Multiple Sclerosis.pdf

The n e w e ng l a n d j o u r na l of m e dic i n e Review Article Dan L. Longo, M.D., Editor Multiple Sclerosis...

The n e w e ng l a n d j o u r na l of m e dic i n e Review Article Dan L. Longo, M.D., Editor Multiple Sclerosis Daniel S. Reich, M.D., Ph.D., Claudia F. Lucchinetti, M.D., and Peter A. Calabresi, M.D.​​ M ultiple sclerosis is the most prevalent chronic inflammatory From the Translational Neuroradiology disease of the central nervous system (CNS), affecting more than 2 million Section, National Institute of Neurologi­ cal Disorders and Stroke, National Insti­ people worldwide (at least 400,000 in the United States),1 and it is currently tutes of Health, Bethesda (D.S.R.), and the incurable. It is punctuated by fully or partially reversible episodes of neurologic Departments of Neurology and Neuro­ disability, usually lasting days or weeks. Typical syndromes at presentation include, science, Johns Hopkins School of Medi­ cine, Baltimore (P.A.C.) — both in Mary­ but are not limited to, monocular visual loss due to optic neuritis, limb weakness land; and the Department of Neurology, or sensory loss due to transverse myelitis, double vision due to brain-stem dysfunc- Mayo Clinic, Rochester, MN (C.F.L.). Ad­ tion, or ataxia due to a cerebellar lesion.2 After typically 10 to 20 years, a progres- dress reprint requests to Dr. Reich at the National Institute of Neurological Dis­ sive clinical course develops in many of the persons affected, eventually leading to orders and Stroke, 10 Center Dr., MSC impaired mobility and cognition; approximately 15% of patients have a progressive 1400, Bldg. 10, Rm. 5C103, Bethesda, MD course from onset. More than a dozen disease-modifying medications are available 20892, or at ­daniel​.­reich@​­nih​.­gov. to reduce the frequency of transient episodes of neurologic disability and limit the N Engl J Med 2018;378:169-80. accumulation of focal white-matter lesions on magnetic resonance imaging (MRI). DOI: 10.1056/NEJMra1401483 Copyright © 2018 Massachusetts Medical Society. No medication fully prevents or reverses the progressive neurologic deterioration, characterized most commonly by impaired ambulation, loss of bladder control, and slowed cognitive processing, but the question of whether disease-modifying medications can delay clinical progression is controversial.3-5 The annual economic cost of multiple sclerosis in the United States is approximately $10 billion.6 Pathol o gy The idea that multiple sclerosis is a disseminated plaque-like sclerosis was estab- lished approximately 150 years ago; indeed, the demonstration of dissemination — in space (disease-related changes in multiple regions of the CNS, including white matter, gray matter, brain stem, spinal cord, and optic nerve) (Fig. 1) and time — forms the cornerstone of diagnosis of the disease. Our understanding of the details of that pathology, and especially how it evolves over time, has been revolutionized with modern techniques such as immunohistochemical staining and MRI. Multiple sclerosis lesions can appear throughout the CNS and are most easily recognized in the white matter as focal areas of demyelination, inflammation, and glial reaction. Evidence from MRI and pathological assessment (biopsies and au- topsies) indicates that the earliest stages of white-matter demyelination (known as early active white-matter lesions) are heterogeneous7 and evolve over the course of months. Regardless of the particular immunologic pattern of early demyelination (Fig. 2), analysis of active lesions, over both time and space, suggests that a single immune-effector mechanism dominates in each person.8 Consistent with this notion are the observations that plasma exchange, which removes pathogenic antibodies from the circulation, ameliorates relapses that are refractory to initial treatment with glucocorticoids in patients whose active lesions contain immunoglobulin and complement9 and that cerebrospinal fluid (CSF) profiles differ according to lesion n engl j med 378;2 nejm.org January 11, 2018 169 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  The n e w e ng l a n d j o u r na l of m e dic i n e A B G C E F D Thalamus Pons Figure 1. Topography of Multiple Sclerosis Lesions. Shown is a schematic of lesion location, calling out imaging and pathological examples, in the periventricular white matter (inset A), subpial cortex (B), leptomeninges (C), thalamus and pons (D), spinal cord (E), optic nerve (F), and retina (G). Insets A, B, and D show a 7­tesla MRI of a 40­year­ old woman with relapsing–remitting multiple sclerosis, with similar pathological findings (in different patients) highlighted by immunohisto­ chemical staining directed against myelin proteolipid protein. Inset C shows a 3­tesla MRI after the administration of gadolinium in a 35­year­old woman with secondary progressive multiple sclerosis, with corresponding pathological findings in the meninges of a different patient (hematox­ ylin and eosin staining). Inset E shows a 3­tesla MRI of a 60­year­old woman with relapsing–remitting multiple sclerosis and corresponding path­ ological findings in a different patient (Luxol fast blue–periodic acid Schiff staining). Inset F shows a 3­tesla MRI of a 31­year­old woman with re­ lapsing–remitting multiple sclerosis and corresponding pathological findings in a different patient (anti–proteolipid protein immunohistochemical staining). Inset G shows a spectral­domain optical coherence tomographic reconstruction illustrating thinning of the peripapillary retinal nerve fiber layer. The normal range of retinal thickness is shown in green, and for this particular patient (black line), the retina is thinner than in 99% of control eyes. The bottom panel of the inset shows corresponding pathological findings in a different patient (immunohistochemical staining for Iba­1, a macrophage and microglial marker, with hematoxylin counterstaining). In all insets, lesions are indicated with arrowheads or circles. 