Search

Immune thrombocytopenia (ITP) is an autoimmune disease with a complex pathogenesis. As in many B cell–related autoimmune diseases, rituximab (RTX) has been shown to increase platelet counts in some ITP patients. From an immunologic standpoint, the mode of action of RTX and the reasons underlying its limited efficacy have yet to be elucidated. Because splenectomy is a cornerstone treatment of ITP, the immune effect of RTX on this major secondary lymphoid organ was investigated in 18 spleens removed from ITP patients who were treated or not with RTX. Spleens from ITP individuals had follicular hyperplasia consistent with secondary follicles. RTX therapy resulted in complete B-cell depletion in the blood and a significant reduction in splenic B cells, but these patients did not achieve remission. Moreover, whereas the percentage of circulating regulatory T cells (Tregs) was similar to that in controls, splenic Tregs were reduced in ITP patients. Interestingly, the ratio of proinflammatory Th1 cells to suppressive Tregs was increased in the spleens of patients who failed RTX therapy. These results indicate that although B cells are involved in ITP pathogenesis, RTX-induced total B-cell depletion is not correlated with its therapeutic effects, which suggests additional immune-mediated mechanisms of action of this drug.

The outcome of classical Hodgkin lymphoma (cHL) patients may be related to the tumor microenvironment, which in turn may be influenced by Epstein-Barr virus (EBV) infection. To characterize the cHL microenvironment, a set of 63 cHL tissue samples was profiled using DNA microarrays. Their gene expression profile differed from that of histiocyte T cell–rich B-cell lymphoma (H/TCRBCL) samples that were used as controls, mainly due to high expression of PDCD1/PD-1 in H/TCRBCL. EBV+ cHL tissues could be distinguished from EBV− samples by a gene signature characteristic of Th1 and antiviral responses. Samples from cHL patients with favorable outcome overexpressed genes specific for B cells and genes involved in apoptotic pathways. An independent set of 146 cHL samples was analyzed using immunohistochemistry. It showed a significant adverse value in case of high percentage of either TIA-1+-reactive cells or topoisomerase-2+ tumor cells, whereas high numbers of BCL11A+, FOXP3+, or CD20+ reactive cells had a favorable influence. Our results suggest an antitumoral role for B cells in the cHL microenvironment and a stronger stromal influence of the PD1 pathway in H/TCRBCL than cHL. The observation of Th1/ antiviral response in EBV+ cHL tissues provides a basis for novel treatment strategies.

In the WHO classification, subcutaneous panniculitis-like T-cell lymphoma (SPTL) is defined as a distinct type of T-cell lymphoma with an aggressive clinical behavior. Recent studies suggest that distinction should be made between SPTL with an α/β T-cell phenotype (SPTL-AB) and SPTL with a γδ T-cell phenotype (SPTL-GD), but studies are limited. To better define their clinicopathologic features, immunophenotype, treatment, and survival, 63 SPTL-ABs and 20 SPTL-GDs were studied at a workshop of the EORTC Cutaneous Lymphoma Group. SPTL-ABs were generally confined to the subcutis, had a CD4-, CD8+, CD56-, βF1+ phenotype, were uncommonly associated with a hemophagocytic syndrome (HPS; 17%), and had a favorable prognosis (5-year overall survival [OS]: 82%). SPTL-AB patients without HPS had a significantly better survival than patients with HPS (5-year OS: 91% vs 46%; P < .001). SPTLGDs often showed (epi)dermal involvement and/or ulceration, a CD4-, CD8-, CD56 +/-, βF1- T-cell phenotype, and poor prognosis (5-year OS: 11%), irrespective of the presence of HPS or type of treatment. These results indicate that SPTL-AB and SPTL-GD are distinct entities, and justify that the term SPTL should further be used only for SPTL-AB. SPTL-ABs without associated HPS have an excellent prognosis, and multiagent chemotherapy as first choice of treatment should be questioned. © 2008 by The American Society of Hematology.

Bcl-2 protein expression has been associated with poor prognosis in patients with noncutaneous diffuse large B-cell lymphomas. In primary cutaneous large B-cell lymphomas, the location on the leg, the round-cell morphology defined as the predominance of centroblasts and immunoblasts over large centrocytes, and multiple skin lesions were identified as adverse prognostic factors. The prognostic value of bcl-2 protein expression has not been studied in large series of patients. We evaluated 80 primary cutaneous large B-cell lymphomas collected by the French Study Group on Cutaneous Lymphomas. The prognostic value of age, sex, number of lesions, cutaneous extent, location, serum lactate dehydrogenase (LDH) level, B symptoms, morphology, and bcl-2 protein expression was studied. The overall 5-year specific survival rate was 65%. In univariate analysis, advanced age, multiple skin lesions (n = 48), location on the leg (n = 25), round-cell morphology (n = 32), and bcl-2 expression (n = 39) were significantly related to death from lymphoma. In multivariate analysis, bcl-2 expression ( P = .0003), multiple skin lesions ( P = .004), and age remained independent prognostic factors. The 5-year specific survival rates in bcl-2–positive and bcl-2–negative patients were 41% and 89%, respectively ( P < .0001). A new prognostic classification of primary cutaneous B-cell lymphoma should be based primarily on bcl-2 protein expression rather than the location of skin lesions.

