Validated HPLC-MS/MS method for quantitation of AMG 510, a KRAS G12C inhibitor, in mouse plasma and its application to a pharmacokinetic study in mice.

AMG 510 is the first-in-class KRASG12C inhibitor, currently in phase 2 clinical trials as an orphan drug to treat non-small cell lung cancer patients. We developed and validated a sensitive, selective, and high-throughput HPLC-MS/MS method for the quantitation of AMG 510 in mouse plasma per the regulatory guideline of the US Food and Drug and Administration. AMG 510 and the IS (MRTX-1257) were extracted from mouse plasma using tert-butyl methyl ether and chromatographed using an isocratic mobile phase (0.2% formic acid:acetonitrile; 25:75, v/v) at a flow rate of 0.65 mL/min on an Atlantis dC18 column. AMG 510 and the IS eluted at ~0.95 and 0.73 min, respectively. AMG 510 and the IS were detected by positive electrospray ionization in multiple reaction monitoring using transition pair (Q1 → Q3) m/z 561.1 → 134.1 and m/z 566.5 → 98.2, respectively. Excellent linearity was achieved in the concentration range of 1.08-5040 ng/mL (r > 0.0996). No matrix effect and carryover were observed. Intra- and inter-day accuracies and precisions were within the acceptance range. AMG 510 was demonstrated to be stable under the tested storage conditions. This novel method has been applied to a pharmacokinetic study in mice.

Discovery of a Potent and Selective PI3Kδ Inhibitor (S)-2,4-Diamino-6-((1-(7-fluoro-1-(4-fluorophenyl)-4-oxo-3-phenyl-4H-quinolizin-2-yl)ethyl)amino)pyrimidine-5-carbonitrile with Improved Pharmacokinetic Profile and Superior Efficacy in Hematological Cancer Models.

PI3Kδ inhibitors have been approved for B-cell malignancies like CLL, small lymphocytic lymphoma, and so forth. However, currently available PI3Kδ inhibitors are nonoptimal, showing weakness against at least one of the several important properties: potency, isoform selectivity, and/or pharmacokinetic profile. To come up with a PI3Kδ inhibitor that overcomes all these deficiencies, a pharmacophoric expansion strategy was employed. Herein, we describe a systematic transformation of a "three-blade propeller" shaped lead, 2,3-disubstituted quinolizinone 11, through a 1,2-disubstituted quinolizinone 20 to a novel "four-blade propeller" shaped 1,2,3-trisubstituted quinolizinone 34. Compound 34 has excellent potency, isoform selectivity, metabolic stability across species, and exhibited a favorable pharmacokinetic profile. Compound 34 also demonstrated a differentiated efficacy profile in human germinal center B and activated B cell-DLBCL cell lines and xenograft models. Compound 34 qualifies for further evaluation as a candidate for monotherapy or in combination with other targeted agents in DLBCLs and other forms of iNHL.

Discovery of isoquinolinone and naphthyridinone-based inhibitors of poly(ADP-ribose) polymerase-1 (PARP1) as anticancer agents: Structure activity relationship and preclinical characterization.

The exploitation of GLU988 and LYS903 residues in PARP1 as targets to design isoquinolinone (I & II) and naphthyridinone (III) analogues is described. Compounds of structure I have good biochemical and cellular potency but suffered from inferior PK. Constraining the linear propylene linker of structure I into a cyclopentene ring (II) offered improved PK parameters, while maintaining potency for PARP1. Finally, to avoid potential issues that may arise from the presence of an anilinic moiety, the nitrogen substituent on the isoquinolinone ring was incorporated as part of the bicyclic ring. This afforded a naphthyridinone scaffold, as shown in structure III. Further optimization of naphthyridinone series led to identification of a novel and highly potent PARP1 inhibitor 34, which was further characterized as preclinical candidate molecule. Compound 34 is orally bioavailable and displayed favorable pharmacokinetic (PK) properties. Compound 34 demonstrated remarkable antitumor efficacy both as a single-agent as well as in combination with chemotherapeutic agents in the BRCA1 mutant MDA-MB-436 breast cancer xenograft model. Additionally, compound 34 also potentiated the effect of agents such as temozolomide in breast cancer, pancreatic cancer and Ewing's sarcoma models.

Prediction of Human Pharmacokinetics of Fomepizole from Preclinical Species Pharmacokinetics Based on Normalizing Time Course Profiles.

Fomepizole is used as an antidote to treat methanol poisoning due to its selectivity towards alcohol dehydrogenase. In the present study, the goal is to develop a method to predict the fomepizole human plasma concentration versus time profile based on the preclinical pharmacokinetics using the assumption of superimposability on simulated time course profiles of animals and humans. Standard allometric equations with/without correction factors were also assimilated in the prediction. The volume of distribution at steady state (Vss) predicted by simple allometry (57.55 L) was very close to the reported value (42.17 L). However, clearance (CL) prediction by simple allometry was at least 3-fold higher to the reported value (33.86 mL/min); hence, multiple correction factors were used to predict the clearance. Both brain weight and maximum life span potential could predict the CL with 1.22- and 1.01-fold difference. Specifically, the predicted Vss and CL values via interspecies scaling were used in the prediction of series of human intravenous pharmacokinetic parameters, while the simulation of human oral profile was done by the use of absorption rate constant (Ka) from dog following the applicability of human bioavailability value scaled from dog data. In summary, the findings indicate that the utility of diverse allometry approaches to derive the human pharmacokinetics of fomepizole after intravenous/oral dosing.

Prediction of Human Pharmacokinetics of Bendamustine from Preclinical Species Pharmacokinetics Based on Normalizing Time Course Profiles.

Bendamustine, an alkylating anticancer agent, is used to treat chronic lymphocytic leukemia by intravenous infusion alone or in combination. The work aimed to develop a method to predict time vs. concentration profile for humans based on preclinical pharmacokinetics using the assumption of superimposability of normalized time course profiles of animals and humans. Standard allometric equations with/without correction factors (CF) were also used in prediction. The Vss was predicted by simple allometry of 0.312W0.871 (r2=0.987), where W is body weight; predicted Vss (19.71 L) was similar to the reported value (20.10 L). However, CL prediction involved both simple and CF allometry. Best proximity CL (543 vs. 598 mL/min) was obtained with maximum life span correction (MLP) [2.46W1.215 (r2=0.988)]. Normalized curves were obtained by normalizing the time (with mean residence time) vs. concentration (with dose/Vss) in animal species. The concentration vs. time profile in humans after intravenous infusion was then simulated using normalized curve for each animal species and the values of CL and Vss were predicted for humans. In summary the findings indicate that normalized time course approach could predict the bendamustine human pharmacokinetics and such an approach could be prospectively applied for analog drugs of this class.