High-Resolution Genomic Scan for Cryptic Chromosomal Lesions in MDS and AML.

Cytogenetic analysis is of eminent importance for the diagnosis and prognosis of hematologic malignancies. Due to limitations of traditional karyotyping, novel technologies which improve resolution and sensitivity are under development. In array-based comparative genomic hybridization (A-CGH), differentially labeled test and reference DNA samples are hybridized to genomic microarrays. Differences in sequence copy number between the samples are reflected in a shift of the fluorescent intensity. The resolution of A-CGH is limited solely by the number of clones; it is theoretically possible to achieve linear coverage of the chromosomes. The principle of the CGH techniques allows for detection of unbalanced chromosomal changes of the whole genome. These types of genomic aberrations are most common in MDS, but may exist and further subclassify malignancies with defined balanced translocations. In MDS, depending on the study, 40–60% of patients have a normal or non-informative karyotype by traditional methods. It is likely that this number may be reduced if the resolution and sensitivity level is increased. Additionally, diagnosis of patients with known chromosomal abnormalities can be further refined.

We first applied A-CGH to the analysis of normal marrow (N=8) to establish whether it will detect chromosomal defects that may acquired and are compatible with normal hematopoiesis. Moreover, defects may be present in healthy elderly. We utilized arrays of up to 2621 clones with a maxium coverage of 1Mb (Vysis, Spectral Genomics). The results were verified by a dye-swap protocol on two arrays per sample. Four controls showed a normal array profile or only changes in clones previously identified as having a polymorphic copy number within the human genome. The remaining controls had changes including a loss of material on 6p (N=1), loss of 6p and 8q material (N=1) and a gain of 4p and loss of 9p sequences (N=1). These changes may reflect unidentified polymorphisms. In contrast, one control had gains of multiple contiguous clones on chromosomes 9, 15 and 22. We also studied the marrow of patients with advanced MDS (N=43) using A-CGH and traditional cytogenetics. The cohort included patients with known singular lesions (N=7) and complex karyotypes (N=1). The remaining patients had either normal or non-informative cytogenetics. For a del 5q patient and a trisomy 21 patient, A-CGH verified the karyotype without identifying further lesions, in a second del 5q patient was a gain of material on 19p, and a monosomy X patient had a gain of 1p36 by CGH. In 3 cases with partially clonal defects, A-CGH did not detect the abnormality. A normal genomic composition was confirmed in a patient with noninformative (N=1) and normal (N=1) karyotypes. Losses of material on 2q and 3q and gains of material on 22q and the 11p telomeric region were identified in a patient with normal cytogenetics, while another "normal" had gains on 2p, 14q and 21q. Additionally, one normal karyotype had loss of chromosome 16 material and one had loss of 6p sequences. This pilot study demonstrates the utility of A-CGH analysis to study chromosomal aberrations in MDS. A-CGH allows for the detection of cytogenetically undetected abnormalities. Analysis of a large number of samples may allow for the detection of consensus defects or global genomic instability with clinical implications.