Background: Diffuse large B-cell lymphomas (DLBCLs) are the most NF-κB-related aggressive non-Hodgkin’s lymphomas. DLBCLs are clinically, molecularly and genetically heterogeneous, suggesting different oncogenic mechanisms. Gene-expression profiling studies led to the proposal of two main cell-of-origin molecular subtypes known as germinal center B-cell-like (GCB) DLBCLs, and activated B-cell like (ABC) DLBCLs. With the Lymphoma Study Association LYSA, we previously published that strong c-Rel DNA-binding activity was mostly found in GCBs and was associated with increased REL mRNA expression. These cases with both overexpression of REL mRNA and c-Rel related gene expression profile were correlated with various recurrent GCB-DLBCLs genetic events, including REL gains, BCL2 translocation, MEF2B, EZH2, CREBBP, and TNFRSF14 mutations and with the EZB GCB genetic subtype. Aims: We propose that c-Rel is a major oncogene of GCB-DLBCLs. We speculate that c-Rel confers a selective advantage in the germinal center (GC)-specific environment. We wish to study the consequences of aberrant overexpression and activation of c-Rel in vivo that could promote tumorigenesis in DLBCL precursor cells by disrupting the normal dynamics of B-cell development in the GC. Methods: We have generated the fl_STOP-REL mouse model: the murine REL transgene preceded by a STOP-Neomycin cassette flanked of loxP sites was introduced in Rosa26 locus on chromosome 6. The REL transgene is followed by an IRES to allow YFP marker concomitant expression. To reach GC B-cell specific expression, we used AID_Cre-ert2 mice: the Cre-ert2 transgene was at the AICDA gene locus on chromosome 6. On the same chromosome, the tdTomato marker preceded by a STOP cassette flanked by loxP sites was introduced at the Rosa26 locus. A pool of REL overexpressing GC B-cell was generated after tamoxifen administration and T-cell-dependent antigen stimulation with sheep red blood cells (SRBC). The fate of these YFP and tdTomato positive GC B-cells was followed over-time until 8 months. The first four months, a SRBC immune booster was performed to maintain the GCs. Results: As expected, our model fl_STOP-REL | AID_Cre-ert2 (i.e. REL mice) allowed for in vivo competition between c-Rel overexpressing (tdTomato+/YFP+) and normal (tdTomato+/YFP-) GC B-cells. In blood, the percentage of YFP positive B-lymphocytes among the tdTomato positive cells was increased over-time: 30% at 14 days, 65% at 2 months and about 80% at 8 months. From blood, spleen and lymph nodes, phenotyping analysis by flow cytometry of REL/YFP positive B-cells indicated an activated profile compared to normal GC-derived B-cells: CD19+, IgM+, IgD-, CD21+, CD80+, and B220low. At 8 months after generation of GC B-cells pool, approximatively 3-5% of B-cells were tdTomato and YFP positives in spleen of REL mice, whereas tdTomato positives cells in control mice were rare hardly detectable (less than 0.8%). Altogether, our result indicated that c-Rel reprograms B-cell to be activated and to better survive in vivo for very long time. Image:Summary/Conclusion: We bring strong experimental arguments showing that increased c-Rel expression in GC B-cell favors their survival for several months, and would thus be a primary event. We suspect that additional oncogenic events are probably needed for tumor clonal expansion such as BCL2, EZH2, CREBBP, TNFRSF14 and/or c-Myc dysregulation.