Brain imaging has provided important insights into the clinical neurobiology of addiction for more than 20 years, beginning with the pioneering Positron Emission Tomography (PET) studies of Volkow et al (1988), and continuing to recent studies showing that cocaine addiction is associated with changes in dopamine function that are predictive of choice to self-administer cocaine (Martinez et al, 2007). However, because of their cost and radiation hazards, PET studies of addicted subjects who are undergoing clinical trials have been limited.

Although BOLD fMRI has some disadvantages, such as lack of an absolute measure of blood flow or neuronal activity, it has the advantages of repeatability, cost, and safety over PET. One area in which fMRI has shown promise in medication development for addiction is as a predictor of treatment response in clinical trials. Recently, Brewer et al (2008) showed that fMRI brain activation during performance of a Stroop task in cocaine-dependent subjects before treatment was predictive of subsequent treatment response. Other studies are underway in several centers using fMRI as a baseline predictor of treatment response in clinical trials for cocaine dependence, which should provide important information about the neurobiology of medication response in addiction.

A second area that shows promise for fMRI in medication development for addictions is in the study of acute effects of medications on brain function in addicted individuals. The use of fMRI in this manner, also termed pharmacoMRI or phMRI, has potential as a tool to drive medication development by providing additional information about the effects of medications before their use in costly, time-consuming phase II clinical trials (Wise and Tracey, 2006). The rationale for using phMRI in medication development for addictions is based on the following: (1) behavioral research has shown that addicted individuals show differences in behavioral performance in tasks such as cue reactivity and behavioral inhibition compared with non-addicted subjects; (2) fMRI of addicted subjects while performing these tasks has shown patterns of brain activation that differ from non-addicted subjects; (3) acute effects of novel compounds on brain activation patterns in addicted subjects while performing these tasks could provide additional information about the pharmacodynamics, and hence the potential utility of these compounds for treatment of addictions. The potential usefulness of this approach has been shown by a recent study showing that an acute dose of methadone reduced fMRI brain activation associated with heroin-related stimuli in opiate addicts (Langleben et al, 2008); however to date, phMRI has not been routinely used as a tool for medication development for addictions.

Basic science advances in understanding addiction raise hope for a future with more effective therapeutic strategies for this chronic, relapsing brain disorder. Yet, the progress of candidate medications on the pathway to patients is slow and of the numerous compounds studied for treatment of addictions, few have reached the FDA approval (Vocci and Elkashef, 2005). Novel brain imaging methods, such as phMRI, offer the potential to provide important information about medications for addictions, which could aid in the development of pharmacotherapies for addictions.