Introduction Primary bone lymphoma (PBL) is an uncommon form of lymphoma. Diffuse large B-cell lymphoma (DLBCL) represents the main histology associated with PBL. Clinicopathological understanding and management of PBL rely only on retrospective series. Methods Cases of DLBCL diagnosed by bone biopsy treated in one referring center between 1993 and 2014 were retrospectively reviewed (Montreal, Canada). Bone biopsies done as part of the bone marrow analysis were excluded. Patients files were analyzed to determine if patient had primary bone lymphoma or systemic lymphoma with bony involvement. We excluded patient with distant lymph node or other extranodal involvement. Data on clinical presentation, staging procedures and treatment management including use of radiotherapy was collected. Staging was performed according to WHO classification of soft tissue tumors (2002). Survival time and time to recurrence were calculated from the date of the first documented treatment until recurrence or death with the Kaplan-Meier method. Prognostic factors of recurrence or death were explored with log-rank tests and Cox proportional hazards models. Competing risks analysis was attempted to isolate deaths from lymphoma and death from other causes. Our study was approved by local research and ethic committees. Results We retrieved 42 cases of PBL with a median age of 63 years (23-83) treated between October 1995 and April 2014. We identified 3/14 (21%) GCB and 11/14 (79%) non-GCB subtypes based on the modified Hans algorithm (Meyer et al., JCO 2011). The most common presenting symptom was pain (88%). 12/42 (33%) patients had an IPI score ≥ 3. We identified 18 (43%) stage I, 11 (26%) stage II and 13 (31%) stage IV. 20/42 (48%) had bulky disease (≥ 10cm). Among the 37 patients treated with curative intent, 36 (97%) received CHOP-based regimen and 23 (62%) received rituximab. 30 of these 37 (81%) patients received additional radiotherapy of which 67% received a dose of radiotherapy between 36-50 Gy. Overall response rate for patients treated with curative intent was 86%. With a median follow up of 64.8 months for the whole cohort, the 5- and 10-year overall survival was 73% and 54%, respectively. The 5- and 10-year progression-free survival was 70% and 49% respectively. Age, LDH, stage (I-II vs IV), ECOG (0-1 vs 2-4), IPI (0-2 vs 3-5), use of radiotherapy or addition of rituximab were associated with differences in survival and progression rates that did not reach statistical significance given the limited number of patients in our cohort and the fact that half of the deaths were attributed to other causes. A larger cohort would be required to demonstrate a benefit with rituximab. Response was based on positron emission tomography-computed tomography (TEP-CT) imaging in 12 patients. There was an insufficient number of patients to evaluate the role of adding radiotherapy in patient complete remission defined by PET-CT. Conclusion Our survival rates reproduce published data from series of PBL (Ramadan et al., Ann Oncol 2007; Bruno Ventre et al., Oncologist 2014), although the benefit of adding rituximab did not reach statistical significance in our cohort. The role of radiotherapy in the era of response defined by PET-CT remains to be defined. Disclosures Fleury: Gilead: Consultancy, Speakers Bureau; Amgen: Consultancy, Speakers Bureau; Roche: Consultancy, Speakers Bureau; Novartis: Consultancy, Speakers Bureau; Lundbeck: Consultancy, Speakers Bureau; Seattle Genetics: Consultancy, Speakers Bureau.

Non-Hodgkin's lymphomas are divided into B- and T-cell neoplasms. The existence and the clinical relevance of lymphomas derived from the third lymphocyte lineage, ie, natural killer (NK) cells are still controversial. NK cells are lymphocytes that mediate cytotoxicity without prior sensitization. NK cells also have phenotypic and genotypic characteristics: they express the NK-related antigen CD56, T- cell markers such as CD2 and CD7, but do not express CD5 and T-cell receptor (TCR) proteins, and their TCR locus is not rearranged. Therefore, if NK cell lymphomas exist, they should express some T-cell markers, but not alpha beta or gamma delta TCR proteins. Such lymphomas are actually called TCR silent peripheral T cell lymphomas (PTCL). To detect and characterize NK cell lymphomas, we investigated the immunophenotype and immunogenotype of 35 patients with TCR silent PTCL. The first group included 16 patients with a lymphoma of CD5-CD56+ phenotype, which is identical to normal NK cells. These patients had either a nasal/nasopharyngeal lymphoma (11 cases) or a lymphoma with predominant non-nasal/non-nodal initial involvement (five cases). Eight of the nine cases for which immunogenotypic data were available lacked clonal rearrangement of the TCR gamma genes. Thus, these tumors are likely to be NK cell lymphomas. The second group of 15 cases had a CD5+ phenotype (14 were CD56-, and 1 was CD56+) and clonal rearrangement of TCR gamma genes, indicating that they were true PTCL with unproductive TCR rearrangement. The four remaining cases were CD5- CD56- lymphomas and disclosed either a clonal (two cases) or no clonal (two cases) rearrangements of the TCR gamma genes. Altogether these findings show that CD5-CD56+ so-called “TCR silent PTCL” bear the immunophenotype and immunogenotype of normal NK cells and display peculiar clinical features distinct from true PTCL.

