Prevalence and prognostic implications of CEBPA mutations in pediatric acute myeloid leukemia (AML): a report from the Children's Oncology Group

Abstract

CEBPA mutations have been associated with improved outcome in adult acute myeloid leukemia (AML). We evaluated the prevalence and prognostic significance of CEBPA mutations in 847 children with AML treated on 3 consecutive pediatric trials. Two types of CEBPA mutations-N-terminal truncating mutations and in-frame bZip-domain mutations-were detected in 38 (4.5%) of 847 patients tested; 31 (82%) of 38 patients with mutations harbored both mutation types. Mutation status was correlated with laboratory and clinical characteristics and clinical outcome. CEBPA mutations were significantly more common in older patients, patients with FAB M1 or M2, and patients with normal karyotype. Mutations did not occur in patients with either favorable or unfavorable cytogenetics. Actuarial event-free survival at 5 years was 70% versus 38% (P = .015) with a cumulative incidence of relapse from complete remission of 13% versus 44% (P = .007) for those with and without CEBPA mutations. The presence of CEBPA mutations was an independent prognostic factor for improved outcome (HR = 0.24, P = .047). As CEBPA mutations are associated with lower relapse rate and improved survival, CEBPA mutation analysis needs to be incorporated into initial screening for risk identification and therapy allocation at diagnosis.

Trial registration: ClinicalTrials.gov NCT00002798 NCT00070174.

Figure 1

CEBPA mutations, CCG-2941/2961 and COG-AAML03P1. (A) Fragment-length analysis, showing representative mutations. NTD: (i) wild type; (ii) 1-bp deletion; (iii) 4-bp insertion. bZip: (iv) wild type; (v) 6-bp insertion; (vi) 24-bp insertion. (B) Location of each functional mutation. Mutations that are one of a pair of CEBPA mutations within the same patient, or “double” mutations, are depicted in red. Polymorphisms are not shown. (C) Thirty-one patients (82% of those with mutations) harbored “double” mutations, pairing an N-terminal truncating mutation with an in-frame bZip mutation. Of the patients with “single” mutations, only one patient had an NTD mutation alone.

Figure 2

Cytogenetics and FAB class of patients with and without CEBPA mutations. (A) CEBPA mutations were found overwhelmingly in the cytogenetically normal cohort, and did not occur in either favorable (CBF AML) or unfavorable risk groups. (B) CEBPA mutations were most commonly found in FAB M1 and M2. Statistically significant differences are denoted with an asterisk.

Figure 3

Clinical significance of CEBPA mutations in pediatric AML. Kaplan-Meier estimates for (A) overall survival and (B) event-free survival from study entry for patients with and without CEBPA mutations. (C) Cumulative incidence of relapse is also presented.

Figure 4

Clinical outcomes for patients with CEBPA mutations versus core binding factor AML versus neither. (A) Overall survival and (B) event-free survival from study entry, as well as (C) relapse risk from first CR for patients with and without CEBPA mutations compared with those with CBF AML.

Figure 5

Comparison of clinical outcome for patients with “single” versus “double” CEBPA mutations. Patients with one CEBPA mutation had nearly identical (A) overall survival and (B) disease-free survival curves to those with 2 mutations.