Duration Of Neutropenia Is Associated With Reduced Relapse Risk In Children With De Novo Acute Myeloid Leukemia: A Report From The Children’s Oncology Group

Children receiving treatment for de novo acute myeloid leukemia (AML) commonly experience infections and prolonged duration of neutropenia. There is great variability in these outcomes and pharmacogenomics may provide one mechanism to explain this variability. We hypothesized that relapse outcomes may be inversely related to the degree of toxicity.

To describe the relationship between the number of sterile site infections and duration of neutropenia during the first four cycles of chemotherapy and the risk of recurrence and overall survival.

AAML0531 was a Children's Oncology Group (COG) randomized phase 3 clinical trial that included 1022 children with de novo AML enrolled between August 2006 and June 2010. Inclusion criteria were age ≥ 1 month to ≤ 30 years. For this specific analysis, we focused on non-Down syndrome patients who completed at least 4 cycles of chemotherapy on protocol therapy without recurrence on therapy and who did not receive HSCT in first remission. Because high-risk patients were allocated to best allogeneic donor HSCT, we focused on favorable and intermediate risk cytogenetic groups. The treatment protocol consisted of 5 cycles of intensive chemotherapy; patients were randomized to receive or not receive gemtuzumab ozogamicin (GMTZ) during Induction I and Intensification II.

The cumulative time to neutrophil recovery between start of Induction II and completion of cycle 4 (Intensification II) was examined; Induction I was not included since many centers did not wait for neutrophil recovery before starting Induction II. Relapsed risk and survival analyses were conducted from the end of Intensification II. Potential confounders explored for this analysis were gender, age, race, weight group, cytogenetic risk group and allocation to GMTZ.

There were 569 patients included; 279 (49.0%) were male and 210 (36.9%) had favorable risk cytogenetics. There were 1116 sterile site infections; 735 (65.9%) were Gram-positive bacteria. The median cumulative time with neutropenia was 96 (range 46 – 204) days. The Table illustrates that infections did not significantly influence the risk of relapse or overall survival. However, increased duration of neutropenia was associated with a lower risk of relapse both in univariate and multiple regression models after adjustment for African-American race, favorable risk cytogenetics and GMTZ allocation.

Longer duration of neutropenia is associated with a reduced risk of relapse for children with favorable and intermediate cytogenetic risk AML. Toxicity may be mediated through pharmacogenomics which suggests that individualized chemotherapy dosing may be an effective strategy.

No relevant conflicts of interest to declare.