Abstract: Abstract 90 Background: rituximab had dramatically improved the prognosis of patients with Diffuse Large B-cell Lymphoma (DLBCL) in combination with chemotherapy. Many biological and clinical studies suggested considerable inter-individual variability in term of anti-CD20 monoclonal antibody (mAb) activity with tumor and host-related influencing factors. Among host-related factors, the presence of functional polymorphisms in FcG receptors genes as FCGR3A-158V/F influences the affinity for IgG1 and consequently the antibody dependant cellular cytotoxicity (ADCC) with therapeutic mAbs such as rituximab. The clinical consequence reported to date consists in a better response rate to rituximab monotherapy for FCGR3A-158V homozygous patients treated for follicular lymphoma compared FCGR3A-158F carriers. In DLBCL and in the context of combination with chemotherapy, the role of FCGR3A and FCGR2A SNPs on treatment response and patient's outcome is not clear with few prospective studies. The aim of this study is to determine the impact of FCGR3A and FCGR2A SNPs on response and outcome of newly diagnosed DLBCL patients included in the prospective trials of the GELA (LNH2003 program). Patients and Methods: 1564 patients from France, Switzerland and Belgium were included in the 5 prospective multicentric trials of the LNH2003 program of the GELA designed for DLBCL patients who were stratified in different subgroups based on age and International Prognostic Index (IPI) score. A sample of peripheral blood lymphocytes was collected before treatment from 760 patients who signed a specific consent form for this genetic study. After pathologic review and exclusion of patients not receiving rituximab (48 patients), 554 DLBCL patients were available for this study. SNPs were genotyped using a TaqMan® based assay. Results: The median age of the 554 patients was 61 years (range, 18–93 years), 57% of them were male and 50% of patients presented at diagnosis a 2–3 age-adjusted IPI score. Chemotherapy regimen consisted in a combination of rituximab with CHOP-21 (110 patients, 20%), CHOP-14 (181 patients, 33%), low dose CHOP for patients older than 80 years (60 patients, 11%), or ACVBP regimen (203 patients, 36%). At the end of treatment, complete response (CR) or unconfirmed CR was observed in 75% of patients. After a median follow-up of 38 months, the 3-year progression free survival (PFS) and overall survival (OS) was 70.2% and 75.7%, respectively. The distribution of the VV, VF and FF FCGR3A alleles was 14.8%, 46.4%, 38.8%, and 27.8%, 48.6%, 23.6% for HH, HR and RR FCGR2A alleles, respectively, and were therefore consistent with Hardy-Weinberg equilibrium. Initial clinical characteristics of patients (age, sex, Performance Status, stage, B-symptoms, number of extra-nodal sites, LDH level, IPI) were not different according to the two FCGR SNPs. CR/CRu after induction therapy was observed in 61%, 66%, 61% for VV, VF and FF carriers (P = .46) and 60%, 64%, 64% for HH, HR and RR carriers (P =.70), respectively. No difference of response after consolidation treatment was observed between each genotype of FCGR3A and FCGR2A SNPs. The 3-year PFS was 65.3%, 71.4%, 70.5% for FCGR3A VV, VF and FF carriers (P = .43) and 69.2%, 67.6%, 76.6% for FCGR2A HH, HR, RR carriers (P =.09), respectively. The 3-year OS was also not different between the three genotypes of each FCGR SNPs. Conclusions: To our knowledge, this is the largest prospective multicentric study that investigates the role of FCGR2A and FCGR3A SNPs on treatment response and outcome in a large series representing the whole spectrum of DLBCL patients. Based on these results, modification of rituximab schedule according to the FCGR3A and FCGR2A genotypes does not appear worth investigating. Others host-related factors influencing the efficiency of immunotherapy need to be investigate. Disclosures: No relevant conflicts of interest to declare.

