Using mouse models to identify actionable molecular liabilities in Cluster 5 DLBCL

• Applicant: Professor Dr. Christian Reinhardt
• Subject Area: Hematology, Oncology
• Term: from 2020 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 442814095

DLBCL remains a clinical challenge, as relapsed and refractory disease is very difficult to treat. However, recent genomics efforts have revealed the landscape of recurrent aberrations in DLBCL. While this area of investigation is still in flux, a rather clear picture of ABC-DLBCL, and particularly C5 DLBCL lymphomagenesis, is emerging: A hallmark feature of C5 DLBCL is inappropriate NFB activation, which is achieved through recurrent mutations within the BCR (CD79B) and/or TLR (MYD88) pathways. A further hallmark of C5 DLBCL are BCL2 copy number gains. Lastly, C5 DLBCL is characterized by recurrent inactivating aberrations in PRDM1, which lead to a block in plasma cell differentiation. These recurrent mutations may offer opportunities for targeted therapeutic interventions, which may allow chemotherapy-free treatment approaches for this disease. However, there is currently a lack of suitable in vivo experimental platforms that faithfully mimic the genomic landscape of human C5 DLBCL. Here, we set out to expand on our existing efforts on C5 DLBCL modeling to develop a comprehensive and versatile in vivo preclinical platform that employs a combination of large-scale in vitro vulnerability screening, and an in vivo validation tool box to develop innovative, chemotherapy-free and genomics-guided therapeutic approaches for the treatment of C5 DLBCL patients. Our proposal builds on substantial preliminary data and we have successfully established a whole range of innovative technologies to fully capture the biology of C5 DLBCL, including longitudinal in vivo imaging, high throughput immuno-histochemistry, CRISPR/Cas9 drop out screening, CyTOF mass cytometry, 3'-RNA sequencing, whole exome and whole genome sequencing, single cell RNA sequencing and cytokine profiling. Our systems can further be exploited to accelerate the search for resistance-mediating genes and pathways, once viable targeted treatment approaches have been identified. Based on these considerations, we have formulated three specific aims:Aim 1: Generation and characterization of an inducible autochthonous mouse model harboring the dominant aberrations defining C5 DLBCLAim 2: Systematic CRISPR/Cas9 loss of function screening to identify genotype-specific vulnerabilities in C5 DLBCLAim 3: Assess the efficacy of combined BCL2-, BTK-, CD20- and PD-1/PD-L1 targeting in autochthonous C5 DLBCL models in vivoThese aims test the hypotheses that 1) oncogenic Myd88 mutations cooperate with Bcl2, Prdm1 and Cd79b aberrations to promote C5 DLBCL, 2) autochthonous mouse models of C5 DLBCL and cell line models derived thereof may serve as experimental platforms to develop and refine novel chemotherapy-free treatment algorithms of C5 DLBCL, and 3) MYD88-driven C5 DLBCL may display an PD-L1 surface expression. The proposed experiments will pave the way for the clinical development of genomics-guided precision medicine approaches to C5 DLBCL.

• DFG Programme: Research Grants