New Disease Concepts
Although genetic vulnerability is probably a prerequisite for development of MS, it is not sufficient in the absence of other contributing factors. With recognized low-, medium-, and high-risk geographic disease zones, environmental factors clearly play a role. Exposures to common viruses and bacteria relatively early in life, in a way that is not yet understood, set the stage for MS. Infections probably act as disease triggers, although continued active neural or extraneural infection has not been ruled out as a disease factor in selected patients. Molecular mimicry (shared epitope sequences between ubiquitous infectious agents and autoantigens, including CNS antigens) is well documented. It is commonly believed, although not proven, that infection-triggered cross-reactivity to a myelin component initiates MS in genetically vulnerable individuals. Epitope spread occurs when CNS damage releases multiple sequestered antigens to the systemic immune system. Two types of epitope spreading may be involved: intermolecular epitope spread is when the immune response moves from 1 myelin antigen to another (ie, started against myelin basic protein (MBP), but with attack of myelin, other antigens such as myelin oligodendrocyte glycoprotein (MOG) are released and the immune response then spreads to MOG); and intramolecular epitope spreading, which occurs when the immune response is directed against a specific peptide of a specific protein (eg, MBP amino acid sequence 82-99) then with the destruction of myelin, it exposes other hidden or "cryptic" epitopes of MBP and the immune response shifts to these new epitopes (eg, MBP 102-118).[1] This expands the immune attack and acts to enhance and perpetuate organ-specific autoimmune disease. Epitope spread occurs in animal models of MS, and preliminary data indicate it is also a factor in MS.[2]

Therapeutic Implications of Epitope Spread
The concept of epitope spread carries important therapeutic implications. It argues for starting effective MS treatment at the earliest possible time point (ideally, at the first attack of definite MS), in order to minimize expansion and reinforcement of the damage process. The concept even provides a rationale for considering initial induction therapy (with broad-spectrum immunosuppression), followed by maintenance therapy. Supporting evidence that the early disease process is critical comes from natural history studies of first-attack, clinically isolated syndrome (CIS) patients. The number and volume of T2 brain lesions on the presenting MRI are the strongest correlates of disability at 14 years, followed by MRI lesion development during the first 5 years.[3]

New Insights Into Pathophysiology: Axons
MS is now believed to involve a biphasic disease process. Early on, inflammation is prominent, corresponding to the relapsing and potentially reversible phase of MS. Later, there is transition to a primarily neurodegenerative phase, corresponding to progressive MS with irreversible deficits. Although inflammation and neurodegeneration are detected at all time points, 1 process appears to dominate. This concept is consistent with natural history studies of MS, since most patients begin with relapsing disease but ultimately transition to secondary progressive disease. The presence of distinct MS phases argues for therapy that is tailored to the nature of the disease process.

The classic view of MS is that it is a disease that involves CNS inflammation and demyelination. It is now clear from direct pathologic data and indirect neuroimaging data that it also involves damage to axons and neurons.[4,5]Both axon density and volume are reduced in MS, not just within the plaque but also in normal-appearing CNS tissue.[6,7]Analysis of n-acetyl aspartate (NAA), an axon/neuron marker measured by MR spectroscopy, indicates that whole brain NAA is reduced even in early MS.[8]Loss or shrinkage of axons is a major contributor to brain and spinal cord volume loss (atrophy). In patients with MS, prominent CNS atrophy is present very early, even at the time of the first clinical attack.[9,10] On a yearly basis, brain volume loss in MS is accelerated 3- to 10-fold over that of matched controls.

The importance of axon damage in MS cannot be overstated; it is believed to be the neuroanatomic substrate of permanent disability and disease progression. Injury to axons undoubtedly reflects multiple factors (Table 3). At least some of the immune and inflammatory elements that injure axons are distinct from those that damage myelin. Axons or axon components (such as ion channels and neurofilaments) could be the target of a direct primary or secondary immune attack. Antibodies to neurofilament components, gangliosides, and myelin oligodendrocyte glycoprotein have been linked to progressive MS.[11-13]Alternatively, damage could reflect secondary bystander effects from inflammatory or toxic factors released into the microenvironment by intrinsic (microglia, astrocyte) or extrinsic (lymphocyte, macrophage) cells. Acute axon injury, measured by amyloid precursor protein expression, correlates with macrophage and CD8+ T-cell infiltration. In addition, ongoing myelin loss harms axons. It disrupts axon transport, with increased metabolic stress on neurons and axons.[14] The symbiotic relationship between myelin and axon means that demyelination itself is an axon damage mechanism. Loss of myelin also affects ion channel expression on the denuded axon surface.[15] These changes can restore nerve conduction, or produce axon destruction.

