Marker chromosomes can arise from chromothripsis and predict adverse prognosis in acute myeloid leukemia

Metaphase karyotyping is an established diagnostic standard in acute myeloid leukemia (AML) for risk stratification. One of the cytogenetic findings in AML is structurally highly abnormal marker chromosomes. In this study, we have assessed frequency, cytogenetic characteristics, prognostic impact, and underlying biological origin of marker chromosomes. Given their inherent gross structural chromosomal damage, we speculated that they may arise from chromothripsis, a recently described phenomenon of chromosome fragmentation in a single catastrophic event. In 2 large consecutive prospective, randomized, multicenter, intensive chemotherapy trials (AML96, AML2003) from the Study Alliance Leukemia, marker chromosomes were detectable in 165/1026 (16.1%) of aberrant non- core- binding-factor (CBF) karyotype patients. Adverse-risk karyotypes displayed a higher frequency of marker chromosomes (26.5% in adverse- risk, 40.3% in complex aberrant, and 41.2% in abnormality(17p) karyotypes, P <.0001 each). Marker chromosomeswere associated with a poorer prognosis compared with other non-CBF aberrant karyotypes and led to lower remission rates (complete remission 1 complete remission with incomplete recovery), inferior event-free survival as well as overall survival in both trials. In multivariate analysis, marker chromosomes independently predicted poor prognosis in the AML96 trial <= 60 years. As detected by array comparative genomic hybridization, about one-third of marker chromosomes (18/49) had arisen from chromothripsis, whereas this phenomenon was virtually undetectable in a control group of marker chromosome-negative complex aberrant karyotypes (1/34). The chromothripsis-positive cases were characterized by a particularly high degree of karyotype complexity, TP53 mutations, and dismal prognosis. In conclusion, marker chromosomes are indicative of chromothripsis and associated with poor prognosis per se and not merely by association with other adverse cytogenetic features. (Blood. 2017; 129(10): 1333-1342)