170 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis Acute white-matter lesions — immunologic patterns Chronic white-matter lesions I Microglia Oligodendrocyte Macrophage T cell Smoldering II Complement Subpial cortical lesions ? Antibody Blood vessel T cell Apoptotic Chronic inactive Macrophage III oligodendrocyte B cell Microglia Region of demyelination Degeneration of inner myelin loops Remyelinated Figure 2. Lesions of the White Matter and Gray Matter. Early active white­matter demyelination falls into three major categories. The most common types (patterns I and II) show a background of mononuclear phagocytes with perivascular and parenchymal T­cell infiltration. Pattern II is further distinguished by immunoglobulin and complement deposition. In approximately 25% of biopsied active lesions (pattern III), oligodendrocyte apoptosis is accompanied by a “dying­back” oligodendrogliopathy, starting at the portion of myelin closest to the axon. These lesions resemble viral, toxic, and ische­ mic processes and can be destructive. After the acute phase, factors that remain poorly understood determine whether surviving axons in a lesion are invested by a thin myelin sheath (remyelinated), whether inflammation resolves without remyelination (chronic inactive), or whether inflammation and slow myelin degeneration persist (smoldering). Smoldering lesions are most common in progressive multiple sclerosis. The subpial cortical lesion, which is also more common in progressive multiple sclerosis, is characterized by demyelination of the superficial cortex, possibly in association with inflammation in the overlying leptomeninges and sparse macrophages and microglia at the border between demyelinated and myelinated neuropil. pattern.10 The identification of noninvasive bio- more effectively,11 a finding consistent with markers that correlate with active lesion pat- preclinical work indicating that age strongly terns will facilitate the design of personalized modulates immune-mediated regenerative pro- therapeutic strategies for multiple sclerosis, cesses.12,13 What remains unclear is whether since current treatment algorithms may not ade- lesions can remyelinate years after a smolder- quately address the underlying pathogenic hetero- ing lesion is established and whether remyelin- geneity of this complex disease. ated lesions have heightened susceptibility to What determines the long-term fate of a recurrent demyelination. High-resolution, ultra- given lesion — whether the inflammation re- high-field (7-tesla) MRI shows promise as a tool solves or “smolders” or whether it remyelinates for noninvasive staging of lesions,14 and it will — is not well understood. Recent data from be important for future studies to investigate longitudinal imaging studies suggest that le- the relationship between lesion outcomes and sions that form in younger people may repair clinical status. n engl j med 378;2 nejm.org January 11, 2018 171 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  The n e w e ng l a n d j o u r na l of m e dic i n e Myelin is not exclusive to white matter, and derived from retinal ganglion cells and which demyelination in multiple sclerosis also involves succumb to a dying-back process after retrobul- gray matter.15-17 Approximately half of cortical bar inflammatory demyelination in acute optic lesions are perivascular. In some cortical lesions, neuritis. Studies have clearly shown concomitant the inflamed vessel may be located near the retinal ganglion-cell loss27 even in the absence of leukocortical junction, in which case demyelin- clinical optic neuritis, presumably reflecting ei- ation also affects the juxtacortical white matter. ther subclinical inflammation of the optic nerve Sometimes, a small penetrating cortical vein is or retrograde transsynaptic degeneration. involved and only central cortical layers are af- fected. Cortical lesions are less inflammatory Epidemiol o gy than their white-matter counterparts and have substantially less permeability of the blood–brain It is not known whether multiple sclerosis has a barrier.18 single or multiple causes, and rarely (if ever) has The remaining cortical lesions do not arise a specific etiologic trigger been identified. None- from a single cortical vessel but rather appear to theless, various genetic and environmental risk proceed inward from the pial surface of the factors have been found (Fig. 3).28 For unknown brain. In autopsies conducted after decades of reasons, approximately three quarters of people disease, most such lesions are found to be inac- with multiple sclerosis are women, as is common tive, in contrast to subpial lesions in early mul- in diseases that are considered autoimmune. tiple sclerosis, which are inflammatory and to- People with an affected first-degree relative have pographically associated with diffuse and focal a 2 to 4% risk of multiple sclerosis (as compared leptomeningeal inflammatory aggregates (espe- with approximately 0.1% risk in the general cially when observed in biopsy specimens).17 Sub- population), and concordance in monozygotic pial lesions can be extensive and are often found twins is 30 to 50%. Genomewide association on flanking cortical banks within a sulcus, studies, based on samples assembled from thou- which strongly suggests a leptomeningeal origin. sands of patients with multiple sclerosis and Leptomeningeal inflammation can organize into matched controls, have identified more than 200 self-sustaining structures akin to tertiary lym- gene variants that raise the risk of the disease, phoid follicles.19 Although findings on MRI sup- of which the most significant remains the HLA port an association between leptomeningeal in- DRB1*1501 haplotype (with an odds ratio of ap- flammation and subpial cortical demyelination,20 proximately 3). Most risk alleles are associated robust detection methods are lacking, and the with immune-pathway genes, a finding consis- natural history of such lesions — and their re- tent with the notion that autoimmune mecha- sponsiveness to therapy — remain unknown. nisms are paramount in the development of Spinal cord lesions are a major source of clinical multiple sclerosis. We are currently un- clinical disability. Perivascular and circumferen- aware of any validated genetic risk factor that tial demyelination is often highly inflammatory strongly influences the clinical course of the and can involve gray matter.21 Spinal cord atro- disease; this limitation reflects the difficulty phy results from focal inflammatory demyelin- of measuring disease severity in a disease that ation and remote neuroaxonal degeneration.22 It evolves over a period of decades. is detectable by MRI, and the cross-sectional Major environmental risk factors include geo- area of the spinal cord is therefore a promising graphic latitude (with a higher incidence in more outcome measure for clinical trials.23,24 temperate climates), which may reflect seasonal As part of the CNS, the optic nerve is also a changes in sunlight exposure influencing vita- major target in multiple sclerosis, and loss of the min D levels or pathogens prevalent in these contiguous retinal ganglion cells is well docu- regions, although a genetic contribution is pos- mented.25 Retinal damage can be assessed in sible as well. Tobacco exposure, obesity, and vivo by means of optical coherence tomogra- mononucleosis are also associated with an en- phy,26 which reveals substantial thinning of the hanced risk of multiple sclerosis. Mononucleosis retinal nerve-fiber and ganglion-cell layers de- results from