Bret S. Stetka, MD, Nora D. Volkow, MD
July 16, 2013
Editor’s Note: In June 2013, the American Medical Association (AMA) declared obesity a disease, a move championed by many clinicians and derided by others. Debate aside, the announcement reinforced that the understanding and appreciation of obesity is evolving, and that in this era of soda bans and school lunch reform, obesity is high in the consciousness of both the public and the medical community.
One area of obesity research receiving a great deal of attention lately is the considerable neurobiological overlap between addiction and some forms of obesity. Nora D. Volkow, MD, Director of the National Institute on Drug Abuse (NIDA), has studied this association extensively. Medscape recently spoke to Dr. Volkow about her ongoing research into the relationship between weight and addiction.
Obesity and Addiction: Introduction
Medscape: Hi, Dr. Volkow. Before we get to the relationship between obesity and addiction, how does human appetite control normally work?
Dr. Volkow: Eating behaviors are obviously necessary for survival, and we’ve evolved complex and redundant systems to ensure that the behaviors necessary to eat occur. The brain receives peripheral signals that relate to the body’s nutritional status: for example, the concentration of glucose, lipids, and certain nutrients. These signals are conveyed to the brain via the plasma and peripheral nerves in the gastrointestinal tract — a process known as “homeostatic regulation” of feeding — and help indicate whether the body needs to consume more food or not and whether there is a need for a specific nutrient. Parallel reward systems in the brain respond to the pleasantness of food and motivate our behaviors to consume them, a process described as “hedonic regulation” of food intake.
It was believed for many years that these peripheral signals were predominantly received by the hypothalamus, and it was initially thought that there were only a few signals (eg, glucose, insulin, leptin). However, in the past 10 years, many other signals (eg, hormones, peptides) have been identified that work in part by acting on and altering the sensitivity of reward systems in the brain to food.
For example, normally the hormone leptin is secreted from adipose cells to tell the brain that enough energy is stored; this decreases appetite and increases energy expenditure. There is a very rare genetic condition where individuals are born without the gene that encodes for leptin, and these individuals overeat and become obese from childhood. In these individuals, brain imaging studies have shown that reward systems in the brain are hypersensitive to the rewarding properties of food. Leptin treatment in these individuals decreases the sensitivity of the reward system, resulting in a decrease in the motivation to eat and a marked reduction in weight.
When peripheral signals — such as leptin or insulin — are not released, or your brain becomes tolerant to them, you don’t have a mechanism to counter the drive to eat. It’s like driving a car without brakes.
Medscape: Which brain regions make up our reward centers?
Dr. Volkow: We describe them as a reward circuit, because there are multiple connected regions involved. A central node in the circuit is the nucleus accumbens (NAc), a region regulated by dopamine, which we always hear about being responsible for the rewarding and addictive effects of drugs. The rewarding effects of drugs result from their ability to activate the NAc through increased dopamine release. Similarly, rewarding effects of food are linked to dopamine release in the NAc.
Before we get to food addiction, I should say that how effective food is at activating these dopamine pathways, and the NAc, is modulated by homeostatic peripheral signals. If you don’t have insulin and leptin, which decrease dopamine signaling in the NAc, the reward circuit will go into overdrive when you eat or are exposed to food; this is because signaling in the reward pathway triggers the motivation to eat. This can result in obesity. As people become obese, they become insulin- and leptin-resistant, thus removing the normal peripheral signals that help inhibit the rewarding effects of food; the more severe the obesity, the worse the brain becomes at preventing excess food intake.
The Addiction Overlap
Medscape: What neurophysiologic changes occur resulting in addiction?
Dr. Volkow: I, along with my colleagues at Brookhaven National Laboratory and the National Institutes of Health, have been studying the changes in the brain dopamine system in people who are addicted and also in morbidly obese individuals. Generally speaking, when dopamine activates D1receptors in the reward system, it increases the drive to participate in a rewarding behavior (eg, eating), whereas D2-receptor stimulation helps dampen and regulate this response, allowing us to exert control over the behavior. What we are seeing in addictive behaviors and in obesity is that signaling through the D2 receptor system is markedly attenuated. As a result, you lose your brakes while the engine just keeps going.
What is very interesting is that addicted patients seem to be hypersensitized to stimuli related to, for example, food or drugs. A surge in dopamine activity occurs in anticipation of the reward but drops markedly with actually participating in the behavior. So it’s the anticipation that is rewarding, and not so much the reward itself.
Medscape: Very interesting. Presumably, this is only one of multiple potential causes of obesity, correct?
Dr. Volkow: Yes. Obesity can also result from lack of physical activity or from enhanced energy efficiency, such that little energy is required by the cells of the bodies to perform their functions. Even if your peripheral homeostatic signals are working fine, if you live a completely sedentary lifestyle, you still may become obese if you are consuming a greater quantity of calories than you need. So the potential addictiveness of food can contribute to or cause obesity, because it drives our consumption of food even when we are not hungry and our body has all of the nutrients it needs.
Medscape: In your mind, then, would the subset of obesity caused by addictive mechanisms qualify as an addiction?
Dr. Volkow: It’s interesting, because in the past we would always say either you have something or you don’t; either you’re addicted or you’re not. However, the new classification system in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5)[1] has a dimensional component that assesses addiction on a continuum. I think the exact same concept applies to food.
