A Non-covalent KRASG12D Allele Specific Inhibitor Demonstrates Potent Inhibition of KRAS-dependent Signaling and Regression of KRASG12D-mutant Tumors

Abstract

The ability to effectively target mutated KRAS has remained elusive despite decades of research. The recent identification of KRASG12C inhibitors has provided an effective treatment option for patients harboring this particular mutation and has also provided insight toward targeting other KRAS mutants, including KRASG12D. MRTX1133 was identified via a structure-based drug design (SBDD) strategy as a potent, selective, and non-covalent KRASG12D inhibitor directed at the switch II binding pocket. MRTX1133 demonstrated a high-affinity interaction with KRASG12D with KD or IC50 values each determined at ~0.2 pM or <2 nM using SPR direct binding or HTRF competition assays, respectively. MRTX1133 also demonstrated ~700-fold selectivity for KRASG12D vs KRASWT binding utilizing SPR. Interestingly, MRTX1133 demonstrated potent inhibition of active KRASG12D using an HTRF effector interaction assay with a IC50 value of 9 nM. Insight toward the structural basis of binding of MRTX1133 to both the inactive GDP-bound and active GMPPCP-bound conformations of KRASG12D is also provided by co-crystal structures. MRTX1133 demonstrated potent inhibition of ERK1/2 phosphorylation and cell viability in KRASG12D-mutant cell lines with median IC50 values of ~5 nM. Consistent with binding affinity determination in cell-free systems, MRTX1133 demonstrated >1000-fold selectivity for inhibition of ERK1/2 phosphorylation in KRASG12D-mutant cell lines compared with KRASWT cell lines. Dose-dependent inhibition of KRAS-mediated signal transduction was also observed in multiple KRASG12D-mutant tumor models in vivo. MRTX1133 demonstrated marked tumor regression (>30%) in a subset of KRASG12D-mutant cell line- and patient-derived xenograft (PDX) models, including 8 out of 11 (73%) pancreatic ductal adenocarcinoma (PDAC) models evaluated. Pharmacological studies and CRISPR-based screens demonstrated co-targeting KRASG12D in concert with putative feedback or bypass pathways including EGFR and PI3Kα led to enhanced anti-tumor activity relative to targeting each individual protein. Together, these data indicate the feasibility of utilizing SBDD approaches to selectively target alternative KRAS mutant variants with non-covalent, high-affinity small molecules targeting either the active or inactive state of KRAS. In addition, these data illustrate the therapeutic susceptibility and broad dependence of KRASG12D mutation-positive tumors, including PDAC, on KRAS for tumor cell growth and survival.

More to the RAS Story: KRASG12C Inhibition, Resistance Mechanisms, and Moving Beyond KRASG12C

“…27,47,[80][81][82] A crystal structure of KRAS G12D bound to MRTX-1133 has recently been resolved or described and will serve as a template for more advanced modeling and drug design for this allele in the coming years. 83 KRAS G12V is the second most common point mutation affecting KRAS in cancer, occurring in approximately 9% and approximately 27% of colorectal and pancreatic adenocarcinomas, respectively. 47,79 Similar to our understanding of KRAS G12D , it is expected that valine will not be susceptible to covalent targeting approaches and that the GTP-bound state of KRAS G12V should be prioritized for putative inhibitors, given its relative rate of hydrolysis and nucleotide exchange.…”