal loss. After the first demyelinating attack, reduced normalized brain volume (NBV) and GMV are predictors of disability progression.^60,77 Prognostic information appears to derive mainly from subcortical grey matter atrophy, whereas changes in white matter volume are not demonstrably predictive.^77–79

Particularly in progressive MS, atrophy of the cervical spinal cord is moderately correlated with EDSS.⁹⁸ In patients with a non-spinal first demyelinating attack, the change in upper cervical cord cross-sectional area (UCCA) over 5 years was independently associated with EDSS at follow-up, whereas baseline UCCA did not predict EDDS 3.0 after 5 years.⁶³ Accelerated atrophy of the upper cervical spinal cord precedes conversion to SPMS by at least 4 years.⁹⁹

While considered a helpful predictor of the future disease course, quantitative MRI so far plays a limited role in the routine clinical care of individual patients, due to difficulties in implementation and numerous potential confounders.⁴⁴

Paramagnetic rim lesions

PRL are identified using susceptibility-based MRI techniques, most commonly susceptibility weighted imaging (SWI) at 3T or 7T. PRL form a relatively small subset of SEL detected in longitudinal studies.⁸¹ On histopathological examination, PRL reflect chronic inflammation at the border of the lesion with iron-laden microglia and macrophages.⁸² Overlapping histopathology terms include chronic inactive-active lesions or ‘smoldering’ lesions. This lesion type is associated with compartmentalized neuroinflammation, failure of remyelination and ongoing tissue damage.^22,82,100 PRL were found in 27% of recently diagnosed RRMS patients.¹⁰¹ The presence of four or more PRL was an independent predictor of disability worsening in a heterogeneous group of MS patients.²² In a cohort of 61 recently diagnosed patients (median disease duration, 0.4 years), more than half had at least one PRL at baseline, and 92% had at least one SEL.⁸² PRL showing slow expansion on follow-up scans were associated with greater EDSS progression after a median of 3.2 years.⁸²

Prognostic value of MRI during early follow-up

Within at least 5 years from diagnosis, T2 lesion load remains a predictor of long-term disability in established RRMS.⁵⁹ Further information can be derived by comparing baseline and follow-up MRI: New T2 lesions after 3 months identify patients with subclinical disease activity and a high risk of relapse.⁵² Development of new or enlarging T2 WML over 2–3 years was weakly correlated to changes in disability.⁵⁸ An increase in T2 WML volume after 2–5 years correlated moderately with physical and cognitive impairment after 13–14 years, and a yearly increase by 2.89 ml (95% CI: 1.78–4.01) predicted SPMS after 20 years.^55,59,102 Moreover, new infratentorial or spinal cord lesions within the first 1–3 years predicted SPMS after 15 years.⁶³ Accumulation of T2 lesions over time does not reflect the total amount of focal inflammation since lesions may remyelinate or shrink over time. In fact, atrophied T2 lesion volume has been suggested as a marker of disease progression.¹⁰³

Current focal inflammation on MRI can be assessed by Gd-enhanced imaging, if required.⁸ Emergence of new Gd⁺ lesions predicts short-term relapses but not confirmed disability progression during the following 2 years.¹⁰⁴

Longitudinal MRI may identify an increased rate of brain volume loss, which is associated with subsequent disability.^60,77,105 Brain atrophy can be difficult to interpret early after the initiation of anti-inflammatory DMT due to pseudoatrophy.¹⁰⁶ Baseline GMV predicts disability progression and conversion to SPMS after 9 years.¹⁰⁷ Central or subcortical grey matter atrophy may be of particular prognostic relevance in predicting clinical impairment.^{78,80,108,109} Accelerated atrophy may be conceptualized as premature aging of the brain, implying reduced functional reserve.¹¹⁰

Prediction of treatment response

In addition to the predictive value of imaging characteristics in terms of disease and disability progression in general, MRI may also aid in predicting treatment response. An increase of T2 lesion volume, or emergence of new T2 lesions or of combined active lesions (new or enlarging T2 lesions and Gd⁺ lesions) on MRI in the first 1–2 years of DMT was first identified as a predictor of disability progression in interferon-treated patients.^111–114 Later, no evidence of disease activity (NEDA) was introduced not only as an outcome in clinical trials, but also as a therapeutic goal and measure of individual treatment response.^115–119 NEDA-3 status requires the absence of relapses, EDSS progression and inflammatory MRI activity (new or enlarging T2 lesions and Gd⁺ lesions).^115,116 NEDA-4 expands this definition by adding the absence of increased whole brain atrophy.¹²⁰ Further modifying the NEDA concept by including liquid biomarkers, such as neurofilament light chain concentrations in the serum, has been suggested.¹²¹ A recent meta-analysis of DMT trials found that a NEDA-3 or NEDA-4 status after 1 or 2 years indicated a lower risk of long-term disability progression after 6 years.¹²²

To weigh the risk of treatment failure, several operationalized derivations of the NEDA concept have been proposed to measure efficacy after 1 year of treatment. These systems aim to define a level of disease activity while on DMT (minimal evidence of disease activity, MEDA) that can be tolerated without risking disability worsening.¹²³ For patients treated with interferon-beta, the Rio score is calculated by adding one point each for (a) at least one relapse, (b) at least three new MRI lesions, and (c) a sustained EDSS deterioration by at least one point; this results in a prediction of low (score 0–1) or high (2–3) risk.¹²⁴ For the modified Rio score, one point is given for (a) at least five new T2 lesions, and (b) one relapse, while two points are given for two or more relapses after 1 year. Again, 0–1 points indicate low risk, while 2–3 indicate high risk.¹²⁵ Long-term follow-up of patients showed that a Rio score of at least 2 obtained after 1 year was associated with EDSS worsening after 6.7 years.¹²⁶ Obtained after 1 year, the MAGNIMS score rates 3 or more new T2 lesions with one point, one relapse with one point and two relapses with two points. A total score of 0–2 is calculated, indicating a low, moderate or high risk of treatment failure (EDSS worsening or treatment switch for inefficacy; 27% and 48% for scores 1 and 2 respectively) and of disability progression (22% and 29% for scores 1 and 2 respectively) at 3 years.¹²⁷ MAGNIMS score predicted a 7-year risk of disability worsening in teriflunomide treated patients, and long-term disability progression for up to 15 years in interferon beta-1a treated patients.^128,129 Application of the MAGNIMS score to ponesimod treatment has so far only been published as an abstract.¹³⁰ In all of these scoring systems, isolated development of new lesions in the absence of clinical relapse or early EDSS progression was not significantly linked to EDSS worsening. However, longer-term follow-ups indicate an increased risk of disability also in the intermediate risk categories of the modified Rio and MAGNIMS scores, calling into question the use of MEDA thresholds in guiding therapy.¹¹ Unless clearly attributable to treatment inadherence, for many MS neurologists any new asymptomatic MRI activity while on platform DMTs prompts treatment escalation.¹³¹

