Background

Historically, disease activity diminished during pregnancy in women with relapsing-remitting MS. Today, women with high disease activity are more likely to attempt pregnancy due to the disease control that new therapies offer. But disease activity during pregnancy in the modern day remains understudied.

Objectives

Describe disease activity in a modern pregnancy cohort, grouped by preconception disease-modifying therapy (DMT) class; determine the predictors of relapse during pregnancy.

Methods

Data were obtained from the MSBase Registry. Term/preterm pregnancies conceived from 2011-2019 were included. DMT were classed by low, moderate and high-efficacy. Annualized relapse rates (ARR) were calculated for each pregnancy trimester and 12 months either side. Predictors of relapse during pregnancy were determined using clustered logistic regression.

Results

We included 1640 pregnancies from 1452 women. DMT used in the year before conception were none (n=346), low (n=845), moderate (n=207) and high-efficacy (n=242). Most common DMT in each class was interferon-beta (n=597), fingolimod (n=147) and natalizumab (n=219) for low, moderate and high-efficacy respectively. Conception EDSS ≥2 was more common in higher efficacy DMT groups (high: 41.3%; moderate 28.5%; low 22.4%; none 20.2%). For low-efficacy and no DMT groups, ARR fell through pregnancy. ARR of the moderate-efficacy group increased in the 1^st pregnancy trimester (0.55 [95% CI 0.36-0.80] vs 0.14 [95% CI 0.10-0.21] on low-efficacy), then decreased to a trough in the third. Conversely, ARR steadily increased throughout pregnancy for those on high-efficacy DMT (3^rd trimester: 0.42 [95% CI 0.25-0.66] vs 0.12 [95% CI 0.07-0.19] on low-efficacy). Higher efficacy DMT groups were associated with higher ARR in the early postpartum period (high: 0.84 [95% CI 0.62-1.1]; moderate: 0.90 [95% CI 0.65-1.2]; low: 0.47 [95% CI 0.38-0.58]). Preconception use of high and moderate-efficacy DMT and higher preconception ARR were predictors of relapse in pregnancy. But, continuation of high-efficacy DMT into pregnancy was protective against relapse (odds ratio 0.80 [95% CI 0.68-0.94]). Age ≥35 years was associated with reduced odds of relapse.

Conclusions

Women with RRMS treated with moderate or high-efficacy DMT are at greater risk of relapse during pregnancy. Careful pregnancy management, and use of long-acting high-efficacy DMT preconception, or continuing natalizumab into pregnancy, may prevent relapse in pregnancy.

Background

Our current understanding of demographic and clinical modifiers of the effectiveness of multiple sclerosis (MS) therapies is limited.

Objectives

To assess whether patients’ response to disease modifying therapies (DMT) in MS varies by disease activity (annualised relapse rate, presence of new MRI lesions), disability, age, MS duration or disease phenotype.

Methods

Using the international MSBase registry, we selected patients with MS followed for ≥1 year, with ≥3 visits, ≥1 visit per year. Marginal structural models (MSMs) were used to compare the hazard ratios (HR) of 6-month confirmed worsening and improvement of disability (EDSS), and the incidence of relapses between treated and untreated periods. MSMs were continuously re-adjusted for patient age, sex, pregnancy, date, time from first symptom, prior relapse history, disability and MRI activity.

Results

Among 23 687 patients with relapsing MS, those on DMT experienced 20% greater chance of disability improvement [HR 1.20 (95% CI 1.0-1.5)], 47% lower risk of disability worsening [HR 0.53 (0.39-0.71)] and 51% reduction in relapses [HR 0.49 (0.43-0.55)]. The effect of DMT on relapses and EDSS worsening was attenuated with longer MS duration and higher prior relapse rate. The effect of DMT on EDSS improvement and relapses was more evident in low EDSS categories. DMT was associated with 51% EDSS improvement in patients without new MRI lesions [HR 1.51 (1.00-2.28)] compared to 4% in those with MRI activity [HR 1.04 (0.88-1.24)]. Among 26329 participants with relapsing or progressive MS, DMT was associated with 25% reduction in EDSS worsening and 42% reduction in relapses in patients with relapsing MS [HR 0.75 (0.65-0.86) and HR 0.58 (CI 0.54-62), respectively], while evidence for such beneficial effects of treatment in patients with progressive MS was not found [HR 1.11 (0.91-1.46) and HR 1.16 (0.91-1.46), respectively].

