Microarray-Based Genomic Profiling Facilitates Genetic Subgrouping and Detects Critical Submicroscopic Aberrations Associated with Prognosis in Pediatric Acute Lymphoblastic Leukemia (ALL)

Risk-adapted therapeutic categories in acute lymphoblastic leukemia (ALL) take into account several key parameters, including cytogenetics. Because accurate conventional chromosome (CC) studies in ALL are hampered by low mitotic indexes and poor chromosome morphology, fluorescence in situ hybridization (FISH) and other molecular methods such as RT-PCR are currently used to complement karyotyping. We evaluated the contribution of oligo/SNP microarrays for providing additive genetic information in ALL that is not obtained by karyotype studies.

Specimens from 24 children and young adults (12 M;12 F), including 3 patients with Down syndrome (DS), were processed for pre-B ALL cytogenetics work-up plus SNP/Oligo microarray. The median age was 4 y (range, 2–21 y); 23 patients had a pre-B-cell immunophenotype. Unstimulated CC (n=23) and pre-B ALL FISH studies (n=20) were performed using standard protocols. Genomic DNA was extracted from the residual bone marrow samples and processed for genome-wide copy number analyses on the Cytoscan HD oligo/SNP microarray (Affymetrix).

CC detected abnormalities that allowed prognostic subgrouping in 19 (83%) of 23 patients tested; the 24th patient was not tested with CC but showed an ETV6-RUNX1 fusion on FISH. Microarray genomic profiling allowed genetic subgrouping in the 4 cases with suboptimal or non-informative CC results. Overall, microarray detected a median of 5 additional copy number aberrations (CNAs) per patient (range, 1–27), including 18 additional CNAs in a T-cell ALL patient with only deletion 9p detected by CC and FISH. The 4 most common deletions detected by array involved CDKN2A (n=10, including 4 biallelic deletions) and ETV6, SESN1/6q16.1, and IKZF1 (6 cases each); sporadic deletions involved genes affecting B-cell development, cell cycle progression, DNA repair, and tumor progression were also seen. Five of the 7 patients with ETV6-RUNX1 translocation also showed deletions or disruptions at or near these 2 loci, suggesting the presence of the “cryptic” t(12;21). No balanced translocations were detected. Clonal diversity was easily detectable by microarray; however, a case with 64 chromosomes and a case with both 2n and 4n clones were difficult to interpret. At least 1 extended area of copy neutral loss-of heterozygosity (>5 Mb) was seen in 8/24 (33%) cases, including a 17q region that encompassed IKZF3; however, in most cases the significance of these CN-LOH changes was not clear. Significant “high risk” prognostic alterations identified by array but not detected by CC included 3 CRLF2-rearragements (found in 2 of the 3 DS patients) and disruption of the IKZF1 locus (6 patients). IKZF1 deletions were detected in a 5-y-old DS-ALL patient with CRLF2-P2PY8, a 20-y-old DS-ALL patient with high hyperdiploidy, an 18-y-old patients with IGH-CRLF2 confirmed by FISH (CC failed), another 18-y-old patient with a normal karyotype, and 1 patient each with iAMP(21) and dic(9;20) ALL.

Submicroscopic IKZF1 deletions have been associated with drug resistance and a high risk of treatment failure in ALL, signifying critically important prognostic information needed for clinical management. Accordingly, OligoSNP arrays provide a comprehensive approach for accurately identifying clinically significant abnormalities in ALL that may be missed by routine chromosome study and targeted FISH panels alone. Array testing is a highly sensitive complementary molecular cytogenetic assay that should be offered to newly-diagnosed ALL patients, especially when CC is non-informative, to facilitate genetic subgrouping and define tumor markers that may help monitor a patient's clinical course.

No relevant conflicts of interest to declare.