Inclusion of MRI in the treatment decision-making process

Need for early risk stratification in MS

Significant progress has been made in understanding the clinico-pathological evolution of MS. Very early in the disease, focal autoimmune demyelination probably lays the groundwork for smouldering inflammation and progressive neuroaxonal loss which may contribute to disability progression independent of relapse activity (PIRA).^13–133 A significant role of chronically activated, pro-inflammatory microglia is increasingly recognized.¹³ With current DMTs focussing on peripheral adaptive immunity, new treatment options targeting microglia are under development, including BTK inhibitors.¹³⁴

Concerning presently available DMTs, data from active-controlled trials, open-label extensions and large patient registries supports the efficacy of early highly effective treatments (HET) not only to reduce relapses and focal demyelination but also long-term disability and conversion to SPMS.^135,136 However, accelerated brain atrophy and disability progression may persist even if relapses and new lesions are successfully controlled by DMT.¹³⁷ It is because of this ‘silent progression’ that the ability of current DMTs to reduce PIRA has been questioned. On the other hand, the potential of early HET to prevent smouldering inflammation and PIRA remains in part unexplored due to the imperfect suppression of focal inflammation even with HET and delayed treatment initiation in already symptomatic disease.¹³⁸

Reserving HET for clinically manifest ‘highly active’ or ‘aggressive’ MS implies foregoing a preventive approach until irreversible damage has accumulated, functional reserve has been compromised, and secondary neurodegeneration has already been induced. Considering the improved long-term safety of modern DMT, many MS neurologists therefore advocate first-line use of HET in active MS rather than following an escalation approach.^121,139 Still, the risk of long-term disability from MS must be weighed against potential side effects from DMT. Tailored therapy therefore should be based on an early prognosis of the individual risk of long-term disability.¹⁴⁰ Most of the available evidence on unfavourable predictors was derived in cohorts that did not yet have access to HET, highlighting the potential to improve the outcome of those at risk.

Early MRI findings remain the strongest biomarkers for mid- and long-term disability outcomes in MS so far.¹⁴¹ Notably, WML dynamics decrease over time and lose predictive quality with longer disease duration and in progressive disease.^{57,100,102,142–144} Loss of correlation of lesion load to present disability and future deterioration reflects an increasing and uncontrollable variance, introduced, for example, by strategic lesions,^145,146 damage to specific networks,^144,147 spinal cord involvement,³⁶ and diffuse pathology.^35–37 For these reasons, the predictive capabilities of MRI should be applied in early RRMS and strongly guide initial treatment decisions.

Integrated interpretation of early MRI findings

Three interconnected levels of information are available from MRI at the time of diagnosis or soon thereafter (Figure 2):

- The number and/or volume of T2-hyperintense lesions indicate the extent of accumulated focal demyelination, complemented by T1-hypointense and cortical lesions. Lesion burden reflects the intensity and duration of MS in the preclinical phase and may be visualized as suggesting the actual position of the individual on a generalized MS trajectory. Hence, these metrics contain staging information.

- The number of Gd⁺ T1 lesions on a single study or the increase in T2 lesion load on short-term follow-up reflect disease activity. The level of disease activity relates to the dynamics of disability increase and time to onset of SPMS and can be interpreted as grading information.

- Functional reserve, that is, the ability to compensate for further neuroaxonal loss, is influenced by lesion accrual, but it also correlates to whole brain and GM atrophy and to damage of non-redundant pathways, for example, motor tracts, brain stem or spinal cord. Lower reserve implies a shorter time to disability milestones and onset of SPMS.

Figure 2.

Integrated view of prognostic MRI findings. The expected course of the disease, in particular the evolution of disability over time, is modulated by the level of disease activity and brain reserve. Initial disease activity can be graded according to the presence of gadolinium-enhancing (Gd⁺) lesions or the number or volume of new or enlarging T2-hyperintense (NET2) lesions on short-term follow-up scans. White matter and cortical lesion load is a product metric of preclinical disease duration and severity, providing staging-related information. In addition to lesion volume, atrophy and presence of lesions in critical systems such as spinal cord and brain stem indicate lower functional reserve and a risk of more rapid development of disability. Together, staging, grading and reserve-related information may be useful in predicting the individual disease course and recommended treatment intensity.

In clinical experience, patients show abnormality typically in not one but several MRI parameters, and this is reflected in the multivariable analyses and correlation studies reviewed above.

Proposed approach to early prognosis

When discussing the role of MRI in initial treatment decisions, similar challenges emerge as with the MEDA concept for treatment escalation. Key questions include: What level of disability risk is considered acceptable? What is the requisite level of accuracy for a predictor to be deemed actionable? Most RRMS patients face a substantial risk of disability,^148,149 and any avoidable impairment would be unacceptable from their perspective. If a ‘non-benign’ course of MS is assumed as the default scenario rather than a ‘benign’ one, establishing safe cutoffs for individual predictors or their combination becomes even more complex.

We propose a simplified, pragmatic approach to MRI in the initial treatment decision and follow-up: newly diagnosed patients should undergo a full MRI assessment at baseline.⁸ This includes high quality 3D FLAIR and Gd-enhanced T1 scans of the brain to determine white matter T2 lesion load as well as presence of hypomyelinated and cortical lesions. If routinely available, DIR (or PSIR) and SWI sequences for improved detection of cortical lesions and PRL, as well as measurements of lesion and brain volumes can be included. All patients should undergo MRI of the entire spinal cord.

