Cell of Origin and Treatment Impact on the Outcome of Monomorphic Post-Transplant Lymphoproliferative Disorder-Diffuse Large B-Cell Lymphoma Subtype

Introduction : Cell of origin (COO) determination of diffuse large B-cell lymphoma (DLBCL) is important, as it has prognostic and therapeutic relevance. Patients with germinal center (GC) like DLBCL have more favorable outcomes than those with non-GC DLBCL, when treated with standard immunochemotherapy. In monomorphic post-transplant lymphoproliferative disorder (PTLD), DLBCL subtype, the biologic significance of COO is not well established. The aims of this study were to evaluate the impact of the COO on clinical presentation, outcome and response to different treatment regimens. Methods : We conducted a retrospective review of all monomorphic PTLD of the DLBCL subtype diagnosed and treated at Columbia University from 2000-2018. COO classification into GC and non-GC subtype was determined by immunohistochemistry using the Hans algorithm. In situ hybridization for Epstein-Barr encoded small RNAs (EBER) was performed to evaluate for EBV infection of the neoplastic lymphocytes. Outcomes according to therapeutic regimens were assessed using the revised response criteria for malignant lymphoma (2007). Overall survival (OS) and progression free survival (PFS) were estimated by the Kaplan-Meier method and compared by the log rank test. Results : Cell of origin: Biopsy material for COO subtyping was available for 49 monomorphic PTLD-DLBCL diagnosed during the study period. By immunohistochemistry 20/49 cases (40.8%) were GC and 29/49 cases (59.2%) were non-GC. Median age of presentation was 52 years (range 3-72) and 55 years (range 3-75), for the GC and non-GC respectively. Median time from transplant to PTLD onset was 3 years, range 7 months-25 years vs. 6 years, range 2 months-24 years for GC vs non-GC respectively. There was no significant difference in the incidence of EBV positivity, immunosuppressive therapy, organ transplanted, stage, LDH, ECOG and IPI between the 2 groups. Of note, extranodal involvement, specifically of the gastrointestinal tract was extremely common in both subtypes (16/20 [80%] in GC and 25/29 [86%] in non-GC). While acknowledging the heterogeneity of therapies administered, a trend suggesting better prognosis for the GC type was noted, although not statistically significant. The median PFS was 151.0 months for GC vs 17.0 months in non-GC (p= 0.1095), while the median OS was 151.0 months for GC vs 36.7 months for non-GC (p=0.1006) (Figure 1A, 1B). Impact of Treatment: Of the 49 cases, 44 were adults (age≥18). The two most common first line therapies administered were R-EPOCH (25) and R-CHOP (9). Ten patients received 6 other therapies including Rituximab monotherapy (3), CHOP (2), palliative care (2), RCEOP (1), EPOCH (1), and radiation (1). When focusing on patients who received R-CHOP or R-EPOCH as initial therapy, no significant differences in age, stage, LDH, extranodal disease, ECOG, IPI, immunosuppressive therapy and COO was identified between the 2 groups. The complete response (CR) rate for R-CHOP was 6/9 (66.6%) vs. 21/25 (84%) for R-EPOCH (p= 0.2701). Primary refractory disease was present in a third of the patients receiving R-CHOP vs. (4/25) 16% of the patients treated with R-EPOCH (p=0.2701). Treatment related mortality was low in both groups (1/9 (0.1%) vs 1/25 (0.04%) patients died during first line therapy with RCHOP vs REPOCH respectively). No difference was seen in terms of PFS and OS in the 2 groups. Median PFS and OS for RCHOP was 80.0 and 83.9 months respectively, the medians for R EPOCH were not reached (log-rank test p= 0.5430, 0.2855 for PFS and OS, respectively) (Figure 1C-1D). When the OS and PFS curves for RCHOP and REPOCH were analyzed based on COO, again no difference was noted. Conclusions : Non-GC subtype is more common than GC in monomorphic PTLD DLBCL. Clinical characteristics, EBV infection and time of onset post transplant is not different in the 2 subtypes. There is a trend suggesting better PFS and OS of PTLD DLBCL GC subtype, although not statistically significant, mirroring the outcome of GC vs non-GC COO in DLBCL of immunocompetent patients. With regard to therapy, R-EPOCH did not improve OS or PFS when compared to R-CHOP, but did not result in increased toxicity or treatment related mortality, which in our series was extraordinarily low for both therapies (<=0.1%). Given the retrospective nature of our analysis, further studies of a larger cohort of patients is ongoing to validate these observations. Disclosures Sawas: Seattle Genetics, Gilead, Daiichi Sanko: Consultancy; Affimed: Research Funding. Lue:Kymera Therapeutics: Honoraria; Astex Pharmaceuticals: Honoraria. Marchi:Spectrum Pharmaceuticals, Verastem Oncology: Research Funding. Radeski:Corvus Pharmaceuticals: Research Funding. O'Connor:Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Mundipharma: Consultancy, Membership on an entity's Board of Directors or advisory committees; TG Therapeutics: Other: Travel Support, Research Funding; ADCT Therapeutics, Affimed, Agensys, Merck, Seattle Genetics, Spectrum, Trillium, and Verastem Oncology.: Research Funding.

