TABLE 1

 Abbreviations: FISH, fluorescent in situ hybridization; IgVH, immunoglobulin heavy-chain variable; ND, not determined.

 According to the manufacturer's protocol, 50 l of RosetteSep antibodies cocktail (Stemcell Technology, Grenoble, France) were added per 1 ml of peripheral blood (collected in heparin-coated tubes). After 20 min of incubation at room temperature, B cells were recovered following Ficoll sedimentation gradient. The purity of CD19⁺ cells was superior to 95% as confirmed by flow cytometry. These cells were used for phenotyping, FISH experiments, RNA and DNA extraction and for the experiments establishing their sensitivity for DNA damage-induced apoptosis (after irradiation of neocarzinostatin treatments). The 39 patients included in this study were diagnosed with typical B-CLL following cytological and immunologic analyses. Matutes scoring system has been established to distinguish between lymphoproliferative disorders.¹⁴ This score is based on the immunophenotypic analysis of five markers, giving a value of 1 for each marker according to whether it is typical of CLL: CD5⁺, CD23⁺, FMC7^- and CD79b^-, weak expression of monotypic or light chain. Diagnosis of typical CLL corresponds to a Matutes score at 4 or 5. Patients were followed up at Pitié-Salpêtrière Hospital (Paris, France) between 1988 and 2006. Approval for this study was obtained from the Institutional Review Boards of the CEA and the Pitié-Salpêtrière Hospital. Among the 39 CLL patients (25 patients having B cells sensitive to DNA damage-induced apoptosis and 14 patients with resistant cells), 27 were men and 12 were women with a quasi-identical average age of 74 years. IgVH mutation status has been determined for 30/39 patients and cytogenetics analysis for 19/39 patients. The somatic mutational status of IgVH genes has been established according to the 98% homology cutoff value to the closest germline gene; that is less than 98% homology was considered as mutated (M), whereas 98% or greater homology was considered as unmutated (U). The analysis of IgVH genes was done on both genomic DNA and complementary DNA amplified by PCR as described elsewhere.¹⁵ Chromosomal aberrations were established by FISH using LSI ATM/LSI p53 and LSI D13S319/LSI 13q34/CEP12 probe sets (Abbott, France) which are complementary to 11q22.3, 17p13.1, 13q14.3, 13q34 and 12p11.1–q11 genome regions, respectively. Presented data take into account the percentage of interphases' nuclei with the same chromosomal anomaly (that is at least 5% of nuclei should contain the same aberration).

^a Sex was defined: M, male; F, female and age is expressed in years.

^b Staging was according to Binet staging system.¹⁶ Treatments (if any) were received at least 3 months before cell sampling for DNA extraction and apoptosis in vitro determination. The serum CD23 level was established at the time of diagnosis (first sCD23 column), 'max sCD23' refers to the maximal level of CD23 in the course of the disease and 'sCD23 PVT' the level at the time of cell sampling for the present study.

^c Apoptosis score was established 24 h after cell exposure to 10 Gy of -rays or after exposure to neocarzinostatin (10 nM), a DNA double-strand breaks-inducing drug, as previously described.⁴ Apoptosis in vitro was considered as S, sensitive, when the number of apoptotic cells after 24 h post-irradiation was at least twofold higher than in controls (24 h of cell culture in absence of irradiation). R, resistant subset, was defined for the cases where no difference was observed between irradiated and non-irradiated cell samples, according to the score of apoptotic cells after 24 h of cell culture following irradiation compared to untreated cells. Patients who changed status are indicated in bold characters.