Chronic Lymphocytic Leukemia Cohort Research Articles

A Molecular Model to Predict Durable Remission after First Line Fludarabine-Cyclophosphamide-Rituximab Treatment in Chronic Lymphocytic Leukemia

Background. Fludarabine, cyclophosphamide, and rituximab (FCR) has represented a significant advancement in the treatment of patients with chronic lymphocytic leukemia (CLL) and has the potential of inducing durable remissions of the disease. In the new scenario of targeted agents for CLL, there is an increasing interest in identifying patients who may gain the maximum benefit in terms of disease control by a single shot of FCR chemo-immunotherapy.Purpose. Here we aimed at identifying predictors of durable remission after first line FCR treatment.Methods. The study was based on 405 progressive and previously untreated CLL patients who consecutively received standard FCR (up to 6 cycles) as first line therapy in 19 hematologic centers between 2001 and 2010. According to an intention to treat approach, all patients who received at least one FCR cycle were registered in the study. This series was representative of a FCR treated CLL cohort with respect to baseline characteristics, including age (median: 61 years; >65 years in 33% of patients), gender (male in 68% of patients), stage (progressive Binet A in 11% of patients; Binet B in 59%; Binet C in 30%) and number of FCR courses (median: 6; <6 in 42% of patients). Most patients were evaluable for IGHV mutation status (unmutated in 192/296, 65%) and genomic aberrations at treatment requirement (17p deletion in 30/306, 9.8%; 11q deletion in 56/300, 19%; +12 in 70/298, 23%; 13q deletion in 108/301, 36%).Results. After a median follow-up of five years, 159 patients have progressed and 72 have died, accounting for a 5-year progression free survival (PFS) according to the IWCLL criteria of 47% (median: 58 months) and for a 5-year overall survival (OS) of 81% (median: not reached). When the demographic effects of age, gender and year of treatment were compensated, the 5-year and 10-year survival of the whole CLL cohort were 85% and 68%, respectively, of those expected in the matched normal general population (p<0.001). By multivariate analysis, unmutated IGHV genes, 11q deletion and 17p deletion maintained independent association with both PFS and OS, thus providing the rationale to utilize them in the development of a model to predict remission duration after FCR. By recursive partitioning analysis, 17p deletion was the most important variable in predicting PFS after FCR, followed by 11q deletion and IGHV mutation status (Fig. 1A). Based on the application of an amalgamation algorithm, cases harboring unmutated IGHV genes and 11q deletion were grouped into a single category because they showed an identical drop of the PFS curve (Fig. 1A). By this approach, three CLL subgroups were hierarchically classified. Clinical features and treatment indication were superimposable across the three risk groups. The low risk category comprised patients harboring mutated IGHV genes but neither 11q or 17p deletion, and accounted for 26% of all cases. Most of the low risk patients (67%) remained free of progression after FCR (Fig. 1B), and their hazard of relapse dropped to zero after 5 years of follow-up. Consistently, the PFS curve of low risk patients plateaued after 5-years from FCR (Fig 1B). The life expectancy of low risk patients (91% at 5 year) was superimposable to that observed in the matched normal general population (Fig. 1D), indicating that neither the disease, nor complications of its treatment affected survival in this favorable group of CLL. Conversely, patients belonging to the intermediate risk (IGHV unmutated and/or 11q deleted) and high risk (17p deleted) categories showed a constant increase of the hazard of progression over time and almost all were projected to relapse after FCR, although at a different rate: 10% per year of follow-up in the intermediate risk group and 17% per year of follow-up in the high risk group (Fig. 1B-C). The 5-year life expectancy of intermediate and high-risk patients was significantly impaired compared to that observed in the matched general population (85% and 60%, respectively) (Fig. 1D), indicating an excess of deaths related to the disease or treatment complications in these unfavorable groups of CLL.Conclusions. The combination of three biomarkers that are routinely tested at treatment allows to segregate a subgroup of CLL (IGHV mutated without 17p or 11q deletion) that may achieve a durable remission after first line FCR treatment and experience an expected survival similar to that of the general population. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Toxicity Is Not Associated with Age or Comorbidity Score in a Randomised Study of Oral Fludarabine and Cyclophosphamide and IV Rituximab (FCR) As First-Line Therapy of Fit, Elderly Patients with Chronic Lymphocytic Leukemia (CLL)

BACKGROUND Fludarabine (F), cyclophosphamide (C) and rituximab (R) gave superior progression free (PFS) and overall survival (OS) versus (vs) FC in the CLL8 Study. The median age in CLL8 was 61 years compared to 72 years for CLL overall. There has been considerable debate regarding the tolerability and toxicity of FCR based therapy in older patients (pts) and those with comorbidities.METHODS Previously untreated fit pts with progressive CLL aged ≥65 were randomised to one of 3 therapy arms: (i) FR5: F 24mg/m2 po D1-5 + R iv D1 (375mg/m2 cycle 1, 500mg/m2 cycles 2-6), (ii) FCR3: F 24mg/m2 po and C 150mg/m2 po D1-3 + R iv D1 or (iii) FCR5: F 24mg/m2 po+ C 150mg/m2 po D1-5 + R iv D1 all at 4 weekly intervals for an intended 6 cycles. Cycles could be delayed up to 2 weeks for grade 3+ toxicity, and if unresolved by 2 weeks, pts were taken off study. Fitness was assessed using the Cumulative Illness Rating Scale (CIRS) score with eligibility restricted to CIRS ≤6.RESULTS Recruitment of 120 pts was completed in July 2012. 117 fulfilled eligibility and 1 had no treatment or follow-up reducing the cohort to 116. Median age was 71 (range 65-82) years (yrs) with 78 males (67%) and 39 females (33%). Patient registration by age was 65-69 yrs – 44, 70-74 yrs – 44, 75-79 yrs – 20 and 80-84 yrs – 8 pts, and by CIRS score was 0-2 – 61, 3-4 – 38, 5-6 – 17 pts. Toxicity data by age and CIRS score are shown in tables 1 and 2 respectively. All 6 protocol cycles were completed in 69% but less on FCR5 44% vs FR5 89% and FCR3 76% (p<0.001). Selected toxicity and early stopping rates in table 3. Overall response and toxicity are presented separately. Table 1:Grade 3+ Adverse Events by AgeAgeTotal65-6970-7475-7980-84(n=116)(N=44)(N=44)(N=20)(N=8)Hematological28 (64%)29 (66%)10 (50%)3 (38%)70 (60%)Neutropenia24 (55%)21 (48%)10 (50%)3 (38%)58 (50%)Thrombocytopenia9 (20%)9 (20%)1 (5%)0 (0%)19 (16%)Anaemia7 (16%)6 (14%)2 (10%)0 (0%)15 (13%)Haemolytic Anaemia1 (2%)2 (5%)1 (5%)1 (12%)5 (4%)Febrile Neutropenia / Infection9 (20%)5 (11%)8 (40%)2 (25%)24 (21%)Skin/Allergy/Fatigue/hypersensitivity3 (7%)2 (5%)4 (20%)2 (25%)11 (9%)Other (Card / resp / neuro / metabolic)13 (30%)10 (23%)3 (15%)1 (12%)27 (23%)At least 1 grade 3+ AE37 (84%)34 (77%)13 (65%)6 (75%)90 (78%)There was no statistically significant difference by age (p=0.429). Table 2:Grade 3+ Adverse Events by CIRS scoreCIRS scoreTotal0-23-45-6(n=116)(N=61)(N=38)(N=17)Hematological38 (62%)21 (55%)11 (65%)70 (60%)Neutropenia31 (51%)16 (42%)11 (65%)58 (50%)Thrombocytopenia11 (18%)6 (16%)2 (12%)19 (16%)Anaemia7 (11%)5 (13%)3 (18%)15 (13%)Haemolytic Anaemia1 (2%)4 (11%)0 (0%)5 (4%)Febrile Neutropenia / Infection11 (18%)9 (24%)4 (24%)24 (21%)Skin/Allergy/Fatigue/hypersensitivity6 (10%)4 (11%)1 (6%)11 (9%)Other (Card / resp / neuro / metabolic)12 (20%)9 (24%)6 (35%)27 (23%)At least 1 grade 3+ AE78 (79%)27 (71%)15 (88%)90 (78%)There was no statistically significant difference by CIRS score (p=0.355). Table 3:Abbreviated Grade 3+ Adverse Events by Treatment ArmTreatment armTotal (n=116)FR5 (N=37)FCR3 (N=41)FCR5 (N=38)Hematological15 (41% )26 (63% )29 (76% )70 (60%)At least 1 grade 3+ AE21 (57%)34 (83%)35 (92%)90 (78%)Early cessation due to toxicity2 (5.6%)1 (2.4%)13 (34%)16 (14%)Early cessation due to toxicity was significantly more common with FCR5 (p<0.001). However, of the 13 patients on FCR5 arm that stopped early due to toxicity, there was no difference by age or CIRS score: by age, 5 were 65-69 yrs, 3 were 70-74 yrs, 4 were 75-79 yrs and 1 was 80-84 yrs and by CIRS score, 3 were 0-2, 6 were 3-4 and 4 were 5-6.CONCLUSIONS Final analysis in this randomised dose de-escalation study shows oral FCR therapy is generally safe and well tolerated in CLL pts aged ≥65 years requiring first-line treatment, when early stopping is utilised if prolonged toxicity occurs. Early cessation due to toxicity was more common with full dose FCR5, but not associated with age or CIRS score within this arm. Overall in this relatively fit elderly CLL cohort, neither age nor CIRS score were associated with toxicity, or early cessation of therapy due to toxicity. The results highlight the difficulty of predicting toxicity based on age and comorbidity in elderly CLL pts. DisclosuresMulligan:Roche: Consultancy, Honoraria, Research Funding, Speakers Bureau; Sanofi Aventis: Research Funding; Janssen: Consultancy, Honoraria, Speakers Bureau; Celgene: Consultancy, Honoraria. Gill:Roche: Research Funding; Sanofi Aventis: Research Funding. Turner:Roche: Research Funding; Sanofi Aventis: Research Funding. Renwick:Roche: Research Funding; Sanofi: Research Funding. Latimer:Roche: Research Funding; Sanofi: Research Funding. Mackinlay:Roche: Research Funding; Sanofi: Research Funding. Berkahn:Roche: Research Funding; Sanofi: Research Funding. Simpson:Roche: Research Funding; Sanofi: Research Funding. Forsyth:Roche: Research Funding; Sanofi: Research Funding. Harrup:Roche: Research Funding; Sanofi: Research Funding. Kuss:Roche: Research Funding; Sanofi: Research Funding.