Assessing the Feasibility of Pathology Informatics Approaches for Subtyping Diffuse Large B-Cell Lymphoma (DLBCL)

Introduction Patients (pts) with DLBCL can experience marked difference in overall survival (OS) based on clinical characteristics, race, and biological subtype. Immunohistochemistry (IHC) algorithms can stratify pts into the germinal center B-cell-like (GCB) or activated B-cell-like (ABC) subtype, which is associated with worse OS with current standard of care therapy with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (RCHOP; Meyer et al. 2011). However, these algorithms can be challenging to perform in clinical practice. We evaluated the frequency of GCB and non-GCB subtypes in a retrospective cohort of patients and evaluated the feasibility of performing automated pathology informatics computer segmentation algorithms to assess DLBCL subtype. Methods We included all patients diagnosed with DLBCL at Emory University from 1995-2016. We collected demographic, clinical, laboratory and pathologic data on all patients. We evaluated GCB and non-GCB status using the Hans algorithm as performed during routine clinical care, collected IHC slides for H&E, CD10, CD20, CD30, BCL2, BCL6, C-MYC, and MUM1, and assessed the feasibility of performing automated computer segmentation algorithms to assess DLBCL subtype using these materials. Results Out of 364 patients with DLBCL in our database, 151 had available data for DLBCL classification by the Hans algorithm and 103 had available slides for computational algorithms. For the entire dataset the median age was 61 years, 51% were male, 31% had stage III/IV disease. Among patients who had classification data, the median age was 61 years, 53% were male, 45% had stage III/IV disease, and 47% GCB DLBCL and 53% had non-GCB. Among patients with slides available for computational algorithms the median age was 64 years, 53% were male, 48% had stage III/IV disease. Conclusions In this retrospective cohort study, the group of patients with available slides for performing computer segmentation algorithms had similar demographics and disease characteristics to patients in the general population of DLBCL at an academic medical center. This dataset and associated pathology images provides a useful resource to evaluate whether computational algorithms can aid in defining the prognosis for DLBCL patients. Future directions for this work will involve the establishment of an online DLBCL digital archive to develop a more precise, repeatable and objective image-analysis based scoring of DLBCL tissues to improve the prognostic accuracy and classification of DLBCL. Disclosures Flowers: NIH: Research Funding; Genentech: Consultancy, Research Funding; Roche: Consultancy, Research Funding; Gilead: Consultancy, Research Funding; TG Therapeutics: Research Funding; Mayo Clinic: Research Funding; Infinity: Research Funding; Millenium/Takeda: Research Funding; ECOG: Research Funding; AbbVie: Research Funding; Acerta: Research Funding; Pharmacyclics, LLC, an AbbVie Company: Research Funding.

Integrating Genomic Alterations in Diffuse Large B-Cell Lymphoma Identifies New Relevant Pathways and Potential Therapeutic Targets

