Predictors and short-term outcomes of hyperleukocytosis in children with acute myeloid leukemia: a report from the Children's Oncology Group

Adults and children with acute myeloid leukemia (AML) who present with a high initial white blood cell count (WBC) have poor outcomes, and a consistent association with induction death has been observed when the initial WBC is 100×10/L or over.1 However, rates of induction death have varied between trials.2,3 Furthermore, major advances in supportive care have been incorporated into routine clinical care such as prompt transfusion of platelets, leukapheresis, hydroxyurea, urate oxidase and hemodialysis. With these advances in supportive care, it is unclear whether poor outcomes with hyperleukocytosis in pediatric AML continue to be problematic.

The Children's Oncology Group (COG) adopted a modified AML Medical Research Council backbone and accrued patients to trials AAML03P1 and AAML0531. In this study, we analyzed risk factors for, and outcomes associated with, hyperleukocytosis for children and adolescents with AML.

This report used data collected from AAML03P14 and AAML0531.5 Both studies were approved by each institutional review board and all parents/participants provided informed consent. Common eligibility were aged one month to 21 years with de novo AML, and infants aged under one month with progressive disease and isolated chloromas. Hydroxyurea was excluded in both. Induction I consisted of cytarabine 100 mg/m/dose intravenous (i.v.) every 12 h on Days 1-10; daunorubicin 50 mg/m/dose i.v. on Days 1, 3 and 5; and etoposide 100 mg/m/dose i.v. on Days 1-5. All patients received gemtuzumab (GMTZ) on AAML03P1 whereas administration was randomized on AAML0531 (3 mg/kg/dose once on Day 6). Uniform guidelines for supportive care were provided. It was recommended that before initiation of chemotherapy, the best possible control of coagulopathy, metabolic derangement, and institution of treatment for fever and/or infection be established. Specific management for hyperleukocytosis consisted of cautious administration of packed red blood cell transfusions and liberal use of platelets as clinically indicated. Both studies suggested using allopurinol while AAML0531 also suggested considering rasburicase.

Hyperleukocytosis was defined as an initial WBC count of 100×10/L or over. All data concerning non-hematologic grade 3 or higher toxicities were obtained prospectively by clinical research associates of participating institutions and graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 3.0. We did not describe outcomes according to GMTZ administration as the outcomes on AAML0531 are still blinded.

We examined the following potential predictors of initial WBC 100×10/L or over: age, race, ethnicity, body mass index (BMI) category at diagnosis, FAB AML classification and cytogenetics. BMI (>2 years) or weight-for-length (1-≤2 years) were used to delineate underweight as 10 percentile or under, overweight as 95 percentile or over, and normal weight as more than 10 to less than 95 percentile.6 We also determined whether patients underwent leukapheresis using data from the Pediatric Health Information System (PHIS), which is an administrative database that includes inpatient data from 43 tertiary children's hospitals. Data from COG and PHIS were merged using a probabilistic merge based on ICD9 diagnosis code, treatment center and date of birth. Leukapheresis was defined by a leukapheresis procedure code (9972).

To determine predictors of hyperleukocytosis, univariate and multivariable logistical regression analyses were performed. Associations between toxicities and initial WBC groups were compared using χ / Fisher's exact test. The Mann-Whitney test was used to determine differences in medians for continuous variables. All statistical analysis was performed using the SAS statistical program (SAS-PC, version 9.2; SAS Institute Inc., Cary, NC, USA). All tests of significance were two-sided.

A total of 1,364 children with de novo AML were included in the study; 256 (18.8%, 95% CI: 16.7-20.9%) had an initial WBC of 100×10/L or over. Characteristics are illustrated by the presence or absence of initial hyperleukocytosis (Table 1). Multivariable regression analysis showed that age one year or under, FAB M1, M4 and M5, inv(16) and FLT3-ITD were independently associated with an increased risk of hyperleukocytosis (Online Supplementary Table S1).

Table 1.Demographic variables and disease characteristics in children with de novo acute myeloid leukemia enrolled on AAML03P1 or AAML0531 who had or did not have hyperleukocytosis.