The Transcription Factor IRF4 Is Crucial for CLL Progression and Regulates Survival and Proliferation in a Microenvironment Related Manner

Introduction: Chronic Lymphocytic Leukemia (CLL) is characterized by an accumulation of CD19+ CD5+ CLL cells in the peripheral blood and the lymphoid compartments. The disease is still incurable despite recent advances in the development of novel therapy concepts (e.g. inhibitors of BCR signaling). Although it is not entirely clear how CLL develops, proliferation centers located in the lymph node and the bone marrow are thought to be the crucial niche that provides the surrounding in which CLL cells originate and re-establish the clone after therapy, leading to relapse in almost 100% of CLL patients. In analogous normal lymphoproliferative environments the transcription factor IRF4 plays an important role in controlling B cell maturation, differentiation, proliferation and survival. A SNP in the human IRF4 3’UTR was linked to CLL susceptibility (DiBernardo Nat.Gen. 2008) and in the mouse model germline deletion of IRF4 in all lineages contributed to CLL development in the New Zealand Black and in the VH11 mouse model (Shukla Blood 2013, Ma J.Biol.Chem. 2013).Methods: CLL samples were collected from the peripheral blood of CLL patients according to the Declaration of Helsinki and with written informed consent. IRF4 mRNA expression analyses were performed in purified CD19+ CLL cells using TaqMan Real Time PCR. In vitro studies were performed in soloculture without stimulation or in coculture systems mimicking the lymph node microenvironment, as previously described (Asslaber Br.J.Haematol. 2013). IRF4 knockdown was performed using lipofection and IRF4 specific siRNAs or non-targeting control siRNAs. IRF4 protein expression was measured by intracellular flow cytometry. Proliferation was detected by Ki-67 intracellular staining. All intracellular stainings were performed after gating on viable CD19+ CD5+ CLL cells. Cell death was assessed by Annexin V / 7AAD staining or using a fixable viability dye.Results: Overall, IRF4 mRNA expression was decreased in CLL patients (N=101) when compared to healthy donors (N=8, P=0.04). In the CLL cohort low IRF4 mRNA expression (cut-off determined by ROC analysis and Youden Index calculation) was significantly correlated with earlier time to treatment (median treatment-free survival: 40 month in the IRF4LOW and 69 month in the IRF4HIGH CLL patient group (P<0.01)). This was proved to be independent of mutation state and other CLL risk factors in multivariate analysis. Using in vitro knockdown of IRF4 in solo- and coculture models we found that lowered expression of IRF4 was associated with increased CLL cell survival and abrogated spontaneous apoptosis (N=11, P<0.01). Moreover, we detected a failure of CLL cells to respond to CD40L signals derived from the microenvironment after IRF4 knockdown, which was associated with abrogated proliferation (N=9, P<0.01). Indeed microenvironmental stimuli (e.g. CD40L ligation or addition of activated T cells) were required to maintain IRF4 expression in vitro (N=17, P < 0.001) which also corresponds to our observation that IRF4 positive CLL cells were highly overrepresented in the CXCR4dim CD5brightCLL population (N = 55, P < 0.0001) which is considered to be the proliferative CLL cell fraction emigrated from the lymph node (Calissano Mol. Med. 2011).Conclusion: We demonstrate that low IRF4 expression is an independent negative prognostic marker for CLL progression. In the microenvironmental context IRF4 plays a dual role, with high expression limiting CLL survival on the one hand but being a crucial mediator of CD40L mediated proliferation on the other hand. While our data suggest that IRF4 is upregulated in response to microenvironmental stimuli the association of low IRF4 levels with bad prognosis suggests that IRF4 levels may serve to limit CLL growth by enhancing cell death responses. The exact mechanisms by which IRF4 impacts CLL pathophysiology is currently being investigated in vivo using B-cell specific IRF4-deficient CLL mouse models. DisclosuresNo relevant conflicts of interest to declare.

Demographics By Age Group (AG) and Line of Therapy (LOT) in Chronic Lymphocytic Leukemia (CLL) Patients (Pts) Treated in US Practices from the Connect® CLL Registry

Introduction: CLL pts who qualify for participation in clinical trials are often thought of as a highly selected population. Therefore whether safety and efficacy data from these trials can be generalized to all CLL pts treated in the community setting is unknown. Prospective, observational studies can provide significant insight into pt and disease characteristics as well as practice patterns in different clinical settings. Here we report on baseline demographics and clinical characteristics for 1495 pts who participated in the Connect CLL observational study and were followed prospectively. Methods: Connect®CLL is a non study-specific intervention treatment registry which collects data on the management and outcomes of CLL pts in clinical, academic and community practices in the US. Clinical information was prospectively collected for pts treated in community (n=1312), academic (n=155), or government (n=28) centers in the US between March 2010 – Jan 2014. Pts eligible for treatment or on treatment not more than 2 months were eligible for enrollment. Here we report on the baseline demographics and clinical characteristics of a large CLL cohort stratified by AG, LOT and type of treatment center. Results: In total,1,495 pts were accrued with a median age at enrollment of 69 (range 22–99). Pts were Caucasian (92.2%), African American (6.9%), and male (63.7%). Median duration from diagnosis until initial therapy was 3.7 years (range 0-32). Age of enrolled pts was as follows; 527 pts age = 70 mL/min, ECOG 0-1; 18% of patients on LOT1 and 12% of pts on LOT=2 would satisfy these criteria. They represent 36% of eligible pts from academic treating institutions and 13% from community/government centers. Only 20% of patients on LOT1 would satisfy enrollment criteria for CLL8 (FCR vs. FC; Hallek; Lancet 2010) CLL 10 (BR vs FCR; [NCT00769522][1]). which did not have enrollment restrictions for age. Conclusions : Baseline characteristics of pts with CLL vary by age, LOT, and type of treatment center. In this large prospective observational study, our data confirms that older pts typically have a worse ECOG PS, more medical comorbidities, more advanced disease when initiating therapy and a lower probability of being treated at an academic medical center. Based on inclusion criteria used for entry in either CLL10 or CLL8, fewer than 20% of pts satisfy age, renal clearance, and fitness profile to be considered suitable for aggressive chemoimmunotherapy in the front line setting. This observation supports the notion that current clinical trial data might not always be applicable to the general CLL population seen in the community and argues for designing clinical trials that are more generalizable to the broader pt population. Disclosures Sharman: Gilead: Research Funding; Calistoga: Honoraria; Phamacyclics: Honoraria, Research Funding; Celgene: Consultancy, Research Funding; TG therapeutics: Research Funding; Roche: Research Funding. Mato: Millennium: Speakers Bureau; Celgene: Speakers Bureau; Genentech: Speakers Bureau. Kay: Celgene: Research Funding. Kipps: Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Lamanna: Genentech-Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Infinity: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Weiss: Celgene: Consultancy; Phamacyclics: Consultancy; Gilead: Consultancy. Nabhan: Celgene: Honoraria, Research Funding. Flinn: Celgene: Research Funding. Grinblatt: Celgene: Honoraria, Speakers Bureau. Kozloff: Celgene: Consultancy. Farber: Alexion: Equity Ownership, Speakers Bureau; Celgene: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Gilead: Speakers Bureau; Janssen/Pharmacyclics: Speakers Bureau; Seattle Genetics: Speakers Bureau; Genentech: Membership on an entity's Board of Directors or advisory committees. Sullivan: Celgene: Employment, Equity Ownership. Street: Celgene: Employment, Equity Ownership. Swern: Celgene: Employment, Equity Ownership. Flowers: TG Therapeutics: Research Funding; Phamacyclics: Research Funding; Acerta: Research Funding; Abbvie: Research Funding; Infinity: Research Funding; Lymphoma Research Foundation: Membership on an entity's Board of Directors or advisory committees; Janssen: Research Funding; Spectrum: Research Funding; Gilead: Research Funding; Genentech: Unpaid consultancy Other; Millennium: Research Funding, Unpaid consultancy, Unpaid consultancy Other; Seattle Genetics: Consultancy; Prescription Solutions: Consultancy; Celgene: Unpaid consultancy Other. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT00769522&atom=%2Fbloodjournal%2F124%2F21%2F3338.atom