infection with Epstein–Barr virus in spite their lack of myelin. Thinning results from the postpubertal population, and multiple scle- injury to axons in the optic nerve, which are rosis eventually develops in only a minority of 172 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis people with a history of mononucleosis (and a tiny minority of all those infected with the nearly EBV and mononucleosis ubiquitous Epstein–Barr virus). Viruses other Other viruses than Epstein–Barr virus have been suggested as Risk genes Temperate latitude potential causes of multiple sclerosis or of mul- Fibrinogen tiple sclerosis–related disease activity, but none Toxins have been definitively proved. Some of these vi- Trauma ruses may act as molecular mimics, whereas Low vitamin D others may interfere with mechanisms that nor- Smoking Obesity mally limit self-reactive cells. Differential sus- Early adulthood ceptibility is reflected in the mouse model of Female sex multiple sclerosis, experimental autoimmune en- cephalomyelitis (EAE), such that specific myelin antigens are required to induce EAE in different strains of mice.29 Along these lines, an interest- ing set of experiments showed that components of the intestinal microbiome can also strongly influence the propensity for the development of Demyelination Axonal loss EAE, especially in genetically predisposed mouse strains with transgenes for myelin recognition by B cells and T cells,30 and evidence for a similar phenomenon in patients with multiple sclerosis Low inflammation, High or chronic inflammation, many spinal cord and cortical lesions, is beginning to emerge.31,32 Overall, the mecha- few spinal cord lesions, poor endogenous repair, good endogenous repair, VS. nisms by which genetic polymorphisms and preserved axons and synapses, mitochondrial dysfunction, extensive axon and synapse loss, environmental exposures raise the risk of mul- early treatment, delayed treatment, younger age tiple sclerosis remain the subject of intense older age investigation. Low Intermediate High Patho gene sis Chance of progression Tissue damage in multiple sclerosis results from a complex and dynamic interplay between the Figure 3. Risk Factors, Triggers, Modifiers, and Disease Courses. immune system, glia (myelin-making oligoden- It is unlikely that multiple sclerosis will ultimately be attributed to a single drocytes and their precursors, microglia, and cause. Rather, the genetic and environmental factor or combination of fac­ tors that result in a predisposition to multiple sclerosis, initiate the disease, astrocytes), and neurons (Fig. 4). Although there and modify its course are highly diverse from one person to the next. The is debate about whether the root cause of mul- top portion of the figure shows the funneling of proposed factors, for which tiple sclerosis is intrinsic or extrinsic to the CNS, varying levels of evidence exist, into the development of inflammatory, de­ studies in animal models, particularly EAE in myelinating lesions with heterogeneous axonal loss (middle portion). The mice and marmosets, together with analysis of bottom portion of the figure lists features of the lesions and their conse­ quences that are generally salutary or deleterious and that modify the risk immune cells and their products in CSF and of progression. EBV denotes Epstein–Barr virus. blood of humans, have disclosed a critical role for adaptive immunity.29 However, despite the fact that some of the disease-modifying thera- the innate immune system (as described below). pies that were first shown to ameliorate EAE Moreover, although some animal models have eventually reached clinical practice, differences clinical progression, none recapitulate the spec- between EAE and multiple sclerosis are myriad trum of critical pathologic features of multiple and have a variety of causes, including the genetic sclerosis.33 Genetic data suggest that the patho- and environmental heterogeneity of humans genesis of multiple sclerosis shares important relative to laboratory mouse strains, as well as a features with a variety of non-CNS autoimmune complex immune process in multiple sclerosis diseases.34 that clearly involves T cells (the major driver of Both helper (CD4+) and cytotoxic (CD8+) T cells EAE) as well as B cells, antibodies, and cells of have been described in multiple sclerosis lesions: n engl j med 378;2 nejm.org January 11, 2018 173 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  The n e w e ng l a n d j o u r na l of m e dic i n e Periphery Cortex Alemtuzumab Daclizumab Oligodendrocyte Dimethyl fumarate precursor cell Fingolimod Glatiramer acetate Interferon beta Lesion Mitoxantrone CD8+ Ocrelizumab T cell Microglia Teriflunomide Glucocorticoids Dalfampridine Complement Oligodendrocyte Neuroprotection Na+ channel Complement Mitochondrial K+ Astrocyte abnormalities channel Macrophage Blood Ocrelizumab vessel CD4+ T cell T cell Natalizumab B cell Figure 4. Cells, Molecules, and Therapies. Shown is a simplified schematic depiction of major cell types within white­matter multiple sclerosis lesions, along with several current and promising therapeutic targets in the central nervous system and in the periphery. CD4+ T cells are more concentrated in the peri- response to B-cell depletion (as early as 8 to 12 vascular cuff, whereas CD8+ T cells are widely weeks), even before the reduction of circulating distributed within the parenchyma.35 Drugs that immunoglobulin, it seems more likely that other limit T-cell access to the CNS can reduce or functions of B cells, including antigen presenta- eliminate new multiple sclerosis lesions. How- tion to helper T cells and cytokine production, are ever, T cells that are reactive to myelin antigens more relevant. have been observed in similar proportions in Cells of the innate immune system are espe- people with and people without multiple sclero- cially important in the pathogenesis of multiple sis, which suggests either that these cells are sclerosis.37 Bloodborne macrophages infiltrate ac- dysfunctional in multiple sclerosis or that other tive multiple sclerosis lesions and remove myelin immune factors also play critical roles. debris and inflammatory by-products; classically Because of the dramatic success of B-cell– and alternatively activated macrophages, as well depleting antibodies in limiting multiple sclero- as mixed populations, have been described in sis lesion formation and clinical disease activity, these lesions. Microglia, the primary endogenous there is renewed attention on the role of B cells.36 phagocytes of the CNS, are abundant in multiple It has long been known that the CSF of most sclerosis lesions, but whether their role is patho- patients with multiple sclerosis harbors unique genic or protective — or both — remains