SEARCH · PubMed Health
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The somatic hypermutation (SMH) status of the immunoglobulin (Ig) V(H) genes can divide chronic lymphocytic leukemia (CLL) into two prognostic subsets, with mutated V(H) genes display superior survival compared to unmutated cases. Biased V(H) gene usage has also been reported in CLL which may reflect antigen selection. In a V(H) gene analysis of 265 CLL cases we confirmed the prognostic impact of the V(H) mutation status and found preferential V(H) gene usage in both the mutated and unmutated subset. Interestingly, CLL cases rearranging one particular V(H) gene, V(H)3-21, displayed poor outcome despite that two-thirds showed mutated V(H) genes. Many of the V(H)3-21 utilizing cases expressed lambda light chains, rearranged a Vlambda2-14 gene, and had homologous complementarity determining region 3s (CDR3s), implying recognition of a common antigen epitope. We thus believe that the cases rearranging the V(H)3-21 gene comprises an additional CLL entity. We further analyzed the V(H) gene rearrangements and, specifically, the heavy chain CDR3 sequences in 346 CLL cases to investigate the role of antigens in CLL. We identified six new subgroups with similar HCDR3 features and restricted VL gene usage as in the V(H)3-21-using group. Our data indicate a limited number of antigen recognition sites in these subgroups and give further evidence for antigen selection in the development of CLL. Different mutational cutoffs have been used to distinguish mutated CLL in addition to the 2% cutoff. Using three levels of somatic mutations we divided 323 CLLs into subsets with divergent survival (<2%, 2-5% and >5% mutations). This division revealed a low-mutated subgroup (2-5%) with inferior outcome that would have been masked using the traditional 2% cutoff. A 1513A/C polymorphism in the P2X(7) receptor gene was reported to be more frequent in CLL, but no difference in genotype frequencies was revealed in our 170 CLL cases and 200 controls. However, CLL cases with the 1513AC genotype showed superior survival than 1513AA cases and this was in particular confined to CLL with mutated VH genes. In summary, we could define new prognostic subgroups in CLL using Ig gene rearrangement analysis. This also allowed us to gain insights in the biology and potential role of antigen involvement in the pathogenesis of CLL.
The activities of 2-chlorodeoxyadenosine (2-CdA) metabolizing enzymes, deoxycytidine kinase (dCK) and cytosolic 5'-nucleotidase (5'-NT) were measured in control and bryostatin 1 treated CLL cells using an EBV-negative WSU-CLL cell line. This cell line was established from a patient with CLL resistant to fludarabine. The results revealed a significant increase in dCK activity in bryostatin 1 treated cells at 48 and 72 h compared with the control. 5'-NT activity decreased significantly at 48 h. The ratio of dCK to 5'-NT activity was significantly increased in bryostatin 1 treated WSU-CLL cells after 48 h. WSU-CLL cells treated with bryostatin 1 exhibited an increase in the percentage of apoptotic and dead cells from control levels of 16% to 40%. This percentage was further increased to 67% following the addition of 11.2 microM 2-CdA to WSU-CLL cells pretreated with bryostatin 1. Results from Western blot analysis indicate that WSU-CLL cells express high levels of Bcl-2, Bcl-xL and c-myc, and a low level of Bax. p53 in untreated WSU-CLL cells is undetectable. WSU-CLL cells treated with bryostatin 1 showed a significant increase in the ratio of Bax to Bcl-2. To demonstrate that the bryostatin 1 mediated enhancement of 2-CdA efficacy was not restricted to in vitro cell culture, we have studied the tumor growth delay of WSU-CLL xenografts treated with placebo, bryostatin 1, 2-CdA, and bryostatin 1 followed by 2-CdA. SCID mice given bryostatin 1 at 75 microg x kg(-1) x d(-1) for 5 days followed by 30 mg x kg(-1) x d(-1) 2-CdA for 5 days in two cycles, had significantly improved tumor growth delay (P = 0.05). We conclude that bryostatin 1 is not only capable of inducing apoptosis by itself, but also sensitizes de novo resistant WSU-CLL cells to the chemo-therapeutic effects of 2-CdA. The bryostatin 1-induced increased ratio of dCK/5'-NT activity and an increased ratio of Bax/Bcl-2 are at least two mechanisms through which this natural compound is able to potentiate the anti-tumor activity of 2-CdA in otherwise resistant CLL cells.
