When I first started writing about the claims made for medical marijuana and the cannabis oil derived from it, it didn’t take long for me to characterize medical claims for cannabis as the “new herbalism,” as opposed to pharmacognosy, the branch of pharmacology devoted to the study of natural products. The reason is simple. Although I support legalization of marijuana for recreational use, when I look at how medical marijuana has been promoted as a “foot-in-the-door” prelude to legalization, I see testimonials and flimsy evidence ruling over all. I see all the hallmarks of alternative medicine herbalism and none of the hallmarks of pharmacology. Here’s what I mean. Pharmacognosy examines an herb, plant, or other natural product and seeks to identify the chemicals within it that have pharmacological activity against a condition or a disease, the better to purify and isolate those chemicals and turn them into drugs. Herbalism, on the other hand, emphasizes the use of whole plants or extracts from plants, rather than the isolation of the most active compounds. Thus, herbal remedies often contain hundreds, or even thousands, of different compounds, of which only one or a few are active. Even extracts, such as cannabis oil, contain many compounds.

In contrast to pharmacognosy, herbalists make the claim that whole herbs and plant components possess a synergy that is missing from the purified active constituents and/or that the mixture is safer than the pure components because one compound can reduce the side effects of another without reducing therapeutic efficacy. When looked at closely neither claim stands up to scrutiny. Synergism between plant constituents is rare and very difficult to demonstrate, for example. In essence, herbalism turns back the clock 200 years to a time before scientists had developed the techniques and abilities to isolate active ingredients with pharmaceutical activity. Moreover, herbalism, in contrast to pharmacognosy, emphasizes anecdotes over scientific evidence.

Indeed, in my previous posts in this series on medical marijuana, one theme has emerged, which is that cannabis—specifically, a class of active chemicals in marijuana known as cannabinoids—has potential for some diseases but is not the panacea claimed by its proponents. It does not cure cancer, for instance, contrary to glowing testimonials promoted by people like Rick Simpson. For other conditions, the evidence is either not particularly compelling or only mildly promising.

So I reacted with considerable dismay on Friday night when I saw this news report on the 11 o’clock news, “Michigan panel recommends allowing marijuana for autism“:

video platformvideo managementvideo solutionsvideo player

And, indeed, there have been stories all over the local news in Michigan like the one above and this one:

Note all the hearty cheering from the audience when the vote tally was read.

Here’s what happened:

Michigan would become the first state to allow medical marijuana for children with severe autism if a senior official follows the recommendation made Friday by an advisory panel.

The state’s Medical Marijuana Review Panel voted 4-2 to recommend autism as a condition that qualifies for the drug.

Supporters say oil extracted from marijuana has been effective in controlling extreme physical behavior by kids with severe autism. Pot wouldn’t be smoked.

The panel was influenced by comments received earlier from some Detroit-area doctors, especially the head of pediatric neurology at Children’s Hospital of Michigan, and parents desperate for relief. Many of the three dozen spectators cheered and applauded after the vote.

This is an utterly horrible idea, but it looks as though our state is nonetheless about to take the plunge into the Brave New World of treating autism with cannabis oil. Indeed, Michael Komorn, the President of the Michigan Medical Marijuana Association and the attorney who brought the petition before the board, basically admitted in the first news clip above that there was no science behind the board’s recommendations when he said: “It’s a no-brainer. And you heard the testimony of these people. They just want a little hope. That’s all they’re asking for.” Notice the distinct lack of any mention of strong science as a rationale for adding autism to the list of qualifying conditions. Komorn even went so far as to tout how the parents “are responsible for growing the plants or acquiring the cannabis, and they are in charge of dosing, frequency of use, and method of ingestion.”

Think about this for a moment. Is there another regulated drug for which this is the case? The board didn’t even provide, along with its recommendation for approving autism as a qualifying condition for treatment with cannabis oil, anything resembling recommendations for dosing, method of ingestion, and frequency of use, something very basic that is required by the FDA for even over-the-counter medications. Yet, here, in the fantasy world of medical marijuana, apparently parents are as knowledgeable as physicians and no guidelines for use are needed. The lack of dosing recommendations is another powerful indication that there is no science behind this recommendation, because if there was science behind it we’d know the optimal dose and method of administration to use cannabis oil to treat autism. Clearly, emotion, not science, ruled the day on Friday. Indeed, how Dr. Harry T. Chugani, the head of pediatric neurology at an institution as respected as Children’s Hospital could testify in favor of this is beyond me, but apparently he did:

“It seems to work. … Wouldn’t that be better than giving them all these psychiatric drugs?” Chugani said. “Not every autistic kid would take this, but if your behavior is wild and you have to be institutionalized, I as a physician would prefer to try medical marijuana. I have at least 50 patients on multiple drugs and still their behaviors are not controlled.”