Prognostic Factors and Outcomes in Allogeneic Hematopoietic Stem Cell Transplant Vs. Non-Transplant Chronic Lymphocytic Leukemia (CLL) Patients: A Comparative Analysis with the Leukemia/BMT Program of British Columbia (BC) and the BC Provincial CLL Database

Background: Chronic Lymphocytic Leukemia (CLL) patients (pts) have significant (sig) heterogeneity; survival ranges from decades to <5 years (yrs). Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is promising treatment (tx) for high-risk pts. Ideally, predictive (pred) tools would allow clinicians to recognize such pts early, permitting transplant performance to maximize benefit and minimize procedure associated risk. Factors with significant (sig) pred capacity are not, however, entirely clarified. Moreover, limited studies compare CLL pts who have/have not received HSCT in terms of differences (diff) in characteristics (char) at diagnosis (dx), population (pop) composition and outcomes. This study evaluates pred factors for outcomes post allo-HSCT, and compares dx char between (bn) tx CLL pts who did /did not receive HSCT by evaluating a large pop-based CLL cohort (n= 1044) from the BC Provincial CLL Database (BCPCD).Methods: 102 CLL pts (71M, 31F) had consecutive allo-HSCT (01-91 to 03-13, L/BMT Program of BC). Median (med) age (range) at dx:HSCT was 50 (26-65):57 (32-68) yrs; med interval dx to HSCT 5.8 yrs (0.5 to 29). Most pts (78, 76%) received non-ablative therapy; (n=61 [60%] reduced-intensity fludarabine /busulfan [flu/bu] based [RIC], n=17 [17%] non-myeloablative flu-cyclophosphamide based [NMA]); 24 pts had myeloablative (MA) conditioning (CON). Donor status was 50% unrelated (UD) (51UD:51RD); 73M, 28 F.Results: With median (med) follow up (FU) (range) post allo-HSCT of 2 yrs (0.5-18); post dx of 9 yrs (1-38), 67 (50%) pts survive. 70 (69%) achieved CR post-HSCT a med of 187 (28-1274) days (d). 27 had CLL PROG a med of 339 (25-4367) d; 18 of 27 (67%) survive a med of 3 (0.5-18) yrs post HSCT. Factors pred OS post HSCT (KM p=; UVA HR=) (p<0.05) were: pre-HSCT FISH deletion 17p (del 17p) (0.005; 2.9), Dohner rank (0.02), HSCT specific comorbidity index (CoI) >3 vs. 0-2 (0.04; 2.5), HLA mismatched (MM) donor (0.03;2.3), pre-HSCT tx with alemtuzumab (alem) (0.005;3.0), CON (MA vs NMA or RIC) (0.046; 3.0), acute (A) graft vs host disease (GVHD) grade (g) 3-4 vs 0-2. (<0.001; 4.5), dn chim <90% (0.001; 5.2), abn FISH not clear post-HSCT (0.009; 2.6), yr of HSCT (pre- vs post-2010) (0.03; 3.13) and lack of CR post HSCT (<0.001; 10.5).The following sig pred for (OR; p=): >90% dn chim: no B symptoms at dx (2.5; 0.004), CON (RIC vs. NMA, (2.6; 0.006); clear FISH abn post-HSCT: CR post-HSCT (4.6; 0.004); CR post-HSCT: B symptoms at dx (0.4; 0.02), <=1 FISH abn (1.7; 0.045), rituximab (R) pre-HSCT (2.5; 0.001), clear FISH abn (2.5; 0.01), flu sensitivity (S) pre-HSCT (1.8; 0.03), S to last tx pre-HSCT (1.7; 0.03), CON (MA vs. RIC or NMA) (3.2; <0.001); PROG: Richter’s transformation ( Rich trans) pre-HSCT (3.5; 0.008), graft failure (3.3; 0.008), CoI >3 vs. 0-2 (6.9; 0.006), no R pre-HSCT (6.7; 0.01), CON (MA vs. NMA or RIC), (0.2; 0.03); NRM: pre-HSCT alem (2.7; 0.03), CoI >3 vs. 0-2 (2.7; 0.049), HLA MM (2.8; 0.01), CON (MA vs. rest) (3.0; 0.007), AGVHD g 3-4 vs. 0-2 (5.9; <0.001), FISH abn not clear (2.6; 0.04), and no CR (6.5; <0.001).Comparison bn allo-HSCT and BCPCD CLL pts showed sig diff at dx for Dohner FISH rank: more del 17p (23% vs.11%) and 11q (23% vs. 9%) in allo-HSCT pts (n=84 with pre-HSCT FISH); less +12 (13% vs. 17%), del 13q (24% vs.41%) or normal (22% vs 18%), p<0.001 than non-HSCT pts (n=952); Age at dx (med, range) was lower in HSCT (50, 26-65) vs non (62, 25-96), p<0.001; lymphocyte (lymph) count higher (14, 1-300 vs.11, 1-662, p=0.03), tx-free survival (TFS) from dx to 1st tx shorter at 0.75 (0-9.3) vs. 2.86 (0-20.6) yrs. Rich trans was more frequent in HSCT pts (8%) vs. non (3%), p=0.015.OS was sig better for HSCT pts (n=103) (med 17.6, SE 4.5, CI 95% 8.8-26) compared to non (n=494) (med 14.4, SE 1.1, CI 95% 12.1-16.6) (p=0.03).Conclusion: CLL allo-HSCT pts have sig diff than non including higher lymph at dx, shorter time to 1st tx, and higher risk FISH abn. 17p del remains high-risk with allo-HSCT. Pre-HSCT R increased post HSCT CR. Strategies to optimize post-HSCT CR and dn chim are important; these milestones are crucial to best outcome. PROG post-HSCT does not confer worse OS; rescue strategies are successful and deserve further study. Comparison of this large allo HSCT and pop-based BPCDB cohort indicate improved OS for allo-HSCT tx CLL pts vs. other, with a survival plateau. This data indicates early recognition of high-risk CLL patients for HSCT is likely to yield best long-term outcome. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Recurrent Mutations within the Nfkbie gene: A Novel Mechanism for NF-κB Deregulation in Aggressive Chronic Lymphocytic Leukemia

Dysregulated NF-κB signaling appears to be particularly important in B-cell malignancies, with recurrent mutations identified within both the canonical and non-canonical NF-κB pathways, as well as in components of the B-cell receptor (BcR) and Toll-like receptor (TLR) signaling pathways. In chronic lymphocytic leukemia (CLL), although recurrent mutations have been identified in MYD88 (TLR signaling) and BIRC3 (non-canonical NF-κB pathway), their frequency is low (<3%) and hence the extent to which genetic aberrations may contribute to constitutional NF-κB activation remains largely unknown. To gain further insight into this issue, we designed a HaloPlex gene panel (Agilent Technologies) and performed targeted next-generation sequencing (NGS) (HiSeq 2000/Illumina) of 18 NF-κB genes in a discovery cohort of 124 CLL patients, intentionally biased towards poor-prognostic patients with either unmutated IGHV genes or high-risk genomic aberrations. Using a conservative cutoff of >10% for the mutant allele, we identified mutations (n=35) within 30/124 (24%) patients in 14/18 NF-κB genes analyzed. IκB genes, which encode for cytoplasmic inhibitor proteins, accounted for 20/35 (57%) mutations, with IκBε (encoded by NFKBIE) mutated in 8 patients; notably, 3/8 cases carried an identical 4bp deletion within exon 1 of NFKBIE. Prompted by these findings, we proceeded to validate our findings in an independent CLL cohort (n=168) using the same methodology as above and primarily focusing on cases with poor-prognostic features. We identified 30 mutations within 28 CLL patients in 11/18 NF-κB genes analyzed. Strikingly, 13/30 mutations were found within IκBε, with 10/13 patients carrying the same 4bp NFKBIE deletion. Notably, investigations into whether additional cases (within both the discovery and validation cohort) may harbor mutations of low clonal abundance (<10% mutant allele), led to the detection of the NFKBIE deletion in another 18 cases. Owing to the prevalence of this 4bp deletion within the NFKBIE gene, we developed a GeneScan assay and screened an additional 312 CLL cases. Collectively, 40/604 (6.6%) CLL patients were found to carry this frame-shift deletion within the NFKBIE gene, which is in line with a recent publication reporting that 10% of Binet stage B/C patients carried this mutation (Damm et al. Cancer Discovery 2014). Remarkably, the majority of these NFKBIE mutations (16/40) were found in a subgroup of patients that expressed highly similar or stereotyped BcRs and are known to have a particularly poor outcome, denoted as subset #1. This finding thus alludes to a subset-biased acquisition and/or selection of genomic aberrations, similar to what has been reported for subset #2 and SF3B1, perhaps as a result of particular modes of BcR/antigen interaction. We utilized proximity-ligation assays to test the functional impact of the NFKBIE deletion by investigating protein-protein interactions. This analysis revealed reduced interaction between the inhibitor IκBε and the transcription factor p65 in NFKBIE-deleted CLL cells; IκBε-knock-down shRNA experiments confirmed dysregulated apoptosis/NF-κB signaling. Finally, to assess whether the NFKBIE deletion could also be present in other B-cell malignancies, we screened 372 mature B-cell lymphoma cases using NGS or the GeneScan assay and found the deletion in 7/136 (5.1%) mantle cell lymphomas, 3/66 (4.5%) diffuse large B-cell lymphomas and 3/170 (1.8%) splenic marginal zone lymphomas. Taken together, our analysis revealed that inactivating mutations within the NFKBIE gene lead to NF-κB activation in CLL and potentially several other B-cell-derived malignancies. Considering the central role of BcR stimulation in the natural history of CLL, the functional loss of IκBε may significantly contribute to sustained CLL cell survival and shape the disease evolution. This novel data strongly indicates that components of the NF-κB signaling pathway may be prime targets for future targeted therapies not only in CLL but also other mature B-cell lymphomas. DisclosuresNo relevant conflicts of interest to declare.