Introduction Diffuse Large B Cell Lymphoma (DLBCL) is the most common lymphoid malignancy, accounting for 30-40% of all Non Hodgkin Lymphomas. Gene expression profiling (GEP) has identified three main subtypes of DLBCL: Germinal Center B-cell like (GCB), Activated B-Cell like (ABC) and Primary Mediastinal B-cell Lymphoma (PMBL). Recently, Next Generation Sequencing (NGS) has enabled a more detailed characterization of DLBCL mutational profiles. Conventional techniques such as immunohistochemistry (IHC) and FISH are also widely used to describe DLBCL. However, no study has yet performed an integrative analysis of the mutational, gene expression, IHC and FISH profiles of DLBCL, in order to provide a comprehensive view of this disease. Methods 215 patients with de novo DLBCL in the prospective, multicenter and randomized LNH-03B clinical trials led by the LYmphoma Study Association (LYSA) were included in this study. Microarray-based GEP identified 81 ABC, 83 GCB, 18 PMBL and 33 other. Mutational profiles of patients' tumor DNA were established using Lymphopanel NGS, designed to identify mutations in 34 genes important for lymphomagenesis. For each recurrently mutated gene, we applied ROMER (Ritchie, Nucleic Acids Res, 2015) to perform gene set enrichment analysis on differential expression profiles of mutant and wild-type patients, using a multifactorial model accounting for subtype. The gene sets were obtained from the MSIGDB Hallmarks (Subramanian A, PNAS, 2005) and Signaturedb (Schaffer, Immunol Rev, 2006) collections. When possible, IHC was performed for IgM (n=150), MYC (n=140), BCL2 (n=148), BCL6 (n=146), CD10 (n=152), FOXP1 (n=147) and MUM1 (n=152); FISH was performed for MYC (n=131), BCL2 (n=133) and BCL6 (n=131). Results As expected, EZH2 mutations were significantly associated with upregulation of GCB gene expression (p An association between BCL6 translocations and ABC subtype was confirmed, via a correlation with ABC-enriched CD79B mutations (p=0.02), although interestingly not with MYD88 mutations. MYD88 mutations were correlated with an upregulation of genes involved in proliferation or repressed by PRDM1 (FDR=0.04 each), as well as with an upregulation of genes involved in checkpoint controls, such as E2F targets and genes involved in DNA repair (FDR=0.03 each). All MYD88 mutants expressed FOXP1 in IHC (p PMBL-enriched mutations in our cohort were frequently associated with IgM and FoxP1 negative IHC, as expected (WHO 2008 and Roschewski, Nat Rev Clin Onc). XPO1 and ITPKB mutations were correlated with JAK-STAT pathway activation in the total cohort, including upregulation of interferon-inducible genes for both gene mutations (FDR=0.02 and FDR=0.08 respectively) and upregulation of BCL6-repressed genes for XPO1 mutations only (FDR=0.02). Interestingly, CD58 mutations were significantly correlated with upregulation of Nfkb pathway target genes (FDR=0.06), perhaps due to their negative impact on CD2 activation and ROS production inhibition. Conclusion The results herein provide steps toward a comprehensive, multi-level overview of DLBCL. We highlight differential gene set expression linked to gene mutation status, as well as driver translocation-associated mutational profiles. By using an integrative analysis approach, this study broadens our understanding of DLBCL subtypes' diverse genetic backgrounds. Disclosures Briere: St. Louis Hospital, Paris, France: Employment. Salles:Celgene Corporation; Roche: Speakers Bureau; Calistoga Pharmaceuticals, Inc.; Celgene Corporation; Genentech, Inc.; Janssen Pharmaceutica Products, L.P.; Roche: Consultancy; Celgene Corporation; Roche and Gilead Sciences: Research Funding.

