Effect of rituximab on immune status in children with mature B-cell non-Hodgkin lymphoma: a prespecified secondary analysis of the Inter-B-NHL Ritux 2010 trial

Background Survival of children and adolescents with high-risk, mature B-cell non-Hodgkin lymphoma is improved by the addition of rituximab to chemotherapy. The effect of rituximab on immune reconstitution after therapy has not been well described. Herein, we evaluate the immune effects of the addition of rituximab to intensive chemotherapy, a prespecified secondary aim of the Inter-B-NHL Ritux 2010 trial. Methods The Inter-B-NHL Ritux 2010 trial was an international, open-label, randomised, phase 3 trial in children (age 6 months to 18 years) with high-risk, mature B-cell non-Hodgkin lymphoma, comparing chemotherapy alone or chemotherapy with rituximab. Measures of immune status were completed at baseline, 1 month from the end of treatment, and 1 year from the start of therapy, and yearly thereafter until normalised. For this secondary analysis, we report on the proportions of patients with low lymphocyte counts and immunoglobulin concentrations at these timepoints with total lymphocyte count, B-cell count, and IgG concentration as the main endpoints. Other endpoints of interest included exposure to immunoglobulin replacement therapy and vaccine serologies. The population assessed for immune endpoints was the eligible per-protocol population with at least one immune parameter at one timepoint. Comparisons of immune status were made between the randomised treatment groups. Safety in the post-therapy period was assessed in the population eligible for the immunity study who were followed up at least 3 months after the end of treatment and without cancer-related events. The Inter-B-NHL Ritux 2010 study was registered with ClinicalTrials.gov, NCT01516580; status completed, with analyses of secondary aims ongoing.Findings From Dec 19, 2011, to June 13, 2017, 421 patients (344 [82%] boys and 77 [18%] girls; mean age was 88 years [SD 41]) were enrolled and had immune data at baseline during follow-up, or both. The study population included randomly assigned patients (n=289) and a non-randomised cohort enrolled after the planned interim analysis (n=132). At baseline, 99 (34%) of 290 patients with available data (excluding patients with bone marrow disease with peripheral blast cells) had lymphopenia, and 178 (48%) of 368 had hypogammaglobulinemia. 1 month from the end of therapy, patients who received chemotherapy with rituximab were more likely than those who received chemotherapy alone to have lymphopenia (86 [81%] of 106 vs 53 (60%) of 89, odds ratio [OR] 292 [95% CI 153-557], p=00011), B-cell lymphopenia (72 [96%] of 75 vs 36 [64%] of 56, OR 1333 [371-4784], p<00001), and hypogammaglobulinemia (67 [71%] of 95 vs 37 [47%] of 79, OR 272 [145-507], p=00017). Differences remained at 1 year for hypogammaglobulinemia only (52 [55%] of 94 vs 16 [25%] of 63, OR 364 [181-731], p=00003). Patients in the chemotherapy with rituximab group were more likely than those in the chemotherapy group to receive immunoglobulin replacement (26 [16%] 164 vs nine [7%] of 158, hazard ratio [HR] 263 [95% CI 123-562], p=0010), mainly due to low immunoglobulin concentration.In the combined treatment groups, including non-randomly assigned patients, the proportion of patients who had loss of protective serologies to a vaccine preventable infection varied from four (9% ) of 47 for polio to 21 (42%) of 50 for Streptococcus pneumoniae (pneumococcus). One patient (chemotherapy with rituximab group) had a life-threatening infectious event of polymicrobial bacterial sepsis reported 2 months after the final chemotherapy administration. Interpretation Children with high-risk mature B-cell non-Hodgkin lymphoma receiving chemotherapy with rituximab were at risk of prolonged hypogammaglobulinemia, although severe infections were rare. Strategies for immunoglobulin replacement and revaccination are needed.