Background Rituximab (RTX) has been used in the treatment of various autoimmune rheumatic diseases (AIRDs) for over a decade. Repeat cycles are effective for maintenance but may lead to hypogammaglobulinaemia. Low IgG at baseline has been associated with post-treatment infection rate but may be confounded by other clinical variables and fully adjusted models with method for handling missing data have not been presented. Importance of post-treatment change in Igs has also not been proven.
Objectives To evaluate risk factors for severe infection in multivariable analysis and assess outcome of hypogammaglobulinaemia.
Methods The first 700 consecutive patients with AIRDs treated with RTX at a single centre were studied. Each cycle of RTX consisted of 2x1000mg infusions repeated on clinical relapse. IgM, IgA and IgG levels were measured at baseline and 4–6 months after each cycle. Multiple imputation was used for missing data. Baseline factors for predicting serious infection and low Ig were tested using univariable and multivariable (MVA) logistic regression analyses.
Results 550 patients were female, median age (IQR) at RTX initiation 58 (46–68) years and median disease duration (IQR) 7.9 (3.4–15.0) years. 506 (72%) had RA, 94 (13%) SLE, 49 (7%) AAV, 14 (2%) DM, 5 (1%) APS, 6 (1%) SSc and 26 (4%) other CTD. 364 (52%) were biologic-naïve and 515 (74%) were on concomitant DMARDs. Total follow-up: 2940 patient-years (PY). 284 serious infections were recorded in 179 patients (9.7/100 PY); 88 cases within 12 months of cycle 1 (C1). In MVA, previous severe infection (OR 10.7, 95% CI 5.8–19.5), low IgG (OR 3.6, 95% CI 1.5–8.6), previous cancer (OR 2.9, 95% CI 1.2–6.6) and chronic lung disease (OR 1.7, 95% CI 0.9–3.1) increased the odds of a severe infection within 12 months of C1. A diagnosis of CTD was associated with lower risk (OR 0.5, 95% CI 0.2–0.9). Low IgG at RTX initiation was predicted by older age, previous cancer, RA diagnosis, previous severe infection and previous treatment with cyclophosphamide. In C1-C3, higher rate of change in IgA and IgG levels were associated with serious infections (Figure 1). Overall, only 7 (1%) of the patients required Ig replacement in this cohort.
Conclusions Factors associated with serious infection at RTX initiation include previous serious infection, low IgG, previous cancer, a diagnosis of RA and chronic lung disease. This is the first study to show the rationale for monitoring the rate of change in Ig levels during repeat cycles of RTX, with reduction in all Ig subclasses being associated with increased risk of post-treatment infection. Further analysis including predictors of serious infections in repeat cycles is in progress and will be used to develop guidelines for safety monitoring of rituximab.
Acknowledgements This research was funded/supported by the National Institute for Health Research (NIHR) and NIHR Leeds Musculoskeletal Biomedical Research Unit based at Leeds Teaching Hospitals NHS Trust; (DRF-2014–07–155). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
Disclosure of Interest None declared