Notice that Dr. Chugani, too, didn’t cite anything other than anecdotal evidence. How, for instance, did he know that the child isn’t calmer because he’s high on cannabis? And if we’re going to drug autistic children to make them behave, why not use opioids or other sedatives as well?

In any event, if you want evidence that medical marijuana is far more akin to herbalism than pharmacology, we need look no further than my state’s dubious position of becoming a “pioneer” in something that no state should be a pioneer in. That’s why I’ll first examine how the Medical Marijuana Review Panel got to where it is now, then the general claim that cannabis is a useful treatment for autism. Finally, I’ll evaluate the existing scientific evidence, which is pretty thin.

Autism biomed quackery embraces medical marijuana

As has been the case with virtually all uses of medical marijuana sanctioned by states, the road to the Michigan Medical Marijuana Review Panel recommending approval of cannabis for autism began not with doctors and scientists clamoring for it based on evidence, but with parents becoming politically active to lobby for it based on emotional anecdotes. Such was the case in Michigan, and, if you’ve been paying attention to the “autism biomed” movement, this quote will sound very familiar:

The review panel voted 4-2 in favor of a petition submitted by Lisa Smith, a Michigan mother who has said cannabis oil has helped improve her severely autistic 6-year-old son’s behavior, sleep patterns and eating schedule.

“The parents I’ve talked to are passionate and adamant that this represents a dramatic improvement in the quality of life for them and their affected children,” said David Crocker, a medical marijuana doctor and member of the panel.

What other treatments for autism have we heard the same claims for? Let me think… Oh, yes. We’ve heard them for quite a few “autism biomed” treatments, with parents being just as passionate. Unfortunately, the vast majority of “autism biomed” is rank quackery. Examples include “Miracle Mineral Solution” (MMS, a.k.a a form of bleach fed to autistic children and administered as enemas for which miraculous results are claimed), chemical castration with Lupron, chelation therapy, GcMAF, and many, many other pseudoscientific “treatments” featured at quackfests like Autism One. I’ve even seen glowing testimonials touting homeopathy for autism and the IonCleanse® “foot detox” bath. As regular readers know, homeopathy is, as I like to say, The One Quackery To Rule Them All, and “foot detox” baths are a scam.

Not surprisingly, segments of the autism biomed movement have embraced medical marijuana. For example, one prominent website advocating autism biomed quackery, the “Thinking Moms’ Revolution” (TMR), has numerous posts on its blog extolling the alleged virtues of medical marijuana for treating autism and advocating “freedom of choice.” If you doubt the increasing embrace of medical marijuana by the autism biomed quackery movement, look no further than this ad for an Medical Marijuana for Children with Autism eConference, sponsored in part by TMR. Indeed, Jeff Bradstreet, the longtime antivaccine autism biomed “practitioner” who committed suicide a month and a half ago after the FDA raided his clinic, was a speaker! So was a key supporter of medical marijuana for autism in Michigan, Lester Grinspoon.

Does this striking resemblance between autism biomed rhetoric and medical marijuana rhetoric mean that cannabis is useless for autism? Not necessarily, but it raises red flags. In any event, it would be more accurate to say that we don’t really know whether cannabis oil is efficacious for treating autism. I mention this similarity between the autism biomed movement and medical marijuana movement to emphasize how anecdotal evidence is incredibly unreliable, particularly for a complex condition like autism spectrum disorder (ASD). If you don’t believe me, consider this. Equally glowing testimonials for the rank quackery that is homeopathy to treat autism are not difficult to come by, which is one reason why testimonials are insufficient evidence upon which to base public policy; yet, that’s just what is happening in Michigan—and has been since medical marijuana was first approved. Remember, autism is a condition of developmental delay, not stasis, and autistic children frequently improve as they get older. Some even improve sufficiently to lose the ASD diagnosis.