Previous studies have demonstrated alkylating (melphalan) resistance in the B-CLL derived WSU-CLL cell line as compared to WIL2 B lymphocytic cells. Nuclear extracts from WSU-CLL cells demonstrate a highly significant increase in DNA topoisomerase II activity as compared to WIL2 cells. Western blot analysis showed the level of topoisomerase II proteins expressed in WSU-CLL cells to be increased as compared to WIL2 cells. WSU-CLL cells were 5.24-fold more sensitive than WIL2 cells to the cytotoxic effect of the topoisomerase II inhibitor doxorubicin. No difference in topoisomerase I activity or of the level of topoisomerase I protein expression was observed comparing the two cell lines. The sensitivity to the cytotoxic effects of topoisomerase I inhibitor, camptothecin, did not differ in WSU-CLL and WIL2 cell lines. Pre-incubation with doxorubicin significantly increased melphalan induced interstrand-DNA-crosslink formation and cytotoxicity in WSU-CLL cells as compared to WIL2 cells. The affinity of topoisomerase II for WSU-CLL UV-irradiated-crosslinked DNA was increased 2.84-fold as compared to that of WSU-CLL native DNA. The affinity of topoisomerase II for both UV-irradiated (crosslinked) and for native DNA was significantly decreased after doxorubicin-pretreatment. Measurement of DNA polymerase beta and DNA polymerase beta revealed significant elevations in DNA polymerase beta (58.82 +/- 3.67 units/mg protein in WSU-CLL cells, as compared to 27.82 +/- 4.39 units/mg protein in WIL 2 cells; p < 0.01) but not DNA polymerase beta (0.82 +/- 0.11 units/mg protein in WSU-CLL cells, compared to 0.74 +/- 0.09 units/mg protein in WIL2, p > 0.05). However, exposure to aphidicolin (an inhibitor of DNA polymerase a) failed to increase melphalan induced cytotoxicity suggesting that although DNA polymerase a activity was increased in WSU-CLL cells the mechanisms of resistance does not involve this specific DNA repair pathway. Elevated topoisomerase II activity and the increased affinity of topoisomerase II for crosslinked DNA in melphalan resistant cells appears to be the major factor responsible for alkylator resistance by changing DNA topology and thereby facilitating DNA repair.
The T1 surface antigen (CD5,p67) expression on blood lymphocytes (PBL) and lymphoid cells from lymph node biopsies (LN) from 31 patients with B-cell chronic lymphocytic leukemia (B-CLL) and 79 with B non-Hodgkin lymphoma (B-NHL), was detected in 25 B-CLL (80 per cent) and in 11 B-NHL (13 per cent) belonging to the following histologic subtypes: lymphocytic of CLL type (DLWD) one case, lymphoplasmacytoid (DLWD) four cases, centrocytic (DLPD) five cases, immunoblastic (DH) one case. All B-CLL and the T1 + B-NHL were also tested with monoclonal antibodies against the Common Acute Lymphoblastic Leukemia Antigen, B cells (FMC7, FMC8, BA1, Y29-55), T cells (OKT11a), HLA-DR and HLA-DQ monomorphic determinants. All the B-CLL and the T1+ B-NHL were CALLA-, BA1+, Y29.55+. FMC7+ cells were detected in large numbers six B-CLL (three T1+ and three T1-) and in four centrocytic lymphomas. FMC8 reacted with 70 per cent of leukemias (where it stained 30 per cent of neoplastic cells) and with 8/9 T+ B-NHL. HLA-DR and HLA-DQ molecules were detected in 100 per cent and 90 per cent of cases respectively. In vitro treatment of HLA-DQ- or T1- B-CLL with phorbol ester TPA led to the expression of these antigens as well as of the receptors for Interleukin 2 and MLR3 activation antigen. Surface membrane Ig (SIg) was detected in 79 per cent of cases, its density measured by FACS analysis varied, even markedly, from case to case. Among the B-CLL, cells with high SIg content were either T1+ or T1- and more likely FMC7+. The SIg- cases were seven B-CLL (five T1+ and two T1-) and two B-NHL, in which, however, cytoplasmic IgM was detected. This study reveals the existence of four major B-CLL subgroups: T1- SIg-, T1+ SIg+, T1+ SIg+, T1- SIg+. It also indicates that the T1 antigen may be transitionally present during B-cell differentiation and that its expression may precede that of SIg as supported by the in vitro studies. In addition, the finding that some B-NHL are T1+ suggests that they derive similarly to the B-CLL from a common progenitor.