Abstract: Abstract 415 Background and aim of the study Genomic gains and losses play a crucial role in the development and progression of DLBCL and are closely related to gene expression profiles (GEP), including the germinal center B-cell like (GCB) and activated B-cell like (ABC) cell of origin (COO) molecular signatures. To identify new oncogenes or tumor suppressor genes (TSG) involved in DLBCL pathogenesis and to determine their prognostic values, an integrated analysis of high-resolution gene expression and copy number profiling was performed. Patients and methods Two hundred and eight adult patients with de novo CD20+ DLBCL enrolled in the prospective multicentric randomized LNH-03 GELA trials (LNH03-1B, -2B, -3B, 39B, -5B, -6B, -7B) with available frozen tumour samples, centralized reviewing and adequate DNA/RNA quality were selected. 116 patients were treated by Rituximab(R)-CHOP/R-miniCHOP and 92 patients were treated by the high dose (R)-ACVBP regimen dedicated to patients younger than 60 years (y) in frontline. Tumour samples were simultaneously analysed by high resolution comparative genomic hybridization (CGH, Agilent, 144K) and gene expression arrays (Affymetrix, U133+2). Minimal common regions (MCR), as defined by segments that affect the same chromosomal region in different cases, were delineated. Gene expression and MCR data sets were merged using Gene expression and dosage integrator algorithm (GEDI, Lenz et al. PNAS 2008) to identify new potential driver genes. Results A total of 1363 recurrent (defined by a penetrance 〉 5%) MCRs within the DLBCL data set, ranging in size from 386 bp, affecting a single gene, to more than 24 Mb were identified by CGH. Of these MCRs, 756 (55%) showed a significant association with gene expression: 396 (59%) gains, 354 (52%) single-copy deletions, and 6 (67%) homozygous deletions. By this integrated approach, in addition to previously reported genes (CDKN2A/2B, PTEN, DLEU2, TNFAIP3, B2M, CD58, TNFRSF14, FOXP1, REL…), several genes targeted by gene copy abnormalities with a dosage effect and potential physiopathological impact were identified, including genes with TSG activity involved in cell cycle (HACE1, CDKN2C) immune response (CD68, CD177, CD70, TNFSF9, IRAK2), DNA integrity (XRCC2, BRCA1, NCOR1, NF1, FHIT) or oncogenic functions (CD79b, PTPRT, MALT1, AUTS2, MCL1, PTTG1…) with distinct distribution according to COO signature. The CDKN2A/2B tumor suppressor locus (9p21) was deleted homozygously in 27% of cases and hemizygously in 9% of cases. Biallelic loss was observed in 49% of ABC DLBCL and in 10% of GCB DLBCL. This deletion was strongly correlated to age and associated to a limited number of additional genetic abnormalities including trisomy 3, 18 and short gains/losses of Chr. 1, 2, 19 regions (FDR 〈 0.01), allowing to identify genes that may have synergistic effects with CDKN2A/2B inactivation. With a median follow-up of 42.9 months, only CDKN2A/2B biallelic deletion strongly correlates (FDR p.value 〈 0.01) to a poor outcome in the entire cohort (4y PFS = 44% [32–61] respectively vs. 74% [66–82] for patients in germline configuration; 4y OS = 53% [39–72] vs 83% [76–90] ). In a Cox proportional hazard prediction of the PFS, CDKN2A/2B deletion remains predictive (HR = 1.9 [1.1–3.2], p = 0.02) when combined with IPI (HR = 2.4 [1.4–4.1] , p = 0.001) and GCB status (HR = 1.3 [0.8–2.3], p = 0.31). This difference remains predictive in the subgroup of patients treated by R-CHOP (4y PFS = 43% [29–63] vs. 66% [55–78], p=0.02), in patients treated by R-ACVBP (4y PFS = 49% [28–84] vs. 83% [74–92], p=0.003), and in GCB (4y PFS = 50% [27–93] vs. 81% [73–90], p=0.02), or ABC/unclassified (5y PFS = 42% [28–61] vs. 67% [55–82] p = 0.009) molecular subtypes (Figure 1). Conclusion We report for the first time an integrated genetic analysis of a large cohort of DLBCL patients included in a prospective multicentric clinical trial program allowing identifying new potential driver genes with pathogenic impact. However CDKN2A/2B deletion constitutes the strongest and unique prognostic factor of chemoresistance to R-CHOP, regardless the COO signature, which is not overcome by a more intensified immunochemotherapy. Patients displaying this frequent genomic abnormality warrant new and dedicated therapeutic approaches. Disclosures: Salles: roche: Consultancy.

