Disease-modifying therapies (DMTs) are the standard care for individuals with multiple sclerosis (MS). A better long-term prognosis is associated with an early initiation of DMTs, highlighted Bruno Stankoff, professor at the University Pierre et Marie Curie (UPMC), neurologist at the Pitié Salpêtrière Hospital in Paris, and vice president of the ECTRIMS Executive Committee, in our recent webinar on how and when to de-escalate or discontinue DMTs, especially in the older population.
Fredrik Piehl, a professor at the Karolinska Insitute in Stockholm, and Elisabeth Maillart, a neurologist at the Pitié Salpêtrière Hospital in Paris, contributed to the discussion.
De-escalation and discontinuation: Why and when?
Professor Piehl started by emphasising the need to adjust the treatment plan after decades of treatment. The risk/benefit analysis performed when initiating the treatment must be reassessed after 10, 20, or 30 years. Generally, older individuals living with MS for a longer time accumulate more disability. Over time, the inflammatory features of MS tend to wane, with relapses and focal MRI (magnetic resonance imaging) activity becoming less frequent. The problem is that most clinical trials have included so far only individuals below 55 years old. AFFIRM, a phase-III clinical trial testing natalizumab against placebo, showed that natalizumab reduces risks of relapses in younger and older individuals with MS but protects from disability only people under the age of 40 years old [1]. Professor Signori and colleagues conducted a meta-analysis of 10 randomised control clinical trials, finding that higher treatment effects are associated with lower age and more active disease [2]. Another meta-analysis identified age as an important modifier of drug efficacy: after the age of 53, higher efficacy treatments are not predicted to be more beneficial than lower efficacy treatments [3].
Older people with MS tend to accumulate more disability. A recent study showed that individuals with late-onset MS accrue more disability in a shorter time compared to those who had onset between 18 and 50 years old [4]. Ageing led to an increased risk of less recovery after relapses and greater disability accrual [5]. Moreover, Professor Piehl illustrated a study (in press) analysing a population-based cohort in Sweden of almost four thousand people. Three years after starting treatment, individuals over 50 accumulated twice the risk of disability worsening compared to the entire cohort, whereas the annualised rate of relapses is reduced [5].
Comorbidities further increased the risk of accumulating disability. Clinical trials on ocrelizumab reported higher rates of cancer or severe infections in people who are older and have a progressive disease [7, 8]. Additionally, treatment duration significantly impacted the risk of severe infections [9].
In conclusion, when starting a treatment at higher age, it is important to look at evidence of inflammatory disease activity. The choice of the disease-modifying therapy (DMT) must be individualised, based on the benefit-risk analysis. De-escalation can be considered when the disease is stable, or when the individual risk profile has changed. Infectious risk should be regularly discussed.
Other than treatment de-escalation, it would be essential to look at lifestyle factors, encourage physical activity, verify the vaccination status, and treat comorbidities.
De-escalation and discontinuation: When and how?
Dr. Maillart presented three practical situations in real life where it is possible to consider de-escalation and discontinuation of treatment: 1) stable ageing individuals with MS, with no activity and no progression; 2) progressive ageing individuals with MS, with no activity but with pure clinical progression; 3) pregnant women, who may also need a temporary de-escalation and discontinuation. In all these situations, it is important to be cautious about the risk of rebound after cessation of anti-trafficking DMTs like natalizumab and fingolimod.
When stopping natalizumab or fingolimod, there is a rapid re-entry of autoreactive T-cells into the central nervous system (CNS), leading to an increased risk of rebound. For natalizumab, the risk of rebound is estimated to be around 8-22%, occurring after 8-24 weeks, and for fingolimod, the risk is between 10 and 33%, occurring after 4-16 weeks [10]. Factors that can lead to an increased risk of rebound include younger age, previous high disease activity, shorter treatment duration, and longer washout periods when switching treatments.
Dr. Maillart highlighted the results of DISCOMS, the first randomised, non-inferiority study including 259 individuals with MS over 55 years old, mostly treated with platform DMTs (with a few individuals on high-efficacy therapies) and with a stable disease course (no relapses for 5 years and no MRI activity for 3 years) [11]. The primary end point – percentage of patients with relapses or MRI activity after discontinuation – was not reached. There was no evidence of non-inferiority of discontinuation compared to continuation. In the discontinuation group, the authors observed no increased risk of relapses, but a slightly increased risk of MRI activity, defined by one new lesion [11].