Despite recent advances, chronic lymphocytic leukemia (CLL) as the most common leukemia remains a largely incurable disease. Modern treatment options include novel drugs like purine analogues, monoclonal antibodies and transplantation strategies. Moreover, gene transfer of immunostimulatory molecules is another, but still experimental approach that can be used to potentiate immune responses against leukemic cells. CD40 ligand (CD40L) was shown to be a promising molecule for immunotherapy of B-CLL playing a critical role in immune activation. However, CLL B cells are resistant to transduction with most currently available vector systems. Improving the efficiency and specificity of gene vectors is critical for the success of gene therapy in this area. Using replication defective adenovirus encoding CD40L (Ad-CD40L), immunologic and clinical responses were seen in CLL patients after infusion of autologous Ad-CD40L-CLL cells in a recent phase I trial. Due to the immunogenic nature of adenovirus vectors, alternative vector systems are currently explored. Recombinant adeno-associated virus (rAAV) was shown to enable efficient transduction of primary B-CLL cells. By use of a library of AAV clones with randomly modified capsids, receptor-targeting mutants with a tropism for CLL cells can be selected. Furthermore, helper-virus free Epstein-Barr virus (EBV)-based gene transfer vectors hold promise for development of CLL-targeted vaccines after remaining safety issues will be resolved. Herpes simplex virus (HSV)-based vectors, especially HSV amplicons, have favorable features for B-CLL gene transfer including high transduction efficiency, ability to infect postmitotic cells and a large packaging capacity. The challenge for the future will be to transfer these alternative vector systems into clinic and allow the detection of a CLL-specific immune response by use of defined tumor antigens. This will make it possible to establish the potential clinical role of gene therapy for CLL patients.
Type 4 cyclic adenosine monophosphate (cAMP) phosphodiesterase (PDE4) inhibitors and other agents that raise intracellular cAMP levels induce apoptosis in B-cell chronic lymphocytic leukemia (B-CLL) but not in T-CLL or peripheral blood T cells. Two principal effector proteins for cAMP are protein kinase A (PKA) and EPAC (exchange protein directly activated by cAMP), a Rap guanosine 5'-diphosphate (GDP) exchange factor. We here examine whether varying expression of EPAC accounts for the discrepant sensitivity of B-CLL and T cells to PDE4 inhibitor-induced apoptosis. B-CLL and peripheral blood B cells express EPAC1 transcript, whereas T-CLL, peripheral blood T cells, monocytes, and neutrophils do not. Treatment with the PDE4 inhibitor rolipram induces Rap1 activation in B-CLL cells but not in peripheral blood B cells, T-CLL, or any of the normal hematopoietic lineages examined. The EPAC-specific cAMP analog 8CPT-2Me-cAMP (8-(4-chloro-phenylthio)-2'-O-methyladenosine-3',5'-cAMP) activates Rap1 in B-CLL cells, but, unlike rolipram/forskolin or 8-Bromo-cAMP, it does not induce PKA activation, as judged by phosphorylation of the transcription factor cAMP-response element binding protein (CREB). Unexpectedly, whereas rolipram/forskolin and 8-Bromo-cAMP induce apoptosis in B-CLL cells, 8CPT-2Me-cAMP decreased basal apoptosis in B-CLL cells by an average of 25% (P<.002). Our results demonstrate that B-CLL cells uniquely activate Rap1 in response to PDE4 inhibitors and suggest that physiologic stimuli that activate EPAC may transmit an antiapoptotic signal.
Several features of chronic lymphocytic leukemia (CLL) suggest that immune-based strategies may have therapeutic potential. A promising approach is provided by the transduction of CLL cells with CD40 ligand (CD40L) by viral vectors to enhance their immunogenicity. We compared the antigen-presenting capacity of CD40L-transduced CLL cells with mock-transduced or CD40L-stimulated CLL cells (CD40-CLL). A significantly higher number of T cells could be expanded using CD40L-transduced CLL cells as antigen-presenting cells (APCs) compared with the control group (P = .008). Using 5 different CLL-associated tumor antigens, including fibromodulin, MDM2 (murine double minute 2), survivin, p53, and KW-13, we show in interferon-gamma (IFN-gamma) enzyme-linked immunospot (ELISPOT) assays after 35 days of in vitro culture that the number of antigen-specific autologous T cells was also significantly higher when CD40L-transduced CLL cells were used as APCs (P < .001). Thus, CD40L-transduced CLL cells are able to induce an antigen-specific T-cell response and might be superior to CD40-CLL cells for immune-based therapeutic strategies in CLL.
