Search

CCR Translation for this Article from Expression of Mutated IGHV3-23 Genes in Chronic Lymphocytic Leukemia Identifies a Disease Subset with Peculiar Clinical and Biological Features

Supplementary Data from Expression of Mutated IGHV3-23 Genes in Chronic Lymphocytic Leukemia Identifies a Disease Subset with Peculiar Clinical and Biological Features

Purpose: B-cell chronic lymphocytic leukemia (CLL) is a clinically heterogeneous disease whose outcome can be foreseen by investigating the mutational status of immunoglobulin heavy chain variable (IGHV) genes. Moreover, a different prognosis was reported for CLL expressing specific IGHV genes in the context or not of stereotyped B-cell receptors. Here we investigated novel associations between usage of specific IGHV genes and clinical features in CLL.Experimental Design: Among 1,426 CLL-specific IG-rearrangements, stereotyped B-cell receptor clusters never utilized the IGHV3-23 gene. Given this notion, this study was aimed at characterizing the IGHV3-23 gene in CLL, and identifying the properties of IGHV3-23–expressing CLL.Results: IGHV3-23 was the second most frequently used (134 of 1,426) and usually mutated (M; 109 of 134) IGHV gene in our CLL series. In the vast majority of M IGHV3-23 sequences, the configuration of the 13 amino acids involved in superantigen recognition was consistent with superantigen binding. Clinically, M IGHV3-23 CLL had shorter time-to-treatment than other M non–IGHV3-23 CLL, and multivariate analyses selected IGHV3-23 gene usage, Rai staging, and chromosomal abnormalities as independent prognosticators for M CLL. Compared with M non–IGHV3-23 CLL, the gene expression profile of M IGHV3-23 CLL was deprived in genes, including the growth/tumor suppressor genes PDCD4, TIA1, and RASSF5, whose downregulation is under control of miR-15a and miR-16-1. Accordingly, relatively higher levels of miR-15a and miR-16-1 were found in M IGHV3-23 compared with M non–IGHV3-23 CLL.Conclusions: Altogether, expression of the IGHV3-23 gene characterizes a CLL subset with distinct clinical and biological features. Clin Cancer Res; 16(2); 620–8

Purpose: Del17p13 predicts poor outcome and chemorefractoriness in chronic lymphocytic leukemia (CLL). Conversely, it is unknown whether TP53 mutations carry any prognostic value independent of del17p13. We tested the independent prognostic value of TP53 mutations in CLL.Experimental Design: The study was based on a consecutive series of 308 CLL. DNA sequencing of TP53 exons 2 to 10 and del17p13 interphase fluorescence in situ hybridization were done at CLL diagnosis. Study end points were survival and chemorefractoriness.Results: At diagnosis, TP53 mutations (n = 32) occurred in 31 of 308 (10.0%) patients. Of all CLL showing TP53 disruption by either mutation and/or deletion (n = 44), 10 cases (22.7%) showed TP53 mutations in the absence of del17p13. Multivariate analysis selected TP53 mutations (hazard ratio, 3.20; P = 0.002) as an independent predictor of overall survival after adjustment for del17p13. Also, multivariate analysis selected TP53 mutations (hazard ratio, 3.97; P < 0.001) as an independent predictor of chemorefractoriness after adjustment for del17p13. Compared with cases without TP53 alterations, CLL harboring any type of TP53 disruption (mutation only, del17p13 only, or both mutation and del17p13) uniformly displayed a high prevalence of unfavorable prognosticators and poor outcome. Analysis of sequential CLL samples showed the acquisition of new or additional TP53 alterations at the time of chemorefractoriness.Conclusions: These data show that (a) TP53 mutations are an independent predictor of short survival and chemorefractoriness, and (b) that CLL presenting with TP53 mutations without del17p13 fare as poorly as CLL carrying del17p13. Because CLL harboring TP53 mutations without del17p13 are currently not recognized by conventional diagnostic strategies, these results may be relevant for a comprehensive prognostic characterization of CLL.

Supplementary Tables S1-S4 from The Prognostic Value of TP53 Mutations in Chronic Lymphocytic Leukemia Is Independent of Del17p13: Implications for Overall Survival and Chemorefractoriness

IGHV1-69/51p1 is expressed by ∼ 30% of unmutated chronic lymphocytic leukemia (U-CLL) and combines with selected IGHD and IGHJ genes generating stereotypes if HCDR3 amino acid homology is > 60%. We had previously revealed stereotypic IGHV1-69/IGHJ6 rearrangements in normal naive B cells, thereby identifying potential counterparts of U-CLL. A different stereotypic IGHV1-69/IGHD3-16(RF2)/IGHJ3 rearrangement carrying the CAR(GGx)YD motif in the N1-region, recurrent in 6% IGHV1-69+ve CLL, is exceptionally sequence restricted, strongly suggestive of shared antigen recognition. We have now analyzed IGHV1-69/IGHJ3 rearrangements in circulating B cells of healthy individuals using several PCR-based approaches with IGHV1-69/IGHJ3 CLL sequences for reference. Stereotypes were found, but all were distinct from CLL. Remarkably, even a highly sensitive semi-nested PCR, specific for the CLL-expressed IGHV1-69/IGHD3-16(RF2)/IGHJ3 stereotype, failed to identify the CAR(GGx)YD sequence, although similar motifs were found. These highly specific B cells are not apparent in the accessible normal repertoire and may expand in response to rarely expressed antigens important in the pathogenesis of CLL.

Richter syndrome (RS) represents the transformation of chronic lymphocytic leukaemia (CLL) to aggressive lymphoma. Risk factors of CLL transformation to RS are only partly known. We explored the role of the host genetic background as a risk factor for RS occurrence. Forty-five single nucleotide polimorphisms (SNPs) known to be relevant for CLL prognosis were genotyped in a consecutive cohort of 331 CLL, of which 21 had transformed to RS. After correcting for multiple testing and adjusting for previously reported RS risk factors, the LRP4 rs2306029 TT variant genotype was the sole SNP independently associated with a higher risk of RS transformation (Hazard Ratio: 4·17; P = 0·001; q = 0·047). The enrichment of LRP4 TT genotype in RS was confirmed in an independent series (n = 44) used for validation purposes. The LRP4 protein was expressed in CLL (n =66). Bioinformatic analysis scored LRP4 rs2306029 as a variant with possible deleterious and damaging variant of LRP4. LRP4 genotyping may help the recognition of patients with increased risk of RS at the time of CLL diagnosis.

Intrinsic apoptosis defects underlie to a large extent the extended survival of malignant B cells in chronic lymphocytic leukemia (CLL). Here, we show that the Shc family adapter p66Shc uncouples the B-cell receptor (BCR) from the Erk- and Akt-dependent survival pathways, thereby enhancing B-cell apoptosis. p66Shc expression was found to be profoundly impaired in CLL B cells compared with normal peripheral B cells. Moreover, significant differences in p66Shc expression were observed in patients with favorable or unfavorable prognosis, based on the mutational status of IGHV genes, with the lowest expression in the unfavorable prognosis group. Analysis of the expression of genes implicated in apoptosis defects of CLL showed an alteration in the balance of proapoptotic and antiapoptotic members of the Bcl-2 family in patients with CLL. Reconstitution experiments in CLL B cells, together with data obtained on B cells from p66Shc−/− mice, showed that p66Shc expression correlates with a bias in the Bcl-2 family toward proapoptotic members. The data identify p66Shc as a novel regulator of B-cell apoptosis which attenuates BCR-dependent survival signals and modulates Bcl-2 family expression. They moreover provide evidence that the p66Shc expression defect in CLL B cells may be causal to the imbalance toward the antiapoptotic Bcl-2 family members in these cells.

The cell of origin of chronic lymphocytic leukemia (CLL) has long been sought, and immunoglobulin gene analysis provides new clues. In the unmutated subset (U-CLL), there is increased usage of the 51p1-related alleles of the immunoglobulin heavy chain variable 1-69 gene, often combined with selected genes and with immunoglobulin heavy chain diversity IGHJ6. Stereotypic characteristics of the HCDR3 result and suggest antigen selection of the leukemic clones. We have now analyzed 51p1/IGHJ6 combinations in normal blood B cells from 3 healthy persons for parallel sequence patterns. A high proportion (33.3% of sequences) revealed stereotypic patterns, with several (15.0%) being similar to those described in U-CLL. Previously unreported CLL-associated stereotypes were detected in 4.8%. Stereotypes (13.6%) not detected in CLL also were found. The HCDR2-IGHJ6 sequences were essentially unmutated. Junctional amino acids in normal B cells were heterogeneous, as in cases of stereotyped CLL. Phenotypically, normal B cells expressing 51p1-derived immunoglobulin M were naive. This snapshot of the naive B-cell repertoire reveals subsets of B cells closely related to those characteristic of CLL. Conserved patterns in the 51p1-encoded immunoglobulin M of normal B cells suggest a restricted sequence repertoire shaped by evolution to recognize common pathogens. Proliferative pressure on these cells is the likely route to U-CLL.

