Abstract
Cytogenetics is important in diagnosis of acute myeloid leukaemia (AML) and a number of chromosomal abnormalities have prognostic significance. AML clinical trials use three prognostic cytogenetic groups: good, intermediate and poor risk, for risk stratification of newly diagnosed patients to the most appropriate treatment arm. Many AML studies have combined data from children and adults, although the biology of their disease and treatment responses differ. For example, chromosomal abnormalities are observed in 80% of childhood AML, compared to only 50% of adults. The recent opening of the International Randomised Phase III Clinical Trial in Children with Acute Myeloid Leukaemia, MyeChild01, required extensive review of existing and novel genetic risk factors for stratification of patients in this study. The final consensus was to include the following abnormalities: Good risk; t(8;21)(q22;q22)/RUNX1-RUNX1T1 (approximate expected incidence in childhood AML, 12%), inv(16)(p13q22)/CBFB-MYH11 (6%); Poor Risk: inv(3)(q21q26)/t(3;3)(q21;q26)/abnormalities of 3q/MECOM rearrangements (~1%), monosomy 5/deletion of the long arm of chromosome 5 (5q) (~1%), monosomy 7 (4%), t(6;9)(p23;q34)/DEK-NUP214 (~1%), t(9;22)(q34;q11)/BCR-ABL1 (~1%), t(6;11)(q27;q23)/MLL-MLLT4/t(4;11)(q21;q23)/MLL-AFF1/t(10;11)(p11-p14;q23)/MLL-MLLT10 (5%), t(5;11)(q35;p15.5)/NUP98-NSD1(<5%),abnormalities of the short arm of chromosome 12 (12p) (~4%), inv(16)(p13.3q24.3)/CBFA2T3-GLIS2 (<2%); Intermediate risk; t(9;11)(p21;q23)/MLL-MLLT3/t(11;19)(q23;p13.3)/MLL-MLLT1/other MLL rearrangements (11%) and all other cases (25%).
A pilot study was performed to evaluate the role of FISH in accurate detection of these abnormalities using a retrospective cohort of 158 paediatric AML patients. Patients were initially classified according to available karyotype, as t(8;21)(q22;q22) (n=15, 9%), inv(16)(p13q22) (n=12, 8%), t(15;17)(q24;q21) (n=8, 5%, although these patients are excluded from MyeChild01), monosomy 7/abnormalities of long arm of chromosome 7 (7q) (n=13, 8%), chromosome 5 abnormalities (n=3, 2%), t(9;22)(q34:q11) (n=1, 0.6%) and MLL rearrangements (n=49, 31%). Those patients with normal karyotype and other abnormalities were grouped together as other (n=57, 36%). They were analysed using a range of specific FISH probes, either commercially available from CytoCell or Kreatech or home grown, for the abnormalities listed above. No additional patients were identified with RUNX1-RUNX1T1, PML-RARα, CBFβ-MYH11 or BCR-ABL1, although copy number changes involving the chromosomal regions covered by these probes were indicated. No patients were found with the poor risk abnormality, inv(16)(p13.3q24.3)/CBFA2T3-GLIS2, likely due to its rarity.However, a number of previously undetected abnormalities were identified: MLL rearrangement, with poor risk translocations excluded (n=2), MLL-MLLT3 (n=1), 12p abnormalities (n=6), NUP98-NSD1 (n=3), DEK-NUP214 (n=1) and MECOM rearrangement (n=1). These latter 11 patients, accounting for 19 % of the other group, originally classified as intermediate risk became re-classified as poor risk following FISH screening.
Cytogenetics is important to identify those significant chromosomal abnormalities involved in paediatric AML, which are used to stratify patients for treatment. FISH enables accurate detection of rare cryptic abnormalities associated with poor risk, meaning that more patients can benefit from appropriate risk stratification. Thus we are confident that FISH provides a reliable detection method to be implemented in MyeChild01, alongside screening for those mutations of prognostic relevance: NPM1, CEBPA and FLT3-ITD.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.