Despite the success of first, second and third generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) in treatment of non-small cell lung cancer (NSCLC) with classical EGFR mutations (L858R or Exon 19 deletions), disease progression often occurs due to the acquisition of additional mutations in the EGFR kinase domain that confer TKI resistance. Specifically, acquisition of both T790M and C797S resistance mutations results in an EGFR variant that is resistant to all approved EGFR TKIs. Herein, we report a physics-based computationally-driven lead identification approach which successfully identified structurally-unique imidazo[3.2-b]pyrazole derivatives as reversible inhibitors of EGFR classical mutations bearing both T790M and C797S. Importantly, they spare EGFR WT to avoid known EGFR WT-driven cutaneous toxicities. During profiling of imidazo[3.2-b]pyrazole derivatives, we elucidated the bioactivation mechanism causing CYP3A4/5 time-dependent inhibition (TDI) and found key modifications to suppress bioactivation and mitigate the TDI risk. Representative lead compound 31 inhibited EGFR L858R/T790M/C797S in biochemical assays with a Ki = 2.1 nM, and EGFR del19/T790M/C797S in a Ba/F3 cellular assay with a IC50 = 56.9 nM. Deuterated analog of 31 (38) demonstrated dose-dependent tumor growth inhibition in a Ba/F3 EGFR del19/T790M/C797S CDX model by 47% at 50 mg/kg BID and 92% at 100 mg/kg BID.
Dual leucine zipper kinase (DLK), expressed primarily in neuronal cells, is a regulator of neuronal degeneration in response to cellular stress from chronic disease or neuronal injury. This makes it an attractive target for the treatment of neurodegenerative diseases such as Alzheimer's, Parkinson's, and amyotrophic lateral sclerosis; and neuronal injury, such as chemotherapy induced peripheral neuropathy. Here we describe the discovery of a potent, selective, brain-penetrant DLK inhibitor, KAI-11101 (59). Throughout the program's progression, medicinal chemistry challenges such as potency, hERG inhibition, CNS penetration, CYP3A time-dependent inhibition, and kinase selectivity were overcome through the implementation of cutting-edge in silico tools. KAI-11101 displayed an excellent in vitro safety profile and showed neuroprotective properties in an ex vivo axon fragmentation assay as well as dose-dependent activity in a mouse PD model.
The blood-brain barrier (BBB) plays a critical role in preventing harmful endogenous and exogenous substances from penetrating the brain. Optimal brain penetration of small molecule CNS drugs is characterized by a high unbound brain/plasma ratio (Kp,uu). While various medicinal chemistry strategies and in silico models have been reported to improve BBB penetration, none were developed to predict Kp,uu directly. We describe a physics-based computational approach, solvation free energy calculations (energy of solvation or E-sol), to predict Kp,uu. Prospective application of this method to internal CNS drug discovery programs highlighted the utility and accuracy of this new method, which showed a categorical accuracy of 79% and a R2 of 0.61 from a linear regression model.