The prevalence of grade 3 or higher metabolic, pulmonary and CNS toxicities by initial WBC count are presented in Table 2; 7 children were excluded due to lack of or conflicting data. Hyperkalemia, hyperphosphatemia, hypercreatinemia and hyperuricemia were rare. Hyperphosphatemia and hyperuricemia were significantly associated with an increased initial WBC. Elevated initial WBC was significantly associated with hypoxia (P<0.001) and pulmonary hemorrhage (P=0.003). CNS ischemia or hemorrhage was significantly associated with initial WBC count, but no seizures occurred in patients with an initial WBC of 200×10/L or over. Induction I death rate was significantly associated with an increased initial WBC (P=0.004) and occurred in 4 of 44 (9.1%, 95% CI: 2.5-21.7%) patients with an initial WBC of 300×10/L or over.

Table 2.Grades 3, 4 or 5 metabolic, pulmonary, and central nervous system toxicities during Induction I by initial white blood cell counts.

Timing of Induction I deaths and severe toxicities by increasing WBC categories, with timing of toxicities being restricted to patients enrolled on AAML0531, are described in Table 2 and in the Online Supplementary Table S2. Although there were few events, those occurring in patients with the highest initial WBC (≥ 400×10/L) tended to occur early. There were 89 children available in PHIS with an initial WBC of 100×10/L or over, representing 35% of the hyperleukocytosis cohort. Sixteen (18%) underwent leukapheresis: initial WBC ≥100-<200×10/L (3 of 49); WBC ≥200-<300×10/L (7 of 24); WBC ≥300-<400×10/L (5 of 10); and WBC ≥400×10/L (one of 6). Among the children who received leukapheresis, one of 16 (6.3%) experienced an Induction I death compared to 3 of 73 (4.1%) for those who did not receive leukapheresis.

We found that almost 20% of children with de novo AML had hyperleukocytosis at presentation, and that patients at higher risk were infants (≤ 1 year) and those with FAB M1, M4 and M5, inv(16) and FLT3-ITD. Tumor lysis syndrome was uncommon and hyperleukocytosis was not associated with acute renal failure. However, hyperleukocytosis continues to be associated with pulmonary toxicities in terms of hypoxia and pulmonary hemorrhage as well as CNS ischemia/hemorrhage, with increasing risk in those with higher initial WBC counts.

Our report suggests that pulmonary and CNS injuries are still the major issues associated with hyperleukocytosis in spite of current supportive care. Our report also suggests that leukapheresis does not reduce induction mortality.

We conclude that outcomes remain poor in children with extremely high initial WBC in spite of the supportive care currently available. Future supportive care trials focused on reducing pulmonary and CNS toxicities should be a priority for children with AML.

Acknowledgments

the authors would like to thank Vani Shanker for editorial assistance.

Funding: this work was supported by research funding in the form of a New Investigator Award from the Canadian Institutes of Health research to LS (Grant n. 87719). This research also was supported by the Chair's Grant U10 CA98543-08 of the Children's Oncology Group from the National Cancer Institute, National Institutes of Health, Bethesda, MD, USA. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NCI or the NIH.

Footnotes

• The online version of this article has a Supplementary Appendix.
• AAML0531/PHIS Group: Laura Burden from the Children's Oncology Group, Arcadia, CA; Susana C.Raimondi from St. Jude Children's Research Hospital Memphis, TN; Betsy A. Hirsch from the University of Minnesota Medical Center, Minneapolis, MN; Soheil Meshinchi from Fred Hutchinson Cancer Research Center, Seattle, WA; Matt Hall and David Bertoch from Child Health Corporation of America, Shawnee Mission, KS; Yuan-Shung V. Huang, Yimei Li, and Brian T. Fisher from Children's Hospital of Philadelphia, Philadelphia, PA, USA.
• The information provided by the authors about contributions from persons listed as authors and in acknowledgments is available with the full text of this paper at www.haematologica.org.
• Financial and other disclosures provided by the authors using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are also available at www.haematologica.org.