FormalPara Key Summary Points

People with multiple sclerosis may have concerns over immune suppression.

We assessed outcomes in patients who switched from ocrelizumab to diroximel fumarate.

No participants relapsed during 1-year follow-up and all remained on treatment

No significant change in peripheral immune parameters, except increased CD19+ cells.

No significant changes in cognitive, ocular, or patient-reported outcomes were observed.

Introduction

There are currently over 20 disease-modifying therapies (DMTs) available for the treatment of relapsing–remitting multiple sclerosis (RRMS) including oral, injectable, and infused medications [1]. The choice of DMT in people with multiple sclerosis (MS) is typically considered on the basis of efficacy and safety, as well as frequency of treatment and route of administration [2]. Escalation, de-escalation, or switching of DMT may be considered if a DMT is no longer optimal for a person with MS [3, 4], and understanding when to make such treatment decisions is of key importance to optimizing the balance of risks and benefits [5]. During the recent global COVID-19 pandemic, other factors related to DMT choice, such as the ability to mount an effective response to vaccination and risk of infection, had to be carefully considered.

In RRMS, B cell-depleting therapies have demonstrated marked reduction in relapses and in disability worsening, and their effect on disease activity as measured by magnetic resonance imaging (MRI) is robust [6]. However, their high efficacy is not without risk. B cell-depleting therapies are reported to be associated with a reduction of protective immunoglobulin synthesis [7], increased rate of infections [8,9,10], and impaired humoral immune response to vaccination [11, 12], albeit T cell response to vaccination may be preserved [13]. Infections are of particular concern as they have been reported to cause pseudo-relapses [14, 15] and, consequently, worsening of neurologic symptoms [16]. Furthermore, there is evidence to suggest that exacerbations associated with systemic infection in people with MS may lead to more sustained neurological damage than non-infection-related exacerbations [17]. Should de-escalation be considered as a result of concerns of immune deficiency, infection risk, impaired vaccine response, immunosenescence or other factors, a DMT from the fumarates class may be an alternative option for disease control. The benefits include a potential reduced risk of infections [8] and improved ability to mount antibody responses to both infections and vaccines compared with B cell-depleting therapies [18, 19]. In addition, a de-escalation strategy may offer the potential for better disease control in some patients compared to DMT discontinuation. In a randomized, controlled, non-inferiority trial, there was an increased risk of new MRI activity in patients older than 55 years with stable disease who discontinued DMT compared to those who remained on treatment [20].

Diroximel fumarate (DRF) is an orally administered fumarate approved in the USA for adults with relapsing forms of MS, and in Europe for adults with RRMS [21, 22]. As of 31 December 2022, approximately 33,989 people had been treated with DRF, representing 35,420 patient-years of exposure. Of these, 1477 people (1718 patient-years) were from clinical trials [23]. Switching treatment to DRF has been reported previously for people with MS who had received glatiramer acetate or interferons as their most recent DMT before initiating DRF in the open-label, phase 3 EVOLVE-MS-1 study (NCT02634307); in these participants, no new safety signals were observed for DRF, and favorable clinical outcomes were reported [24].

During the COVID-19 pandemic, in the pre-COVID vaccine era, several studies have shown increased risk of severe COVID-19, hospitalization, ICU admission, the need for artificial ventilation, and mortality among people with MS who were on anti-CD20 medications [25, 26]. At the height of the pandemic, many patients raised concern about continuing ocrelizumab (OCRE) in the setting of increased risk of severe COVID-19 infection. For those who were stable for at least 1 year on OCRE and for whom the risk of severe COVID-19 was deemed greater than the risk of MS exacerbation, a shared decision was made during a routine follow-up meeting to de-escalate treatment. DRF was offered as a suitable substitute owing to its short elimination time, if needed during active infection and because no COVID-19-related safety concerns had been raised with regard to this DMT.

The purpose of this study was to evaluate and track changes in functional ability and overall disease stability in people with RRMS who were switched from the B cell-depleting DMT OCRE to DRF by examining clinical outcomes as well as changes found in peripheral immune parameters, cognitive functioning, optical coherence tomography (OCT) testing, MRI, and patient-reported outcomes (PROs).

Methods

Study Design and Participants

In this real-world, single-center study, all people with RRMS under our care, who switched from treatment with OCRE to DRF, as a result of COVID-19-related concerns were included.