Introduction: DLBCL is biological and clinically highly heterogeneous. Although different genetic aberrations, including recurrent somatic mutations, have been described in this tumor, their clinical impact remains to be clarified. The aim of the present study was to determine somatic mutations and copy number alterations of a selected group of genes in patients with DLBCL, in order to assess their prognostic importance and to identify potential personalized targeted drugs for these patients. Methods: 150 patients (78M/72F; median age, 66 years) diagnosed with de novo DLBCL no otherwise specified at Hospital Clínic and other institutions of the GELCAB, treated with immunochemotherapy, were included in the study. An independent series of 111 patients (54M/57F; median age, 63 years), diagnosed at different Japanese and Spanish institutions, was used to validate the significant findings. Targeted next generation sequencing (NGS) of 106 representative genes related with DLBCL and Copy Number Alterations (CNA) assessment were performed. Ten functional pathways were pre-defined, including NOTCH, tumor suppressor genes, JAK/STAT, epigenome/chromatic modifier, BCR signaling, PI3K-AKT-mTOR, MAP-kinase, B-cell differentiation, immune surveillance and cell cycle alterations. Cell of origin (COO) of the tumors was established using gene expression or the Lymph2Cx assay. Genomic-guided potential therapeutic opportunities for each patient were identified in silico by a Cancer Genome Interpreter platform. Results: A total of 765 potential driver mutations were identified in 89 of the 106 genes with a slightly higher number in germinal center B-cell like (GCB) than activated B-cell-like (ABC) DLBCL subtype. The most frequently mutated genes found in >15% of the cases were KMT2D (MLL2), MYD88, CREBBP and TP53, with other 27 genes being mutated in >5% of the cases. Several genes were differentially mutated in GCB DLBCL subtype (KMT2D, CREBBP, TNFRSF14, B2M, EZH2, GNA13, FOXO1, ACTB and SOCS1) or ABC subtype (MYD88, PIM1, CD79B and PRDM1). No relevant differences were observed in the clinical features according to individual mutations or CNA. No single gene mutation predicted response to therapy. Genetic alterations in KLHL6, ETV6, SGK1, L8q12.1, CD79B, PIM1 and TP53 predicted poor OS, whereas mutations of SOCS1 were associated with better outcome. Alterations in NOTCH pathway and tumor suppressor pathway were associated with poor outcome, whereas those of JAK/STAT pathway showed favorable prognosis (see table for detailed data). NOTCH pathway (HR 2.8; p=0.006) and tumor suppressor pathway (HR 2.4; p=0.005) maintained independent significance for OS along with R-IPI (H 4.0; p=0.006) in a multivariate analysis that also included COO and beta2-microglobulin. In addition, the prognostic value of NOTCH and tumor suppressor pathways was confirmed in the independent validation series. Finally, we identified 69 cases (46%) carrying at least one genomic alteration in 9 genes considered a biomarker of drug response supported by data of early clinical trials or pre-clinical assays; tumors of additional 26 patients (17%) had at least one gene alteration that could be exploited by a drug repurposing strategy. Conclusions: Integrating the deep sequencing analysis of a panel of selected genes and CNA, we have recognized novel target genes and defined the clinical relevance of alterations of NOTCH and tumor suppressor pathways in DLBCL. Using an in silico prescription pipeline we have also identified a number of candidate drugs with potential therapeutic interactions with driver oncogenic proteins. All these findings may orient future preclinical and clinical intervention strategies in DLBCL. Table Initial features, response to therapy and outcome according to pathways´ status Table. Initial features, response to therapy and outcome according to pathways´ status Disclosures Sancho: Celltrion, Inc: Research Funding; Gilead: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Sanofi: 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: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Gonzalez Barca:Janssen: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Roche: Speakers Bureau; Gilead: Speakers Bureau. Ohshima:Kyowa Hakko Kirin Co., Ltd.: Research Funding, Speakers Bureau; CHUGAI PHARMACEUTICAL CO.,LTD.: Research Funding, Speakers Bureau. Akashi:Sunitomo Dainippon Pharma: Consultancy; Celgene: Research Funding; Kyowa Hakko Kirin: Consultancy, Research Funding; Bristol Meyers Squibb: Research Funding; Asahi Kasei Pharma Corporation: Research Funding; Chugai Pharmaceutical Co., Ltd.: Research Funding; Shionogi & Co., Ltd: Research Funding; Astellas Pharma: Research Funding.

Subtype-specific addiction of the activated B-cell subset of diffuse large B-cell lymphoma to FOXP1

High expression of the forkhead box P1 (FOXP1) transcription factor distinguishes the aggressive activated B cell (ABC) diffuse large B-cell lymphoma (DLBCL) subtype from the better prognosis germinal center B-cell (GCB)-DLBCL subtype and is highly correlated with poor outcomes. A genetic or functional role for FOXP1 in lymphomagenesis, however, remains unknown. Here, we report that sustained FOXP1 expression is vital for ABC-DLBCL cell-line survival. Genome-wide analyses revealed direct and indirect FOXP1 transcriptional enforcement of ABC-DLBCL hallmarks, including the classical NF-κB and MYD88 (myeloid differentiation primary response gene 88) pathways. FOXP1 promoted gene expression underlying transition of the GCB cell to the plasmablast—the transient B-cell stage targeted in ABC-DLBCL transformation—by antagonizing pathways distinctive of GCB-DLBCL, including that of the GCB “master regulator,” BCL6 (B-cell lymphoma 6). Cell-line derived FOXP1 target genes that were highly correlated with FOXP1 expression in primary DLBCL accurately segregated the corresponding clinical subtypes of a large cohort of primary DLBCL isolates and identified conserved pathways associated with ABC-DLBCL pathology.

