Alcohol addiction is a psychiatric disorder characterized by a chronic relapsing trajectory and continued drinking despite negative consequences, also referred to as compulsive drinking. Craving for alcohol, a hallmark of addiction, is a predictor of relapse, and can be triggered by social stress. The mechanisms of alcohol addiction are complex, and involve environmental and genetic susceptibility factors. Social marginalization, commonly experienced by people with alcohol addiction, can promote a vicious circle of drinking and further social rejection that exacerbates the addictive state. Several studies implicate the endogenous opioid system as a link between these behaviors.

The endogenous opioid system acting through mu-opioid receptors (MORs), plays a key role in the modulation of pain and reward, but has also been proposed as a neural substrate of social attachment. Functional genetic variation in the gene encoding the MOR, OPRM1 A118G, has been reported to moderate rejection sensitivity in both primates and humans, but previous studies have been criticized due to small sample sizes. Therefore, the aim of Paper I was to test if OPRM1 A118G influenced dispositional sensitivity to rejection in an online game setting, with a sample large enough to detect small effects. However, we did not find support for this hypothesis.

The anterior cingulate cortex (ACC) and the insula are neural structures typically activated during somatic pain, but have also been reported to become activated during experimental social rejection.  This has led to the hypothesis that physical and social pain share a common alarm system, a notion consistent with high concentrations of MORs in these shared areas. However, the evidence for a social pain construct has primarily relied on fMRI studies showing task-based activations, or correlations with MOR-ligand displacement during positron emission tomography (PET) scans. These studies are merely correlational, and Paper II therefore aimed to test if MOR activation using exogenous opioids would alleviate rejection related distress in a randomized controlled trial (RCT). We found that the experimental social exclusion was associated with negative affect and the expected neural activity in the insula and the orbitofrontal cortex (OFC). However, contrary to our hypothesis, we found no influence of the prototypical MOR agonist morphine on self-reported distress or neural activity. 

Childhood maltreatment, a form of severe social stress, is a risk factor for developing addiction, but clinical assessments commonly rely on retrospective self-reports that are subject to bias. In Paper III, we examined the performance of the Childhood Trauma Questionnaire (CTQ) to detect prospectively documented childhood maltreatment in a unique cohort, that allowed us to test how Substance Use Disorder (SUD) influenced the retrospective reports. We found that the CTQ had an excellent ability to identify childhood maltreatment in individuals without SUD, but performed only slightly better than chance in people with SUD.

The insula has also been proposed to play an important role in processing alcohol craving. The insular cortex has only recently become accessible for non-invasive neuromodulation, and treatment using deep repetitive transcranial magnetic stimulation (rTMS) was reported to reduce both cigarette craving and smoking. Paper IV aimed to test if deep rTMS would be efficacious against alcohol craving and drinking. We found imaging-based evidence of neuromodulation of insula function, but despite this, we found no significant treatment effects of deep rTMS targeting this structure on alcohol craving or use.

Recent advances in brain imaging have shown that emotional and motivational states, including craving, recruit networks across the whole brain rather than being represented in isolated brain areas. The newly developed Neurobiological Craving Signature (NCS) has been shown to correlate with self-reported drug craving, an established predictor of relapse. This prompted the question whether the NCS could also predict clinical outcomes in alcohol addiction. This was evaluated in Paper V in the same patient cohort as in Paper IV. Here, we showed that the NCS was strongly correlated with self-reported craving. Most importantly, it predicted relapse and alcohol use during the three month follow-up phase.