Richter's syndrome (RS) represents the transformation of chronic lymphocytic leukaemia (CLL) to aggressive lymphoma and is mostly represented by diffuse large B-cell lymphoma (DLBCL), with a post-germinal centre (GC) phenotype, clonally related to the pre-existing CLL. RS has a very poor prognosis and its pathogenetic mechanisms are poorly understood. In order to gain additional hints in RS pathogenesis, we performed a genome-wide DNA profiling study of 13 RS phases and eight matched CLL phases using the Affymetrix Human Mapping 250K NspI SNP arrays. Individual genomic profiles were heterogeneous, with no individual lesions occurring in more than half of the cases. However, several observations suggest that MYC pathway might be involved in RS. The 13q13.3-qter region containing MIRHG1 (MIR-17-92), a cluster of microRNA interacting with c-MYC, was acquired at the time of transformation. The 13q gain was coupled with the gain of c-MYC and loss of TP53. Translocation of c-MYC was acquired at transformation in a fraction of cases and this event appeared mutually exclusive with gain of MIRHG1. MYCN, a c-MYC homologue, was also recurrently gained. By comparing RS with 48 de nova DLBCL, RS presented a significantly lower prevalence of deletions affecting the PRDM1 and TNFAIP3, genes on 6q, known to be associated with a post-GC phenotype. In conclusion, the genomic profile of RS seems to differ from what observed in de novo DLBCL and in other transformed DLBCL. Genomic lesions occurring in RS are heterogeneous suggesting the existence of different RS subsets, possibly due to different transforming mechanisms. A deregulation of MYC pathway might represent one of the main transformation events in the pathogenesis of a subset of RS clonally related to the previous CLL. Copyright (C) 2009 John Wiley & Sons, Ltd.

Fludarabine plus cyclophosphamide (FC) at conventional doses is an effective treatment for chronic lymphocytic leukaemia (CLL). However, FC at standard doses may give hematological and non-hematological toxicity, predominantly in the elderly. Intravenous or oral low-dose FC regimens remain highly effective in elderly patients with Low-Grade Lymphomas other than CLL and are well tolerated. We tested efficacy and toxicity of oral FC at reduced doses in 26 elderly patients (median 71 years) with previously untreated (UT-CLL, n = 14) or relapsed/refractory CLL (R-CLL, n = 12), unfit for conventional treatments. Twentyfour-of-26 (92%) patients (14/14, 100% UT-CLL; 10/12, 83.5% R-CLL) obtained a response, with 12/26 (46%) complete responses (9/14, 64.2% in UT-CLL; 3/12, 25% in R-CLL). Non-hematological toxicity was mild and myelosuppression was documented in 8/26 (31%) patients (4/14, 28% UT-CLL; 4/12, 33% R-CLL). With a median follow-up of 24 months, median event-free survival was 48 months with no differences between UT-CLL and R-CLL and all responders were alive. Low-dose oral FC treatment showed good efficacy in both untreated and refractory/relapsed CLL. The treatment is useful in elderly patients who cannot benefit of more aggressive schedules and is easy to administer on an outpatient basis.

Hairy cell leukemia (HCL) is a distinct clinicopathologic entity that responds well to purine analogs but is sometimes difficult to differentiate from HCL-like disorders (eg, splenic marginal zone lymphoma and HCL variant). We recently identified the BRAF-V600E mutation as the disease-defining genetic event in HCL. In this study, we describe a new, simple, and inexpensive test for genetics-based diagnosis of HCL in whole-blood samples that detects BRAF-V600E through a sensitive allele-specific PCR qualitative assay followed by agarose-gel electrophoresis. This approach detected BRAF-V600E in all 123 leukemic HCL samples investigated containing as few as 0.1% leukemic cells. BRAF-V600E was detected at different time points during the disease course, even after therapy, pointing to its pivotal role in HCL pathogenesis and maintenance of the leukemic clone. Conversely, 115 non-HCL chronic B-cell neoplasms, including 79 HCL-like disorders, were invariably negative for BRAF-V600E. This molecular assay is a powerful tool for improving the diagnostic accuracy in HCL.

The B cell receptor (BCR) is the functional distinguishing unit that defines any B cell. Immunoglobulin gene (IG) status is preserved in the neoplastic B cell clone and can provide an indicator of the maturation stage reached by the B cell prior to transformation. In hairy cell leukemia (HCL), several data from IG analysis provide clear hints regarding the cell of origin and the ongoing selective interactions of the tumor BCR with environmental stimuli: HCLs (i) have variable levels of IG somatic mutations, (ii) continue somatic mutations at low levels, (iii) have active processes of IG class switching after transformation, and (iv) have a relatively high frequency of selective events in the light chain of the BCR. It has recently emerged that an unmutated status of the HCL-derived IG can be associated with failure to respond to cladribine, genetic abnormalities indicative of poor outcome, and aggressive disease. These observations suggest that IG gene analysis may have biological and prognostic relevance in HCL and merits further characterization.

The B cell receptor (BCR) is the functional distinguishing unit that defines any B cell. Immunoglobulin gene (IG) status is preserved in the neoplastic B cell clone and can provide an indicator of the maturation stage reached by the B cell prior to transformation. In hairy cell leukemia (HCL), several pieces of data from IG analysis provide clear hints regarding the cell of origin and the ongoing selective interactions of the tumor BCR with environmental stimuli. HCLs have variable levels of IG somatic mutations, and continue somatic mutations at low levels as well as IG class switching after transformation. More recent data also show the occurrence of selective events in the light chain of the BCR, suggesting a dominant role for IG status in the pathogenesis of HCL. Moreover, it has recently emerged that an unmutated status of the HCL IG can be associated with failure to respond to cladribine, genetic abnormalities indicative of poor outcome, and aggressive disease. These observations suggest that IG analysis may have biological and prognostic relevance in HCL and merits further characterization.

Analyses of the tumour immunoglobulin (Ig) gene (IG) heavy (H) and light chains show heterogeneity of mutational status, but reveal common features of ongoing IGH isotype-switching with multiple IGH isotype expression and preference of IG lambda (IGL) light chain with selective use of IGLJ3. Phenotypic and immunogenetic analyses were performed in a series of 105 HCL patients to estimate prevalence of multiple IG light chain expression by the tumour cells. By phenotype, 3/105 HCL (2.9%) expressed double tumour-related Ig kappa (K) and L light chain proteins. By immunogenetic analysis, functional mutated double IGK(I) /IGK(II) , IGK(I) /IGL(I) and IGL(I) /IGL(II) transcripts were cloned and sequenced in 3/71 (4.2%) HCL. These latter three HCL expressed multiple IGH isotypes with mutated IGHVDJ rearrangements at the time of AID transcript expression. Most interestingly, the three cases had reinduced RAG1 transcript. In the double IGL expresser, single-cell analysis documented co-expression of the tumour-related IGLs in 5/6 cells (83%). In the IGK/IGL co-expresser, evidence of surface IgK/IgL isotype proteins confirmed functionality of the tumour-derived transcripts. The evidence of double light chain expression in single HCs and the new observation of RAG re-induction suggest ongoing selective influences on the BCR that may promote or maintain the HCL clone in the periphery.

To the editor: Sellick et al[1][1] showed the prognostic role of single nucleotide polymorphisms (SNPs) in chronic lymphocytic leukemia (CLL). The primary end point of the study was progression-free survival (PFS) after first treatment.[1][1] Several factors may limit the general applicability of

Purpose: B-cell chronic lymphocytic leukemia (CLL) is a clinically heterogeneous disease whose outcome can be foreseen by investigating the mutational status of immunoglobulin heavy chain variable (IGHV) genes. Moreover, a different prognosis was reported for CLL expressing specific IGHV genes in the context or not of stereotyped B-cell receptors. Here we investigated novel associations between usage of specific IGHV genes and clinical features in CLL. Experimental Design: Among 1,426 CLL-specific IG-rearrangements, stereotyped B-cell receptor clusters never utilized the IGHV3-23 gene. Given this notion, this study was aimed at characterizing the IGHV3-23 gene in CLL, and identifying the properties of IGHV3-23–expressing CLL. Results: IGHV3-23 was the second most frequently used (134 of 1,426) and usually mutated (M; 109 of 134) IGHV gene in our CLL series. In the vast majority of M IGHV3-23 sequences, the configuration of the 13 amino acids involved in superantigen recognition was consistent with superantigen binding. Clinically, M IGHV3-23 CLL had shorter time-to-treatment than other M non–IGHV3-23 CLL, and multivariate analyses selected IGHV3-23 gene usage, Rai staging, and chromosomal abnormalities as independent prognosticators for M CLL. Compared with M non–IGHV3-23 CLL, the gene expression profile of M IGHV3-23 CLL was deprived in genes, including the growth/tumor suppressor genes PDCD4, TIA1, and RASSF5, whose downregulation is under control of miR-15a and miR-16-1. Accordingly, relatively higher levels of miR-15a and miR-16-1 were found in M IGHV3-23 compared with M non–IGHV3-23 CLL. Conclusions: Altogether, expression of the IGHV3-23 gene characterizes a CLL subset with distinct clinical and biological features. Clin Cancer Res; 16(2); 620–8