Estimated risk stratification for each metric is listed in Table 1. T2 lesion count or volume should be regarded as the strongest predictor. The specificity of any T2 lesions should be ascertained, if necessary, by demonstration of a central vein.⁸ Except for atrophy measures, all remaining predictors imply the presence of T2 lesions.

In patients presenting with low-risk findings on each available staging, grading and reserve-related metric, a favourable prognosis is suggested, and these patients may be offered a moderately effective DMT as their initial treatment. If one or more predictors indicate an unfavourable prognosis, HET should be recommended first-line.

Following the initiation of a DMT, it is advisable to consistently observe patients for any signs of disease activity. A re-baselining MRI is obtained after the expected onset of effect at 3–6 months. In the absence of clinical relapse, treatment effectiveness is determined by a further MRI after 1 year of therapy. A relevant (⩾3) increase in T2 lesions predicts treatment failure and should prompt escalation to a DMT with higher efficacy. Gd-enhanced acquisitions are not recommended for routine follow-up MRI but may add predictive value when included in the re-baselining study.^8,150

Treatment decisions should take into account additional risk factors for a more rapid progression of disability or earlier conversion to SPMS, including older age,^{89,148,151,152} male sex,¹⁴⁸ clinical disability,¹⁵³ early relapse frequency,^{89,148,154,155} presence of oligoclonal bands in the CSF,¹⁵⁶ elevated serum concentration of neurofilament or glial fibrillary acidic protein (GFAP),^157,158 vitamin D deficiency¹⁵⁹ and comorbidities.¹⁶⁰ Moreover, individual treatment risks, patient preferences (e.g. planned pregnancy), local availability and licensing of DMTs must be considered.^161,162

In view of recent advances in risk stratification of patients with RIS,^48,49,80,163 it is likely that most MRI predictors also apply to preclinical MS. As we still await further evidence on DMT use in RIS,^164,165 we have limited our discussion here to patients with early established MS according to the current McDonald diagnostic criteria.

Footnotes

Acknowledgements

The authors wish to thank Dr. Jacob Henry Gogarten for proofreading the manuscript.

Declarations

ORCID iDs

Olaf Hoffmann

Ralf A. Linker

Heinz Wiendl

References

Reich

Lucchinetti

Calabresi

. Multiple sclerosis. N Engl J Med 2018; 378: 169–180.

Thompson

Banwell

Barkhof

, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018; 17: 162–173.

Barnett

Reddel

. MRI and laboratory monitoring of disease-modifying therapy efficacy and risks. Curr Opin Neurol 2022; 35: 278–285.

Wijburg

Warnke

McGuigan

, et al. Pharmacovigilance during treatment of multiple sclerosis: early recognition of CNS complications. J Neurol Neurosurg Psychiatry 2021; 92: 177–188.

Rovira

Wattjes

Tintoré

, et al. Evidence-based guidelines: MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis-clinical implementation in the diagnostic process. Nat Rev Neurol 2015; 11: 471–482.

Wattjes

Rovira

Miller

, et al. Evidence-based guidelines: MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis – establishing disease prognosis and monitoring patients. Nat Rev Neurol 2015; 11: 597–606.

Traboulsee

Simon

Stone

, et al. Revised recommendations of the consortium of MS centers task force for a standardized MRI protocol and clinical guidelines for the diagnosis and follow-up of multiple sclerosis. AJNR Am J Neuroradiol 2016; 37: 394–401.

Wattjes

Ciccarelli

Reich

, et al. 2021 MAGNIMS-CMSC-NAIMS consensus recommendations on the use of MRI in patients with multiple sclerosis. Lancet Neurol 2021; 20: 653–670.

Rovira

Doniselli

Auger

, et al. Use of gadolinium-based contrast agents in multiple sclerosis: a review by the ESMRMB-GREC and ESNR Multiple Sclerosis Working Group. Eur Radiol. Epub ahead of print 2 September 2023. DOI: 10.1007/s00330-023-10151-y.

10.

Cortese

Giorgio

Severa

, et al. MRI prognostic factors in multiple sclerosis, neuromyelitis optica spectrum disorder, and myelin oligodendrocyte antibody disease. Front Neurol 2021; 12: 679881.

11.

Río

Rovira

Gasperini

, et al. Treatment response scoring systems to assess long-term prognosis in self-injectable DMTs relapsing-remitting multiple sclerosis patients. J Neurol 2022; 269: 452–459.

12.

Tiu

Enache

Panea

, et al. Predictive MRI biomarkers in MS: a critical review. Medicina (Kaunas) 2022; 58: 377.

13.

Kuhlmann

Moccia

Coetzee

, et al. Multiple sclerosis progression: time for a new mechanism-driven framework. Lancet Neurol 2023; 22: 78–88.

14.

Geurts

JJG

Calabrese

Fisher

, et al. Measurement and clinical effect of grey matter pathology in multiple sclerosis. Lancet Neurol 2012; 11: 1082–1092.

15.

Bö

Geurts

JJG

Mörk

, et al. Grey matter pathology in multiple sclerosis. Acta Neurol Scand Suppl 2006; 183: 48–50.

16.

Petrova

Nutma

Carassiti

, et al. Synaptic loss in multiple sclerosis spinal cord. Ann Neurol 2020; 88: 619–625.

17.

Eden

Gros

Badji

, et al. Spatial distribution of multiple sclerosis lesions in the cervical spinal cord. Brain 2019; 142: 633–646.

18.

Bodini

Veronese

García-Lorenzo

, et al. Dynamic imaging of individual remyelination profiles in multiple sclerosis. Ann Neurol 2016; 79: 726–738.

19.

Chen

Collins

Atkins

, et al. Magnetization transfer ratio evolution with demyelination and remyelination in multiple sclerosis lesions. Ann Neurol 2008; 63: 254–262.

20.