Data were collected at baseline (DRF initiation) and 1 year after the switch. The washout period was calculated as the duration between the last infusion date of OCRE and the first dose of DRF. Although the exact day a patient took the medication based on chart notes alone is difficult to confirm, the closest DRF start dates were noted on the basis of all available documentation. The washout period between treatment switch ranged from 4.2 to 18.4 months (Supplementary Table 1). Those eligible for inclusion were patients treated at South Shore Neurologic Associates Comprehensive MS Care Center, aged ≥ 18 years, had a confirmed diagnosis of RRMS according to McDonald criteria [27], had been clinically stable for ≥ 1 year on OCRE, with an Expanded Disability Status Scale (EDSS) score between 0 and 6.5. Participants were excluded if they had a history of head injury, seizures, or neurological conditions other than MS; a history of drug or alcohol abuse; active psychosis; immunoglobulin (Ig) G < 300 mg/dL; or absolute lymphocyte count (ALC) < 0.5 × 109/L. The exclusion criterion for ALCs was set because fumarates have a recommendation to discontinue therapy if patients have persistent severe lymphopenia (ALC < 0.5 × 109/L).

Data Collection

Data were collected for relapse history; MRI; blood work for assessment of peripheral immune parameters, including concentrations of IgG, IgA, IgM, and IgG subclasses 1–4, and levels of CD4+ , CD8+ , and CD19+ cells; a Cognitive Assessment Battery (CAB) validated for people with MS (NeuroTrax™; NeuroTrax, Medina, NY) assessing memory, processing speed, visuospatial and executive function, among others [28]; OCT using a Heidelberg SPECTRALIS® OCT imaging platform (Heidelberg Engineering Inc., Franklin, MA); and Patient Determined Disease Steps (PDDS), a PRO developed by the North American Research Committee on MS (NARCOMS). PDDS scores range from 0 (normal) to 8 (bedridden). Clinical and PROs are summarized in Table 1.

Table 1 Clinical and patient-reported outcomes assessed
Full size table

Brain MRI images were obtained on a 1.5 T, Hitachi Echelon scanner. High-resolution 3D T1-weighted sequences were used to count gadolinium-enhancing lesions (acquisition voxel, 1 × 1 × 1 mm3; field of view, 224 × 224 nm2; inversion time, 666 ms; repetition time, 10.8 ms; echo time, 3.7 ms; flip angle, 10°). Fluid-attenuated inversion recovery (FLAIR) sequences were used for hyperintense lesion count (acquisition voxel, 1 × 1 × 2 mm3; field of view, 256 × 256 mm2; inversion time, 2600 ms; repetition time, 10,179 ms; echo time, 90 ms; flip angle, 90°).

Statistical Analysis

Protocol-defined MS relapse was defined as new or recurrent neurologic symptoms (not associated with fever/infection) lasting ≥ 24 h and accompanied by new neurological findings and change in EDSS score. Paired t tests were used to determine statistically significant differences between data collected at baseline and 1 year post switch (p < 0.05).

Ethics

The study was conducted according to local regulations and in compliance with the principles of the Declaration of Helsinki 1964. All instructions, regulations, and agreements of the protocol, and applicable International Conference on Harmonisation guidelines, were adhered to, and the study protocol was approved.

Results

Participants

Twenty-five people with RRMS who switched from OCRE to DRF met the eligibility criteria and were included in the study. The reason for DMT change for all participants was concern over immunosuppression during the COVID-19 pandemic. Initially 32 patients were identified to switch; however, five were excluded as a result of having < 1 year of being clinically stable on OCRE, and two were excluded for having EDSS scores > 6.5. On average, participants were mean (SD) age of 52 (9) years, and 16/25 (64%) were female. Sixteen percent of patients had a relapse in the year prior to OCRE initiation, but none had active MRI (Table 2). OCRE was started in some people with MS with inactive disease because it is well known that MS-related disease progression may continue independent of relapse and MRI activity and high-efficacy DMTs, such as OCRE, can prevent this silent progression more efficiently than low-efficacy DMTs [29].