Abstract: Background. R-CHOP is the standard first-line treatment for elderly patients with diffuse large B-cell lymphoma (DLBCL). However 30% of patients will relapse and 70% of relapsed patients will die within 2 years of diagnosis. The REMARC study (clinicalTrials.gov NCT01122472) is an international, multicenter, double-blind, randomized, placebo controlled, phase III trial that assessed the benefit of lenalidomide (LEN) maintenance after response to R-CHOP in patients aged 60 to 80 years with untreated DLBCL, FL3b or transformed lymphoma. Patients achieving CR or PR at the end of 6 or 8 cycles of R-CHOP21 or R-CHOP14 were stratified by CR/PR status and country and randomized 1:1 to receive 2 years of LEN maintenance (25 mg/day for 21 of every 28 days) or placebo (PBO). The primary endpoint of the study was progression-free survival (PFS). Secondary endpoints were safety, PR to CR conversion rate, and overall survival (OS). Diagnosis was retrospectively centrally reviewed. In patients with adequate samples, GCB/nonGCB profile was assessed by the Hans algorithm and GCB/ABC/unclassified profile was assessed using NanoString gene expression profiling technology. Methods. From 05/2009 to 05/2014, 784 patients were enrolled either before R-CHOP (n= 437) or after completion of 6 or 8 cycles of R-CHOP (n= 347). At the end of R-CHOP therapy, 650 patients were randomized to maintenance, either in CR (n= 495) or in PR (n= 152). Central review found that 3 patients were randomized in SD or PD, all in LEN arm. At time of diagnosis, median age was 68 y (range 58-80), 43.5% were older than 70 y, and 56% were male. aaIPI was low in 38.5% and high in 57.5% of patients (missing data 4%). COO analyses are ongoing for both Hans algorithm and NanoString technology. Results. With a median follow-up of 40 months, median PFS (according to independent centralized radiology review) was not reached in the LEN group versus 68 months in the PBO group (hazard ratio favoring the LEN group, 0.708 (95% CI 0.537-0.932; p=0.0135))(See Figure). In the LEN group, 18 patients (21%) converted from PR to CR during maintenance compared to 13 patients (14%) in the PBO group. Immature overall survival data did not show any benefit for LEN arm, a lack of difference not attributable to an excess of lymphoma relapse, secondary cancer or safety problems in LEN arm. Deaths generally occurred off study drug (median time from last dose of study drug to death was 277 days (range 20, 1291) in LEN arm and 334 (41, 1594) in control arm. During maintenance, the most common observed grade 3 or 4 AEs were neutropenia (56% vs. 22%), rash (5% vs. 1%), infections (8% vs. 6%), and thrombocytopenia (2.5% vs. 0.6%) in LEN and PBO arms, respectively. Dose adjustments were necessary in 72% of the LEN patients and 42% of PBO patients. 59% of patients stopped LEN and 40% stopped PBO for toxicity (p 〈 0.001). Median number of cycles was 15 in LEN and 25 in PBO (p 〈 0.001). Secondary primary malignancies occurred in 33 patients receiving LEN and in 42 patients on PBO. Conclusion. This analysis of the REMARC study shows that 2 years of LEN maintenance in patients responding to R-CHOP significantly improved PFS (primary endpoint) without an early significant impact on OS. The COO analysis is currently ongoing. This is the first report finding that using an immunomodulatory agent as maintenance therapy prolongs PFS for patients with DLBCL after first line treatment with R-CHOP. Figure 1. Progression-free survival of elderly patients with diffuse large B-cell lymphoma in response to R-CHOP treated in maintenance with either lenalidomide or placebo Figure 1 Figure 1. Disclosures Thieblemont: Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche: Research Funding; Bayer healthcare: Honoraria, Membership on an entity's Board of Directors or advisory committees; Abbvie: Honoraria, Membership on an entity's Board of Directors or advisory committees. Gomez da Silva:Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; ROche: Consultancy, Membership on an entity's Board of Directors or advisory committees; takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bristol Meyer Squibb: Consultancy, Membership on an entity's Board of Directors or advisory committees. Morschhauser:Celgene: Consultancy, Honoraria; Roche: Consultancy, Honoraria; Gilead Sciences: Consultancy, Honoraria; Janssen: Honoraria; Servier: Consultancy, Honoraria. Haioun:Sandoz: Honoraria, Membership on an entity's Board of Directors or advisory committees; Gilead: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees. Cabecadas:celgene: Consultancy, Honoraria. Salles:Gilead: Honoraria, Research Funding; Janssen: Consultancy, Honoraria; Celgene: Consultancy, Honoraria; Mundipharma: Honoraria; Roche/Genentech: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria; Novartis: Consultancy, Honoraria. Coiffier:Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Mundipharma: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Astra-Zeneca: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celltrion: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees.