Conclusions

DMTs are associated with reduction in relapse frequency, progression of disability, and increased chance of recovery from disability. In general, the effectiveness of DMTs was most pronounced in subgroups with shorter MS duration, lower EDSS, lower relapse rate and relapsing MS phenotype.

Background

Prevention of disability over the long-term is the main treatment goal in multiple sclerosis (MS), however, randomized clinical trials provide only short-term (3-6-month) treatment effect on disability.

Objectives

To externally validate the previously developed sustained progression score which established 6-month confirmed disability progression events as indicators of long-term disability worsening in randomized double-blind CLARITY (Cladribine Tablets Treating MS Orally) trials. A higher score indicates a greater probability of a progression event to be sustained over the long-term.

Methods

Data from the CLARITY and CLARITY 2-year Extension study were used. The 3-and 6-month confirmed disability progression events and the time over which progression remained sustained were identified. Patient characteristics at the time of progression previously associated with the likelihood of improvement after confirmed progression event (age, primary progressive MS, a relapse in previous month, expanded disability status scale score≥6 and its change from baseline, number of affected functional system domains, and worsening in pyramidal, brainstem and sensory domains) were used to calculate the sustained progression score for each progression event. A Cox proportional hazards model was used to validate the association of the score calculated for each 6-month confirmed disability progression event with time to improvement. To demonstrate application of the score in trial setting, the analysis of the CLARITY study were repeated estimating the effect of therapy on long-term disability outcomes. Effect of cladribine on 3-month confirmed progression events with any score and score˃1.5 was evaluated.

Results

A total of 667 6-month confirmed disability progression events were identified in the combined dataset, of which only 12 were followed by a 6-month confirmed disability improvement. The external validation of the score illustrated a trend towards a decrease in the likelihood of disability improvement in progression events with higher score (Hazard Ratio HR=0.72, 95% CI: 0.21-2.42). Results of the original trial were confirmed in addition to that oral cladribine reduced the risk of disability progression that was likely to be sustained over the long-term (only events with score˃1.5; HR: 0.64; 95% CI: 0.47-0.87).

Conclusions

The sustained progression score, estimated using the characteristics of confirmed progression events, provides important complementary information that will allow future trials to establish the effect of therapy not only on short-term but also on long-term disability accrual.

Background

Natalizumab 300 mg every 4 weeks (Q4W) is an effective therapy for relapsing-remitting multiple sclerosis (MS) but is also associated with increased risk of progressive multifocal leukoencephalopathy (PML) in anti–JC virus seropositive patients. Analysis of the TOUCH Prescribing Program safety database showed that natalizumab extended interval dosing (EID; average dosing interval approximately 6 weeks) is associated with lower risk of PML than Q4W dosing. Previous analysis of TYSABRI Observational Program (TOP) data showed no difference in relapse outcomes for patients on Q4W and every-6-week (Q6W) dosing. Comparative disability outcome data in well-matched real-world populations are lacking.

Objectives

Compare relapse and disability outcomes in propensity-score (PS)–matched TOP patients who switched to Q6W dosing with outcomes in patients who remained on Q4W dosing.

Methods

Intentional dosing data collected in TOP as of November 2019 were used to identify patients with ≥1 year of Q4W dosing who remained on Q4W or switched to Q6W dosing. Patients with dosing intervals ≥12 weeks or <3 weeks were excluded. Patients with similar exposures were PS-matched 1:1 with age, sex, Expanded Disability Status Scale score, time from MS onset, exposure duration, and relapse activity as covariates. Between-group comparisons were made for the post-switch follow-up period for Q6W patients and the matching time period for Q4W patients. Adjusted relapse rates (ARRs) were calculated using negative binomial regression with robust standard error estimation. Hazard ratios (HRs) for time to first relapse and 24-week confirmed disability worsening (CDW) were estimated with Kaplan-Meier and Cox methods.

