Search

Acute lymphoblastic leukemia (ALL) represents the most common childhood malignancy. Although survival for pediatric B-ALL has approached 90%, variability in outcome among and within cytogenetically defined subgroups persists. While G-banding and fluorescence in situ hybridization (FISH) have been used to characterize leukemic clones, there is added value of chromosomal microarray and next generation sequencing in screening genome-wide for copy number aberrations, copy neutral loss of heterozygosity and nucleotide variations. Evaluation of novel genetic aberrations can provide information about the biologic mechanisms of cytogenetically defined subgroups associated with poor prognosis, explain heterogeneity in patient outcome and identify novel targets for therapeutic intervention. The high risk B-ALL intrachromosomal amplification of chromosome 21, (iAMP21), subtype is characterized by amplification of a region of chromosome 21 that typically encompasses the RUNX1 gene and is associated with poor prognosis. Analysis of chromosomal microarray and FISH data revealed that deletions of SH2B3, encoding a negative regulator of multiple tyrosine kinase and cytokine signaling pathways, are enriched among leukemias harboring iAMP21. Enrichment of SH2B3 aberrations in the iAMP21 subtype may indicate that loss of SH2B3 contributes to disease progression and raises the possibility that these leukemias may be sensitive to tyrosine kinase inhibitors.

Hyperdiploidy with greater than 50 chromosomes is usually associated with favorable prognosis in pediatric acute lymphoblastic leukemia (ALL), whereas hypodiploidy with ≤43 chromosomes is associated with extremely poor prognosis. Sometimes, hypodiploidy is "masked" and patients do not have a karyotypically visible clone with ≤43 chromosomes. Instead, their abnormal karyotypes contain 50-78 or more chromosomes from doubling of previously hypodiploid cells. When the hypodiploid and doubled hyperdiploid clones are both present, patients can be identified by traditional test methods [karyotype, DNA Index (DI), fluorescence in situ hybridization (FISH)], but the incidence of masked hypodiploid cases in which only the doubled clone is visible is unknown. We analyzed 7013 patients with B-ALL enrolled in COG AALL03B1 (2003-2011) for whom chromosome studies were available. Of 115 patients with hypodiploidy (25-39 chromosomes), karyotypes of 40 showed only the hypodiploid clone, 47 showed mosaicism with both hypodiploid and hyperdiploid (doubled) karyotypes, and 28 with masked hypodiploidy showed only a hyperdiploid (doubled) clone. Unique karyotypic signatures were identified, and widespread loss of heterozygosity (LOH) was seen in the microsatellite panel for all patients with masked hypodiploidy. An increased awareness of the unusual karyotypic profile associated with a doubled hypodiploid clone and coordinated use of DI, FISH, and LOH studies when indicated can identify patients with masked hypodiploidy and allow appropriate treatment selection.

Copy number alterations (CNAs) are a hallmark of pediatric cancer genomes. An increasing number of research groups use multiple platforms and software packages to detect and analyze CNAs. However, different platforms have experimental and analysis-specific biases that may yield different results. We sought to estimate the concordance of CNAs in children with de novo acute myeloid leukemia between two experimental platforms: Affymetrix SNP 6.0 array and Illumina OmniQuad 2.5 BeadChip. Forty-five paired tumor-remission samples were genotyped on both platforms, and CNAs were estimated from total signal intensity and allelic contrast values using the allele-specific copy number analysis of tumors (ASCAT) algorithm. The two platforms were comparable in detection of CNAs, each missing only two segments from a total of 42 CNAs (4.6%). Overall, there was an interplatform agreement of 96% for allele-specific tumor profiles. However, poor quality samples with low signal/noise ratios showed a high rate of false-positive segments independent of the genotyping platform. These results demonstrate that a common analytic pipeline can be utilized for SNP array data from these two platforms. The customized programming template for the preprocessing, data integration, and analysis is publicly available at https://github.com/AplenCHOP/affyLumCNA.