The clinical relevance of angiopoietin-2 (Ang2) in chronic lymphocytic leukemia (CLL) was previously suggested by the association between high Ang2, and shorter progression-free survival reported in small series of patients. Here, we evaluated Ang2 glycoprotein levels in plasma samples collected from a multicentric cohort of CLL patients (n = 316) using an enzyme-linked immunosorbent assay method, and we investigated its prognostic role in relation to time to first treatment (TTFT) and overall survival. Based on a cutoff equal to 2459 pg/mL, we divided our cohort in 2 subsets (high and low Ang2) composing 100 (31.6%) and 216 (68.4%) patients, respectively. High Ang2 was predictive of reduced TTFT (P < .001) and overall survival (P = .002). Multivariate analysis confirmed that high Ang2 was an independent prognosticator for TTFT (hazard ratio = 1.739; 95% confidence interval, 1.059-2.857; P = .029). Significant associations were found between high Ang2 and advanced Binet stages (P < .001), high β2-microglobulin (P < .001), unmutated variable region of immunoglobulin heavy chain gene status (P < .001), high CD38 and ζ-chain-associated protein kinase 70 expression (P < .001 and P = .003), and intermediate/high cytogenetic risk (P = .005). Moreover, Ang2 added prognostic power to other conventional prognosticators and helped to refine prognosis among CLL subsets with both high and low vascular endothelial growth factor plasma levels. Ang2 plasma level may be a useful independent prognosticator for CLL.

Antigen stimulation may be important for splenic marginal zone lymphoma pathogenesis. To address this hypothesis, the occurrence of stereotyped B-cell receptors was investigated in 133 SMZL (26 HCV+) compared with 4,414 HCDR3 sequences from public databases. Sixteen SMZL (12%) showed stereotyped BCR; 7 of 86 (8%) SMZL sequences retrieved from public databases also belonged to stereotyped HCDR3 subsets. Three categories of subsets were identified: i) "SMZL-specific subsets" (n=5), composed only of 12 SMZL (9 HCV-from our series); ii) "Non-Hodgkin's lymphoma-like subsets" (n=5), comprising 5 SMZL (4 from our series) clustering with other indolent lymphomas; iii) "CLL-like subsets" (n=6), comprising 6 SMZL (3 from our series) that belonged to known CLL subsets (n=4) or clustered with public CLL sequences. Immunoglobulin 3D modeling of 3 subsets revealed similarities in antigen binding regions not limited to HCDR3. Overall, data suggest that the pathogenesis of splenic marginal zone lymphoma may involve also HCV-unrelated epitopes or an antigenic trigger common to other indolent lymphomas.

Monoclonal B-cell lymphocytosis (MBL) is an asymptomatic monoclonal expansion of5.0 x 10(9)/l circulating CLL-phenotype B-cells. The relationship between MBL and Rai 0 CLL, as well as the impact of biological risk factors on MBL prognosis, are unknown. Out of 460 B-cell expansions with CLL-phenotype, 123 clinical MBL (cMBL) were compared to 154 Rai 0 CLL according to clinical and biological profile and outcome. cMBL had better humoral immune capacity and lower infection risk, lower prevalence of del11q22-q23/del17p13 and TP53 mutations, slower lymphocyte doubling time, and longer treatment-free survival. Also, cMBL diagnosis was a protective factor for treatment risk. Despite these favourable features, all cMBL were projected to progress, and lymphocytes1.2 x 10(9)/l and3.7 x 10(9)/l were the best thresholds predicting the lowest and highest risk of progression to CLL. Although IGHV status, CD38 and CD49d expression, and fluorescence in situ hybridization (FISH) karyotype individually predicted treatment-free survival, multivariate analysis identified the presence of +12 or del17p13 as the sole independent predictor of treatment requirement in cMBL (Hazard ratio: 5.39, 95% confidence interval 1.98-14.44, P = 0.001). Overall, these data showed that cMBL has a more favourable clinical course than Rai 0 CLL. Given that the biological profile can predict treatment requirement, stratification based on biological prognosticators may be helpful for cMBL management.

Purpose: Del17p13 predicts poor outcome and chemorefractoriness in chronic lymphocytic leukemia (CLL). Conversely, it is unknown whether TP53 mutations carry any prognostic value independent of del17p13. We tested the independent prognostic value of TP53 mutations in CLL. Experimental Design: The study was based on a consecutive series of 308 CLL. DNA sequencing of TP53 exons 2 to 10 and del17p13 interphase fluorescence in situ hybridization were done at CLL diagnosis. Study end points were survival and chemorefractoriness. Results: At diagnosis, TP53 mutations (n = 32) occurred in 31 of 308 (10.0%) patients. Of all CLL showing TP53 disruption by either mutation and/or deletion (n = 44), 10 cases (22.7%) showed TP53 mutations in the absence of del17p13. Multivariate analysis selected TP53 mutations (hazard ratio, 3.20; P = 0.002) as an independent predictor of overall survival after adjustment for del17p13. Also, multivariate analysis selected TP53 mutations (hazard ratio, 3.97; P < 0.001) as an independent predictor of chemorefractoriness after adjustment for del17p13. Compared with cases without TP53 alterations, CLL harboring any type of TP53 disruption (mutation only, del17p13 only, or both mutation and del17p13) uniformly displayed a high prevalence of unfavorable prognosticators and poor outcome. Analysis of sequential CLL samples showed the acquisition of new or additional TP53 alterations at the time of chemorefractoriness. Conclusions: These data show that (a) TP53 mutations are an independent predictor of short survival and chemorefractoriness, and (b) that CLL presenting with TP53 mutations without del17p13 fare as poorly as CLL carrying del17p13. Because CLL harboring TP53 mutations without del17p13 are currently not recognized by conventional diagnostic strategies, these results may be relevant for a comprehensive prognostic characterization of CLL.

A fraction of chronic lymphocytic leukaemia (CLL) cases carry highly homologous B-cell receptors (BCR), i.e. characterized by non-random combinations of immunoglobulin heavy-chain variable (IGHV) genes and heavy-chain complementarity determining region-3 (HCDR3), often associated with a restricted selection of IGVK/L light chains. Such 'stereotyped' BCR occur more frequently in CLL with unmutated (UM) than mutated (M) IGHV genes. We analysed 1426 IG rearrangements (from 1398 CLL cases) by a clustering driven by HCDR3 similarities. Molecular findings were correlated to time-to-treatment (TTT) and presence of known prognosticators. Sixty-nine clusters (319 IG-rearrangements, 22.4%) with stereotyped BCR were identified. Among 30 confirmed clusters (>or=3 IG-rearrangements/cluster), we found 14 novel clusters, of which 11 had M IG rearrangements (M clusters) and predominantly (8/11) used IGHV3 subgroup genes. Recurrent cluster-biased amino acid changes were found throughout IGHV sequences of these 'M clusters'. Regarding clinical outcome: (i) UM CLL from the IGHV1-2/1-3/1-18/1-46/7-4-1/IGKV1-39 cluster had poorer prognosis than UM/M cases, or UM cases using the same IGHV genes but not in clusters; (ii) M CLL from the IGHV3-21/IGLV3-21 cluster had TTT similar to UM CLL, and shorter than M CLL expressing IGHV3-21 but not in cluster. Altogether, our analysis identified additional molecular and clinical features for CLL expressing stereotyped BCR.

Hairy cell leukemia (HCL) is generally responsive to single-agent cladribine, and only a minority of patients are refractory and with poor prognosis. HCLs generally express mutated (M) and, in a minority, unmutated (UM) IGHV. In a multicenter clinical trial in newly diagnosed HCL, we prospectively investigated clinical and molecular parameters predicting response and event-free survival after single-agent cladribine. Of 58 HCLs, 6 expressed UM-IGHV (UM-HCL) and 52 M-IGHV (M-HCL). Beneficial responses were obtained in 53 of 58 patients (91%), whereas treatment failures were observed in 5 of 58 patients (9%). Failures were associated significantly with UM-IGHV (5 of 5 failures vs 1 of 53 beneficial responses had UM-IGHV, P < .001), leukocytosis (3 of 5 vs 3 of 53, P = .006), and bulky spleen (4 of 5 vs 4 of 53, P < .001). The UM-HCL not benefiting from cladribine characteristically had bulky spleen (4 of 5, 80%), leukocytosis (3 of 5, 60%), and TP53 defects (2 of 5, 40%), and progressed rapidly after first treatment (median event-free survival, 7.5 months). Our data suggest that UM-HCLs identify the minor subgroup failing cladribine treatment and with more aggressive disease. High incidence of TP53 dysfunction indicates a potential mechanism of resistance to cladribine in the UM-HCL group. Overall, our data provide new molecular elements relevant for treatment concerns in HCL.