High Resolution Genomic Profiling of Primary “Ultra High Risk” and Refractory Chronic Lymphocytic Leukemia: Results from the CLL2O Trial

Introduction:“Ultra high-risk” chronic lymphocytic leukemia (CLL) can be defined by TP53 loss and/or mutation, and refractory or early relapsing disease after intense chemo(immuno)therapy. Since its pathogenic mechanisms are not fully understood and TP53 alteration is only found in approx. half of the cases, it was our aim to identify novel genetic alterations that might contribute to the pathogenesis of “Ultra high-risk CLL”.Methods:We screened 112 samples [36 treatment naïve patients with del(17p), 26 with del(17p) at relapse, and 50 refractory to purine analogue-based therapy] from the CLL2O trial of the GCLLSG by Affymetrix® 6.0 single nucleotide polymorphism (SNP) arrays [N=79 paired to intraindividual reference DNA, N=33 unpaired].Results:In total, 663 copy number alterations (CNA) were detected corresponding to a mean of 5.9 CNAs per case [6.0 paired, 5.7 unpaired]. Compared to an unselected CLL cohort at time of first treatment this number was strikingly high [CLL8 trial: 1.8 CNAs per case].The most frequent aberrations, apart from the well established ones, were: +(8)(q24.21) [15%], del(8p) [14%], del(18p) [14%], del(10)(q24.32) [12%], del(15)(q15.1) [11%], del(6)(q21) [10%], del(3)(p26.1-p25.3) [10%], del(9)(p21.3) [10%], del(3)(p21.31) [9%], +(2)(p16.1-p15) [8%], del(4)(p15.2-p15.1), del(9)(q13 q21) [8%].The high number of CNAs per case was associated with alterations in DNA damage response (DDR) genes. Patients with both del(17p) and TP53 mutation [N=80] had a mean of 6.4 CNAs, whilst del(17p) cases without TP53 mutation had 4.7 [N=7]. Cases without TP53 alteration [N=20] had a CNA score of 4.4.Apart from TP53, we found that loss of three other components of the DDR pathway, ATM, RAD18 and SMC5, had impact on the CNA score. In particular, in del(17p) / TP53 mutated cases [N=80], the number of CNAs per case was further increased by a loss of one of the genes: ATM 8.8 CNAs per case [del(11)(q22.3), N=8]; RAD18 9.8 CNAs per case [del(3)(p26.1-p25.3), N=9]; SMC5 10.0 CNAs per case [del(9)(q13 q21), N=7].Within the group with no TP53 alteration [N=20], cases without ATM, RAD18 or SMC5 loss [N=9] had a mean value of only 2.0 CNAs per case.Frequent occurrence of discontinuous lesions [N=168 in 71 cases] implicated a high frequency of chromothripsis. Referring to its definition of 10 switches between 2 or more copy number states, chromothripsis was found in 11 cases [10%]. With a cut point of ≥ 5 switches, 37 cases were affected [33%]. Chromothripsis further increased the mean number of CNAs per case [≥ 10 switches: 9.6 vs. 5.5 CNAs per case; ≥5 switches: 8.8 vs. 4.7 CNAs per case].Interestingly, prior therapy was not associated with an increased number of CNAs in this cohort [7.3 CNAs per case in treatment-naïve cases, 5.4 in refractory and 6.0 in relapsed cases]. The number of prior chemotherapy lines did not have impact on genomic complexity [1 line (N=23): 5.8 CNAs; 2 (N=16): 4.6 CNAs; 3 (N=15): 4.4 CNAs; ≥ 4 (N=22): 5.7 CNAs). This did also apply to the group of del(17)(p13) / TP53 mutated cases without loss of ATM, RAD18 or SMC5 [N=62] [no previous therapy (N=22): 7.4 CNAs; 1 or 2 previous therapy lines (N=20): 4.7 CNAs; ≥ 3 (N=20): 4.6].Neither genomic complexity (> 3 CNAs per case) nor chromothripsis had any impact on the clinical outcome. We hypothesised whether the recurrent lesions mentioned above are associated with the refractory nature of the disease. Clinical significance tests revealed that only del(9)(p21.3) encompassing the CDKN2A/CDKN2B gene locus and +(8)(q24.21) had significant impact on the clinical outcome [del(9)(p21.3): shorter OS (p=0.038), tendency to shorter PFS (p=0.12); +(8)(q24.21): shorter PFS (p=0.0016), tendency to shorter OS (p=0.064), inferior response (p=0.0079)]. Results for +(8)(q24.21) were in line with an accumulation of cases with 8q gain in the group of refractory [16%, N=8 of 50] and relapsed cases [23%, N=6 of 26]. The minimally amplified region on 8q24.21 [16.6 kb] could be delineated to a region previously linked to the enhancement of MYC expression.Conclusion:The majority of “ultra high-risk” CLL showed high genomic complexity. Alteration of genes involved in DNA damage response [TP53, ATM, RAD18, SMC5] and chromothripsis increased the number of aberrations per case, while, somewhat surprisingly, prior therapy did not. Prognostic impact of additional genomic lesions can be suggested for +(8)(q24.21) [MYC] and del(9)(p21.3) [CDKN2A/CDKN2B]. DisclosuresLeblond:Roche: Honoraria, Speakers Bureau. Gribben:Celgene: Research Funding; Pharmacyclics: Honoraria; Roche: Honoraria. Stilgenbauer:Pharmacyclics, Janssen: Honoraria, Research Funding.

Novel Putative Driver Gene Mutations in Chronic Lymphocytic Leukemia (CLL): Results from a Combined Analysis of Whole-Exome Sequencing of 262 Primary CLL Samples

