ALBERT

Description: Abstract 4609 Active NOTCH signaling is observed in an increasing number of human neoplasias including chronic lymphocytic leukemia (CLL) and represents a potential therapeutic target. We have recently reported that the γ-secretase inhibitor DAPT may not be effective in all cases of CLL (Hubmann et al; BJH 2010). The aim of this study was to evaluate the therapeutic value of a newly identified NOTCH transactivation inhibitor (gliotoxin) and to compare its efficiency with the highly selective γ-secretase inhibitor (GSI) DAPT in CLL cells. Electrophoretic mobility shift assays (EMSA) showed that gliotoxin completely blocked the formation of DNA-bound NOTCH2 in CLL cells independent of their sensitivity to DAPT. The inhibition of NOTCH2 signaling by gliotoxin was associated with down regulation of its target gene CD23 and induction of apoptosis. The IC50 of Gliotoxin was variable between patients (n=20) and ranged between 50–100 nM irrespective to their sensitivity to GSI. Short term (4 hours) exposure of CLL cells to gliotoxin revealed that gliotoxin modulates the mRNA expression of several NOTCH-related genes including JAG1, PRKCD (PKC-δ), and NR4A1. We (Shehata et al; BLOOD 2010) and others have reported on the supportive effect of primary bone marrow stromal cells (BMSC) for CLL cells. Therefore, we tested whether gliotoxin may overcome this supportive effect. FACS analysis and MTT assays showed that gliotoxin abolished the supportive effect of BMSC under co-culture conditions. The effect of gliotoxin was dose dependent and selectively induced apoptosis in CLL cells accompanied by down regulation of NOTCH2 and CD23 transcription. Western blotting analysis demonstrated that gliotoxin also decreased the phosphorylation of Akt-pSer473 suggesting a link to PI3-K signaling. NOTCH1 has been recently shown to be affected by “gain of function” mutations in a subset of CLL patients. We observed that co-culture of CLL cells with BMSC was associated with increased NOTCH1 mRNA expression which could be decreased upon exposure to gliotoxin. In addition, gliotoxin inhibited DNA-bound NOTCH1 complexes in the T-ALL cell line SupT1 which is known to express high NOTCH1 activity. This indicates that gliotoxin may target both NOTCH1 and NOTCH2 isoforms. In conclusion, the data show that gliotoxin effectively targets NOTCH activity in CLL cells in a mechanism which is independent of γ-secretase. Thus, gliotoxin might have a beneficial effect in a wider range of CLL patients and warrants further evaluation. Disclosures: No relevant conflicts of interest to declare.

Description: The prognosis of fludarabine-refractory CLL is poor with a median survival time of 10 months. Intravenous Campath-1H (Alemtuzumab) was approved for fludarabine refractory CLL based on a remission rate of 33% and a median survival time of 16 months (Keating et al., Blood 2002). While the standard route of Campath-1H administration is by 2 hour intravenous infusion, the subcutaneous route was shown to be safe and efficacious in first line treatment (Lundin et al., Blood, 2002). The CLL2H trial of the GCLLSG was initiated to evaluate the subcutaneous application of 3 x 30 mg Campath-1H weekly for a maximum of 12 weeks in fludarabine refractory CLL after intravenous dose escalation (3, 10, 30 mg). The current interim analysis is based on the first 50 consecutive patients enrolled until April 2004. Median age was 63 (range 35–79) years, 70% were male, and a median number of 4 (range 1–7) prior lines of therapy had been given. Dose escalation was given in 46, subcutaneous continuation of therapy in 44, and treatment has not been started in 4 patients, respectively. Intravenous dose escalation was accompanied by no or mild (grade I-II) rigors and fever after premedication with paracetamol and antihistamines in the majority of patients, while grade III/IV infusion reactions were rare (n=3). Subcutaneous treatment was continued on an outpatient basis in all cases but had to be temporarily interrupted in 28 patients due to neutropenia (n=18), anemia (n=1), thrombocytopenia (n=3), infections (n=6; CMV reactivations n=3), and was stopped early in 22 cases due to progression (n=9), CMV reactivation (n=3), hematotoxicity (n=7), hemolysis (n=1), side effects (n=1), and infection (n=1). There were 6 cases of CMV reactivation, 5 of which promptly responded to oral valgancyclovir. During subcutaneous treatment toxicity was mostly grade I/II apart from hematotoxicity (grade III/IV anemia: 14%, thrombocytopenia: 34%, neutropenia: 66%) and grade III/IV infections (24%). After a median follow up time of 9.3 months 18 deaths have occurred (progression n=12, sepsis n=3, not CLL related n=2, before treatment start n=1). The overall response rate was 36% (CR 2%, PR 34%), the median progression free survival time was 9.7 months, and median overall survival time was 13.1 months. Analyses of genetic risk factors showed an unmutated VH status in 62% and high-risk genomic aberrations in the majority of patients (17p-: 29%, 11q-: 29%, +12q: 18%, 13q- single: 16%). Responses (CR or PR) were observed in 14 of 27 VH unmutated, 5 of 13 11q-, and 7 of 13 17p- cases. In conclusion, Campath-1H given via the subcutaneous route appears to be feasible in an outpatient setting in a high risk population of fludarabine-refractory CLL and appears to be of similar efficacy as compared to intravenous administration. Most importantly, genetic high risk subgroups with unmutated VH, 11q- or 17p- appear to respond to Campath-1H. An amendment has been activated including prophylactic pegfilgrastim and allowing subcutaneous dose-escalation. An update of this trial with ongoing recruitment and extended follow-up will be presented.

