New Molecular and Therapeutic Insights into Canine Diffuse Large B Cell Lymphoma Elucidates the Role of the Dog As a Model for Human Disease

Background. Diffuse large B-cell lymphoma (DLBCL) is the commonest lymphoma in both humans and dogs. Canine DLBCL (cDLBCL) is considered an ideal comparative model for drug development, but a complete genomic characterization of this tumor is still lacking. In this study, we report an integrated analysis to comprehensively define the molecular mechanisms of cDLBCL and possible associations with clinical outcome.

Methods. Fifty cDLBCLs were analyzed by RNA-Seq, methyl-CpG-binding sequencing and array comparative genomic hybridization. Normal B-cells derived from lymph nodes of 11 healthy dogs were used as controls.Additionally, immunohistochemistry, in vitroand in vivoexperiments were performed as validation analyses.

Results.Compared to normal B-cells, cDLBCL showed a marked up-regulation of genes involved in the PI3K/mTOR and NF-κB pathways, including several TLRs in association with MYD88, indicating mechanisms similar to the human activated B cell-like subtype DLBCL.

Both RNA-Seq and methylation sequencing led to the identification of two groups of cDLBCLs bearing different clinical outcome. The two groups did not overlap with the human germinal center B-cell (GCB) and the activated B-cell-like (ABC) DLBCL subtypes or the human DLBCL consensus clusters. The dogs with the poorest outcome presented a signature largely defined by markers of T-cell-mediated immune responses, with a high expression of PDL-1, PD-1 and CTLA-4, also validated in an independent cohort of cDLBCL by immunohistochemistry. These data provide a strong rationale for the use of cDLBCL to study immune checkpoint modulators.

The observed high expression of PI3K/mTOR pathway genes was confirmed and validated achieving a clear anti-tumor activity with the use of the PI3K-delta inhibitor idelalisib and of the novel dual PI3K/mTOR inhibitor bimiralisib in the cDLBCL cell line CLBL-1.

The cDLBCLs showed an up-regulation of MYC and of its targets, sustained by recurrent gains in the chromosome 13, where the oncogene is located, in approximately half of the cases. Thus, we have exposed the cDLBCL cell line CLBL-1 to the BET inhibitor birabresib (OTX015) and to the BRD4 degrader MZ1. Both compounds caused a significant reduction in the proliferation of tumor cells, and this effect was stronger especially with the second compound. Exposure to MZ1 determined an important downregulation of MYC and also of LIN28B, the most overexpressed transcript in cDLBCL when compared to controls. While LIN28B does not seem to be a relevant gene for human DLBCL, its overexpression causes murine T-cell lymphomas (Beachy et al, Blood 2011), and there is a direct association of MYC with LIN28B promoter resulting in transcriptional transactivation (Chang et al, PNAS 2009). Here, LIN28B genetic silencing in the CLBL-1 lead to a reduction in cell growth, opening new therapeutic target perspectives in canine lymphoma.

Conclusions. We have reported the first large next generation sequencing study investigating the cDLBCL transcriptome, methylome and the genome-wide CNVs. We identified deregulated pathways and individual transcripts providing therapeutic targets, including an immune-related signature affecting the outcome of a subgroup of cDLBCL. Our data sustain the use of cDLBCL as comparative models for human DLBCL but also highlight differences that must be kept in consideration.