Background We previously reported ongoing mutational and isotype switch events in the immunoglobulin (Ig) heavy chain (H) locus in hairy cell leukemia. Those analyses raised questions on the incidence and type of selective influences occurring on the tumor B-cell receptor of hairy cell leukemia.Design and Methods To further investigate this issue, we examined the full IGH and κ and λ light chains (IGκ and IGλ) variable and constant region transcripts expressed in a large cohort of patients with hairy cell leukemia (n=88).Results Multiple IgH isotypes were expressed in 46/56 (82%) cases of hairy cell leukemia. Comparison of tumor with normal B-cell repertoires revealed preferential usage of IGHV3-21, IGHV3-30 and IGHV3-33 in hairy cell leukemia (p=0.001, p=0.003 and p=0.001, respectively). Light chain analysis demonstrated preferential Igl use with an inverted IGk:IGl ratio (0.7:1) and universal usage of IGLJ3. Analysis of LCDR3 junctions revealed highly homologous motifs in 40% of IGL. Parallel analysis of IGH and IGL showed selective pairing of IGHV3-21/30/33 segments to specific LCDR3-J3 subsets (p=0.008). Of 40 cases of hairy cell leukemia, 38 had mutated IGHV and/or IGK/LV, with variations in 13/13 cloned cases, while two had 100% unmutated IGHV and IGK/LV.Conclusions Overall, biased IGV usage, preference for Igλ with universal IGLJ3 usage and a high incidence of LCDR3 homologous motifs suggest selective influences on the B-cell receptor of hairy cell leukemia. Ongoing mutations and isotype switching suggest that influences occur on the tumor B-cell receptor at ectopic sites.

Hairy cell leukaemia (HCL) is a rare B-cell neoplasm for which the molecular mechanisms are largely unknown. High-density genome-wide DNA profiling was performed with Affymetrix 250K arrays to analyse copy number (CN) changes and loss of heterozygosity (LOH) in 16 cases of HCL. Four of 16 cases (25%) demonstrated gross non-recurrent CN deletions. Within the affected regions, we identified genes involved in bone marrow fibrosis (FGF12) and response to treatment (TP53) in individual cases. Large regions (> 5 Mb) of LOH without any concomitant DNA CN changes were identified in 5/16 (31%) HCL and were indicative of uniparental disomy UD. The germline origin of UD was demonstrated in one case for which a matched normal sample was available. Overall analysis of LOH showed that identical loci were recurrently targeted in chromosomes 1, 2 and 6. As a whole, however, HCL showed a remarkably stable genome. This finding adds to several other features that are unique to HCL among mature B-cell tumours.

Hairy cell leukemia (HCL) is a rare B-cell neoplasm associated with a good outcome after treatment with single-agent purine-analogs (PAs) Pentostatin or Cladribine [1]. However, responses are not u...

Hairy cell leukemia (HCL) is a rare B-cell neoplasm highly responsive to purine analogues 2Chlorodeossiadenosine (2CdA) or Desossicoformicin or Interferons as single agents, and only a minority are refractory. Patients who obtain any response (either complete or partial) tend to have survivals as normal healthy subjects and/or will benefit from repeating the treatments in case of relapse. Conversely, the minority of patients who do not respond to one of the drugs often do not respond to the others and have a poor prognosis. We have recently observed that the majority of HCL have mutated VH genes, while a minority have unmutated VH genes. In the most common B-cell neoplasm chronic lymphocytic leukemia (CLL), VH gene status has prognostic impact and correlates with progression, treatment-response and surivival. In the process of identifying prognostic parameters of responsiveness to 2CdA, we prospectively investigated the VH and VL genes expressed by the tumor cells and response to treatment in patients receiving subcutaneous 2CdA. In newly diagnosed HCL requiring treatment, enrolled in an Italian multicenter trial (ICGHCL2004), peripheral blood mononuclear cells were obtained prior to treatment, and the expressed tumor VH and VL transcripts were identified by RT-PCR and cloning. Tumor sequences with > 98% homology to germline VH and VL genes were defined as “unmutated”. Patients received 0.1 mg/kg subcutaneous 2CdA (Litak) for 5 or 7 consecutive days and responses were evaluated by immunohistochemistry of trephine bone marrow biopsies 2 and/or 6 months after the end of treatment. Of 56 patients recruited, 22 patients were evaluable for response. Definition of response was according to consensus resolution criteria. We observed that 19/22 patients responded to subcutaneous 2CdA (15 CR, 4 PR), while 3/22 patients demonstrated refractory or progressive disease, indicating similar efficacy of subcutaneous to intravenous administration. Leukocytosis was observed in 2/3 refractory, but also in 2/21 responsive patients. In one of one patient in CR, molecular remission was also documented in the bone marrow by PCR and capillary electrophoresis. Most remarkably, the 3/3 refractory HCL shared the common feature of expressing unmutated VH and VL genes, in contrast to the responsive patients that all carried mutated VH and/or VL genes. From our series, there are indications that mutational status may relate with tumor burden (leukocytosis) and, more importantly, with response to 2CdA. Overall, the interim data suggest that HCL patients with unmutated VH genes may not benefit from single agent subcutaneous 2CdA and provide elements to build new clinical trials with combined strategies in cases of refractory/non responsive HCL where the immunogenenetic tumor profile is provided.

Abstract 1369 Introduction: Single-agent cladribine (2CdA) given intravenously or subcutaneously (sc) is highly effective in Hairy cell leukemia (HCL). However, HCL patients are projected to progress in their life-span or, in a minority, are refractory and may benefit from other treatment strategies. Parameters predicting efficacy of (sc)2CdA in HCL are scarcely known. Aims: In a national multicentre clinical trial (EudraCT code: ICGHCL2004) we prospectively sought parameters able to predict efficacy of sc2CdA in classic HCL requiring first treatment (1987 Consensus Criteria). Methods: Clinical data and samples were collected centrally for diagnostic revision according to the WHO criteria and for molecular analyses prior to treatment. Tumor IGH and TP53 analyses or Genome-wide DNA profiling was performed in the cases with >10% or >50% hairy cells (HC) in the test sample, respectively (Forconi, Blood 2009). Patients entering the study received 0.5–0.7 mg/kg sc2CdA as a single course. Efficacy endpoints were response to sc2CdA, treatment free interval (TFI), relapse free survival (RFS), and overall survival (OS). Responses were assessed on month 2 and 6 after treatment and defined according to the 1987 Consensus criteria. Complete Remissions (CR) and Partial Remissions (PR) were rated as beneficial responses (CR/PR), while minor Responses (mR) and No Responses (NR) were rated as failures (mR/NR). TFI was measured in all patients from sc2CdA initiation to time of new treatment requirement. RFS was measured in patients with CR/PR from sc2CdA initiation to relapse. OS was measured from sc2CdA initiation to death. Results: Of 148 patients, 116 were males (78,4%). Median values were: age 52 years (range 30–83), splenomegaly below costal margin 2 cm (0-16), hemoglobin 11,8 (5-17 g/dL), PLT 77×10e9/L (14-255), WBC 2,7×10e9/L (0,7-48), HC 0,2×10e9/L (0-38). Serum B2microglobulin was twice above upper limit (B2M>2X) in 9/122 patients (7,4%), IGH unmutated (UM) in 7/62 (11,3%) and TP53 mutated (mutTP53) in 2/62 (3,2%). Onehundred-forty patients (94,6%) had a CR/PR (101 CR, 68,2%; 39 PR, 26,4%) and 8 (5,4%) a mR/NR (5 mR, 3,4%; 3 NR, 2%). Risk factors of sc2CdA failure were splenomegaly (p=.001, best-cutoff> 3 cm bcm by ROC and Youden's Test), WBC (p10×10e9/L), HC counts (p5×10e9/L), and B2M>2X (p=.013). In 62 patients investigated for molecular features, UM-IGH and TP53 were confirmed as risk factors for mR/NR (p10×10e9/L (p5×10e9/L (p5×10e9/L, spleen>3cm, UM-IGH and mutTP53), UM-IGH (HR=8.1, 95%CI 1.7–38.1, p=.008, median TFI=8.5 months), high HC (HR=6.9, 95%CI 1.5–31.6, p=.012, TFI=27,4 months) and splenomegaly (HR=3.7, 95%CI 1.1–12.8, p=.039, TFI=49,3 months) scored as independent risk factors. Quality of response also predicted risk of short TFI (NR=mR>PR>CR, p5×10e9 (p=.016) and PR (p=.001)as risk factors of short RFS. Cox multivariate analysis revealed PR as the sole independent risk factor of relapse after a median of 46 months (HR=4.5, 95%CI 1,7-12,3, p=.003). 2CdA dose schedule was among other factors that did not associate with TFI or RFS. Conclusion: This data confirm that sc2CdA has potent activity in HCL. Tumor UM-IGH status, splenomegaly, high HC count (or leukocytosis) are confirmed as independent risk factors of treatment failure and rapid progression. RFS analysis identify PR as the sole independent factor of relapse risk in the patients initially responsive to sc2CdA. This analysis may have important implications for the selection of HCL patients that will require treatments alternative to single-agent 2CdA. Disclosures: No relevant conflicts of interest to declare.

