Toward Personalized Care for Multiple Sclerosis

The face of multiple sclerosis (MS) is changing. If asked to describe the typical person and journey of MS, not too long ago, the answer would likely have conjured up the image of a young woman with northern European ancestry who initially had episodic neurologic symptoms with increasing severity over time and, later, with passing years, more gradually advancing disability. MS was considered rare outside the white population, and it was viewed as unusual for someone with MS to live into their 60s and beyond. Although it holds true that about 75% of people newly diagnosed with MS are women, it is now recognized that MS occurs in all ethnicities, with a particularly high incidence in the US among Black women (see Multiple Sclerosis in the Black Population in the United States).¹ The average age for people with MS in the US has increased as well, with highest prevalence in people over age 50 (see Multiple Sclerosis in People Over Age 55).² Aging, in addition to lifestyle, comorbid conditions, environment, and genetics, plays an increasing role in the overall health of a person with MS. These changes in “who” has MS are superimposed on the inherent variability of the clinical presentation and the treatment response both amongst individuals and within a given individual’s disease course with likely greater malleability of the process in the early phases of disease when inflammatory processes are thought predominant.

Genetic, environmental, and lifestyle factors are thought to contribute to an individual’s risk of developing MS and are believed to affect the disease course. Faced with this complexity, there have been many efforts to develop and validate biomarkers and evaluation tools that promise to provide more definitive answers regarding prevention, diagnosis, prognosis, treatment selection, and response monitoring. Having such information could allow disease classification based on intrinsic biology criteria in addition to clinical signs and symptoms. Incorporating pharmacogenetics and pharmacogenomics might permit the development of tailored treatment approaches for groups of patients. Inherent in such a precision medicine approach is also continuous improvement of care through adjustment of best practices based on measured outcomes. Our understanding, however, of MS pathogenesis and variability at the level required for precision medicine with robust proven domain-specific biomarkers and precise outcome measurement tools are still lacking (Table 1). Although progress has been made, precision medicine for MS is still an ultimate goal rather than an implementable reality. The variable clinical presentations, the multitude of factors that affect the disease process at the individual level, and a long time between intervention and outcome continue to require a degree of personalized care approaches in MS that go beyond the concept of precision medicine.³

A key steppingstone on the path to precision medicine is the availability of domain-specific biomarkers and outcomes assessment tools. Although genome-wide association studies have identified more than 200 alleles associated with a higher risk of MS, there have been few advances in understanding how these genes are related to the etiology and pathophysiology of MS. An association between a more typical MS disease course and DRB1*15 in Black individuals remains the strongest example of such an association.⁴ A comprehensive review of the state of the biomarker research in MS and of the multiple tools developed to assess outcomes is beyond the scope of this article. The following sections cover select aspects of these topics as they relate to the development of personalized medicine for all individuals with MS.

Disease Definitions

The journey to reliable biomarkers for different domains of MS has been challenging. Although MS can be viewed as a single disease entity, it is still possible that MS, as defined today, is an umbrella diagnosis for conditions that share many common clinical and pathologic features with separate intrinsic biologic underpinnings. On the basis of this appraisal of MS, specific diagnostic, prognostic, treatment selection, treatment response, and more aspirational disease prevention biomarkers may only become evident as separate clinical conditions currently included under the MS umbrella have been further elaborated. The recent history of neuromyelitis optica spectrum disorder (NMOSD) serves as an illustrative example of this concern.⁵

NMOSD was considered a variant form of MS until the discovery that antibodies to aquaporin-4 (antiAQP4), also termed NMO-IgG, are diagnostic for NMOSD and, to a degree, both prognostic and useful for measuring treatment response. People with NMOSD who have a positive NMO-IgG test are at an increased risk of relapses and, in phase 3 clinical trials, appear to respond more favorably to studied interventions. Positive NMO-IgG tests have also led to the recognition of area postrema and diencephalic syndromes as part of NMOSD and the discovery of myelin oligodendrocyte glycoprotein (MOG) and glial fibrillary acidic protein (GFAP) antibodies as diagnostic biomarkers for conditions previously grouped with NMO-IgG–negative NMOSD. NMO-IgG as a biomarker has helped to further define the clinical manifestations of NMOSD that predominantly affect women and have higher incidence in individuals with Asian, African, or mixed ancestry, leading to the discovery of further disease entities distinct from both NMOSD and MS.⁶