Results

The analysis included 236 matched pairs of Q6W and Q4W patients. Mean (SD) follow-up times for Q6W and Q4W patients were 2.00 (1.30) and 1.89 (1.15) years, respectively. ARRs (0.146 vs 0.139; P=0.796), time to first relapse (HR [95% CI] 1.078 [0.723–1.608]; P=0.711), and time to CDW (HR [95% CI] 0.749 [0.270–2.074]; P=0.578) did not differ significantly for Q6W and Q4W patients.

Conclusions

Relapse and disability outcomes in TOP were similar for PS-matched patients who switched to Q6W or remained on Q4W dosing. These results are consistent with prior matched and unmatched analyses in real-world settings and underscore the need for the ongoing, prospective, randomized efficacy trial of natalizumab Q6W vs Q4W dosing (NOVA, clinicaltrials.gov NCT03689972).

The TOP study was supported by Biogen.

Background

Disease-modifying therapies (DMT), which modify, mediate or suppress the immune system, are a major medication class for treating people with relapsing-onset multiple sclerosis (MS). However, our knowledge about these medications is largely limited to their short-term effects.

Objectives

To determine: 1) the impact of national-level DMT subsidy policy on DMT use and disability in people living with MS (PwMS); and 2) the long-term effects of DMT on disability (EDSS and MSSS) and quality of life (EQ5D5L utility score).

Methods

This project was an ecological, observational cohort study comparing populations in Australia and New Zealand with similar demographics, but markedly different levels of DMT use 10-20 years post-diagnosis. Differences between countries were assessed using standardized differences (Cohen’s d), phi coefficient and Cramer’s V. Associations were assessed with univariable and multivariable (mediation) linear regression models.

Results

We recruited 328 Australian participants, 93.9% of whom had been treated with DMT, and 256 New Zealand participants, 50.4% of whom had been treated with DMT. The Australian cohort had a longer median treatment duration (148 vs 0 months), greater proportion of disease course treated (86% vs 0%), and shorter time between diagnosis and first DMT (3 vs 24 months). The Australian cohort also had lower median EDSS (3.5 vs 4.0) and MSSS (3.05 vs 3.71), and higher quality of life (0.71 vs 0.65) at follow-up. In multivariable models, differences in DMT use significantly mediated the effect of country on disability and quality of life.

Conclusions

This large ecological study provides evidence for the impact of national level policy on DMT use and subsequent disability outcomes in PwMS. It also demonstrates that the protective effect of DMT may mediate the effect of national policy on disability progression and quality of life 10-20 years post-diagnosis in people with relapsing-onset MS.

Background

Damage to the cerebellum is common in people with multiple sclerosis (pwMS) and associated with a worse prognosis and, cerebellar relapses are associated with poorer recovery and earlier onset of progressive disease. Studies examining the importance of the cerebellum are hampered by incomplete characterisation of cerebellar damage and its relation to cerebellar function.

Objectives

We aim to examine axonal loss in the cerebellum using diffusion imaging and compare the degree of cerebellar axonal loss with cerebellar dysfunction in pwMS.

Methods

We prospectively recruited 55 pwMS and 14 healthy controls (HC). Clinical assessments included scale for the assessment and rating of ataxia (SARA) and Bain tremor ratings. Subjects underwent 3-tesla volumetric, lesion and diffusion magnetic resonance imaging. Cerebellar axonal loss was examined with fibre-specific markers. Fibre density and cross-section (FDC) accounts for microscopic and macroscopic changes in a fibre bundle.

Results

Significant loss of cerebellar FDC was found in pwMS compared to HC (p=0.03). Lower FDC was associated with increased SARA (r=-0.42, p<0.01) and tremor severity (rho=-0.35, p=0.01). Cerebellar lesion volume correlated with SARA (r=0.49, p<0.01) and tremor severity (rho=0.41, p=0.01). Cerebellar volume showed no correlation with cerebellar clinical assessments.

Conclusions

Fibre-specific measures of cerebellar pathology could provide a functionally relevant marker of cerebellar damage in pwMS. Future trials using fibre-specific markers are needed to further characterise cerebellar pathology and understand its significance in disease progression.