Dal-Bianco

Grabner

Kronnerwetter

, et al. Slow expansion of multiple sclerosis iron rim lesions: pathology and 7 T magnetic resonance imaging. Acta Neuropathol 2017; 133: 25–42.

21.

Calvi

Tur

Chard

, et al. Slowly expanding lesions relate to persisting black-holes and clinical outcomes in relapse-onset multiple sclerosis. Neuroimage Clin 2022; 35: 103048.

22.

Absinta

Sati

Masuzzo

, et al. Association of chronic active multiple sclerosis lesions with disability in vivo. JAMA Neurol 2019; 76: 1474–1483.

23.

Howell

Reeves

Nicholas

, et al. Meningeal inflammation is widespread and linked to cortical pathology in multiple sclerosis. Brain 2011; 134: 2755–2771.

24.

Ineichen

Tsagkas

Absinta

, et al. Leptomeningeal enhancement in multiple sclerosis and other neurological diseases: a systematic review and meta-analysis. Neuroimage Clin 2022; 33: 102939.

25.

Absinta

Vuolo

Rao

, et al. Gadolinium-based MRI characterization of leptomeningeal inflammation in multiple sclerosis. Neurology 2015; 85: 18–28.

26.

Bergsland

Ramasamy

Tavazzi

, et al. Leptomeningeal contrast enhancement is related to focal cortical thinning in relapsing-remitting multiple sclerosis: a cross-sectional MRI study. AJNR Am J Neuroradiol 2019; 40: 620–625.

27.

Bonnan

Money

Desblache

, et al. Focal cortical atrophy following transient meningeal enhancement in a progressive multiple sclerosis. Neurol Sci 2021; 42: 1959–1961.

28.

Ighani

Jonas

Izbudak

, et al. No association between cortical lesions and leptomeningeal enhancement on 7-Tesla MRI in multiple sclerosis. Mult Scler 2020; 26: 165–176.

29.

Zurawski

Tauhid

Chu

, et al. 7T MRI cerebral leptomeningeal enhancement is common in relapsing-remitting multiple sclerosis and is associated with cortical and thalamic lesions. Mult Scler 2020; 26: 177–187.

30.

Makshakov

Magonov

Totolyan

, et al. Leptomeningeal contrast enhancement is associated with disability progression and grey matter atrophy in multiple sclerosis. Neurol Res Int 2017; 2017: 8652463.

31.

Zivadinov

Ramasamy

Vaneckova

, et al. Leptomeningeal contrast enhancement is associated with progression of cortical atrophy in MS: a retrospective, pilot, observational longitudinal study. Mult Scler 2017; 23: 1336–1345.

32.

Titelbaum

Engisch

Schwartz

, et al. Leptomeningeal enhancement on 3D-FLAIR MRI in multiple sclerosis: systematic observations in clinical practice. J Neuroimaging 2020; 30: 917–929.

33.

Friedli

Wagner

Hammer

, et al. Leptomeningeal enhancement under different MS immunotherapies: a monocentric retrospective cohort study of 214 patients. Mult Scler 2023; 29: 63–73.

34.

Bhargava

Kim

Reyes

, et al. Imaging meningeal inflammation in CNS autoimmunity identifies a therapeutic role for BTK inhibition. Brain 2021; 144: 1396–1408.

35.

Seewann

Vrenken

van der Valk

, et al. Diffusely abnormal white matter in chronic multiple sclerosis: imaging and histopathologic analysis. Arch Neurol 2009; 66: 601–609.

36.

Nijeholt

van Walderveen

Castelijns

, et al. Brain and spinal cord abnormalities in multiple sclerosis. Correlation between MRI parameters, clinical subtypes and symptoms. Brain 1998; 121: 687–697.

37.

Cairns

Vavasour

Traboulsee

, et al. Diffusely abnormal white matter in multiple sclerosis. J Neuroimaging 2022; 32: 5–16.

38.

Herranz

Giannì

Louapre

, et al. Neuroinflammatory component of gray matter pathology in multiple sclerosis. Ann Neurol 2016; 80: 776–790.

39.

Brier

Taha

. Measuring pathology in patients with multiple sclerosis using positron emission tomography. Curr Neurol Neurosci Rep 2023; 23: 479–488.

40.

Sastre-Garriga

Pareto

Battaglini

, et al. MAGNIMS consensus recommendations on the use of brain and spinal cord atrophy measures in clinical practice. Nat Rev Neurol 2020; 16: 171–182.

41.

Barkhof

Filippi

Miller

, et al. Comparison of MRI criteria at first presentation to predict conversion to clinically definite multiple sclerosis. Brain 1997; 120: 2059–2069.

42.

Tintoré

Rovira

Martínez

, et al. Isolated demyelinating syndromes: comparison of different MR imaging criteria to predict conversion to clinically definite multiple sclerosis. AJNR Am J Neuroradiol 2000; 21: 702–706.

43.

Tintoré

Rovira

Brieva

, et al. Isolated demyelinating syndromes: comparison of CSF oligoclonal bands and different MR imaging criteria to predict conversion to CDMS. Mult Scler 2001; 7: 359–363.

44.

Sastre-Garriga

Tintoré

Rovira

, et al. Conversion to multiple sclerosis after a clinically isolated syndrome of the brainstem: cranial magnetic resonance imaging, cerebrospinal fluid and neurophysiological findings. Mult Scler 2003; 9: 39–43.

45.

McDonald

Compston

Edan

, et al. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann Neurol 2001; 50: 121–127.

46.

Polman

Reingold

Edan

, et al. Diagnostic criteria for multiple sclerosis: 2005 revisions to the ‘McDonald Criteria’. Ann Neurol 2005; 58: 840–846.

47.

Polman

Reingold

Banwell

, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011; 69: 292–302.

48.

Okuda

Mowry

Beheshtian

, et al. Incidental MRI anomalies suggestive of multiple sclerosis: the radiologically isolated syndrome. Neurology 2009; 72: 800–805.

49.

Lebrun-Frénay

Okuda

Siva

, et al. The radiologically isolated syndrome: revised diagnostic criteria. Brain 2023; 146: 3431–3443.

