Abstract: Approximately 20%–50% of patients with large B‐cell lymphoma (LBCL) experience poor outcomes. We aimed to evaluate the combined prognostic value of circulating tumour DNA (ctDNA) and total metabolic tumour volume (TMTV) in LBCL. This observational single‐centre study included 112 newly diagnosed LBCL patients, receiving R‐CHOP/R‐CHOP‐like chemotherapies. CtDNA load was calculated following next‐generation sequencing of cell‐free DNA (cfDNA) using a targeted 40‐gene lymphopanel. TMTV was measured using a fully automated artificial intelligence‐based method for lymphoma lesion segmentation. CtDNA was detected in cfDNA samples from 95 patients with a median concentration of 3.15 log haploid genome equivalents per mL. TMTV measurements were available for 102 patients. The median TMTV was 501 mL. High ctDNA load ( 〉 3.57 log hGE/mL) or high TMTV ( 〉 200 mL) were associated with shorter 1‐year PFS (44% vs. 83%, p   〈  0.001 and 64% vs. 97%, p  = 0.002, respectively). When combined, three prognostic groups were identified. The shortest PFS was observed when both TMTV and ctDNA load were high ( p   〈  0.001). Even with a short follow up, combining ctDNA load with TMTV improved the risk stratification of patients with aggressive LBCL. In the near future, very high‐risk patients could benefit from CAR T‐cell therapy or bispecific antibodies as first‐line treatments.

Abstract: Introduction: AITL, one of the most common peripheral T-cell Lymphoma portends a poor prognosis. AITL is characterized by neoplastic T cells with a follicular helper immunophenotype, frequent mutations in epigenetic regulators TET2, IDH2, DNMT3A and in RHOA, and a prominent tumor microenvironment that could contribute to lymphomagenesis. Aiming to target this microenvironment and given the promising activity of lenalinomide (Len) in a relapsed setting (PMID: 23731832), we postulated that AITL patients (pts) might benefit from a treatment with Len combined with a classical CHOP regimen. This multicenter, open label, phase 2 trial (NCT01553786) investigates this treatment in previously untreated elderly pts. Patients and methods: Patients older than 59 years were treated with 8 cycles of Len + CHOP 21 (Len 25 mg/day (d), d1 to 14) and received intrathecal methotrexate prophylaxis. Thromboprophylaxis with low molecular weight heparin was mandatory. PET CTs at diagnosis and at the end of treatment were centrally reviewed. The primary objective was to evaluate the complete metabolic response rate according to the Lugano 2014 Classification. Secondary endpoints were safety, progression-free (PFS) and overall survival (OS). Mutations in TET2, IDH2, DNMT3A, RHOA, CD28, PLCG1, STAT3 and STAT5B were analyzed by deep sequencing (1000X) using DNA extracted from formalin-fixed paraffin-embedded tumor samples by PGM technology and were correlated to clinical parameters. Results: Between November 2011 and March 2017, 80 pts were enrolled, and 78 were evaluable. Central pathology review confirmed the diagnosis of AITL in 72 cases (92%). Median age was 69 (59-80), 52 % were female, 68% had a performance status of 0 to 1, 94% an Ann Arbor stage ≥III, 82% IPI≥3. Forty-five patients (58 %) completed the 8 planned cycles (mean number of cycles delivered, 5.9). Of the 624 planned treatment cycles, 458 (72 %) were completed. Treatment was stopped in 8 pts because of progressive disease, and in 15 because of adverse events. Toxicity was within the range expected of R-CHOP therapy with 70% grade 4 neutropenia