Background

Longitudinal cognitive trajectories in MS are heterogenous and difficult to measure. Computerised tests designed to screen broad cognitive domains may be a reliable method to detect discrete trajectories of cognitive performance

Objectives

To validate a latent class model to identify trajectories of reaction time change in people with relapsing remitting MS (pwRRMS).

Methods

The MSReactor computerised cognitive battery is a self-administered, online set of reaction time tasks assessing psychomotor function (PsychoM), visual attention (VisAtt) and working memory (WorkingM). Participants completed both 6-monthly in-clinic testing and additional remote testing. Latent class analysis was used to model the longitudinal reaction times and identify discrete cognitive trajectories within the heterogeneous data. We applied a cross validation method to confirm the optimal models for all three reaction time tasks. Briefly, the cohort was split into training and test sets (50:50) and the mean root mean square error (RMSE) calculated for the difference between training and test trajectories calculated at each day of follow up, over 100 repetitions. To determine the minimum number of tests required to reliably classify an individual into a longitudinal trajectory the dataset for each task was reduced to 3, 4 and 5 tests per participant and classification compared to the complete dataset.

Results

We included 478 pwRRMS who had completed at least 3 MSReactor tests over a minimum of 30 days follow up. In total, 3846 individual tests were included in each model (median tests/participant=5 (range 3-332), mean follow up of 774 +- 359.5 days). Three latent classes were identified for each task, with the PsychoM task identifying a group of pwRRMS with a mean predicted trajectory of slowing reaction times. In validation, the PsychoM task was the most consistent with the smallest RMSE for each of the 3 classes (68 milliseconds (ms), 95%CI 59-77ms; 61ms, 95% CI 50-72ms and 16ms, 95% CI 14-18ms) followed by the VisAtt task (RMSE = 114ms, 95ms and 29ms) and WorkingM task (RMSE = 137ms, 138ms and 46ms). Reduced datasets of 3, 4 and 5 tests per participant in the PsychoM task were able to classify participants into the trajectories identified in the full dataset model, with 83%, 86% and 90% accuracy respectively.

Conclusions

Latent class modelling of longitudinal reaction times collected with MSReactor was able to detect discrete trajectories of cognitive performance. The PsychoM task latent class model identified a group of RRMS with a mean predicted slowing of reaction times and was the most consistent model in cross validation. Performing the modelling on just 3, 4 or 5 tests per participant was highly accurate in defining latent trajectories, giving the battery clinical utility where multiple years of follow up may not be realistic.

Background

Some cohort studies have reported similar onset age and disability accrual in primary and secondary progressive MS (PPMS, SPMS); others have reported later onset and faster disability accrual in SPMS. Comparisons are complicated by differences in baseline disability and exposure to disease-modifying therapies (DMT), and by lack of a standardized definition of SPMS.

Objectives

We compared hazards of disability accrual in PPMS and SPMS patients from the MSBase cohort using multivariable Cox models, applying validated diagnostic criteria for SPMS (Lorscheider et al., Brain 2016).

Methods

Inclusion required adult-onset progressive MS; ≥ 3 recorded Expanded Disability Status Scale (EDSS) scores; and, for SPMS, initial records with EDSS ≤ 3 to allow objective identification of SPMS conversion. Phenotypes were subgrouped as active (PPMS-A, SPMS-A) if ≥ 1 progressive-phase relapse was recorded, and inactive (PPMS-N, SPMS-N) otherwise. Disability accrual was defined by sustained EDSS increases confirmed over ≥ 6 months. Hazard ratios (HR) for disability accrual were obtained using Andersen-Gill Cox models, adjusted for sex and time-varying age, disability, visit frequency, and proportion of time on DMT or immunosuppressive therapy. Sensitivity analyses were performed using (1) PPMS and SPMS diagnosed since 1995, and (2) physician-diagnosed SPMS. Cumulative probability of reaching EDSS ≥ 7 (wheelchair required) was assessed (Kaplan-Meier).

