TRANSFUSION-RELATED ALLOIMMUNIZATION IN CHILDREN: EPIDEMIOLOGY AND EFFECTS OF CHEMOTHERAPY (TRACE-EC)

Abstract

Background: Red cell transfusions can lead to alloimmunization; however, pediatric alloimmunization frequency has not been well studied, and it may vary among diagnostic subgroups. Subgroups such as pediatric hematology/oncology patients require numerous transfusions during chemotherapy, but the immunosuppressive effect of chemotherapy on alloimmunization is unknown. One study demonstrated reduction in IgM and IgG in pediatric leukemia patients (Martín Ibáñez 2003); hence, one could hypothesize that chemotherapy suppresses alloimmunization.

Objectives: This study aimed to: 1) describe alloimmunization frequency and antibody specificities in transfused pediatric patients; and 2) determine if chemotherapy affects the frequency of alloimmunization.

Methods: A retrospective cohort study of pediatric patients (transfused between April 2002 and November 2011 at Hamilton Health Sciences) was performed. Data were extracted from 3 sources: a blood utilization database; the Laboratory Information System; and chart reviews. The chemotherapy treatment group included pediatric hematology/oncology patients stratified by HSCT status and diagnosis; control cohort included other pediatric patients not diagnosed with cancer. Control patients with hemoglobinopathies were analyzed separately due to increased alloimmunization. Alloimmunization was defined as clinically significant IgG alloantibody formation.

Results: There were 1273 patients in the study: 949 in control group; 324 in study group. Alloimmunization was 1.6% overall: 0.6% (95% CI: 0, 1.47) in study group; 1.3% (95% CI: 0.58, 2.06) in control group. The association between chemotherapy and alloimmunization was not significant (p value = 0.38 Fisher’s exact test; OR 0.46, 95% CI: 0.10, 2.09). Due to low outcome rate, logistic regression to explore the association was not feasible.

Conclusions: This is the first study exploring the frequency of alloimmunization in pediatric patients by diagnosis and the association between chemotherapy and alloimmunization. The frequency of alloimmunization was low and no association between chemotherapy and alloimmunization was observed. Low event rate would have contributed to low power.