Optimal dosage of rituximab for children with Burkitt lymphoma

Abstract

The current chemotherapy treatments have led to an improvement in survival rates for pediatric Burkitt’s lymphoma (BL). Survival in children with high-grade, mature B-cell non-Hodgkin’s lymphoma (B-NHL) has been prolonged by six rituximab doses combined with chemotherapy, whereas the efficacy of four doses has not been reported. This study aimed to explore optimal therapeutic strategies—the number of doses of rituximab based on different risk groups—and also aim to investigate the clinical characteristics of Chinese pediatric BL. This study consecutively enrolled children with BL in Beijing Children’s Hospital who received French-American-British mature B-cell lymphoma 96 (FAB/LMB96). The patients were divided into three groups: R0 group (chemotherapy alone), R6 group (chemotherapy combined with six rituximab doses), and R4 group (chemotherapy combined with four rituximab doses). The clinical characteristics and outcomes were evaluated. Univariate and multivariate analyses and prognostic nomogram were used to assess prognostic factors. A nomogram was developed that predicted overall survival based on the Cox proportional hazards model, and the concordance index (C-index) and a calibration curve were used to determine its predictive and discriminatory capacity. We enrolled 385 boys and 71 girls, with a median age of 6 years (1–14 years). Of these, 296 patients (65%) had initial abdominal symptoms, 182 (40%) had bulky disease, 46 (10%) had B symptoms, 77 (16.9%) had BL-ALL (blasts ≥ 25% in bone marrow (BM)), 96 (21%) had central nervous system (CNS) disease, 406 (89%) were in stages III–IV, 378 (83%) were in group C, 170 (37.2%) had lactate dehydrogenase (LDH) levels ≥ 1000 U/L at initial diagnosis, and 137 (30%) had tumor lysis syndrome. The R0, R6, and R4 groups included 79, 144, and 227 patients, respectively. Six patients were excluded due to treatment withdrawal for various reasons. The 3-year overall survival (OS) and event-free survival (EFS) percentages were 92% ± 1.3% and 91.3% ± 1.3%, respectively, in all cohorts, whereas the 3-year EFS percentage was 83.5% ± 4.2%, 93% ± 2.1%, and 92.9% ± 1.8% in the R0, R6, and R4 groups, respectively (P = 0.025). The nomogram included four important variables based on a multivariate analysis of the primary cohort: course of disease ≤ 20 days, presence of bulky disease at the beginning of diagnosis, central nervous system(CNS) invasion, and dosage of rituximab. The calibration curve showed that the nomogram was able to predict 3-year OS accurately. The C-index of the nomogram for OS prediction was 0.79 for both cohorts. In our hospital, pediatric BL was more commonly observed in school-age boys with an abdominal mass and mostly in advanced stages at initial diagnosis. The FAB/LMB96 regimen combined with rituximab significantly increased survival outcomes. We observed no significant differences between four and six doses of rituximab in terms of treatment outcomes. The proposed nomogram provides an individualized risk estimate of OS in patients with BL and may assist treatment decision-making or rituximab dose design.