Results

5461 patients were included (1257 PPMS-N; 1308 PPMS-A; 1731 SPMS-N; 1165 SPMS-A). Age at progression onset was older in SPMS than PPMS (47.2 ± 10.2, vs. 41.5 ± 10.7 [mean ± SD]), and in the inactive subgroups of each phenotype. Hazard of disability accrual was decreased in SPMS relative to PPMS (HR 0.85; 95% CI 0.78–0.92); decreased by proportion of time on DMT (HR 0.99 per 10% increment; 0.98–0.99); and higher in males (1.18; 1.12–1.25). Relative to PPMS-N, hazard was decreased in SPMS-A (0.79; 0.71–0.87) but similar for PPMS-A (1.01; 0.93–1.10) and SPMS-N (0.94; 0.85–1.05). Sensitivity analyses corroborated these results. However, patients with SPMS-A reached EDSS ≥ 7 at younger ages (cumulative probability 30% by 57, vs. 64–66 for SPMS-N, PPMS-A, PPMS-N).

Conclusions

Progressive phase onset is later in SPMS than PPMS. Hazard of disability accrual during the progressive phase is lower in SPMS than PPMS. However, patients with SPMS-A reach wheelchair requirement younger than other progressive phenotypes, reflecting earlier progression onset versus SPMS-N, and greater disability at onset versus PPMS

Background

In secondary progressive multiple sclerosis (SPMS), reduction in the rates of disability accrual after starting disease modifying therapy (DMT) has largely been limited to patients with ongoing inflammatory activity. A delayed treatment effect, termed therapeutic lag, may obscure therapeutic benefits in SPMS.

Objectives

To compare the effect of high and low efficacy DMT on disability outcomes in patients with recently active and inactive SPMS after accounting for therapeutic lag.

Methods

Using data from MSBase, a multinational MS registry, and OFSEP, the French MS registry, we identified patients with SPMS as per a previously validated objective definition. We identified patients treated with high- (natalizumab, alemtuzumab, mitoxantrone, ocrelizumab, rituximab, cladribine, fingolimod) or low-efficacy (interferons, glatiramer acetate, teriflunomide) DMT after SPMS onset. Based on our previous work, an individualised estimate of duration of therapeutic lag was calculated for each patient. Only events that occurred after the estimated therapeutic lag period were included in the analysis. Propensity score matching was used to select groups with comparable baseline characteristics. Disability and relapse outcomes were compared in paired, pairwise-censored analyses adjusted for visit density.

Results

Of 7359 patients with SPMS, 1000 patients fulfilled the criteria for study inclusion (510 active SPMS, 490 inactive SPMS). For the relapse outcomes, patients with active SPMS treated with high-efficacy DMTs experienced lower probabilities of relapses than low-efficacy DMTs (hazard ratio [HR] 0.7 [95%CI 0.5-0.9], p=0.006). Patients with inactive SPMS had similar probabilities of relapses in the high and low efficacy DMT groups (0.8 [0.6-1.2], p=0.39). No difference in the risk of 6-month sustained disability accumulation, or proportion of patients reaching EDSS>=7, was observed between groups when accounting for therapeutic lag.

Conclusions

The risk of disability accumulation in SPMS seems to be comparable in patients treated with high- and low- efficacy DMT. High efficacy DMT is superior to low efficacy therapy in reducing relapse activity in patients with active SPMS, but not those with inactive SPMS. Pre-treatment inflammatory activity, clinical or radiological, is a treatable target in SPMS which may benefit from higher-efficacy anti-inflammatory therapies.

Background

Unemployment rates can be high among patients with multiple sclerosis (MS), and a return to work after unemployment can be difficult, highlighting the importance of treatment in preventing a departure from the workforce due to MS. Natalizumab is a highly effective treatment for patients with relapsing-remitting MS (RRMS) and was associated with positive employment outcomes in real-world studies.

Objectives

To evaluate changes in employment status in RRMS patients treated with natalizumab in the TYSABRI Observational Program (TOP), a large observational study assessing the long-term safety and effectiveness of natalizumab.

Methods

This retrospective analysis included patients aged ≤65 years at TOP enrolment (i.e., baseline [BL]) who were surveyed on their employment status in the year before natalizumab initiation and in the period since treatment initiation (N=2004). Multivariate logistic regression tested the association between BL characteristics and employment outcomes.

