The year 2021 saw several important shifts in how secondary-progressive multiple sclerosis (SPMS) is viewed and which may lead to improvements in diagnosis and management.
SPMS is characterized as the onset of the progressive phase of a relapsing-remitting course. Almost a decade ago, the phenotype was further modified somewhat curiously as active with or without progression, not active with progression, and stable disease (no activity or progression), in effect using the presence or absence of disease activity as the primary descriptor of progression (Lublin et al. Neurology 2014;83:278-286). This phenotype approach, developed primarily to standardize clinical trial enrolment for drugs targeting inflammation, was necessarily limited by what could be seen or detected. This shortfall became more apparent with a number of studies this year (see also Is all MS progressive? NeuroSens, September 8, 2021).
The MS-EPIC group initially described what was called “silent progression”, or progression not attributable to focal inflammation, which occurred in a majority of RRMS patients. In this context, while relapses were informative, they served to mask early progression (UCSF MS-EPIC team et al. Ann Neurol 2019;85:653-666). But it was a year before the concept caught on as progression independent of relapse activity (PIRA). The analysis of the OPERA studies differentiated PIRA, considered a feature of progressive disease, and relapse-associated worsening (RAW) (Kappos et al. JAMA Neurol 2020;77:1132-1140), maintaining the distinction that Lublin and colleagues had made between “worsening” and true “progression” (Lublin 2014). In the OPERA cohort, the authors found that PIRA accounted for >80% of 24-week confirmed disability accumulation (CDA) in RRMS patients. Suppressing inflammatory activity with ocrelizumab or beta-interferon essentially unmasked the underlying progression that was present even early in the disease course.
Similarly, an Italian registry study of CIS/early MS (mean EDSS 1.7) reported this year that 45% of patients had a CDA event. Overall, 61% of events were PIRA; mean age at PIRA onset was 34 years (Fonderico M. ECTRIMS 2021; 037). The Barcelona group reported PIRA events in 28% of CIS patients, with one-third occurring in the first five years after CIS diagnosis (Tur C. ECTRIMS 2021; 155). These data suggested that the distinction between the RRMS and SPMS phenotypes was somewhat arbitrary: relapses occurred in the progressive phase and progression occurred in the relapsing phase. Progressive disease was evident early, even in RRMS patients with little disability (EDSS <3.0).
Another interesting finding this year came from a post-hoc analysis of siponimod data (Giovannoni et al. ECTRIMS 2021; P001). Active SPMS was defined as relapses in the two years prior to screening with or without ≥1 gadolinium-enhancing lesion at baseline, criteria that had been used in EXPAND (Kappos et al. Lancet 2018;391:1263-1273). In the non-active group assigned to the placebo arm, 41.8% had new MRI activity, 4.6% relapsed only, and 9.2% had both relapse and MRI activity. This suggested that periodic clinical or radiological observations would miss the disease activity that occurs in a majority of patients categorized as non-active SPMS.
Classifying a patient as non-active SPMS has real-world consequences. In the Giovannoni analysis, a higher proportion of patients considered to be non-active were left untreated (45.1%), which may reflect, in part, the view of insurers that non-active SPMS is untreatable (or at least not reimbursable). Moreover, a non-active designation can be self-fulfilling: a German MS registry reported this year that only a minority of SPMS patients routinely received MRIs (Frahm et al. Mult Scler Relat Disord 2021;56:103281). The rate of MRIs dropped below 20% once a patient had reached age 50 years. A limitation of the study was that a majority of patients could not be categorized as active or inactive SPMS because of missing MRI data.
The year also saw several groups attempt to identify predictors of progression. Of increasing importance is early cortical changes (Pisani et al. Eur J Neurol 2021;28:2503-2512). A 20-year study found that a higher number of cortical lesions at CIS/MS diagnosis was predictive of early disability accumulation and SPMS onset (Schiavi et al. ECTRIMS 2021; P105). A change in cortical grey-matter volume was also predictive of worsening progression in SPMS patients in the EXPAND study (Fox RJ. AAN 2021; 092).
Of emerging interest is spinal cord lesions. A 12-year longitudinal study of CIS patients found that the presence of SC lesions was a significant predictor of SPMS (odds ratio 2.2) (Lacruz-Ballester et al. ECTRIMS 2021; 137). Similarly, a 5-year study by the MAGNIMS group found that the number of SC lesions was an independent predictor of SPMS (OR 2.0) (Rocca et al. ECTRIMS 2021; P417).
Some progress was made this year with to respect serum neurofilament-light (NfL), a biomarker of axonal damage. NfL levels correlate with disease activity (Novakova et al. Neurology 2017;89:2230-2237. Damasceno et al. Mult Scler Relat. Disord 2019;30:149-153), so it may substitute for MRI in some instances or indicate a need for further investigation. In addition, NfL levels may predict patients at risk of disability progression, provide an indication of therapeutic response and serve as a guide to switching treatments (Kuhle et al. Neurology 2018;90(suppl 15):S8.006. Kuhle et al. Mult Scler 2020;26:1691-1699. Reyes et al. Neurol Neuroimmunol Neuroinflamm 2020;7:880). While some MS clinics now evaluate serum NfL, its optimal use in routine care remains to be determined.
The year also saw the trend to expanding the concept of progression to incorporate other domains, such as cognition, patient-reported symptom worsening and quality of life, in part a reflection of the limitations of EDSS in detecting change. In a NeuroSens survey this year, 50% of respondents said their criterion for defining SPMS was a 1-point change in EDSS at six or 12 months; however, 43% said that the most useful measures were worsening physical and/or cognitive function, patient-reported declines in QoL and activities of daily living or other changes (complete the survey to view the results at https://neuro-sens.com/is-all-ms-progressive/). Such a broader perspective may help to identify the onset of progression earlier and more adequately meet the needs and concerns of patients. The Canadian Prospective Cohort Study to Understand Progression in MS (CanProCo) announced this year should provide additional insights on how these and other factors interact to drive disease progression and how to target therapies more effectively (Oh et al. BMC Neurol 2021;21:418).