50.

Schwenkenbecher

Wurster

Konen

, et al. Impact of the McDonald criteria 2017 on early diagnosis of relapsing-remitting multiple sclerosis. Front Neurol 2019; 10: 188.

51.

Swanton

Fernando

Dalton

, et al. Early MRI in optic neuritis: the risk for disability. Neurology 2009; 72: 542–550.

52.

Swanton

Fernando

Dalton

, et al. Early MRI in optic neuritis: the risk for clinically definite multiple sclerosis. Mult Scler 2010; 16: 156–165.

53.

Tintore

Rovira

Río

, et al. Defining high, medium and low impact prognostic factors for developing multiple sclerosis. Brain 2015; 138: 1863–1874.

54.

Tintoré

Rovira

Río

, et al. Baseline MRI predicts future attacks and disability in clinically isolated syndromes. Neurology 2006; 67: 968–972.

55.

Fisniku

Brex

Altmann

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

56.

Kuhle

Disanto

Dobson

, et al. Conversion from clinically isolated syndrome to multiple sclerosis: a large multicentre study. Mult Scler 2015; 21: 1013–1024.

57.

Tintore

Arrambide

Otero-Romero

, et al. The long-term outcomes of CIS patients in the Barcelona inception cohort: looking back to recognize aggressive MS. Mult Scler 2020; 26: 1658–1669.

58.

Filippi

Paty

Kappos

, et al. Correlations between changes in disability and T2-weighted brain MRI activity in multiple sclerosis: a follow-up study. Neurology 1995; 45: 255–260.

59.

Brex

Ciccarelli

O’Riordan

, et al. A longitudinal study of abnormalities on MRI and disability from multiple sclerosis. N Engl J Med 2002; 346: 158–164.

60.

Popescu

Agosta

Hulst

, et al. Brain atrophy and lesion load predict long term disability in multiple sclerosis. J Neurol Neurosurg Psychiatry 2013; 84: 1082–1091.

61.

Giorgio

Stromillo

Bartolozzi

, et al. Relevance of hypointense brain MRI lesions for long-term worsening of clinical disability in relapsing multiple sclerosis. Mult Scler 2014; 20: 214–219.

62.

Di Filippo

Anderson

Altmann

, et al. Brain atrophy and lesion load measures over 1 year relate to clinical status after 6 years in patients with clinically isolated syndromes. J Neurol Neurosurg Psychiatry 2010; 81: 204–208.

63.

Brownlee

Altmann

Prados

, et al. Early imaging predictors of long-term outcomes in relapse-onset multiple sclerosis. Brain 2019; 142: 2276–2287.

64.

Rovira

Swanton

Tintoré

, et al. A single, early magnetic resonance imaging study in the diagnosis of multiple sclerosis. Arch Neurol 2009; 66: 587–592.

65.

Patrucco

Rojas

Cristiano

. Assessing the value of spinal cord lesions in predicting development of multiple sclerosis in patients with clinically isolated syndromes. J Neurol 2012; 259: 1317–1320.

66.

Sombekke

Wattjes

Balk

, et al. Spinal cord lesions in patients with clinically isolated syndrome: a powerful tool in diagnosis and prognosis. Neurology 2013; 80: 69–75.

67.

Arrambide

Rovira

Sastre-Garriga

, et al. Spinal cord lesions: a modest contributor to diagnosis in clinically isolated syndromes but a relevant prognostic factor. Mult Scler 2018; 24: 301–312.

68.

Brownlee

Altmann

Alves Da Mota

, et al. Association of asymptomatic spinal cord lesions and atrophy with disability 5 years after a clinically isolated syndrome. Mult Scler 2017; 23: 665–674.

69.

Rocca

Valsasina

Meani

, et al. Spinal cord lesions and brain grey matter atrophy independently predict clinical worsening in definite multiple sclerosis: a 5-year, multicentre study. J Neurol Neurosurg Psychiatry 2023; 94: 10–18.

70.

D’Amico

Patti

Leone

, et al. Negative prognostic impact of MRI spinal lesions in the early stages of relapsing-remitting multiple sclerosis. Mult Scler J Exp Transl Clin 2016; 2: 2055217316631565.

71.

Tintore

Rovira

Arrambide

, et al. Brainstem lesions in clinically isolated syndromes. Neurology 2010; 75: 1933–1938.

72.

Minneboo

Barkhof

Polman

, et al. Infratentorial lesions predict long-term disability in patients with initial findings suggestive of multiple sclerosis. Arch Neurol 2004; 61: 217–221.

73.

Chung

Altmann

Barkhof

, et al. A 30-year clinical and magnetic resonance imaging observational study of multiple sclerosis and clinically isolated syndromes. Ann Neurol 2020; 87: 63–74.

74.

Calabrese

Filippi

Gallo

. Cortical lesions in multiple sclerosis. Nat Rev Neurol 2010; 6: 438–444.

75.

Calabrese

Poretto

Favaretto

, et al. Cortical lesion load associates with progression of disability in multiple sclerosis. Brain 2012; 135: 2952–2961.

76.

Scalfari

Romualdi

Nicholas

, et al. The cortical damage, early relapses, and onset of the progressive phase in multiple sclerosis. Neurology 2018; 90: e2107–e2118.

77.

Pérez-Miralles

Sastre-Garriga

Tintoré

, et al. Clinical impact of early brain atrophy in clinically isolated syndromes. Mult Scler 2013; 19: 1878–1886.

78.

Azevedo

Cen

Khadka

, et al. Thalamic atrophy in multiple sclerosis: a magnetic resonance imaging marker of neurodegeneration throughout disease. Ann Neurol 2018; 83: 223–234.

79.

Eshaghi

Prados

Brownlee

, et al. Deep gray matter volume loss drives disability worsening in multiple sclerosis. Ann Neurol 2018; 83: 210–222.

80.

Filippi

Preziosa

Copetti

, et al. Gray matter damage predicts the accumulation of disability 13 years later in MS. Neurology 2013; 81: 1759–1767.

81.