Results

At BL, patients had a mean (standard deviation [SD]) Expanded Disability Status Scale (EDSS) score of 3.5 (1.5). At the survey, patients had a mean (SD) of 5.5 (3.3) years of natalizumab treatment. Survey responses indicated that in the year before natalizumab initiation, 1107 patients (55.2%) were working; 814 patients (40.6%) were working full time, 53 (2.6%) were working part time due to MS, 265 (13.2%) were not working due to MS, and 240 (12.0%) and 632 (31.5%) were working part time or not at all, respectively, for other reasons. After natalizumab initiation, 861 patients (43.0%) improved (1.3%) or maintained (41.6%) their employment level, whereas 170 (8.5%) experienced a decline in employment level due to MS and 256 (12.8%) remained unemployed due to MS. Significant predictors of improving/maintaining employment status were younger age (adjusted odds ratio [aOR]: 0.756; P=0.005), lower BL EDSS score (aOR: 0.747; P<0.001), and fewer relapses in the year before natalizumab initiation (aOR: 0.829; P=0.042), but did not include sex, prior therapy use, or RRMS duration.

Conclusions

Of patients who were working prior to natalizumab initiation, 77.0% maintained or improved their employment level with an average follow up of 5.5 years. Overall, favourable outcomes were predicted by younger age and less BL disease activity, supporting the importance of natalizumab initiation early in the disease course to help prevent patients from leaving the workforce due to MS.

The TOP study is funded by Biogen. Biogen funded the analyses and writing support for this abstract. Writing support was provided by Ashfield Healthcare Communications (Middletown, CT, USA).

Background

A strategy for timely multiple sclerosis (MS) care was described in the policy report, Brain health: time matters in multiple sclerosis. Building on this report, multiple stakeholder groups participated in a modified Delphi process to define acceptable, good and high-quality brain health-focused MS care. These benchmarks were incorporated into an Excel-based quality improvement (QI) tool. The first prototype of this tool was piloted in three MS centers; local analysis of results led to improvements in clinical practice in those centers.

Objectives

We aimed to improve the clinical usability of the QI tool and to test the applicability of a refined version in different healthcare settings.

Methods

The recommendations from all three centers that participated in the initial pilot study were gathered and used to prepare a refined prototype of the QI tool (prototype II). MS centers worldwide have been invited to conduct a service evaluation using prototype II as part of a larger pilot study of 10–20 MS centers across a broad geographical area. Each participating site will review the medical records of 36 adults with MS (at representative stages of the care pathway) and input the data requested into the tool. To assess whether the QI tool can be applied in MS centers globally, study sites will be asked to complete a survey following their service evaluation. The survey asks about ease of use of the tool, its usefulness for facilitating local change, relevance of the data captured and key data for repeated use.

Results

Prototype II has separate spreadsheets for entering information on patients at different stages of the care pathway; fields are tailored to the different patient populations so there is less data to input per patient. Data validation programming prevents the insertion of invalid information. To assist MS centers in analyzing their findings, improved visual summaries of clinic-level and patient-level results are generated within the tool; these auto-populate when the required fields in the data input spreadsheets are completed. Prototype II will also support future language translations. More than 18 MS centers have so far expressed interest in trialing prototype II of the QI tool; preliminary insights from selected study sites will be presented.

Conclusions

Following further refinements, widespread roll-out of the QI tool will enable MS centers to collect data to benchmark their clinical standards and to support service improvement.

Background

Disease activity has been investigated in pregnant women with RRMS treated with low-efficacy or no therapy. How newer, more efficacious therapies affect relapse and disability progression risk after pregnancy remains understudied.

Objectives

To describe disease activity in a modern pregnancy cohort contrasted with historical cohorts. To determine the predictors of postpartum relapse and the predictors of six-month confirmed disability progression events in a contemporary pregnancy cohort.

Methods

Data were obtained from the MSBase Registry. Term/preterm pregnancies conceived from 2011-2019 (modern cohort) were compared with those conceived between 2005-2010 and pre-2005. Annualised relapse rates (ARR) were calculated for each pregnancy trimester and 12 months either side. Predictors of time-to-relapse postpartum (1^st 3 months) and time to 6-month confirmed disability progression event were determined with clustered Cox regression analyses. Breastfeeding duration and time to DMT reinitiation were modelled as time-varying covariates.

