Practice Points
- Clinicians should be informed about and able to discuss the changing risks and benefits of disease-modifying therapies (DMTs) as they relate to older people with multiple sclerosis (MS).
- Discontinuation or de-escalation approaches may be reasonable for older people with MS with no recent relapse or MRI activity, those who have continued to experience slow progression despite ongoing use of DMT, and those with enhanced risks due to a combination of comorbidities and the DMTs themselves (eg, sphingosine-1-phosphate modulators).
- DMT discontinuation in people with MS older than 60 years with no recent relapse or MRI activity primarily results in a small increase in MRI-detected lesions of unclear long-term clinical significance.
- Ongoing clinical and MRI monitoring is essential in older people with MS, especially if they discontinue or de-escalate DMT. Findings from future studies may reveal serum or other nonimaging markers of disease prognosis to help guide decision-making.
- More data are needed to ascertain the actual risks and benefits of DMT use, discontinuation, and de-escalation in older people with MS.
Often diagnosed as young adults, people with multiple sclerosis (MS) may live for decades with variable manifestations of the disease, including either stability or slow worsening of neurological disability, as they age. Since 1993, regulatory agencies around the world have approved over 25 disease-modifying therapies (DMTs) that alter the natural history of the disease in profound ways, resulting in longer lives and less disability for those affected by MS. The expanding armamentarium of DMTs, with their variable benefits, adverse effects, and risks, has resulted in a dramatic change in discussions of risks and benefits of these powerful drugs, especially because as people with MS age, the pathological, clinical, and radiological manifestations of MS change,1 as has been discussed in other manuscripts in this series. The consequences of these changes and the limitations in our knowledge about the use of MS DMTs in older adults are highlighted in this article.
DMT Use in Older Adults With MS
Several factors are relevant to the use of DMT in aging people with MS. Relapses and active gadolinium-enhancing and new T2-hyperintense MRI-detected lesions diminish with age.2,3 There is, however, no maximum known age at which there is no risk of new or ongoing relapses, inflammatory MRI-detected lesions, or disability progression. Although relatively common even in younger people, the clinical progression of disability independent of relapses and/or active new MRI-detected lesions is substantially greater in older patients.4
Current DMTs are primarily directed against the adaptive immune system, which is most active in younger people with relapsing MS,5 and have shown minimal benefit in older people with MS who primarily experience progression independent of acute inflammatory activity manifested by relapses and/or new, active MRI-detected lesions.6 Although 46% of adults with MS living in the United States are 55 years or older,7 there are minimal data available about the risks and benefits of DMT use in this population, as those older than 55 years have been excluded from the vast majority of controlled clinical trials. Thus, although most subgroup analyses and meta-analyses of these studies have concluded that there is diminished benefit of current DMTs above the age of 40 years8-12 and no clear benefit in reducing progression of disability above the age of 53 years,12 all these reviews are limited by minimal data in those older than 55 years. Notably, and for many different reasons, many older people with MS have discontinued DMT use.13-15 Conversely, among middle-aged people with MS who remain on DMT, the vast majority are unwilling to consider a trial off their DMT.16 Thus, use of MS DMT remains highly variable in people with MS older than 55 years.