uncer- antibodies (oligoclonal bands) that are produced tain.38 Microglial activation, often remote from within the CNS. There is evidence that the anti- established lesions, has been found in the white body-producing function of B-lineage cells is matter of autopsy specimens from patients with important in some multiple sclerosis lesions.7 multiple sclerosis39 and may represent the earli- However, because of the rapidity of the clinical est stage of lesion development (as is the case in 174 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis animal models40). Once activated, microglia and up glutamate, providing metabolic support to macrophages are pathologically indistinguish- axons, and maintaining the blood–brain barrier.49 able, but progress with the use of gene-expres- An underemphasized but surprisingly common sion technology has opened the door to unraveling cell (approximately 5% of all CNS cells) is the their separate contributions, potentially enabling oligodendrocyte precursor cell, which expresses the development of targeted therapy.41 Studies in the proteoglycan NG2.50 Oligodendrocyte pre- animals have suggested that monocyte and mac- cursor cells can differentiate into oligodendro- rophage populations strongly influence myelin cytes and are present even late in life,51 but in regeneration.13,42 patients with multiple sclerosis they are often Disturbance in the blood–brain barrier is an arrested at the plaque edge, or they may differ- important step in the development of white- entiate into premyelinating oligodendrocytes but matter lesions, which show evidence of gado- fail to wrap myelin.52 Thus, promoting oligoden- linium extravasation on MRI early in their devel- drocyte precursor-cell differentiation is an attrac- opment. Abnormal vascular permeability precedes tive strategy to enhance endogenous remyelin- inflammatory demyelination in EAE40 and po- ation, but such a strategy must be balanced tentially in multiple sclerosis.43 Studies in mice against the potential of oligodendrocyte precur- have shown that leakage of a key plasma protein sor cells to respond to cytokines and thereby (fibrinogen),44 or even secretion of a bacterial participate in inflammation themselves.53,54 Fur- toxin,45 can trigger inflammatory demyelination thermore, oligodendrocytes may become dysfunc- by a cascade that involves microglial activation tional even without dying, causing tissue dam- and subsequent adaptive immunity. In early mul- age through loss of trophic support to axons; tiple sclerosis lesions, vessels near the lesion whether such dysfunctional oligodendrocytes can center become permeable to gadolinium, which participate in repair is unclear. then diffuses passively into enlarged interstitial spaces; days later, the central breach in the Axon Biology blood–brain barrier begins to repair, while small Although relative axonal sparing in the face of capillaries at the lesion edge become permeable profound demyelination is a hallmark of multi- — perhaps as part of the early wound-healing ple sclerosis pathology, axonal transections are process.46 Leptomeningeal inflammation can also frequent, especially acutely.55 Studies with two- contribute to vascular permeability, but this ap- photon microscopy in animal models have be- pears to be a chronic process.20 gun to elucidate relevant cellular and molecular processes, some of which are potentially revers- Glial-Cell Biology ible.56 In chronically demyelinated lesions, denud- Acute multiple sclerosis plaques show activation of ed axons remain vulnerable and can degenerate astrocytes and microglia and sometimes caspase- slowly; possible mechanisms include impaired independent oligodendrocyte apoptosis.7 Microg- axonal transport, mitochondrial dysfunction, and lia are prominent in white-matter lesions but are increased energy demands related to the up- less activated in gray matter.18 Importantly, mi- regulation of ion channels.57 Adaptive immunity croglia play dual roles, sometimes mediating — which is critical for the formation of new inflammation but in other circumstances pro- white-matter lesions — is much less prominent moting repair through clearance of myelin de- in the slow neurodegeneration of progressive bris.47 In gray matter, microglia may limit dam- multiple sclerosis, which highlights the impor- age through pruning of dysfunctional synapses tance of glial activation and secondary mecha- that express classical complement cascade pro- nisms of injury. teins (C1q and C3). This pruning process may become pathologic if activated astrocytes promote Biom a r k er s aberrant expression of complement at synapses, thereby accelerating degeneration.48 Since astro- Magnetic Resonance Imaging cytes are a major component of the multiple The slow rate of disease progression in time sclerosis plaque, they are well positioned to en- frames that are relevant for clinical monitoring hance inflammation by releasing effector mole- or clinical trials, together with heterogeneous cules, but they may also limit damage by taking pathogenic mechanisms and the impracticality n engl j med 378;2 nejm.org January 11, 2018 175 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  The n e w e ng l a n d j o u r na l of m e dic i n e of directly sampling CNS tissue (as opposed to not shown a strong correlation with clinical blood or CSF), have limited the development of status at the population level, probably because biomarkers for progressive multiple sclerosis. The of the heterogeneous presentation and course of most important diagnostic and prognostic tech- multiple sclerosis and the inherent variability of nique for assessing multiple sclerosis — particu- clinical measures, there has been a trend toward larly early in the disease course — is MRI, which the use of imaging to investigate multiple scle- is currently the only technique that can interro- rosis pathology and pathogenesis, including peri- gate the entire CNS in vivo. vascular inflammation, the development of cor- Inflammatory demyelination is easily visible on tical and spinal cord lesions, myelin loss and MRI, as are changes in the blood–brain barrier regeneration, innate immune activation, lepto- that accompany its early development. Figure 1 meningeal inflammation, and network func- shows the in vivo appearance on MRI of lesions tion.14 Such research has been facilitated by the in the periventricular white matter, thalamus and advent of 7-tesla MRI and, to a lesser extent, brain stem, spinal cord, and optic nerve. Since molecular tracers detectable by positron-emission 2000, MRI has been the key diagnostic test when tomography. A particularly exciting innovation patients present with a clinical syndrome that is has been the use of optical coherence tomography suggestive of multiple sclerosis, and