and 31 % thrombocytopenia. Deep vein thrombosis occurred in 8 pts. Four secondary primary malignancies were reported. Five patients died from toxicity (4 from infection). Len dose reductions or interruptions were applied in 37 (5%) and 59 (9%) cycles, respectively, related to toxic effects. The median dose of Len per patient was 2275 mg (IQR 95-2825)-i.e. 81% of the planned dose of 2800 mg. Doxorubicin and cyclophosphamide were administrated at 98% and 97% of the planned dose. Complete metabolic response was observed in 34 patients (43.6%) (90%CI = [34.0%; 53.5%]), partial metabolic response in 3 (3.8%), no metabolic response in 2 (2.6%) and progressive metabolic disease in 16 (20.5%), the other being not evaluated because of progression (N=20) or death (N=3). With a median follow-up duration of 31.5 months (95%CI = [23.0; 43.7] ) at the time of the cut-off, 2-year PFS is 42.3% (95%CI = [30.9%; 53.2%]) and 2-year OS is 60.1% (95%CI = [47.4%; 70.7%] ). IPI was strongly associated with the survival (figure 1A). Mutational status was successfully determined in 64 pts with confirmed AITL diagnosis. TET2 mutations were detected in 49 cases (77%), with 28 (43%) pts bearing ≥2 TET2 mutations. RHOAG17V mutations in 34 pts (53%), DNMT3A mutations in 20 (31%) pts, including 6 with the DNMT3AR882X variant and IDH2 mutations in 14 (22%). CD28, PLCCG1 and STAT mutations were detected in less than 10% of pts (figure 1B). TET2 mutations correlated to age 〉 65 years (p=0.006) and IPI 3-5 (p=0.007). Interestingly, DNMT3A mutations were associated with a decreased response rate (p=0.003), a shorter PFS (p=0.04) and a trend toward a shorter OS (0.08). It is noteworthy that none of the 6 pts with the DNMT3AR882X mutant had response, suggesting that the resistance to anthracycline reported in DNMT3AR882X mutated acute myeloid leukemia (PMID: 27841873) could also occur in DNMT3AR882X mutated AITL. No correlation between other detected mutations and outcome was observed (table 1C). Conclusion A combination of 25 mg of Len for 14 days with CHOP cycles gives acceptable toxicity in AITL elderly pts. However, response rate and outcome appear similar to previous studies. We also confirmed in a prospective study the frequency of mutations in epigenetic regulators and RHOA in AITL and clarified their prognostic impact. Figure Figure. Disclosures Bachy: Gilead Sciences: Honoraria; Sandoz: Consultancy; Janssen: Honoraria; Roche: Research Funding; Takeda: Research Funding; Amgen: Honoraria; Celgene: Consultancy. Cartron:Celgene: Consultancy, Honoraria; Sanofi: Honoraria; Gilead Sciences: Honoraria; Janssen: Honoraria; Roche: Consultancy, Honoraria. Casasnovas:Merck: Honoraria; Takeda: Honoraria; Roche: Honoraria; Gilead Sciences: Research Funding; Roche: Research Funding; Janssen: Consultancy; Gilead Sciences: Consultancy; MSD: Consultancy; merck: Consultancy; takeda: Consultancy; Roche: Consultancy; Janssen: Honoraria; Celgene: Honoraria; Gilead Sciences: Honoraria; MSD: Honoraria. Tilly:Celgene: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; BMS: Honoraria; Astra-Zeneca: Membership on an entity's Board of Directors or advisory committees. Gaulard:Celgene: Research Funding; Roche: Honoraria; Takeda: Consultancy, Honoraria, Research Funding. Haioun:Takeda: Consultancy, Honoraria; Amgen: Consultancy, Honoraria; Sciences: Consultancy, Honoraria; Roche: Consultancy, Honoraria; Gilead Sciences: Consultancy, Honoraria; Celgene: Consultancy, Honoraria; Janssen: Consultancy, Honoraria.

