PartⅠ: Selective O-alkylation based practical and efficient synthesis of Gefitinib, an EGFR kinase inhibitor PartⅡ: Practical synthesis of Levocabastine, a H1 receptor antagonist

Abstract

PartⅠ. 4-Anilinoquinazolines, have been prevalent motifs for pharmaceutical molecules in recent years, notably as EGFR inhibitors. Many novel EGFR inhibitors have been designed and synthesized through incorporating an alkoxy moiety via O-alkylation of hydroxyl group at the C-6 or C-7 position of the 4-anilinoquinazoline scaffold. However, in most case of alkylation of anilinoquinazolines with alkyl halide, a mixture of O- and N-alkylated, and N,O-dialkylated derivatives were generated without decent selectivity. It is a tedious work to separate and purify them. Thus, the development of new synthetic strategies which allow access to selective O-alkylation of 4-anilinoquinazolines is of interest to both laboratory and industry preparation of EGFR inhibitors. Gefitinib (marketed by AstraZeneca) is the first selective inhibitor of EGFR kinase domain and was approved in May 2003 for treatment of recurrent NSCLC (Non Small Cell Lung Cancer). Since then, it has been widely used worldwide and synthetic methods for gefitinib consisting of 4-anilinoquinazoline moiety have been continuously reported. According to the procedure described in previous reports, the generation of an excess of an N-alkylated impurity is an inevitable hurdle for the preparation of gefitinib. We introduced a novel concept, that trimethylsilyl (TMS) group was used as a transient protection group capable of significantly suppressing the N-alkylated impurity generation. The TMS group was quite attractive to us in terms of highly N-alkylation suppressing effect and facile removable character by a simple work-up. Indeed, this new and unique process minimized the generation of the N-alkylated side product. We also confirmed that our process is well applicable to a variety of 4-anilinoquinazolines.

PartⅡ. Levocabastine (marketed by Janssen) is a selective second generation H1 receptor antagonist and was discovered in 1979 for allergic conjunctivitis. As well as acting an antihistamine, levocabastine has also subsequently been found to act as a potent and selective antagonist for the neurotensin receptor NTS2, and was the first drug used to characterise the different neurotensin subtypes. Levocabastine is composed of piperidine substituted with phenyl and cyclic hexane, and has three chiral centers. According to the procedure described in a previous report, piperidine intermediate 9 was synthesized using N,N-bis(2-chloroethyl)-4-methylbenzene-sulfonamide as a starting material in a sequence of piperidine ring formation, the chiral resolution of intermediate 5, and detosylation with electrolysis. Subsequently, reductive amination with cyclic hexanone ring and hydrolysis afford to levocabastine. However, this procedure is considered not industrially suitable because it requires a chiral resolution, which lowered the yield and limited the utility of this synthesis in an economical process. Furthermore, the use of expensive platinum catalyst for reductive amination and electrolysis for detosylation remained as drawbacks for the commercial production. Therefore, we studied an efficient and practical procedure for the preparation of levocabastine. The key part of our strategy involves an introduction of chiral starting material to afford an optically pure key intermediate 8, replacement of electrolysis for detosylation, and removal of the use of platinum catalyst for reductive amination. By employing our process, we produced levocabastine hydrochloride with a 14.2% overall yield from a commercially available starting material, (S)-propylene oxide (11), on a multigram scale.