The Genomic Landscape of <b><i>PAX5</i></b>, <b><i>IKZF1</i></b>, and <b><i>CDKN2A/B </i></b>Alterations in B-Cell Precursor Acute Lymphoblastic Leukemia

Abstract

We present a comprehensive comparison of PAX5,IKZF1, and CDKN2A/B abnormalities in 21 B-cell precursor acute lymphoblastic leukemia (B-ALL) patients studied by aCGH and gene-specific FISH assays. In our cohort of B-ALL patients, alterations of IKZF1, PAX5, and CDKN2A/B were detected by aCGH analysis in 43, 52, and 57% of samples, respectively. Deletions of IKZF1 were present in 9 samples, including 5 cases positive for both PAX5 and IKZF1 deletions, implying digenic impairment. Furthermore, all cases with IKZF1 deletions also had additional genomic alterations, including BCR-ABL1 gene fusions, PAX5 deletions, CDKN2A/B deletions, and FLT3 amplification. Deletions of CDKN2A/B represented the most frequent abnormalities in our group of patients. Our study demonstrates the high incidence of PAX5, IKZF1, and CDKN2A/B alterations in B-ALL detected by aCGH analysis. Due to the small size and variability in the deletion breakpoints, FISH studies showed false-negative results in 10, 40, and 28% of the samples tested for the IKZF1,PAX5, and CDKN2A/B gene deletions, respectively. The PAX5 and IKZF1 abnormalities are highly specific to B-ALL and can be used as diagnostic markers. Moreover, IKZF1 alterations frequently coexist with a BCR-ABL gene fusion. Our study revealed multiple additional B-ALL-specific genomic alterations and showed that aCGH is a more sensitive method than FISH, allowing whole genome profiling and identification of aberrations of diagnostic and prognostic significance in patients with B-ALL.

Chromosomal microarray analysis is superior in identifying cryptic aberrations in patients with acute lymphoblastic leukemia at diagnosis/relapse as a single assay

“…This is concordant with previous findings suggesting that IKZF1 deletion is a characterization of BCR-ABL1 ALL. 13,21,28 Deletions at 9p21.3 involving CDKN2A/2B are well known in adult ALL, and the frequencies are similar among other studies and ours. 25,26,30 However, the high occurrence of homozygous deletions (55.6%) revealed in this study has not been reported by others.…”

“…Among the deleted cases, 3 patients (ALL5, 13, and 32) had hemizygous deletions, 4 patients (ALL7, 10, 20, and 23) had homozygous deletions, and 2 patients (ALL25 and 28) had biallelic deletions of different size on each chromosome 9 homologue. Such complexity was evidenced by a previous study . Gain of 1q23.3 to q44 in 2 cases was due to a derivative chromosome 19 which resulted from translocations between chromosomes 1 and 19, whereas in the third case (ALL8), it was considered an additional finding by CMA as CC for this case was incomplete due to poor chromosome morphology.…”

“…In addition, the low resolution, the relatively subjective nature of CC, the need for dividing cells, the number of metaphases studied, and the limited number of FISH probes available are considerable limitations when compared to CMA. Even commercially available FISH probes yielded false‐negative results in 10%‐40% of ALL cases due to the small sizes of CNAs . Moreover, the labor‐intensive specimen processing and time‐consuming demands of CC and FISH analyses contribute to a longer turnaround time.…”

“…For example, as shown in ALL13, a 103 kb hemizygous deletion involving CDKN2A/2B spans only about half of the 190 kb LSI p16 probe. The FISH assay would invariably yield a false‐negative result …”

“…Even commercially available FISH probes yielded false-negative results in 10%-40% of ALL cases due to the small sizes of CNAs. 21 Moreover, the labor-intensive specimen processing and time-consuming demands of CC and FISH analyses contribute to a longer turnaround time. To date, deletions of at least 10 loci, including PAX5, IKZF1, CDKN2A/2B, ETV6, BTG1, and RB1, which were all recurrent CNAs in this study, have been integrated into diagnostic classification and risk stratification in ALL.…”