Abstract: Introduction Aggressive B-cell lymphomas are heterogeneous in their clinical course and biological characteristics. They include diffuse large B-cell lymphoma not otherwise specified (DLBCL-NOS), high grade-B-cell lymphoma with double/triple-hit or NOS (HGBL), primary mediastinal B-cell lymphoma (PMBL). In order to better differentiate these entities, the WHO classification recommends using immunohistochemistry (IHC), FISH, targeted sequencing and gene expression profiles (GEP). However, these techniques are most often performed retrospectively in clinical trials, which is not representative of real life. In order to use these information to improve the current standard of treatment with targeted therapies adapted to lymphoma biology, we have set up a national network on behalf of the LYSA called RT3 (for Real-time tailored therapy) to demonstrate that we are able to comprehensively characterize aggressive B cell lymphomas in a clinically relevant timeline. Materials and methods Patients & gt; 18 years of age with untreated aggressive B-cell lymphoma were included prospectively from 21 LYSA centers. FFPE specimens were analyzed and classified according to WHO classification, benefiting from an IHC profile (Hans algorithm, BCL2, MYC), FISH analysis (BCL6, BCL2, MYC break apart and MYC-IGH/IGL/IGK fusion probes), targeted NGS analysis and a targeted GEP using RT-MLPA (Bobée et al. J Mol diagn 2017). A first part of the study phase was carried out in an oligocentric manner with only 2 referral platforms for pathological analysis and molecular characterization. The second phase was the implementation of 7 RT3 platforms spread over France. The main objective was to evaluate the capacity of the network to provide an exhaustive histopathological and molecular characterization 4 days before theoretical R-CHOP21 C3 administration (within 38 days after D1 cycle 1). Results The oligocentric cohort 1 prospectively included 72 patients in 6 months: 19 had insufficient material or inappropriate diagnosis to be qualified and 53 benefited from the complete analysis on referral platforms. A complete characterization 4 days before RCHOP-C3 was provided to the clinician in 47 cases (88.7%), allowing to further implement the network with 7 platforms and 23 clinical investigation centers. 183 patients were included in the second phase in 9 months, 35 were excluded for inadequate diagnosis/material. On this population, 143 (96.6%) complete patient-reports were provided with a median time of 32 days (1-50). Finally, 201 cases were retained in the Full Analysis Set for which tumor samples fulfilled all prerequisites and diagnosis of DLBCL was confirmed by RT3 platforms. The clinical characteristics were as follows: median age of 61 y (20-92), with 45.2% of pts with aaIPI 2-3and 67.1% with Ann Arbor stage III-IV; 1L treatment consisted mainly of RCHOP (RCHOP14/21 = 136, 72.3%). 7% of patients were treated with experimental drugs. 76% presented with extranodal involvement. After pathological review, 139 (69%) patients were classified as DLBCL-NOS (41% GCB; 58% nGCB), 11 (5%) as HGBL-NOS, 8 (4%) as HGBL-DH and 18 (9%) as PMBL, 3 (1%) as EBV+ DLBCL-NOS, 9 (4%) as FL-3B, 7 (3%) transformed DLBCL, 2 (1%) plasmablastic, 1 (0.5%) unclassified, 1 (0.5%) DLBCL with IRF4 rearrangement. By immunohistochemistry, 22 cases were CD5+, 74% were BCL2+, and 53% MYC+ and 49% as double expressors. BCL2/18q21, BCL6/3q27 and MYC/8q24 breaks were observed in 14% (29/201), 25% (50/201) and 12% (25/201) of the cases respectively. MYC partner gene was available in 10/29 cases (5 MYC-IG, 5 non-IG). At the molecular level, PIM1 was the most frequently mutated gene in the entire cohort (36%). In the ABC subtype, MYD88 was mutated in 48% of cases. In the GCB