Chronic Lymphocytic Leukemia (CLL) is clinically divided into IGHV mutated (M-CLL) and IGHV unmutated (U-CLL) subtypes, which are thought to arise from distinct cells of origin along the B-cell differentiation pathway. We measured genome-scale DNA methylation in purified CLL samples ( n = 89) and utilized reference-based cell deconvolution techniques to develop a continuous metric of epigenetic similarity across a B-naive-like to B-memory-like scale (B-Index). B-Index accurately classifies CLL into clinical subtypes (98.8%), has a stronger epigenetic signal than IGHV gene percent identity, and demonstrates additional epigenetic signal within the M-CLL subgroup. We demonstrate that U-CLL is epigenetically more similar to B-memory than B-naive cells and reconcile previous reports of a B-naive-like epigenetic signal. The B-memory-like program of U-CLL is enriched for binding sites of transcription factors related to the germinal center activation pathway. Our findings provide epigenetic evidence for discerning CLL mechanisms of initiation and cell of origin. We also identified an epigenetic signal associated with tumor burden, which may have some relation to viral infections such as Epstein-Barr-Virus. Our cell-type deconvolution-based approach to developing a continuous metric for CLL epigenetic differentiation state can be applied to other tumors with multiple subtypes across differentiation stages.
We have studied the apparent kinetic parameters of the ecto-nucleotide triphosphatase from CLL B lymphocytes and compared them to blood and tonsillar B and T cells. The Vmax of the ecto-ATPase activity in CLL B lymphocytes, was 65 +/- 10 fmol Pi/cell per 30 min compared to 37 +/- 2.1 in blood B lymphocytes, and 8.5 +/- 1.7 in blood T lymphocytes. The ATPase of membranes prepared from CLL, tonsillar B and T, and blood T lymphocytes had a relationship among the cell types similar to that seen in intact cells. However, no difference in the km for ATP, .17 mM, or the km for magnesium, .15 mM was found in the ecto-ATPase of CLL lymphocytes as compared to blood or tonsillar B cells. The ectoenzyme of CLL cells hydrolyzed GTP, ITP, CTP, and UTP as well as ATP. Further, ATP added to an enzyme assay containing an alternative nucleotide did not result in increased phosphate release. Nucleotide acceptance of blood B and T lymphocytes was very similar to that of CLL B cells. ATP inhibited phosphate release when present in excess of magnesium in both CLL and blood B lymphocytes. These data indicate that there is greater ectonucleotide triphosphatase activity in tonsillar and blood B lymphocytes, including CLL, as compared either to blood or tonsillar T lymphocytes. However, CLL cells showed no qualitative difference from blood or tonsillar B cells in ectonucleotidase activity. Thus, the higher activity in CLL cells is "B cell-like" and might reflect, also, their maturation stage or monoclonal origin.
The response of lymphocyte subpopulations to staphylococcal Protein A (SPA) was studied in patients with B-cell chronic lymphocytic leukemia (CLL) and normal volunteers. The kinetics of the proliferative response, optimal dose and sera requirements were determined. Of 92 experiments conducted in 60 patients, SPA induced peripheral blood mononuclear cells (PBMC) proliferation in 81.5% of cases studied. The mean proliferative response of CLL PBMC was significantly lower than normal PBMC, 5823 vs. 30,364 cpm. However, enriched CLL B-cells failed to respond to SPA compared to normal enriched B-cells, with mean cpm 444 vs. 6438 respectively. As PBMC from CLL consist of increased proportions of B-cells and decreased proportions of T-cells, enriched CLL B-cells were cultured at 1: 1 ratio with autologous or allogeneic normal T-cell enriched fractions. CLL B-lymphocytes were stimulated by SPA when cultured in the presence of T-lymphocytes, indicating a T-helper defect in the B-CLL proliferative response to SPA, rather than an intrinsic inability of CLL B-cells to proliferate. These observations are of import for further studies of CLL B-lymphocyte function, cytogenetics, and establishment of CLL B-cell lines in culture.
