Deciphering epigenetically-regulated pathways to improve targeted therapy for invasion and metastasis in head and neck cancer

PROJECT SUMMARY Head and neck squamous cell carcinoma (HNSCC) is the sixth leading cause of cancer-related mortality, with most deaths attributable to metastasis and treatment failure. Unfortunately, our understanding of the pathways that underlie invasion and metastasis is incomplete, but some hints were uncovered by our recent single cell sequencing analyses which revealed that many HNSCC tumors contained cells that were neither fully epithelial nor fully mesenchymal, but were in a unique, hybrid epithelial-mesenchymal (hybrid-EM) state. The presence of hybrid-EM cells was more predictive of poor response than smoking history, and functional studies established hybrid-EM as a key driver of invasion and metastasis. Together, these observations establish hybrid-EM as a central pathway in HNSCC progression. However, hybrid-EM marker proteins are not druggable, so it remains challenging to therapeutically target this state. Cell states like hybrid-EM are regulated by a variety of mechanisms, but super-enhancers, in particular, have been identified as essential for maintaining cell identity. We therefore targeted super-enhancers in HNSCC using the BET inhibitor JQ1 and observed a reduction in HNSCC invasion of ~2-fold and a suppression of hybrid-EM. This result was encouraging, but effect size was modest due to the indirect and non-specific mode of action of JQ1. We hypothesized that if we could elucidate genes regulated by super-enhancers in HNSCC, we could identify molecular pathways that are more directly and specifically involved with invasion and metastasis and uncover more potent targets. Our initial experiments found that super-enhancers regulate cholesterol biosynthesis genes in hybrid-EM cells and that their knockdown (KD) robustly disrupts hybrid-EM phenotypes, with a >2000-fold and >8000-fold reduction in invasion upon KD of two mevalonate synthesis genes. Importantly, the protein products of many of the cholesterol genes we identified can be inhibited by commercially available small molecules. Indeed, we found that statins, which inhibit cholesterol biosynthesis, potently inhibit invasion (>10-fold reduction). These data provide strong support for our original hypothesis. We now propose to use CRISPRi to systematically investigate the role of cholesterol metabolism in invasion in HNSCC using sophisticated in vitro and in vivo models (Aim 1). In Aim 2, we will evaluate 9 small molecule inhibitors of cholesterol biosynthesis in cell lines, patient-derived organoids (PDOs), and in vivo pre-clinical models. Finally, in Aim 3, we will identify the full regulatory network of super-enhancers that control hybrid-EM, specifically focusing on Brd2-4 which are inhibited by JQ1. For these experiments, we will use single cell approaches to account for tumor heterogeneity and thereby uncover additional pathways beyond cholesterol biosynthesis that direct invasion and metastasis. Together, these experiments will provide critical insights into how super-enhancers regulate hybrid-EM, while exploring the possibility that by targeting the super-enhancer-regulated cholesterol biosynthesis pathway using existing FDA-approved drugs, we could circumvent a length drug development process and move quickly into human trials.