Na. Heerema et al., Hypodiploidy with less than 45 chromosomes confers adverse risk in childhood acute lymphoblastic leukemia: A report from the children's cancer group, BLOOD, 94(12), 1999, pp. 4036-4045
We have determined the prognostic significance of hypodiploidy (<46 chromos
omes) in a large cohort of children with acute lymphoblastic leukemia (ALL)
treated by the Children's Cancer Group. Among 1,880 patients, 110 (5.8%) h
ad hypodiploid karyotypes: 87 had 45 chromosomes, 15 had 33 to 44 chromosom
es, none had 29 to 32 chromosomes, and 8 had 24 to 28 chromosomes (near-hap
loidy). Six-year event-free survival (EFS) estimates for patients with 45 c
hromosomes, 33 to 44 chromosomes, or 24 to 28 chromosomes were 65% (standar
d deviation [SD], 8%), 40% (SD, 18%), and 25% (SD, 22%), respectively (log
rank, P = .002: test for trend, P = .0009). The combined hypodiploid group
had worse outcome than nonhypodiploid patients, with 6-year EFS of 58% (SD,
7%) and 76% (SD, 2%), respectively (P < .0001). EFS for the subgroup with
45 chromosomes was similar to that of patients with pseudodiploidy (P = .43
) or 47 to 50 chromosomes (P = .76). None of the patients with 24 to 28 chr
omosomes had a t(4:11), a t(9;22), or a t(1;19), and most received highly i
ntensive therapy. The adverse risk associated with 33 to 44 and 24 to 28 ch
romosomes remained significant in multivariate analyses adjusted for import
ant risk factors including age, white blood cell count, and Philadelphia ch
romosome status. Thus, hypodiploidy with less than 45 chromosomes, particul
arly 24 to 28 chromosomes, is a significant adverse risk factor despite tre
atment with contemporary intensive therapies. (C) 1999 by The American Soci
ety of Hematology.