Table 2 Baseline demographics and disease characteristics
Full size table

The mean (SD) duration of treatment with OCRE before the switch to DRF was 26 (8) months. No relapses were recorded on OCRE. Only one patient had several non-enhancing new T2 lesions on MRI, performed on OCRE, before switching to DRF, 12 months following OCRE start, compared to a previous scan, 2 months prior to OCRE start. This patient had a clinical relapse immediately prior to OCRE initiation. The mean (SD) interval from the last dose of OCRE to initiation of DRF was 8.3 (3.4) months. All patients received OCRE infusions every 6 months. Study follow-up was 1 year for all 25 patients; zero patients were lost to follow-up.

Relapse Rate, MRI, Persistence, and Patient-Reported Outcomes

No participants relapsed on DRF during follow-up, and all remained persistent on DRF at 1 year (Table 3). MRI activity, as assessed 6–12 months post DRF start, remained stable, and none of the patients had active MRI after de-escalation to DRF. There was no significant difference observed between scores on any of the PROs assessed at baseline (DRF initiation) and at 1 year following the switch to DRF (n = 17; all p > 0.05) (Fig. 1 and Supplementary Fig. 1).

Table 3 Relapse, MRI, and persistence outcomes 1 year post switch from OCRE to DRF
Full size table
Fig. 1
figure 1

Median PRO scores at baseline on OCRE and 1 year post switch to DRF. Median and interquartile range is shown. Higher scores indicate worse outcome for PROs, except for PROMIS Social Roles. DRF diroximel fumarate, HADS-A Hospital Anxiety and Depression Scale—Anxiety, HADS-D Hospital Anxiety and Depression Scale—Depression, MFES Modified Falls Efficacy Scale, MSWS-12 12-Item Multiple Sclerosis Walking Scale, OCRE ocrelizumab, PDSS Patient Determined Disease Steps, PRO patient-reported outcome, PROMIS Patient-Reported Outcome Measurement Information System

Full size image

Cognitive Assessment Battery

There was no significant difference observed between CAB assessments at baseline (DRF initiation) and those carried out 1 year after switching to DRF (n = 17; all p > 0.05), with most CAB parameters representing average cognitive function at both assessment points (Fig. 2).

Fig. 2
figure 2

Mean CAB parameters at baseline on OCRE and 1 year post switch to DRF. Error bars represent SD. Cognitive Assessment Domain Scores: > 115 = above average; 115–100 = average; 99–85 = below average; < 85 = abnormal. CAB Cognitive Assessment Battery, DRF diroximel fumarate, OCRE ocrelizumab, SD standard deviation

Full size image

Ocular Tomography

OCT assessments showed no significant difference between retinal nerve fiber layer (RNFL) thickness measured at baseline following treatment with OCRE and at 1 year after the switch to DRF (n = 12; all p > 0.05) (Table 4).

Table 4 OCT outcomes at baseline and 1 year post switch
Full size table

Blood Work

Peripheral immune parameters were available for 23 patients before starting DRF, as well as for 15 patients after a year on DRF (Table 5). There were no significant changes in levels of IgG, IgA, IgM, IgG subclasses 1–4, CD4+ cells, CD8+ cells, or CD4+/CD8+ ratio between baseline (DRF initiation) and 1 year post switch (all p > 0.05 for the 15 patients with available measurements at two time points). There was, however, a significant (p < 0.05) increase in the percentage of CD19+ cells 1 year after switching to DRF compared with baseline measurements following treatment with OCRE (Table 5). Notably, 7 out of 15 patients (47%) for whom CD19+ count on DRF was available still had CD19+ count below the lower limit of normal, i.e., < 70 cells/µL. Mean (SD) ALC count was 1327 (343) cells/µL at baseline versus 1480 (704) cells/µL at 1 year post switch.

Table 5 Peripheral immune parameters at baseline on OCRE and 1 year post switch to DRF
Full size table

Adverse Effects

One patient on OCRE reported significant post-infusion fatigue. Five patients on DRF reported flushing and four patients on DRF reported gastrointestinal discomfort.

Discussion

In this study of people with RRMS who switched from OCRE to DRF treatment as a result of concerns around immunosuppression during routine clinical care, all participants remained relapse free and had no MRI activity after 1 year of follow-up. After the switch in treatment, there were no significant changes in peripheral immune parameters apart from the expected rise in CD19+ B cell count, cognitive outcomes, or OCT outcomes (all p > 0.05); PROs also remained stable. Mean PDDS showed a small increase from 3.83 at baseline to 4.23 after 1 year on DRF, but this was not statistically significant.