Table 3. Potential Contributors to Axon Damage in MS
Damage Mechanisms
Direct immune attack
- primary

- secondary

Indirect (bystander) immune damage
- from intrinsic CNS components

- from extrinsic components

Loss of myelin
- disrupted axonal transport

- damaging ion channel remodeling

- metabolic stress

Loss of neurons

Damage Factors
CD8+ T cells


Macrophages


Antibodies


- myelin oligodendrocyte glycoprotein

- neurofilament components

- gangliosides

Neurotoxic factors



Therapeutic Implications of Axonal Damage
Because insertion of new sodium channels allows conduction, enhanced or more rapid expression of such channels is a desirable therapeutic goal. By contrast, calcium channels (alpha 1-beta), normally expressed only at the presynaptic axon terminal, may also be inserted into demyelinated membrane. Because the insertion of these channels leads to axon damage (mediated by calcium-dependent proteases called calpains), one would want to prevent their expression. Ion-channel manipulation is likely to be a future therapeutic focus in MS.

New Insights Into Pathophysiology: Neurons
Axons are also lost when neurons die. A recent study of brain autopsy material examined the thalamus, a gray matter region rich in neurons. MS brains showed 30% to 35% fewer neurons than control brain tissue.[4] NAA measurements indicate diffuse gray matter disturbances even in patients with early MS. Loss of neurons likely reflects primary damage, from direct and indirect inflammatory injury, as well as secondary neurodegeneration, from a variety of toxic factors (Table 4).

Table 4. Potential Neurotoxic Factors in MS
Proinflammatory cytokines


Excitotoxins


Free radicals


Oxidative and metabolic stress factors


Disturbed extracellular ionic milieu


Immune system cells/immunoglobulins



The full significance of neuron involvement in MS is unclear, but it could well play a role in cognitive disturbances. The fact that axons and neurons are damaged, with proven correlates to clinical disability and MS disease progression and possible correlates to cognitive loss, highlights their importance as a therapeutic target. Neuroprotective strategies, neurotrophic growth factors, immunomodulatory therapies, cell transplantation, and even gene transfer are future approaches to make neurons and axons less vulnerable to injury; to promote remyelination and axon repair; and to replace lost oligodendrocytes and neurons.

The Subclinical Nature of MS
By definition, relapsing MS patients are clinically stable between disease attacks. Previously, scientists and clinicians believed that MS went into remission and that active disease was confined to periods of clinical relapse. Another common belief was that the MS disease process burned out over time. We now know that virtually all patients with MS experience ongoing subclinical disease activity and CNS damage, even when no obvious worsening or changes in the patient's neurologic examination have occurred.

Frequent MRI studies indicate that 80% to 90% of new brain lesions are not associated with clinical relapse or detectable examination changes.[16] In untreated MS populations, brain lesion burden increases by 5% to 10% each year.[16] Accelerated brain atrophy is present even in early, mild, relapse-free patients with stable Kurtzke Expanded Disability Status Scale (EDSS) scores.[17] Clearly, then, clinical criteria underestimate disease activity.

With the exception of benign relapsing MS, natural history studies indicate that all untreated patients develop disability. Furthermore, benign MS has never been formally defined and probably involves no more than 5% to 10% of patients.[18] This is a retrospective diagnosis, looking back after several decades of mild disease. In 1 recent study, half of patients diagnosed with benign MS at 10 years had significant disability at 20 years.[19] When MS is viewed as an active, ongoing disease process with accumulating permanent damage to the CNS, the importance of treatment at the earliest recognizable time point becomes even more apparent.