Peripheral T-cell lymphomas (PTCLs) are rare and have a dismal prognosis. The most frequent subtype is PTCL, unspecified. Epstein-Barr virus (EBV) has been detected in around 40% of cases, but its prognostic significance is not fully established. Lymph node samples from 110 patients with PTCL, unspecified included in LNH87 and LNH93 trials were available. EBV status was studied by EBV-encoded small RNA in situ hybridization (EBER-ISH). EBER-ISH showed positive cells in 45 (41%) of 110 patients. Pretreatment characteristics were comparable between positive and negative cases, except for male sex (80% versus 60%, respectively, P = .02). Only 50% of patients achieved complete remission with a 5-year event-free survival (EFS) and overall survival (OS) of 21% and 30%, respectively. EBER-ISH positivity was the sole factor linked with worse EFS, with a 5-year probability of 11% for positive patients. In univariate analysis, factors affecting OS were EBER-ISH positivity, high LDH level, and age older than 60 years. In multivariate analysis, EBER-ISH was associated with a worse OS in the elderly population. Time-dependent analysis showed that the negative impact of EBV was essentially seen in the first 2 years following diagnosis. These results warrant further studies regarding pathogenesis and specific treatment approaches for EBV-associated PTCL patients.

Although mycosis fungoides (MF) is considered to be an indolent lymphoma, survival is highly influenced by TNM stage. At diagnosis, most MF patients present with early stage disease and a high probability of long-term survival. Treatment is generally directed towards skin lesions, and achievement and duration of complete responses are variable. A dominant T-cell clone is detectable in the cutaneous lesions of 60% of patients. The aim of this study was to determine whether the presence of a T-cell clonal population influences the clinical course of the disease after topical therapy. Cutaneous biopsies from 68 patients were histologically diagnosed as MF and T-cell clonality was analyzed by in vitro amplification of TCR-γ chain gene rearrangements (polymerase chain reaction γ [PCRγ]). After a median follow-up of 48 months, response to treatment was clinically assessed. Age, sex, duration of symptoms before diagnosis, type of cutaneous lesions (T stage), TNM stage, and PCRγ were evaluated as predictive factors of response to treatment in univariate and multivariate analyses. Univariate analysis demonstrated that T1 cutaneous lesions (P = .05) and PCRγ negativity (P = .007) were associated with a higher complete remission rate. Using multivariate analysis, T stage (relative risk, 3.13; P = .06) and PCRγ (relative risk, 4.4; P = .01) remained independent significant predictive parameters of response. In conclusion, T stage and cutaneous PCRγ at diagnosis are the two predictive parameters of treatment response for MF. Therefore, the cutaneous PCRγ findings should be considered in the analysis of future therapeutic trials. © 1998 by The American Society of Hematology.

Anaplastic, CD30+, large-cell lymphoma is now a well-recognized pathologic entity that accounts for 2% to 8% of all lymphomas. Recent progress has been made in the understanding of certain biologic features found in anaplastic large-cell lymphoma, but information about its clinical behavior, in comparison to other large-cell lymphomas, is limited. The pathologic review of a large multicenter study of the treatment of aggressive lymphoma identified 146 cases of anaplastic large-cell lymphoma (ALCL) on the basis of morphology and CD30 expression. We compared initial presentation, immunophenotype, and clinical outcome of these cases with those of the 1,695 nonanaplastic diffuse large-cell lymphomas (non-ALCL) included in the same trial. Patients with ALCL were more likely to be male (P = .018) and were younger (P < .0001) than those with non-ALCL. B symptoms were more frequent in ALCL (P = .006). Skin (P < .0001) and lung (P < .05) involvement was also more frequent in ALCL, but frequency of bone marrow involvement was identical (P = .5). Tumor cell phenotype was B in 56 cases (38%), T in 49 cases (34%), and null in 33 cases (22%). Response to chemotherapy (P = .001), event-free survival (P = .006), and overall survival (P = .0004) were better for ALCL than for non-ALCL. Multivariate analyses identified anaplastic character as an independent factor that predicted a longer survival. Tumor cell phenotype did not influence event-free survival (P = .72) or overall survival (P = .83). ALCL in adults is a clinicopathologic entity which, independent of its phenotypic characteristics, has a better outcome than other diffuse large-cell lymphomas.

