Comparative effectiveness, safety and persistence of ocrelizumab versus natalizumab in multiple sclerosis: A real-world, multi-center, propensity score-matched study.

Authors: Elena Barbuti^a∙ Alessia Castiello^b ∙ Valeria Pozzilli^c ∙ Antonio Carotenuto^b ∙ Ilaria Tomasso^a ∙ Marcello Moccia^d ∙ Serena Ruggieri^e ∙ Giovanna Borriello^f,g ∙ Roberta Lanzillo^b ∙ Vincenzo Brescia Morra^b ∙ Carlo Pozzilli^a ∙ Maria Petracca^a.

a. Department of Human Neurosciences, Sapienza University, Rome, Italy, b Department of Neurosciences, Reproductive Science and Odontostomatology, University of Naples “Federico II”, Naples, Italy, c. Unit of Neurology, Neurophysiology and Neurobiology, Campus Bio-Medico University, Rome, Italy, d Department of Molecular Medicine and Medical Biotechnology, University of Naples “Federico II”, Naples, Italy, e Department of Neurosciences, San Camillo-Forlanini Hospital, Rome, Italy, f MS Center, San Pietro Fatebenefratelli Hospital, Rome, Italy, g Department of Public Health- University of Naples “Federico II”, Naples, Italy

Abstract

Ocrelizumab (OCR) and Natalizumab (NTZ) are highly effective treatments widely used in Multiple Sclerosis (MS). However, long-term, real-world comparative data on clinical effectiveness, safety and treatment persistence are limited. This retrospective analysis included relapsing and progressive MS patients initiating treatment at two Italian Universities (“La Sapienza” and “Federico II”). Propensity-score nearest-neighbour matching with a caliper of 0.1 was conducted to adjust for between-group differences in age, sex, previous treatment status, MS phenotype, disease duration, clinical and MRI activity at baseline. Differences in follow-up duration were adjusted with pairwise censoring. Cox proportional hazard regression models were used with Evidence of disease activity (EDA-3) and its components (relapses, MRI activity, and confirmed disability progression) as outcomes. Treatment discontinuation rate and occurrence of adverse events (AEs) were tested using logistic regression. We identified 308 patients (140 on OCR, 168 on NTZ) with a mean (SD) follow-up of 75.7 (30.8) months. Patients treated with OCR were older and less active and less frequently naïve at baseline than NTZ-treated patients. The PS-matching procedure retained 140 patients (70 pairs) with a mean follow-up of 55.9 (14.3) months. No significant differences were found between NTZ and OCR in terms of relapses, MRI activity or confirmed disability progression. OCR treatment was associated with a higher incidence of mild to moderate AEs, and higher to comparable treatment persistence. This study provides real-world evidence of comparable effectiveness between OCR and NTZ over a 5-year observation period, with OCR being associated with a higher incidence of AEs and, possibly, higher treatment persistence.

Conflict of interest statement

Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests. Many authors (see manuscript) report a relationship with Roche, Teva, Janssen, Merck, Biogen, Novartis, Mylan, Viatris, Bayer, Sanofi, Genzyme, BMS, Bristol, Almirall, Celgene, Actelion, CSL, Behring, HEALTH&LIFE, AIM Education, FARECOMUNICAZIONE E20 that includes: board membership, consulting or advisory, funding grants, speaking and lecture fees, and travel reimbursement.

Results

Study population

308 patients (168 on NTZ, 140 on OCR), with a mean (standard deviation) follow-up of 75.7 (30.8) months were eligible for analysis [Fig. 1]. Demographic and clinical characteristics of eligible patients are reported in Table 1. Patients treated with OCR were older, less active at baseline, less frequently naive than NTZ-treated patients and showed higher neurological disability (p ​< ​0.05, see Table 1 for details). One-hundred and forty patients (70 pairs) were retained after PS-matching. No difference in baseline clinical and demographic data was present in the matched cohort (p ​> ​0.05 for all, see Table 1 for details). As treatment duration was longer in NTZ than in OCR-treated patients (mean (sd): 77.8(28.8) vs 71.2(32.8); p ​= ​0.045), pairwise censoring was applied before any other analysis to avoid bias due to differences in follow-up. After the procedure, mean follow-up time was 60 months (55.9 ± 14.3) for both groups.

The Article can be found on Neurotherapeutics website.

Meet the expert

Prof. Carlo Pozzilli

We value your opinion.

How likely are you to recommend this material to a friend or colleague?

Selected Value: 1

Give a score from 1 (= not at all likely) to 10 (extremely likely)

Tell us what you think or ask a question to the expert.

Give your opinion on the quality of the material and send us your comments, questions or feedback. Where relevant, we'll liaise with the Expert and get back to you.

friend a Email

Email

If you have submitted a question above, leave your email so we can get back to you.

Submit