The Ultimate Long-term Outcome in Multiple Sclerosis

Gavin Giovannoni


Brain. 2019;142(5):1166-1167. 

When the initial positive results of the pivotal interferon beta-1b trial in relapsing-remitting multiple sclerosis were published (Paty and Li, 1993; IFNB Multiple Sclerosis Study Group, 2001), they were met in some quarters with scepticism. Many commentators wondered if the efficacy was clinically meaningful and questioned whether interferon beta should be prescribed at all. The subsequent uptake of interferon beta as an 'effective treatment' for multiple sclerosis was slow in many areas and led to wide variation in prescribing and access to treatments, which persists to this day both regionally and globally. In the UK, NICE—formerly called the National Institute for Clinical Excellence—ruled that interferon beta treatments were not cost-effective and hence could not be prescribed under the National Health Service. This led the Department of Health to introduce a risk-sharing scheme (RSS) to allow people with multiple sclerosis to access interferon beta as part of a national study with the aim of ascertaining whether or not the three interferon beta formulations and glatiramer acetate were cost-effective in clinical practice. The RSS cohort, when compared with a modelled untreated control population from British Columbia, showed that declines in Expanded Disability Status Scale (EDSS) scores and in utility, or quality of life, were significantly reduced for relapsing-remitting patients, with an approximately 4-year delay to needing a walking stick (EDSS 6.0) (Palace et al., 2019). As a reasonable proxy for secondary progressive multiple sclerosis, this result indicates that interferon beta delays the onset of the more advanced stages of multiple sclerosis. Subgroup analyses of the data show that outcomes are better for patients treated earlier and with lower EDSS scores (Palace et al., 2019).

In this issue of Brain, Kingwell and co-workers add to these data by revealing the long-term impact of interferon beta on survival in a 'real-world' setting (Kingwell et al., 2019). In a population-based observational study of patients with relapsing multiple sclerosis naïve to disease-modifying therapies (DMTs) and other immunosuppressant treatments, they show an association between all-cause mortality and interferon beta exposure. They used a nested case-control design with up to 20 controls, matched to cases or deaths by country, sex, age, year and disability level at study entry. The odds of interferon beta exposure were 32% lower among cases than controls (odds ratio 0.68; 95% confidence interval 0.53–0.89) and survival increased with more than 3 years of interferon beta exposure, although not with between 6 months and 3 years of exposure. The investigators assessed the cause of death from the death certificates and found the majority of deaths to be multiple sclerosis-related. However, they were unable to assess whether the reduction in mortality could have been due to the pleiotropic effects of interferon beta on, for example, infections and malignancies. The latter, however, seems unlikely in view of the causes of death.

These results now confirm the, often criticized, 21-year observational follow-up study on all-cause mortality in the cohort of 372 subjects who participated in the pivotal clinical trial of interferon beta-1b (Goodin et al., 2012). Remarkably, 98.4% (366 of 372) of the original patients were identified, and, of these, 81 deaths were recorded. Subjects originally randomized to interferon beta-1b (250 μg) had a significant reduction in all-cause mortality over the 21-year period compared with placebo (P = 0.02, hazard ratio 0.53; 95% confidence interval 0.31–0.90). The majority of deaths were also considered to be multiple sclerosis-related, making off-target effects of interferon beta unlikely (Goodin et al., 2012).

These Canadian and French results should surely establish beyond reasonable doubt the real-life efficacy of DMTs for the treatment of multiple sclerosis, which in turn should support the early effective treatment philosophy to maximize long-term outcomes (Giovannoni et al., 2016).

Can we extrapolate these findings further? Numerous phase three programmes have shown that several of the newer and more effective DMTs result in better disease outcome, in terms of disability progression, when compared to interferon beta. In addition, these controlled data are supported by several real-life datasets showing that on average patients with multiple sclerosis do better when started on, or switched to, higher efficacy therapies compared to the injectable first-generation DMTs, interferon beta and glatiramer acetate. These results are likely to impact on time to higher EDSS scores, including time to secondary progressive multiple sclerosis, and time to EDSS 10 or death. It is reassuring to note that in a recently reported international study, with prospective data from 68 neurology centres in 21 countries, commencing therapy between 1988 and 2012 on the higher efficacy DMTs natalizumab, fingolimod or alemtuzumab was associated with a lower risk of conversion to secondary progressive multiple sclerosis compared to initial treatment with glatiramer acetate or interferon beta (Brown et al., 2019). In an epoch analysis using a time frame of over 30 years, Capra and colleagues (2017) in Brescia, Italy, demonstrate that patients diagnosed in more recent epochs need a walking aid (EDSS = 6) at an older age, which is associated with changes in the treatment patterns experienced across the period.

It is clear that in the modern era, DMTs have not only changed the short-term outcomes of people with multiple sclerosis, but have had a long-term impact on time to secondary progressive multiple sclerosis and hard disability outcomes, in particular time to needing a walking stick, and death. It is also becoming clear that the earlier DMTs are started, the better the outcome, and that high efficacy therapies are superior to interferon beta and glatiramer acetate (He et al., 2015; Spelman et al., 2015; Kalincik et al., 2017; Harding et al., 2019). Based on these observations, treating to a target of 'no evident disease activity' and flipping the pyramid, i.e. starting with the most effective therapies first-line or early, is likely to improve disability outcomes further (Giovannoni, 2018). Whether these gains will be viewed as marginal and come with unacceptable risks and costs remains to be determined. Kingwell and colleagues, however, must be congratulated on filling in these knowledge gaps, which cannot be addressed by the pivotal phase 3 trials. The question now is not whether or not to treat multiple sclerosis, but whether or not we are undertreating the disease. Is the new therapeutic nihilism 'undertreated' or 'smouldering' multiple sclerosis?