Abstract 541 Introduction Diffuse large B-cell lymphomas (DLBCL) represent a heterogeneous group of tumors with variable clinical, morphological, and molecular features. The International Prognostic index (IPI) imperfectly stratifies patients and does not reflect the molecular heterogeneity of the disease. New tools for stratifying DLBCL patients in routine practice are needed. The aim of the study was to analyse the prognostic relevance of CMYC, BCL2 and BCL6 rearrangements and the correlation with CMYC protein expression in a large series of 776 CD20+ de novo DLBCL treated with R-chemotherapy included in the Groupe d'Etude des Lymphomes de l'Adulte (GELA) trials. Methods Patients with de novo CD20+ DLBCL were enrolled in the randomized LNH-03 GELA trials (LNH-03-1B, -2B, -3B, 39B, -6B, 7B-, 5B). Seven hundred seventy six patients treated with anthracyclin based chemotherapy (R-CHOP/R-miniCHOP=483 or R-ACVBP=293) were eligible for the study. Tumor samples were analyzed by Fluorescence in situ hybridization (FISH) using split signal FISH DNA probes for BCL2/18q21, BCL6/8q27 and CMYC/8q24 and were immunostained for the expression of CMYC protein on tissue microarrays. Results Five hundred seventy six DLBCL samples were evaluable for MYC rearrangement. Rearrangement of MYC, BCL2, and BCL6 genes were observed in 9.2% (53/576), 15.9 % (82/515), and 23.8% (129/541) of DLBCL, respectively. Fifteen cases showed both BCL2/BCL6 gene rearrangement. MYC rearranged cases (MYC-R+) included 21 MYC simple-hit (MYC-SH) and 32 MYC double or triple hit with concurrent BCL2 and/or BCL6 rearrangement (MYC-DH) DLBCL. There were 33 male and 20 women, with a median age of 62 years [range, 18–93], and 5 were stage I-II and 48 stage III-IV according to Ann Arbor. The common histological feature of the MYC-SH cases was an increased mitotic rate but a starry sky pattern with tingible body macrophages was observed only in 2 cases. MYC protein expression was found positive in 30/45 (66%) evaluable MYC-R+ cases using a cut off of ≥70% of positive tumor cells. In univariate analysis, IPI had a significant impact on OS (P< .0001). BCL2, BCL6 gene alterations did not significantly predict survival (PFS, EFS, OS). MYC-R+ breakpoint was associated with a shorter 5-year OS rate (57.7% vs 73.7%, P= 0.01) which remained significant in the R-CHOP treated group of patients (P=0.04). When separating MYC-SH and MYC-DH DLBCL, MYC-SH maintained a significant impact on survival in the global population as well as in the R-CHOP group of patients. By contrast, MYC-DH had no significant impact on OS in the global population and in the R-CHOP group of patients. In multivariate analysis, the prognostic impact of the IPI, MYC-R+ and MYC-SH breakpoint remained significant. Conclusion This is the first large extensive study on the prognosis impact of MYC rearrangement in controlled trial of DLBCL patients treated with R-chemotherapy. MYC-SH had an adverse prognostic effect on survival. Interestingly, in contrast to previous published studies, this impact was not maintained in the MYC-DH DLBCL suggesting different biological response to treatment between the two groups. Immunostaining for MYC protein was not efficient to detect MYC rearranged cases. FISH for MYC, BCL2 and BCL6 should be performed systematically to differentiate MYC-SH and MYC-DH in future clinical trials. Disclosures: No relevant conflicts of interest to declare.

Abstract 1607 Background: Lymphoid malignancies derived from T and NK cells (PTCLs) constitute a heterogeneous group of uncommon disease entities, with marked geographic variation in their distribution. The most recent data from the International T-cell Lymphoma Project based on a retrospective analysis of PTCLs diagnosed between 1990 and 2002 (Blood. 2011;117(25): 6756–67) indicate that PTCL, not otherwise specified (PTCL,NOS) represents the most common PTCL worldwide (25,9%), followed by angioimmunoblastic T-cell lymphoma (AITL) (18,5%). Over the last few years, a better characterization of the cellular origin and pathophysiology of PTCLs has led to the development of new diagnostic markers. Aim of the study: To characterize the epidemiology of mature T/NK-cell malignancies in Western Europe (France and bordering countries) and to examine whether the availability of new tools/antibodies and changing concepts over the past years might have translated into an apparent change in the relative distribution of PTCL entities. Materials and methods: The histopathological diagnosis of PTCL entities according to the 2008 WHO classification were collected through two independent sets of PTCLs in France and bordering countries. Results: Over the past two years (2010–2011), 933 newly diagnosed non-cutaneous PTCLs were reviewed in reference centers through the prospective network “Lymphopath” aiming to review any newly diagnosed lymphoma in France. According to the 2008 WHO classification, the 933 PTCLs comprised: 314 AITL (33,6%), 239 PTCL,NOS (25,6%), 78 ALK-positive anaplastic large cell lymphoma (ALCL) (8,3%), 72 ALK-negative ALCL (7,7%), 59 extranodal NK/T-cell lymphomas (ENKTL 6%), 33 enteropathy-associated T-cell lymphoma (4%), 32 HTLV1+ adult T leukemia/lymphoma (3%), 7 hepatosplenic T-cell lymphoma (1%) and 99 cases of other entities or unclassifiable (11%). A high prevalence of AITL was also found in an independent set of PTCLs retrospectively collected in the framework of a multicentric T-cell lymphoma research consortium “Tenomic” where non-cutaneous PTCL with frozen material available (n=623) from 1999 to 2009 in France and Belgium were retrieved and collegially reviewed for consensus diagnosis. In this collection, AITL (n=288, 46%) also outnumbered PTCL, NOS (n=215, 35%). Of the 196 AITL cases extensively investigated for CD10, TFH markers (PD1, CXCL13), CD21/CD23 follicular dendritic cells (FDC) and EBV, the initial diagnosis was recorded in 178 cases as: AITL in 155 cases (87%), PTCL, NOS in 21 cases (12%), and intermediate between PTCL,NOS and AITL (PTCL,NOS/AITL) in 2 cases, indicating the impact of additional stainings for the diagnosis of AITL. The 107 PTCL,NOS cases also extensively immunostained included 9 follicular variant of PTCL,NOS, 8 PTCL,NOS/AITL cases, 5 cases intermediate between PTCL,NOS and ALK-negative ALCL (of which 2 had been diagnosed as such and two as PTCL,NOS), and 85 remaining cases truly unspecified. Of these, 60 had been initially diagnosed as PTCL,NOS; 2 as PTCL,NOS/AITL; 4 as ALK-negative ALCLs and 1 as ENKTL. Conclusion: This study based on two independent large cohorts of non-cutaneous PTCLs highlights AITL as the most prevalent entity in Western Europe. It shows that extensive studies including investigation for CD10, TFH markers, FDC and EBV can at least partly contribute to the reclassification of some PTCL, NOS into the AITL spectrum category. Disclosures: No relevant conflicts of interest to declare.