OBJECTIVE: CD229, a cell-surface molecule being involved in cell adhesion, is overexpressed in B-CLL cells. In this study we wanted to explore whether CD229 might function as B-CLL-specific tumor-associated antigen (TAA). PATIENTS AND METHODS: Autologous, CD229-specific HLA-A2-restricted T cells were identified using IFN-gamma-ELISPOT assays and HLA-A2/dimer-peptide staining after 4 weeks of in vitro culture. RESULTS: We were able to expand autologous T cells from 9/11 B-CLL patients using native B-CLL cells as antigen presenting cells (APCs) in 5 cases, whereas for 4 samples an autologous T-cell response could only be evoked by use of CD40L-stimulated B-CLL cells as APCs. The number of CD8+ T cells could be expanded during 4 weeks of in vitro culture with native or CD40L-activated B-CLL cells while the amount of specific T cells recognizing CD229 peptides bound to HLA-A2 dimers increased on average 12-fold (native CLL) and 13-fold (CD40L-activated CLL), respectively. Using IFN-gamma-ELISPOT assays we could demonstrate that the expanded T cells were able to secrete IFN-gamma upon recognition of the antigen. These T cells not only recognized HLA-A0201-binding CD229-derived peptides presented by T2 cells, but also CD229-overexpressing autologous B-CLL cells in an MHC-I-restricted manner. CONCLUSION: In summary, CD229 was shown to be naturally processed and presented as TAA in primary B-CLL cells, enabling the expansion of autologous tumor-specific T cells.
Relative protein expression levels were compared in leukemic B cells from two patients with chronic lymphocytic leukemia (CLL) having either mutated (M-CLL) or unmutated (UM-CLL) immunoglobulin variable heavy chain genes (IgV(H)). Cells were separated into cytosol and membrane protein fractions then labeled with acid-cleavable ICAT reagents (cICAT). Labeled proteins were digested with trypsin then subjected to SCX and affinity chromatography followed by LC-ESI-MS/MS analysis on a linear ion trap mass spectrometer. A total of 9 proteins from the cytosol fraction and 4 from the membrane fraction showed a 3-fold or greater difference between M-CLL and UM-CLL and a subset of these were examined by Western blot where results concurred with cICAT abundance ratios. The abundance of one of the proteins in particular, the mitochondrial membrane protein cytochrome c oxidase subunit COX G was examined in 6 M-CLL and 6 UM-CLL patients using western blot and results showed significantly greater levels (P < 0.001) in M-CLL patients vs UM-CLL patients. These results demonstrate that stable isotope labeling and mass spectrometry can complement 2D gel electrophoresis and gene microarray technologies for identifying putative and perhaps unique prognostic markers in CLL.
Initial work has shown that clonal B cells from B-chronic lymphocytic leukemia (B-CLL) are able to synthesize pro-angiogenic molecules. In this study, our goal was to study the spectrum of angiogenic factors and receptors expressed in the CLL B cell. We used ELISA assays to determine the levels of basic fibroblast growth factors (bFGF), vascular endothelial growth factor (VEGF), endostatin, interferon-alpha (IFN-alpha) and thrombospondin-1 (TSP-1) secreted into culture medium by purified CLL B cells. These data demonstrated that CLL B cells spontaneously secrete a variety of pro- and anti-angiogenic factors, including bFGF (23.9 pg/ml +/- 7.9; mean +/- s.e.m.), VEGF (12.5 pg/ml +/- 2.3) and TSP-1 (1.9 ng/ml +/- 0.3). Out of these three factors, CLL B cells consistently secreted bFGF and TSP-1, while VEGF was expressed in approximately two-thirds of CLL patients. Of interest, hypoxic conditions dramatically upregulated VEGF expression at both the mRNA and protein levels. We also employed ribonuclease protection assays to assay CLL B cell expression of a variety of other angiogenesis-related molecules. These analyses revealed that CLL B cells consistently express mRNA for VEGF receptor 1 (VEGFR1), thrombin receptor, endoglin, and angiopoietin. Further analysis of VEGFR expression by RT-PCR revealed that CLL B cells expressed both VEGFR1 mRNA and VEGFR2 mRNA. In summary, these data collectively indicate that CLL B cells express both pro- and anti-angiogenic molecules and several vascular factor receptors. Because of the co-expression of angiogenic molecules and receptors for some of these molecules, these data suggest that the biology of the leukemic cells may also be directly impacted by angiogenic factors as a result of autocrine pathways of stimulation.