Unbiased high-throughput massively parallel sequencing methods have transformed the process of discovery of novel putative driver gene mutations in cancer. In chronic lymphocytic leukemia (CLL), these methods have yielded several unexpected findings, including the driver genes SF3B1, NOTCH1 and POT1. Recent analysis, utilizing down-sampling of existing datasets, has shown that the discovery process of putative drivers is far from complete across cancer. In CLL, while driver gene mutations affecting >10% of patients were efficiently discovered with previously published CLL cohorts of up to 160 samples subjected to whole exome sequencing (WES), this sample size has only 0.78 power to detect drivers affecting 5% of patients, and only 0.12 power for drivers affecting 2% of patients. These calculations emphasize the need to apply unbiased WES to larger patient cohorts.To this end, we performed a combined analysis of CLL WES data joining together our previously published cohort of 159 CLLs with data from 103 CLLs collected by the International Cancer Genome Consortium (ICGC). The raw sequencing reads from these 262 primary tumor samples (102 CLL with unmutated IGHV, 147 with mutated IGHV, 13 with unknown IGHV status) were processed together and aligned to the hg19 reference genome. Somatic single nucleotide variations (sSNVs) and indels were detected using MuTect. Subsequently, inference of recurrently mutated genes was performed using the MutSig algorithm. This method combined several characteristics such as the overall mutation rate per sample, the gene specific background mutation rate, non-synonymous/synonymous ratio and mutation clustering to detect genes that are affected by mutations more than expected by chance.This analysis identified 40 recurrently mutated genes in this cohort. This included 22 of 25 previously identified recurrently mutated genes in CLL. In addition, 18 novel candidate CLL drivers were identified, mostly affecting 1-2% of patients. The novel candidates included two histone proteins HIST1H1D and HIST1H1C, in addition to the previously identified HIST1H1E. Another was IKZF3, affected by a recurrent sSNV resulting in a p.L162R change in its DNA binding domain, in close proximity to a region recently identified as critical for lenalidomide resistance in multiple myeloma (MM). An additional recurrently mutated gene was nuclear RNA export factor 1 (NXF1), which along with previously known recurrently mutated genes (SF3B1, XPO1, DDX3X), highlights the importance of RNA processing to CLL biology. Finally, this search for putative CLL driver genes also identified ASXL1 and TRAF3, already characterized as drivers in acute myeloid leukemia and MM, respectively. Of the 59 of 262 samples for which RNA-seq data were available, 76% of the identified driver mutations were detected and thereby validated. Validation using RNAseq detection of driver mutations and targeted sequencing within the entire cohort are ongoing.The larger size of our cohort enabled the separate application of the somatic mutation discovery process to samples with mutated or unmutated IGHV. Among the 147 samples with mutated IGHV, only 5 driver genes (TP53, SF3B1, MYD88, CHD2, RANBP2) retained significance. In contrast, analysis of the 102 IGHV unmutated samples revealed a distinct and more diverse pattern of recurrently mutated genes (lacking MYD88 and CHD2, and including NOTCH1, RPS15, POT1, NRAS, EGR2, BRAF, MED12, XPO1, BCOR, IKZF3, MAP2K1, FBXW7 and KRAS). This extended cohort also allowed for better resolution of the clinical impact of those genetic variants with greater than 4% prevalence in the cohort. For example, samples with POT1 mutations were found to be associated with shorter time from sample to therapy compared with those with wild-type POT1 (P= 0.02).Our study demonstrates that with larger cohort size, we can effectively detect putative driver genes with lower prevalence, but which may nonetheless have important biological and clinical impact. Moreover, our interrogation shows that subset analysis can reveal distinct driver patterns in different disease subsets. In particular, the marked clinical difference between CLLs with mutated and unmutated IGHV may reflect the higher likelihood of the latter group to harbor a broader spectrum of driver mutations with a more complex pattern of co-occurrence. DisclosuresBrown:Sanofi, Onyx, Vertex, Novartis, Boehringer, GSK, Roche/Genentech, Emergent, Morphosys, Celgene, Janssen, Pharmacyclics, Gilead: Consultancy.

The impact of SF3B1 mutations in CLL on the DNA-damage response.

Mutations or deletions in TP53 or ATM are well-known determinants of poor prognosis in chronic lymphocytic leukemia (CLL), but only account for approximately 40% of chemo-resistant patients. Genome-wide sequencing has uncovered novel mutations in the splicing factor sf3b1, that were in part associated with ATM aberrations, suggesting functional synergy. We first performed detailed genetic analyses in a CLL cohort (n=110) containing ATM, SF3B1 and TP53 gene defects. Next, we applied a newly developed multiplex assay for p53/ATM target gene induction and measured apoptotic responses to DNA damage. Interestingly, SF3B1 mutated samples without concurrent ATM and TP53 aberrations (sole SF3B1) displayed partially defective ATM/p53 transcriptional and apoptotic responses to various DNA-damaging regimens. In contrast, NOTCH1 or K/N-RAS mutated CLL displayed normal responses in p53/ATM target gene induction and apoptosis. In sole SF3B1 mutated cases, ATM kinase function remained intact, and γH2AX formation, a marker for DNA damage, was increased at baseline and upon irradiation. Our data demonstrate that single mutations in sf3b1 are associated with increased DNA damage and/or an aberrant response to DNA damage. Together, our observations may offer an explanation for the poor prognosis associated with SF3B1 mutations.

Proteomics-based strategies to identify proteins relevant to chronic lymphocytic leukemia.

Chronic lymphocytic leukemia (CLL), a malignant B-cell disorder, is characterized by a heterogeneous clinical course. Two-dimensional nano liquid chromatography (2D-nano-LC) coupled with matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF/TOF MS) (LC-MALDI) was used to perform qualitative and quantitative analysis on cellular extracts from 12 primary CLL samples. We identified 728 proteins and quantified 655 proteins using isobaric tag-labeled extracts. Four strategies were used to identify disease-related proteins. First, we integrated our CLL proteome with published gene expression data of normal B-cells and CLL cells to highlight proteins with preferential expression in the transcriptome of CLL. Second, as CLL's outcome is heterogeneous, our quantitative proteomic data were used to indicate heterogeneously expressed proteins. Third, we used the quantitative data to identify proteins with differential abundance in poor prognosis CLL samples. Fourth, hierarchical cluster analysis was applied to identify hidden patterns of protein expression. These strategies identified 63 proteins, and 4 were investigated in a CLL cohort (39 patients). Thyroid hormone receptor-associated protein 3, T-cell leukemia/lymphoma protein 1A, and S100A8 were associated with high-risk CLL. Myosin-9 exhibited reduced expression in CLL samples from high-risk patients. This study shows the usefulness of proteomic approaches, combined with transcriptomics, to identify disease-related proteins.

Low Frequency Of IGHV4-39/IGHD6-13/IGHJ5 Rearrangement With Stereotyped HCDR3 (subset #8) In Russian and Ukrainian Chronic Lymphocytic Leukemia Patients

The leukemia cells of unrelated patients with chronic lymphocytic leukemia (CLL) display a restricted repertoire of immunoglobulin gene rearrangements and very similar structure of B-cell receptor (stereotyped BCR). Numerous different subsets of stereotyped BCRs were identified. Some associations between specific BCRs subsets, clinical course of CLL, and molecular abnormalities were found. Namely, short overall survival, accumulation of SF3B1 mutations in subset #2, high risk for Richter's transformation, trisomy 12, NOTCH1 mutations in subset #8, increased frequency of NOTCH1 mutations in subset #1 were revealed (Rossi et al., 2009, 2013; Streffoed et al., 2013).The aim of this study was to evaluate distribution of different stereotyped BCR subsets in Russian and Ukrainian CLL cohort with special attention on prognostic unfavorable subsets. MethodsImmunoglobulin heavy chain variable (IGHV) genes of 969 CLL patients (564 from Russia and 405 from Ukraine) were amplified according to BIOMED-2 rules, sequenced and analyzed with IMGT and IgBlast databases. Additionally, NOTCH1 c.7544_7545delCT mutations were analyzed in 301 patients (randomly collected and in the prognostic unfavorable subsets) by amplification refractory mutation system (ARMS) PCR as proposed by Rossi D. et al., 2012. In 19 cases NOTCH1 mutational status was confirmed by direct DNA Sanger sequencing. ResultsJoined Russian-Ukrainian CLL cohort was presented mainly by unmutated (UM) IGHV gene cases (64.2%). The distribution of major clusters of stereotyped BCR did not differ from those presented in the literature (comparison was made separately for cases with mutated (M) and UM IGHV genes, taking into account the differences in the mutational status of IGHV genes between our CLL cohort and CLL cohorts from European countries and USA), except for subset #8, which was found in one case only (0.16% of UM IGHV gene cases). At the same time, subset #8 (IGHV4-39/IGHD6-13/IGHJ5 rearrangement) belongs to the group of “major” subsets in CLL (Agathangelidis et al., 2012), and its frequency varies from 0.9% to 2% among cases with UM IGHV genes in the most of presented CLL cohorts (Murray et al., 2008; Messmer et al., 2009; Maura et al., 2011). Frequency of subset #2 had tendency to be lower among Russian CLL patients (1.4% of M CLL cases), but not among Ukrainian patients (4.5% of M CLL cases) comparing with European CLL cohorts.NOTCH1 c.7544_7545delCT was identified in 27 out of 301 patients (9.0%), which was in accordance with other CLL studies. Twenty five cases with NOTCH1 mutations had UM IGHV genes and 2 cases expressed M IGHV3-21 gene (both subset #2). As well, majority of NOTCH1 mutated cases (17/27, 63%) showed stereotyped BCRs, which belonged to the following subsets: #1 (2 of 12), #2 (2 of 7), #3 (1 of 6), #7 (2 of 10), #9 (2 of 10), #25 (1 of 2), #59 (1 of 3) (subset's designation according Murray et al., 2008), UA/ref2 (1 of 2), UA/ref11 (2 of 2) (subset's designation according Bilous et al., 2010), or having a homology with the CLL sequences of the public CLL database. No NOTCH1 mutations were found in 11 IGHV4-39-used cases, including one case from subset #8. In 10 of 27 (37.0%) NOTCH1 mutated cases IgHV1-69 gene was used, while among wild type NOTCH1 patients IgHV1-69 was presented in 65/274 (23.7%) cases (p=0.12). ConclusionThe distribution of the major subsets of stereotyped BCRs in CLL patients from Russia and Ukraine (eastern slavic countries) was similar to the western European countries and USA, except for low frequency of subset #8, associated with unfavorable clinical course. Our preliminary data suggested that NOTCH1 c.7544_7545delCT mutation is typical for UM CLL cases with stereotyped BCRs, but no clear association with specific subsets was found in the absence of subset #8. Disclosures:No relevant conflicts of interest to declare.

