Persistent vulnerability to relapse despite complete extinction of cocaine craving

Craving often precedes relapse into cocaine addiction. This explains why considerable research effort is being expended to try to develop anti-craving strategies for relapse prevention. Recently, we discovered using the classic reinstatement model of cocaine craving that the reinstating or priming effect of cocaine can be extinguished with repeated priming in rats, a phenomenon dubbed extinction of cocaine priming. Here we sought to measure the potential beneficial effect of this novel extinction strategy on subsequent relapse (i.e., return to the pre-extinction pattern of cocaine self-administration once the drug is made again available after extinction). Overall and contrary to our initial hope, extensive and complete extinction of cocaine priming had no major impact on relapse. This lack of effect occurred despite evidence for post-extinction loss of neuronal responses to cocaine priming in brain regions critically involved in cocaine-induced reinstatement (i.e., the dorsomedial prefrontal cortex and the core of the nucleus accumbens). An effect of extinction of cocaine priming on relapse was only observed when cocaine was available for self-administration under more demanding conditions. However, this effect was modest and short-lived. Finally, we succeeded to trace the origin of our failure to prevent relapse to a persistent, extinction-resistant form of operant behavior that is not directly induced by cocaine. This extinction-resistant behavior is commonly reported, though generally ignored as causally irrelevant, in many other reinstatement studies. We propose that this behavior should become both a novel marker for long-term vulnerability to relapse and a novel target for preclinical development of potential relapse prevention interventions.