Preclinical Activity of RAS(ON) Multi-Selective Inhibitor RMC-7977 and Therapeutic Combinations in AML with Signaling Mutations

Constitutive hyperactivation of RAS/MAPK pathway is a limitation of targeted therapies in AML. In FLT3-mutant AML, the emergence of clones harboring RAS mutations is a major mechanism of resistance to FLT3 inhibitors (FLT3i) leading to clinical relapse. RAS mutations are also associated with resistance to the BCL2 inhibitor venetoclax and IDH1/2 inhibitors. In 5-10% of patients with de novo AML, RAS mutations are the only signaling mutations identified. Despite the critical need to inhibit RAS signaling in AML, there are no clinically-approved targeted therapies for RAS-mutant AML. RMC-7977 is a potent, oral small molecule inhibitor of both wild-type and mutant GTP-bound RAS (RAS(ON) multi-selective) and is a preclinical tool compound representative of the investigational drug RMC-6236. RMC-7977 non-covalently binds to the intracellular chaperone cyclophilin A, generating a neomorphic interface with high affinity for all RAS isoforms. The resulting tricomplexes sterically block RAS-effector interactions required for oncogenic signaling. We hypothesized that RMC-7977 would have activity in AML cells resistant to FLT3i and venetoclax due to activating RAS mutations.

Using in vitro cell viability assays, we show that RMC-7977 has single agent antiproliferative activity in AML cell lines driven by FLT3-ITD, KIT^N822K, NRAS^Q61L or KRAS^G13D mutations and restores sensitivity to gilteritinib in Molm-14 cells that acquired secondary NRAS^G12C or NRAS^Q61K resistance mutations after long-term exposure to FLT3i. Furthermore, we demonstrate that RMC-7977 has synergy with venetoclax in inhibiting proliferation and inducing apoptosis in AML cell lines resistant to single agent venetoclax.

To validate our findings in vivo, we tested the activity of RMC-7977 in a patient derived xenograft (PDX) model of FLT3-ITD AML after relapse on gilteritinib with homozygous FLT3-ITD and NRAS^Q61K co-mutations. Thirty six NSG mice were engrafted and treated orally for 24 days with: RMC-7977 (25 mg/kg, q.o.d, n=6), gilteritinib (30 mg/kg, q.d., 5 days/week, n=6), venetoclax (100 mg/kg, q.d., 5 days/week, n=6), RMC-7977 + gilteritinib (n=6), RMC-7977 + venetoclax (n=6), and vehicle (n=6). At study termination, bone marrow samples were analyzed for the presence of hCD45+ cells via flow cytometry. Mice receiving vehicle, venetoclax or gilteritinib had an average residual leukemic burden of 88.9% (SD=±23.3), 98.9% (SD=±0.4), and 79.7% (SD=±5.3) respectively, while those receiving RMC-7977 or RMC-7977 + venetoclax exhibited 46.12% (SD=±23.9) and 44.27% (SD=±15.8), respectively. Strikingly, mice treated with the combination of RMC-7977 + gilteritinib exhibited only 0.95% (SD=±0.7) hCD45+ cells (98.9% leukemic burden reduction compared to vehicle-treated controls), in keeping with our in vitro observation that RMC-7977 resensitizes FLT3-ITD and NRAS co-mutant AML to FLT3i.

We next tested the activity of RMC-7977 in a second PDX model of AML driven by a KMT2A-PICALM fusion and a NRAS^Q61L mutation. Twenty four NSG mice were engrafted and treated with RMC-7977 (10 mg/kg, q.d., n=6), venetoclax (100 mg/kg, q.d., n=6), RMC-7977 + venetoclax (n=6), and vehicle (n=6). At study termination, the bone marrow samples of mice receiving vehicle and venetoclax had a mean of 59.21% (SD=±16.9) and 50.7% (SD=±29.0) hCD45+ cells, respectively. Mice receiving RMC-7977 and RMC-7977 + venetoclax had only 0.27% (SD=±0.31) and 0.05% (SD=±0.04), respectively (99.5% and 99.9% leukemia reduction compared to vehicle). These data show that RMC-7977 +/- venetoclax significantly suppresses outgrowth of RAS-mutant AML in vivo.

Next, we performed a CFU assay with mobilized peripheral blood CD34+ cells collected from a healthy donor and observed that RMC-7977 (100 nM) alone or in combination with giltertinib (100 nM) and venetoclax (500 nM) did not significantly reduce colony formation compared to vehicle. These preclinical data suggest that hematopoietic toxicity in vivo will be limited.

Our data suggest that RAS(ON) multi-selective inhibition is an effective therapeutic approach to overcome RAS/MAPK-mediated resistance to FLT3i and is active in AML driven by oncogenic RAS mutations, addressing limitations of current standard of care treatments for AML with signaling mutations. These results provide a strong preclinical rationale for the clinical investigation of RAS(ON) multi-selective inhibition in FLT3- and RAS- mutant AML.