Abstract 1371 Background: IGHV1-69 gene identifies the most frequent IGHV gene in chronic lymphocytic leukemia (CLL) and identifies the paradigm of unmutated CLL (U-CLL), being used in roughly 1/3 U-CLL. It is often rearranged to form subsets of stereotyped HCDR3 patterns, likely selected and transformed from the natural naïve B-cell repertoire (Blood. 2010; 115:71-7). Being unmutated, IGHV1-69 CLL are hypothetically expected to have competent tumor B-cell receptors (BCR) and to progress more rapidly. However, it has not been investigated if progression occurs similarly in all the subsets. Aims: we aimed to investigate the prognostic significance of mutational status and of stereotypic B-cell receptors in IGHV1-69+ CLL. Methods: Nucleotide sequences of the tumor IGHV1-69/D/J rearrangement, clinical and molecular prognostic parameters at diagnosis and clinical status at follow-up of 294 IGHV1-69+ CLL patients were obtained from 22 hematological Institutions in Italy. CLL B-cell derived IGHV1-69 rearrangements were scanned for HCDR3 stereotypic patterns and assigned to subsets according to the criteria by Murray et al (Blood. 2008; 111: 1524–1533). Enpoint of outcome was time to progression requiring first treatment according to NCI criteria (TTFT) in Rai stage 0 CLL. Results: Of 294 IGHV1-69+ CLL, 264 (89,8%) were unmutated, 168 (57,1%) were assigned to subsets, of which subsets 7 (n=23, 7,8%), 6 (17, 5,8%), 3 (13, 4,4%), 5 (10, 3,4%) and 9 (10, 3,4%) were the most frequent. CD38, ZAP70, normal or sole del13, +12, del11 and del17p scored positive in 109/264 (58,7%), 139/245 (56,7%), 128/248 (50,5%), 51/248 (20,6%), 43/248 (17,3%) and 34/248 (13,7%). CLLs were reclassified as 18 (6,1%) clinical MBL, 101 (34,4%) Rai 0, 155 (52,7%) Rai I-II and 20 (6,8%) Rai III-IV CLL. Subset 6 was also UM in 16/17 (94,1%) cases. Prevalence of CD38 (p Conclusions: our analysis documents and confirms that unmutated status of IGHV, and not stereotypy, is a relevant prognosticator of outcome (TTFT) in CLL. In the IGHV1-69 CLL it exclude a role of IGHV gene use for CLL progression. However, the good prognosis of Rai 0 U-CLL assigned to subset 6 suggests a differential clinical benign course of this particular subset, irrelevant of unmutated status. One possibility is that the IGHV1-69/D3-16/J3 rearrangements of subset 6 produce a tumor-specific BCR with stereotypic HCDR3 patterns that are anergized by antigen while circulating in the peripheral blood in early stage (Rai 0) CLL. Disclosures: No relevant conflicts of interest to declare.

Abstract 701 Introduction: Hairy cell leukemia (HCL) is generally responsive to intravenous Cladribine (iv2CdA). Subcutaneous Cladribine (sc2CdA) is an alternative route with 100% bioavailability. In indolent non-Hodgkin lymphomas other than HCL, at the dose of 0.7 mg/kg/cycle, efficacy of sc2CdA is similar to iv2CdA and reduction to 0.5 mg/kg/cycle determines equivalent efficacy and lower toxicity. Aims: In a national multicentre clinical trial (protocol EudraCT code: ICGHCL 2004), we prospectively evaluated toxicity and efficacy of sc2CdA given 0.1mg/kg/die for 5 (total dose 0.5 mg/kg, arm A) or for 7 consecutive days (total dose 0.7 mg/kg, arm B) as a single course in classic HCL requiring first treatment. Method: Clinical data and diagnostic samples were collected from all patients for central revision of classical HCL diagnosis (WHO criteria) and for molecular analyses at the University of Siena, prior to study entry. Early grade 3–4 toxicity was assessed on days 7 to 30 after treatment, according to the 2003 NCI/CTCAE v3 criteria and occurrence of second tumors was assessed at every subsequent follow-up. Enpoints of sc2CdA efficacy were response to treatment, treatment free interval (TFI), relapse free survival (RFS), time to second tumor (TTST) and overall survival (OS). Responses to treatment were assessed on days 60 and 180 after treatment and defined according to the 1987 Consensus criteria. Complete Remissions (CR) and Partial Remissions (PR) were rated as beneficial responses (CR/PR), while minor Responses (mR) and No Responses (NR) were rated as treatment failures (mR/NR). TFI was measured as the time elapsed from sc2CdA initiation to second treatment because of new progression or failure to sc2CdA. RFS was measured in patients with a CR/PR from treatment inititation to relapse. OS was measured from sc2CdA initiation to death for any cause. Result: of 156 patients screened centrally, 148 scored as classical HCL and entered the study. Gender (male total 116/148, 78,4%; arm A: 62/77, 80,5% vs arm B: 54/71, 76,1%), age (total: median age 52 years, range 30–83; arm A: median 56, range 30–83; arm B: median 51, range 33–82), clinical and laboratory parameters prior to treatment (including spleen, hemoglobin, platelet, leucocyte and hairy cell counts) were equally balanced in the two arms. Dose reduction was required in no patients from arm A and in 2 patients from arm B due to concurrent infection. Overall hematological toxicity was no different among the two treatment arms. Requirement of platelet transfusions (total 7/148, 4,7%; arm A: 2/77, 2,6% vs arm B: 5/71, 7%), red blood cell transfusion (total 32/148; 21,6%, arm A: 15/77, 19,5% vs arm B: 17/71, 23,9%), or G-CSF (total 102/148; 68,9%, arm A: 55/77, 71,4% vs arm B: 47/71, 66,2%) was no different among the two arms. However, a significantly higher non hematological toxicity (total 28/148; 18,9%, arm A: 9/77, 11,7% vs arm B: 19/71, 26,8%) was observed in arm B (p=.019). Non haematological toxicity was mainly represented by infections or FUO (total 25/148; 16,9%) that were significantly more frequent in arm B (arm A: 8/77, 10,4% vs arm B: 17/71, 23,9%, p=.028). Higher prevalence of toxicity resulted in a higher hospitalization rate in arm B than in arm A (total 29/148; 19,6%, arm A: 9/77, 11,7% vs arm B: 20/71, 28,2%, p=.012) and one early death was experienced because of lung aspergillosis in arm B. Onehundred-forty patients (94,6%) had a beneficial response (101 CR, 68,2%; 39 PR, 26,4%) and 8 (5,4%) failed treatment (5 mR, 3,4%; 3 NR, 2%). Responses were equivalent in the two arms, with 72/77 (93,5%) beneficial responses (49/77 CR, 63,6%; 23/77 PR, 29,9%; 4/77 mR, 5,2% and 1/77 NR, 1,3%) in arm A versus 68/71 (95,8%) beneficial responses in arm B (52/71 CR, 73,2%; 16/71 PR, 22,5%; 1/71 mR, 1,4%; 2/71 NR, 2,8%). After a median follow-up of 36 months (range 12–66), 5 year TFI, RFS, TTST and OS were 67%, 71% 87% and 94%, respectively. Causes of late death were 2 cardiac events and 3 second tumors. No differences of TFI, RFS, TTST and OS were observed in the 2 arms of treatment. Conclusion: The present data indicate that overall activity of sc2CdA is similar to iv2CdA (Cheson, 1998). Furthermore, this study indicates that sc2CdA given at 25% reduced doses (0.5 mg/kg) has equivalent activity and significantly lower toxicity than sc2CdA at standard doses (0.7 mg/kg). The reduced infection rates and hospitalization rates of sc2CdA have important pharmaco-economic implications. Disclosures: No relevant conflicts of interest to declare.

Abstract 672 BACKGROUND: CLL is the most common adult-onset leukemia in the Western world. The most common known genetic lesion is the 13q14.3 deletion targeting MIR15/MIR16. We applied a very high resolution array to identify new genetic lesions in CLL. METHODS: 266 CLL samples were analyzed with Affymetrix Human Mapping 6.0 arrays, comprising over 1,8 million probes with a median distance of less than 1 Kb. Copy number was inferred using the circulary binary segmentation (CBS) algorithm. Minimal common regions (MCR) were defined using a modified version of the algorithm by Lenz et al. (PNAS 2008), specifically altered to identify very small genomic losses covered by only 2-9 probes and occurring in at least 5% of the cases (mMCRs). mMCRs having 100% overlap with known copy number variations were discarded. RESULTS: mMCRs occurred in 75 known genes. The most commonly affected genes were CDC73 (cell division 73; 63% of the cases, 3 probes), RREB1 (ras responsive element binding protein 1; 60%, 5 probes), JAK2 (47%, 8 probes), CCDC88A (AKT-phosphorylation enhancer,; 47%, 3 probes), AKT3 (43%, 4 probes). Other affected genes at a lower frequency were PIK3CA (26%), EGFR (25%), XRCC4 (18%), JAK1 (18%), PTPRK (15%), RB1 (14%), ERBB2 (10%), PDGFRA (8%), FHIT (7%). A functional analysis performed with DAVID 2008 (http://david.abcc.ncifcrf.gov/) identified the terms “anti-oncogene” and “tyrosine-protein kinase” and five KEGG (http://www.genome.jp/kegg/) pathways (“prostate cancer”, “non-small cell lung cancer”, “pancreatic cancer”, “endometrial” cancer”) as enriched among the 75 genes with a statistically significant p-value CONCLUSIONS: The application of high resolution arrays on a large series of CLL samples has shown frequent small interstitial deletions targeting a discrete number of genes, highly enriched for transcripts coding for kinases. A potential mechanism of action might be the loss of regulatory regions determining gene activation. Once validated, the current data would provide the basis to explore the rationale for the use of kinase inhibitors in the treatment of CLL. Disclosures: No relevant conflicts of interest to declare.

