The Neurobiology & Treatment of Cannabis Dependence

David A. Gorelick, MD, PhD; Adjunct Professor of Psychiatry
University of Maryland School of Medicine Chief, Pharmacotherapy Section, Treatment Branch, NIDA IRP

---

Q:You received your MD, PhD from Albert Einstein College of Medicine.  What was your specialty or research focus?

A:I was involved in the Medical Scientist Training program, focusing on psychopharmacology.  My undergraduate work was in psychology at Cornell and in med school, pharmacology and psychopharmacology for the PhD.

Q:How did you become involved with NIDA?

A:I was at UCLA for 15 years.  Then NIDA called me up and it was a chance to work in one of the world’s leading drug abuse research facilities . A typical academic pathway involves getting the graduate degree, do a post-doctoral fellowship (sometimes several), get a job for 3-5 years or the duration of the grant – there’s often a lot of moving around, which is not always good for family. I took a different path – my wife is also a physician, we trained at UCLA and did not intend to move until this position came up.  NIH/NIDA has so many advantages for a researcher - who would turn it down?

Q:There are lots of big things happening in the field of cannabinoids.  We think a lot about tobacco and second hand smoke (SHS) here but we don’t know much about marijuana (MJ) and its ensuing SHS? Anything new?

A:I  am not familiar with any clinical research...I’m sure it exists.  One of my former colleagues at NIDA, Dr. Ed Cone, has done published research showing that if you are in a room with people smoking marijuana, but you are not smoking yourself, marijuana still gets absorbed – in fact it will show up in your urine.  (For example, see Cone & Johnson, Clinical Pharmacology & Therapeutics, Sept., 1986, vol. 40, no. 3, pp. 247-256.).

Q:When Dr. David Smith was here, he said that during Woodstock 1 marijuana was <1% THC and by Woodstock 2 it was >6%. Aside from it being 6-12 times the dose, what have been the consequences? 

A:I wasn’t at either Woodstock. I’m not a toxicologist, but I would be careful about the basis of statements on changes in potency of marijuana.  Aside from the small supply from NIDA used for research purposes, marijuana is not FDA regulated - no one is measuring the potency before it is sold on the streets.  The DEA has a program where they analyze batches of marijuana submitted to them as it is seized by police, but this does not give a representative or random sample.  Some people in the field are not convinced there have been any significant changes in marijuana potency over the years.  It is true that marijuana plants can vary widely in THC potency based on the variety of cannabis, where it is grown, how it is prepared - whether leaves, stems, or resin are used, etc.  I know of one study that analyzed the potency of almost 30,000 marijuana samples seized in the U.S. between 1980 and 1997.  The study found that the potency (concentration of THC) increased from less than 1.5% in 1980 to around 3% from 1983 to and to 4.2% in 1997 (ElSohly et al., Journal of Forensic Science, Jan., 2000, vol. 45, no. 1, pp. 24-30). 

Q:Can you explain how THC is different or the same as marijuana? 

A:THC is the shorthand abbreviation for Δ-9-tetrahydrocannabinol, which is believed to be the primary psychoactive ingredient in marijuana.  The plant, however, contains hundreds of other chemicals – so called cannabinoids – many of which have not been well studied and some of which probably have biologic activity, e.g. cannabidiol, cannabinol. These may have biologic activity of a different spectrum than THC. There is good experimental evidence, both in animals and humans that THC produces the major psychological effects – it is what makes people feel “high” or gives euphoria.  Marijuana causes many physiological effects, e.g. increased heart rate, increased appetite.  You can reproduce many of the effects of marijuana by administering pure THC.  You can take THC and other chemicals out of a marijuana cigarette and it will be like a placebo.

Q:How about the cannabinoid receptor antagonists, we know about the CB1 antagonist rimonabant, which has been shown to be effective in obesity and cigarette smoking in Europe - does it also cause cannabis withdrawal in man like naloxone (Narcan®) causes opiate withdrawal? 

A:In animals it does.  You can make an animal tolerant to chronic THC and they will go into withdrawal when THC is removed.  If you give them an injection of rimonabant, they go into precipitated or antagonist-elicited withdrawal. It’s very similar to giving Naloxone or Naltrexone to an animal that is dependent on an opiate.  It’s not known whether the same would happen in humans.

Q:What are the main differences, if any, between provoked or precipitated withdrawal for cannabis and naturally occurring withdrawal?

A: It hasn’t been tested in people, but theoretically, based on what occurs with opiates, the antagonist-elicited withdrawal to a drug as compared to spontaneous or natural withdrawal comes on more quickly and is more intense, but also resolves more quickly.  

Q:What is the resistance to the idea that cannabis is addicting? Why is cannabis dependence not in the DSM-IV? 

A:Actually it is – what’s not in the DSM-IV is cannabis withdrawal.  There’s a section on substance use disorders and their criteria for abuse and dependence are not s substance-specific.  

Q:Can you explain this slide? 

A:The groups of symptoms shown after stopping marijuana use are a true withdrawal syndrome.  If you look at the symptoms reported, (similar to previous self-report studies) they tend to be nonspecific, e.g. boredom, irritability, change in appetite, anxiety, upset stomach…lots of things can cause that, even the flu.  As a matter of fact, one study compares marijuana withdrawal to having a mild case of the flu – so how do we know it is marijuana withdrawal? There are several criteria that make this a true withdrawal syndrome, conceptually similar to withdrawal from alcohol or opiates.  One criterion our data show is consistent time course – consistent onset of symptoms, consistent duration, and offset.  The slide shows that the physical symptoms tend to cluster together starting within a day or two of ceasing marijuana use and resolving within the first week.  A factor analysis shows distinct groups of physical and psychological symptoms.  The psychological symptoms tend to have later onset and can persist for weeks.