Misguided activism advances cannabis for autism in Michigan

I understand (at least as much as someone who hasn’t actually experienced having a child with special needs can) how parents would be desperate to do everything they can for their children, but personal experience and anecdotes, contrary to what is claimed, can be extremely misleading, as we’ve described here many times before. That’s why anecdotes are not enough, and carefully controlled randomized clinical trials are needed, which, contrary to the claims of advocates, don’t really exist. Indeed, the science cited in the petition, as we will see in the final section of this post, is inadequate to make such a major policy change, consisting as it does of mainly preclinical evidence and case reports.

With that in mind, it’s useful to note that the road to this “victory” began years ago and actually represents a rebound from a defeat two years ago. In 2013, the Michigan Medical Marihuana Review Panel voted against adding autism to the list of indications for medical marijuana by a 7-2 vote:

Jenny Allen, whose 6-year-old son was diagnosed with autism several years ago, has tried giving him “mind-bending” medications, signed him up for behavior therapy and changed his diet. But his problems, including self-destructive behavior and biting, continue.

Now the 32-year-old Lansing mom wants to try giving him part of a brownie — a pot brownie — but was brought to tears Tuesday when the Michigan Medical Marihuana Review Panel rejected a petition that would have given her the legal means to do so.

The panel, in a 7-2 vote, gave a final recommendation against adding autism to a list of debilitating conditions suitable for treatment under Michigan’s voter-approved medical marijuana law.

“I’m incredibly disappointed,” Allen told MLive after the hearing, going on to question whether all panelists had thoroughly researched the topic. “I’m pretty shocked, actually, that nobody even brought up what the base condition is. It’s rather appalling.”

The panel made this decision in 2013 based on a correct assessment of the state of the evidence, which is that there is “not much quality, peer-reviewed research exploring marijuana as a treatment for autism,” that the case for adding autism to the list of approved conditions for which medical marijuana can be prescribed consists almost entirely of anecdotes and testimonials, and that not enough is known about the effects of long term use of cannabis on the developing brain. This assessment was correct in 2013, and nothing has happened in terms of the existing science during the intervening two years to change that assessment. So what really happened to reopen the case?

Not surprisingly, it was litigation that forced the Department of Licensing and Regulatory Affairs to submit a new petition for autism to the Medical Marihuana Review Panel. It was submitted on behalf of a woman named Lisa Smith for her son Noah, who has severe autism:

Lisa Smith says her son’s behavior was dangerous: hair pulling, kicks, punches, all related to a severe form of autism. But it began to change more than a year ago when he was given daily oral doses of oil extracted from marijuana.

“That’s all stopped. He’s more focused, he’s calmer,” Smith said of 6-year-old Noah. “He sleeps better through the night. He has a better appetite. You can tell he’s growing, gaining weight.”

Another parent who features in several of the news stories about this issue is Dwight Zahringer:

Dwight and Ixchel Zahringer’s son Brunello is going on 4 but has yet to speak. Last fall, his parents heard the chilling diagnosis — autism.

“We’ve had a hard, fast education in this for the last nine months,” Dwight Zahringer said. “Think of it like always being at a rock concert — everything really loud — and then you’re trying to have conversations or focus on things but you can’t because everything is overwhelming,” he said.

The dealings with health professionals form a familiar tale — advice to use powerful prescription drugs that are costly and may have worrisome side effects, the failure to see that those drugs are helping and the decision to stop them.

The Zahringers have thus far been disappointed with the progress their son Brunello has made with conventional therapy, such as Applied Behavioral Analysis. Seeking faster progress, they found medical marijuana and latched on to it for hope:

“We’ve been watching a lot of videos, a lot of documentaries, and we’ve seen proof that it can help,” said Ixchel Zahringer.

“One family had a kid who was very severe” with autism symptoms, “and they started giving him some of that (cannabis) oil, and they saw the child calming down.”

Dwight Zahringer came to his belief that he needs medical marijuana for his Brunello the same way that many in the autism biomed movement come to the conclusion that their child needs, for example, MMS. He pored over Internet sites and marijuana-themed literature, which convinced him that he needed to try cannabis on his son ASAP. He started using hemp oil, which apparently gets around the state’s current ban on treating autism with marijuana, but really wants the real thing. Parents like the Zahringers and Lisa Smith are clearly loving parents who want the best for their children, just as most parents who fall for “autism biomed” quackery love their children and want the best for them. Unfortunately, as we have seen and will see, the pro-cannabis literature and websites that tout cannabis as a treatment for autism provide a very biased and cherry-picked view of the medical literature.

