Key Points• TRAF3 is genetically inactivated in a substantial fraction of cBCLs.• Focal genetic loss of TRAF3 is recurrent in human DLBCLs.Non-Hodgkin lymphomas (NHLs) are the most common cancer to affect pet dogs. In contrast to the many genes whose mutation contributes to lymphomagenesis in humans, relatively little is known about the acquired genetic alterations that lead to canine B-cell lymphomas (cBCLs). We performed a survey of 84 canine NHL tumors to identify genes affected by somatic point mutations. We found mutations affecting TRAF3, which encodes a negative regulator of nuclear factor (NF)-kB, to be a common feature of cBCLs, with mutations observed in 44% of tumors including a combination of somatic and rare germ-line variants. Overall, 30% of the tumors contained ‡1 somatic TRAF3 mutation. The majority of mutations are predicted to cause loss of TRAF3 protein including those impacting reading frame and splicing. To determine whether TRAF3 loss might be relevant to human NHL, we also analyzed 148 human diffuse large B-cell lymphoma (DLBCL) tumors and identified loss of TRAF3 as a common event, affecting ∼9% of DLBCLs, and reduced expression of TRAF3 among deleted cases. This study implicates mutations affecting NF-kB activity as a novel genetic commonality between human and canine NHLs and supports the potential utility of cBCLs with mutated TRAF3 as a model of the more aggressive activated B-cell subgroup of DLBCL. (Blood. 2015;125(6):999-1005) Introduction Domestic dogs have potential as a clinical model for a variety of human cancers. In contrast to transgenic laboratory animal models in which cancers have been induced, dogs spontaneously develop tumors at a rate comparable to humans.1 Some of the tumors that commonly arise in canines appear histopathologically similar to those of humans, suggesting that they may also share similar genetic features to their human counterparts. Additional benefits of dogs as models of human cancer include a larger body size than other model organisms, a shared living environment with humans, and a higher sequence homology of known cancer genes with humans relative to mice.2 Dogs also age fiveto eightfold faster than humans, and this accelerated lifespan affords the opportunity to observe more rapid response to experimental cancer treatments. 3 Before canine B-cell lymphoma (cBCL) can be considered as a relevant model of human non-Hodgkin lymphoma (NHL) for the purpose of testing investigative compounds, it is important to capture the genetic commonalities and differences between these diseases and particularly any similarities in genetic pathways that are targeted by emerging therapeutics.NHLs collectively represent the seventh most common group of cancers among humans in the United States, and their incidence continues to rise. 4,5 NHLs are the most common cancer to afflict dogs, with a strong enrichment of certain malignancies in individual breeds. 6,7 Human diffuse large B-cell lymphoma (hDLBCL) is an aggressive type of NHL that can be subclassified into 2 m...

No abstract