The Development Of A New Hypnotic Agent

David T. Beattie

Ph.D. and Thomas Jenkins Ph.D., Theravance, Inc.,

901 Gateway Boulevard, South San Francisco, CA 94080, USA

The ideal intravenous hypnotic agent, amongst other qualities, should have a rapid, uncomplicated and pain-free onset of action, and be devoid of any effects on the cardiovascular and respiratory systems. It should be associated with a prompt alteration in the depth of hypnosis upon titration of the infused dose, and have a rapid and predictable recovery of consciousness and patient function, independent of the duration of its infusion. Furthermore, it should be formulated in aqueous media and not support bacterial growth. No such intravenous hypnotic agent exists, although significant progress is being made in addressing several of the deficiencies of existing drugs. Two main strategies have been adopted by the pharmaceutical industry, namely modifications to the formulation or route of administration of existing drugs, and identification of novel agents.

In an attempt to avoid injection pain, hyperlipidemia or microbial growth (issues associated with Diprivan^â), alternative, modified lipid or lipid-free, formulations of propofol are available, or in late-stage development. Propofol-Lipuro^â 1% (Braun), a medium and long chain triglyceride lipid emulsion is associated with less injection pain than Diprivan, which contains only long chain triglycerides, while micellar solubilization of propofol (Maelor Pharmaceuticals, Inc.), confers bactericidal and improved physicochemical properties. A water-soluble prodrug of propofol (Aquavan^â), which relies on conversion by alkaline phosphatase in the blood for hypnotic activity, is in clinical development, while a cyclodextrin-based (Captisol^â) formulation of propofol has been evaluated preclinically. Transmucosal etomidate (Anesta Corp) and lingual propofol (Manhattan Pharmaceuticals Inc and NovaDel Pharma) may have value for premedication.

Efforts to identify and develop novel hypnotic agents are progressing. The approach taken at Theravance, Inc. sought to identify a metabolically-labile hypnotic agent which would be cleared efficiently, resulting in a rapid and predictable offset of drug effect, regardless of the duration of infusion or depth of hypnosis. The desired clearance mechanism (i.e. esterase metabolism) would be fundamentally different from that of propofol, which relies upon redistribution kinetics, and in consequence can be associated with delayed and unpredictable recovery upon prolonged infusion or inadvertent overdose. The product of this research, TD-4756 (THRX-918661), is a metabolically-labile, positive allosteric modulator of the GABA_A receptor, which is hydrolyzed rapidly by blood and tissue esterases to an inactive carboxylate metabolite (Beattie et al., 2004; Jenkins et al., 2004). In pre-clinical studies, TD-4756 produces hypnosis following intravenous bolus administration or infusion in rats, mice, guinea pigs, ferrets, dogs, cats, pigs and minipigs. Synergy between TD-4756 and fentanyl or remifentanil, occurs to at least the same degree as that observed with propofol and the opioids. In rats, pigs and minipigs, bolus administration or infusion of TD-4756 produces a dose-dependent and short-lived suppression of the EEG with a more rapid recovery profile than that of propofol. Emergence from TD-4756-induced hypnosis is rapid (duration <10 minutes) in all species evaluated. Furthermore, the duration of recovery from hypnosis is independent of the duration of infusion in rats, cats and pigs. In contrast, the time for emergence from propofol-induced hypnosis increases with the duration of infusion.

In vitro studies indicate that TD-4756 is metabolized rapidly by esterases in either the liver or blood, and it is anticipated that in humans, the liver will play the predominant role in its metabolism. Data from experiments using selective inhibitors, indicate that butyrylcholinesterase, and not acetylcholinesterase, is responsible for hydrolysis of TD-4756. With respect to its tolerability profile, in the majority of species tested, TD-4756 has a similar effect on the cardiovascular system to propofol, but a more rapid recovery profile is evident upon termination of infusion. TD-4756 is associated with less respiratory depressant activity than propofol in animals. It is postulated that TD-4756 may provide a significant advance over existing hypnotic agents in humans with respect to ease of titration to the desired hypnotic depth, and more rapid and predictable recovery, resulting in potential safety and cost benefits.

The development of novel hypnotic agents targeting, selectively, the specific loci in the CNS and mechanisms responsible for hypnosis, may also be on the horizon (Harrison, 2003). Thus, recent research efforts suggest that hypothalamic and thalamocortical regions are the origins of GABA_A receptor-mediated hypnotic activity, and that GABA_A receptor subtypes are heterogeneously expressed in the CNS, with distinctive physiological and pharmacological properties. Allosteric modulation of GABA_A-mediated responses, such as sedation and hypnosis, is dependent on the specific receptor subunit assembly. A single amino acid substitution in either the b₂ or b₃ subunits, for example, markedly reduces the sensitivity of the GABA_A receptor to propofol and etomidate, but not to neurosteroids. Furthermore, recent animal data suggest that sedation and hypnosis may be mediated by different GABA_A receptor subtypes (Reynolds et al., 2003). As our understanding of the mechanisms responsible for hypnosis increases, and the pharmacology of existing hypnotic agents becomes clearer, it is possible that novel strategies, unrelated to GABA_A receptor modulation, will also be identified and pursued by the pharmaceutical industry. Such approaches would follow the precedent set by ketamine and dexmedetomidine which produce hypnosis via antagonism of glutamate and agonism of a₂-adrenergic receptors, respectively. Ongoing rational research efforts should lead to the introduction of novel hypnotics in the future, with profiles of activity closer to that expected of the perfect agent.

References

Beattie, D. et al. (2004). Anaesthesia, 59, 101.

Jenkins, T. et al. (2004). Anaesthesia, 59, 100.

Harrison, N.L. (2003). Anesth. Analg., 97, 616-618.

Reynolds, D.S. et al. (2003). J. Neurosci., 23, 8608-8617.