Abstract: Abstract 3676 Background: Diffuse Large B-cell Lymphoma (DLBCL) is a heterogeneous disease based on biological analyses of tumor cells, with for instance the determination of two molecular subgroups, one with a profile close to germinal center B-cells and the second to activated B-cells. Data from the biology of tumor microenvironment could also be helpful to define different prognostic subgroups of DLBCL. The role of the biology of the patient with DLBCL, i.e. the host-related factors is little explored. Epidemiological studies showed that some inherited genetic variation in immune genes could increase the risk of lymphoma development and some retrospective studies indicated that patient's outcome could be predicted by some germ line polymorphisms (SNPs) in cytokine genes. The aim of this study is to determine the prognostic impact of 13 SNPs in 7 immune genes, IL10 (4 SNPs), Tumor Necrosis factor α (TNFA), lymphotoxin α (LTA), BAFF (3 SNPs), IL1A, IL8RB, IL4R (2 SNPs) in a cohort of newly diagnosed DLBCL patients included in the GELA LNH2003 prospective trials all treated by rituximab combined with chemotherapy. Patients and Methods: 1564 patients from France, Switzerland and Belgium were included in the 5 prospective multicentric trials of the LNH2003 program of the GELA designed for DLBCL patients who were stratified in different subgroups based on age and International Prognostic Index (IPI) score. A sample of peripheral blood lymphocytes was collected before treatment from 760 patients who signed a specific consent form for this genetic study. After pathologic review and exclusion of patients not receiving rituximab (48 patients), 554 DLBCL patients were available for this study. SNPs were genotyped using a TaqMan® based assay. Results: The median age of the 554 patients was 61 years (range, 18–93 years), 57% of them were male and 50% of patients presented at diagnosis a 2–3 age-adjusted IPI score. Chemotherapy regimen consisted in a combination of rituximab with CHOP-21 (110 patients, 20%), CHOP-14 (181 patients, 33%), low dose CHOP for patients older than 80 years (60 patients, 11%), or ACVBP regimen (203 patients, 36%). At the end of treatment, complete response (CR) or unconfirmed CR was observed in 75% of patients. After a median follow-up of 38 months, the 3-year progression free survival (PFS) and overall survival (OS) was 70.2% and 75.7%, respectively. All the polymorphism distributions of the SNPs analyzed were consistent with Hardy-Weinberg equilibrium. The main initial clinical characteristics of patients were not different according to the 13 studied SNPs, except for IL4R (rs2107356), CC carriers presented less frequently B symptoms than CT and TT carriers (28% vs. 41% and 61%, P = .01). No correlation was observed between the quality of the response at the end of the treatment and each genotype of the 13 studied SNPs. The 3-year PFS and OS were overall not influenced by the genotyping of the 13 SNPs. However, a trend for a better 3-year PFS for LTA +252GG carriers compared to LTA +252AG and AA carriers was observed (79.7% vs. 64.6% and 72.7%, P = .04). Conclusions: To our knowledge, this is the largest prospective multicentric study that investigates the role of immune SNPs on treatment response and outcome in a large cohort of newly diagnosed patients with DLBCL receiving rituximab. No correlation between SNPs and treatment response was observed. Analyses of the impact on outcome of each individual SNP showed only a trend for a prognostic role on PFS of LTA A252G SNP, which is already described as a functional SNP. The results will be further precise by the correlation of IL10, BAFF and TNFA /LTA full haplotypes with response and outcome. Disclosures: No relevant conflicts of interest to declare.

