The spread of medical marijuana to treat chronic pain is welcome to some people, and alarming to others. On the plus side, marijuana, or cannabis, has a long history of use and some evidence of benefits—many patients swear by the herb to relieve their pain, and physicians see it helping those who have few other options. To others, medical marijuana—which lacks both standardized formulations and rigorous clinical testing—represents a challenging deviation from the ideal of evidence-based medicine.
Should physicians recommend cannabis to their patients with intractable pain? That was the underlying question put forward in a session at the World Congress on Pain in Buenos Aires. The turnout for the debate-style session was high, a reflection of the interest in the topic, said session chair Dwight Moulin, Western University, London, Canada. (The debate did not consider the political or legal aspects of medical marijuana, which, in fact, remains against the law in most places in the world.)
The proposal on the table for debate was “Cannabis is an efficacious and safe analgesic for neuropathic pain.” To be precise, the term “cannabis” specified the plant, Cannabis sativa, commonly known as marijuana, delivered by smoking or via a vaporizer. The debate was not about synthetic cannabinoids that are approved or are making their way through clinical trials.
Taking the pro side was Mark Ware, a pain clinician at McGill University, Montreal, Canada. Ware is the director of clinical research at the Alan Edwards Pain Management Unit at McGill and director of the nonprofit Canadian Consortium for the Investigation of Cannabinoids (CCIC). On the con side was Andrew S.C. Rice, Professor of Pain Research at Imperial College, London, UK. Rice, who is also a practicing pain clinician, has done research on cannabinoids and their use for pain.
Both Ware and Rice generously provided PRF with copies of their slide presentations from the debate, which can be downloaded for viewing below.
Ware argued that clinical trial results and the experience with approved cannabis-related drugs offer sufficient evidence that herbal cannabis is safe and efficacious. On top of that, he said the clinical need is dire. In Canada, the government provides a standardized cannabis product, and it should be part of the physician’s armamentarium against pain, he argued.
“People are afraid of cannabis,” Ware told the audience. “But it’s important that you check that feeling at the door and think carefully about the issues.”
Arguing against the use of cannabis for pain, Rice referred to a new meta-analysis of clinical trials in this area by the IASP Neuropathic Pain Special Interest Group, which found that overall, a review of robust trials does not support efficacy of cannabis for treatment of neuropathic pain (see PRF related news blog). However, Rice’s main arguments against cannabis related to the safety side of the equation. For Rice, there are too many unanswered questions about long-term consequences of smoked or inhaled cannabis to support the conclusion that the therapy is safe. “There are long-term adverse effects we have to pay attention to,” Rice said. He cautioned that a haphazard approach to medical marijuana that does not take into account emerging safety concerns could “poison the well” for future development of cannabinoids, which he considers a promising new class of drugs for pain relief.
A vote for efficacy
The scientific rationale for the use of cannabinoids in pain is clear, Ware explained: Throughout the nervous system, endogenous cannabinoids act as synaptic circuit breakers. These small lipid neuromodulators bind to specific receptors on presynaptic terminals, where they block the release of neurotransmitters. Because their receptors are widespread, cannabinoids, whether endogenous, plant-derived, or synthetic, regulate neurotransmission in many circuits, including in pain-producing pathways (for a complete overview of cannabinoids in pain, see Hohmann and Rice’s Chapter 38 in the Wall and Melzack’s Textbook of Pain, Sixth Edition, 2013). To the extent that neuropathic pain stems from abnormal neuronal activity, it makes sense that cannabinoid receptor (CB) agonists should be analgesic, and their effectiveness has been shown in many models of neuropathic pain in animals, using both natural and synthetic cannabinoids.
The active ingredients in Cannabis sativa include the main psychoactive cannabinoid, delta9-tetrohydrocannabinol (delta9-THC), the non-psychoactive cannabidiol (CBD), and others. Prescription cannabinoids have been on the market for decades, starting in 1985 when the US Federal Drug Administration approved a purified THC preparation for the treatment of nausea. Most cannabinoid-based therapies are approved to treat cancer- or HIV-related anorexia and nausea, but they are also prescribed off-label for pain. The only cannabinoid approved specially for pain is nabiximols, a standardized plant extract containing THC and CBD marketed in Canada for pain in people with multiple sclerosis (MS). In other countries, nabiximols is approved for MS-related spasticity, but not pain. That may change, as nabiximols is now progressing through a Phase 3 trial in the US for cancer pain.
The development of cannabinoids has been a patient-driven exercise, Ware said. Many patients self-medicate with herbal cannabis for conditions including pain, spasticity, seizures, and anxiety. Many smoke small amounts that are not sufficient to get high, but do ease their complaints. “Cannabinoids have evolved as medicines because people say that they work,” he said.
Among the literature cited as proof of efficacy, Ware highlighted a review from none other than Rice and colleagues (Phillips et al., 2010), where the authors concede evidence of efficacy for smoked cannabis against painful HIV-associated sensory neuropathy, based on a meta-analysis of two trials.
But overall, clinical trial data on herbal cannabis are sparse compared to purified or synthetic cannabinoids. It has been very hard to do proper clinical trials for smoked cannabis, Ware said. The actual product is one barrier. After Canada legalized medical marijuana in 2001, patients sued the government for access to cannabis. Now, the national government contracts with growers to provide standardized, quality-controlled cannabis for medicinal use. But even with that, Ware said, the pharmaceutical industry is not interested in cannabis, and so there have been only a handful of randomized clinical trials. In the US, the biggest hindrance to trials is law: it remains a federal crime to possess or use cannabis, despite laws in some states decriminalizing medical or recreational use.