Abstract 4370 The cell of origin of chronic lymphocytic leukemia (CLL) has long been sought and immunoglobulin gene analysis provides new clues. The immunoglobulin heavy variable gene (IGHV) status has clinical relevance in CLL, where two subsets, delineated by the absence or presence of somatic mutation, have markedly different prognoses. The unmutated subset (U-CLL), of inferior prognosis, appears to derive from a pre-germinal center B cell. In U-CLL, there is strikingly increased usage of the 51p1-related alleles of the IGHV1-69 gene, often combined with selected IGHD genes and with IGHJ6. Shared sequence “stereotypic” characteristics of the HCDR3 result, and suggest antigen selection of the leukemic clones. In this study, we have analyzed 147 51p1/IGHJ6 rearrangements from 3 healthy individuals (>51yr) and sought sequence patterns parallel to those of U-CLL. A pre-established dataset of 313 51p1/IGHJ6 rearrangements from patients with U-CLL was used as a reference. A high proportion (49/147, 33.3%) of normal sequences revealed stereotypic patterns, several (22/147, 15%) being similar to those described in U-CLL. Additional CLL-associated stereotypes, not yet reported, were detected in 7/147 sequences (4.8%). Stereotypes (13.6%) not detected in CLL were also found in 20/147 (13.6%) 51p1/IGHJ6 combinations. The HCDR2-IGHJ6 sequences were almost exclusively unmutated (143/147, 97,3% sequences had ≥98% homology to germline). Junctional amino acids in normal B cells were heterogeneous, as in the cases of CLL with stereotyped 51p1/IGHJ6 B-cell receptors. Normal B cells expressing 51p1-derived IgM (4.8% of all B-cells) had a phenotype of naïve B-cells, similar to 51p1-negative (CD27-) B cells, i.e. IgM+ IgD+ CD23+ CD38+, with a small percentage of CD5+ B cells, not found in the memory B-cell subset. This snapshot of the naïve B-cell repertoire reveals subsets of B cells closely related to those characteristic of CLL. Conserved patterns in the 51p1-encoded IgM of normal B cells suggest a restricted sequence repertoire shaped by evolution to recognize common pathogens. Proliferative pressure on these cells is the likely route to U-CLL. Disclosures: No relevant conflicts of interest to declare.

Abstract 801 Intrinsic defects in the apoptotic circuitry underlie to a large extent the extended survival of malignant B cells in chronic lymphocytic leukemia (CLL) and are moreover believed to be responsible for their resistance to chemotherapy. We have recently demonstrated that p66Shc, a member of Shc family of protein adapters, acts as a promoter of apoptosis in T cells. Here we show that p66Shc uncouples the B-cell antigen receptor (BCR) from the Erk and Akt dependent survival pathways, thereby enhancing B-cell apoptosis. Expression of p66Shc was found to be profoundly and consistently impaired in CLL B cells compared to peripheral blood B cells form healthy donors. Moreover, significant differences in p66Shc expression were observed in patients with favorable or unfavorable prognosis, classified on the basis of the mutational status of the IGHV genes, with the lowest expression in the unfavorable prognosis group. Analysis of the expression of genes previously implicated in the apoptosis defects of CLL B cells revealed a selective alteration in the balance of pro- and anti-apoptotic members of the Bcl-2 family in these patients. Reconstitution experiments in CLL B cells, as well as data obtained on B cells from p66Shc-/- mice, showed that p66Shc expression correlates with a bias in the Bcl-2 family towards the pro-apoptotic members. Collectively, the data identify p66Shc as a novel regulator of B cell apoptosis which attenuates survival signals emanating from the BCR and modulates expression of the Bcl-2 family. They moreover provide evidence that the defect in p66Shc expression identified in CLL B cells may be causally related to the imbalance towards the anti-apoptotic Bcl-2 family members observed in these cells. Disclosures: No relevant conflicts of interest to declare.

Abstract 2339 Poster Board II-316 BACKGROUND: CLL, the most common adult-onset leukemia in the Western world, has a heterogeneous clinical course. Many advances have led to a better understanding of its pathogenesis and to improvements in treatment strategies, but striking solutions are still missing. We conducted a study to evaluate the impact of genomic aberrations on the clinical course. METHODS: From January 1980 to May 2008, 395 frozen samples of CLL patients, were prospectively collected in four centers. Extracted DNA was analyzed with Affymetrix Human Mapping 6.0 arrays. Normal matched DNA was analyzed for one fourth of the cases. Correlations between minimal common regions (MCR) and clinical parameters were evaluated with the Fisherôs-exact test and their impact on OS with the log-rank test. A p-value after Bonferroni multiple test correction (MTC) (p-adj.) RESULTS: Analysis of the clinical parameters (CPs) and known risk factors (Rai/Binet, age, doubling time, LDH, beta2, IGVH status, p53 mutations, telomere length, CD38, 11q, 17p) was consistent to previous published series. ZAP70 did not affect the clinical course, likely due inter-laboratories variability. After a median follow up of 53 months, 143/239 (60%) of the patients have started therapy and 63/261 (24%) died. 5-yr OS was 82%. Fisher test between the MCRs and CPs revealed an inverse relation between the presence of trisomy 12 by FISH and del13q14.3, an association between del17p and losses of 8p regions and between CD38 and 12q gain. Before MTC, 46 MCRs had a significant impact on OS and 67. After MTC, 3 regions maintained their role: 8p22 loss (38/248, 15%, p-adj.=0.002, median OS: 26 months vs. 48), 17p13.3-11.2 loss (20/248, 8%, p-adj.=0.001; median OS: 10 months vs. 48). In univariate analysis, the log-rank test among pts with 8p-/17p- (8/248, 3%), 8p- (30/248, 12%), 17p- (12/248, 5%), wild type (198/248, 80%) was statistically significant (p CONCLUSIONS: Loss of 8p22 designated a CLL subgroup with a worse outcome among all patients and in the subset with 17p loss. Our data suggested that this aberration might constitute an independent prognostic factor to be evaluated in independent studies. Results, including a Cox regression model, will be presented on all 395 cases. Disclosures: No relevant conflicts of interest to declare.

Chronic lymphocytic leukemia (CLL) with deletion of 17p(p53) locus (17p- CLL) identifies the most aggressive CLL subset, due to active disease course and resistance to current treatments. Intrinsic mechanisms of the poor prognosis of 17p- CLL are scarcely known. In order to identify additional lesions occurring in 17p- CLL which contribute to prognosis, we performed genome-wide DNA profiling in 18 cases of 17p- CLL (median follow-up 25 months, range 4–107), using high-resolution single-nucleotide polymorphism (SNP) arrays (Affymetrix Human Mapping Nsp 250 k arrays). Chromosomal 11q, 13q or 17p deletions and trisomy 12 were determined by FISH. Tumor IGHV usage and mutational status were determined by PCR/sequencing of the tumor gDNA/cDNA. Median interval from diagnosis to progression (11 months, p=0.000122) and treatment (16 months, p=0.000827) were representative of the bad behavior of our 17p- CLL cases if compared to 95 non17p- CLL patients from our internal cohort (68 and 69 months, respectively). Interestingly, 3/18 (16%) 17p- CLL expressed the IGHV3–74 segment which is rare in normal B-cells (1,4%, p=0.02) and in non17p- CLL (2.1%, p=0.02). Tumor IGHV genes were unmutated in 10/18 and mutated in 8/18 17p- CLL. All cases carried 17p13.1 deletion in more than 60% nuclei, as assessed by FISH. Genome-wide DNA profiling was 99% concordant with FISH (71/72 data-points; one 13q deletion detected by FISH in 20% of the nuclei was missed by the microarray) and confirmed 17p deletion in all cases. Deletions at the 13q14.3 locus recurred in 50% cases. Only one case had an extra-copy of chromosome 12. No cases carried ATM deletion. Other recurrent lesions identified by genome-wide DNA profiling were: gains of 2p16.1-pter (30%), 3q24-qter (20%), 8q23.3-qter (30%), 17q21.2-q22 (30%), 14q32 (30%); losses of 8p12-pter (30%), 9q21.33-q22.1 (20%), 14q32 (40%), 20p12.1-pter (20%). Approximately half of the cases had concomitant losses of 8p or 17p and gains of the 8q or 17q regions, respectively, suggesting the presence of isochromosomes 8 or 17. Indeed, isochromosome 17 was evident in one case analyzed by conventional cytogenetics. Six patients had one or more regions of loss of heterozygosity (LOH) longer than 5 Mb in the absence of copy number reduction, suggestive of uniparental disomy. Several lesions, such as those involving chromosomes 3, 8 and 9, associate with poor clinical behavior in other B-cell tumors. Thus, we investigated the prognostic significance of the lesions identified in our series. Despite the small number of cases, we found that 17p- CLLs with loss of 8p12-pter, which contains tumor necrosis factor and defensin family genes, associated with worse treatment free interval from diagnosis (12 vs 46 months in 8p- vs non 8p- CLL, p=0.04). Overall, our analysis identified new recurrent genomic abnormalities at high frequencies that may specifically contribute to a different biologic and clinical behavior within 17p- CLL. Among these, isochromosomes 8 and 17 were particularly common, and 8p loss might identify a new aggressive subentity within 17p- CLL.