Abstract: Anemia is a common feature at diagnosis of patients with lymphoid malignancies and has been previously described as an important prognostic factor (Moullet et al. Ann Oncol 1998). Recent guidelines recommend evaluating iron parameters as part of the initial assessment of cancer associated anemia in order to propose iron therapy for patients with functional or absolute iron deficiency. However, incidence of iron parameters abnormalities and impact on prognosis is largely unknown. Methods The phase III randomized LNH 03-6B protocol included elderly patients from 60 to 80-year with untreated diffuse large B-cell lymphoma and an age-adjusted International Prognostic Index (aaIPI) of 1 or more. Patients were randomized between 8 cycles of R-CHOP given every 2 or 3 weeks (Delarue et al. Lancet Oncol 2013) and between a prophylactic treatment with Darbepoetin alfa in order to maintain hemoglobin level between 13 and 14 g/dL versus an usual management of chemotherapy-induced anemia including ESAs and transfusions. Iron parameters including serum iron, transferrin and ferritin levels and transferrin saturation (TSAT) were measured at screening and at the end of study treatment. Iron deficiency was defined by TSAT below 20% and was considered either as true or functional if ferritin level was below or above normal or value, respectively. Results At diagnosis, median serum iron level was 9 µmol/L (n=358 patients, range: 0.14-62.6), median transferrin level was 2.1 g/L (n=262 patients, range: 0.3-29), median ferritin level was 325 ng/mL (n=321 patients, range: 6-6071) and median TSAT was 17% (n=258 patients, range: 2-57). Among patients with available TSAT data, 163/258 (63%) presented a coefficient lower than 20%. When comparing with the whole study population, patients with TSAT ≤ 20% had similar baseline characteristics including aaIPI. In univariate analysis, there is no difference of PFS (HR: 1.078, 95%CI: 0.734-1.584, p=0.7006) and OS (HR: 0.963, 95%CI: 0.622-1.491, p=0.8664) according to TSAT. By contrast, in univariate and multivariate analysis, patients with ferritin level lower than normal and patients with ferritin level higher than normal presented, compared with patients with normal ferritin level, worse PFS (respectively, HR: 4,973, 95%CI: 1.673-14.780, p=0.039; HR: 1,654, 95%CI: 1.094-2.499, p =0,017) and OS (respectively HR: 6.204, 95%CI: 1.713-22.475, p=0.0055; HR: 1.8, 95%CI: 1.108-2.923, p =0.0175). Conclusion Elderly patients with untreated aggressive lymphoma and iron deficiency as measured by TSAT showed similar characteristics compared to the whole population. TSAT level did not impact prognosis in contrast to ferritin level, whose variations could be independent of iron metabolism. The high frequency of iron deficiency at diagnosis raise the question of the use of IV iron as a frontline treatment of chemotherapy-associated anemia. A prospective, placebo-controlled, phase III trial is planned within our group, whose primary objective will be to demonstrate efficacy of ferric carboxymaltose alone compared to placebo as measured by the diminution of percentage of patients requiring red blood cell transfusion and/or ESA administration. Disclosures: No relevant conflicts of interest to declare.

Abstract: There are some evidences that blood transfusions, either red blood cell (RBC) or platelet, could impact survival of patients affected with various medical conditions. Mechanisms are largely unknown but immune dysfunction, storage duration, cytokines or iron releases could be involved. These concerns led to the publication of revised guidelines with a more restrictive approach for RBC and platelet transfusions in various conditions, mostly surgery and emergency. However, less was known for cancer patients receiving therapy with a curative attempt. Methods The phase III randomized LNH 03-6B protocol included elderly patients from 60 to 80-year with untreated diffuse large B-cell lymphoma and an age-adjusted International Prognostic Index (aaIPI) of 1 or more. Patients were randomized between 8 cycles of R-CHOP given every 2 or 3 weeks (Delarue et al. Lancet Oncol 2013) and between a prophylactic treatment with Darbepoetin alfa in order to maintain hemoglobin level between 13 and 14 g/dL and an usual management of chemotherapy-induced anemia. In both arms, RBC and platelet transfusions were given according to physician decision without predefined threshold, but analysis of current practices showed that most patients were transfused with RBC for grade 3 or more anemia and with platelet for grade 4 thrombocytopenia. Results A total of 602 patients were included in the study and 599 have been analyzed for safety issues. Overall, 236 patients received at least one RBC transfusion (39%) with a higher incidence in the R-CHOP14 group (47% versus 31%, p=0.0001) and 60 patients (10%) received at least one platelet transfusion. Comparing with the whole study population, patients who were transfused presented with more aggressive baseline characteristics including percentage of patients with an aaIPI of 2 or 3 (72% versus 58%). Occurrence of RBC transfusion was associated with a worse outcome regarding progression-free (HR: 0.696, 95%CI: 0.547-0.885, p=0.0031) and overall survival (HR: 0.544, 95%CI: 0.414-0.714, p= 〈 0.0001). Moreover, the number of episode of red blood cell transfusion (0 vs. 1-2 vs. 