Abstract 1539 Introduction Diffuse large B-cell lymphomas (DLBCL) represent a heterogeneous disease with variable clinical outcome. Identifying phenotypic biomarkers of tumor cells on paraffin sections that predict different clinical outcome remain an important goal that may also help to better understand the biology of this lymphoma. Differentiating non-germinal centre B-cell-like (non-GCB) from Germinal Centre B-cell-like (GCB) DLBCL according to Hans algorithm has been considered as an important immunohistochemical biomarker with prognostic value among patients treated with R-CHOP although not reproducibly found by all groups. Gene expression studies have also shown that IgM expression might be used as a surrogate for the GCB and ABC subtypes with a strong preferential expression of IgM in ABC DLBCL subtype. ImmunoFISH index based on the differential expression of MUM-1, FOXP1 by immunohistochemistry and on the BCL6 rearrangement by FISH has been previously reported (C Copie-Bergman, J Clin Oncol. 2009;27:5573-9) as prognostic in an homogeneous series of DLBCL treated with R-CHOP. In addition, oncogenic MYC protein overexpression by immunohistochemistry may represent an easy tool to identify the consequences of MYC deregulation in DLBCL. Our aim was to analyse by immunohistochemistry the prognostic relevance of MYC, IgM, GCB/nonGCB subtype and ImmunoFISH index in a large series of de novo DLBCL treated with Rituximab (R)-chemotherapy (anthracyclin based) included in the 2003 program of the Groupe d'Etude des Lymphomes de l'Adulte (GELA) trials. Methods The 2003 program included patients with de novo CD20+ DLBCL enrolled in 6 different LNH-03 GELA trials (LNH-03-1B, −B, −3B, 39B, −6B, 7B) stratifying patients according to age and age-adjusted IPI. Tumor samples were analyzed by immunohistochemistry using CD10, BCL6, MUM1, FOXP1 (according to Barrans threshold), MYC, IgM antibodies on tissue microarrays and by FISH using BCL6 split signal DNA probes. Considering evaluable Hans score, 670 patients were included in the study with 237 (35.4%) receiving intensive R-ACVBP regimen and 433 (64.6%) R-CHOP/R-mini-CHOP. Results 304 (45.4%) DLBCL were classified as GCB and 366 (54.6%) as non-GCB according to Hans algorithm. 337/567 cases (59.4%) were positive for the ImmunoFISH index (i.e. two out of the three markers positive: MUM1 protein positive, FOXP1 protein Variable or Strong, BCL6 rearrangement). Immunofish index was preferentially positive in the non-GCB subtype (81.3%) compared to the GCB subtype (31.2%), (p0.4) for both cut-offs. Progression-free Survival (PFS) was significantly worse among patients with high IPI score (p Overall Survival (OS) was also significantly worse among patients with high IPI score (p All significant parameters were included in a multivariate analysis using Cox Model and in addition to IPI, only the GCB/non-GCB subtype according to Hans algorithm predicted significantly a worse PFS among non-GCB subgroup (HR 1.9 [1.3—2.8] p=0.002) as well as a worse OS (HR 2.0 [1.3–3.2], p=0.003). This strong prognostic value of non-GCB subtyping was confirmed considering only patients treated with R- CHOP for PFS (HR 2.1 [1.4–3.3], p=0.001) and for OS (HR 2.3 [1.3–3.8], p=0.002). Conclusion Our study on a large series of patients included in trials confirmed the relevance of immunohistochemistry as a useful tool to identify significant prognostic biomarkers for clinical use. We show here that IgM and MYC might be useful prognostic biomarkers. In addition, we confirmed in this series the prognostic value of the ImmunoFISH index. Above all, we fully validated the strong and independent prognostic value of the Hans algorithm, daily used by the pathologists to subtype DLBCL. Disclosures: No relevant conflicts of interest to declare.