Cannabis for autism: The evidence (or, more correctly, the lack thereof)

Perusing the news reports on the vote of the Michigan Medical Marijuana Review Panel, one point I’ve seen is that the reason why this vote came down in favor of adding severe autism to the list of qualifying conditions and the vote in 2013 did not is because the science was so much better described in the petition this time around. Indeed, advocates tout having bolstered the petition with “over 75 peer reviewed articles with over 800 pages of research on the issue of cannabis as a viable option for the treatment of autism” in addition to the “nineteen families, as well as physicians from MI and around the country.” Apparently, two advocates had a major hand in picking these articles:

The individuals who navigated the deepest into the science behind Autism’s riddling labyrinth of theories are without doubt Joe Stone and Dr. Christian Bogner. They were able to provide peer reviewed evidence that cannabis not only has the potential to provide palliative relief of symptoms related to autism, but may also have the potential to target the underlying causes of autism itself.

Note that nowhere in this report is mentioned evidence from well-designed clinical trials. In any event, I could not find the exact petition text, but there is a MoveOn.org petition to add autism to the list of qualifying conditions for medical marijuana in Michigan posted by Joe Stone and signed by Dr. Christian Bogner, Chad Carr, Dr. Harry Chugani, M.D., Dr. Lester Grinspoon, Michael A. Komorn, and Joe Stone. At the end, it lists links to research, specifically a paper written by Stone and Bogner that, one notes, was published online on the Cannabis for Autism blog rather than in a peer-reviewed medical journal, entitled “The Endocannabinoid System as it Relates to Autism“. This paper is also available on Scribd, with a complete list of references. It is clear to me that this discussion, along with its references, was the basis of the science presented to the panel.

Before I address this paper, which appears to be the best evidence advocates can put forward, let me just refer you to a good solid analysis of the state of the evidence for cannabis for neurological disorders by Skeptical Raptor. You can read the whole thing if interested, but the point I want to emphasize is that there is weak evidence that cannabis could be useful for epilepsy, as I discussed in my first post on this topic, but that a Cochrane review concluded that high quality evidence was insufficient to recommend it. There is also some evidence for the use of medical marijuana for spasticity and pain. The reason that this is relevant is because advocates for using medical marijuana for autism frequently point to cases where cannabis is used to control seizures, which many autistic children suffer from. This is a separate issue than whether cannabis is specifically therapeutic in autism, but the two issues are often conflated.

Sadly, Stone and Bogner’s “paper” is one of the most blatant examples of cherry-picked research I’ve seen in a long time, and that’s saying something. Indeed, plowing through the list and looking up key papers was a tediously predictable endeavor. Basically, Stone and Bogner take papers that look at some aspect of cannabinoid function, whether it’s about autism or not, and extrapolate to autism. Let’s take a look at their most convincing piece of evidence first (to me, at least). In this case, by “most convincing,” I mean least unconvincing:

“Rare mutations in neuroligins and nerexins predispose to autism” (Földy 2013). Neuroligin-3 is the only known protein required for tonic secretion of endocannabinoids that include AEA and 2-AG (Földy 2013). Neuroligin-3 mutations have been shown to inhibit tonic endocannabinoid secretion (Földy 2013). These alterations in endocannabinoid signaling may contribute to autism pathophysiology (Földy 2013, Krueger 2013, Onaivi 2011, Siniscalco 2013). These finding have in part prompted researchers to apply to conduct research with nonhuman primates in order to further elucidate this link (Malcher-Lopes 2013).