MS Biomarkers

MRI

MRI is a proven diagnostic biomarker for MS that can also aid with prognosis and measurement of treatment response (Table 2). MRI helps exclude other conditions in the differential diagnosis. The presence of enhancing and nonenhancing lesions in characteristic locations—often seen on initial scans during or shortly after a first symptomatic event—can also confirm an MS diagnosis, representing disease activity dissemination in both space and time, a cardinal principle of MS diagnosis.^7,8 The prognostic qualities of MRI are evident in radiologically isolated syndrome (RIS), in which the presence of lesions in both the spinal cord and brain meets the diagnostic criterion of dissemination in space and predicts the development of a first clinical event.⁹ Findings on MRIs obtained after a first demyelinating event also provide prognostic information because the number of brain lesions, cerebellar involvement, T1 hypointensities, and atrophy (particularly of the gray matter) on early MRIs predict a less favorable disease course.^10,11 A higher proportion of persons with African ancestry have such early MRI findings concerning for more involved disease at an earlier stage.¹² Brain and spinal cord atrophy measures also provide information regarding disease effects. Spinal cord MRI, however, is not always part of the routine MRI ordered for suspicion of MS, in part because spinal cord imaging requires certain minimal imaging standards not always available and also because there is a need for consistent subsequent image processing.¹³ Cortical, gray matter, and cerebral deep gray matter atrophy has greater correlation with disability measures than white matter volume changes and correlates with cognitive impairment, fatigue, and depression.^14,15 Automated quantitative analysis of white and gray matter volumes on routine MRI scans has become technically feasible through widely available free-standing and integrated software. The availability of quantitative volumetric measurement during routine care can be expected to more reliably assess disease activity and outcomes.

Magnetization transfer (MT) and diffusion tensor imaging (DTI) also have the potential to serve as domain-specific imaging biomarkers available in the regular care environment for routine care of people with MS. Both sequences assess tissue integrity, neurodegeneration, and repair. As noted for atrophy, automated image analysis is needed for these imaging techniques to have broader application in regular MS care.⁷ Susceptibility-weighted imaging (SWI), which takes advantage of the paramagnetic properties of venous deoxygenated hemoglobin, may also have use in MS care because it allows visualization of the relationship between small blood vessels and inflammation. This technique permits detection of a central vein within T2 hyperintense white matter lesions, which can help differentiate MS from other conditions that present with white matter lesions. SWI may be of particular value when assessing individuals with comorbid conditions.¹⁶ A concern with SWI is that it requires field strength higher than the 1.5 T strength most widely used in clinical neuroimaging practice today.

High-field (3 T) and ultrahigh-field (7 T) MRI increase detection of white matter, cortical, deep central gray matter, and other lesions not typically seen at lower field strength and are needed to assess cortical pathology.¹⁷ Such imaging advances may reduce the relative difference in imaging findings and clinical presentation in MS referred to as the clinical-imaging paradox. Enhanced visualization of gray matter involvement will affect diagnosis,⁷ assessment of disease effects, and the study of treatment efficacy. A greater role for MR sequences currently considered unconventional to assess tissue integrity and lesions in gray matter and the spinal cord can be expected as image acquisition techniques improve and higher field strength scanners become more widely available.

Although the value of MRI in MS is well documented and holds much promise to improve diagnostic certainty, disease monitoring, and treatment response measurement, there are challenges of MRI in routine practice that can affect care. Regular monitoring with MRI is now a recommended part of value-based care in MS, but no quality standards for MRIs in MS are included in that definition. Minimal imaging standards have been proposed by the Consortium of MS Centers, but it is not known how widely these are implemented,¹² and episodic MRI monitoring for MS is often obtained at different imaging facilities with different scanners and different protocols at different points in time for the same person. Technical differences make scan-to-scan comparison difficult and imaging-based assessment of disease stability less reliable.

Fluid-Based Biomarkers

Neurofilament Light. Neurofilament light chain (NfL) is expressed in the cytoplasm of neurons, and blood serum NfL levels increases correlate with the degree of axonal damage in people with a variety of neurologic disorders, including first demyelinating events and MS.¹⁸ In people with MS, higher serum NfL levels are observed following a recent relapse and, thereafter, decline with passing time. Sensitive quantitative tests for NfL levels have been developed and validated, making longitudinal monitoring of NfL feasible. Recent clinical trials have included serial measurements of serum NfL levels in the study design and reduction of serum NfL levels have been observed after therapeutic intervention. The increase in NfL during relapses and subsequent decrease that correlates with time since relapse could potentially be used to qualify clinical trial participants as well.

Correlation between serum NfL levels and markers of disease activity on routine MRI is relatively weak. NfL levels may provide additional information regarding axonal injury in normal-appearing white matter and in the gray matter.¹⁹ Some have even proposed integrating serum NfL levels into the no evidence of disease activity (NEDA) measure that currently includes no new MRI lesions, no disability progression, no relapses, and no accentuated atrophy. There are several considerations still to be explored before NfL is used for monitoring disease activity, however, because age and comorbidities (including cardiovascular disease and diabetes) also correlate with increased serum NfL levels. The age range of people with the highest prevalence of MS has increased (see Multiple Sclerosis in People Over Age 55) as has the proportion of people with MS who have comorbid conditions. The effects of these factors need to be understood in order to evaluate and discern an age-adjusted normal vs elevated NfL level in a given individual. Another consideration is that decreases in NfL levels seen after starting study treatments in clinical trials may, in part, simply reflect regression to the mean from last disease activity, because many trials require recent disease activity as an enrollment criterion. Because serum NfL lacks disease specificity and does not provide anatomic information, it may add to rather than displace MRI as a routine monitoring tool for MS (at least where regular imaging is feasible).¹⁷ Serum NfL as a disease marker shares certain qualities with measurement of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), which are nonspecific markers alerting to the presence of an unspecified inflammatory process.

