Part 1

Secondary-progressive multiple sclerosis (SPMS) is generally defined as the onset of irreversible neurological disability in the absence of relapses in patients with a prior relapsing-remitting (RRMS) course. The invention of the SPMS phenotype has proved problematic. The pathology of RRMS and SPMS differs in degree, not kind, leading some to argue that there is no biological rationale for distinguishing phenotypes at all (Scalfari A. Mult Scler 2021;27:1002-1007). So it is not surprising that no biomarker signalling the onset of SPMS has been identified (Cree et al. Neurology 2021;97:378-388). The transition from RRMS to SPMS is of variable length, which likely reflects the limitations of detection and the diagnostic criteria used rather than the evolution of a distinct biology. The transition period is typically 2-3 years, which is the average duration of diagnostic uncertainty (Katz Sand et al. Mult Scler 2014;20:1654-1657).

The traditional view was that an SPMS diagnosis was contingent on two phenomena: an absence of relapses and the onset of irreversible disability. This created a false binary of relapsing disease versus progressive disease. The more contemporary view is that both inflammation and neurodegeneration are present to a greater or lesser degree during the two phases. About one-third of relapsing patients will not progress to SPMS; and a similar proportion of SPMS patients will have ongoing disease activity (Paz Soldan et al. Neurology 2015;84:81-88). Thus, the two phenotypes might be more properly considered as an admixture rather than as a sequence of pathologic events.

Indeed, the trigger in the evolution of SPMS may be time itself. Some studies have suggested that age may be the most relevant indicator of SPMS onset (Tutuncu et al. Mult Scler 2013;19:188-198). This has prompted the notion of MS phenotypes based on age rather than biology (Scalfari 2021). In line with this is the observation that current age – independently of disease duration – is a risk factor for SPMS (Cree et al. Neurology 2021;97:378-388).

An analysis of the British Columbia MS database estimated that 25% of RRMS patients reached SPMS by age 45 years, or less than 11.4 years after MS symptom onset (Koch et al. J Neurol Neurosurg Psychiatry 2010;81:1039-1043). Other studies have also reported SPMS onset around age 45 (Confavreux et al. Brain 2006;129[Pt 3]:595-605; Tutuncu 2013). The London group found that 25% of patients developed SPMS at six years after MS onset, and 50% reached SPMS onset at 15 years after onset (age 44 years) (Scalfari et al. J Neurol Neurosurg Psychiatry 2014;85:67-75). These findings suggest that using age as the sole diagnostic criterion, a majority of RRMS patients will reach SPMS before age 50 years.

Worsening disability is generally attributed to ongoing inflammatory activity, however, a number of recent studies have shown that most progression is not due to relapses. In an OPERA analysis, >80% of disability accumulation in treated RRMS patients was due to PIRA rather than relapse-associated worsening (RAW) (Kappos et al. JAMA Neurol 2020;77:1132-1140). Relapses may mask the underlying process so that progression only becomes apparent as relapses diminish. This creates the impression of a sequence from what are actually co-occurring events.

A more interesting finding from this analysis was that 19-23% of patients experienced progression at a younger age (mean 39 years at endpoint), or about 8 years after MS symptom onset. As PIRA essentially represents SPMS onset, this suggests that SPMS might be diagnosed earlier in a subgroup of patients.

There are conflicting data on whether treatment during the RRMS phase will appreciably delay the onset of SPMS. An MSBase analysis of data cobbled from three sources reported that SPMS was diagnosed later in treated patients; the most substantial difference was seen in patients receiving fingolimod versus placebo (hazard ratio 0.37) (Brown et al. JAMA 2019;321:175-187). In contrast, Coret and colleagues reported that 36.3% of treated MS patients developed SPMS at a mean age of 42.6 years; this was consistent with natural history studies and suggested that treatment had little impact on delaying SPMS onset (Coret et al. Mult Scler J Exp Transl Clin 2018;4:2055217318783347). It is uncertain if a more aggressive treatment plan would have produced an effect.

An SPMS diagnosis was once a means of identifying MS patients who were no longer likely to respond to a disease-modifying therapy, a situation that no longer applies with newer therapies demonstrating a benefit in progressive MS trials (Kappos et al. Lancet 2018;391:1263-1273; Hauser et al. N Engl J Med 2017;376:221-234). While the impact of treatment on disability accumulation appears to be greater in patients with ongoing disease activity, this may be due in part to greater disease severity. Relapses during the SPMS phase have been shown to be associated with a shorter time to EDSS 6.0 (Paz Soldan et al. Neurology 2015;84:81-88). However, it is noteworthy that a clinical benefit is also seen in patients with non-active disease (Cree et al. Mult Scler 2021;27:1564-1576).

In the EXPAND trial, patients were older (mean age 48 years), had lived with SPMS for a mean of 3.9 years, and already had significant disability (mean EDSS 5.4). The benefits seen in that study would likely have been more substantial in SPMS patients who were diagnosed sooner. The early diagnosis of SPMS will be addressed in Part 2 of this article.