Blood-brain barrier models: in vitro to in vivo translation in preclinical development of CNS-targeting biotherapeutics.

INTRODUCTION: The majority of therapeutics, small molecule or biologics, developed for the CNS do not penetrate the blood-brain barrier (BBB) sufficiently to induce pharmacologically meaningful effects on CNS targets. To improve the efficiency of CNS drug discovery, several in vitro models of the BBB have been used to aid early selection of molecules with CNS exposure potential. However, correlative studies suggest relatively poor predictability of in vitro BBB models underscoring the need to combine in vitro and in vivo BBB penetration assessment into an integrated preclinical workflow. AREAS COVERED: This review gives a brief general overview of in vitro and in vivo BBB models used in the pre-clinical evaluation of CNS-targeting drugs, with particular focus on the recent progress in developing humanized models. The authors discuss the advantages, limitations, in vitro-in vivo correlation, and integration of these models into CNS drug discovery and development with the aim of improving translation. EXPERT OPINION: Often, a simplistic rationalization of the CNS drug discovery and development process overlooks or even ignores the need for an early and predictive assessment of the BBB permeability. Indeed, past failures of CNS candidates in clinical trials argue strongly that the early deployment of in vitro and in vivo models for assessing BBB permeability, mechanisms of transport and brain exposure of leads, and the co-development of BBB delivery strategies will improve translation and increase the clinical success of CNS pipelines.

A peripherally restricted cannabinoid receptor agonist produces robust anti-nociceptive effects in rodent models of inflammatory and neuropathic pain.

Cannabinoids are analgesic in man, but their use is limited by their psychoactive properties. One way to avoid cannabinoid receptor subtype 1 (CB1R)-mediated central side-effects is to develop CB1R agonists with limited CNS penetration. Activation of peripheral CB1Rs has been proposed to be analgesic, but the relative contribution of peripheral CB1Rs to the analgesic effects of systemic cannabinoids remains unclear. Here we addressed this by exploring the analgesic properties and site of action of AZ11713908, a peripherally restricted CB1R agonist, in rodent pain models. Systemic administration of AZ11713908 produced robust efficacy in rat pain models, comparable to that produced by WIN 55, 212-2, a CNS-penetrant, mixed CB1R and CB2R agonist, but AZ11713908 generated fewer CNS side-effects than WIN 55, 212-in a rat Irwin test. Since AZ11713908 is also a CB2R inverse agonist in rat and a partial CB2R agonist in mouse, we tested the specificity of the effects in CB1R and CB2R knock-out (KO) mice. Analgesic effects produced by AZ11713908 in wild-type mice with Freund's complete adjuvant-induced inflammation of the tail were completely absent in CB1R KO mice, but fully preserved in CB2R KO mice. An in vivo electrophysiological assay showed that the major site of action of AZ11713908 was peripheral. Similarly, intraplantar AZ11713908 was also sufficient to induce robust analgesia. These results demonstrate that systemic administration of AZ11713908, produced robust analgesia in rodent pain models via peripheral CB1R. Peripherally restricted CB1R agonists provide an interesting novel approach to analgesic therapy for chronic pain.