Old DAT / New Data : Positive Direct Antiglobulin Test Identifies a Specific Subgroup Of CLL Patients

Autoimmune cytopenias represent a well-known complication of chronic lymphocytic leukemia (CLL). The most frequent one is autoimmune haemolytic anemia (AIHA), occurring in about 5% of the patients during the course of the disease. The prognostic significance of autoimmune cytopenias remains uncertain and controversial. In previous studies, a positive direct antiglobulin test (DAT) was found in up to 35% of CLL cases, however only a minority of the patients would develop AIHA at any time of the disease. The aim of this monocentric study was to explore the clinical and biological characteristics and the outcome of CLL patients showing a positive direct antiglobulin test at any time during the course of the disease. We reviewed all CLL patients seen at our institution who had at least one positive DAT between January 2007 and May 2013.Fifty-four patients were found, representing about 10% of our CLL cohort. The sex ratio was 1,8 (35M/19F). Median age at diagnosis was 66,2 years (range 44,7 to 87,3). According to the Binet classification, 41 (76 %) patients were in stage A, 7 in stage B, and 6 in stage C. Study of usual prognostic parameters evidenced that these patients represented a very high risk group: 82% of cases harbored unmutated IGHV, and CD38 was expressed in 60% cases. Cytogenetic alterations were studied by FISH analysis: 11q deletion, trisomy 12 and 17p deletion were found in 12,5%, 22,9% and 14,3% cases respectively.In this cohort, with a median follow up of 85 months, 41 (76 %) patients required treatment. The first line of treatment was initiated for CLL progression in 33 cases, for symptomatic AIHA along with CLL progression in 5, and for symptomatic AIHA (without CLL progression) in 3. The median time to first treatment was very short (16 months), and the median overall survival of this cohort was 84 months. In 22 cases, the DAT was positive from the time of diagnosis, while in 34 patients it became positive later during the course of CLL. Only 19 patients (35%) developed symptomatic AIHA. DAT specificity was IgG, IgG+ C or C for respectively 36, 10 and 8 patients. Interestingly, there was no impact of the time of first positive DAT during the course of the disease (at diagnosis or later). Moreover, occurrence of a symptomatic AIHA had no impact on treatment free survival or overall survival when compared with cases with positive DAT only.Among these cases, 41 were in stage A at diagnosis and up to 78% harbored unmutated IGHV (as compared with 30 % in our institutional stage cohort). In order to evaluate the impact of a positive DAT on a homogenous population, we focused on the 31 IGHV unmutated stage A cases and compared them with our cohort of IGHV unmutated stage A with consistently negative DAT. We found a significant prevalence of VH1-69 and VH 3-21 use (43 %) and a high percentage belonging to a stereotyped subset (34%). The majority of the cases (19/28 tested) were responders to in vitro B-cell receptor stimulation by anti IgM. In this population, time to first treatment (26 months) was slightly shorter but median overall survival (55 months) was significantly reduced highlighting the poor response to treatment.In conclusion, occurrence of a positive DAT at any time of the course of CLL is associated with poor outcome and has the same adverse prognostic impact as the onset of a symptomatic AIHA. Moreover, study of B-Cell receptor (BCR stimulation by anti-IgM, CDR3 stereotypy) show that these patients may represent a pathophysiological distinct subgroup in which antigen stimulation is likely to play a major role. Disclosures:No relevant conflicts of interest to declare.

Heritable Predisposition To Richter Syndrome In Patients With Chronic Lymphocytic Leukemia

BackgroundApproximately 2-8% patients with chronic lymphocytic leukemia (CLL) will transform to diffuse large B-cell lymphoma (DLBCL, Richter syndrome [RS]). Clinical characteristics and molecular markers at the time of CLL diagnosis are associated with the risk of RS; however, there are no data regarding germline genetic variations and the risk of RS. Genomewide association studies (GWAS) have shown several single-nucleotide polymorphisms (SNPs) that are associated with a higher risk of familial CLL. It is not known whether any of these polymorphisms also predispose to RS. MethodsSince 2002, all consecutive patients with newly diagnosed (<9 months diagnosis) CLL at Mayo Clinic were offered enrollment into a prospective genetic epidemiology study. Patients completed extensive epidemiologic questionnaires and baseline clinical, laboratory, and biologic data were abstracted using a standard protocol. Genotyping of germline tissue at diagnosis was performed using an Illumina iSelect panel and Affymetrix 6.0 SNP chip. All patients were prospectively and longitudinally followed at defined time-points with systematic collection of data on treatments, second cancers, and RS. All patients with biopsy-proven DLBCL during follow-up were considered to have undergone transformation into RS. Time to RS was calculated from CLL diagnosis date until RS or until last follow-up date for those with no RS. SNPs were modeled in two ways: ordinal and dominant. Cox regression was used to estimate hazard ratio (HR) for individual SNPs with time to transformation. ResultsThirteen of the GWAS-discovered SNPs associated with risk of developing CLL were available and genotyped on 620 CLL patients. Median age at diagnosis of CLL was 62 years (range 27-88), and 428 (69%) were male. Three hundred and ten (51%) patients were low (0) Rai stage, 271 (45%) were intermediate (I-II) Rai stage, and 22 (4%) were advanced (III-IV) Rai stage. The immunoglobulin heavy chain gene was unmutated in 189 (40%) patients; 157 (32%) patients expressed ZAP-70, 163 (29%) expressed CD38 and 104 (31%) expressed CD49d. Fluorescence in-situ hybridization (FISH) revealed that 210 (41%) patients had del13q, 90 (18%) patients had trisomy 12, 37 (7%) had del11q, 23 (5%) had del17p and 141 (28%) had no detectable FISH abnormalities. As of last follow-up, 239 (39%) patients received therapy for CLL.After a median follow-up of 5.9 years (range 0-11), 15 (2.4%) patients developed biopsy-proven RS. The median time to RS in these 15 patients was 4.5 years (range 1.0-8.7 years). The ordinal HR for the 13 SNPs tested, their corresponding genes, and p-values are shown in Table 1. Germline polymorphisms in a single SNP, rs4987852, encoding for BCL2 (chromosome 18), was significantly associated with an increased risk of RS (ordinal HR=3.9; 95% CI=1.6-9.8; p-value=0.004). This allele was present in 48/605 (8%) non-transformed CLL patients compared to 4/15 (27%) of patients with RS. Time to RS according to the Kaplan-Meier analysis for rs4987852 is shown in Figure 1. This SNP is located in a region in which t(14;18) translocation breakpoints commonly occur in follicular lymphoma and overexpression of BCL2 leads to an increased incidence of B-cell lymphomas in mice.Table 1Reference SNP Cluster IDChromosomeChromosome regionGeneOrdinal Hazard Ratiop-valuers1339798522q13SP140.SP1100.790.59rs1748346622q13ACOXL.BCL2L111.520.29rs376982522q33CASP10.CASP80.590.16rs757819922q37HDLBP1.730.16rs75797822q37FARP20.710.58rs937880566p25IRF40.920.83rs79440041111p15C11orf21.TSPAN321.080.82rs7356651111q24GRAMD1B.SCN3B1.150.74rs71694311515q21RFXDC21.450.42rs22929821616q24IRF80.720.47rs43682531818q21PMAIP11.250.58rs49878521818q21BCL23.900.004rs48023221919q13PRKD2.STRN41.60.21 [Display omitted] ConclusionOur results suggest that inherited genetic polymorphisms predispose CLL patients to develop RS. Specifically, SNP (rs4987852) present on the BCL2 gene on chromosome 18 in CLL is associated with an increased risk of transformation to RS. These observations require replication in other CLL cohorts. Disclosures:Shanafelt:Genentech: Research Funding; Glaxo-Smith-Kline: Research Funding; Cephalon: Research Funding; Hospira: Research Funding; Celgene: Research Funding; Polyphenon E International: Research Funding.