This study is the first to report outcomes in a cohort of clinically stable people with MS who switched from OCRE to DRF. Although the results should be interpreted with caution because of a limited sample size and duration of follow-up, they are encouraging. The findings from this study support DRF as a potential effective treatment option for people who have been stable on OCRE but may need to switch for reasons unrelated to effectiveness.

Making decisions about when to escalate, de-escalate, or switch DMTs should be based on disease activity. Disease activity quantified by MRI is a strong determinant of treatment switch, particularly in younger people with RRMS [30]. In our study of adults with RRMS, we used OCT as one of the measures to monitor disease progression. OCT is a rapid, noninvasive, high-resolution imaging technique [31] that can be applied to provide objective, examiner-independent measurements that do not rely on EDSS, MRI, or neurologic assessment. OCT measurements have previously demonstrated a thinner RNFL in individuals with MS versus healthy volunteers [32]; moreover, cross-sectional and longitudinal monitoring of RNFL thickness has been shown to predict both physical and cognitive decline in MS [33,34,35]. Our OCT findings of no significant change in RNFL thickness 1 year following the switch to DRF are therefore reassuring, given that previous studies report OCT cutoffs for cross-sectional measurements of the RNFL of ≤ 87 or ≤ 88 μm (depending on OCT device used) [34, 35], and rate of peripapillary RNFL thinning of > 1.5 μm/year [34], as increasing the risk for disease progression in subsequent years. Supporting the outcomes of the quantitative measures, PROs were assessed across several domains, and all remained stable 1 year after switching to DRF. PROs allow for better understanding of people’s interests and needs regarding their treatment and may also encourage patient participation in treatment management decisions [24, 36, 37].

As expected, we observed a significant increase in CD19+ B cells after 1 year of DRF treatment (Table 5), although, notably, about half of available CD19 counts were still below the lower limit of normal after 1 year on DRF. The slow rate of B cell repopulation undoubtedly contributed to maintained stability. Still, it is reassuring that recurrence of disease activity was not observed, despite the clear increase in CD19 count after switching to DRF.

The importance of B cells in MS pathophysiology has been widely published, and previous studies of anti-CD20 therapies have reported a prolonged duration of action beyond their continuous administration period. Depletion data have shown a median time to B cell repletion of 72 weeks (range 27–175) following a final infusion of 600 mg OCRE; lower limit of normal was defined as 80 cells/μl [38]. A phase 2 study of 104 people with RRMS treated with a single dose of rituximab resulted in remission from both clinical and MRI disease activity over 48 weeks [39]. Rituximab maintenance treatment leads to sustained B cell depletion, although early B cell repopulation has been observed in people with MS during the first 2 years of treatment. A recent study reported that early repopulation of B cells (> 5 cells/mm3) in rituximab-treated people occurred at least once in 38.5% of patients (n = 70/182 patients) with MS and was not associated with a risk of relapse or clinical worsening. Interestingly, in some patients it was correlated with increased clinical benefit [40]. In a retrospective observational study of 225 people with MS conducted in Sweden, no statistically significant differences in relapse rate or MRI activity were observed between people with MS who stopped using rituximab and those who remained on regular treatment [41].

In the phase 2 study of OCRE, 133 people with MS were monitored for 48 weeks without treatment, after previously receiving three cycles of therapy [42]. The reduced relapse rate and MRI activity that was reported persisted into the treatment-free period despite CD19+ B cell repopulation. Our data further support a potential prolonged period of effectiveness of anti-CD20 therapies beyond their regular administration period, in addition to the beneficial effects of DRF administration. Some studies have addressed the differential repopulation of B cells in patients with RRMS after treatment with DRF—including increased numbers of circulating B cells with naïve and regulatory capacity [43,44,45] and altered cytokine production [44]. Therefore, an altered repopulation of B cells has been suggested as a potential mechanism for this long-standing efficacy [46]. Additionally, a possible combination mechanism between the two treatments might account for the longer-term efficacy as people transition from anti-CD20 to DRF. OCRE has been found to be safe and effective in treating people with MS [47, 48]. However, it is important to consider the differences between DMTs in terms of longer-term risk of infection [10] and patient response to vaccination—particularly when looking at when to switch treatments. A retrospective analysis of national data collected prospectively and longitudinally in England found a substantial increase in the risk of SARS-CoV-2 infection in people with MS treated with OCRE compared with the general population, despite widespread vaccination, although potential confounders such as age, comorbidities, and neurological disability were not taken into account in this study [49]. Studies have shown that people taking OCRE and other B cell-depleting DMTs have reduced antibody and memory B cell responses to SARS-CoV-2 vaccines [12, 50,51,52,53]. However, these people can mount a vaccine-specific T cell response [50,51,52], and T cells might provide protection from severe disease by limiting viral replication to the upper respiratory tract [51, 54]. The failure to mount a humoral response to SARS-CoV-2 vaccination with anti-CD20 therapies is likely due to the mode of action of this class of DMTs. In contrast, the humoral response to SARS-CoV-2 vaccination appears to be well preserved in people with MS treated with fumarates [19, 53].