the most to rapidly assess the retina at micron-level reso- recent criteria58 — when applied carefully59 — lution. Retinal ganglion-cell axon loss results in allow for accurate diagnosis with a single scan. easily detectable retinal thinning, which tracks MRI diagnostic criteria are revised as new data with MRI changes in the brain74 and can predict accumulate, and standardized protocols for rou- the evolution of disability at the cohort level.75 tine use have been proposed.60,61 MRI is also critical in the development of new disease-modi- Blood and CSF fying therapies, because new lesions are an order Clonal expansion of immunoglobulin-secreting of magnitude more frequent than clinical re- B cells and plasma cells in the CNS results in the lapses.62 Indeed, the effect of a therapy on the characteristic finding of CSF-specific oligoclonal formation new lesions, as detected by MRI, in bands.76 Although the targets of these immuno- small proof-of-concept studies strongly predicts globulins are probably multifaceted, their pres- the effect of the therapy on rates of relapse in ence implies a CNS-restricted immune response. definitive trials.63 Furthermore, MRI findings that However, the specificity of oligoclonal bands for are consistent with multiple sclerosis have been multiple sclerosis is poor, and infections can observed in healthy people who underwent scan- cause the same pattern. Currently, no externally ning for other purposes (such as research), and validated blood immune marker has adequate clinical multiple sclerosis develops in up to 50% sensitivity and specificity to be used for the di- of people with this so-called radiologically isolat- agnosis of multiple sclerosis, which probably re- ed syndrome, sometimes with a primary progres- flects the genetic and environmental heterogene- sive course.64,65 ity of the disease. CSF and serum neurofilament Neurodegeneration in multiple sclerosis is best light chains are promising in their ability to re- captured on MRI by measuring the size of the flect axonal pathologic processes in the CNS at brain or spinal cord. An abnormally low brain the cohort level,77 and there is ongoing interest parenchymal fraction — a measure of brain size in various types of noncoding RNA molecules relative to intracranial capacity — can be taken that can affect gene expression.78 The extent to as surrogate evidence of previous disease-related which these approaches will be useful in pa- atrophy of the brain. In cohort studies, CNS at- tients remains unclear. rophy has been documented even before clinical presentation.66,67 Atrophy complements lesion- Ther a pie s based biomarkers,68 and proof-of-concept clini- cal trials using atrophy as the primary outcome As of December 2017, the Food and Drug Ad- have been published.69,70 Studies of CNS atrophy ministration has approved 15 medications for have focused on specific gray-matter structures modifying the course of multiple sclerosis: (the neocortex and thalamus).71-73 5 preparations of interferon beta; 2 preparations Because conventional MRI biomarkers have of glatiramer acetate; the monoclonal antibodies 176 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis natalizumab, alemtuzumab, daclizumab, and specific inhibition, clonal deletion, or induction ocrelizumab (the first B-cell–targeted therapy); of immunotolerance. Previous attempts at target- the chemotherapeutic agent mitoxantrone; and the ing cytokines have been unsuccessful85 or even small-molecule oral agents fingolimod, dimethyl deleterious,86 probably because of an incomplete fumarate, and teriflunomide. Dalfampridine has understanding of the roles of different forms of been approved as a symptomatic therapy to im- cytokines and their receptors, as well as com- prove walking speed. It is beyond the scope of pensatory pathways. The innate immune system this article to discuss the relative benefits, risks, has not been specifically targeted in large-scale modes of action, and routes of administration of trials of treatment for multiple sclerosis, and these various medications (although some tar- given the high likelihood that this system can be gets are shown in Fig. 4), except to say that all both protective and deleterious, such efforts must are approved for relapsing–remitting multiple be approached cautiously. Nonetheless, the ubiqui- sclerosis and reduce, to various extents, the like- ty of innate immune cells in and around mul- lihood of the development of new white-matter tiple sclerosis lesions underscores the need for lesions, clinical relapses, and stepwise accumu- further research. lation of disability. On the basis of the ability of Beyond the immune system, a great deal of several of these medications to delay a formal work has revolved around tissue repair and pro- diagnosis of multiple sclerosis after an initial tection. On the repair side, small studies have attack, there has been a general move toward preliminarily reported mixed results for therapies early treatment, although, as discussed above, that promote endogenous remyelination through the long-term value of this approach with re- various pathways.87 On the basis of preclinical spect to preventing progressive multiple sclero- data, including in vitro screens and testing in sis remains uncertain. The recent approval of models such as EAE, several approved drugs ocrelizumab for primary progressive multiple (targeting, e.g., nuclear hormone receptor, hista- sclerosis is a promising step, but the reasons for minic, cholinergic [muscarinic], and adrenergic the ability of ocrelizumab to slow progression79 pathways) are being tested for remyelination or remain uncertain. Another important trend has myelin protection. Transplantation of neural or been to escalate treatment with a target of “no oligodendrocyte precursor cells into the brain evidence of disease activity,” as evidenced by the is effective in animal models, but well-designed absence of new lesions, relapses, disability pro- clinical trials involving patients with multiple gression and, more recently, tissue atrophy80,81; sclerosis have not been undertaken, and it is however, it is doubtful that multiple sclerosis likely that promotion of endogenous remyelin- can be fully arrested with current therapies. ation will prove more fruitful and feasible, espe- Several incipient multicenter studies will com- cially if the inhibitory factors inherent in the pare early intensive treatment with more conven- multiple sclerosis plaque can be overcome.52 A tional treatment-escalation approaches. challenge for remyelination trials is the lack of a Small-scale studies have shown that immuno- robust, easily deployable biomarker of success. ablation followed by autologous hematopoietic Visual evoked potentials have been used in small stem-cell