Endocannabinoid system deficiencies are suggested to be involved in the pathophysiology of a growing number of diseases (Marco 2012, Russo 2003). Pacher and Pertwee both cover the endocannabinoid system in detail (Pacher 2006, Pertwee 2010). The number of functions that endocannabinoid signaling regulate in the human body is extensive and beyond the scope of this paper (Pertwee 2010). For sake of brevity only a few potentially relevant aspects will be listed:

So what are these elements? Basically, Stone and Bogner cite a whole bunch of papers supporting the conclusions that endocannabinoids:

  • Modulate synaptic function
  • Regulate GI functions
  • Suppress proliferation and cytokine release in the central nervous system (CNS)
  • Regulate stress responses
  • Increase cerebral blood flow
  • Modulate neural and glial cell function

…you get the idea. So, yes, cannabinoids are important molecules in the CNS and elsewhere. No one argues against that. It’s also true that it was recently shown that mutations in neuroligins and nerexins appear to predispose children to autism and that neuroligin-3 mutations inhibit endocannabinoid secretion, suggesting that bolstering endocannabinoid secretion might antagonize or reverse the abnormalities associated with such mutations. (A decent overview of this research suitable for a lay person can be found here, and the original paper is here.) Let’s just put it this way. This is a rodent model, and right there it’s a question of how relevant it is to real, human ASDs. This is very preclinical evidence, meaning that it might or might not turn out to be relevant to human ASDs. At best, it justifies further study. Moreover, the primate model referenced in Stone and Bogner’s paper appears to be merely a proposal presented to a conference in Qatar. This is thin gruel indeed, as a recent review article on medical marijuana published in the Journal of Developmental & Behavioral Pediatrics by Hadland et al. points out:

Many advocates cite scientific literature regarding benefits of cannabis for the treatment of pediatric behavioral conditions, but often, data cited are from animal model-based research that does not yet have translation to human subjects. For example, a 2013 study 80 from Stanford University showed that mice with a specific and rare gene mutation linked to autism showed altered endocannabinoid signaling in the central nervous system. These data were then cited by online and print media supporters of medical marijuana (e.g., the High Times 81) as evidence that cannabis could be used as a treatment for autism. As another example, when another recent study 72 based on a mouse model of Fragile X syndrome (described earlier in this review) showed alterations in endocannabinoid signaling pathways, these data were referenced (in this case, by more mainstream media outlets, such as the Huffington Post 8 and Fox News 82) as evidence for a promising role for cannabis as treatment. Although these and other high-impact studies share important insights into the pathogenesis of autism spectrum disorders (ASD) and Fragile X syndrome, based on their results alone, it is erroneous and potentially harmful to conclude that cannabis should be used as treatment for either of these disorders at this time.

Indeed. Not surprisingly, Stone and Bogner’s article references that very same study on Fragile X syndrome as well. Much of the rest of their article boils down to the list I enumerated above plus other correlations, their arguments reduced to, in essence, this:

  1. The endocannabinoid system is important in [insert important CNS function or signaling pathway here] in preclinical cell culture and animal models.
  2. Autism and ASDs involve abnormalities of this important CNS function or signaling pathway.
  3. Ergo, cannabinoids can be used to treat autism and ASDs.
  4. Q.E.D.

In other words, there’s a whole lot of confusing correlation with causation and assuming pathogenesis when what is being observed might just be an epiphenomenon. There might be a direct role for these correlations in causing autism, and some of the signaling pathways might even represent promising targets, including the endocannabinoid pathway or a subset of it. Stone and Bogner go wrong by assuming all of these studies indicate a critical role for endocannabinoids in autism, such as elevated cytokine levels. For instance, take a look at this list of effects of cannabidiol (CBD):

  • CB1/CB2 agonist blocker (can inhibit overstimulation of CB1 by THC)
  • FAAH inhibition increases endocannabinoid levels (including AEA, 2-AG)
  • AEA reuptake inhibitor
  • 5-HT1a agonist
  • Suppressor of tryptophan degradation
  • PPAR alpha and gamma agonist Positive allosteric modulator at glycine receptors
  • TRPV1 and TRPV2 agonist
  • Adenosine uptake competitive inhibitor
  • Antagonist at abnormal-CBD receptor
  • Regulator of intracellular Ca 2+ T-type Ca 2+ channel inhibitor (Izzo 2009)

This is a wide range of effects, some of which could be relevant to autism/ASD. The problem, of course, is that we don’t know which ones are the most relevant and which ones actually involve promising therapeutic targets for intervention.

Extrapolate, extrapolate, extrapolate!