A related issue is the role and potential benefit of DMTs in those with late-onset MS (LOMS), typically defined as people with first symptoms at 50 years or older, which occurs in approximately 5% of MS incident cases.17 A recent retrospective propensity-matched study of 436 participants with relapsing-onset disease 55 years or older in the Observatoire Français de la Sclérose en Plaques (OFSEP) registry compared outcomes between those who were treated with DMT and those who were not (218 in each group). Mean age at onset was 59.3 years, and median follow-up was over 5 years. Mean time to first relapse (the primary outcome) and time to first MRI activity were longer in those treated with DMT; however, median time to confirmed disability progression, progression independent of relapse activity (PIRA), and secondary progressive MS (SPMS) diagnosis were not delayed by treatment. The risk of serious infection was no different between groups.18
Risks of DMTs in Older Adults With MS
Compared with no DMT or lower-efficacy DMTs, the risks of using at least some MS DMTs, especially those with greater immunosuppression, may be higher for older people with MS. Infections, cancers, and diminished responses to vaccinations are more common in older people but may be even higher in those with MS, particularly older people with MS who use specific classes of DMTs, especially cell-depleting agents such as anti-CD20 therapies (eg, ocrelizumab, rituximab, ofatumumab, and ublituximab) and the rarely used alemtuzumab and mitoxantrone.19
Common comorbidities in older people with MS, such as diabetes mellitus, hypertension, vascular disease, and cancer, are associated with worse clinical outcomes in MS20 and may also worsen safety outcomes and contribute to early discontinuation of MS DMTs.21 Thus, clinical equipoise appears to be achieved in the discussion about how, or whether, to use or discontinue DMTs in aging people with MS. The challenge is to balance DMT efficacy in relapses and disability progression with the risks of infections (in multiple drug classes), worsening vascular risks (as with sphingosine-1-phosphate [S1P] modulators), or malignancies (with S1P modulators and cell-depleting agents) in aging people with MS.
DMT Discontinuation in Older Adults With MS
At least 50 observational, retrospective, and cohort studies have examined outcomes after discontinuation of MS DMTs. Many are series of variable sizes, including all ages, many reasons for discontinuation, and DMTs with differing mechanisms of action. Others have specifically studied older patients, certain DMTs (especially fingolimod and natalizumab due to concerns of rebound disease activity with sequestering agents), or particular contexts such as pregnancy. Findings from these studies, as well as those from meta-analyses22,23 and comprehensive reviews,24 have concluded that the highest risk of MS relapse after DMT discontinuation is seen in younger people with MS, those with recent active disease (relapse or new MRI-detected lesions), and those with lower disability. Enhanced risk of disability progression upon DMT discontinuation is somewhat less clear but may be associated with older age, higher level of disability, and recent progression.
More recently, 5 randomized controlled trials (RCTs) have investigated disease activity recurrence risks in DMT discontinuation. Two have been published (DISCOMS [NCT03073603]25 and DOT-MS [NCT04260711]26); 3 are ongoing (STOP-I-SEP [NCT03653273],27 TWINS [NCT06663189],28 and AMS05 [NCT05285891]29). In addition, 3 propensity-matched trials addressing DMT discontinuation have been published. See Table 1 and Table 2 for a comparison of study design (all 8 trials, Table 1) and primary outcomes (the 5 published studies, Table 2).
The DISCOMS trial25 enrolled 259 participants 55 years or older with no relapse for at least 5 years and no new MRI-detected lesions for at least 3 years. Participants were evenly divided into continuers and discontinuers and followed for a median of 24.5 months. The primary outcome of this rater-blinded study was a relapse or any new brain lesion. There were multiple secondary outcomes, including progression of disability confirmed at 6 months, and several patient-reported outcomes focused on quality of life, MS symptoms, and patient satisfaction. The statistical approach was a noninferiority analysis with an 8%
noninferiority margin.
The median age in DISCOMS was 63 years, and the average time since last relapse was almost 14 years. Findings from this study were unable to demonstrate noninferiority in the combined primary outcome measure, primarily due to a small increased risk of new brain MRI activity in discontinuers, especially 1 to 2 new asymptomatic lesions. Time to first event was also faster in the discontinuers. Relapses were rare in both groups, and there were no substantial differences in many other clinical measures and safety outcomes. Importantly, progression of disability confirmed at 6 months was very similar in continuers (11.1%) and discontinuers (12.3%). Patient satisfaction with treatment was the same in both groups at the outset but significantly higher among discontinuers at the end of the study. Notably, among discontinuers with new relapse or MRI activity, mean age was 60.8 years vs 63.3 years among discontinuers without new disease activity, and the number of years since disease onset since last relapse was also very similar in those discontinuers with and without new disease activity.
