Structure-Based Optimization of the Third Generation Type II Macrocycle TRK Inhibitors with Improved Activity against Solvent-Front, xDFG, and Gatekeeper Mutations.

The solvent-front (SF), gatekeeper, and xDFG motif mutations of tropomyosin receptor kinase (TRK) mediating acquired resistance of larotrectinib and entrectinib represent an unmet clinical need. To date, no effective drugs are being approved to overcome these mutants. Thus, a series of macrocycle compounds were designed and synthesized as new type II TRK inhibitors to combat clinically relevant mutations. The representative compound 10g exhibited excellent potency against wide type TRKA/C, TRKAG595R, TRKAG667C, and TRKAF589L with IC50 values of 5.21, 4.51, 6.77, 1.42, and 6.13 nM, respectively, and a good kinome selectivity against 378 kinases. 10g also strongly suppressed the proliferation of Ba/F3 cells transfected with SF, GK, xDFG, and others (Val to Met) single mutants with IC50 values of 1.43-47.56 nM. Moreover, 10g demonstrated ideal antitumor efficacy in both BaF3-CD74-NTRK1G595R and BaF3-CD74-NTRK1G667C xenograft models. The study provides a promising lead compound for pan-anticancer drug discovery.

Switch type I to type II TRK inhibitors for combating clinical resistance induced by xDFG mutation for cancer therapy.

TRK xDFG mutation-induced acquired resistance of 1st generation inhibitors larotrectinib and entrectinib remains an unmet clinical need. Here we report a series of 6-(pyrrolidin-1-yl)imidazo[1,2-b]pyridazine-based derivatives as selective type II TRK inhibitors by hybridization. A representative compound 12d potently inhibited TRKA/B/C and TRKAG667C with IC50 values of 3.3, 6.4, 4.3 and 9.4 nM, respectively. 12d potently suppressed proliferation of a panel of Ba/F3 cells stably transformed with wild type, xDFG as well as solvent-front (SF) mutant TRK fusion proteins. Compared with larotrectinib and selitrectinib, 12d displayed superior inhibitory activity towards Ba/F3 cells harboring CD74-TRKAG667C and ETV6-TRKCG696C with IC50 values of 2.6 and 6.1 nM, respectively. Moreover, 12d also exhibited potent antiproliferation activity against Ba/F3-ETV6-TRKCG623R and Ba/F3-ETV6-TRKCG623E mutants with IC50 values of 31.0 and 28.2 nM, respectively. This work provided a new potential type II TRK inhibitor-based lead compound for the treatment of TRK driven cancers.

Discovery of the First Highly Selective and Broadly Effective Macrocycle-Based Type II TRK Inhibitors that Overcome Clinically Acquired Resistance.

Tropomyosin receptor kinase (TRK) secondary mutations mediating acquired resistance, especially at the solvent-front (SF) and the DFG motif, represent an unmet clinical need. Small-molecule macrocyclic kinase inhibitors have displayed significant advantages in overcoming clinical resistance driven by kinase mutations; however, all reported small-molecule macrocyclic TRK inhibitors are all type I inhibitors and are therefore much more sensitive to SF than xDFG mutations. Novel therapeutics for patients with xDFG resistance mutations are urgently needed. We report the first highly selective macrocycle-based potent type II TRK inhibitor, 7b, that exhibits high inhibitory potency toward various TRK fusion protein variants as well as wild type. 7b exhibited potent antiproliferative activity against Ba/F3 cells harboring CD74-TRKAG667C and ETV6-TRKCG696C with half-maximum inhibitory concentration (IC50) values of 6 and 1.7 nM, respectively. More importantly, 7b also showed potent antiproliferative activity against a panel of SF mutants (IC50 = 5.6-110 nM) and displayed extraordinary kinome selectivity.

Discovery of novel TrkA allosteric inhibitors: Structure-based virtual screening, biological evaluation and preliminary SAR studies.

Tropomyosin receptor kinases A (TrkA) is a potential therapeutic target for the treatment of numerous tumor types and chronic pain. However, most of the reported TrkA inhibitors are ATP competitive pan-Trks inhibitors that lack subtype selectivity. A selective TrkA inhibitor may provide valuable therapeutic benefits. Here, we described the discovery of novel TrkA allosteric inhibitors by structure-based virtual screening. A promising hit (D5261, TrkA cell IC50 = 3.32 µM) was selected for further studies. The binding free energy between TrkA and D5261 was calculated. In addition, the preliminary structure-activity relationship (SAR) studies with D5261 were investigated. The results suggest that D5261 can be used as a starting point for development of TrkA allosteric inhibitors.

Deep learning-driven scaffold hopping in the discovery of Akt kinase inhibitors.

Scaffold hopping has been widely used in drug discovery and is a topic of high interest. Here a deep conditional transformer neural network, SyntaLinker, was applied for the scaffold hopping of a phase III clinical Akt inhibitor, AZD5363. A number of novel scaffolds were generated and compound 1a as a proof-of-concept was synthesized and validated by biochemical assay. Further structure-based optimization of 1a led to a novel Akt inhibitor with high potency (Akt1 IC50 = 88 nM) and in vitro antitumor activities.