FOXO1-p300-Txn Circuit Regulates Oxidative Stress Responses in Diffuse Large B-Cell Lymphomas Characterized By Enhanced Oxidative Phosphorylation

Diffuse large B-cell lymphoma (DLBCL) is a clinically and molecularly heterogeneous disease. The comparison of DLBCLs transcriptional profiles using multiple clustering algorithms led to the identification of distinct DLBCL subtypes reflecting tumor-intrinsic features. The OxPhos-DLBCL subtype is characterized by enhanced mitochondrial oxidative phosphorylation, a major source of potentially toxic reactive oxygen species (ROS). Therefore, we investigated the role of potential mechanisms attenuating ROS toxicity in this DLBCL subtype. We found significantly increased thioredoxin (TXN) mRNA abundance in DLBCLs classified as OxPhos subtype compared to other subtypes. The overall survival (OS) of patients with high TXN mRNA expression was significantly shorter than of those with low TXN mRNA expression, regardless of treatment regimen (R-CHOP and CHOP). TXN overexpression in OxPhos-DLBCLs was also confirmed in a cell line panel at protein level using immunoblotting. To explain transcriptional mechanisms responsible for differential TXN expression in different DLBCLs, we analyzed the TXN promoter and identified two BCL6 binding sites within 1kb upstream of TXN transcription start site. Unlike in other molecular subtypes, BCL6 does not exhibit a repressor activity in OxPhos-DLBCLs (PNAS, 2007; 104: 3207-12). Using luciferase reporter assays and shRNA-mediated gene expression knock-down, we demonstrated that relative differences in TXN abundance between DLBCL subtypes are at least in part caused by the lack of BCL6 transcription repressor activity. We next tested the consequences of TXN depletion in DLBCLs. We found that OxPhos cells with silenced TXN expression were uniformly more sensitive to apoptosis induced by ROS production than control cells. TXN inhibition sensitized all tested cell lines to doxorubicin, the fundamental drug used in DLBCL chemotherapy acting in part as a ROS inducer. In addition to its role in maintaining redox homeostasis, TXN also regulates transcriptional responses to ROS. For example, TXN reduces disulfide bonds between FOXO4 and acetyltransferase p300, which results in reduced FOXO4 acetylation and impaired proapoptotic signaling. For this reason, we assessed whether p300 and TXN are involved in acetylation of FOXO1, a major FOXO member expressed in DLBCLs. In cells with blocked TXN activity, FOXO1 acetylation was significantly higher compared to cells overexpressing wild-type TXN. TXN decreased p300-mediated FOXO1 acetylation, reduced its proapoptotic activity and expression of FOXO1-dependent genes (TRAIL, p27, BIM). We found that FOXO1 and p300 interact in redox and TXN-dependent manner and identified a conserved FOXO1's Cys612 to be responsible for FOXO1-p300 binding. We mutated FOXO1 C612 to Ala and found that when cotransfected with p300, C612A FOXO1 exhibited dramatically suppressed ROS-induced acetylation. Blockade of FOXO1 acetylation resulted in markedly lower expression of FOXO1 target genes, higher cellular proliferation and lower apoptosis. Furthermore, TXN inhibited FOXO1 nuclear translocation in response to oxidative stress in OxPhos-DLBCLs. Finally, silencing FOXO1 in OxPhos cells with knocked-down TXN expression markedly inhibited DLBCL cell line apoptosis in response to oxidative stress, suggesting that FOXO1 is an essential TXN-regulated sensor and effector of ROS toxicity in OxPhos-DLBCL cells. Taken together, these results demonstrate that TXN is overexpressed in a subset of DLBCLs, and high TXN mRNA abundance is related to shorter OS of DLBCL patients. TXN knock-down enhances oxidative stress toxicity in OxPhos cell lines at least in part by facilitating FOXO1 nuclear retention and increasing acetylation of this transcription factor, thus augmenting its proapoptotic activity. Disclosures No relevant conflicts of interest to declare.