〉 2) was also associated with PFS and OS. Occurrence of platelet transfusion was also associated with worse PFS (HR: 2.658, 95%CI: 1.922-3.677, p 〈 0.0001) and OS (HR: 3.257, 95%CI: 2.302-4.608, p 〈 0.0001), with an impact of the number of episode (0 vs. 1 vs. 〉 1). In a multivariate analysis including LDH level, ECOG performans status and β2-microglobulin level, RBC and platelet transfusions were both predictive for overall survival (HR: 1.575, 95%CI: 1.158-2.143, p=0.0038 and HR: 2.608, 95% CI: 1.736-3.917, p 〈 0.0001). Finally, no adverse event was associated with transfusion. Conclusion In elderly patients who received a first line immunochemotherapy for aggressive B-cell lymphoma, occurrence of transfusions as well as the number of blood products transfused appear to be significantly and independently associated with lower survival rates. These results suggest a direct effect of transfusion and will prompt us to explore possible mechanistic explanations in animal models and to validate this hypothesis in others cancer populations. Disclosures: No relevant conflicts of interest to declare.

Abstract: Abstract 406 Introduction: In 2000, the Gela demonstrated the survival advantage of adding Rituximab to CHOP21 over CHOP21 in the treatment of diffuse large B-cell lymphoma (DLBCL) in elderly patients. Two consecutive studies from the German group have shown an improvement of survival with CHOP14 compared to CHOP21, and then after with R-CHOP14 compared to CHOP14. Here, we report the results of the planned interim analysis of the LNH03-6B, a multicentric, phase III open-label, randomized trial evaluating the efficacy of R-CHOP given every 14 days compared to R-CHOP given every 21 days, held after the inclusion of the first 202 patients, with a median follow-up of 24 months. Patients and methods: Patients between 60 and 80 years old with DLBCL and aaIPI ≥ 1 were eligible. They were randomized between two immunochemotherapy regimens combining Rituximab and CHOP given every 2 (R-CHOP14, arm A) or 3 weeks (R-CHOP21, arm B) for 8 cycles. They were subsequently randomized between a prophylactic treatment with Darbepoetin alfa and a conventional treatment of chemotherapy-induced anemia. G-CSF was given according to physician decision. The primary objective was to evaluate the efficacy of R-CHOP14 compared to R-CHOP 21 as measured by the EFS, events being defined as death from any cause, relapse for complete responders and unconfirmed complete responders, progression during or after treatment and changes of therapy during allocated treatment. Secondary objectives were OS, PFS, DFS, response rate and analysis of dose-intensity and toxicity. According to previous LNH98-5 protocol, sample size was calculated to demonstrate an improvement of 2-year EFS from 55% to 65% with R-CHOP14. Six-hundred patients, randomized 1:1 between the two treatment groups recruited over 4 years and followed for a minimum of one year, will provide 80% power at the overall 5% (2-sided) significance level to detect the expected difference. Results: In this planned interim analysis, 202 patients were randomized and 201 received study treatment, 103 with R-CHOP14 and 98 with R-CHOP21. Median age was 72 years. Patients' characteristics were similar in both groups with a slightly higher proportion of patients with aaIPI 2-3 in R-CHOP14 arm (67% vs 59%) whereas a higher proportion of patients in R-CHOP21 arm presented with B symptoms (43% vs 37%). The median interval between cycles was 15 days in R-CHOP14 group and 21 days in R-CHOP21 group; 73 patients (71%) in R-CHOP14 group and 74 patients (76%) in R-CHOP21 group completed 8 cycles without progression. In the R-CHOP14 group, the increase of dose-intensity at the end of treatment, calculated according to 3-week interval as a reference, was 125% for cyclophosphamide and doxorubicin. Ninety percent of patients treated with R-CHOP14 received G-CSF, whereas only 66% in R-CHOP21 group. Response rate (CR+CRu) was 67% in R-CHOP14 arm and 75% in R-CHOP21 arm (p=NS). The 2-year EFS was 48% in R-CHOP14 arm compared with 61% in R-CHOP21 (p=NS). A similar trend was observed for 2-year PFS (49% vs 63%), 2-year DFS (57% vs 70%) and 2-year OS (67% vs 70%) (p=NS for all). Grade 3-4 hematological toxicity was more frequent in R-CHOP14 group, with a higher proportion of patients receiving red cell or platelet transfusions and/or experiencing febrile neutropenia, resulting in higher proportion of patients hospitalized for adverse events. In contrast, there was no difference for extra-hematological grade 3-4 toxicities. Conclusions: The results of this interim analysis of the LNH03-6B trial favor treatment with R-CHOP21 in elderly patients with DLBCL, with trends toward higher efficacy and lower toxicity compared to R-CHOP14. These results should be confirmed by the final analysis, concerning the 602 patients included, planned in 2010. Disclosures: No relevant conflicts of interest to declare.

