Background/Purpose: We were alerted to the possibility of very long term B cell depletion (BCD) in Systemic Lupus Erythematosus (SLE) by a patient treated with rituximab in 2001 whose CD19 counts remain <0.001x10⁹/L 12 years later. The duration of B cell depletion is variable between SLE patients. Because most relapses occur after the return of B cells, our purpose was to analyse clinical and serological features and outcome in patients considering the duration of B cell depletion.

Methods: We analysed our lupus cohort retrospectively to identify those BCD treated patients. We collected data noting the time to return of the B cells, clinical and serological features and classic British Isles Lupus Assessment Group (BILAG) scores and baseline, 6 months and 12 months after the treatment. Logistic regression analysis was made using SPSS Statistics Data Editor software.

Results: A total of 190 courses of BCD in 101 patients in whom we had full serological and clinical data prior to December 2012 were considered.

Among the 101 patients, 94 were female and ethnicities included 40 Caucasian, 28 Afro-Caribbean, 23 Asian, 7 Oriental and 3 others.

57 patients had more than 1 treatment. One patient and one of the infusions of another patient were excluded because of incomplete BCD after the treatment.

32.1% repopulated between 6 to 9 months and 28.6% repopulated after 12 months (figure 1). Two groups were analysed based on the time to repopulation at a defined threshold of 12 months. We included infusions for which the patients had not repopulated but were depleted for at least 12 months. We excluded the infusions for which a follow up was less than 12 months and for which the patients remained depleted. 144 treatments were analysed. 41.7% (group 1) repopulated in less than 12 months and 58.3% (group 2) were depleted for at least 12 months.

An association with longer time to repopulate and lymphopenia (p=0.008) at any point in the course of the patient’s disease was noted. Inverse association with alopecia (p=0.033) and oral ulcers (p=0.039) was also noted. No association was found between serological features such as the presence of anti-dsDNA or anti-Sm antibodies or low complement.

The cohort’s mean classic BILAG numerical score at baseline was 13.44 (SD=7.55). Group 2 was associated with a higher BILAG score at baseline (p=0.026). At 6 months group 2 patients had lower numerical BILAG score (p=0.002); at 12 months the same tendency was observed but with no statistical significance. Likewise, there was an association between group 2 patients and no BILAG As nor Bs at 6 months (p=0.012). Also a decrease of the BILAG score at 6 months (p=0.012) and 12 months (p=0.012) in these patients was noted.

Conclusion: Despite higher disease activity at baseline, as measured by classic BILAG, patients who were B cell depleted for longer showed some differences in clinical features and, most importantly, had a better outcome at 6 and 12 months.