Clonal Abundance Determines The Prognostic Relevance Of a NOTCH1 Mutation In Chronic Lymphocytic Leukemia

IntroductionThe transmembrane receptor NOTCH1 operates as a ligand-activated transcription factor controlling developmental processes, proliferation and apoptosis. In the context of cancer, activating NOTCH1 mutations are the most frequent oncogenic events in T-cell acute lymphoblastic leukemia and have been implicated in chronic lymphocytic leukemia (CLL) as well. The most prevalent CLL NOTCH1 mutation (N1ΔCT) leads to a truncation of the protein (p.P2515Rfs*4) and has been associated with impaired overall survival (OS). Here, we applied three different methods to study the N1ΔCT prevalence and subclone size in a cohort of n=275 CLL patients. MethodsPresence of the N1ΔCT mutation was analyzed using newly established restriction fragment length polymorphism (RFLP) and allele-specific PCR (AS-PCR) methodologies. A novel real-time PCR (qRT-PCR) assay was used to precisely quantify the N1ΔCT allele frequency. Presence of the N1ΔCT mutation was confirmed by conventional Sanger sequencing. ResultsUsing RFLP analysis we detected the N1ΔCT mutation in n=17 CLL patients. In parallel, we used a more sensitive AS-PCR and identified n=12 additional N1ΔCT-mutated cases resulting in a total N1ΔCT mutation rate of 10.5% (n=29/275) in our cohort. The OS of RFLP-positive patients (RFLP+) was significantly shorter than the OS of N1ΔCT-unmutated patients (wt) (mean OS; RFLP+, 87 months vs. wt, 218 months; p=0.017). In contrast, OS of AS-PCR-positive cases (AS-PCR+) did not differ significantly from the OS of wt patients (mean OS; AS-PCR+, 175 months vs. wt, 218 months; p=0.42). These data prompted us to design a quantitative real-time PCR (qRT-PCR) assay, which is capable of precisely quantifying the size of the N1ΔCT-mutated subclones (allele frequency, %) in our CLL cohort. As expected, significantly different allele frequencies between RFLP+ (mean±SEM 27.1±3.4%), AS-PCR+ (3.7±0.6%) and wt patients (0.6±0.04%) were revealed by qRT-PCR (p<0.0001). In order to determine a methodology-independent cut-off which correlates with the clinical significance of the N1ΔCT mutation, we employed Receiver Operating Characteristics (ROC) analysis based on survival status and calculated a N1ΔCT allele frequency cut-off of 15.2% (AUC=0.71). Next, we determined N1ΔCT allele frequencies over time to investigate clone dynamics within individual patients (n=15 patients, mean observation period 87.4 months; range 5-186 months). Unexpectedly, the N1ΔCT allele frequencies remained relatively constant and none of the patients with N1ΔCT allele frequencies below 15.2% rose above this cut-off over time. ConclusionsOur data demonstrate that high abundance of a N1ΔCT-mutated CLL clone correlates with an aggressive disease course. In our CLL cohort a N1ΔCT allele frequency below 15% was of negligible clinical relevance. Thus, mere qualitative detection of a N1ΔCT mutation by PCR is not inevitably associated with shortened survival. Surprisingly, we did not observe that a minor N1ΔCT clone became dominant over time. Disclosures:No relevant conflicts of interest to declare.

NOTCH1 Mutations Are Associated With The 14q Deletion In Chronic Lymphocytic Leukemia (CLL) and Small Lymphocytic Lymphoma (SLL)

Deletions of the long arm of chromosome 14 are rare (<5%) but recurrently observed in mature B-cell neoplasms, particularly in CLL. The size of the deletions is variable, and the molecular consequences are unknown. The aim of our study is to characterize this abnormality in a large cohort of patients.The Groupe Francophone de Cytogenetique Hematologique (GFCH) collected data from 89 patients with CLL or Small Lymphocytic Lymphoma (SLL), harboring a del14q. Morphological review was performed for 65 of them and immunological review for 48. All karyotypes were reviewed by members of the GFCH. Fluorescence in situ hybridization (FISH) analysis was performed with CEP12, 13q14, TP53, ATM, 6q21, and IGH probes, and BACs RP11-35D12 and RP11-226F19 covering the ZFP36L1 gene on 14q24.1. SNP-array analysis (illumina omni1, omin2.5) was performed for 37 patients. The IGHV status was analyzed locally, or in our laboratory. The mutation hotspots of NOTCH1 (exon 34), SF3B1 (exons 14-16), XPO1 (exons 14-15), MYD88 (exon 5) and TP53 (exons 4-10) were analyzed by direct sequencing.Patients were classified as 49 CLL Matutes score 4-5, 5 atypical CLL score 3 (all CD5+, CD23+), and 27 SLL. Eight patients were excluded with a Matutes score<3 or not evaluable. The sex ratio M/F was 1.02, the median age at diagnosis was 64.5 years [29-87]. Regarding all CLL, there were 35 (66%) stages A, 10 (19%) stages B and 8 (15%) stages C. The median time between diagnosis and the first treatment (TTT) was 17 months [0-144], the overall survival (OS) at 5 years was 93.7%. For the 27 SLL patients, the median TTT was 1 month [0-85], and the OS at 5 years was 88.1%.The karyotype showed chromosomal translocation in 26/79 (33%) (10 balanced, 16 unbalanced) and was complex (> 3 abnormalities) in 26/79 (33%) cases. Using karyotype and FISH, we observed 28/79 (35%) trisomy 12, 12/79 (15%) 13q14 deletions, 11/80 (14%) TP53 deletions, 5/79 (6%) ATM deletions, 3/76 (4%) 6q21 deletions. The whole CLL cohort showed 15/53 (28%) tri 12, 11/53 (21%) del13q, 8/54 (15%) delTP53, 4/53 (7%) delATM, 2/50 (4%) del6q.IGHV status was not mutated in 41/53 (77%) patients, and the gene IGHV1-69 was rearranged in 21/52 (40%) cases. NOTCH1 gene was mutated in 14/45 (31%) patients, SF3B1 in 2/45 (4%), XPO1 in 2/45 (4%), MYD88 in 0/43, TP53 in 6/43 (14%) cases. A significant association was observed between TP53 disrupted and NOTCH1 mutations (p=.04). NOTCH1 mutation was not correlated with tri12.Regarding all the cytogenetic and molecular data, there was no significant difference between CLL score 4-5 and CLL score 3, and between CLL and SLL.Compared with common CLL in the literature, CLL with del14q dysplayed more tri12 (p<.03) and NOTCH1 mutations (p<.003), less del13q (p<.0006), more unmutated IGHV (p<.00001) and V1-69 repertoire (p<.00001). Compared with common SLL in the literature, SLL with del14q also exhibited more tri12, unmutated IGHV (p<.01) and V1-69 (p<.04).The 14q deletions appeared distributed along chromosome 14 from bands q11 to q32. The centromeric and telomeric breakpoints of the 14q deletion were investigated by FISH or SNP-array when material was available, and allowed us to identify a main group (Group 1) among 4 groups: Group 1: 37/77 (48%) patients, with IGH and ZFP36L1 loci rearranged; Group 2: 17/77 (22%), with ZFP36L1 deleted and IGH not deleted; Group 3: 7/77 (9%), with ZFP36L1 and IGH not deleted; Group 4, 16/77 (21%), all the other combinations. Group 1 showed a deletion del14q24.1-14q32.3 of about 38 megabases, which broke in or near ZFP36L1, and in the IGH gene. It included 26 (70%) CLL score 4-5 and 11 (30%) SLL. When compared to the other groups, Group 1 showed significantly more tri12 (p=.004), and NOTCH1 mutations (p=.02). IGHV status was unmutated in 22/24 (92%) patients in Group 1, 12/24 (50%) harboring the gene V1-69.In conclusion, the size of the 14q deletion is variable, with in about half of the cases a recurrent interstitial deletion 14q24.1-14q32.3. Both CLL and SLL with del14q, particularly with the 14q24.1-14q32.3 deletion, are associated with poor prognosis parameters, like short time to treatment, trisomy 12, NOTCH1 mutated, and unmutated IGHV status, with an overrepresentation of the V1-69 gene. Disclosures:No relevant conflicts of interest to declare.

Single Agent Ibrutinib (PCI-32765) Achieves Equally Good and Durable Responses In Chronic Lymphocytic Leukemia (CLL) Patients With and Without Deletion 17p

