Background: The first biosimilar rituximab (RTX-B), Truxima® was launched in our department in October 2017. Patients therefore underwent non-medical switch (i.e. changed for reasons other than the patient’s health and safety). We previously showed that non-medical switch from originator rituximab (RTX-O) to RTX-B was largely effective in rheumatoid arthritis (RA) with comparable 18-month retention rates between those who switched vs remained on RTX-O, 76% and 82% respectively [1]. However, the uptake of non-medical switch in autoimmune connective tissue diseases and vasculitis (CTD-VAS) has been slow due to a concern with cross-reactivity of antibodies. Data pertaining to effectiveness of switching between biosimilars are also lacking.

Objectives: To evaluate the effectiveness of non-medical switch from RTX-O to RTX-B or between biosimilars RTX in CTD-VAS.

Methods: We conducted a retrospective observational cohort study of RMD patients in a single centre between October 2017 (Index date) and November 2024. During this period, all patients were encouraged to switch to RTX-B (Truxima®) unless declined by the patient or specified by the treating clinician. Furthermore, between 2021-2023, patients on Truxima® were switched to Rixathon® and then reverted to Truxima® in 2024 due to contractual agreement. Due to differences in disease activity tools, clinical responses were graded into full response; partial; and non-response. Other measures of effectiveness include the depth of CD20+ cells depletion by highly sensitive flow cytometry and 5-year RTX retention rate between patients who underwent non-medical switch (Group 1) vs those who remained on RTX-O or started on RTX-B (Group 2).

Results: At Index date, of 827 RMD patients treated with RTX, 300 (36%) were given for CTD-VAS, while the remaining for RA. Of these, 90/300 (30%) underwent non-medical switch, Group 1 [RTX-O to RTX-B = 61 (68%); between biosimilars = 29 (32%)]. They had mean (SD) age 51 (15) years, 61 (72%) were female, 64 (71%) had European ancestry, median (IQR) SLE duration of 8.7 (5-15) years, median (range) number of comorbidities of 1 (0-5), and diagnoses were SLE (53%), AAV (31%), Sjogren (4%), Myopathies (3%), Scleroderma (1%) and other CTD (8%). 16/300 (5%) patients remained on RTX-O, while 194/300 (65%) initiated treatment with RTX-B; therefore Group 2 = 210/300 (70%). At the last follow-up, of 90 patients in Group 1, 78/90 (87%) remained on RTX-B including 9/78 who underwent further biosimilar switches due to contractual agreement without problems and 1/78 (3%) reverted to the previous RTX-B brand due to development of a rash following biosimilar switch. 5/90 (6%) of patients on RTX-B reverted to RTX-O and regained response. Reasons were infusion reaction=1, serum sickness=1; neutropenic sepsis 5 days post-RTX switch=1; skin lesion=1; incomplete depletion and inferior response=1. Of 87/90 and 77/90 patients in Group 1 with paired clinical response and B-cells data in the RTX cycle before and after switch respectively, there was no difference in response rate (partial or full) and CD20+ cell complete depletion, p=0.289 and p=1.00 respectively (McNemara’s test). The delta difference in CD20+ cells depletion at 2 weeks from RTX infusion between before and after switch were no different, p=0.995 (Wilcoxon Sign Ranked Test). At Index Date, patients in Group 2 had shorter disease duration, less number of RTX cycles, and higher daily oral prednisolone dose than those in Group 1. At 5 years, 9/90 (10%) patients discontinued RTX in Group 1 [inefficacy=7 including 2 who reverted to RTX-O); 2 deaths due to pneumonia=1 and COVID pneumonitis=1], while in Group 2, 28/210 (13%) discontinued therapy [inefficacy=16; side effects=3; and deaths due to cardiac=4, pneumonia=2; pulmonary fibrosis=2; and multi-organ failure=1]. Kaplan-Meier analysis showed no difference in 5-year RTX retention survival between Group 1 and Group 2; HR 0.71 (95% CI 0.30-1.71), p=0.448, adjusted for disease duration, number of RTX cycles and oral prednisolone dose (Figure 1).

Conclusion: Our findings support the non-medical switch either from RTX-O to RTX-B or between biosimilars in CTD-VAS with no difference in clinical response and the depth of B-cell depletion before and after switch. 5-year rituximab retention rate was very good regardless of non-medical switch and appeared higher than in RA. Given its low cost and durability, future studies should assess the cost-effectiveness of first-line use of rituximab biosimilars in patients with CTD-VAS.

REFERENCES: [1] Melville A et al. Rheumatology. 2021 Aug 2;60(8):3679-3688.

Figure 1.

Acknowledgements: NIL.

Disclosure of Interests: Rusyai Zalynda Ramli: None declared, Charles Joseph Bithell: None declared, Mohmedshahid Patel: None declared, Paul Emery Roche, Shouvik Dass: None declared, Edward M. Vital Roche, Roche, Sandoz, Md Yuzaiful Md Yusof Roche.

© The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.