Blunted Dopamine Release as a Biomarker for Vulnerability for Substance Use Disorders

Imaging of the striatal dopamine system continues to dominate studies using positron emission tomography (PET) in substance use disorders (SUD). A key reason for this is the stability of the findings: most studies imaging the dopamine D₂ family of receptors (D2R) and stimulant-induced dopamine release show blunting of striatal dopamine transmission in subjects with addiction. This phenotype is seen across SUDs, including cocaine, nicotine, alcohol, opiate, and methamphetamine. Reduced binding at the D2R persists independently of many clinical factors, and this effect is maintained following days to months of abstinence. The functional implications of blunted striatal dopamine transmission in the pathophysiology of addiction are becoming more evident, as shown by animal and human studies, such as that reported in this issue of Biological Psychiatry by Casey et al. (1).

Studies in animals have revealed that drug self-administration leads to a decrease in dopamine release and D2R expression (2,3) suggesting that blunted dopamine transmission directly results from drug exposure. However, other studies show that decreases in both D2R and dopamine levels could also constitute vulnerability for the development of addiction. In rats and nonhuman primates with no history of drug exposure, the lower the levels of striatal D2R, the higher the subsequent drug-seeking behaviors (2). Similarly, alcohol-preferring rodents display lower dopamine release in the nucleus accumbens (2,4). The behavioral consequences of blunted dopaminergic transmission are still unclear, but accumulating data suggest that the facilitation of subsequent drug-seeking behavior could originate from impaired inhibitory control resulting in increased impulsivity, which is a well-known feature of addiction (2).

To date, only indirect evidence supports a similar neurobiological phenotype in humans. It was previously shown that a pleasurable response to stimulant administration—a potential risk for addiction—in subjects without SUD is inversely correlated with striatal D2R availability (5). However, direct evidence that blunted striatal transmission precedes the development of addiction in humans has been lacking, partly because of ethical and practical constraints. The study by Casey et al. provides an innovative approach to identifying a specific neurobiological phenotype associated with high risk of SUD. In this study, PET imaging with the D2R radiotracer [¹¹C]-raclopride was conducted before and after the administration of amphetamine, a well-validated method that provides an estimate of both D2R availability and presynaptic dopamine release. Subjects were young adults who were included in three groups: 1) subjects using drugs (although not dependent) with multigenerational family history of addiction; 2) subjects using drugs (also not dependent) without a family history of addiction; and 3) subjects without significant drug use or family history of addiction. The results showed that subjects with the highest risk of developing a future addiction—as indicated by their drug use and family history—had the greatest blunting of dopamine release. This finding is consistent with the studies of substance dependence, indicating that at-risk individuals and individuals who have an established addiction share common neurobiological features.

A strength of the study by Casey et al. is the inclusion of young adults with personal drug use that did not reach criteria for dependence, with and without a significant family history of addiction. Drug exposure for all three groups included use of alcohol, tobacco, and cannabis, which was higher in the two “drug-using” groups but not absent in the “nonusing” control group, which is representative of the general low-risk population. The two drug-using groups were closely matched for exposure to other drugs, mostly stimulants with some psychedelics and limited opiate use. The high-risk subjects reported a more pleasurable response to the amphetamine challenge used for the PET scans compared with the controls, which has also been reported to predict future SUD and strengthens the idea that blunted dopamine release may be predictive of developing addiction.

The authors note that previous PET studies imaging subjects at risk for addiction included a group of subjects with a strong family history of substance abuse but who were not using drugs themselves. One study showed that subjects not using drugs and with a family history of SUD had higher D2R levels than control subjects not using drugs without a family history, suggesting that high D2R conferred resilience to SUD (6). The design of the study by Casey et al. complements these previous publications by including family history and drug-taking behavior, both of which confer a risk for future addiction. Taken together, these results suggest that low striatal dopamine release is associated with a family history of SUD and drug-taking behavior, whereas high D2R may confer resilience in the setting of a strong family history of addiction.

However, in contradiction to numerous imaging studies of addiction, the current study showed no difference in baseline D2R receptor availability, which is generally observed together with blunted dopamine release. One theory, mentioned by the authors, is that low D2R may follow drug exposure rather than confer risk for SUD, which is supported by some animal data (2). However, other studies have demonstrated that low D2R levels precede and predict subsequent drug self-administration (2). One caveat of these latter studies is that they were performed on animal groups selected a priori on the basis of behaviors known to predict greater drug-taking behavior (e.g., impulsivity or social stress) (2), which could accentuate the differences in D2R levels. For example, in monkeys, low D2R precedes cocaine self-administration in the setting of social stress, but differences in D2R were seen only when comparing the extremes of monkey groups (i.e., most subordinate vs. most dominant monkeys, with no differences seen in the intermediate animals) (7). Similarly, high impulsivity in rodents, which predicts higher self-administration, is accompanied by decreased D2R levels compared with animals with low impulsivity (2). Based on these data, it is possible that human studies might reveal a difference in D2R only when extreme groups are compared—such as addicted individuals versus healthy control subjects—because of their greater endophenotypic differences. In the study by Casey et al., the comparison of subjects who are at risk but do not meet criteria for SUD and control subjects could fail to show a difference in D2R because the at-risk SUD group likely includes a heterogeneous population (will or will not develop SUD).

Another possibility to consider is the effect of endogenous dopamine on measuring D2R. [¹¹C]-raclopride binding is affected by levels of endogenous dopamine, and dopamine depletion unmasks D2R, increasing its binding potential (8). Using dopamine depletion, lower levels of endogenous dopamine have been reported in subjects with cocaine abuse (8). If the high-risk group imaged in the study by Casey et al. had lower levels of endogenous dopamine, this could appear as no difference in D2R. Given that low D2R is generally seen in conjunction with low dopamine release, it is possible that these are interrelated. Owing to the complex bidirectional interactions between the striatum and midbrain, it is possible that low striatal D2R expression affects midbrain dopamine transmission. This possibility could be tested in animal models of low striatal D2R expression by measuring dopamine release in the striatum.

Overall, the results of Casey et al. show that subjects at risk for future addiction, based on family history and current drug-taking behavior, have blunted dopamine release compared with control subjects without a family history, including subjects who also engage in behavior associated with addiction risk. This is an important finding for addiction research because it demonstrates that blunted dopamine release, which is a biomarker for addiction, may precede the development of addiction. However, the authors note that their study cannot tease apart the contribution of an inherited risk, early life experience risk, or a very rapid adaptation to drug use. This question would be difficult to answer using PET because it would require a large study in subjects <18 years old, which is problematic. However, a large European study (IMAGEN) has set out to address this question partly using functional magnetic resonance imaging to investigate behaviors closely associated with the development of addiction in adolescents. Functional magnetic resonance imaging cannot detect changes in dopamine specifically, but it can be used to measure striatal activation in response to reward, which likely includes a dopamine signaling component. Some of the IMAGEN results show that adolescents at risk for substance use display greater risk-taking behavior and lower striatal activation in response to reward (9). Another study showed that activation of the ventral striatum in response to reward anticipation was lower in adolescent smokers compared with control subjects, even in subjects who had smoked <10 times, suggesting that blunted activation cannot be attributed only to nicotine exposure (10). Because the reward tasks depend at least in partly on striatal dopamine transmission, these findings suggest that low striatal signaling is associated with an increased risk of addiction, which is consistent with the PET findings of Casey et al. Together, these studies show how PET and functional magnetic resonance imaging studies can complement each other, and they indicate that blunted striatal dopamine response may serve as a neurobiological marker for individuals at risk for developing addiction.