cohort, SOCS1 was the most frequently mutated gene (30%). In the HGBL cohort, MYC was the most mutated gene (43%). Conclusion This study demonstrates that this RT3 LYSA network allowed a real time pathological and molecular characterization of aggressive B-cell lymphomas according to the WHO recommendations. The RT3 network provides relevant informations to the clinician before R-CHOP21 C3, that might contribute to modify the treatment eventually adding targeted therapies or tailor a second line treatment. This network should be used as a basis for future clinical trials. Inter platforms reproducibility and prognostic relevance of the RT3 project will be presented. Disclosures Haioun: Gilead: Honoraria; Janssen: Honoraria; Novartis: Honoraria; Amgen: Honoraria; Takeda: Honoraria; Miltenyi: Honoraria; Servier: Honoraria; Roche: Honoraria; Celgene: Honoraria. Le Gouill:Roche Genentech, Janssen-Cilag and Abbvie, Celgene, Jazz pharmaceutical, Gilead-kite, Loxo, Daiichi-Sankyo and Servier: Honoraria; Loxo Oncology at Lilly: Consultancy. Ribrag:arGEN-X-BVBA: Research Funding; BAY1000394 studies on MCL: Patents & Royalties; Institut Gustave Roussy: Current Employment; argenX: Current equity holder in publicly-traded company, Research Funding; Epizyme: Consultancy, Current equity holder in publicly-traded company, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Honoraria, Membership on an entity's Board of Directors or advisory committees; Infinity: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; MSD: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; BMS: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel Expenses; Nanostring: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; F. Hoffmann-La Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel Expenses; AstraZeneca: 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; Servier: Consultancy, Honoraria; Pharmamar: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; AZD: Honoraria, Other; Eisai: Honoraria; Incyte: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Immune Design: Consultancy, Membership on an entity's Board of Directors or advisory committees; Roche/Genentech: Consultancy, Membership on an entity's Board of Directors or advisory committees. Salles:Takeda: Honoraria; Karyopharm: Honoraria; Genmab: Honoraria, Other; Debiopharm: Consultancy, Honoraria, Other: consultancy or advisory role; Autolos: Other: consultancy or advisory role; Abbvie: Other: consultancy or advisory role; Roche: Honoraria, Other: consultancy or advisory role; Novartis: Honoraria, Other: consultancy or advisory role; MorphoSys: Honoraria, Other: consultancy or advisory role; Janssen: Honoraria, Other: consultancy or advisory role; Epizyme: Honoraria, Other: consultancy or advisory role; Kite, a Gilead Company: Honoraria, Other: consultancy or advisory role ; BMS/Celgene: Honoraria, Other: consultancy or advisory role. Sujobert:Sunesis: Research Funding; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead/Kyte: Membership on an entity's Board of Directors or advisory committees. Bouabdallah:Gilead Sciences: Consultancy, Honoraria; Takeda: Consultancy, Honoraria; Roche: Consultancy, Honoraria. Sibon:takeda france: Consultancy.

