Abstract

Introduction

The World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia was revised in 2008 and is based on genetic characteristics and morphology. The classification incorporates newly recognized entities and emphasizes, more than previously, the pivotal role of cytogenetic abnormalities aiming at identifying biological and clinical entities as basis for a stratified treatment. The classification of acute myeloid leukemia (AML) is primarily based on adult studies. However, the frequency of cytogenetic changes varies among children and adults and the relevance of the 2008 revised WHO classification in pediatric AML has not yet been elucidated. The aim of this study was to evaluate the usability of the WHO classification when applied to a large population-based pediatric AML cohort.

Methods

We included children 0-18 years of age diagnosed with de novo AML in the Nordic countries (Sweden, Norway, Finland, Iceland and Denmark) and Hong Kong treated according to the NOPHO-AML-1993 and 2004 protocols in the period January 1993 to December 2012. Clinical, morphologic, and genetic data were retrieved from the database. Patients with myeloid leukemia of Down syndrome, acute promyelocytic leukemia, and therapy-related AML were excluded. The karyotypes from the Nordic countries were centrally reviewed with annual evaluation of the karyograms by the NOPHO cytogenetic working group. Morphological and molecular analyses and immunophenotyping were carried out at regional centers. The patients were classified according to the revised 2008 WHO classification using genetic parameters and the French-American-British (FAB) classification.

Results

In the NOPHO database we identified 609 children aged 0-18 years diagnosed with AML. In 13 cases, the karyotype analyses either failed or considered uninformative and were excluded. Among the remaining 596 informative cases (98%), 241 (40%) were categorized as AML with recurrent genetic abnormalities; t(8;21)(q22;q22);RUNX1-RUNX1T1 (n=69), inv(16)(p13.1q22) or t(16;16)(p13.1;q22);CBFB-MYH11 (n=47), t(9;11)(p21;q23);MLLT3-MLL (n=61), t(6;9)(p22;q34);DEK-NUP214 (n=5), inv(3)(q21q26.2) or t(3;3)(q21;q26.2);RPN1-EVI1 (n=1), t(1;22)(p13;q13)/RBM15-MKL1 (n=4), NPM1 mutated (n=13), and CEBPA mutated (n=10). Furthermore, 23 children were classified in the provisional entity AML with FLT3-ITD. Based on the karyotype, 92 children (16%) were classified as AML with myelodysplasia-related cytogenetical changes (AML-MDS). This group was dominated by unbalanced abnormalities (9%) with -7/del(7q) being the most common (4% of the total cohort) and complex karyotypes (6% of the total cohort). AML with balanced myelodysplasia-related changes was found in 0.5%. The largest group was the highly heterogeneous AML not otherwise specified (AML-NOS) (n=263) (44%), and included some of the well-defined 11q23 abnormalities. In the 2001 version of the WHO classification, “AML with abnormalities of 11q23; MLL,” was listed separately, but in the 2008 edition only AML with t(9;11)(p22;q23);MLLT3-MLL was listed as an entity. The AML-NOS cases were sub-classified according to morphology, with FAB M5 being most common, found in 22% of NOS, followed by FAB M1, FAB M2, and FAB M4, all found in 18%.

In conclusion, 40% of the children were classified in clinically relevant entities, but the WHO classification allocated 16% of the children to AML-MDS, the relevance is unknown in children, and 44% to the NOS group. This large and unspecified group limits the applicability of the WHO classification in children with AML suggesting that additional considerations are warranted for relevant classification of pediatric AML.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.