Neuroimaging and MS: Present and Future
Not only does clinical observation underestimate the true MS damage process, but so too does the best current clinical MRI analysis. Conventional MRI techniques detect T2 hyperintense lesions (which have little to no pathologic specificity), T1 hypointense lesions (which, when chronic, indicate greater tissue damage and axon loss), and gadolinium contrast-enhancing lesions (indicating a focal major breach of the blood brain barrier and current disease activity). However, unconventional techniques can detect microscopic and physiologic changes in normal-appearing CNS tissue (Table 5).

Table 5. Neuroimaging Techniques That Detect Abnormalities in Normal-appearing CNS Tissue of MS Patients
Magnetic resonance spectroscopy


Magnetic transfer imaging


Diffusion-weighted and diffusion-tensor imaging


Functional MRI


High magnet (>/= 3 Tesla) MRI


Positron emission tomography (PET) scanning





In brain MRIs of patients with MS, up to 70% of normal-appearing white matter may actually be abnormal. Indeed, tracking changes in normal-appearing tissue may be a sensitive marker for disease severity and response to therapy. In a recent analysis, yearly whole brain NAA changes differentiated early relapsing patients into 3 groups. Approximately 20% of patients had stable NAA levels, 55% showed a modest reduction, and 25% showed a marked reduction in these levels.[8] Presumably those patients with marked reductions in NAA levels, consistent with greater axon damage, have more severe disease, a worse prognosis, and will ultimately develop more rapidly progressing disability.

Magnetic transfer imaging measures signal changes in fluid and fixed phase molecules within regions of interest. This generates a magnetic transfer ratio (MTR). MTR allows lesion severity to be measured; the lower the MTR, the greater the tissue damage. Such nonconventional neuroimaging techniques are being used in research protocols, but ultimately some will come to routine clinical use.

Heterogeneity of the Disease
One final new concept about MS is that it is probably heterogeneous. The disease demonstrates clear clinical variability, based on distinct clinical subtypes and disease severity (Table 6).

Table 6. MS Clinical Heterogeneity
Subclinical (asymptomatic) MS
Based on autopsy studies


May account for up to 20% of MS



Clinical (symptomatic) MS
Relapsing subtype
- 85% of MS at onset

- 55% of MS overall

- characterized by disease attacks, clinical stability in between

Primary progressive subtype
- 10% of MS

- Slow worsening from onset

- Distinctive disease onset features (older age onset, progressive myelopathy, equal gender ratio, no relapses)

Progressive relapsing subtype
- 5% of MS

- Indistinguishable from primary progressive except for later superimposed relapses

Secondary progressive subtype
- 30% of all MS

- ultimately 90% of untreated relapsing MS

- prior relapsing patient who transitions to slow worsening



There is also genetic heterogeneity, with distinct disease-associated genes based on race. MS also shows heterogeneity based on ability to remyelinate; only 70% of patients experience remyelination of MS plaques. The factors that prevent remyelination in 30% of patients are not known, although studies suggest that the presence of oligodendrocyte precursors, axon integrity, and axon surface molecule expression are important contributors.[20-24] Finally, recent studies suggest immunopathologic heterogeneity. A multinational consortium of neurologists and neuropathologists has studied acute plaque pathology in MS brain tissue samples obtained at biopsy or autopsy.[25] Results of this study reveal 4 distinct immunopathologies (Table 7). These observations await confirmation but, if true, suggest 4 distinct categories of MS based on primary damage mechanisms, and would have profound therapeutic implications.

Table 7. Heterogeneous MS Brain Plaque Immunopathology
Pattern Frequency Damage Mechanism Animal
Model Oligodendrocyte Numbers Remyelination Clinical Correlations
I 19% Macrophage
mediated + (myelin-
induced EAE) Preserved + Seen in all MS subtypes
II 53% Antibody,
complement
mediated + (MOG-
induced EAE) Preserved + Neuromyelitis optica, + plasma exchange response
III 26% Distal dying back oligodendrogliopathy with apoptosis (ischemic, toxic, virus induced) --- --- Balo's concentric sclerosis
IV 2% Oligodendrocyte degeneration (metabolic defect) --- --- Atypical primary progressive MS


EAE = experimental allergic/autoimmune encephalomyelitis; MOG = myelin-associated glycoprotein