Abstract: Abstract 2723 Introduction: Elderly patients with diffuse large B-cell lymphoma (DLBCL) frequently receive reduced dose-intensity (DI) chemotherapy (Lyman GH et al. J Clin Oncol 2004) but impact on outcome in the era of classical or dose-dense immunochemotherapy is unknown. Final results of the LNH03–6B did not show any difference in PFS (HR: 0.99 [95%CI: 0.78–1.26]; p=0.90) and OS (HR: 0.96 [95%CI: 0.73–1.26] ; p=0.75) between R-CHOP14 and R-CHOP21 (Delarue et al. ASCO 2012). Methods: Patients between 60 and 80 years old with DLBCL and aaIPI≥1 were eligible. They were randomized between R-CHOP14 and R-CHOP21 for 8 cycles. Dose-intensity (DI) was calculated for all patients for cyclophosphamide (CPM), doxorubicin (DOX) and rituximab (RTX) and patients were separated into 4 quartiles for each of these drugs. Outcome (PFS and OS) according to final DI was evaluated for all patients and in R-CHOP14 and R-CHOP21 arms. Results: 602 pts were randomized, 600 were evaluable, 304 with R-CHOP14 and 296 with R-CHOP21. Median age was 70 years. Patient characteristics were similar in both arms. The percentage of cycles administered with G-CSF was 89% in R-CHOP14 and 66% in R-CHOP21. Median interval between 2 cycles was 14 d [9–94] in R-CHOP14 arm and 21 d [15–66] in R-CHOP21 arm. Median dose-intensity was 88% in R-CHOP14 and 97% in R-CHOP21 for CPM, and 88% in R-CHOP14 and 96% for R-CHOP21 for DOX. There was no difference of median dose-intensity for CPM and DOX according to G-CSF use at C1 in R-CHOP14 arm. In the R-CHOP14 arm, the increase of DI at the end of treatment, calculated according to 3-week interval as a reference, was 133% for CPM and DOX. When separating patients in 4 quartiles according to final dose intensity, there was no impact for CPM, DOX and RTX for the entire cohort in term of PFS and OS. Moreover, decreased DI for CPM, DOX and RTX did not impact negatively PFS and OS in patients randomized in the R-CHOP14 arm. On the other hand, for patients included in the R-CHOP21 arm, PFS was negatively impacted by lower DI of CPM (p=0.02), DOX (p=0.007) and RTX (p=0.006). OS was also negatively impacted by lower DI of CPM (p=0.0002), DOX (p 〈 0.00001) and RTX (p 〈 0.0001). In the R-CHOP21 arm, the negative impact was pronouncedly marked when comparing patients in the first and the second quartiles with those in the fourth quartile. Conclusion: While low DI has a major impact on PFS and OS in patients who receive conventional dose immunochemotherapy, consequences for patients receiving dose-dense immunochemotherapy seem less pronounced. As a consequence, immunochemotherapy could be safely decreased for elderly patients who receive R-CHOP14 and experienced adverse events. Disclosures: No relevant conflicts of interest to declare.