Other Cerebrospinal Fluid Biomarkers. Intrathecal immunoglobulin (Ig) synthesis is a hallmark of MS that can be observed quantitatively as an elevated Ig index or qualitatively as cerebrospinal fluid (CSF)-restricted oligoclonal bands (OCBs). Over 90% of people with MS have CSF-restricted OCBs, and the presence of OCB at the initial clinical event predicts a subsequent clinical event. The 2017 diagnostic criteria for MS consider IgG index or presence of CSF OCB as evidence of dissemination in time.⁷

CSF levels of the proteins GFAP and chitinase-3-like protein-1 (CH3L1) also correlate with ongoing disease activity and tissue injury and may correlate with disability.²⁰ Although these proteins both hold promise as biomarkers, the inability to measure levels in the blood rather than the CSF limit clinical use at present. Like NfL, OCB, GFAP, and CH3L1 levels alone are not specific for MS and thus also require consideration of the clinical context.

Measuring Outcomes in MS

Reliable metrics linked to important outcomes of MS, including the ability to function in everyday life and overall well-being, are integral to precision medicine approaches. The Expanded Disability Status Scale (EDSS) is the most broadly used clinical disability outcome measure. The EDSS, however, relies significantly on ambulatory functioning.²¹ Further measures including the Multiple Sclerosis Functional Composite (MSFC) and its proposed iterations incorporate an assessment of cognition—the paced auditory serial addition test (PASAT) or the Symbol Digit Modalities Test (SDMT)—and assessment of low-contrast visual acuity (LCVA) to fill the assessment gaps in the EDDS.^22-24 Other measures for additional MS symptoms including fatigue, depression, and bladder and sexual dysfunction also need to be integrated into outcome assessment. Ideally, a new tool would be developed that retains the intuitiveness of the EDSS while incorporating measures of cognitive deficits, depression, fatigue, bladder and sexual dysfunction, and vision. Digital technologies using biosensors or capabilities already integrated into smartphones may help continuously assess functioning during daily life in contrast to periodic monitoring during clinic visits.²⁵

Patient-centered care models, including personalized care and precision medicine, rely upon having a participatory patient who plays an active role in their own care. This is ideally done, in part through patient-reported outcomes (PROs) that capture information about a person’s health without requiring interpretation by a clinician or anyone else. The MS Quality of Life-54 (MSQOL-54) scale is a multidimensional health-related quality-of-life measure that combines generic and MS-specific items. The Neuro-QoL (Quality of Life in Neurological Disorders) is a measure of the physical, mental, and social effects of neurologic conditions. These 2 measurement scales are the most frequently used PROs for people with MS.²⁶ Food and Drug Administration (FDA) guidance regarding PROs suggests these will play a greater role in MS care in the future.²⁷ An example of using this guidance in practice is the Fatigue Symptoms and Impacts Questionnaire-Relapsing Multiple Sclerosis (FSIQ-RMS), a validated 20-item questionnaire regarding symptoms of fatigue in relapsing forms of MS. This FSIQ-RMS is the first PRO developed studied as a secondary outcome in a phase 3 clinical trial that was developed in response to this FDA guidance.²⁸ Considering a potentially greater role for PROs in MS care, it will be important to have transparency regarding how these tools are developed to better understand how factors such as ethnicity and socioeconomic status are considered in validating these tools.

Summary

MS is the first neurodegenerative disorder for which there are an increasing number of disease-modifying treatments (DMTs) with different proposed modes of action. Despite—or perhaps because of—such considerable advances, however, there are still hurdles on the path towards precision medicine in MS. Although MRI-guided diagnosis and monitoring have largely become standard of care, biomarker discovery remains an area of unmet need. Fluid-based markers, such as serum NfL, are poised to complement imaging rather than replace it. There is also a need for a clinical outcome measure that retains the intuitiveness of the EDSS and incorporates the additional MS-related domains affecting patients’ lives that are not included in the EDSS. PRO measures are likely to play an increasing role not only for disease and safety monitoring but also in the clinical development of treatments to establish efficacy. It will be important when addressing these unmet needs in MS that a truly representative population of people with MS is duly considered. This starts with the need for a better understanding of pathogenesis across groups, including why Black people are at higher risk for MS than white people and more insight into how factors such as health literacy, poverty, lack of continued adequate insurance coverage, and other social determinants of health influence the course of MS. This is an area where inclusive, adequately developed PROs may be helpful considering that historically minoritized populations continue to be underrepresented in scientific research and clinical trials.