▪ IntroductionChemoimmunotherapy has markedly improved the outcomes of patients (pts) with CLL. However, pts having deletion of the short arm of chromosome 17 (DEL 17p) have inferior outcomes with current standard treatments (with the possible exception of allogeneic stem cell transplantation), and elderly pts are in need of less toxic regimens. Bruton's tyrosine kinase (BTK) is essential for B cell receptor (BCR) signaling, which has emerged as a key driver of CLL pathogenesis and progression. Ibrutinib (PCI-32765) is a covalent inhibitor of BTK with significant antitumor activity in CLL (Advani et al, JCO 2013, Byrd et al NEJM 2013). Patients and methodsThis investigator-initiated phase II, single-center trial of ibrutinib monotherapy prospectively addressed the possible role of ibrutinib in DEL 17p CLL irrespective of the pts' prior treatment history (NCT01500733). Elderly pts without a chromosomal deletion at 17p (normal (NL) 17p) were concurrently enrolled into a second cohort. The primary endpoint of the study is response after 6 months (mo) assessed by computed tomography (CT), bone marrow (BM) biopsy, and routine clinical and laboratory studies. The spleen volume was calculated from CT scans using a General Electric Advanced Workstation Server. Del 17p was assessed by interphase fluorescence in situ hybridization (FISH). Differences in response rates were calculated by a proportional test. ResultsWe report on the first 53 patients (n=24 (NL 17p); n=29 (DEL 17p)) with a median follow up of 14 months (mo). Median age was 66 yrs (33-85) and 70% had Rai stage III/IV. Most adverse events were grade ≤2, most commonly diarrhea, fatigue, arthralgias/myalgias, and rash. Treatment related non hematologic toxicities grade ≥3 occurred in < 5%, grade ≥3 infections or cytopenias were uncommon and reported in 15% of pts regardless of causality. Four deaths on study were not treatment related. A total of 47 pts were restaged at 6 mo (6 patients did not reach this endpoint; 2 unrelated deaths, 3 unrelated secondary malignancies, and 1 progressive disease due to presumed transformation at 2 weeks in a pt with DEL 17p). The estimated event free survival at 14 mo is 93%. At 6 months, 31(66%) pts had a partial response (PR by IWCLL criteria), and 13 (28%) pts a PR with lymphocytosis (PRL), i.e. pts fulfill criteria of PR except for the absolute lymphocyte count. Responses by treatment cohort (Table 1) were for NL 17p vs DEL 17p: 81% vs 53% PR and 9% vs 43% PRL. The apparent difference in response rates is due to slower clearance of the treatment-induced lymphocytosis in DEL 17p. However, clinical benefit and disease control in all tissue sites was equal for NL 17 and DEL 17p (Table 2): nodal response was 100% in both cohorts (median tumor reduction 75% in NL 17p and 70% on DEL 17p), reduction in spleen volumes of 269 ml (40%, range 30-1079 ml) vs 446 ml (46%, range 44-1716 ml), and reduction in tumor infiltration in the bone marrow by 76% and 84%, respectively. Median ALC decreased from 79 k/µl (0.5–402) to 30 k/µl (0-167) for a median reduction of 71% and 60%, respectively.Table 1Responses at 6 monthsResponsesAll patientsNL 17p N=21DEL 17p N=26P =PR66%81%53%0.04PRL28%9%43%0.02SD4%9%0%0.11PD2%0%4%0.35Nodal response (>50% reduction)100%100%100%NATable 2Degree of disease reduction (tumor volume at 6 months / pre-treatment)Median tumor reductionAll patientsNL 17p N=21DEL 17p N=26P =Nodal73%75%70%0.75Spleen44%40%46%0.50BM80%76%84%0.51ALC62%71%60%0.42To obtain a direct measure of the relative impact of ibrutinib on tumor cells carrying a DEL 17p, we repeated FISH testing at 6 mo (n=20). In the individual pts, DEL 17p was present in 12-97% of the tumor cells pre treatment and in 0-92% at 6 months. Interestingly, in 80% of pts the relative size of the DEL 17p subclone decreased (n = 20, median reduction 34%; P< 0.02); 4 pts (20%) had no evidence of 17p after 6 mo. ConclusionIbrutinib as a single agent appears to be equally effective against CLL with or without DEL 17p. This conclusion is based on a comparison of responses in two concomitantly treated CLL cohorts and supported by the absence of a treatment-related increase in the 17p clone in individual patients.Research supported by the Intramural Research Program of NHLBI. We thank our patients for participation and their willingness to have additional research studies done. We acknowledge Pharmacyclics for providing study drug. Disclosures:Off Label Use: Ibrutinib not FDA approved for CLL.

CD49d is overexpressed by trisomy 12 chronic lymphocytic leukemia cells: evidence for a methylation-dependent regulation mechanism

AbstractCD49d is a negative prognosticator in chronic lymphocytic leukemia (CLL), expressed by ∼40% of CLL cases and associated with aggressive, accelerated clinical courses. In this study, analyzing CD49d expression in a wide CLL cohort (n = 1200) belonging to different cytogenetic groups, we report that trisomy 12 CLL almost universally expressed CD49d and were characterized by the highest CD49d expression levels among all CD49d+ CLL. Through bisulfite genomic sequencing, we demonstrated that, although CD49d+/trisomy 12 CLL almost completely lacked methylation of the CD49d gene, CD49d–/no trisomy 12 CLL were overall methylated, the methylation levels correlating inversely to CD49d expression (P = .0001). Consistently, CD49d expression was recovered in CD49d– hypermethylated CLL cells upon in vitro treatment with the hypomethylating agent 5-aza-2′-deoxycytidine. This may help explain the clinicobiological features of trisomy 12 CLL, including the high rates of cell proliferation and disease progression, lymph node involvement, and predisposition to Richter syndrome transformation.

Progranulin Is a Novel Independent Predictor of Disease Progression and Overall Survival in Chronic Lymphocytic Leukemia

Progranulin (Pgrn) is a 88 kDa secreted protein with pleiotropic functions including regulation of cell cycle progression, cell motility, wound repair and tumorigenesis. Using microarray based gene expression profiling we have recently demonstrated that the gene for Pgrn, granulin (GRN), is significantly higher expressed in aggressive CD38+ZAP-70+ as compared to indolent CD38−ZAP-70− chronic lymphocytic leukemia (CLL) cases. Here, we measured Pgrn plasma concentrations by enzyme-linked immunosorbent assay (ELISA) in the Essen CLL cohort of 131 patients and examined Pgrn for association with established prognostic markers and clinical outcome. We found that high Pgrn plasma levels were strongly associated with adverse risk factors including unmutated IGHV status, expression of CD38 and ZAP-70, poor risk cytogenetics (11q-, 17p-) as detected by flourescence in situ hybridization (FISH) and high Binet stage. Pgrn as well as the aforementioned risk factors were prognostic for time to first treatment and overall survival in this series. Importantly, these results could be confirmed in the independent multicentric CLL1 cohort of untreated Binet stage A patients (n = 163). Here, multivariate analysis of time to first treatment revealed that high risk Pgrn (HR = 2.06, 95%-CI = 1.13–3.76, p = 0.018), unmutated IGHV status (HR = 5.63, 95%-CI = 3.05–10.38, p<0.001), high risk as defined by the study protocol (HR = 2.06, 95%-CI = 1.09–3.89, p = 0.026) but not poor risk cytogenetics were independent prognostic markers. In summary our results suggest that Pgrn is a novel, robust and independent prognostic marker in CLL that can be easily measured by ELISA.

ANGPT2 promoter methylation is strongly associated with gene expression and prognosis in chronic lymphocytic leukemia

Increasing evidence suggests a key role for angiopoietin-2 (ANGPT2) in influencing the aggressiveness of chronic lymphocytic leukemia (CLL). In the presence of vascular endothelial growth factor (VEGF), ANGPT2 causes vessel destabilization leading to neoangiogenesis. Accordingly, high expression levels of ANGPT2 and high degree of angiogenesis have consistently been associated with poor prognosis in CLL; however, the molecular mechanisms behind the variability in ANGPT2 expression are still to be discovered. Here, for the first time, we investigated the DNA methylation status of the ANGPT2 promoter in a large CLL cohort (n = 88) using pyrosequencing and correlated methylation data with ANGPT2 expression levels, prognostic factors and outcome. Importantly, methylation levels of the ANGPT2 gene correlated inversely with its mRNA expression levels (p < 0.001). Moreover, low ANGPT2 methylation status was highly associated with adverse prognostic markers, shorter time to first treatment and overall survival. Finally, treatment with methyl inhibitors induced re-expression of ANGPT2 in two B-cell lymphoma cell lines, underscoring the importance of DNA methylation in regulating transcriptional silencing of this gene. In conclusion, we believe that the known variability in ANGPT2 expression among CLL patients could be explained by differential promoter DNA methylation and that low methylation levels of the ANGPT2 promoter have an adverse prognostic impact in CLL.

Clonal Evolution, Genomic Drivers, and Effects of Therapy in Chronic Lymphocytic Leukemia

The identification of gene mutations and structural genomic aberrations that are critically involved in chronic lymphocytic leukemia (CLL) pathogenesis is still evolving. One may postulate that genomic driver lesions with effects on CLL cell proliferation, apoptosis thresholds, or chemotherapy resistance should increase in frequency over time when measured sequentially in a large CLL cohort. We sequentially sampled a large well-characterized CLL cohort at a mean of 4 years between samplings and measured acquired copy number aberrations (aCNA) and LOH using single-nucleotide polymorphism (SNP) 6.0 array profiling and the mutational state of TP53, NOTCH1, and SF3B1 using Sanger sequencing. The paired analysis included 156 patients, of whom 114 remained untreated and 42 received intercurrent therapies, predominantly potent chemoimmunotherapy, during the sampling interval. We identify a strong effect of intercurrent therapies on the frequency of acquisition of aCNAs in CLL. Importantly, the spectrum of acquired genomic changes was largely similar in patients who did or did not receive intercurrent therapies; therefore, various genomic changes that become part of the dominant clones are often already present in CLL cell populations before therapy. Furthermore, we provide evidence that therapy of CLL with preexisting TP53 mutations results in outgrowth of genomically very complex clones, which dominate at relapse. Using complementary technologies directed at the detection of genomic events that are present in substantial proportions of the clinically relevant CLL disease bulk, we capture aspects of genomic evolution in CLL over time, including increases in the frequency of genomic complexity, specific recurrent aCNAs, and TP53 mutations.