Results

We included 1640 pregnancies from 1452 women (modern cohort). Disease-modifying therapy (DMT) used in the year before conception included interferon-beta (n=597), natalizumab (n=219) and fingolimod (n=147). Continuation of DMT up to conception increased over time (31% pre-2005 vs 54% modern cohort). Preconception ARR decreased across epochs (pre-2005: 0·58 [95% CI 0·49-0·70]; 2005-2010: 0·40 [95% CI 0·36-0·45]; modern: 0·29 [95% CI 0·27-0·32]). In all epochs, ARR decreased during pregnancy to reach similar troughs in the 3^rd trimester, and rebounded in the 1^st 3-months postpartum. Preconception use of high-efficacy DMT predicted early postpartum relapse (hazard ratio (HR) 2.1 [1.4-3.1]); although those on no DMT were also at risk of postpartum relapse, relative to women on low-efficacy DMT (HR 2.7 [1.2-5.9]). Conception EDSS ≥2, higher preconception and in-pregnancy ARR were also risk factors. DMT reinitiation, particularly of high-efficacy DMT (HR 0.17 [0.07-0.38]), was protective against postpartum relapse. Women who breastfed were less likely to relapse (HR 0.63 [0.42-0.94]). 4.5% of modern pregnancies had confirmed disability progression after delivery. This was predicted by higher pregnancy and postpartum ARR, with postpartum ARR remaining independently predictive in multivariable analysis (HR 1.5 [1.2-2.0]).

Conclusions

The early postpartum period remains a period of vulnerability for disease rebound in women with MS in the modern era. Early DMT reinitiation, particularly with high-efficacy treatment, is protective against postpartum relapse. Confirmed disability progression events after pregnnacy are uncommon in the modern era. Relapse activity is the key driver of these events.

Background

Natalizumab 300 mg every 4 weeks (Q4W) is an effective therapy for relapsing-remitting multiple sclerosis (MS) but is also associated with increased risk of progressive multifocal leukoencephalopathy (PML) in anti–JC virus seropositive patients. Analysis of the TOUCH Prescribing Program safety database showed that natalizumab extended interval dosing (EID; average dosing interval approximately 6 weeks) is associated with lower risk of PML than Q4W dosing. Previous analysis of TYSABRI Observational Program (TOP) data showed no difference in relapse outcomes for patients on Q4W and every-6-week (Q6W) dosing. Comparative disability outcome data in well-matched real-world populations are lacking.

Objectives

Compare relapse and disability outcomes in propensity-score (PS)–matched TOP patients who switched to Q6W dosing with outcomes in patients who remained on Q4W dosing.

Methods

Intentional dosing data collected in TOP as of November 2019 were used to identify patients with ≥1 year of Q4W dosing who remained on Q4W or switched to Q6W dosing. Patients with dosing intervals ≥12 weeks or <3 weeks were excluded. Patients with similar exposures were PS-matched 1:1 with age, sex, Expanded Disability Status Scale score, time from MS onset, exposure duration, and relapse activity as covariates. Between-group comparisons were made for the post-switch follow-up period for Q6W patients and the matching time period for Q4W patients. Adjusted relapse rates (ARRs) were calculated using negative binomial regression with robust standard error estimation. Hazard ratios (HRs) for time to first relapse and 24-week confirmed disability worsening (CDW) were estimated with Kaplan-Meier and Cox methods.

Results

The analysis included 236 matched pairs of Q6W and Q4W patients. Mean (SD) follow-up times for Q6W and Q4W patients were 2.00 (1.30) and 1.89 (1.15) years, respectively. ARRs (0.146 vs 0.139; P=0.796), time to first relapse (HR [95% CI] 1.078 [0.723–1.608]; P=0.711), and time to CDW (HR [95% CI] 0.749 [0.270–2.074]; P=0.578) did not differ significantly for Q6W and Q4W patients.

Conclusions

Relapse and disability outcomes in TOP were similar for PS-matched patients who switched to Q6W or remained on Q4W dosing. These results are consistent with prior matched and unmatched analyses in real-world settings and underscore the need for the ongoing, prospective, randomized efficacy trial of natalizumab Q6W vs Q4W dosing (NOVA, clinicaltrials.gov NCT03689972).

The TOP study was supported by Biogen.