Integration of genomic and gene expression data of childhood ALL without known aberrations identifies subgroups with specific genetic hallmarks

Abstract

Pediatric acute lymphoblastic leukemia (ALL) comprises genetically distinct subtypes. However, 25% of cases still lack defined genetic hallmarks. To identify genomic aberrancies in childhood ALL patients nonclassifiable by conventional methods, we performed a single nucleotide polymorphisms (SNP) array-based genomic analysis of leukemic cells from 29 cases. The vast majority of cases analyzed (19/24, 79%) showed genomic abnormalities; at least one of them affected either genes involved in cell cycle regulation or in B-cell development. The most relevant abnormalities were CDKN2A/9p21 deletions (7/24, 29%), ETV6 (TEL)/12p13 deletions (3/24, 12%), and intrachromosomal amplifications of chromosome 21 (iAMP21) (3/24, 12%). To identify variation in expression of genes directly or indirectly affected by recurrent genomic alterations, we integrated genomic and gene expression data generated by microarray analyses of the same samples. SMAD1 emerged as a down-regulated gene in CDKN2A homozygous deleted cases compared with nondeleted. The JAG1 gene, encoding the Jagged 1 ligand of the Notch receptor, was among a list of differentially expressed (up-regulated) genes in ETV6-deleted cases. Our findings demonstrate that integration of genomic analysis and gene expression profiling can identify genetic lesions undetected by routine methods and potential novel pathways involved in B-progenitor ALL pathogenesis.

DNA copy-number abnormalities do not occur in infant ALL with t(4;11)/MLL-AF4

“…The 100K SNP array platform used in this study has a median intra-markers resolution of 22.5 Kb, lower than that of other studies (Table 1). Although it can be argued that the low resolution of the SNP array platform could fail in detecting CNA, on the contrary, several arguments sustain that the sensitivity of the method applied in this study does not impair the results: (1) large CNAs were observed at the relapse phase; (2) the same platform was recently used in a subgroup of older childhood ALL, 27 allowing us to identify the same CNA detected by other groups with higher resolution; [13][14][15][16] (3) the comparison of our infant ALL 100K data set with Mullighan's 100K subset (same resolution), further showed that the genotypic profile of infant and older childhood ALL subgroups are different (mean CNA in infant MLL-AF4 ¼ 0.2 versus mean CNA in ETV6-RUNX1 ¼ 6.0 and mean CNA in BCR-ABL ¼ 4.2 in Mullighan et al); (4) the frequency of CNA calculated in this study resulted to be similar to that observed using a SNP platform with higher sensitivity by Mullighan et al in a limited series of 11 MLL-positive cases (Table 1 and Mullighan et al 13 ), although the age of the patients and the identity of the MLL partner genes were not clearly defined in that study; (5) in a subset of 7 infant ALL diagnostic samples of this study, we carried out a more dense SNP analysis using the 250K SNP array technology and confirmed the results obtained by the 100K platform, both for CNA and for UPD (data not shown); (6) finally, the SNP array resolution applied in this study allowed us to detect deletions associated with the somatic rearrangement of Immunoglobulin/T-cell Receptor genes, although not included in the analysis of results, because of their irrelevance to the study. However, as a technical limitation of the method applied, we cannot exclude the presence of balanced chromosomal abnormalities or very small CNA, as well as point mutations.…”

IKZF1 status as a prognostic feature in BCR-ABL1–positive childhood ALL

“…Deletions were validated by either an independent MLPA SALSA p335, SALSA MLPA p202, array comparative genomic hybridization (Sureprint G3 Human CG 180K arrays; Agilent Technologies, Santa Clara, CA) as previously described, 24 or by single nucleotide polymorphism array analysis (GeneChip Human Mapping, 100K Array Set; Affymetrix, Santa Clara, CA), as previously performed. 27 Deletions were classified by assumed effect on protein function in 3 different groups: the "dominant negative group," including all cases with at least one exons 4-7 deleted allele 22 ; the "haploinsufficiency group," including whole gene deletions and deletions affecting exon 2 13,22 ; and the "miscellaneous group," representing all remaining variants.…”

Genome-wide profiling of genetic alterations in acute lymphoblastic leukemia: recent insights and future directions

“…38,[44][45][46][47][48] Using Affymetrix SNP arrays interrogating over 350 000 markers at an average intermarker resolution of less than 5 Kb, we examined 242 diagnosis B-progenitor (N ¼ 192) and T-lineage (N ¼ 50) ALL samples. 38 Several key methodological aspects enabled accurate identification of somatic (tumor-acquired) lesions in this study.…”