Description: Fludarabine-refractory CLL has a poor prognosis with a median overall survival time of less than 12 months despite salvage chemotherapy and intravenous alemtuzumab (Campath-1H) is the approved treatment based on a remission rate of 33% and a median survival time of 16 months (Keating et al., Blood 2002). The CLL2H trial of the GCLLSG was initiated to evaluate the subcutaneous application of 3 × 30 mg alemtuzumab weekly in fludarabine refractory CLL. The current analysis is based on 109 consecutive patients enrolled until completion of the trial in April 2006. Median age was 63 (36–81) years, 71% were male. A median number of 3 (1–9) prior lines of therapy had been given. Subcutaneous treatment was performed on an outpatient basis in all cases and had to be temporarily interrupted in 68 patients due to neutropenia (43%), anemia (6%), thrombocytopenia (3%), infections (40%, CMV reactivations 30%), and was stopped early in 63 cases due to insufficient response (44%), hematotoxicity (16%), infection (17%), and CMV reactivation (13%). The median alemtuzumab dose given was 722 (3–2203) mg. Toxicity was mostly grade I/II apart from hematotoxicity (grade III/IV anemia: 42%, thrombocytopenia: 52%, neutropenia: 54%) and grade III/IV infections (25%). After a median follow up time of 21.4 months, 56 deaths have occurred (due to progression 52%, infections 39%, not CLL related 9%). The overall response rate was 33% (CR 4%, PR 27%), the median progression free survival time was 7.7 months, and median overall survival time was 19.1 months. Genetic high-risk factors were present in the vast majority of cases (unmutated VH 66%, 17p–29%, 11q–19%, TP53 mutation 39%). Responses (CR or PR) were observed in 22% of VH unmutated, 24% of 11q-, 39% of 17p-, and 33% of TP53 mutated cases. Progression free survival and overall survival were not significantly different when comparing the genetic subgroups, particularly TP53 mutated, 11q-, and 17p- (see figure). In conclusion, subcutaneous alemtuzumab is feasible in an outpatient setting in a high-risk population of fludarabine-refractory CLL and appears to be of similar efficacy as by intravenous administration. Most importantly, genetic high risk subgroups with unmutated VH, 11q- or 17p- appear to respond to alemtuzumab. Figure Figure

Description: The mutation status and usage of specific VH genes such as V3-21 and V1-69 are potentially independent pathogenic and prognostic factors in chronic lymphocytic leukemia (CLL). To investigate the role of antigenic stimulation, we analyzed the expression of genes involved in B-cell receptor (BCR) signaling/activation, cell cycle, and apoptosis control in CLL using these specific VH genes compared to VH mutated (VH-MUT) and VH unmutated (VH-UM) CLL not using these VH genes. V3-21 cases showed characteristic expression differences compared to VH-MUT (up: ZAP70 [or ZAP-70]; down: CCND2, P27) and VH-UM (down: PI3K, CCND2, P27, CDK4, BAX) involving several BCR-related genes. Similarly, there was a marked difference between VH unmutated cases using the V1-69 gene and VH-UM (up: FOS; down: BLNK, SYK, CDK4, TP53). Therefore, usage of specific VH genes appears to have a strong influence on the gene expression pattern pointing to antigen recognition and ongoing BCR stimulation as a pathogenic factor in these CLL subgroups.