Doctors and others may balk at recommending their patients smoke cannabis because of possible lung damage. The absorption and pharmacokinetics of orally delivered forms of cannabis (“edibles”) is notoriously unpredictable, which is why many users prefer smoking. But Ware pointed out that the use of vaporizers can overcome the need to smoke—vaporizers heat the herb enough to volatilize the THC without actually burning it and deliver THC just as effectively as smoking (Abrams et al., 2007). A newer technology, a cannabis inhaler, is being developed that uses cannabis processed into granules. Ware pointed to a new study showing that the device offers the possibility of more controlled dosing that keeps blood THC way below what he called “recreational levels”—the levels required to get high (Eisenberg et al., 2014).
What of adverse effects? The lethal dose of cannabis is unknown, and Ware argued that data show most side effects are short term (dizziness, dry mouth, drowsiness), despite the fact that the population of users is very ill, have both medical and psychiatric comorbidities, use other medications, and may have substance abuse issues (Wang et al., 2008). He claimed most adverse events are derived from recreational users trying to get stoned, including impaired driving, psychosis, sedation, anxiety, tolerance, dependence, changes in cognitive function, and other effects.
Taking the podium, Rice noted that a lot of animal data clearly support the use of cannabinoids for pain. But even so, he characterized the human trial data for neuropathic pain using existing formulations, including smoked cannabis, as “equivocal.” Cannabis is not approved for treating neuropathic pain in any country or locale, he stressed.
Rice agreed that minor adverse events are frequent in short-term trials and are not a concern. But, even given apparent short-term safety, Rice could not endorse cannabis for patients with neuropathic pain. The reason, he said, is the accumulating evidence of long-term harms associated with cannabis use. “Is it safe?” Rice asked. “I don’t know.”
Studies have tied cannabis use to mental illness and cognitive impairment. Rice reviewed studies of recreational cannabis that revealed a link between pot smoking and an increased risk of psychosis and schizophrenia. A meta-analysis of seven studies gave an overall threefold dose-related increased risk of psychosis among cannabis users (Semple et al., 2005). The risk has been mainly studied in adolescent users, and the generalizability of the findings to other populations is unknown, Rice said.
“We should all be worried about that [risk],” he said, noting that a twofold increase in risk of heart events sank the COX2 inhibitors. The causal links between pot and psychosis are unclear, but genetic factors play a role. So does dose, which has been going up over time—the THC content in street supplies of cannabis has increased by fourfold over the last 30 years (Volkow et al., 2014).
Studies also show cannabis use is tied to long-term cognitive decline. Like the psychosis results, the worst effects are seen in people who start cannabis use in adolescence and who use cannabis persistently. In one study, persistent use was associated with broad and noticeable cognitive decline in a cohort followed from birth to age 38 (Meier et al., 2012).
Most of the studies on the long-term effects of cannabis involve adolescent use, a time when the risk of psychosis is high, so the relevance for long-term therapeutic use in older and sicker people is not clear. But Rice argued that the cognitive decline seen in adolescents might be something that is relevant to older populations and cannot be ignored.
As a cautionary tale, Rice pointed to the experience with rimonabant, which blocks brain CB receptors and was developed to treat obesity and aid smoking cessation. The drug was approved for obesity in the US in 2006, but had to be withdrawn from the market in 2009 because post-marketing research found it increased depression and suicidality. That shows the dangers of tinkering with the cannabinoid system in the brain, Rice said. It also highlights the critical need for long-term follow up in future clinical trials of cannabis or cannabinoids.
As an additional safety concern, Rice pointed to the unknown potential for abuse or diversion of new and rapidly acting oral mucosal spray formulations of potent brain-penetrant cannabinoids.
Ware was unconvinced by these arguments. “We are not talking about recommending cannabis to young people. We are talking about patients with severe, intractable pain who have failed other therapies,” he countered.
On the issue of psychosis, Ware stressed that psychosis is currently a contraindication for prescription of any cannabinoids. “We know how to use cannabis, and we know how to monitor it. Let’s use it—cannabis could be a potentially safe and efficacious treatment for neuropathic pain.”
Rice replied, “Physicians need to think back to the Hippocratic oath and ask if we should be prescribing drugs which are of uncertain efficacy for which we do not know the long-term safety profile.”
A better approach, Rice proposed, one that the pharmaceutical industry is pursuing, is to develop alternative ways to modulate the cannabinoid system to produce analgesia. Going forward, Rice said, researchers should aim at improving the therapeutic index of cannabinoids, including identifying non-psychoactive analogues and non-brain-penetrant analogues that can produce analgesia at sites outside of the central nervous system. In addition, he pointed to ongoing clinical trials with fatty acid amide hydrolase (FAAH) and monoacylglycerol (MAG) lipase inhibitors, which prevent breakdown of endogenous cannabinoids.
“I wouldn’t have worked on this for 10 years in my lab if I haven’t believed there’s an analgesic mechanism,” Rice said. “We started this because of what our patients are telling us. But we don’t want a very useful drug target being destroyed like we saw with COX2 because we didn't pay proper attention to adverse effects.”
Ware said the pharmaceutical approach is “promising, highly protected with patents, and highly financially rewarding,” and “will take five to 10 years.”
“We have patients struggling now and need to figure out how to use this today,” he said.
During the discussion period, several audience members mentioned the opioid-sparing effects of cannabis, where anecdotal reports and their experience hold that cannabis users needed much lower doses of opioids to control their pain. “In Canada and other countries, we have licit supplies of cannabis of known quality, so maybe we can reduce harms associated with the use of opioid drugs,” Ware said.
Debates have winners and losers, and a decidedly unscientific show of hands before and after the session seemed to favor Ware’s argument. When asked at the beginning of the session, the attendees appeared evenly divided among yes, no, and not sure. By the end, the yes group had grown ever so slightly.
Where do you fall on this issue? Please join the debate by leaving a comment below.
Image: Wikimedia Commons