One of the most stubborn challenges in oncology is that tumors engineer their own survival environment — starving themselves of oxygen while simultaneously rewiring immunity, invading new tissue, and resisting treatment. The proteins at the center of this adaptive machinery, HIF-1 and HIF-2, have been recognized as high-value targets for decades, yet selective inhibition of only one isoform has proven insufficient for many cancers. A new approach may fundamentally shift that calculus.
Using computer-aided drug discovery paired with cell-based reporter screening, investigators identified a class of dual HIF-1α/HIF-2α inhibitors that bind directly to highly conserved structural domains shared across both proteins. The lead compound, 1.21S9N, prevents dimerization with HIF-1β and routes both HIF-α subunits toward proteasomal degradation, shutting down transcriptional programs governing angiogenesis, immune evasion, epithelial-mesenchymal transition, and metastasis simultaneously. Tested across breast, colorectal, head and neck, melanoma, and prostate tumor models, 1.21S9N demonstrated monotherapy efficacy and — critically — restored responsiveness to anti-CTLA-4 and anti-PD-1 checkpoint immunotherapy, yielding an aggregate complete response rate exceeding 50%. Against breast and colorectal cancers specifically, 1.21S9N outperformed both belzutifan and PT2385, the approved and investigational HIF-2-selective agents, while avoiding the breathing abnormalities associated with PT2385. The compound showed oral bioavailability and no identified safety signals at extended or supratherapeutic doses.
The breadth of tumor types addressed here is notable, and the immunotherapy synergy data are particularly compelling given the now-central role of checkpoint blockade in oncology. Dual HIF inhibition addresses a mechanistic gap: HIF-1 dominates early hypoxic response and glycolytic reprogramming, while HIF-2 governs stem cell phenotype and some immune evasion pathways — targeting only one leaves the other to compensate. That said, all data reported here appear to be preclinical, likely murine tumor models, meaning translation to human pharmacokinetics, toxicity, and efficacy requires rigorous clinical validation. This is nonetheless one of the more mechanism-rich and cross-tumor-type preclinical packages to emerge in cancer biology this year, and the checkpoint-rescue effect alone warrants accelerated clinical investigation.