The intravenous regimen of fludarabine plus cyclophosphamide (FC) at conventional doses is highly effective in Chronic lymphocytic leukemia (CLL), where it is currently indicated in first line therapy, and low-grade non Hodgkin lymphomas (LG-NHL). We have shown that intravenous or oral FC regimens at reduced doses remain highly effective in elderly patients with LG-NHL other than CLL. We tested tolerability and efficacy of oral FC at reduced doses in elderly patients with previously untreated (UT-CLL) or relapsed/refractory CLL (R-CLL). Twenty-five patients (>60 years) with UT-CLL (n=13) or R-CLL (n=12) were given oral F (25mg/m2/day, 40 mg total dose) and C (150mg/m2/day, 200 mg total dose) in an outpatient regimen for 4 consecutive days every 4 weeks for a maximum of 4 cycles. Patients were evaluated after every cycle for toxicity and hematological response. Toxicity was defined according to NCI criteria. Responses were defined as Complete (CR) when a normalization of blood count and misurable masses (lymph nodes and spleen) was documented, partial (PR) when at least a 50% reduction of lymphocytosis or nodes/splenomegaly occured, no response (NR) when disease was stable (less than PR) or progressed (>25% increment of tumor manifestations). Progression, new treatment or death was defined as “event”. Median age of the whole population was 70 years (range 61–80), 70 (61–80) in the UT-CLL, 71 (62–79) in the R-CLL. M:F distribution was 14:11 (7:6 in UT-CLL, 7:5 in R-CLL). Performance status was ≥1 in 80% patients. Of 25 patients, 16 (64%) were diagnosed CLL in stage A, 8 (32%) in stage B, 1 (4%) in stage C. Tumor IGHV gene was unmutated in 90% cases (in 85% UT-CLL, in 100% R-CLL). CD38 was positive in 68% (77% UT-CLL, 55% R-CLL) and ZAP-70 in 68% (82%, 67%). Chromosomal deletions of 11q or 17p regions were documented in 25.5% patients (27% UT-CLL, 25% R-CLL). The R-CLL had received 1–4 (median 2) treatment lines prior to FC. Median time from diagnosis or from prior treatment to FC was 24 months (range 1–77), 37 (1–77) in the UT-CLL, 12 (3–48) in the R-CLL. Overall, biological risk and clinical characteristics were indicative of an aggressive behavior of the investigated cohort. Patients received median 3 cycles (4 in UT-CLL, 3 in R-CLL). Nine-of-25 (36%) reduced the number of cycles because of fatigue (1 patient), heart failure (1), idiopathic thrombocytopenic purpura (1) in the UT-CLL (3/13) or infection (2), femur fracture (1), prolonged isolated thrombocytopenia or pancitopenia (2) and idiopathic thrombocytopenic purpura (1) in the R-CLL (6/12). Overall, 23/25 patients (92%) obtained a response (11/25 CR, 44%; 12/25 PR, 48%). Among UT-CLL, all responded to treatment (8/13 CR, 61.5%; 5/13 PR, 38.5%). Among R-CLL, 10/12 (83.5%) responded (3/12 CR, 25%; 7/12 PR, 58.5%; 2/12 NR, 16.5%). With a median follow-up of 23 months, 14/25 patients (56%) were event-free and 23/25 (92%) were alive. Among UT-CLL (median follow-up 12 months), 9/13 (69%) were event-free and all were alive. Among R-CLL (median follow-up 25 months), 5/12 (42%) were event-free and 10/12 (83%) were alive. Deaths occurred in the 2 R-CLL that had not responded to treatment, after 2 and 4 years, respectively. Despite the poor risk of the investigated population, this low-dose oral FC treatment showed good efficacy both in untreated and refractory/relapsed CLL. The treatment may be useful in elderly patients who cannot benefit of more aggressive or eradicating strategies and is easy to administer on an outpatient basis.

Abstract 760 Splenic marginal zone B-cell lymphoma (SMZL) is frequently associated with HCV infection and with autoimmune disorders. Subsets of “stereotyped” Complementarity-Determining Region 3 (HCDR3) sequences are a frequent feature of CLL patients. The aim of this study was to perform clustering analysis of HCDR3 aminoacid (AA) sequences in SMZL. We characterized IGHVDJ rearrangements from 133 SMZL patients diagnosed according to Matutes criteria (Leukemia, 2008) and performed clustering analysis of SMZL HCDR3 AA sequences. Sources for analysis were bone marrow and peripheral blood in 106 cases and spleen tissue + lymph node in 27 cases. SMZL HCDR3 sequences were also compared to 27,413 productive, non redundant HCDR3 AA sequences retrieved from public databases (EMBL, NCBI, IMGT/LIGM-DB; 25,655 sequences) and from our unpublished laboratory databases (1,758 sequences), including sequences from malignant B-cell clones (3,197 CLL, 86 SMZL, 1,131 other B-cell malignancies) and from non malignant B-cell repertoire (22,999 sequences). Alignments were performed using the multiple sequence alignment software ClustalX (2.0). Criteria for assigning HCDR3 AA sequences to stereotyped subsets included AA identity ≥60%, sharing of junction residues and/or usage of the same IGHD-J genes (Stamatopoulos et al, 2007). Male/female ratio of the 133 SMZL patients was 0.7:1; median age at diagnosis was 67 years (40-82 years). HCV serology was positive in 25/125 cases (20%). Median follow-up was 2.7 yrs (range 1-20 yrs); 5-yrs OS for the entire series was 80% and 5-yrs PFS was 39%. The most frequent IGHV families were IGHV3 (n=66, 50%), IGHV1 (n=40, 30%) and IGHV4 (n=23, 17%); the most frequent IGHV genes were IGHV1-2 (n=26, 20%), IGHV 3-23 (n=24, 18%), and IGHV 4-34 (n=10, 8%). Using the 98% identity cut-off value, 40/133 (30%) sequences were “unmutated” (15 with 100% identity, 10 with 99-99.9%, 15 with 98-98.9%). By means of clustering analysis, 16/133 SMZL cases (12%) met the minimal criteria to be included in subsets with “stereotyped” HCDR3. Additional 7/86 (8%) SMZL public sequences were included in subsets. We identified 3 major groups of HCDR3 subsets: 1) “SMZL-biased subsets”: 5 novel subsets were composed only by HCV-negative SMZL cases: 9 from this study cohort and 3 from public databases. Two subsets included both SMZL cases from this study series and SMZL cases from public databases, while 3 subsets were composed by 3 pairs of SMZL from the present study series; 2) “CLL subsets”: 4 SMZL cases (2 from our series) were assigned to 4 subsets previously reported in CLL (subsets 1, 6, 9, 25). In addition, 2 SMZL cases (1 from our series) formed 2 novel subsets with CLL cases; 3) “Non-Hodgkin's lymphoma subsets”: in 5 new subsets HCDR3 of 5 SMZL cases (4 from present series) were homologous to HCDR3 of non-splenic MZL and extranodal indolent lymphomas; in particular 2 HCV+ SMZL clustered with 1 HCV-related lymphoma and 1 HCV+ low grade B-cell lymphoma with splenomegaly, respectively. Distribution of mutated and unmutated cases did not differ between SMZL cases belonging or not to HCDR3 subsets. PFS was statistically shorter for cases belonging to “SMZL-biased subsets” (5-yrs PFS 14%) compared to the remaining cases of the series (5-yrs PFS 41%) (p=0.04). Mutational status was not predictive of OS and PFS. Multivariate analysis selected “SMZL-biased” clustering [hazard ratio (HR), 2.4; 95% CI: 1.003-5.684; p = 0.04], but not mutational status, as predictor of PFS. Sequencing of light chains and 3D modeling of cases belonging to subsets are ongoing. In conclusion, 1) “SMZL-biased” stereotyped HCDR3 sequences appear to be unique to HCV-negative SMZL and suggest a potential role for antigen(s) different from HCV in SMZL lymphomagenesis, and may determine a worse outcome in a subset of SZML patients; 2) only a small number of SMZL cases is related to previously reported CLL subsets; 3) SMZL also share stereotyped HCDR3 sequences with non-splenic MZL and extranodal indolent lymphomas, suggesting a role for a common antigen stimulation in the pathogenesis of these B-cell malignancies. Disclosures: No relevant conflicts of interest to declare.