Calvi

Haider

Prados

, et al. In vivo imaging of chronic active lesions in multiple sclerosis. Mult Scler 2022; 28: 683–690.

82.

Calvi

Clarke

Prados

, et al. Relationship between paramagnetic rim lesions and slowly expanding lesions in multiple sclerosis. Mult Scler 2023; 29: 352–362.

83.

Morrissey

Miller

Kendall

, et al. The significance of brain magnetic resonance imaging abnormalities at presentation with clinically isolated syndromes suggestive of multiple sclerosis: a 5-year follow-up study. Brain 1993; 116: 135–146.

84.

Filippi

Horsfield

Morrissey

, et al. Quantitative brain MRI lesion load predicts the course of clinically isolated syndromes suggestive of multiple sclerosis. Neurology 1994; 44: 635–641.

85.

O’Riordan

Thompson

Kingsley

, et al. The prognostic value of brain MRI in clinically isolated syndromes of the CNS: a 10-year follow-up. Brain 1998; 121: 495–503.

86.

O’Riordan

Losseff

Phatouros

, et al. Asymptomatic spinal cord lesions in clinically isolated optic nerve, brain stem, and spinal cord syndromes suggestive of demyelination. J Neurol Neurosurg Psychiatry 1998; 64: 353–357.

87.

Giorgio

Battaglini

Rocca

, et al. Location of brain lesions predicts conversion of clinically isolated syndromes to multiple sclerosis. Neurology 2013; 80: 234–241.

88.

Galassi

Prosperini

Logoteta

, et al. A lesion topography-based approach to predict the outcomes of patients with multiple sclerosis treated with Interferon Beta. Mult Scler Relat Disord 2016; 8: 99–106.

89.

Sechi

Keegan

Kaufmann

, et al. Unilateral motor progression in MS: association with a critical corticospinal tract lesion. Neurology 2019; 93: e628–e634.

90.

Pareto

Garcia-Vidal

Groppa

, et al. Prognosis of a second clinical event from baseline MRI in patients with a CIS: a multicenter study using a machine learning approach. Neuroradiology 2022; 64: 1383–1390.

91.

Barkhof

. The clinico-radiological paradox in multiple sclerosis revisited. Curr Opin Neurol 2002; 15: 239–245.

92.

Lebrun-Frenay

Kantarci

Siva

, et al. Radiologically isolated syndrome: 10-year risk estimate of a clinical event. Ann Neurol 2020; 88: 407–417.

93.

Okuda

Mowry

Cree

BAC

, et al. Asymptomatic spinal cord lesions predict disease progression in radiologically isolated syndrome. Neurology 2011; 76: 686–692.

94.

van Walderveen

Barkhof

Pouwels

, et al. Neuronal damage in T1-hypointense multiple sclerosis lesions demonstrated in vivo using proton magnetic resonance spectroscopy. Ann Neurol 1999; 46: 79–87.

95.

Rocca

Comi

Filippi

. The role of T1-weighted derived measures of neurodegeneration for assessing disability progression in multiple sclerosis. Front Neurol 2017; 8: 433.

96.

Minneboo

Uitdehaag

BMJ

Jongen

, et al. Association between MRI parameters and the MS severity scale: a 12 year follow-up study. Mult Scler 2009; 15: 632–637.

97.

Geurts

JJG

Roosendaal

Calabrese

, et al. Consensus recommendations for MS cortical lesion scoring using double inversion recovery MRI. Neurology 2011; 76: 418–424.

98.

Song

Qiu

, et al. Correlation between EDSS scores and cervical spinal cord atrophy at 3T MRI in multiple sclerosis: a systematic review and meta-analysis. Mult Scler Relat Disord 2020; 37: 101426.

99.

Bischof

Papinutto

Keshavan

, et al. Spinal cord atrophy predicts progressive disease in relapsing multiple sclerosis. Ann Neurol 2022; 91: 268–281.

100.

Absinta

Sati

Reich

. Advanced MRI and staging of multiple sclerosis lesions. Nat Rev Neurol 2016; 12: 358–368.

101.

Ng Kee Kwong

Mollison

Meijboom

, et al. Rim lesions are demonstrated in early relapsing-remitting multiple sclerosis using 3 T-based susceptibility-weighted imaging in a multi-institutional setting. Neuroradiology 2022; 64: 109–117.

102.

Rudick

Lee

J-C

Simon

, et al. Significance of T2 lesions in multiple sclerosis: a 13-year longitudinal study. Ann Neurol 2006; 60: 236–242.

103.

Dwyer

Bergsland

Ramasamy

, et al. Atrophied brain lesion volume: a new imaging biomarker in multiple sclerosis. J Neuroimaging 2018; 28: 490–495.

104.

Kappos

Moeri

Radue

, et al. Predictive value of gadolinium-enhanced magnetic resonance imaging for relapse rate and changes in disability or impairment in multiple sclerosis: a meta-analysis. Gadolinium MRI Meta-analysis Group. Lancet 1999; 353: 964–969.

105.

Radue

E-W

Barkhof

Kappos

, et al. Correlation between brain volume loss and clinical and MRI outcomes in multiple sclerosis. Neurology 2015; 84: 784–793.

106.

De Stefano

Arnold

. Towards a better understanding of pseudoatrophy in the brain of multiple sclerosis patients. Mult Scler 2015; 21: 675–676.

107.

Lavorgna

Bonavita

Ippolito

, et al. Clinical and magnetic resonance imaging predictors of disease progression in multiple sclerosis: a nine-year follow-up study. Mult Scler 2014; 20: 220–226.

108.

Lukas

Minneboo

de Groot

, et al. Early central atrophy rate predicts 5 year clinical outcome in multiple sclerosis. J Neurol Neurosurg Psychiatry 2010; 81: 1351–1356.

109.

Schoonheim

Hulst

Brandt

, et al. Thalamus structure and function determine severity of cognitive impairment in multiple sclerosis. Neurology 2015; 84: 776–783.

110.

Cole

Raffel

Friede

, et al. Longitudinal assessment of multiple sclerosis with the brain-age paradigm. Ann Neurol 2020; 88: 93–105.