Abstract: Background : High grade B cell lymphoma (HGBL) either Not Otherwise Specified (NOS) or either Double Hit (DH MYC/BCL2 or DH MYC/BCL6) or Triple Hit (TH) have been considered as separate entities from Diffuse Large B-Cell Lymphoma (DLBCL) NOS in the 2016 WHO classification. Until now, treatment of these particularly aggressive histological sybtypes remains unclear and most of patients (pts) still have a poor outcome because of chemoresistance or early relapse. Except IPI score, prognosis factors are not completely clarified. Since few years, treatment with Chimeric Antigen Receptor T (CAR-T) cells has changed the prognosis of relapsed refractory DLBCL and could represent a promising therapeutic approach - early in the course of the disease - to improve survival of HGBL. The aim of the present study is to describe a « real life » cohort of pts treated in our institution and to evaluate the prognostic impact of clinical, histological, genetic and imaging characteristics and treatment. Methods : We collected data of 45 pts diagnosed between January 2009 and October 2020. BCL2, BCL6 and MYC break were analyzed by Interphase Fluorescence In Situ Hybridization (FISH) with break-apart probes. When MYC was rearranged, MYC partner gene was analyzed with double fusion probes (MYC/IgH, MYC/IgL, MYC/IgK). Total metabolic tumor volume (TMTV) was calculated using fully automatic segmentation of lymphoma lesions by an artificial intelligence (P. Blanc-Durand, Eur J Nucl Med Mol Imaging, 2021). All pts were treated with rituximab R-CHOP/CHOP-like regimens. Treatment was reinforced (intensive regimen) with high dose methotrexate (14 pts treated with RCOPADM and one with R-ACVBP) when they were considered fit enough. Autologous stem cell transplantation at 1st relapse was performed in 3 pts and 5 pts were treated with CAR-T cells at 2 nd relapse. Results : Baseline characteristics were as follows: median age 60 y (28-80), IPI 3-5 : n=33 (73%), CNS localization : n=8/42 (19%), bone marrow involvement : n=10/34 (22%). Twenty-seven pts were DH lymphoma (MYC/BCL2: 18; MYC/BCL6: 9), 8 patients were TH-lymphoma and 10 were HGBL-NOS. MYC rearrangement was detected in 40 cases with an immunoglobulin partner gene (IgH, IgL or IgK) in 14 (52%) of 27 evaluable cases. TMTV was calculated for 29 pts with a median TMTV of 1019 ml (2-4536 ml). Fifteen pts were treated with an intensive regimen. With a median follow-up of alive pts of 17 months, 2-y progression-free survival (PFS) and overall survival (OS) were 42% [26.4 ; 56.8] and 62% [43.7 ; 75.5] respectively. PFS and OS were significantly affected by IPI 3-5 (PFS: p= 0.006, HR=3.37 ; OS: p= 0.01, HR=3.71) and CNS involvement (PFS : p=0.006, HR=6.93 ; OS: p=0.006, HR=7.44). TMTV & gt;1000ml (PFS: p=0.051, HR 2.99 ; OS: p=0.033, HR=4.13) was associated with a significantly inferior OS. Prognosis of HGBL DH/TH and HGBL-NOS was similar (PFS: p= 0.89, HR=0.93 ; OS: p=0.64, HR=1.32). Immunoglobulin as MYC partner gene had no significant impact on prognosis (PFS: p=0.47, HR=1.48 ; OS: p=0.32, HR=1.73). Intensive regimens did not improve significantly prognosis (PFS: p=0.26, HR=1.71 ; OS: p=0.08, HR=2.86). Conclusion: This monocentric study confirms previous reports on the poor prognosis of HGBL (DH/TH and NOS) especially for patients for both high IPI or high TMTV and in case of initial CNS involvement. We confirm that intensive regimen don't improve prognosis supporting the need of alternative strategies incorporating bispecific monoclonal antibodies or consolidative CAR-T cells in very high risk pts. Disclosures Roulin: Janssen: Other: Travel and meetings. Lemonnier: Institut Roche: Research Funding; Gilead: Other: travel grant. Le Bras: Novartis: Honoraria; Takeda: Honoraria, Research Funding; Kite Gilead: Honoraria; Celgene BMS: Research Funding. Gaulard: Innate Pharma: Research Funding; Sanofi: Research Funding; Alderaan: Research Funding; Gilead: Consultancy; Takeda: Consultancy, Honoraria. Haioun: Miltenyi: Honoraria, Research Funding; Gilead: Honoraria, Research Funding; Janssen: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; F. Hoffmann-La Roche Ltd: Honoraria, Research Funding; Servier: Honoraria, Research Funding; Takeda: Honoraria, Research Funding; Celgene: Honoraria, Research Funding; Amgen: Honoraria, Research Funding.