Mirna-181a expression Lead to Longer Animal Survival and Slower Tumor-Growth Rate in Diffuse Large B-Cell Lymphoma Xenograft Models

Introduction: Diffuse large B-cell lymphoma (DLBCL) is subdivided into the germinal center B-like (GCB) and activated B cell-like (ABC) subtypes by gene expression profiling, and these subtypes exhibit different clinical outcomes and signaling pathway deregulations. Compared to the GCB, the ABC-DLBCL subtype displays a more aggressive clinical course and shorter patient survival. Constitutive nuclear factor kappa-B (NF-kB) activity is often associated with the ABC-DLBCL subtype, however recent studies suggest that NF-kB signaling activation is also observed to a lower extent in the GCB-DLBCL subtype (Lina Odqvist et al. 2014). miRNAs have diagnostic and prognostic value in disease classification, and growing evidence implicates miRNAs in tumorigenesis, tumor maintenance, and dissemination through their ability to modulate the expression of critical genes and signaling networks. We previously demonstrated that miRNA-181a expression correlates with longer survival in patients treated with R-CHOP, independent of established clinical and molecular predictors. However, the molecular and cellular mechanisms underlying the association between miRNA-181a expression and improved prognosis in DLBCL patients are currently unknown. Herein we analyzed the role of miRNA-181a in DLBCL pathogenesis. Results:Quantitative RT-PCR analyses demonstrate higher endogenous miRNA-181a levels in centroblasts than in plasmablasts. Concordantly, endogenous miRNA-181a levels were significantly higher in GCB DLBCL cell lines and primary tumors compared with ABC DLBCL. These expression differences could not be attributed to distinct DNA methylation signatures in the miRNA-181a promoters (Chromosomes 1, 9) or regulatory elements as analyzed by Mass Array Sequenom Epityping. In search for putative miRNA-181a targets we identified 5 genes (CARD11, NFKB1A (IKBα), NFKB1 (p105/p50), RELA (p65), and REL (CREL)) within the NF-kB signaling pathway. Analyses of these targets show a decrease in the levels of these proteins and mRNAs in ABC and GCB DLBCL cell lines ectopically expressing miRNA-181a compared with scramble control plasmid. Luciferase reporter analyses encoding the respective wild type or mutated 3′UTR sequences demonstrate direct and specific targeting of these transcripts with the exception of RELA. Analysis of the net effect of miRNA-181a on NF-kB signaling using NF-kB luciferase reporter demonstrate significant decrease in NF-kB signaling. Concordantly, anti-miRNA-181a transfection led to increased NF-kB luciferase reporter activity. Moreover, western blot analyses of cytoplasmic and nuclear fractions showed a decrease in the levels of the transcription factors CREL and p50 in both cellular compartments, a decrease in the binding to DNA at NF-kB binding motifs, and a consequent decrease in NF-kB target gene transcription in the miRNA-181a expressing cells compared with scramble control. Together these studies point to miRNA-181a-mediated repression of NF-kB signaling in DLBCLs. Ectopic miRNA-181a expression led to a decrease in cell proliferation and an increase in cell death in both DLBCL subtypes, but this effect was more pronounced in the ABC DLBCL cell lines. The miRNA-181a-mediated increase in cell apoptosis could not be rescued by BCL2 co-transfection, an anti-apoptotic protein that was previously established as a direct miRNA-181a target. Analyses of miRNA-181a effects in NOD/SCID mice demonstrated that in vivo miRNA-181a induction in GCB and ABC human DLBCL xenografts led to decreased tumor growth and significantly longer animal survival. Notably, survival was prolonged in both GCB and ABC DLBCL bearing animals. Figure 1 Figure 1. Conclusions: miRNA-181a directly suppress the NF-kB signaling pathway and lead to increased tumor cell death in both DLBCL subtypes suggesting that NF-kB deregulation is present in both tumor subtypes. However, the lower miRNA-181a expression level in the ABC DLBCL subtype may contribute to the higher NF-kB signaling activity that is observed in this subtype. Furthermore, our study provides a plausible explanation for the association between high miRNA-181a expression and longer survival of DLBCL patients. Disclosures No relevant conflicts of interest to declare.

