Treatment Persistence of Cladribine Tablets Versus Other Oral Disease-modifying Treatments for Multiple Sclerosis: Results from Danish, Norwegian, and Swedish Registries
Timothy Spelman1, Jan Harald Aarseth2, Elisabeth Celius3, Stig Wergeland4, Melinda Magyari5, Elisabeth Framke5, Namita Tundia6, Schiffon Wong6, Jan Hillert1
1Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden, 2Norwegian MS Registry and Biobank, Department of Neurology, Haukeland University Hospital, Bergen, Norway, 3Department of Neurology, Oslo University Hospital and Institute of Clinical Medicine, University of Oslo, Oslo, Norway, 4Norwegian MS Registry and Biobank, Department of Neurology, Haukeland University Hospital, Bergen, Norway; and Department of Clinical Medicine, University of Bergen, Bergen, Norway, 5The Danish Multiple Sclerosis Registry, Copenhagen University Hospital, Rigshospitalet, Glostrup, Denmark; and Danish Multiple Sclerosis Center, Department of Neurology, Copenhagen University Hospital Rigshospitalet, Glostrup, Copenhagen, Denmark, 6EMD Serono, Billerica, MA, USA
Objective:
Compare real-world time-to-treatment switch and discontinuation in adult patients receiving cladribine tablets (CladT; 3.5mg/kg cumulative dose over 2 years) versus other oral disease-modifying therapies (DMTs).
Background:
Generating Learnings In MultiPle SclErosis (GLIMPSE) is a longitudinal study of prospectively collected registry data to compare real-world treatment outcomes in patients with relapsing multiple sclerosis treated with CladT versus other oral DMTs (fingolimod [FING], dimethyl fumarate (DMF), and teriflunomide [TER]).
Design/Methods:
Data from Danish, Norwegian, and Swedish multiple sclerosis registries were pooled for treatment-naïve and first-switch patients. Differences in treatment persistence between first-line CladT and comparators were analysed using a propensity score based Inverse Probability of Treatment Weighted marginal Cox model. Outcome definitions: treatment switch=initiation of a new DMT within six months of interrupting index treatment; discontinuation=any treatment interruption with or without initiation of a subsequent DMT at any point.
Results:
In the treatment-naïve cohort (n=3621), CladT was significantly associated with a 73%, 72%, and 79% reduction in the rate of treatment switching vs FING (Hazard Ratio [HR] 0.27; 95% confidence interval [CI] 0.17–0.43), DMF (HR 0.28; 95%CI 0.20–0.40), and TER (HR 0.21; 95%CI 0.13–0.32), respectively. CladT was associated with 75%, 78%, and 82% reductions in the rate of treatment discontinuation vs FING (HR 0.25; 95%CI 0.16–0.39), DMF (HR 0.22; 95%CI 0.16–0.32), and TER (HR 0.18; 95%CI 0.12–0.28), respectively. In the first-switch cohort (n=1721), CladT was significantly associated with a 41% reduction in the rate of a subsequent switch vs FING (HR 0.59; 95%CI 0.40–0.88), a 56% reduction vs DMF (HR 0.44; 95%CI 0.31–0.64), and a 50% reduction vs TER (HR 0.50; 95%CI 0.34–0.74).
Conclusions:
After correcting for confounder imbalance, CladT was associated with significantly longer time-to-treatment switch and lower rate of discontinuation vs comparators, in both treatment-naïve and first-switch cohorts.