Abstract: Even with the introduction of rituximab, some patients with lymphoma continue to relapse or progress during treatment. To better define these patients, we looked at all patients with aggressive lymphoma included in the GELA trials during the last 20 years or 4 generations of studies: LNH-87, -93, -98, and -03. Each study generation comprise several randomized studies according to different groups of patients, i.e., young or old, low or high risk. ACVBP, the high-dose regimen used by GELA since 1984 was usually one of the 2 arms, except in elderly patients. Rituximab was first introduced in the LNH-98.5 study and was part of nearly all arms in the LNH-03 study. A total of 7806 patients were included in this retrospective analysis, 3116 being treated with ACVBP and 4880 with other regimens. Two analyses were done: one for the 7198 patients treated without rituximab and one for the 608 patients treated with rituximab. Only patients included in a published study were included explaining the lower number for rituximab-treated patients, most of the LNH-03 studies being not yet published. 4 groups of patients were defined: refractory or non-responding pts (NR, progression during treatment); patients in PR at the end of treatment (persisting lymphoma cells or PET fixing tumor); relapsing pts with early relapses (ER, during the first year) and late relapses (LR, after one year). 7-year OS was 56% and 7-year PFS was 47.5%, meaning that only 8.5% of the patients were rescued by any treatment at time of progression. Identical results were found in all study generations with 2 groups of patients: those with PR or LR with 7-year OS at 38%, and those with NR or ER, 7-year OS at 12%. Therefore, all generations were grouped (see figure 1). Patients treated with rituximab had a gain of 9% in OS at 6 years (69% vs. 60%). However, they had the same pattern for progression: 6-year OS around 40% for PR and LR patients and around 20% for NR and ER patients. The difference between patients treated with or without rituximab being the percentage of patients in each group: 61% vs. 50% for those without progression and 16% vs. 29% for those with NR or ER. The IPI score does not allow the identification of these poor risk patients. A study is ongoing to characterize the NR+ER patients at diagnosis. This analysis allowed recognizing 2 patterns of failing therapy. Patients with PR and LR have a disease sensible to chemotherapy while NR and ER pts have a disease not responding to treatment. These patients probably must be treated differently with the introduction of new therapeutic agents into the first line regimen. Figure Figure

Abstract: 8021 Background: In 2000, the GELA established a new standard for elderly patients with DLBCL, demonstrating survival advantage of R-CHOP21 over CHOP21. Based on RICOVER-60 results, the DSHNHL proposed R-CHOP14 as a standard. Comparison between both regimens is lacking. We report the results of the final analysis of the randomized phase III trial LNH03-6B, with a median follow-up of 56 months. Methods: Pts between 60 and 80 years old with DLBCL and aaIPI≥1 were eligible. They were randomized between R-CHOP14 and R-CHOP21 for 8 cycles. G-CSF prophylaxis was given according to physician decision. Primary objective was to evaluate the efficacy of R-CHOP14 compared to R-CHOP21 as measured by the EFS. Results: 602 pts were randomized, 600 were evaluable, 304 with R-CHOP14 and 296 with R-CHOP21. Median age was 70 years. Pts characteristics were similar between the two arms. Percentage of pts with baseline IPI3-5 was 72% in R-CHOP14 arm and 78% in R-CHOP21 arm. Median interval between 2 cycles was 14 d in R-CHOP14 arm and 21 d in R-CHOP21 arm. In R-CHOP14 arms, 89% of cycles were administered with G-CSF. Median dose-intensity for R-CHOP14 arm was 88% for cyclophosphamide and doxorubicin. There was no difference in median dose-intensity according to G-CSF administration at first cycle. Response rate (CR+CRu) was 71% in R-CHOP14 arm and 74% in R-CHOP21 arm (p=0.42). The 3-y EFS was 56% in R-CHOP14 arm and 60% in R-CHOP21 (HR 1.04; CI95% 0.82-1.31; p=0.79). Moreover, there is no difference between both arms regarding 3-y PFS (60% vs. 62%; HR 0.99; CI95% 0.78-1.26; p=0.90), DFS (72% vs. 67%; HR 0.80; CI95% 0.58-1.10; p=0.80) and OS (69% vs. 72%; HR 0.96; CI95% 0.73-1.26; p=0.75). Finally, percentage of patients with at least one serious adverse event (R-CHOP14: 51%; R-CHOP21: 47%) and rate of toxic death (4.6% and 4.7% respectively) were similar. Conclusions: Results of the final analysis of LNH03-6B demonstrate similar efficacy and safety profile between R-CHOP14 and R-CHOP21.