583 Novel, potent, and selective inhibitors of hypoxia-inducible factor (HIF)-2α reverse pro-tumorigenic transcriptional programming in cancer, stromal, and immune cells

Background The microenvironment of solid tumors is hypoxic and requires induction of genes associated with metabolism, growth, proliferation, and angiogenesis for cancer cells to survive and metastasize. The master transcriptional regulators of hypoxia-induced genes are the HIF proteins, consisting of three distinct oxygen-regulated α monomers (HIF-1α, -2α, and -3α). In normoxia, hydroxylation of HIF-2α allows for recognition by the pVHL E3-ubiquitin ligase complex and proteasomal degradation. Exposure to hypoxia, or VHL mutation or silencing, leads to HIF-2α stabilization, dimerization with HIF-1β/ARNT, and transcription of pro-tumorigenic gene sets in a variety of cancer and non-cancer cell types in the tumor microenvironment. In patients, overexpression of HIF is associated with poor prognosis, and both preclinical and clinical evidence suggests that inhibiting HIF-2α is an effective strategy to mitigate tumor growth, particularly in clear cell renal cell carcinoma (ccRCC), warranting further development of HIF-2α inhibitors and investigation into the role of HIF-2α in various cellular and combinatorial settings. Methods Using a suite of assays to evaluate HIF-2α-specific effects, herein we describe pharmacological properties associated with novel, potent, and selective small-molecule inhibitors of HIF-2α. Results Optimized compounds inhibited HIF reporter transcription and VEGF secretion. Compounds that were biochemically confirmed to bind HIF-2α also inhibited HIF-2α-, but not HIF-1α-, mediated gene expression. Characterization of HIF-2α inhibition was expanded to human stromal and immune cell subsets. While compounds inhibited pro-angiogenic gene sets in endothelial cells, inhibiting HIF-2α in activated hypoxic T cells did not affect proliferation or cytokine secretion, suggesting that HIF-2α inhibitors would not impede T cell functionality in tumors. In contrast, in a M2-polarized macrophage model for suppressive tumor-associated macrophages, HIF-2α drove hypoxia-induced changes in the chemokine secretome that favored granulocytic rather than lymphocytic infiltration, an effect that was effectively reversed by HIF-2α inhibition. At the transcriptional level, mRNA-sequencing was used to define global gene sets impacted by HIF-2α inhibition in M2 macrophages. Additionally, in a set of liver, kidney, pancreatic, and colon cancer lines, CRISPR/Cas9-mediated gene editing was used to differentiate the transcriptomic profile driven by HIF-2α from that of HIF-1α or HIF-3α, allowing for the derivation of a HIF-2α-specific gene signature. Cancer cell and macrophage-derived signatures were applied to publicly available datasets to investigate cancer types, other than ccRCC, in which HIF-2α may play an important pathological role. Conclusions Collectively, these data support the development of our novel and selective HIF-2α inhibitors for the treatment of cancer and expand the indications that may benefit beyond ccRCC.