Infection Outcomes and Hypogammaglobulinemia in Patients with Non-Hodgkin Lymphoma Treated with Immunoglobulin Replacement Therapy

Abstract

CONCLUSIONS Patients diagnosed with non-Hodgkin lymphoma (NHL) histologies often develop hypogammaglobulinemia (HGG), a secondary immunodeficiency (SID) that is associated with exposure to NHL-related treatments. Those with HGG frequently experience infections, which represent the leading cause of death in NHL (33% of cases). Patients with HGG who experience recurrent infections may benefit from immunoglobulin replacement therapy (IgRT). We aimed to assess real-world IgRT utilization and associated infection outcomes in patients with NHL. This retrospective, longitudinal, observational study evaluated HGG (immunoglobulin G [IgG] < 500 mg/dL) status and infection outcomes among adult patients diagnosed with NHL after 2010 who had ≥ 12 months of clinical data and ≥ 3 visits/year after NHL diagnosis and received immune globulin infusion (human), 10% (IG10% [ie, Gammagard Liquid]). De-identified clinical data from the Massachusetts General Brigham Research Patient Data Registry was used. IgG testing, infection outcomes, and antimicrobial use were compared 3 months before versus 3 months after initiation of IG10%. Generalized estimating-equation logistic regression models were used to obtain odds ratios (ORs), 95% confidence intervals (CIs), and P values. A total of 13,232 patients with NHL were identified. Among patients with NHL, 623 (4.7%) received ≥ 1 IgRT during the follow-up period, of whom, 563 (90.4%) received IG10%. The majority of patients with NHL treated with IG10% were male (61.3%), White (90.1%), with a median (interquartile range [IQR]) age of 62 (53-69) years, and median (IQR) follow-up of 4.8 (2.3-7.7) years. Median (IQR) administrations of IG10% across recipients were 2 (1-4) and over half of patients (55.6%) received ≥ 2 IG10% administrations. Median (IQR) time from initiation of IG10% to end of follow-up was 25.6 (6.8-55.2) months. Most patients had ≥ 1 IgG test in the follow-up period (n = 532 [94.5%]), median (IQR) number of IgG tests was 10 (4-27), and 84.2% of patients had ≥1 IgG test result available before initiation of IG10%. Of the patients treated with IG10% who had IgG test results available (n = 532), 81.6% were identified with HGG (< 500 mg/dL). Few patients had IgG testing at the time of first NHL diagnosis (13.3%), and median (IQR) time from diagnosis to first IgG test was 4.9 (0.8-17.4) months. Of the patients with ≥ 2 IG10% administrations (n = 313), median (IQR) time from the first administration of IG10% to subsequent IgG testing was 1.0 (0.3-2.8) month. Most patients (91.3%) who received ≥ 2 IG10% administrations and ≥ 1 IgG testing had an IgG test between first and second administrations of IG10%. Median (IQR) serum IgG levels were higher in the 3 months following initiation of IG10% versus 3 months before initiation of IG10% (520.0 mg/dL [413.0-670.0] vs 339.0 mg/dL [228.0-405.0]), and significantly fewer patients had HGG (34.7% vs 83.0%, P < 0.0001). In addition, significantly lower odds of infection (OR 0.60, 95% CI 0.47-0.75), severe infection (OR 0.40, 95% CI 0.32-0.50), and associated antimicrobial use were observed 3 months after IG10% initiation (Figures 1 and 2). In this real-world study of patients with NHL, treatment with IG10% was shown to be associated with significantly lower rates of HGG as well as lower odds of infection, severe infection, and antimicrobial use 3 months after IG10% initiation. Guidelines and consensus-based recommendations are needed to optimize the use of IgG testing, initiation of IgRT, and SID management in patients with NHL.