Whenever writing a research paper that is basic science that could potentially be translated into a treatment for a disease, it is generally considered mandatory to speculate at the end just how this could come about. For instance, when scientists write about cannabinoids in the context of models of neurodevelopmental disorders, after all the basic science, cell culture, and animal model work, naturally they try to describe how their results could be pursued so that they translate into a clinical treatment. It’s known as showing clinical relevance to your findings, no matter how basic science they are. Stone and Bogner quote several of these sorts of speculative statements in the discussion or introduction of papers as though they were Gospel truth, then conclude:

Given the known role of the endocannabinoid system in ASD it seems entirely possible, if not likely, that cannabinoid rich botanical extracts from cannabis can be utilized as useful agents targeting the pathophysiology of ASD, as well as the many debilitating symptoms and conditions associated with it. The wealth of options that cannabis has to offer those that suffer from ASD in MI is not currently legally permitted. We believe that needs to change.

As I said before, Q.E.D. (Yes, that’s sarcasm.)

Where’s the beef (i.e., the clinical evidence)?

What all these 75 references really mean is that there is some correlative evidence that the endocannabinoid system is abnormal in autism. However, it’s not at all clear whether these abnormalities are causative or downstream effects from the true cause or causes, whatever they might be. What this evidence means is that it’s probably worthwhile to study the endocannibinoid system in autism and whether modulating its activity can have an effect on autistic symptoms. What it most definitely isn’t is compelling evidence to authorize any doctor in the state who wants to do so to use cannabis oil to treat autism. Yes, because the law states two doctors have to sign off for use of medical marijuana in children, parents will have to find two doctors, but, really, does anyone think that will be very difficult?

To recommend a treatment for general use, we need high quality clinical evidence. Is there any such evidence for cannabis oil for autism? The answer is a resounding and unequivocal no. Indeed, an excellent indication of the paucity of evidence regarding cannabis oil and autism is the way Stone and Bogner dance around the issue by citing anecdotal reports about the use of cannabis oil and cannabinoids to treat epilepsy:

How can combinations of cannabinoids be put into practical use by individual families? For our purposes let’s review the anecdotal reports of cannabinoid based treatments currently being utilized in MI (and around the world) for pediatric epilepsy. I think this is a good comparison due to the range and complexity of both conditions. CBD continues to prove its effectiveness in treating many types of epilepsy, but not all (Porter 2013). Anecdotal reports provided in online groups with families that share dosing and other related information to cannabinoid based pediatric epilepsy treatments reveal that in many cases parents (and physicians) find that an increased ratio of THC is required to increase the efficacy of treatment. The range seems to vary significantly from 24:1 to 1:1 (CBD:THC). Some partial explanations for this might include the ability of THC to increase GABAergic transmissions via CB1 activation, its modulation of ion channels, and that it’s a PPAR gamma agonist which is neuroprotective in epilepsy (Stone 2014).

Due to the range of ASD it seems possible that, similar to cannabinoid based epilepsy treatments, varying ratios of cannabinoids (specifically CBD:THC) will prove to have a greater efficacy overall when compared to individual cannabinoid based treatments (like Dronobinal). This concept has been further supported by the research and clinical use of Sativex, a 1:1 (CBD:THC) botanical extract marketed for use in a range of treatments throughout the world (Hazekamp 2013, Russo 2006). The ability to specifically tailor cannabinoid ratios in botanical extracts from cannabis in a case specific manner may prove an even greater efficacy.

This is nothing more than handwaving, comparing two different conditions and assuming that what is observed in one condition will apply to another condition. Worse, as Skeptical Raptor reminds us, despite these anecdotal reports of benefit due to cannabinoids or cannabis oil in epilepsy, a recent Cochrane Review concludes that “no reliable conclusions can be drawn at present regarding the efficacy of cannabinoids as a treatment for epilepsy.” In other words, there’s no high quality evidence that cannabinoids are efficacious in treating epilepsy. There’s even less evidence that cannabinoids can be used to treat autism. Indeed, as the aforementioned review by Hadland et al. notes:

Regarding human data on use of cannabis for developmental and behavioral conditions, to the best of our knowledge, the only available data are from small case series or single studies. For example, one 6-year-old boy with autism was treated with daily dronabinol for 6 months and was noted to have improvement in hyperactivity, irritability, lethargy, stereotyped behaviors, and speech, as measured by the Aberrant Behavior Checklist.83 This single case study was uncontrolled and unblinded. In another single case study 84 of a cannabis-using adult male with attention-deficit hyperactivity disorder (ADHD) off stimulants, the subject’s driving skills in a simulated test during a time of abstinence improved after smoking marijuana (What is unclear is whether this subject may have actually been experiencing cannabis withdrawal from his abstinence, with alleviation of his symptoms through subsequent use of marijuana.85). Another small case series 86 showed an improvement in self-injurious behaviors among adolescents after dronabinol therapy, but to date, the study has not been published, leaving protocol details scarce. In sum, none of these studies provides sufficient, high-quality data to suggest that cannabis should be recommended for treatment of ASD or ADHD at this time.

