A structure-based modelling approach identifies effective drug combinations for RAS-mutant acute myeloid leukemia

Area of Science:

• Oncology
• Molecular Biology
• Pharmacology

Background:

• Activating mutations in RAS/RAF/MEK/ERK signaling pathways are linked to poor prognosis in acute myeloid leukemia (AML).
• Targeting this pathway therapeutically in AML presents significant challenges.

Purpose of the Study:

• To utilize a structure-based, dynamic RAS pathway model to predict synergistic RAF inhibitor (RAFi) combinations for RAS-mutant AML.
• To validate the efficacy of predicted RAFi combinations in vitro and in vivo models.

Main Methods:

• Development of a structure-based, dynamic RAS pathway model for *in silico* prediction of RAFi synergy.
• Validation of predicted synergistic combinations (Type I½ + Type II, Type I + Type II) using AML cell lines and patient samples.
• Assessment of combination efficacy in a pre-clinical *NRAS*-mutant AML patient-derived xenograft (PDX) model, evaluating leukemia growth delay and survival.

Main Results:

• The *in silico* model successfully predicted synergistic RAFi combinations, including Lifirafenib (Type II) + encorafenib (Type I½) and Lifirafenib (Type II) + SB590885 (Type I).
• Both combinations demonstrated significant synergy *in vitro* and reduced RAS pathway activation.
• In *in vivo* PDX models, the combinations significantly improved leukemia growth delay and event-free survival compared to single agents, with site-specific efficacy observed.

Conclusions:

• A structure-based modeling approach can effectively identify novel and synergistic RAFi combinations for RAS-mutant AML.
• These identified combinations show potent anti-leukemic activity *in vitro* and *in vivo*, offering potential therapeutic strategies for high-risk AML.
• The study highlights the utility of computational modeling in drug discovery for targeted cancer therapies.