Pediatric acute myeloid leukemia with t(8;16)(p11;p13), a distinct clinical and biological entity: a collaborative study by the International-Berlin-Frankfurt-Munster AML-study group

Abstract

In pediatric acute myeloid leukemia (AML), cytogenetic abnormalities are strong indicators of prognosis. Some recurrent cytogenetic abnormalities, such as t(8;16)(p11;p13), are so rare that collaborative studies are required to define their prognostic impact. We collected the clinical characteristics, morphology, and immunophenotypes of 62 pediatric AML patients with t(8;16)(p11;p13) from 18 countries participating in the International Berlin-Frankfurt-Münster (I-BFM) AML study group. We used the AML-BFM cohort diagnosed from 1995-2005 (n = 543) as a reference cohort. Median age of the pediatric t(8;16)(p11;p13) AML patients was significantly lower (1.2 years). The majority (97%) had M4-M5 French-American-British type, significantly different from the reference cohort. Erythrophagocytosis (70%), leukemia cutis (58%), and disseminated intravascular coagulation (39%) occurred frequently. Strikingly, spontaneous remissions occurred in 7 neonates with t(8;16)(p11;p13), of whom 3 remain in continuous remission. The 5-year overall survival of patients diagnosed after 1993 was 59%, similar to the reference cohort (P = .14). Gene expression profiles of t(8;16)(p11;p13) pediatric AML cases clustered close to, but distinct from, MLL-rearranged AML. Highly expressed genes included HOXA11, HOXA10, RET, PERP, and GGA2. In conclusion, pediatric t(8;16)(p11;p13) AML is a rare entity defined by a unique gene expression signature and distinct clinical features in whom spontaneous remissions occur in a subset of neonatal cases.

Figure 1

Age at diagnosis of pediatric cases with t(8;16)(p11;p13) as compared with a pediatric AML reference cohort. Shown are two histograms depicting the age at diagnosis for two cohorts: t(8;16)(p11;p13) in dark gray, and a reference cohort of unselected pediatric AML patients treated by AML-BFM study group between 1995-2005 in light gray. (A) Age at diagnosis is shown in categories of months for the first 2 years after birth. (B) Age at diagnosis is shown in categories of 2 years until the age of 18. In t(8;16)(p11;p13) AML, after an initial peak in the occurrence shortly after birth, occurrence is stable through childhood. In the reference cohort, the frequency of congenital cases is significantly lower.

Figure 2

Survival t(8;16)(p11;p13) patients as compared with a pediatric AML reference cohort. Survival curves of t(8;16)(p11;p13) patients diagnosed after January 1, 1993, and treated up front with chemotherapy compared with 543 unselected pediatric AML patients treated by the AML-BFM study group between 1995 and 2005.

Figure 3

Unsupervised clustering of the gene expression profiles of 297 pediatric AML patients. Pairwise correlations between 297 samples of pediatric patients with acute myeloid leukemia. The cells in the visualization are colored by Pearson's correlation coefficient values with deeper colors indicating higher positive (red) or negative (blue) correlations. Cytogenetic data are depicted in the columns along the original diagonal of the heatmap. Cytogenetic group classification is depicted in the first column (MLL-rearranged-green, inv(16)-red, t(8;21)-yellow, t(15;17)-purple, t(7;12)-brown, cytogenetically normal-blue, other-orange, unknown-aqua). Presence of t(8;16)(p11;p13) is depicted in the second column (t(8;16)-blue, no t(8;16)-yellow).

Figure 4

GGA2 and PERP mRNA expression. RET indicates relative expression measured by RT-qPCR. Shown is GGA2 and PERP mRNA expression levels of pediatric AML patient samples and AML cell lines. (A,B) Levels from GEP for GGA2 and PERP, respectively. (C,D) Levels as determined by RT-qPCR. (E,F) show the correlation between both RT-qPCR and GEP.