The COVID-19 pandemic introduced new uncertainties into clinical decision-making regarding the balance between efficacy and safety of DMTs. Recent epidemiological studies have shown that despite increased risk of breakthrough COVID-19 on high efficacy DMTs, such as OCRE, vaccination may still be effective in preventing severe infection regardless of DMT [55, 56]. Taken together, DMT choice in the post-vaccination COVID-19 era should still be guided first and foremost by MS disease activity; however, the aforementioned signals of increased risk of breakthrough infection on high efficacy DMTs, such as OCRE, call for considering personal risk of infection during therapeutic decision-making, especially if MS is not highly active.

Another explanation of the apparent stability of this cohort despite de-escalation may be related to age. Indeed, transitioning from a highly efficacious DMT (i.e., OCRE) to a less potent one (i.e., DRF) could entail a reduced level of risk, especially within an older population akin to the study sample. It has been suggested by a meta-analysis of randomized, blinded, MS clinical trials that age has a defining effect on the therapeutic efficacy of immunomodulatory DMTs, so that the differential efficacy of DMTs in inhibiting MS disability is lost with ageing [57]. It was reported that differences between higher-efficacy and moderate-efficacy treatments are typically only demonstrated during earlier stages of MS; therefore, the predicted benefit of receiving immunomodulatory DMTs after the age of 53 years may be limited when considered alongside the exposure to treatment-associated risks [57]. This highlights that higher-efficacy treatments may have more potential benefit in younger people with MS, which should be taken into consideration when initiating and switching therapies. Furthermore, clinical trials evaluating anti-CD20 therapy in progressive MS primarily found more benefit in younger, less disabled people with more inflammatory disease activity. One study reported that 40% of older people with progressive MS still progressed on anti-CD20 therapy (60% remained stable or improved) after 2 years on OCRE, impacting the overall benefit/risk ratio of this treatment [58].

Limitations of this study include those inherent to real-world studies of people with MS in routine clinical care, such as confounding factors, the absence of a comparison arm, and outcomes not being available for all participants. The examined population, being a limited sample, at least partly comprising patients with modest disease activity, impacts the generalizability of these findings. The brief follow-up duration further limits assessing the net value of switching to DRF compared to OCRE’s prolonged effects. This time constraint limits the comprehensive understanding of the long-term implications of the treatment switch. In addition to the limited follow-up time of the study, the sample size of the study was modest since the study was performed proactively during the COVID-19 pandemic, before much of the data became available indicating increased risks of infections and complications on B cell therapies. In addition, longer follow-up time may be required in order to detect significant changes on RNFL thickness.

Conclusion

This analysis characterizes outcomes in people with RRMS who switched from OCRE to DRF in routine care as a result of concerns over immunosuppression during the global COVID-19 pandemic, with a median washout duration of 7 months (range 4–18 months) since the last OCRE infusion. After 1 year of DRF treatment, all patients remained relapse free as well as free from MRI activity. No participants discontinued DRF after 1 year of treatment. PROs, cognition, and OCT outcomes remained stable and there were no significant changes in peripheral immune parameters, aside from increased CD19 counts. These findings suggest that transition to DRF from OCRE because of safety considerations is a potential reasonable treatment strategy for people with RRMS who are stable on OCRE. On the basis of the limited data presented herein, the switch might be feasible primarily for patients exhibiting modest disease activity and of relatively advanced age. Further exploration is warranted prospectively in a larger patient population with longer follow-up.