An extension of DISCOMS with 74 participants from the original study has also been completed and published.30 It was performed at the 10 largest recruiting sites and included those who completed at least 18 months in DISCOMS, did not meet the primary end point of relapse or new MRI-detected lesion, were willing to participate in 1 additional visit at least 1 year after the last DISCOMS visit, and remained in their original assignment group. The primary outcome was time to relapse or new brain MRI-detected lesion. One of 30 continuers and 2 of 44 discontinuers had new MRI-detected lesions, but there were no relapses in either group. As in the original DISCOMS trial, other clinical outcomes were minimally different between groups. Additionally, faster time to new disease activity (from the start of the primary study) and higher treatment satisfaction continued among discontinuers.
DOT-MS26 was a similarly designed study but potentially included patients as young as 18 years. Participants had to be free of relapses and could have a single new MRI-detected lesion in the past 5 years prior to enrollment. The primary outcome measure was a new relapse or any significant new MRI activity, meaning 3 or more new brain T2 lesions or 2 or more contrast-enhancing brain MRI-detected lesions. The median age was 54 years, and median time from last relapse was approximately 9 years. This study was stopped prematurely after 89 of 150 (59.3%) anticipated participants were enrolled and followed for a median of 15.3 months, as there was significant new activity in the discontinuer group (8 of 45; 17.8%) vs none in the continuer group. Importantly, the median age of those with significant new disease activity was 46 years (mean, 49.5 years). When the outcome was assessed as in DISCOMS, new activity was 24.4% higher over a shorter follow-up time in DOT-MS vs 12.2% in DISCOMS. Thus, the DOT-MS participants were younger and had more recent disease activity at baseline than those in DISCOMS and had twice as much recurrent disease as discontinuers in DISCOMS. Disability progression was confirmed in 17% of continuers vs 16% of discontinuers in DOT-MS. Comparing the DOT-MS and DISCOMS data reinforces the idea that younger age and more recent disease activity are key risks for disease recurrence in those who discontinue DMT and suggests that although there may be no tipping point where disease recurrence is so low that all people with MS with characteristics similar to those of DISCOMS participants should strongly consider this option, the risks are low enough that discussions about discontinuation should be routine in people with MS fulfilling these criteria.
One significant limitation of some of the retrospective observational trials and both published RCTs examining DMT discontinuation in MS has been the lack of inclusion of more highly effective drugs, especially cell-trafficking agents such as fingolimod and natalizumab, which have been associated with significant return of disease activity upon discontinuation of DMT in younger patients. An interrogation of the OFSEP registry31 has addressed this through a propensity-matched study of 154 participants continuing and 154 participants discontinuing fingolimod, natalizumab, and anti-CD20 agents (mostly rituximab, a small number of ocrelizumab) combined, with a minimum 2-year follow-up. Importantly, these participants had a median age of 57.5 years, a median time since last relapse or new MRI-detected lesion of approximately 4 years, and significantly more disability than DISCOMS or DOT-MS participants. Moreover, more than 35% had 3 or more prior DMT failures. For those discontinuing fingolimod and natalizumab, there was faster time to relapse (both drugs) and progression of disability (significant only for natalizumab) compared with continuers, and this was not seen in the anti-CD20 DMT discontinuers. A similar German propensity-matched cohort of those continuing or discontinuing only ocrelizumab had a similar primary outcome, but there were discordant baseline features and a degree of new activity in both groups.32 The ages at time of discontinuation were not reported but, based on extrapolating from age at onset of ocrelizumab use and median time on the drug before discontinuation, appear to be in the mid-40s. The median Expanded Disability Status Scale score (2.5) and time since last relapse (3.4 years) suggest this group was younger and had less impairment overall at baseline compared with those in the OFSEP study. As in the OFSEP study, the time to new inflammatory activity (relapse or MRI-detected lesion) and PIRA was not faster in discontinuers (following a median of 28.3 months); however, there appeared to be substantially more PIRA in findings from the French study. This difference likely reflects the older French cohort with more disability, with a high number of people with SPMS on anti-CD20 agents. The German trial data also seemed to show that new disease activity in discontinuers began to exceed that in continuers at approximately 24 months, consistent with a potential slow wearing-off of benefit after 2 years. Interestingly, findings from a small, open-label trial looking at a 2-dose induction of ocrelizumab in recently diagnosed people with relapsing-remitting MS reported that 6 of 19 participants had MRI-only disease reactivation starting at month 19 of observation.33 A similar induction RCT29 but with 4 doses of ocrelizumab is underway and should be completed in 2030.