111.

Rudick

Lee

J-C

Simon

, et al. Defining interferon beta response status in multiple sclerosis patients. Ann Neurol 2004; 56: 548–555.

112.

Río

Rovira

Tintoré

, et al. Relationship between MRI lesion activity and response to IFN-beta in relapsing-remitting multiple sclerosis patients. Mult Scler 2008; 14: 479–484.

113.

Prosperini

Gallo

Petsas

, et al. One-year MRI scan predicts clinical response to interferon beta in multiple sclerosis. Eur J Neurol 2009; 16: 1202–1209.

114.

Traboulsee

DKB

Cascione

, et al. Predictive value of early magnetic resonance imaging measures is differentially affected by the dose of interferon beta-1a given subcutaneously three times a week: an exploratory analysis of the PRISMS study. BMC Neurol 2018; 18: 68.

115.

Havrdova

Galetta

Hutchinson

, et al. Effect of natalizumab on clinical and radiological disease activity in multiple sclerosis: a retrospective analysis of the Natalizumab Safety and Efficacy in Relapsing-Remitting Multiple Sclerosis (AFFIRM) study. Lancet Neurol 2009; 8: 254–260.

116.

Giovannoni

Cook

Rammohan

, et al. Sustained disease-activity-free status in patients with relapsing-remitting multiple sclerosis treated with cladribine tablets in the CLARITY study: a post-hoc and subgroup analysis. Lancet Neurol 2011; 10: 329–337.

117.

Banwell

Giovannoni

Hawkes

, et al. Editors’ welcome and a working definition for a multiple sclerosis cure. Mult Scler Relat Disord 2013; 2: 65–67.

118.

Giovannoni

Turner

Gnanapavan

, et al. Is it time to target no evident disease activity (NEDA) in multiple sclerosis? Mult Scler Relat Disord 2015; 4: 329–333.

119.

Smith

Cohen

Hua

. Therapeutic targets for multiple sclerosis: current treatment goals and future directions. Neurotherapeutics 2017; 14: 952–960.

120.

Kappos

De Stefano

Freedman

, et al. Inclusion of brain volume loss in a revised measure of ‘no evidence of disease activity’ (NEDA-4) in relapsing-remitting multiple sclerosis. Mult Scler 2016; 22: 1297–1305.

121.

Giovannoni

. Disease-modifying treatments for early and advanced multiple sclerosis: a new treatment paradigm. Curr Opin Neurol 2018; 31: 233–243.

122.

Rotstein

Solomon

Sormani

, et al. Association of NEDA-4 with no long-term disability progression in multiple sclerosis and comparison with NEDA-3: a systematic review and meta-analysis. Neurol Neuroimmunol Neuroinflamm 2022; 9: e200032.

123.

Prosperini

Mancinelli

Haggiag

, et al. Minimal evidence of disease activity (MEDA) in relapsing-remitting multiple sclerosis. J Neurol Neurosurg Psychiatry 2020; 91: 271–277.

124.

Río

Castilló

Rovira

, et al. Measures in the first year of therapy predict the response to interferon beta in MS. Mult Scler 2009; 15: 848–853.

125.

Sormani

Signori

Stromillo

, et al. Refining response to treatment as defined by the Modified Rio Score. Mult Scler 2013; 19: 1246–1247.

126.

Río

Rovira

Tintoré

, et al. Disability progression markers over 6–12 years in interferon-β-treated multiple sclerosis patients. Mult Scler 2018; 24: 322–330.

127.

Sormani

Gasperini

Romeo

, et al. Assessing response to interferon-β in a multicenter dataset of patients with MS. Neurology 2016; 87: 134–140.

128.

Sormani

Truffinet

Thangavelu

, et al. Predicting long-term disability outcomes in patients with MS treated with teriflunomide in TEMSO. Neurol Neuroimmunol Neuroinflamm 2017; 4: e379.

129.

Sormani

Freedman

Aldridge

, et al. MAGNIMS score predicts long-term clinical disease activity-free status and confirmed disability progression in patients treated with subcutaneous interferon beta-1a. Mult Scler Relat Disord 2021; 49: 102790.

130.

Keenan

Gandhi

, et al. Prognostic value of the magnetic resonance imaging in MS Score for assessment of disease-modifying therapy in patients with relapsing multiple sclerosis treated with ponesimod. Paper presented at CMSC Annual Meeting 2022, 1–4 June, National Harbor, MD.

131.

Min

Spelman

Lugaresi

, et al. Silent lesions on MRI imaging – shifting goal posts for treatment decisions in multiple sclerosis. Mult Scler 2018; 24: 1569–1577.

132.

Giovannoni

Popescu

Wuerfel

, et al. Smouldering multiple sclerosis: the ‘real MS’. Ther Adv Neurol Disord 2022; 15: 17562864211066751.

133.

Kappos

Wolinsky

Giovannoni

, et al. Contribution of relapse-independent progression vs relapse-associated worsening to overall confirmed disability accumulation in typical relapsing multiple sclerosis in a pooled analysis of 2 randomized clinical trials. JAMA Neurol 2020; 77: 1132–1140.

134.

Krämer

Bar-Or

Turner

, et al. Bruton tyrosine kinase inhibitors for multiple sclerosis. Nat Rev Neurol 2023; 19: 289–304.

135.

Lublin

Häring

Ganjgahi

, et al. How patients with multiple sclerosis acquire disability. Brain 2022; 145: 3147–3161.

136.

Horáková

Boster

Bertolotto

, et al. Proportion of alemtuzumab-treated patients converting from relapsing-remitting multiple sclerosis to secondary progressive multiple sclerosis over 6 years. Mult Scler J Exp Transl Clin 2020; 6: 2055217320972137.

137.

University of California, San Francisco MS-EPIC Team, Cree

BAC

Hollenbach

, et al. Silent progression in disease activity-free relapsing multiple sclerosis. Ann Neurol 2019; 85: 653–666.

138.

Amato

Portaccio

. A first step towards preventive medicine in multiple sclerosis. Nat Rev Neurol 2023; 19: 134–135.

