Development of cytogenetic methods has contributed to the understanding that ALL is not a homogenous disease. Detection of structural and numerical alterations in the chromosomes of lymphoblasts has identified several different ALL subgroups. G-banding reveals about 60–70% of these changes. The development of FISH and PCR methods has decreased the proportion of apparently normal karyotype to less than 20%. Still a part of ALL patients have no chromosomal aberrations detected with conventional cytogenetics. It seems likely, however, that ALL with a normal karyotype reflects rather our inadequate capability to detect all possible aberrations than a true normality of the lymphoblast genome. Microarray methods offer an effective tool to define novel cytogenetic changes in ALL. In our study characterizing and evaluating ALL in adolescents and young adults aged 10–25 years in Finland, we analyzed patients diagnosed during 1990–2007 (n=231). Eighty-nine patients had normal (n=80) or failed (n=9) karyotype at diagnosis. DNA from initial samples was available for 27 of these 89 patients. 26 patients had normal karyotypes, for one patient the karyotype analysis failed by G-banding at diagnosis. The key clinical characteristics of the 27 patients did not differ from the rest of the patients with normal or failed karyotype. Genomic DNA was extracted from diagnostic bone marrow samples. Digestion, labeling and hybridization of DNA was performed according to the Agilent protocol version 2.0 for 44K arrays. Labeled samples were hybridized against gender matched reference DNAs to Human Genome CGH 44B oligo microarray slides (Agilent Technologies Santa Clara, CA, USA). For data-analysis Agilent’s CGH Analytics software version 3.5 was used. The starting and ending points of the aberrations were confirmed by the ADM-2 algorithm with 10.0 threshold. The immunophenotype of the patients was as follows: T-cell ALL 8/27 patients, precursor B ALL 13/27, mixed lineage 5/27 (according to the European Group for the Immunological Characterization of Leukemias), not known 1/27. Seventeen patients had normal karyotype and no other marker for MRD follow-up, while 9 patients had either immunoglobulin and/or T-cell receptor gene rearrangement (n=8) or over-expression of Willms Tumor gene 1 (WT1) (n=1). In total 58 aberrations were detected in the 27 patient samples (1–7 aberrations per sample, mean 2.1) (Figure 1). Four samples (15%) did not show any aberration (two with immunoglobulin and/or T-cell receptor gene rearrangements). Losses were detected in 20/27 cases and gains in 10/27 cases. Cases with losses only were more frequent (n=13, 48%) than those with gains only (n=3, 11%). Losses were more numerous than gains (44/20 vs. 14/10). Single aberrations were seen in seven patients. Five of these were deletions affecting 9p21.3, two were gains in 21q. In the 27 cases, the most commonly detected aberrations were deletions involving 9p21.3 (n=10), 5/10 (50%) being T-ALL. In all the 10 cases the CDKN2A gene was affected. Other aberrations seen more than once were deletion of 6q (n=4), amplification of the terminal part of 21q (n=3), amplification of 1q (n=2), deletion of 12p (n=2), deletion in 12q23-q24 (n=2), deletion in 16q22 (n=2), deletion in 17q11 (n=2) and deletion in 22q11 (n=2). Nineteen relatively small aberrations (about 2 Mb or less in size) were detected in 15 cases and such deletion was found to be the only aberration in 4/15 cases. Our data indicate that a subgroup of ALL with fully normal cytogenetics may not exist. Microarray CGH shows a clear benefit in more detailed examination of the blast cell DNA. In 85% (23/27) of the patients with initially normal karyotypes we determined single or multiple aberrations with array CGH. Losses were more frequent than gains. Seven patients (26%) had only a single aberration, three of these being submicroscopic (<200 kb).We conclude that microarray CGH enables to detect molecular-genetic changes also in ALL cases having a “normal” karyotype using conventional cytogenetics. We are getting closer to the point where normal molecular-genetic findings do not exist in leukemic lymphoblasts.
Disclosures: No relevant conflicts of interest to declare.