Addiction is characterized by loss of control in relation to a substance or behavior. Now, I lose control over certain foods all the time; however, I am not obese. You could say I have a vulnerability to the rewarding effects of certain foods. But is it severe enough to be called a disease? Probably not. So I would say that in extreme cases of obesity — in which patients want to stop eating but they just can’t, even with awareness of all the adverse physical and psychological consequences (obese people are stigmatized, which is a powerful social stressor) — then yes, I would say there is an addictive dimension to their behavior.
The DSM, Video Games, and DC Traffic
Medscape: On a related note, I know it was somewhat controversial that DSM-5 included specific addictive behaviors, such as Internet gaming addiction. Do you feel that calling out specific addictions like this is warranted? Or in theory, aren’t there an infinite number of behaviors one could be addicted to?
Dr. Volkow: There actually appear to be a limited number of behaviors which can result in addiction, much like drugs. There are only a few chemical compounds that can produce addiction. You can’t get addicted to antibiotics, or antidepressants. Only chemical compounds that increase dopamine activity in reward pathways can produce addiction. Similarly, only behaviors that increase dopamine in the NAc (ie, gambling) can result in addiction with repeated exposures in those that are vulnerable. The same is true with food; as a result, not all foods are equally rewarding.
If you eat an overboiled piece of chicken, it can be rewarding, but only if you are hungry. But if you are not hungry, it may not be inherently rewarding. On the other hand, many would find a chocolate chip cookie inherently rewarding, even in the absence of hunger. This makes some foods much more dangerous in promoting reward-system activation and causing compulsive eating behaviors.
The same goes for such behaviors as video gaming, which is inherently reward-based. You get the reward of winning, or beating your enemy. And gambling is obviously reward-based. But driving a car, for example, in the middle of rush hour in Washington, DC, will never generate an addictive behavior!
Medscape: I know you were a proponent of including obesity caused by addiction in the DSM-5. In the end, it wasn’t included in the revised manual, but do you still feel it should have been? And do you think ultimately it will be included as more data come to light?
Dr. Volkow: During the DSM-5 development, Chuck O’Brian and I brought forward the similarities between addiction to drugs and obesity from overconsumption of palatable food, both in terms of clinical presentation and neurocircuitry. We saw it as an opportunity for psychiatrists to consider the addictive component of obesity and help in developing therapeutic interventions. Specifically, we thought psychiatry could play an important role in the behavioral element of overeating behaviors, but obesity didn’t get incorporated into the DSM-5. Some were concerned that including obesity as a mental illness would increase its stigmatization. On the other hand, I was delighted when the AMA declared obesity a disease.
Medscape: Is there much support in the medical community for one day including obesity in the DSM?
Dr. Volkow: I don’t know. Societies that develop diagnostic criteria have committees with long-standing procedures and opinions regarding the diseases they are classifying. There is a section in the DSM on eating disorders, which includes bulimia and anorexia nervosa. Of note, DSM-IV (and DSM-5) consider binge eating disorder a mental illness, but not obesity. But I do foresee that eventually — and particularly now that the AMA considers obesity a disease — perhaps in the next round of revisions, certain types of obesity (ie, that not associated with endocrine dysfunction) may be considered for inclusion in DSM.
Medscape: Do you have any sense as to what percentage of obesity is caused by addictive pathology?
Dr. Volkow: I do not have any data, but I would predict that most of the problems of being overweight and obesity in our country are driven by excessive consumption of rewarding food. However, just as is the case for drugs, for which most of the consumption is not by people who are addicted, I would predict that most cases of overeating are in people who are not addicted to food.
Therapeutic Implications and Future Directions
Medscape: What does the improved understanding of addictive obesity mean for treatment? Are there therapies under investigation targeting the reward pathways in obesity?
Dr. Volkow: In animal models, there are medications that, for example, interfere with the consumption of high quantities of food. In many instances, these same drugs are useful in interfering with drug consumption. This is not surprising, as these drugs work by acting on the reward system. People are looking for drugs that can act on this system and decrease their addictive behaviors.
As the Director of NIDA, this is an area of great interest to me, because there has been very limited interest by the pharmaceutical industry in developing medications for drug addiction, whereas there is much more interest in developing obesity medication treatments.
Medscape: What antiaddiction medications have shown efficacy?
Dr. Volkow: One is naltrexone, a new opioid receptor antagonist that is very useful in treating heroin addiction and alcoholism. A combination of sustained-release bupropion/naltrexone therapy is currently being reviewed by the US Food and Drug Administration for obesity. There has also been a lot of interest in cannabinoid receptor antagonists.
Rimonabant — a drug that was approved in Europe for the treatment of obesity, and then removed from the market because it caused suicidal thinking in some patients — had been shown to interfere with the administration of drugs in animal models of addiction. It also showed potential benefits in treating marijuana abuse. Unfortunately, its side effects will preclude its clinical use.
More researchers are working on the neurobiological overlap between drug and food reward, and the adaptions that ensue with repeated use, as a strategy to help identify molecular targets for addiction medications.
Medscape: What are you and your group at NIDA currently working on?
Dr. Volkow: We are studying many facets of obesity. We are interested in understanding the mechanism by which glucose and other peripheral signals (including vagal stimulation) activate the reward system. We are also studying the neurocircuitry overlap between food and alcohol consumption; this has generated a lot of clinical interest, because bariatric surgery for the treatment of obesity is associated with a higher risk for developing an alcohol use disorder. Specifically, we are studying the mechanisms by which bariatric surgery modifies the rewarding effects of alcohol and other drugs.