transplantation may be a highly dura- studies, but standardization is difficult and tech- ble and effective — and increasingly safe — nical variability high. The specificity of high- therapy.82 The favorable side-effect profile and resolution imaging-based markers for myelin re- high efficacy of B-cell–inhibiting therapies is generation remains questionable. Nevertheless, likewise a welcome development, although op- MRI is highly sensitive to changes in myelin, portunistic infections can occur in rare cases, and such sensitivity can be exploited in early and postmarketing studies will need to monitor proof-of-concept trials.88 long-term side effects. There are early-stage ef- Axonal protection is actively being examined. forts to interfere with specific T-cell populations Results from initial clinical trials of a wide vari- that are thought to drive multiple sclerosis, stem- ety of drugs have been published or reported, ming from data indicating that certain key sub- with several medium-to-large studies currently sets of helper T cells, including those that ex- under way.89 There is an emerging consensus press both interferon gamma and interleukin-17, that slowing the rate of cerebral or spinal cord are important.83,84 Such approaches may involve atrophy is a feasible goal, which, at the proof-of- n engl j med 378;2 nejm.org January 11, 2018 177 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  The n e w e ng l a n d j o u r na l of m e dic i n e concept stage, can be undertaken in several have sometimes strayed too far from the caus- hundred people over a period of a few years.90 ative biology. The richest conception of multiple However, definitive proof of neuroprotection — sclerosis will allow appreciation of common pa- an elusive goal in many neurologic conditions thology, which, in the context of variable triggers — awaits larger studies with clinical end points. and clinical courses, makes multiple sclerosis among the most remarkable of all neurologic disorders. C onclusions a nd F u t ur e Dir ec t ions Dr. Reich reports having cooperative research and develop- ment agreements with Vertex Pharmaceuticals, holding a patent Meaningful advances in basic immunology, my- (US9607392) on a system and method of automatically detect- ing tissue abnormalities, and having a pending patent (PCT/ elin biology, and neuroscience, together with a US2013/033334) on a method of analyzing multisequence MRI global focus on halting progressive accumula- data for analyzing brain abnormalities in a patient; Dr. Lucchinetti, tion of disability,91 have opened the promise of a receiving grant support from Novartis, Sanofi-Synthelabo, Biogen, Mallinkrodt, and Alexion; and Dr. Calabresi, receiving grant multipronged understanding of, and therapeutic support, paid to his institution, and consulting fees from Biogen attack on, multiple sclerosis. At the same time, Idec, grant support paid to his institution from Novartis, Med- a renewed focus on lesion development and re- Immune, Teva, and Annexon, and consulting fees from AbbVie, Merck, Vaccinex, Vertex, and Disarm Therapeutics. No other po- pair — more broadly conceived to include lesions tential conflict of interest relevant to this article was reported. in white matter, gray matter, and leptomeninges Disclosure forms provided by the authors are available with — should ultimately unify lines of research, the full text of this article at NEJM.org. We thank Ms. Erina He for drafting earlier versions of the particularly on the side of fluid and imaging- figures and Drs. Martina Absinta, Carlos Pardo, and Yong Guo related biomarkers and clinical outcomes, which for providing radiologic and histopathological images. References 1. GBD 2015 Neurological Disorders cal changes in multiple sclerosis and re- cortical multiple sclerosis lesions. Ann Collaborator Group. Global, regional, and sponse to therapeutic plasma exchange. Neurol 2001;​50:​389-400. national burden of neurological disorders Lancet 2005;​366:​579-82. 19. Serafini B, Rosicarelli B, Magliozzi R, during 1990-2015: a systematic analysis 10. Jarius S, König FB, Metz I, et al. Pat- Stigliano E, Aloisi F. Detection of ectopic for the Global Burden of Disease Study tern II and pattern III MS are entities dis- B-cell follicles with germinal centers in 2015. Lancet Neurol 2017;​16:​877-97. tinct from pattern I MS: evidence from the meninges of patients with secondary 2. Brownlee WJ, Hardy TA, Fazekas F, cerebrospinal fluid analysis. J Neuroin- progressive multiple sclerosis. Brain Pathol Miller DH. Diagnosis of multiple sclero- flammation 2017;​14:​171. 2004;​14:​164-74. sis: progress and challenges. Lancet 2017;​ 11. Absinta M, Sati P, Schindler M, et al. 20. Absinta M, Vuolo L, Rao A, et al. Gad- 389:​1336-46. Persistent 7-tesla phase rim predicts poor olinium-based MRI characterization of 3. Zhang T, Shirani A, Zhao Y, et al. Beta- outcome in new multiple sclerosis patient leptomeningeal inflammation in multiple interferon exposure and onset of second- lesions. J Clin Invest 2016;​126:​2597-609. sclerosis. Neurology 2015;​85:​18-28. ary progressive multiple sclerosis. Eur J 12. Rawji KS, Mishra MK, Yong VW. Re- 21. Gilmore CP, Geurts JJG, Evangelou N, Neurol 2015;​22:​990-1000. generative capacity of macrophages for et al. Spinal cord grey matter lesions in 4. Signori A, Gallo F, Bovis F, Di Tullio remyelination. Front Cell Dev Biol 2016;​4:​ multiple sclerosis detected by post-mortem N, Maietta I, Sormani MP. Long-term im- 47. high field MR imaging. Mult Scler 2009;​ pact of interferon or glatiramer acetate in 13. Ruckh JM, Zhao J-W, Shadrach JL, et al. 15:​180-8. multiple sclerosis: a systematic review Rejuvenation of regeneration in the aging 22. DeLuca GC, Williams K, Evangelou N, and meta-analysis. Mult Scler Relat Dis- central nervous system. Cell Stem Cell Ebers GC, Esiri MM. The contribution of ord 2016;​6:​57-63. 2012;​10:​96-103. demyelination to axonal loss in multiple 5. Cree BA, Gourraud PA, Oksenberg JR, 14. Absinta M, Sati P, Reich DS. Advanced sclerosis. Brain 2006;​129:​1507-16. et al. Long-term evolution of multiple MRI and staging of multiple sclerosis le- 23. Liu W, Nair G, Vuolo L, et al. In vivo sclerosis disability in the treatment era. sions. Nat Rev Neurol 2016;​12:​358-68. imaging of spinal cord atrophy in neuro- Ann Neurol 2016;​80:​499-510. 15. Bø L, Vedeler CA, Nyland HI, Trapp inflammatory diseases. Ann Neurol 2014;​ 6. Adelman G, Rane SG, Villa KF. The BD, Mørk SJ. Subpial demyelination in the 76:​370-8. cost burden of multiple sclerosis in the cerebral cortex of multiple sclerosis pa- 24. Kearney H, Miller DH, Ciccarelli O. United States: a systematic review of the tients. J Neuropathol Exp Neurol 2003;​62:​ Spinal cord MRI in multiple sclerosis — literature. J Med Econ 2013;​16:​639-47. 