139.

Wiendl

Gold

Berger

, et al. Multiple Sclerosis Therapy Consensus Group (MSTCG): position statement on disease-modifying therapies for multiple sclerosis (white paper). Ther Adv Neurol Disord 2021; 14: 17562864211039648.

140.

Rotstein

Montalban

. Reaching an evidence-based prognosis for personalized treatment of multiple sclerosis. Nat Rev Neurol 2019; 15: 287–300.

141.

Hapfelmeier

Mühlau

, et al. Retrospective cohort study to devise a treatment decision score predicting adverse 24-month radiological activity in early multiple sclerosis. Ther Adv Neurol Disord 2023; 16: 17562864231161892.

142.

DKB

Held

Petkau

, et al. MRI T2 lesion burden in multiple sclerosis: a plateauing relationship with clinical disability. Neurology 2006; 66: 1384–1389.

143.

Mostert

Koch

Steen

, et al. T2 lesions and rate of progression of disability in multiple sclerosis. Eur J Neurol 2010; 17: 1471–1475.

144.

Rocca

Valsasina

Meani

, et al. Network damage predicts clinical worsening in multiple sclerosis: a 6.4-year study. Neurol Neuroimmunol Neuroinflamm 2021; 8: e1006.

145.

Kincses

Ropele

Jenkinson

, et al. Lesion probability mapping to explain clinical deficits and cognitive performance in multiple sclerosis. Mult Scler 2011; 17: 681–689.

146.

Filli

Hofstetter

Kuster

, et al. Spatiotemporal distribution of white matter lesions in relapsing-remitting and secondary progressive multiple sclerosis. Mult Scler 2012; 18: 1577–1584.

147.

Fleischer

Gonzalez-Escamilla

Pareto

, et al. Prognostic value of single-subject grey matter networks in early multiple sclerosis. Brain 2024; 147: 135-146.

148.

Confavreux

Vukusic

Adeleine

. Early clinical predictors and progression of irreversible disability in multiple sclerosis: an amnesic process. Brain 2003; 126: 770–782.

149.

Vukusic

Confavreux

. Natural history of multiple sclerosis: risk factors and prognostic indicators. Curr Opin Neurol 2007; 20: 269–274.

150.

Dallera

Affinito

Caliendo

, et al. The independent contribution of brain, spinal cord and gadolinium MRI in treatment decision in multiple sclerosis: a population-based retrospective study. Mult Scler Relat Disord 2023; 69: 104423.

151.

Scalfari

Neuhaus

Daumer

, et al. Age and disability accumulation in multiple sclerosis. Neurology 2011; 77: 1246–1252.

152.

Guillemin

Baumann

Epstein

, et al. Older age at multiple sclerosis onset is an independent factor of poor prognosis: a population-based cohort study. Neuroepidemiology 2017; 48: 179–187.

153.

Brown

Glasmacher

Kearns

PKA

, et al. Systematic review of prediction models in relapsing remitting multiple sclerosis. PLoS One 2020; 15: e0233575.

154.

Eriksson

Andersen

Runmarker

. Long-term follow up of patients with clinically isolated syndromes, relapsing-remitting and secondary progressive multiple sclerosis. Mult Scler 2003; 9: 260–274.

155.

Scalfari

Neuhaus

Degenhardt

, et al. The natural history of multiple sclerosis: a geographically based study 10: relapses and long-term disability. Brain 2010; 133: 1914–1929.

156.

Magliozzi

Cross

. Can CSF biomarkers predict future MS disease activity and severity? Mult Scler 2020; 26: 582–590.

157.

Kuhle

Kropshofer

Haering

, et al. Blood neurofilament light chain as a biomarker of MS disease activity and treatment response. Neurology 2019; 92: e1007–e1015.

158.

Meier

Willemse

EAJ

Schaedelin

, et al. Serum glial fibrillary acidic protein compared with neurofilament light chain as a biomarker for disease progression in multiple sclerosis. JAMA Neurol 2023; 80: 287–297.

159.

Ascherio

Munger

White

, et al. Vitamin D as an early predictor of multiple sclerosis activity and progression. JAMA Neurol 2014; 71: 306–314.

160.

Magyari

Sorensen

. Comorbidity in multiple sclerosis. Front Neurol 2020; 11: 851.

161.

Montalban

Gold

Thompson

, et al. ECTRIMS/EAN Guideline on the pharmacological treatment of people with multiple sclerosis. Mult Scler 2018; 24: 96–120.

162.

Rae-Grant

Day

Marrie

, et al. Practice guideline recommendations summary: disease-modifying therapies for adults with multiple sclerosis: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology 2018; 90: 777–788.

163.

Kantarci

Lebrun

Siva

, et al. Primary progressive multiple sclerosis evolving from radiologically isolated syndrome. Ann Neurol 2016; 79: 288–294.

164.

Okuda

Kantarci

Lebrun-Frénay

, et al. Dimethyl fumarate delays multiple sclerosis in radiologically isolated syndrome. Ann Neurol 2023; 93: 604–614.

165.

Lebrun-Frénay

Siva

Sormani

, et al. Teriflunomide and time to clinical multiple sclerosis in patients with radiologically isolated syndrome: the TERIS randomized clinical trial. JAMA Neurol 2023; 80: 1080–1088.

Prognostic relevance of MRI in early relapsing multiple sclerosis: ready to guide treatment decision making?

Abstract

Plain language summary

Keywords

Introduction

Current concept of MS pathology and MRI correlates

Prognostic value of MRI from the historical perspective

Prognostic value of brain and spinal cord MRI at disease onset

T2-hyperintense white matter lesions

T1-hypointense lesions

Gd+ lesions

Cortical grey matter lesions

Volumetric data

Paramagnetic rim lesions

Prognostic value of MRI during early follow-up

Prediction of treatment response

Inclusion of MRI in the treatment decision-making process

Need for early risk stratification in MS

Integrated interpretation of early MRI findings

Proposed approach to early prognosis

Footnotes

Acknowledgements

Declarations

ORCID iDs

References

Gd⁺ lesions