Hairy cell leukemia (HCL) is a rare B-cell neoplasm with largely unravelled features of lymphomagenesis. Molecular mechanisms of disease are largely unknown and genomic aberrations have infrequently been investigated in HCL. In order to elucidate the genetic mechanisms of HCL pathogenesis, we analyzed copy number changes and loss of heterozigosity (LOH) by genome wide DNA profiling with the Affymetrix Human Mapping 250k Nsp arrays in 16 HCL. All investigated cases expressed multiple pre- (IgM+/−D) and post-switch (IgG+/−A) isotypes. Tumor IGHV gene transcript homology to germline ranged from 88.4 to 98.6%, with 4/16 HCL having >98% homology. Four of 16 patients were refractory or had progressive disease after first line purine-analogue-based treatments. CD103+ve hairy cells were >60% in all test samples. Genome wide DNA profiling demonstrated gross non recurrent copy number deletions in only 4/16 cases (25%), affecting 14q24-q32, 3q23, 3q27.3-q29, 8p12-pter 10q21-q23, 10q25-q26, 17p11-pter, 17q11-q21, 17q22–23, Xp21.2-p22.11, Xq13.3, Xq21.1-q21.33, Xq22.2–q28, the whole chromosomes 21 and X. Five of 16 patients (31%) presented LOH. Most interestingly, the 5 patients showed LOH regions, with a region of overlap in two patients at 6q25.1, without any copy number changes, indicative of uniparental disomy. Genes involved in bone marrow fibrosis, multiple isotype expression and response to treatment were specifically investigated. Interestingly, among genes critical for fibrosis, FGF12, a member of the fibroblast growth factors family, was specifically targeted by a complex chromosomal rearrangement at 3q27.3-q29, and reverse transcript-polymerase chain reaction confirmed FGF12 transcript expression. Among genes associated with poor response, p53 deletion was documented in a non-responder HCL. Conversely, no abnormalities at 14q32 locus (duplications or gains) were observed at IgH 14q32 region in these multiple IgH isotype expressing HCL. Overall, our analysis showed a highly stable genome with lack of gross recurrent abnormalities, which is an additional unique feature of HCL among B-cell neoplasms. However, our genomic analyses were able to identify specific chromosomal rearrangements targeting genes involved in fibrosis. Lack of gains of 14q32 locus was consistent with our hypothesis that multiple IgH isotypes are generated at RNA level in multiple IgH expressing HCL. The present analysis provides the basis for further molecular and epigenetic investigations.

Deletion of 17p (TP53) identifies a rare subset of chronic lymphocytic leukaemia (17p- CLL) with aggressive behaviour. Genome-wide DNA-profiling was performed to investigate 18 patients with 17p- CLL. All cases had multiple copy-number (CN) changes. Among the several recurrent CN changes identified, 8q24.13-q24.1-gain (MYC), 8p-loss (TNFRSF10A/B, also known as TRAIL1/2) and 2p16.1-p14-gain (REL/BCL11A) appeared frequently represented. 8p-loss and 2p16.1-p14-gain also appeared clinically relevant and predicted significant shorter time from diagnosis to treatment (8p-loss) and overall survival (8p-loss and 2p16.1-p14-gain, P < 0.05). These observations document a highly unstable genome in 17p- CLL and suggest that additional genes outside the TP53 locus may be important for tumour behaviour.

CLL harboring del17p13 displays a dismal prognosis and a high risk of chemorefractoriness. We tested whether TP53 inactivation through mutation harbors the same prognostic role as del17p13 on 308 consecutive CLL. TP53 mutation status was assessed at diagnosis by direct sequencing of TP53 exons 2–10. Thirty-two TP53 mutations were observed in 31/308 (10.0%) CLL. Among 26 missense mutations, the median residual transactivation activity compared to germline TP53 was 8.9% (25th–75th percentile: 0.6–13.0%), and was ≤20% in 22/26 (84.6%) mutations. All microdeletions and microinsertions (4/32; 12.5%) led to frameshift. One non-sense mutation introduced a stop codon at position 213, and one mutation affected the 3′ splicing site of exon 8. Despite the statistical association between TP53 mutations and del17p13 in the whole cohort, the overlap between the two genetic lesions was restricted to a fraction of patients. By combining results of FISH and TP53 mutation analysis, 44/297 (14.8%) CLL carried TP53 inactivation: 18/44 (40.9%) showed TP53 mutation paired to del17p13; 10/44 (22.7%) showed TP53 mutations in the absence of del17p13; and 16/44 (36.3%) showed del17p13 in the absence of TP53 mutations. Univariate log-rank analysis identified both TP53 mutations and del17p13 as risk factors for short overall survival (OS). Median OS for patients with TP53 mutations was 79.6 months (5-year OS 66.2%), whereas median OS for patients without TP53 mutations was not reached and the 5-year OS was 85.2% (p65 years (HR 4.98; p mutation in the absence of del17p13 is the sole mechanism of TP53 inactivation in 20% CLL harboring TP53 disruption; TP53 mutations carry the same prognostic relevance as del17p13 in terms of CLL progression, survival and risk of chemorefractoriness; because of the practical implications for choice of therapy, screening for TP53 mutations, in addition to del17p13 assessment, should be included in the initial prognostic assessment of CLL.

CLL subsets with stereotyped HCDR3 and IGH/IGK-L gene combinations have been reported among cases with both mutated (M) and unmutated (UM) IGHV gene status. Here we evaluated the characteristics of stereotyped combinations in 1428 IG sequences (848 M and 580 UM) from 1399 CLL patients ordered according to a HCDR3-driven clustering, performed aligning HCDR3 amino acid sequences (codons 107–117 IMGT) by the multiple sequence alignment software ClustalX (1.83). Clusters with stereotyped HCDR3 were selected for having a mean alignment score>60/100. We identified 73 different clusters with stereotyped HCDR3, 32/73 with >3 cases (range 3–32) for a total of 247 cases (17.4%), 81 with M and 166 with UM IGHV genes; 17/32 clusters (200 cases) were previous described, while 15 (47 cases) were novel. Seventeen/32 clusters (156 cases) were composed by UM cases (UM/M = 153/3), 14/32 clusters (59 cases) were of M cases (UM/M = 3/53), and a single cluster (32 IGHV3-21 cases) was a mixed one (UM/M=10/22). In our series, 64% of cases (158/247) utilized four specific IGHV genes, i.e. 1–69 (60/169 cases, all UM, split into 6 clusters), 3–21 (32/55 cases, all in a single cluster), 4–34 (19/129 cases, 17 M and 2 UM, split into 2 clusters), 4–39 (17/55 cases, 14 M and 3 UM, split into 3 clusters) or mixed IGHV genes of the IGHV1 family/clan (IGHV1-2/1-3/1-18/1-46/7-4), all in a single cluster (30/153 cases, all UM). Clinical data (time-to-treatment, TTT) were available for 637/1399 patients. UM cases (251) had significantly shorter TTT (p

Abstract 1256 Poster Board I-278 Introduction In the last years, the B cell receptor (BCR) has become a key molecule in chronic lymphocytic leukemia (CLL), given the correlation between mutational status of immunoglobulin heavy chain variable (IGHV) genes and disease prognosis. Recently, a fraction of CLL has been shown to preferentially express specific IGHV genes, often in a non-random combination with homologous heavy chain complementarity-determining region-3 (HCDR3) and peculiar light chains. Some of these stereotyped BCR mark CLL subsets with peculiar clinical behavior regardless of IGHV mutations. These data suggest a role for BCR in defining the clinical and biological features of CLL, also beyond the mutational status of IGHV genes. Patients and Methods A HCDR3-driven clustering of 1,426 IG sequences (1,398 patients) was performed using ClustalX(1.83). Time to treatment (TTT) intervals, Rai staging, IGHV mutational status, CD38, ZAP-70, and karyotype abnormalities evaluated by FISH were available for 617 patients. Gene expression profiling (GEP) and quantitative real-time PCR experiments (QRT-PCR) were performed on purified CLL cells. Results IGHV3-23 was totally absent in 71 identified stereotyped clusters despite being the second most frequently used IGHV gene, such distribution was significantly skewed (p Conclusion Expression of IGHV3-23 marks a subset of M CLL with a worse prognosis; such a peculiar clinical behavior may be related to superantigen stimulation combined with down-regulation of specific growth/tumor suppressor genes and up-regulation of miR-15a and miR-16-1. Disclosures No relevant conflicts of interest to declare.