Background/Purpose: B-cell depletion therapy has increasingly been used for the treatment of childhood-onset rheumatic diseases. Previous studies investigating whether rituximab results in increased infections have yielded variable results. We therefore conducted a preliminary analysis of the incidence and risk factors for the development of hypogammaglobulinemia and infectious complications in a cohort of children with rheumatic disease treated with rituximab at Boston Children’s Hospital.

Methods: We performed a retrospective analysis of 96 subjects (ages 2 – 25) who received rituximab for the treatment of a rheumatic disease (years 2009 – 2019), excluding those with cancer or hypogammaglobulinemia prior to rituximab use. We quantified the incidence of hypogammaglobulinemia and infectious complications in the 12 months following rituximab use. Bivariate and multivariate analyses were conducted to identify the associated risk factors.

Results: Most patients (n=81, 84%) were treated with one or more immunosuppressives (cyclophosphamide, biologics, other DMARDS, corticosteroids) in the year prior to rituximab use (Table 1). Hypogammaglobulinemia developed in 25 (26%) patients within 12 months following rituximab use. In the 12 months following first recorded use of rituximab, an infectious complication developed in 44% of patients who developed hypogammagloblinemia vs 20% who did not develop hypogammaglobulinemia. Serious infections included sepsis (n=6), aspergillosis (n=2), and pneumocystosis (n=1). The majority of infections (n=15; 68.2%) presented within the first month after rituximab treatment, possibly due to a synergistic effect of cyclophosphamide and corticosteroids.

Hypogammaglobulinemia post-rituximab was associated with male gender (OR 4.22; 95% CI 1.16 – 15.9; p=0.021) and diagnosis of vasculitis (OR 4.66; 95% CI 1.61 – 14.1; p=0.002), and inversely associated with age (OR 0.85; 95% CI 0.75 – 0.95; p=0.006) (Table 2a). These relationships persisted in multivariate analysis (Table 2b). We also observed a borderline-significant association between exposure to cyclophosphamide in the 12 months prior to rituximab use (OR 2.85; 0.70 – 11.3; p=0.094) and development of hypogammaglobulinemia.

Infection risk post-rituximab was associated with hypogammaglobulinemia (OR 4.21; 95% CI 1.36 – 13.3; p=0.006) and inversely with age (OR 0.80; 95% CI 0.69 – 0.90; p< 0.001) (Table 3a). These associations persisted in multivariate analysis (Table 3b). Indication for treatment and the use of other immunosuppressives in the year prior to rituximab use were not associated with infection risk. However, a borderline-significant association of infection risk with corticosteroid use was observed (OR 6.18; 95% CI 0.86 – 273.7; p=0.064).

Conclusion: We observed high rates of hypogammaglobulinemia and infection, including serious infections, in pediatric patients treated with rituximab, particularly in younger children. The development of hypogammaglobulinemia in this cohort is strongly associated with increased infectious risk, suggesting a role for the development of screening guidelines for hypogammaglobulinemia and study of the efficacy of infection prophylaxis in this population.

« Back to 2020 Pediatric Rheumatology Symposium

ACR Meeting Abstracts - https://acrabstracts.org/abstract/incidence-and-risk-factors-of-hypogammaglobulinemia-and-infectious-complications-associated-with-rituximab-use-in-pediatric-rheumatic-diseases/