723-32. diagnostic, prognostic and clinical value. 7. Lucchinetti C, Brück W, Parisi J, 16. Kutzelnigg A, Lucchinetti CF, Stadel- Nat Rev Neurol 2015;​11:​327-38. Scheithauer B, Rodriguez M, Lassmann H. mann C, et al. Cortical demyelination and 25. Green AJ, McQuaid S, Hauser SL, Allen Heterogeneity of multiple sclerosis lesions: diffuse white matter injury in multiple IV, Lyness R. Ocular pathology in multi- implications for the pathogenesis of de- sclerosis. Brain 2005;​128:​2705-12. ple sclerosis: retinal atrophy and inflam- myelination. Ann Neurol 2000;​47:​707-17. 17. Lucchinetti CF, Popescu BFG, Bunyan mation irrespective of disease duration. 8. Metz I, Weigand SD, Popescu BFG, RF, et al. Inflammatory cortical demye- Brain 2010;​133:​1591-601. et al. Pathologic heterogeneity persists in lination in early multiple sclerosis. N Engl 26. Petzold A, Balcer LJ, Calabresi PA, et early active multiple sclerosis lesions. J Med 2011;​365:​2188-97. al. Retinal layer segmentation in multiple Ann Neurol 2014;​75:​728-38. 18. Peterson JW, Bö L, Mörk S, Chang A, sclerosis: a systematic review and meta- 9. Keegan M, König F, McClelland R, et Trapp BD. Transected neurites, apoptotic analysis. Lancet Neurol 2017;​16:​797-812. al. Relation between humoral pathologi- neurons, and reduced inflammation in 27. Syc SB, Saidha S, Newsome SD, et al. 178 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis Optical coherence tomography segmenta- 44. Ryu JK, Petersen MA, Murray SG, et al. lines for the diagnosis and follow-up of tion reveals ganglion cell layer pathology Blood coagulation protein fibrinogen pro- multiple sclerosis. AJNR Am J Neuroradiol after optic neuritis. Brain 2012;​135:​521-33. motes autoimmunity and demyelination 2016;​37:​394-401. 28. Ascherio A, Munger KL. Epidemiology via chemokine release and antigen pre- 61. Rovira À, Wattjes MP, Tintoré M, et al. of multiple sclerosis: from risk factors to sentation. Nat Commun 2015;​6:​8164. Evidence-based guidelines: MAGNIMS con- prevention — an update. Semin Neurol 45. Linden JR, Ma Y, Zhao B, et al. Clos- sensus guidelines on the use of MRI in 2016;​36:​103-14. tridium perfringens epsilon toxin causes multiple sclerosis-clinical implementation 29. Kipp M, van der Star B, Vogel DYS, selective death of mature oligodendro- in the diagnostic process. Nat Rev Neurol et al. Experimental in vivo and in vitro cytes and central nervous system demye- 2015;​11:​471-82. models of multiple sclerosis: EAE and be- lination. MBio 2015;​6(3):​e02513. 62. Harris JO, Frank JA, Patronas N, yond. Mult Scler Relat Disord 2012;​1:​15-28. 46. Gaitán MI, Shea CD, Evangelou IE, ­McFarlin DE, McFarland HF. Serial gado- 30. Berer K, Mues M, Koutrolos M, et al. et al. Evolution of the blood-brain barrier linium-enhanced magnetic resonance im- Commensal microbiota and myelin auto- in newly forming multiple sclerosis lesions. aging scans in patients with early, relapsing- antigen cooperate to trigger autoimmune Ann Neurol 2011;​70:​22-9. remitting multiple sclerosis: implications demyelination. Nature 2011;​479:​538-41. 47. Aguzzi A, Barres BA, Bennett ML. Mi- for clinical trials and natural history. Ann 31. Cekanaviciute E, Yoo BB, Runia TF, croglia: scapegoat, saboteur, or something Neurol 1991;​29:​548-55. et al. Gut bacteria from multiple sclerosis else? Science 2013;​339:​156-61. 63. Sormani MP, Bruzzi P. MRI lesions as patients modulate human T cells and ex- 48. Liddelow SA, Guttenplan KA, Clarke a surrogate for relapses in multiple sclero- acerbate symptoms in mouse models. Proc LE, et al. Neurotoxic reactive astrocytes sis: a meta-analysis of randomised trials. Natl Acad Sci U S A 2017;​114:​10713-8. are induced by activated microglia. Nature Lancet Neurol 2013;​12:​669-76. 32. Berer K, Gerdes LA, Cekanaviciute E, 2017;​541:​481-7. 64. Okuda DT, Siva A, Kantarci O, et al. et al. Gut microbiota from multiple scle- 49. Ludwin SK, Rao VTs, Moore CS, Antel Radiologically isolated syndrome: 5-year rosis patients enables spontaneous auto- JP. Astrocytes in multiple sclerosis. Mult risk for an initial clinical event. PLoS One immune encephalomyelitis in mice. Proc Scler 2016;​22:​1114-24. 2014;​9(3):​e90509. Natl Acad Sci U S A 2017;​114:​10719-24 50. Bergles DE, Richardson WD. Oligo- 65. Kantarci OH, Lebrun C, Siva A, et al. 33. Lassmann H, van Horssen J, Mahad D. dendrocyte development and plasticity. Primary progressive multiple sclerosis Progressive multiple sclerosis: pathology Cold Spring Harb Perspect Biol 2015;​8(2):​ evolving from radiologically isolated syn- and pathogenesis. Nat Rev Neurol 2012;​8:​ a020453. drome. Ann Neurol 2016;​79:​288-94. 647-56. 51. Chang A, Nishiyama A, Peterson J, 66. De Stefano N, Giorgio A, Battaglini M, 34. Farh KK-H, Marson A, Zhu J, et al. Prineas J, Trapp BD. NG2-positive oligo- et al. Assessing brain atrophy rates in a Genetic and epigenetic fine mapping of dendrocyte progenitor cells in adult hu- large population of untreated multiple causal autoimmune disease variants. Na- man brain and multiple sclerosis lesions. sclerosis subtypes. Neurology 2010;​ 74:​ ture 2015;​518:​337-43. J Neurosci 2000;​20:​6404-12. 1868-76. 35. Lassmann H. Mechanisms of white 52. Franklin RJM, Goldman SA. Glia dis- 67. Azevedo CJ, Overton E, Khadka S, et al. matter damage in multiple sclerosis. Glia ease and repair — remyelination. Cold Early CNS neurodegeneration in radiologi- 2014;​62:​1816-30. Spring Harb Perspect Biol 2015;​ 7(7):​ cally isolated syndrome. Neurol Neuro- 36. Michel L, Touil H, Pikor NB, Gommer- a020594. immunol Neuroinflamm 2015;​2(3):​e102. man JL, Prat A, Bar-Or A. B cells in the 53. Dimou L, Gallo V. NG2-glia and their 68. Sormani MP, Arnold DL, De Stefano N. multiple sclerosis central nervous system: functions in the central nervous system. Treatment effect on brain atrophy corre- trafficking and contribution to CNS-com- Glia 2015;​63:​1429-51. lates with treatment effect on disability in partmentalized inflammation. Front Im- 54. Kitic M, Karram K, Israel N, et al. multiple sclerosis. Ann Neurol 2014;​75:​ munol 2015;​6:​636. NG2 plays a role in neuroinflammation 43-9. 37. Mishra MK, Yong VW. Myeloid cells but is not expressed by immune cells. 69. Chataway J, Schuerer N, Alsanousi A, — targets of medication in multiple scle- Acta Neuropathol 2017;​134:​325-7. et al. Effect of high-dose simvastatin on rosis. Nat Rev Neurol 2016;​12:​539-51. 55. Trapp BD, Peterson J, Ransohoff RM, brain atrophy and disability in secondary 38. Prinz M, Priller J, Sisodia SS, Ranso- Rudick R, Mörk S, Bö L. Axonal transec- progressive multiple sclerosis (MS-STAT): hoff RM. Heterogeneity of CNS myeloid tion in the lesions of multiple sclerosis. a randomised, placebo-controlled, phase 2 cells and their roles in neurodegenera- N Engl J Med 1998;​338:​278-85. trial. Lancet 2014;​383:​2213-21. tion. Nat Neurosci 2011;​14:​1227-35. 