Another study based on the French registry OFSEP looked at 1,620 individuals with MS, over 50 years old, treated with high efficacy therapies, with no relapses and no MRI activity for at least 2 years [12]. The results varied according to the DMT used. The probability of a first relapse one year after the discontinuation was 33.6% in the group who discontinued natalizumab, 16.3% in the group who stopped fingolimod, and 0% when discontinuing an anti-CD20 [12].
In the second part of her presentation, Dr. Maillart focused on how to de-escalate or discontinue treatment. This process depends significantly on the type of DMT being discontinued. For platform therapy, no risk of rebound has been observed in older, stable individuals with MS. Similarly, no risk of rebound has been observed for anti-CD20 therapies. However, extending interval dosing can be considered before stopping an anti-CD20. Discontinuing lymphocyte anti-trafficking therapies requires an “exit strategy,” often involving bridging therapies like anti-CD20.
Individuals with MS who stop a therapy should be closely monitored, especially when stopping anti-trafficking DMTs. Young individuals with a short disease duration should be monitored for a longer period. Classic outcomes to be assessed include the presence of relapses, scores on the Expanded Disability Status Scale (EDSS) and the Multiple Sclerosis Functional Composite (MSFC), as well as radiological monitoring of the brain and spinal cord with MRI. In the future, digital biomarkers might also play a role in monitoring.
Finally, an important topic to consider is managing the anxiety of individuals with MS and neurologists who intend to stop treatment. A survey of 377 individuals living with MS revealed that only 11.9% of participants were willing to stop the treatment if the disease was considered stable [13]. Therefore, it is fundamental for neurologists to discuss the situation with individuals with MS and make shared decisions.
In case of pregnancy, injectable platform DMTs might be stopped at the beginning of pregnancy or continued during pregnancy. Other treatments must be stopped when pregnancy starts or even 2 months before. Immune reconstitution therapy must be completed before authorising pregnancy. Anti-trafficking DMTs cannot be abruptly stopped. Therefore, treatment is continued until the third trimester. For fingolimod, which can increase the risk of cardiovascular malformations in the foetus, a bridging therapy (e.g., anti-CD20 or natalizumab) is needed. More details are offered in our recent webinar on treatment during pregnancy for women with MS, NMOSD, and MOGAD.
In conclusion, de-escalation and discontinuation can be considered in specific situations, particularly in patients older than 55-60 years who have been stable for a long time. An exit strategy with a bridging therapy is recommended after the cessation of natalizumab and fingolimod. Further studies are needed to investigate the benefit/risk analysis in individuals with clinical progression who discontinue treatment, and a continuous monitoring is necessary.
A poll conducted during the webinar revealed that 88% of participants do not feel confident about the current knowledge on the benefit-risk analysis when stopping or de-escalating treatment. More clinical trials are needed to better explore this topic.
References
[1] Hutchinson M et al. J. Neurol. 2009; 256: 405-415.
[2] Signori A et al. Eur. Neurol. 2015; 22(6): 960-966.
[3] Weideman AM et al. Front. Neurol. 2017; 8: 296533.
[4] Knowles S et al. Ann. Neurol. 2024; 95(3): 471-486.
[5] Lublin FD et al. Brain 2022; 145(9): 3147-3161.
[6] Piehl F Ann. Neurol. in press.
[7] Hauser SL et al. NEJM 2017; 376(3): 221-234.
[8] Montalban X et al. NEJM 2017; 376(3): 209-220.
[9] Virtanen S, Piehl F and Frisell T JNNP 2024; 0:1-8.
[10] Roos I et al. Neurol. 2022; 99(17): e1926-e1944.
[11] Corboy JR. et al. Lancet Neurol. 2023; 22(7): 568-577.
[12] Jouvenot G et al., JAMA Neurol 2024.
[13] McGinley MP et al. Mult. Scler. J. 2020; 26(12): 1581-1589.