CUE–REACTIVITY IN RODENTS—HOW PREDICTABLE?
Article first published online: 25 JUN 2009
© 2009 The Authors. Journal compilation © 2009 Society for the Study of Addiction
Volume 104, Issue 10, pages 1619–1620, October 2009
How to Cite
SHOAIB, M. (2009), CUE–REACTIVITY IN RODENTS—HOW PREDICTABLE?. Addiction, 104: 1619–1620. doi: 10.1111/j.1360-0443.2009.02626.x
- Issue published online: 8 SEP 2009
- Article first published online: 25 JUN 2009
- Animal models;
Perkins  has reviewed a series of cogent arguments that undermine the utility of smoking cue–reactivity studies and the basis for measuring cue-induced craving to devise long-term treatment strategies. For many years, the terms ‘craving’ and ‘relapse’ have been used frequently; significant effort has been directed at defining these measures and then modelling them within the context of drug dependence and acknowledging the crucial roles played by cues [2,3].
From a pre-clinical perspective, non-pharmacological stimuli/cues are involved in nicotine-seeking and nicotine-taking behaviours . For a variety of abused substances, conditioned stimuli have been demonstrated to exert control over behaviour which can be captured using the cue-induced reinstatement of intravenous self-administration  and also by the conditioned place preference paradigm, which assesses approach responses to contextual/environmental stimuli associated previously with drug administration. Both these models assess reactions to cues [4,6] and thus provide an analogous measure to the cue–reactivity responses reviewed by Perkins. The situation with regard to predictive validity of these models is no different to that argued by Perkins; while our understanding on the neurobiological processes underlying these cue-induced changes on behaviour has advanced significantly, the therapeutic utility of these two cue-dependent models to develop smoking cessation aids has yet to emerge. Here are some practical reasons why these models do not translate. First, cue-induced reinstatement of nicotine-seeking behaviour is observed only when the cues are presented contingent upon a lever press response. Further, when contextual cues are presented non-contingently in the combination of visual and tactile cues, levels of reinstatement are small and not robust . Secondly, to validate the model, only non-nicotinic compounds may be evaluated, because nicotine and nicotinic agonists will reinstate behaviour by themselves prior to the contingent presentation of the cues. This obviously limits the validation process of the cue-induced reinstatement model. Of the smoking cessation aids tested so far, only rimonabant has shown efficacy against cue-maintained responding , while bupropion is ineffective . Surprisingly, the nicotinic receptor antagonist mecamylamine can attenuate cue-induced reinstatement of nicotine-seeking behaviour ; however, this finding has yet to be fully translated clinically.
So where do we go from here? Clearly, there are far too many gaps and unknowns with regard to the utility of cue–reactivity studies from both pre-clinical and clinical perspectives. As Perkins states, this area is ‘premature until such time that cue-induced craving can be shown to predict relapse risk’. A more comprehensive database is required, with Food and Drug Administration (FDA)-approved smoking medications tested in both pre-clinical animal models and clinical models of cue–reactivity. Furthermore, there is greater scope with animal model development to incorporate elements of conditioned reactivity to make them more clinically relevant which hopefully will, in the long term, generate novel pharmacological targets for intervention. With the current models available, such as the cue-induced reinstatement of nicotine-seeking and place conditioning paradigms, there is the concern that these models will generate only compounds with clinical efficacy either similar or slightly better than those observed currently for rimonabant or bupropion. We must move away from this approach and to meet this challenge will require significant time and financial investment. To render this target a little more feasible, it is essential that animal model development is based on our advanced understanding of neurobiological mechanisms mediating smoking cue–reactivity. Thus, a good starting-point would be to examine the role of the anterior cingulate in cue–reactive animal models using either neurotoxin lesions or localized injections. These strategic routes will allow us to harness further the therapeutic utility of cue–reactive processes or, if unsuccessful, to banish them as an artefact of the tobacco smoking conundrum.
Declaration of interest
Dr Shoaib is a consultant to pharmaceutical companies: Abbott and Neurosearch.