56. Nikić I, Merkler D, Sorbara C, et al. 70. Fox RJ, Coffey CS, Cudkowicz ME, et al. 39. van der Valk P, Amor S. Preactive A reversible form of axon damage in ex- Design, rationale, and baseline character- ­lesions in multiple sclerosis. Curr Opin perimental autoimmune encephalomyeli- istics of the randomized double-blind Neurol 2009;​22:​207-13. tis and multiple sclerosis. Nat Med 2011;​ phase II clinical trial of ibudilast in pro- 40. Maggi P, Macri SMC, Gaitán MI, et al. 17:​495-9. gressive multiple sclerosis. Contemp Clin The formation of inflammatory demye- 57. Mahad DH, Trapp BD, Lassmann H. Trials 2016;​50:​166-77. linated lesions in cerebral white matter. Pathological mechanisms in progressive 71. Fisher E, Nakamura K, Lee JC, You X, Ann Neurol 2014;​76:​594-608. multiple sclerosis. Lancet Neurol 2015;​14:​ Sperling B, Rudick RA. Effect of intra- 41. Butovsky O, Jedrychowski MP, Moore 183-93. muscular interferon beta-1a on gray mat- CS, et al. Identification of a unique TGF- 58. Thompson AJ, Banwell BL, Barkhof F, ter atrophy in relapsing-remitting multi- β-dependent molecular and functional sig- et al. Diagnosis of multiple sclerosis: 2017 ple sclerosis: a retrospective analysis. nature in microglia. Nat Neurosci 2014;​17:​ revisions of the McDonald criteria. Lancet Mult Scler 2016;​22:​668-76. 131-43. Neurol 2017 December 21 (Epub ahead of 72. Zivadinov R, Havrdová E, Bergsland N, 42. Miron VE, Boyd A, Zhao J-W, et al. M2 print). et al. Thalamic atrophy is associated with microglia and macrophages drive oligo- 59. Solomon AJ, Bourdette DN, Cross AH, development of clinically definite multi- dendrocyte differentiation during CNS re- et al. The contemporary spectrum of mul- ple sclerosis. Radiology 2013;​268:​831-41. myelination. Nat Neurosci 2013;​16:​1211-8. tiple sclerosis misdiagnosis: a multicenter 73. Schlaeger R, Papinutto N, Panara V, 43. Absinta M, Nair G, Sati P, Cortese study. Neurology 2016;​87:​1393-9. et al. Spinal cord gray matter atrophy cor- ICM, Filippi M, Reich DS. Direct MRI de- 60. Traboulsee A, Simon JH, Stone L, et al. relates with multiple sclerosis disability. tection of impending plaque development Revised recommendations of the Consor- Ann Neurol 2014;​76:​568-80. in multiple sclerosis. Neurol Neuroimmu- tium of MS Centers Task Force for a Stan- 74. Saidha S, Al-Louzi O, Ratchford JN, et nol Neuroinflamm 2015;​2(5):​e145. dardized MRI Protocol and clinical guide- al. Optical coherence tomography reflects n engl j med 378;2 nejm.org January 11, 2018 179 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.  Multiple Sclerosis brain atrophy in multiple sclerosis: a four- 81. Kaunzner UW, Al-Kawaz M, Gauthier tion in MS: results of a randomized, year study. Ann Neurol 2015;​78:​801-13. SA. Defining disease activity and response placebo-controlled multicenter study. Neu- 75. Martínez-Lapiscina EH, Arnow S, Wil- to therapy in MS. Curr Treat Options Neu- rology 1999;​53:​457-65. son JA, et al. Retinal thickness measured rol 2017;​19:​20. 87. Tran JQ, Rana J, Barkhof F, et al. Ran- with optical coherence tomography and 82. Muraro PA, Martin R, Mancardi GL, domized phase I trials of the safety/toler- risk of disability worsening in multiple Nicholas R, Sormani MP, Saccardi R. Au- ability of anti-LINGO-1 monoclonal anti- sclerosis: a cohort study. Lancet Neurol tologous haematopoietic stem cell trans- body BIIB033. Neurol Neuroimmunol 2016;​15:​574-84. plantation for treatment of multiple scle- Neuroinflamm 2014;​1(2):​e18. 76. Housley WJ, Pitt D, Hafler DA. Bio- rosis. Nat Rev Neurol 2017;​13:​391-405. 88. Reich DS, White R, Cortese IC, et al. markers in multiple sclerosis. Clin Immu- 83. Carbajal KS, Mironova Y, Ulrich-Lewis Sample-size calculations for short-term nol 2015;​161:​51-8. JT, et al. Th cell diversity in experimental proof-of-concept studies of tissue protec- 77. Kuhle J, Barro C, Disanto G, et al. Serum autoimmune encephalomyelitis and mul- tion and repair in multiple sclerosis le- neurofilament light chain in early relapsing tiple sclerosis. J Immunol 2015;​195:​2552-9. sions via conventional clinical imaging. remitting MS is increased and correlates 84. Kwong B, Rua R, Gao Y, et al. T-bet- Mult Scler 2015;​21:​1693-704. with CSF levels and with MRI measures of dependent NKp46(+) innate lymphoid cells 89. Nandoskar A, Raffel J, Scalfari AS, disease severity. Mult Scler 2016;​22:​1550-9. regulate the onset of TH17-induced neu- Friede T, Nicholas RS. Pharmacological 78. Guerau-de-Arellano M, Alder H, Ozer roinflammation. Nat Immunol 2017;​18:​ approaches to the management of second- HG, Lovett-Racke A, Racke MK. miRNA 1117-27. ary progressive multiple sclerosis. Drugs profiling for biomarker discovery in multi- 85. Segal BM, Constantinescu CS, Ray­ 2017;​77:​885-910. ple sclerosis: from microarray to deep se- chaud­ huri A, Kim L, Fidelus-Gort R, 90. Altmann DR, Jasperse B, Barkhof F, et quencing. J Neuroimmunol 2012;​248:​32-9. Kasper LH. Repeated subcutaneous injec- al. Sample sizes for brain atrophy outcomes 79. Montalban X, Hauser SL, Kappos L, tions of IL12/23 p40 neutralising antibody, in trials for secondary progressive mul- et al. Ocrelizumab versus placebo in pri- ustekinumab, in patients with relapsing- tiple sclerosis. Neurology 2009;​ 72:​ 595- mary progressive multiple sclerosis. N Engl remitting multiple sclerosis: a phase II, 601. J Med 2017;​376:​209-20. double-blind, placebo-controlled, random­ 91. Fox RJ, Thompson A, Baker D, et al. 80. Giovannoni G, Turner B, Gnanapavan ised, dose-ranging study. Lancet Neurol Setting a research agenda for progressive S, Offiah C, Schmierer K, Marta M. Is it 2008;​7:​796-804. multiple sclerosis: the International Col- time to target no evident disease activity 86. The Lenercept Multiple Sclerosis Study laborative on Progressive MS. Mult Scler (NEDA) in multiple sclerosis? Mult Scler Group, the University of British Columbia 2012;​18:​1534-40. Relat Disord 2015;​4:​329-33. MS/MRI Analysis Group. TNF neutraliza- Copyright © 2018 Massachusetts Medical Society. images in clinical medicine The Journal welcomes consideration of new submissions for Images in Clinical Medicine. Instructions for authors and procedures for submissions can be found on the Journal’s website at NEJM.org. At the discretion of the editor, images that are accepted for publication may appear in the print version of the Journal, the electronic version, or both. 180 n engl j med 378;2 nejm.org January 11, 2018 The New England Journal of Medicine Downloaded from nejm.org at AUCKLAND UNIVERSITY OF TECHNOLOGY on January 28, 2019. For personal use only. No other uses without permission. Copyright © 2018 Massachusetts Medical Society. All rights reserved.