Q:Do you think cannabis dependence should be treated like opiate dependence and have a maintenance option like buprenorphine or methadone or like alcohol and have no such option? 

A:I can’t really give an answer right now.  As a scientist who believes in evidence-based medicine, I’d say we’d need to do clinical trials/studies to see which is best. There is one published study by Dr. Haney and colleagues at Columbia University showing that oral THC (Marinol®) suppresses withdrawal, but that’s a short-term (6-day) study (Haney et al., Neuropsychopharmacology, Jan., 2004, vol. 29, no. 1, pp. 158-170).  We know from experience with other drugs that treating withdrawal by itself is not usually an adequate treatment for addiction.  I am unaware of studies as of yet on long-term agonists or substitution maintenance treatments for marijuana.  Substitution maintenance treatment is not just an empirical question (is there evidence for it?). There are also ethical and moral issues to consider; some say you are just substituting one addiction for another.  You can take an empirical approach and say, yes, it is a moral issue, but, do patients on methadone do better than patients with no treatment? My view of the evidence is yes, they use heroin less, they have less HIV risk, less criminal activity, better employment status, etc.  Some drugs are not as well studied. In the 1960’s when benzodiazepines came out, there were very few studies on benzodiazepine maintenance treatment for alcohol dependence.  Conceptually, it would be no different than methadone treatment for opiate addiction. I am aware of only one or two studies and those are mildly positive, but the field as a whole never took that up as a consensus.  The consensus view now – in my opinion – is that benzodiazepine maintenance treatment for alcohol dependence would not be appropriate.  Of course, we do have other types of medications now: naltrexone, acamprostate, disulfiram.  You touched on something - that’s my field, pharmacological treatment of addiction. Similar issues are going to come up with other substances.  How about substitution maintenance treatment of stimulant abuse? There are published, controlled clinical trials suggesting that sustained-release amphetamine might be safe and effective for cocaine addiction. There’s some clinical experience in Britain of prescribed amphetamine to treat amphetamine and cocaine abuse and addiction.  But again, these are more than just empirical questions; they do require some sensitivity, and create controversy. The risk with agonist substitution treatment can be engendering drug craving and drug abuse, e.g. the early trials of methylphenidate for cocaine addiction in the 1980’s were not very positive. In fact, some trials were stopped because the patients just turned to abusing methylphenidate.  This gets me back to my original point – as a scientist and physician I am an empiricist.  Is there a clinical trial showing the treatment is safe and effective? What is best for my patients?  It would be nice to have an effective medication for cannabis dependence, but we have none that’s FDA approved currently.  This is an area that NIDA is trying to promote research in. So hopefully in the future we will have effective medications.  Right now the focus of treatment remains psychosocial.

Q:Is there such a thing as late onset...starting over 30 years of age, cannabis dependence? 

A: Again, I go with the data. I can’t say those people don’t exist, but they are certainly uncommon. As with other psychoactive drugs: tobacco, alcohol, cocaine (most people); if they are going to start using a psychoactive drug, they start early in life: adolescence, early 20s.

Q:What do you think the CB1 receptors and cannabinoids do normally in the brain? 

A: CB1 receptors and endogenous cannabinoids are not just found in the brain – it is a fairly widespread system.  They are found in the lining of blood vessels, the GI tract, sperm, the uterus, the liver, so they may mediate more than brain function. This makes CB1 receptors and cannabinoids very interesting therapeutic targets. In the brain at the neuronal level, endogenous cannabinoids are retrograde synaptic modulators. (Please see figures below.)  They are released from the postsynaptic neuron and diffuse back across the synapse to activate CB1 receptors in the presynaptic neuron to reduce presynaptic neurotransmitter release. This provides sort of a negative feedback. The neuronal activity presynaptically releases neurotransmitters that diffuse across the synapse to activate postsynaptic receptors – leading to release of anandamide, which diffuses back and activates the CB1 receptors to modulate further neurotransmitter release.  

Q:Do you think chronic antagonists may cause serious side effects? 

A:We have an example of use of a chronic antagonist for opiate dependence in naltrexone.  Theoretically (as there are no studies yet), one can hypothesize that chronic blockade of the CB1 receptor with a compound like rimonabant would keep someone from feeling the effects of using marijuana, therefore their drug-taking behavior would extinguish.  So, chronic antagonist treatment may be an effective therapy.  Again, studies would have to be done to see if this is actually true, but it is a plausible hypothesis based on what is seen with naltrexone and heroin use.  Now with oral treatment you’d be blocking CB1 receptors throughout the body, not just those in the brain responsible for cannabis use. So in theory, there might be other significant side effects. The antagonist rimonabant is being developed for other indications, specifically obesity and cigarette  smoking. More than 12,000 people worldwide have received rimonabant and the side effect profile suggests that it is fairly well tolerated vs. placebo. There are a small percentage of patients with psychological side effects like depressed mood, some people suffer GI side effects, but overall it is well tolerated.