Chronic lymphocytic leukemia (CLL) cells develop chemo-resistance over time. Most anticancer agents function through induction of apoptosis, and therefore resistance against these agents is likely to be caused by selection for CLL cells with defects in the particular apoptosis pathway that is triggered by these drugs. Anticancer agents that function through alternative apoptotic pathways might therefore be useful in treating chemo-resistant CLL. Triterpenoids represent a class of naturally occurring and synthetic compounds with demonstrated antitumor activity. We examined the effects of CDDO (triterpenoid 2-cyano-3,12-dioxoolean-1,9-dien-28-oic acid) on CLL B cells in vitro. CDDO induced apoptosis in a dose-dependent manner in all (n = 30) CLL samples tested, including previously untreated and chemo-resistant CLL specimens. CDDO induced rapid proteolytic processing of caspase-8, but not caspase-9, in CLL B cells, suggesting activation of a mitochondria-independent pathway. CDDO-induced apoptosis of CLL B cells was blocked by cytokine response modifier A (CrmA), a suppressor of caspase-8, but not by X-linked inhibitor of apoptosis protein-baculovirus IAP repeat-3 (XIAP-BIR3), a fragment of XIAP, which selectively inhibits caspase-9. Examination of CDDO effects on expression of several apoptosis-relevant genes demonstrated significant reductions in the levels of caspase-8 homolog Fas-ligand interleukin-1-converting enzyme (FLICE)-inhibitory protein (c-FLIP), an endogenous antagonist of caspase-8. However, reductions of FLIP achieved by FLIP antisense oligonucleotides were insufficient for triggering apoptosis, indicating that CDDO has other targets in CLL B cells besides FLIP. These data suggest that the synthetic triterpenoid CDDO should be further explored as a possible therapeutic agent for treatment of chemo-resistant CLL.
Nucleotide insertions/duplications or deletions in immunoglobulin heavy chain genes have been found in 24/760 patients (3.15%) with chronic lymphocytic leukemia (CLL). In 21/24 cases, the inserted/duplicated or lost nucleotides occurred in multiples of 3; therefore, the original reading frame was maintained and a potentially intact receptor was coded. The pattern and location of insertions/duplications or deletions in CLL and their restriction to mutated IGHV rearranged genes strongly suggests that they resulted from somatic hypermutation. Their incidence in CLL is consistent with previous reports in normal, auto-reactive and neoplastic human B cells, thus seemingly indicating that these modifications generally arise without any particular disease-specific associations. A striking exception to this rule was identified in CLL IGHV3-21-expressing cases: one amino acid was deleted from the CDR2 region in 16/63 (25.4%) mutated CLL IGHV3-21 sequences (including public database-derived IGHV3-21 CLL cases + the present series) vs. only 2/257 (0.78%) public database-derived mutated non-CLL IGHV3-21 sequences; 15/16 CLL IGHV3-21 sequences carrying this deletion belonged to a subset with unique, shared HCDR3 and light chain CDR3 motifs. This finding further supports the idea of selective antigenic pressures playing a pathogenetic role in some CLL cases.
Analysis of peripheral blood (>85% CD19+/CD5+ B) lymphocytes, obtained from 44 patients affected by B chronic lymphoid leukemia (B-CLL), showed that surface TNF-related apoptosis inducing ligand (TRAIL) was expressed in all samples and at higher levels with respect to unfractionated lymphocytes and purified CD19+ B cells, obtained from 15 normal blood donors. Of note, in a subset of B-CLL samples, the addition to B-CLL cultures of a TRAIL-R1-Fc chimera, which binds at high affinity to surface TRAIL, significantly decreased the percentage of viable cells with respect to untreated control B-CLL cells, suggesting that surface TRAIL may play an unexpected role in promoting B-CLL cell survival. In spite of the majority of B-CLL lymphocytes expressed variable surface levels of "death receptors" TRAIL-R1 and TRAIL-R2, the addition in culture of recombinant TRAIL increased (>20% vs. controls) the degree of spontaneous apoptosis in only 11/44 of the B-CLL samples, had no effect in 19/44, while it significantly increased leukemic cell survival in 14/44. Taken together, these findings suggest that an aberrant expression of TRAIL might contribute to the pathogenesis of B-CLL by promoting the survival in a subset of B-CLL cells.