BACKGROUND: Ocrelizumab (OCR) is the only disease-modifying therapy (DMT) for both relapsing and primary progressive forms of multiple sclerosis (MS). OCR is given by intravenous (IV) infusion twice a year, which may improve adherence to the dosing schedule relative to other MS DMTs that require more frequent administration. Real-world evidence on the persistence and adherence of patients with MS to OCR compared with other DMTs is limited. OBJECTIVE: To examine the persistence and adherence to OCR compared with other DMTs for MS in the United States. METHODS: This analysis was conducted in the PharMetrics Plus commercial claims database and included patients with MS who initiated a new DMT between April 2017 and September 2018. Patients were required to have health plan enrollment for ≥ 1 year before and after DMT initiation (a subgroup analysis was performed for those with ≥ 18 months' continuous enrollment after DMT initiation). Persistence was defined as not switching to another DMT and having no gap in coverage of the initiated DMT for ≥ 60 days during the postinitiation period. The proportion of days covered (PDC) was calculated as the total days covered by the DMT during the postinitiation period divided by the length of the time period (12 or 18 months); PDC ≥ 0.8 was considered adherent. Multivariable Poisson regression models compared discontinuation (nonpersistence) and nonadherence between OCR users and users of other DMTs grouped by administration route. RESULTS: A total of 4,587 patients (OCR, 1,319; injectable, 1,051; oral, 1,876; other IV, 341) were included. The OCR group had the lowest proportion of patients discontinuing at 12 months (8% vs. 28%, 32%, and 43% for other IV, oral, and injectable, respectively) and the highest mean PDC (93% vs. 76%, 74%, and 69%, respectively). Compared with patients initiating OCR, adjusted relative risks (RR) of 12-month discontinuation were 3.3 (95% CI = 2.3-4.6), 3.8 (95% CI = 3.0-4.9), and 5.5 (95% CI = 4.1-7.5) for patients initiating other IV, oral, and injectable DMTs, respectively. Similarly, patients initiating other IV, oral, and injectable DMTs had RRs of 4.9 (95% CI = 3.6-6.8), 5.1 (95% CI = 3.9-6.6), and 6.8 (95% CI = 5.0-9.3) for 12-month nonadherence compared with OCR. A subgroup of 2,913 patients with 18 months of continuous enrollment had similar trends, with 17% in the OCR group discontinuing compared with 40%, 41%, and 55% in the other IV, oral, and injectable groups, respectively. Trends over 18 months were consistent with the 12-month analysis in adjusted models. CONCLUSIONS: Patients initiating OCR had superior persistence and adherence at 12 and 18 months of follow-up compared with patients initiating other MS DMTs. Long-term persistence and adherence should be monitored as OCR experience accrues in a real-world setting. DISCLOSURES: This study was funded by Genentech (South San Francisco, CA), a member of the Roche Group. Engmann, Sheinson, Bawa, and Ng are employees of Genentech and shareholders of F. Hoffman-La Roche (Basel, Switzerland).