Abstract 2679 Extranodal nasal type NK/T-cell lymphoma (ENKTL) is a rare and severe disease, more frequent in Asia and South America. Previous studies showed that the 5-year overall survival (OS) rate was less than 40%, with a progression occurrence within 2 years. The early use of radiotherapy sequentially may cure approximately half of the patients with localized disease. In most patients with advanced disease (stage III/IV), the clinical course is highly aggressive with frequent chemotherapy resistance and a poor prognosis. CHOP-based therapy is frequently used and was associated with poor outcome. Considering the rarity of this lymphoma in Europe, we conducted this retrospective study that represents one of the largest studies of NK/T cell lymphoma 'nasal type'. Between January 1989 and September 2010, 36 patients were recruited with a median age of 49 years (22 to 80). 80% the patients was classified as upper aerodigestive tract NK/T-cell lymphoma (UNKTL) and 20% as extra-upper aerodigestive tract NK/T-cell lymphoma (EUNKTL). 72% of UNKTL patients had a local invasiveness superior to T3. 10 (28%) patients were in stage I, 6 (17%) in stage II and 20 (55%) in stage IV. Most frequent symptoms at diagnosis were obstructive symptomatology, epistaxis or cervical lymphadenopathy. 19 (53%) were in low risk versus 17 (47%) in high risk according to the Korean prognostic index. Chemotherapy (CT) alone was the primary treatment for advanced stage (84% CT and 10% CT+ radiotherapy (RT)) with 82% of patients receiving anthracyclin-based regimen. For the early stage, 50% of patients have received the combination CT plus RT and 50% CT. In the CT group, there was no statistical difference (p=0.77) in the percentage of CR according to the type of regimen. Complete response (CR) was observed in 48% of patient while the other 52% were in less than partial response (PR) or in progressive disease (PD) at the end of treatment. The CR rate was improved with the combination CT+RT (90%) versus 33% with CT alone (p The early use of radiotherapy concomitantly or sequentially with chemotherapy improves the outcome of patients, especially in localized disease but still with aggressive forms in this study. We confirm than quality of initial response is the most important factor for the 5 years overall survival. We think Korean T/NK lymphoma prognostic index to be probably more interesting than IPI score to distinguish high risk group and could be the most appropriate prognostic score for NK/T lymphoma nasal type. Finally, the use of agent like asparaginase in first line should be considered for patients with disseminated disease; this has to be evaluated in prospective trials in order to improve the survival in this disease. Disclosures: Salles: Roche and/or Genetech: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen-Cilag: Honoraria, Membership on an entity's Board of Directors or advisory committees; Genzyme: Membership on an entity's Board of Directors or advisory committees; Calistoga/Gilead: Membership on an entity's Board of Directors or advisory committees.

Abstract 2687 Classical Hodgkin lymphomas (cHL) are B lympho-proliferative disorders that can be cured in 80% of cases. The goal of current research is to identify, at diagnosis, the 20% patients, who are likely to relapse or be refractory to treatment. As cHL have an extensive micro-environment implicated in tumour growth, the prognostic value (disease-free survival) of 2 important subsets of lymphoid cells (CD45RO+ effector memory T cells and FOXP3+ regulatory T cells) of the micro-environment has been evaluated using Tissue Micro Array (TMA). Patients (n=96) treated and followed for cHL between 1998 and 2005 were included in this study. Six TMA with 3 spots per case, were built (3 formalin and 3 Bouin fixative). Immunostaining with anti-CD45RO (Dako, UCHL1, 1/400) and anti-FOXP3 (Abcam, 236A/E7 1/100) antibodies was performed on the Benchmark Ultra (Ventana). After scanning of the slides, the staining was quantified automatically using a function (pixel count) of Scanscope software (Aperio). ROC curves were used in order to calculate the best threshold for each marker, depending on the number and intensity of pixels. Survival curves were built according to these thresholds using the Kaplan Meier method and compared with the log-rank test. Then, a Cox model was used to estimate the risk of occurrence of an event in univariate and multivariate analyses. The demographic characteristics of patients and sub-types of HL were not different in the << formalin >> and << Bouin >> groups. Median follow-up of patients was 69 months (1-135). In univariate analysis, low quantities of CD45RO+ cells was associated to a 40% 5-years progression free survival (PFS) compared to a 87% 5-years PFS for patients with high number of CD45RO+ cells (p=0.003), and low number of FOXP3+ cells was related to a 56% 5-years PFS compared to a 86% 5-years PFS for patients with high number of FOXP3+ cells (p=0.026). In multivariate analysis, taking into account the International Prognostic Score (IPS) and the expression of CD15 by neoplastic cells, the association of CD45RO and FOXP3 was the only independent prognostic factor identified (p This preliminary study underlines the importance of certain subsets of cells from the microenvironment of HL in the control of tumour growth. These results are encouraging. When used in association with validated criteria, they could make it possible to identify patients at risk of relapse and/or drug resistance. Nevertheless, due to the small number of patients in our study, these findings need to be validated in a larger and/or independent cohort. Disclosures: No relevant conflicts of interest to declare.