DMT De-Escalation
An alternative to the either/or method of stopping or continuing the same MS DMT in an older person with MS might be risk reduction by de-escalation, especially for people who have been receiving infrequently administered and highly effective (but potentially riskier) agents for long periods. De-escalation would be based on the principle that prolonged use of highly effective DMTs may not be indicated or offer enhanced benefit29 due to an overall decline in MS disease activity with age, especially when risks are likely higher with some more highly effective DMTs in older individuals.34-37 De-escalation might be accomplished by reducing the dose34 or administration frequency38,39 of the current medication or switching to an alternative with lower risks.40 De-escalation might also be a useful strategy for those taking fingolimod or natalizumab to hopefully avoid disease recurrence or rebound with abrupt discontinuation. The information base for de-escalation is rapidly evolving, but as of yet, there is no broadly accepted algorithm or strategy for implementation and there are minimal data specifically in older people with stable MS.
Future Directions
Future DMT RCTs should include adequate subgroups of participants older than 55 years to ascertain more completely the benefits and hazards associated with DMT use in this population. Similarly, treatment discontinuation algorithms41 based on presently available data offer the opportunity to design RCTs examining different approaches. It will be especially important to define the long-term consequences of modest silent MRI activity after discontinuation of MS DMTs.42 Development of postdiscontinuation, patient-specific recurrence risk profiles with machine learning would aid decision-making in the office. Present data43,44 do not suggest that analyzing serum biomarkers (eg, serum neurofilament light chain and glial fibrillary acidic protein) at the time of DMT discontinuation is predictive of disease recurrence, but this approach should be pursued further and measurements should be made longitudinally following discontinuation. Finally, DMT utility in those older than 55 years and those with LOMS needs to be examined in more detail.
Conclusions
Among people with MS older than 55 years, there is a relative dearth of data, one way or the other, arguing that presently available MS DMTs have a positive risk-benefit ratio. Subgroup analyses from phase 3 clinical trials, with the limitation of including only a few participants older than 55 years, argue that there is minimal benefit for those 55 years or older, especially if there is no recent relapse or active MRI-detected lesion activity. In older people with MS who discontinue DMT, findings from observational studies and RCTs also show that the highest risk of recurrence of MS activity is in relatively younger patients with recent disease activity. However, even among people with MS older than 60 years with long-time stable disease, available data suggest that a small risk of recurrent MS activity upon DMT discontinuation remains, especially modest new MRI activity of unclear long-term clinical significance. The risk of stopping agents such as fingolimod and natalizumab in this older group with stable disease may also be higher, though this is less clear.
As a real-world example, the regimen at the University of Colorado, Aurora, for people with stable MS older than 60 years who discontinue DMT, especially DMT not associated with rebound activity, is to examine them every 6 months and schedule brain MRI scans once a year for 2 years. Assuming 2 years of stable examination and MRI scan results, we then transition to annual examinations and MRI scans every other year until they are 65 years or older, based on discussions
with them.
Based on all available data, older people with MS may stay on their DMT if tolerated, accept some slight enhanced risk of disease recurrence with DMT discontinuation, or consider de-escalation to a DMT with lower risk, especially if they are not tolerating their present DMT or if they have had recent MS disease activity. In all 3 scenarios, ongoing monitoring for new disease activity remains a key component of MS care.
