EGFR kinase mutations in NSCLC: Unexpected structural and kinetic insights into the mechanism of the T790M resistance mutation and its inhibition by irreversible inhibitors including HKI-272

Cai-Hong Yun,Michelle Woo,Heidi Greulich,Matthew Meyerson,Kwok-Kin Wong,Michael J. Eck
IF: 11.2
2007-01-01
Cancer Research
Abstract:LB-367 Somatic mutations in the EGFR kinase are a cause of non-small cell lung cancer, and the presence of these mutations correlates with response to small-molecule tyrosine kinase inhibitors. Despite initial responses, longer-term efficacy of TKIs targeting the EGFR has been limited by the emergence of drug resistance, often conferred by an additional mutation of Threonine 790 in the EGFR to Methionine (T790M). This mutation, which lies in the ATP binding pocket of the kinase, confers resistance to anilinoquinazoline inhibitors including gefitinib and erlotinib. Importantly, the T790M mutant remains sensitive to a number of irreversible inhibitors, including HKI-272, Cl-387,785 and Cl-1033. These irreversible inhibitors are expected to bind in the ATP binding cleft in a manner analogous to similar reversible inhibitors, but they have an additional reactive group that forms a covalent bond with Cys 773 at the edge of the ATP binding cleft. The activity of these inhibitors against the T790M mutant therefore presents a paradox â\#8364;“ the steric effects of the T790M mutation that are presumed to underlie resistance to reversible anilinoquinazoline inhibitors should also interfere with the targeting of the irreversible inhibitors. To better understand the effects of the T790M mutant and the mechanism by which it confers resistance, we have determined the structure of this mutant alone and in complex with irreversible inhibitor HKI-272. Additionally, we have studied the kinetics and inhibitor-binding properties of the T790M mutant, the L858R/T790M double mutant, and the wild-type EGFR kinase. Surprisingly, using a direct binding assay we find that gefitinib binds to the T790M and L858R/T790M mutants with low nanomolar affinity (Kd=~5 nM), indicating that the T790M substitution does not sterically block gefitinib binding. These results are further supported by the crystal structure of HKI-272 in complex with the T790M mutant, which shows that the binding mode of HKI-272 is the same as that expected in the wild-type kinase. Kinetic analysis of the T790M mutant shows that it is approximately 5-fold more active that the wild-type kinase, with a similar Km for ATP. Interestingly, the L858R/T790M double mutant also has a Km for ATP equivalent to that of the wild-type kinase, whereas the L858R and G719S single mutants have markedly impaired affinity for ATP. Collectively, our findings indicate that much of the resistance induced by the T790M mutation may stem from its ability to restore wild-type affinity for ATP, rather than from its steric effects on inhibitor binding as was previously supposed. Since TKIs must compete with ATP to achieve their intended effect, their effective potency is diminished by the enhanced ATP-affinity. Irreversible inhibitors, as a class, overcome this effect through covalent binding.
What problem does this paper attempt to address?