Description: Abstract 2826 Background: The addition of rituximab to CHOP-21 significantly improved clinical outcome in elderly patients with DLBCL (Coiffier et al., 2002). In young patients, the MInT trial, where young good-prognosis patients were randomized to receive a CHOP-like regimen or the same CHOP-like regimen plus rituximab, had to be stopped early because of superiority of the rituximab arm, and results were published with a median follow-up of 34 months (Pfreundschuh et al., Lancet Oncology 2006;379–91). Objective: Because the MInT study was the first study to show a survival benefit of the addition of rituximab to a CHOP-like regimen in young good-prognosis patients, it is important to analyze the effect of different chemotherapy regimens on long-term outcome. Methods: In a phase III intergroup study with participating cooperative groups from 18 countries, previously untreated young (18-60 years) patients with good-prognosis DLBCL (age-adjusted IPI 0 or 1, stages II-IV and stage I with bulky disease) were randomized to receive 6 cycles of a CHOP-like regimen (CHEMO) or the same chemotherapy plus rituximab 375 mg/m2, given on day 1 of each 3-weekly regimen and on days 1, 22, 43, 64, 85 and 106 of the 2-week regimens, respectively (R-CHEMO). Radiotherapy (30-40 Gy) was planned to sites of initial bulky disease and/or extranodal involvement. The primary endpoint was event-free survival (EFS) with events defined as failure to achieve complete remission, progressive disease, relapse, death or additional (unplanned) therapy. Results: Between 05/2000 and 10/2003 a total of 823 patients were recruited of whom 396 were allocated to receive CHOP-21, 361 to CHOEP-21, 34 to MACOP-B, and 32 to PMitCEBO with or without rituximab. Patients`characteristics were not different between the treatment arms with the exception that patients in the MACOP-B and R-MACOP-B arm had a more favorable prognostic profile. Toxicity, incidence of adverse events and severe adverse events in the different CHEMO and the R-CHEMO arms were not significantly different. After a median observation time of 70 (0.03-117) months, the 6-year EFS rates of patients assigned to CHEMO only were 50.4%, 60%, 41.7%, and 78.6% for CHOP-21, CHOEP-21, PMitCEBO, and MACOP-B, respectively. In an adjusted multivariate Cox regression model for EFS restricted to CHEMO patients, only the CHOEP-21 hazard ratio (HR) was significant (HR=0.73; p=0.05) compared to CHOP-21, while the hazard ratios of the different chemotherapies were not significantly different for PFS and OS (6 year PFS-rates: 60.2%, 64.8%, 67.5%, 86.2%; and 6 year OS-rates: 78.8%, 80.2%, 65.5%, and 100.0%, respectively). In patients assigned to R-CHEMO, 6-year EFS rates were 74.9%, 75%, 36.5%, and 86.7% for CHOP-21, CHOEP-21, PMitCEBO and MACOP-B, respectively. Likewise, PMitCEBO patients tended to have lower PFS (6-year rates: 79.1%, 81.9%, 58.9, and 86.7%, respectively) and OS (6-year rates: 91.9%, 89.4%, 80.0 and 94.1%, respectively). The poor outcome after R-PMitCEBO in contrast to R-CHOP was confirmed by multivariable Cox regression restricted to R-CHEMO and adjusting for aaIPI and bulky disease. Hazard ratio of PMitCEBO was significant for EFS (HR 4.35, p

Description: The PAX-5 gene codes for the transcription factor BSAP, which is expressed throughout B-cell development. Although loss-of-function mutation in the mouse showed an essential role forPax-5 in early B lymphopoiesis, gain-of-function mutations have implicated the human PAX-5 gene in the control of late B-cell differentiation. PAX-5 (on 9p13) has been involved together with the immunoglobulin heavy-chain (IgH) gene (on 14q32) in the recurring t(9;14)(p13;q32) translocation that is characteristic of small lymphocytic lymphoma with plasmacytoid differentiation. Here we have characterized a complex t(2;9;14)(p12;p13;q32) translocation present in a closely related non-Hodgkin’s lymphoma referred to as splenic marginal zone lymphoma (MZL). In this MZL-1 translocation, the two promoters of PAX-5 were replaced on the derivative chromosome 14 by an immunoglobulin switch Sμ promoter that was linked to the structural PAX-5 gene upstream of its translation initiation codon in exon 1B. Expression analyses confirmed thatPAX-5 transcription was upregulated due to efficient initiation at the Sμ promoter in the malignant B lymphocytes of patient MZL-1. For comparison we have analyzed PAX-5 expression in another B-cell lymphoma, KIS-1, indicating that transcription from the distalPAX-5 promoter was increased in this tumor in agreement with the previously characterized translocation of the immunoglobulin Eμ enhancer adjacent to PAX-5 exon 1A. In both lymphomas, the J-chain gene, which is thought to be under negative control by BSAP, was not expressed, whereas transcription of the putative target genep53 was unaffected by PAX-5 overexpression. Together these data indicate that the t(9;14)(p13;q32) translocation contributes to lymphoma formation as a regulatory mutation that leads to increasedPAX-5 expression in late B-cell differentiation due to promoter replacement or enhancer insertion.

