Abstract B046: Pseudo-irreversible, pan-KRAS inhibitor BH-501284 achieves prolonged inhibition of KRAS signaling leading to deep and durable antitumor efficacy in mouse tumor models
Wei Deng, Nancy Ling, Zhenping Wang, Yue Hu, Danan Li, Joshua Choi, Zhengwei Peng, Eugene Rui, Geoffrey Oxnard, Jean CuiAbstract
Background:
KRAS is a frequently mutated oncogene in pancreatic cancer. Trial success with Switch-II covalent KRAS G12C inhibitors and panRAS tricomplex inhibitors validates KRAS mutations as targetable oncogenic drivers. Yet the development of non-covalent KRAS inhibitors has remained challenging due to the shallow Switch-II pocket, resulting in many clinical molecules with high molecular weight and limited oral bioavailability. Many of these non-covalent Switch-II inhibitors share similar chemical scaffolds and consequently exhibit common clinical liabilities, such as high doses with limited clinical activity. We believe a novel Switch-II inhibitor design will be needed to achieve clinical success, specifically through long residence leading to prolonged inhibition of KRAS signaling and sustained pharmacologic effect for deeper and more durable efficacy.
Methods & Results:
BH-501284 was designed and optimized as a potent, non-covalent, pseudo-irreversible, pan-KRAS inhibitor with long residence time. In SPR assays, BH-501284 exhibited high binding affinity and long residence time on KRAS mutant proteins (>54 hours on KRAS G12D). In cell line studies, BH-501284 potently inhibited cell proliferation with a median IC50 of 0.81 nM and suppressed ERK phosphorylation similarly in KRAS-mutant (G12V/C/D/A/F/S, G13D) and KRAS amplified models but was not active against HRAS- and NRAS-mutant models. Studying H358 NSCLC cell lines carrying KRAS G12C, BH-501284 demonstrated prolonged inhibition of KRAS signaling (consistent with its long residence time) and induced deep tumor regression as effectively as the covalent KRAS G12C inhibitor divarasib. In xenograft tumor models of PDAC, NSCLC or CRC harboring various KRAS mutations, BH-501284 showed durable antitumor activity at relatively low dose levels, in some models as low as 5mg/kg BID. In the Panc 04.03 CDX PDAC model carrying KRAS G12D, BH-501284 induced deeper tumor regression compared to daraxonrasib or zoldonrasib given at similar doses. In the CT26 syngenetic murine CRC model carrying KRAS G12D, the combination of BH-501284 plus anti-PD1 antibody showed synergistic activity with survival prolongation compared to each single agent alone.
Conclusions:
Building upon the success of covalent Switch-II KRAS G12C inhibitors, we designed and discovered BH-501284 to be a novel non-covalent Switch-II inhibitor that exhibits high binding affinity and pseudo-irreversible characteristics, resulting in prolonged suppression of KRAS signaling and potent antitumor activity across multiple KRAS-mutant cancer models – while sparing H/NRAS – in vitro and in vivo. These data support the potential of novel non-covalent Switch-II pocket KRAS inhibitors to address the limitations of available KRAS inhibitors and strongly support advancement of BH-501284 into first-in-human clinical studies.
Citation Format:
Wei Deng, Nancy Ling, Zhenping Wang, Yue Hu, Danan Li, Joshua Choi, Zhengwei Peng, Eugene Rui, Geoffrey Oxnard, Jean Cui. Pseudo-irreversible, pan-KRAS inhibitor BH-501284 achieves prolonged inhibition of KRAS signaling leading to deep and durable antitumor efficacy in mouse tumor models [abstract]. In: Proceedings of the AACR Conference on Pancreatic Cancer: New Frontiers in Biology and Therapeutic Development; 2026 Sep 25-28; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(18_Suppl_2):Abstract nr B046.