NF-κB Subunit c-Rel Cooperates with Myc and Mutated p53 to Confer Significantly Worse Survival in Patients with Diffuse Large B-Cell Lymphoma: A Report from the International DLBCL Rituximab-CHOP Consortium Program

Objective :To evaluate the prognostic significance of RELamplification, expression and c-Rel activation in DLBCL patients and to identify potential mechanisms for impact of c-Rel activation on patient survival. Patients and Design : The study cohort consists 460 de novo DLBCL patients (median follow-up, 46.8 moths) treated with R-CHOP. We assessed the nuclear expression/activation of c-Rel and other NF-κB subunits by immunohistochemistry, REL gene amplification by fluorescence in situ hybridization, and gene expression profiling using Affymetrix GeneChips array. Correlations between expression of nuclear c-Rel, REL mRNA, and expression of TP53, MDM2, MDM4, MYC, BCL2, AKT1, NFKB1, RELA, NFKB2, and RELB, both at the mRNA and protein levels were analyzed using t-tests. The prognostic significance of c-Rel activation, REL mRNA expression, and REL amplification was evaluated in the overall cohort, and different subgroups stratified by COO, status of TP53 mutation (wide type/WT, or mutated/MUT) and expression, Myc, Bcl-2 overexpression, and nuclear expression of other NF-κB subunits. Results :Nuclear c-Rel expression was observed in 29.6% of DLBCL patients and did not correlate with REL mRNA levels (P=0.95) and COO (P=0.77). In contrast, REL mRNA was significantly higher in germinal center B-like (GCB) subtype (P<0.0001). In GCB-DLBCL, nuclear c-Rel expression was associated with significantly lower Myc, p53, MDM4, and pAKT protein levels but not at the transcriptional level. In contrast, in ABC-DLBCL, c-Rel activation was associated with significantly higher MUT-TP53mRNA level and reduced pAKT expression. Correspondingly to the lack of associations with reduced Myc, pAKT, and p53 in ABC-DLBCL, nuclear c-Rel expression showed prognostic impact only in ABC- but not in GCB- DLBCL, especially when it was concurrent with Myc overexpression (P<0.0001 for OS and P=0.0012 for PFS). Importantly, patients with c-Rel activation and p53 mutations had significantly worse survival (for OS, hazard ratio, 3.56; P=0.0011; median survival, 16.2 vs 87.3 months. For PFS, hazard ratio, 3.976; P=0.0004; median survival, 10.4 vs 55.5 months) compared with other MUT-p53 DLBCL patients. The additive impact of c-Rel activation to TP53 mutations was more apparent in the ABC-DLBCL subtype, in which c-Rel activation was associated with significantly upregulated MUT-TP53 transcription (P=0.0087). Conversely, MUT-p53 expression was associated significantly with upregulated REL mRNA expression (P=0.0021), predominantly in the ABC-DLBCL subtype (P=0.0047). Only in ABC-DLBCL patients, TP53 mutations were associated with elevated nuclear c-Rel levels with a borderline P value (P=0.05). In addition to the association of nuclear p65 in GCB-DLBCL (P=0.003), and p50, NFKB1 and RELA mRNA in ABC-DLBCL (P=0.0023), the prognostic significance of c-Rel appears to depend on p50 (P=0.08) and p65 expression (P=0.12). Also, supporting that c-Rel transactivates anti-apoptotic BCL2L1/BCLXL by previous studies, there was no significant difference in survival of ABC-DLBCL patients with isolated BCL2 overexpression and with nuclear c-Rel expression. Comprehensive gene-expression profiling analysis and cell line study are undergoing in order to identify pathways to activate c-Rel and oncogenic mechanisms for c-Rel–mediated chemoresistance. REL amplification was predominantly observed in GCB-DLBCL (frequency, 7%) and correlated with significantly higher mRNA level (P<0.0001). However, REL amplification did not correlation with nuclear c-Rel expression or patient survival, illustrating the importance of posttranslational regulations in c-Rel activation and function. Consistent with the adverse impact of c-Rel activation at the protein level, a strong trend toward poor survival was observed for elevated RELmRNA in ABC- but not in GCB-DLBCL. Conclusions : Nuclear c-Rel activation was associated with reduced level of proteins whose degradation involves with ubiquitin-proteasome in GCB-DLBCL, and upregulated TP53 transcription in ABC-DLBCL. Reciprocal regulation of c-Rel and MUT-p53 at the transcriptional level may underlie the synergetic adverse effect of c-Rel activation and TP53 mutations. c-Rel cooperated with Myc and conferred significantly worse survival in ABC-DLBCL patients. These subsets of c-Rel+ DLBCL patients likely will benefit from c-Rel targeted therapies. Disclosures No relevant conflicts of interest to declare.