Description: Background: The course of chronic lymphocytic leukemia (CLL) is highly variable. Therefore, there is a need for prognostic factors that are readily performed and have a high predictive power. Methods: The occurrence of translocations, a recently identified prognostic factor in CLL (Blood2006;107:742–751), was studied in 148 previously untreated, mostly early-stage patients and compared with respect to treatment-free survival (TFS) to several prognostic factors (Binet stage, mutational status of immunoglobulin genes, CD38, thymidine kinase serum concentration and cytogenetic aberrations detected by interphase FISH). To investigate chromosomal translocations, we applied a new method, CpG oligodeoxynucleotide stimulation that allows efficient preparation of metaphase spreads from CLL cells. Results: The occurrence of translocations classified the majority of patients with poor prognosis. If translocations were investigated in addition to the currently used prognostic factors they identified those patients who were classified to be in a low-risk group based on traditionally used criteria, who had in fact a high risk for progression. Vice versa, patients in the high-risk groups for progression who did not have translocations had a long TFS. There was a substantial overlap of patients who had translocations and additional risk factors. But when we omitted patients who had translocations in addition to a given risk factor, we found that the respective risk factor lost its prognostic significance for the remaining patients. The two factors that retained their prognostic power in these patients were translocations and the Binet stage. This could suggest that the prognostic significance of the currently used factors derives from their frequent co-occurrence with translocations. Finally, multivariate analyses demonstrated that Binet stage (p=0.02) and translocations (p=0.0005) are the factors with the highest impact on TFS in our study cohort. Conclusion: We present a method for efficient preparation of metaphase spreads in CLL cells in order to investigate chromosomal translocations. The occurrence of translocations is an independent prognostic marker in CLL. Finally, translocations occur not as a late event in the course of the disease and may define a new biological subgroup in this disease entity.

Description: Abstract 3512 Introduction T-cell large granular lymphocytic leukemia (T-LGL) is a rare lymphoproliferative disease characterized by an expansion of large granular lymphocytes involving blood, bone marrow, spleen and liver. T-LGL cells are mature CD3, CD8 and T-cell receptor (TCR) αβ positive cells exhibiting the immunophenotype of activated cytotoxic T lymphocytes (CTLs). CD4 or TCR γδ positive variants occur rarely. T-LGL affects adults at a median age of 55–60 years and arises commonly in patients with a preexisting autoimmune disorder. Many patients remain asymptomatic for years and do not require treatment. Palliative therapy with immunosuppressant agents such as low dose methotrexate, ciclosporin and fludarabine is used for the correction of severe immune-mediated cytopenias, which often complicates the course of the disease. The molecular pathogenesis of T-LGL remains unclear. No recurrent karyotypic anomalities but several numeric and structural chromosomal alterations have been identified. Recently, activating somatic mutations in the signal transducer and activator of transcription 3 gene (STAT3) have been described by Koskela et al. in approximately 40% of T-LGL patients. STAT3 mutations lead to an increased transcriptional activity and were more prevalent in patients with neutropenia and rheumatoid arthritis than in patients without these conditions. As these findings only explain part of the pathogenesis in the fraction of patients affected by STAT3 mutations, we here aimed to identify novel mutations which may help to better understand the mechanisms of disease development. Methods We sorted tumor- and non-tumor cells from peripheral blood samples of T-LGL patients by using fluorescence activated cell sorting (FACSDiva®, Becton Dickinson) to perform single nucleotide polymorphism (SNP) chip analysis and next-generation RNA sequencing. SNP chips were analyzed in 10 patients (Affymetrix, Mapping 250K Sty Array®). To identify somatic mutations in patients with T-LGL, we compared CD8/CD57 positive tumor cells with non-tumor cells as germline control. Sample libraries for RNA sequencing of 5 patients were generated with NuGEN Encore®, sequencing was performed on Illumina HiSeq 2000® yielding 100 million 100 basepair single reads, and alignment was realized on TopHat2 against hg19 as reference genome. For quantification and analysis of variants Partek GS 6.6 was used. Results High resolution copy number determination employing SNP chips in 10 patients revealed both gains and losses on different chromosomes, among others 1q, 7q, 14q and chromosome X. The affected chromosomal regions included genes with potential relevance to the disease process such as WNT and RASSF gene family members in deleted regions and PIM3 and MAPK family members in gained regions. However, in line with previous reports no recurrent chromosomal aberrations were detected. Preliminary analysis of RNA sequencing data revealed activating STAT3 Y640F mutations in 2 out of 5 patients tested (40%). Interestingly, one of the STAT3 mutated T-LGL clones also exhibited an inactivating mutation of the NFKB inhibitory gene TNFAIP3 (A20), which has been reported to play an important role in the molecular pathogenesis of different B cell lymphomas but has as yet not been described in T-LGL. Detailed analysis of sequencing data is currently ongoing and further results will be presented at the conference. In conclusion, combined RNA sequencing and molecular cytogenetic profiling identified novel specific chromosomal loci and genes that could help to better understand the molecular pathogenesis of T-LGL and develop novel targeted treatment modalities for this disease. Disclosures: No relevant conflicts of interest to declare.

