Abstract
The effectiveness of MEK inhibitors with distinct binding modes depends on cancer cell genotype.
Major finding: The effectiveness of MEK inhibitors with distinct binding modes depends on cancer cell genotype.
Concept: Distinct MEK activation and regulatory states predominate in KRAS-mutant and BRAF-mutant cells.
Impact: Use of clinically available allosteric MEK inhibitors should be guided by KRAS or BRAF mutation status.
Activating mutations of KRAS and BRAF both lead to constitutive MAPK pathway activation through the shared downstream effector MEK. However, MEK inhibitors in clinical development have had far greater single-agent activity in BRAF-mutant cancers than KRAS-mutant cells. Hatzivassiliou and colleagues profiled three allosteric MEK inhibitors in a panel of KRAS- and BRAF-mutant cancer cell lines and xenografts and found that clinically relevant doses of GDC-0973 (also known as cobimetinib) were consistently more effective against BRAF-mutant cells than KRAS-mutant cells, whereas GDC-0623 and G-573 were similarly effective in KRAS- and BRAF-mutant cells. Unlike GDC-0623 and G-573, which prevented feedback-mediated phosphorylation of MEK by RAF in both genotypes, GDC-0973 allowed feedback MEK activation in KRAS-mutant cells. Consistent with these observations, in vitro dimerization assays showed that GDC-0973 promoted BRAF–CRAF heterodimerization and induced RAF translocation to the plasma membrane, whereas GDC-0623 and G-573 did not because they stabilized MEK–RAF complex formation. Based on molecular modeling of the MEK inhibitor binding modes, the aromatic nitrogens of GDC-0623 and G-573 were predicted to form a hydrogen bond with serine 212 (S212) in the MEK activation loop, whereas the aromatic fluorine in GDC-0973 did not. The relatively weak interaction between GDC-0973 and S212 might therefore allow MEK activation loop flexibility and RAF phosphorylation of MEK as part of the negative feedback response in KRAS-mutant cells. In contrast, in BRAF-mutant cells, which generally have higher basal MEK activity than KRAS-mutant cells, GDC-0973 was more potent than GDC-0623 or G-573 due to its higher affinity for the active phosphorylated form of MEK. These findings suggest that use of MEK inhibitors that specifically target the predominant state of MEK activation in KRAS-mutant and BRAF-mutant cancers might improve clinical responses.