Patients. A total of 71 acute leukemia patients were referred to our laboratory between January 2002 and August 2004. The study was approved by the Institutional Review Board and Ethics Committee. Samples were taken from patients only after informed consent was obtained. Heparinized bone marrow and/or peripheral blood samples were obtained from patients at initial diagnosis and before any specific therapy was initiated, and were sent to our laboratory for further analysis.

Cultures and G-Banding. Direct, overnight and 72-hour lymphocyte cultures were set up from the samples. Metaphase chromosomes were obtained according to standard procedures and analyzed after G-banding [4]. Karyotypes were set up according to the 1995 International System for Human Cytogenetic Nomenclature (ISCN) [5]. To qualify as a structural clonal aberration, at least two cells with the same chromosomal change had to be found and at least three abnormal metaphases had to be identified to qualify as a chromosomal aneuploidy.

Fluorescent In Situ Hybridization Studies. The FISH studies were done using commercial probe sets including t(9;22) dual color, dual fusion DNA probe (Q-BIO Gene, Cambridge, MA, USA) t(15;17) dual color translocation probe (Cytocel, Cambridge, Cambridgeshire, UK), inversion 16 dual color breakpoint region probe (Cytocel) and locus specific identification (LSI) MLL (11q23) break apart rearrangement dual color probe (Vysis, Downers Grove, IL, USA). Information about the probes is shown in Table 1. Probes were selected according to the type of leukemia and hybridization was performed according to the manufacturer’s protocols. Two hundred nuclei or available metaphase spreads, were analyzed per patient, per probe. Two experienced operators, who were unaware of each other’s results or of the conventional cytogenetics results, analyzed the slides. Nuclei that were overlapping or in which the number of signals was ambiguous were not scored. Cut-off values for each probe were calculated in peripheral blood samples of 10 healthy volunteers by adding three times the standard deviation to the mean percentage of abnormal cells analyzed. For the t(9;22) transloca?tion, t(15;17) translocation, MLL (11q23) break apart and inv(16) breakpoint region probes the cut-off values were calculated to be 4.15, 3.02, 3.16 and 1.93%, respectively.

Table 1. Fluorescent in situ hybridization probes, gene regions they include and signal colors used in the study according to leukemia type.