Abstract: Background: Angioimmunoblastic T cell lymphoma (AITL) is one of the most frequent peripheral T cell lymphoma, and has a poor prognosis. Neoplastic T cells originate from T follicular helper cells, and are admixed among a prominent microenvironment, making their identification sometimes difficult. IDH2 mutations are present in 20-30% AITL patients, where they often co-exist with TET2, DNMT3A and RHOA mutations. They affect almost exclusively the codon R172 of IDH2, providing to the IDH2 enzyme a neo-activity that converts α ketoglutarate (αKG) to D 2-hydroxyglutarate (2HG). D-2HG, the dextrogyre form of 2HG, is an oncometabolite present at very low level under physiological condition, which inhibits many αKG dependent dioxygenases, including TET proteins and is involved in oncogenesis of various cancers such as gliomas or acute myeloid leukemias (AML). Preliminary data, based on small series, showed that increased level of 2HG was detectable in tumor and in serum of IDH2 mutated AITL. However, 2HG level, as well as D/L ratio, has not been evaluated as a surrogate marker of IDH2 mutation in AITL, at diagnosis or during the follow-up. Patients and Methods: Serum from 69 AITL patients, collected in REVAIL trial (NCT00169156) (N=48), RAIL trial (NCT01553786) (N=9) or TENOMIC collection (N=12) were included in this study. IDH2 mutations were assessed in formalin fixed paraffin embedded tumor tissue by deep next generation sequencing of exon 4, using PGM technology (N=63) or allele specific PCR (N=6). For the purpose of the study the cohort was enriched in mutated patients. Serum was collected at diagnosis and at the end of the frontline treatment in 6 patients, 5 of them being IDH2 mutated. D and L 2HG was measured in serum using a liquid tandem mass spectrometry method as previously described (Poinsignon et al. J Chromatogr B 2016) to determine total (D+L) 2HG and D/L 2HG ratio. Results: Twenty-four patients (35%) were IDH2 mutated. Median IDH2 variant allele frequency (VAF) was 7% (IQR, 4%-12.5%). Median total 2HG was 3.63 µM (IQR, 1.6-6.1) in mutated patients versus 1.17 µM (IQR, 0.85-1.68) in wild type patients (p 〈 0.001) (figure). D/L was also found significantly higher in mutated [4.7 (IQR 3.52-7.74)] than in WT patients [1.5 (IQR 0.94-2.23)] (p 〈 0.001). IDH2 mutations included 11 IDH2R172K mutations, 5 IDH2R172G mutations, 4 IDH2R172S, 2 IDH2R172T, 1 IDH2R172M and 1 IDH2R140Q, all showing similar level of 2HG, without variation of 2HG level depending on the mutation. 2HG level correlated with IDH2 mutation VAF (Spearman r=0.4590, p=0.024) likely paralleling the tumor cells mass. Using the cut off established in AML, 2µM and 2.5 for respectively total 2HG and D/L ratio, we found that 17/24 (71%) mutated patients versus 7/45 (15%) had a total 2HG ≥ 2, and that 21/24 (87.5%) mutated versus 8/45 (18%) WT patients had D/L 2HG ratio ≥2.5, leading to a sensibility of 71% and 84%, and a specificity of 87% and 82%, respectively for total 2HG and D/L ratio to detect IDH2 mutations. End of treatment 2HG dosage was available in 5 IDH2 mutated patients with elevated 2HG at diagnosis. All patients had a total 2HG decreased below 2µM. Three patients normalized their D/L ratio, 2 of them being in complete response and one being not evaluable. In contrary, two patients kept a D/L 2HG ratio 〉 2.5, one having progressive disease while the other remains in complete response 6 months after the end of treatment. Conclusion: Serum total 2HG and D/L 2HG ratio can reliably predict the presence of IDH2 mutation in AITL patients. Most patients in complete remission normalize their dosage and ratio, suggesting that serum 2HG could be a marker of residual disease that could be used to monitor patients. Specific cut-off values of 2HG and D/L 2HG ratio for AITL are under investigation. Figure. Figure. Disclosures Ribrag: Servier: Consultancy, Honoraria; pharmamar: Other: travel; Gilead: Consultancy, Honoraria; Amgen: Research Funding; epizyme: Consultancy, Honoraria; Roche: Honoraria, Other: travel; argenX: Research Funding; NanoString Technologies: Consultancy, Honoraria; BMS: Consultancy, Honoraria, Other: travel; Incyte Corporation: Consultancy; MSD: Honoraria; Infinity: Consultancy, Honoraria. Tilly:Astra-Zeneca: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; BMS: Honoraria. Haioun:Roche: Consultancy, Honoraria; Celgene: Consultancy, Honoraria; Amgen: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Gilead Sciences: Consultancy, Honoraria; Sciences: Consultancy, Honoraria; Takeda: Consultancy, Honoraria. Gaulard:Roche: Honoraria; Takeda: Consultancy, Honoraria, Research Funding; Celgene: Research Funding.