Description: The 110-kD lung resistance protein (LRP) is overexpressed in P-glycoprotein–negative multidrug-resistant cell lines and most likely involved in the multidrug resistance (MDR) of these cell lines. To determine the clinical significance of LRP, we have studied LRP expression of leukemic blasts and its association with clinical outcome in patients with de novo acute myeloid leukemia (AML). LRP expression of leukemic blasts obtained from peripheral blood or bone marrow of previously untreated patients (n = 86) was determined by immunocytochemistry by means of monoclonal antibody LRP-56. LRP expression at diagnosis was detected in 31 (36%) patients. LRP expression was independent of age and sex of the patients, French-American-British subtype, cytogenetic abnormalities, and lactate dehydrogenase levels, but correlated with white blood cell count (P = .01). Eighty-two patients received standard induction chemotherapy that included cytarabine and MDR drugs (daunorubicin in most patients, additional etoposide in the majority of patients). The complete remission rate of induction chemotherapy was 72% (95% confidence interval [CI] = 61% to 82%) for the total study population. The complete remission rate was 81% (95% CI = 67% to 91%) for patients without LRP expression but only 55% (95% CI = 36% to 74%) for patients with LRP expression (P = .01). Overall survival and disease-free survival were estimated according to Kaplan-Meier in 82 and 59 patients, respectively. Overall survival was significantly longer in patients without LRP expression than in patients with LRP expression. At a median follow-up of 16 months, median overall survival was 17 months (95% CI = 12 to 38 months) for LRP-negative patients but only 8 months (95% CI = 4 to 12 months) for -positive patients (P = .006). Disease-free survival was 9 months (95% CI = 7 to 11 months) for LRP-negative patients and 6 months (95% CI = 5 to 8 months) for -positive patients (P = .078). Outcome was best in patients lacking both LRP and P-glycoprotein expression. In conclusion, LRP predicts for poor outcome and thus theLRP gene appears to be another clinically relevant drug resistance gene in AML.

Description: Abstract 1267 Poster Board I-289 The international multicenter randomized CLL8 trial evaluated 1st line treatment with FC or FCR in 817 CLL patients. Analysis of TP53 mutation status by a re-sequencing chip (Amplichip, Roche Molecular Systems) and confirmatory direct DNA sequencing were performed in a central reference laboratory. Samples were available for 628 (76.9%) patients and this cohort was representative of the full trial population regarding other baseline prognostic factors and demographics. Outcome was analyzed for subgroups defined by genetic parameters in univariate and multivariate analyses. The incidence of the TP53 mutations was 11.9% (71/628; 41 in FC arm, 30 in FCR arm). Forty-two of 51 patients (82.4%) with 17p deletion had a TP53 mutation. 5% of patients without 17p deletion (28/553) had a TP53 mutation. Patients with TP53 mutation showed lower complete response (CR) and overall response (OR) rates as compared to the group without TP53 mutation (6.9 vs. 36.4% and 62.1% vs. 95.3% (p