Abstract 3122 Poster Board III-59 Objective The Registry of Haematological Malignancies (HM) has been established on January 1st, 1980 in the department of Côte d'Or (pop 500 000 inhabitants). It was the first specialized registry in haematology in the world. During the course of 25 years (1980-2004), 5026 cases of HM were recorded including 1553 Myeloid malignancies (MM) in which entities not initially considered as malignant were taken in account such as Myelodysplastic syndrome (MDS) and some Myeloproliferative neoplasms (MPN). This allow us to present trends in incidence and survival of myeloid malignancies by entities since 1980. Method MM diagnosed in the population between 01/01/1980 and 31/12/2004 were registered. They were coded according to ICD-O-3 and following the principles of the 2001 WHO classification i.e. new threshold of definition of acute myeloid leukaemia (AML). World population standardized incidence rate were calculated by sex, age, by urban /rural repartition and by period of time. Five-year, 10-y, 15-y and 20-y relative survival was calculated using Estève's method and based on vital status updated at the end of 2007. Results Incidence rate for whole MM were 3.73/100 000/y in men and 2.74 in women. They were 2.82 for AML, 3.73 in MPN and 2.83 in MDS. Sex ratio was 1.5/1 as a whole and urban/ratio was always close to 2, data being statistically significant for numerous entities. Incidence rates increase with age until 75y-o for AML and 80y-o for MPN and then decreased except for MDS in which incidence continue to increase. Along the period, only incidence of MDS increases significantly with an annual medium rate of 3% (2.7 in men and 3.3 in women). In MPN, a significant increase was observed only for women (2%). Within AML, surprisingly a significant increase was found for AML with recurrent cytogenetic abnormality in men (8.03%). 15-years relative survival was the best for MPN (46%), being significantly better in women (59% vs 31%), in which Essential Thrombocythemia was the best (74%). On the contrary, the worst 15-y survival was for was for MDS (9%). Within AML, the best survival was found for AML recurrent cytogenetic abnormalities (15-y: 54%) compare to other categories (4-8%). Survival has not increased along the period, has increased significantly in MPN. In AML, 10-y survival has increased until 1999 to reach 39% but has decreased for the 2000-2004 period (8-y: 11%). Conclusion Analysis of the largest epidemiologic database on HM allow to produce information such as a significant increase of incidence of MDS in which survival remained very poor in both sex, a good survival of MPN compared to other categories and a worrying decreased of survival of AML in the more recent period. Disclosures No relevant conflicts of interest to declare.

Background: Angioimmunoblastic T-cell lymphoma (AITL), the second most frequent peripheral T-cell lymphoma, is characterized by a distinct clinical presentation and a spectrum of biological and morphologic features with little available data on prognostic factors in the literature. Objective: to evaluate the prognostic significance of clinical, biological and pathological features in AITL. Methods: between 1987 and 1998, 158 AITL patients were retrieved from the GELA LNH87 and LNH93 trials. Most patients received an anthracycline based polychemotherapy. Histologically, cases were reviewed by a panel of expert hematopathologists. For each case, at least 2 hematoxylin eosin slides were available for review together with an appropriate immunohistochemical staining including CD3, CD20 and CD21/CNA42. Cases were classified as “transformed” (>10% large B and/or T cells) and “classic” (including cases rich in clear cells, rich in epithelioid cells and with hyperplastic germinal centers). 62 cases were evaluated for CD10 and CXCL13 expression and scored according to the number of positive lymphoid cells. Results: there was a slight male predominance (M/F=1.53). Median age was 62 years. 81%, 50%, 67% and 39% of patients had respectively an advanced stage, a non ambulatory PS, an IPI>2 and a PIT>2 (age>60y, PS>=2, LDH>=normal and bone marrow involvement). Anemia, positive coombs test and bone marrow involvement were observed in 65%, 33% and 47% of cases respectively. Hypergammaglobulinemia, elevated LDH levels and skin rash were observed in 15%, 66% and 44% of patients respectively. 41 cases were classified as transformed and 117 as classic including 20 epithelioid-cell rich, 14 clear-cell rich and 4 with hyperplastic germinal centers. CD10 and CXCL13 were clearly positive, in a fraction of the lymphoid infiltrate, in 71% and 73% of cases respectively. CR/CRu remission was observed in 46% of patients after induction therapy. With a median follow-up of 69 months, the 5y overall survival rate was 33%, reaching a plateau level around 6 years. Concerning PIT, patients with 0–1 factors and 3–4 factors had a 5y OS of 37% and 38% respectively. In univariate analysis, absence of CR/CRu (p Conclusion: despite of various intensive regimens with an anthracycline based chemotherapy, AITL pursues an aggressive clinical course and carries a poor prognosis. Our large series emphasizes the morphologic heterogeneity of AITL, without any clinical impact, and further confirms the common expression of CD10 and CXCL13.