And neither does Stone and Bogner’s analysis, which was the basis of the evidence submitted with the petition to the Michigan Medical Marijuana Review Board to add autism as a qualifying condition for medical marijuana treatment. Their analysis is the very definition of cherry picking studies and extrapolating wildly from preclinical cell culture and animal studies and studies that address other conditions to conclude that cannabinoids are efficacious treating autism, while ignoring the dearth of evidence that counts: Actual clinical trial evidence. As the Cochrane review I cited pointed out, there were four randomized trials including a total of 48 patients using cannabinoids to treat autism. One report was just an abstract; another a letter to the editor. None of the trials provided randomization details, and there was no description of whether the control and treatment groups were equivalent. The studies were thus of incredibly low quality. All there are, are a handful of uninformative single patient case reports like this one.

Yet, when this this incredibly thin gruel was combined with emotional testimonials of distraught parents of severely autistic children, it was the emotional testimonials of distraught parents that won out. There isn’t another drug for which the FDA or a state would give doctors the go-ahead to use to treat humans for conditions like autism or cancer based on such slim to nonexistent evidence.

Approving the use of cannabis oil for autism: “Premature” doesn’t even begin to describe it

There is no doubt that the approval of medical marijuana for various medical conditions is driven far more by politics than science or clinical observations. For no condition is that more true than autism, for which even the anecdotal evidence is weaker than it is for other qualifying conditions such as chemotherapy-induced nausea, chronic pain, and epilepsy. Moreover, in this case, the Michigan Medical Marijuana Review Panel abdicated its responsibility to interpret existing science with respect to medical marijuana and autism. It is a travesty that the panel could take the cherry picked list of studies annotated with unjustified extrapolation of preclinical studies and conclude that there was a compelling case for adding autism to the list of qualifying conditions. The panel failed even to require that only relevant specialists, such as pediatric neurologists, be allowed to prescribe cannabis oil for autism and instead let any licensed physician (OK, two licensed physicians) do it. Worse, these physicians don’t even have to monitor how often or how much is given to an autistic child for whom they prescribe medical marijuana. They can leave it all up to the parents, the vast majority of whom have no medical training.

This lack of oversight is a big deal because, contrary to what medical marijuana advocates claim, science still doesn’t have a good understanding of what the long term effects of chronic daily cannabis use are on the developing brain. We do know that teenagers who were found to be dependent on marijuana before age 18 and continued using it into adulthood lose IQ points. One can argue that IQ is a poor surrogate for intelligence, but nonetheless such findings are worrisome. We do know that marijuana use is associated with abnormalities in the brain in young users in an exposure-dependent manner. There are other potential adverse health effects as well. In any decision to use a drug, there is a risk-benefit analysis, and thus far in autism there’s almost no evidence for benefit and troubling evidence of risk when cannabis is used long term in children.

Fortunately, the recommendation of the Michigan Medical Marijuana Review Board is not binding. That recommendation will now go to Mike Zimmer, director of the Michigan Department of Licensing and Regulatory Affairs. He will have the final say over whether autism is added to the list of qualifying conditions. We in Michigan can only hope he realizes what a massive mistake the review board made and overrules their recommendation. Although new evidence might change this in the future, at present, medical marijuana for autism is unscientific herbalism, not pharmacognosy, and has no place in science-based medicine or state policy.



Posted by David Gorski

Dr. Gorski's full information can be found here, along with information for patients. David H. Gorski, MD, PhD, FACS is a surgical oncologist at the Barbara Ann Karmanos Cancer Institute specializing in breast cancer surgery, where he also serves as the American College of Surgeons Committee on Cancer Liaison Physician as well as an Associate Professor of Surgery and member of the faculty of the Graduate Program in Cancer Biology at Wayne State University. If you are a potential patient and found this page through a Google search, please check out Dr. Gorski's biographical information, disclaimers regarding his writings, and notice to patients here.