Genetically engineered mouse model to improve therapy of NUT carcinoma

NUT carcinoma (NC) is an aggressive squamous carcinoma driven by the BRD4-NUT fusion oncoprotein. NC affects all ages and is highly lethal (>90%), a median survival of 6.5 months. There are no effective treatment options for NC; thus, this disease represents an extreme unmet need. The overarching goal of this proposal is

to improve survival of these patients through mechanism-driven identification and testing of therapeutic targets. In response to the need for an immunocompetent animal model in which both NC cell-autonomous and -non- autonomous pathogenic mechanisms can be investigated rigorously, we have developed the first genetically

engineered mouse (GEM) model of NC. As a squamous carcinoma, NC serves as a paradigm for fusion oncoprotein-driven solid tumors. The NC GEM will expand that paradigm to understanding how tumor intrinsic and extrinsic interactions sustain NC growth. Mechanistically, BRD4 is a BET family protein whose dual bromodomains bind acetyl-histones that when fused

to NUT, recruits the histone acetyltransferase, p300, forming enormous super-enhancers called megadomains. BRD4-NUT megadomains maintain expression of pro-growth, anti-differentiation transcription factors including MYC, SOX2, and TP63. Our demonstration that treatment with BET bromodomain inhibitors (BETi), small

molecules that competitively inhibit binding of BET bromodomains to chromatin, can inhibit growth of NC in humans, led to a new field investigating the role of BRD4 in cancer. However, it has become clear that monotherapy with BETi does not fully address NC biology. We have recently found that repression of tumor

suppressor gene expression, such as that of CDKN2A/B, by the histone methyltransferse, EZH2, highly complements oncogenic activation by BRD4-NUT in maintaining NC growth. Targeting of this pathway with tazemetostat (taz) is highly synergistic with BETi and will be explored in the proposed GEM model.

It is now recognized that NC harbors an immune-evasive tumor microenvironment (TME) and can respond to immune modulation therapy. Epigenetic modulators such as taz and BETi are known to promote an anti-tumor immune TME. Thus, pre-clinical animal models with intact immune systems such as our NC GEM are needed to

fully evaluate effects of epigenetic modifiers, and the role of immune therapy in this disease. Our GEM has a tamoxifen-inducible, conditional knock-in fusion of murine Brd4 with human NUTM1, encoding a BRD4-NUT fusion oncoprotein. Invasive tumors formed in our NC GEM (`mNC') have provided the most definitive evidence

that BRD4-NUT is the sole driver of this cancer. mNC closely mimics human NC, demonstrating rapid growth, metastatic spread, and an indistinguishable histopathology and immunophenotype. Moreover, the immune cell composition of the mNC TME also resembles that of human NC. We will make use of our novel GEM to address

the following aims: 1. establish the applicability of the NC GEM (mNC) to human NC (hNC) biology.; 2. devise improved primary therapy for NUT carcinoma.; 3. explore approaches to prevent relapse of NC.