Please note the information in this publication may no longer be current but is retained on our website for historical or research purposes.
<< Previous Chapter | Contents | Next Chapter>>
8. The therapeutic effects of cannabinoids
8.2 Cannabinoids as anti-emetic agents
Profound nausea and vomiting can be such serious complications of chemotherapy and radiotherapy for cancer that patients may discontinue potentially life-saving treatment (Institute of Medicine, 1982). Although various types of drugs (e.g. the phenothiazines) have been shown to be effective in controlling nausea and vomiting in cancer patients, substantial minorities of patients do not benefit from these drugs. The seriousness of the problem of chemotherapy-induced nausea, and the incomplete success of existing treatments, prompted oncologists in the late 1970s and early 1980s to take a particular interest in the anti-emetic properties of cannabinoids (Institute of Medicine, 1982).
8.2.1 Clinical trials
One of the earliest trials of the effectiveness of THC as an anti-emetic was prompted by patient reports that smoking marijuana relieved nausea and vomiting (Sallan et al, 1975). In this study, 22 patients (10 males and 12 females, average age 30 years) with a variety of neoplasms were studied. In 20 patients, the nausea and vomiting had proven resistant to existing anti-emetic drugs. A randomised placebo-controlled trial with crossover was used, in which patients were randomly assigned to receive oral THC (10mg per m2) and placebo in one of four different orders (THC-placebo-THC; THC-placebo-placebo; placebo-THC-placebo; placebo-THC-THC). Outcome was assessed by grading patients' self-reports of nausea and vomiting after THC and placebo into three categories: complete response if there was vomiting after placebo but not after THC; partial response if there was a greater than 50 per cent reduction in nausea and vomiting after THC compared to placebo; and no response if there was a less than 50 per cent reduction in nausea and vomiting.
Ten patients completed all three courses of THC and placebo and vomited on at least one trial. After excluding one trial because of a variation in the chemotherapy dose, there were 29 trails available for analysis, 14 of placebo and 15 of THC. All 14 placebo trials resulted in no response, while in the 15 THC trials there were five complete responses, seven partial responses, and three no responses. This difference was statistically significant when full and partial responses were combined. Most patients (13/16) reported a "high" after receiving THC, an experience which was correlated with the anti-emetic effect. The most common side-effect was somnolence which curtailed activities for two to six hours in a third of patients. Only two patients experienced any symptoms of toxicity, (both after receiving 20mg doses of THC), namely, visual distortions and hallucinations and depression lasting several hours. Sallan et al reported "preliminary" observations from several patients that smoking marijuana produced an equivalent anti-emetic effect to oral THC.
A trial by Chang et al (1979) largely supported the findings of Sallan et al. In this study 15 patients with osteogenic sarcoma (10 males and five females, average age 24 years) served as their own controls in the course of monthly high dose methotrexate therapy. They were assigned to receive three THC and three placebo trials in randomised order during six treatment sessions. THC (10mg per m2 of body area) and placebo were administered orally five times at three hourly intervals, beginning two hours before chemotherapy. If the patients vomited, the remaining doses of either THC or placebo were administered by smoking a cigarette using a standardised smoking technique. The effect of THC and placebo were assessed by nursing staff who rated various endpoints (e.g. number of vomiting and retching episodes, volume of emesis, degree and duration of nausea) without being aware of which treatment patients had received. Patient response was graded into three categories: excellent (greater then 80 per cent reduction after THC by comparison with placebo in each of these endpoints); fair (greater than 30 per cent and less than 80 per cent reduction), and no response (less than 30 per cent reduction).
The results showed that eight patients had an excellent response, six a fair response, and one had no response. On all endpoints THC produced a statistically significant reduction in nausea and vomiting by comparison with placebo. There was also a dose-response relationship between blood levels of THC and the incidence of nausea and patient reports of feeling "high". Generally, higher THC blood levels were achieved when marijuana was smoked than when THC was taken orally. There were few side effects reported, with sedation being the most common (12/15 patients). Four patients experienced five dysphoric reactions in the course of 281 THC drug doses (2 per cent), none of which lasted more than 30 minutes, and all of which were successfully managed with simple reassurance.
In a second phase of the study, four patients who had an excellent response to THC in the first phase were retested under double-blind conditions using two placebo trials in the next 10 treatments. A small number of patients who had a fair response were also studied using an increased dose of THC. All patients showed a reduction in the average anti-emetic benefit of THC, decreasing from excellent to fair in the case of previous excellent responders, and from fair to no response in the case of the fair responders. Chang et al hypothesised that the decline in effect reflected either the development of tolerance to the effects of THC, or the development of conditioned nausea and vomiting that was resistant to the anti-emetic effects of THC.
Since these early studies, a large number of controlled clinical studies have been conducted which compared the effectiveness of THC with either a placebo or with other anti-emetic drugs (see Carey et al, 1983; Poster et al, 1981; Levitt, 1986 for reviews). The results of this literature have sometimes been unfairly described as "confused" (e.g. Carey et al, 1983; Nahas, 1984). This description betrays an unreasonably high expectation of the consistency of results from studies which have generally used small samples of heterogenous patients who have received various forms of chemotherapy. It also ignores the fact that the cross-over studies comparing the anti-emetic effects of THC with placebo have generally reported greater anti-emetic effects for THC than placebo (Poster et al, 1981); the single exception to this finding was a study which had a sample size of only eight patients.
Comparisons of the effectiveness of oral THC with that of existing anti-emetic agents have been less consistent than the results of comparisons with placebo. Nonetheless, the results have generally indicated that THC is at least equivalent in effectiveness to the widely used anti-emetic drug prochlorperazine (Carey et al, 1983; Levitt, 1986). The inconsistencies in this case arise because some studies have shown THC to be superior, probably because of the practice in some trials of enlisting patients whose nausea had previously proven resistant to prochlorperazine (Carey et al, 1983).
The equivalence of THC and prochlorperazine has been supported by the results of one of the largest and best conducted studies (Ungerleider et al, 1982). In this study 214 patients with a variety of forms of cancer (carcinomas, sarcomas, lymphomas and leukemias) were recruited if they had already undergone chemotherapy and experienced nausea and vomiting, or they were to receive a form of chemotherapy which had a high emetic potential. Patients were randomly assigned to receive a paired trial of either oral THC followed by prochlorperazine or vice versa. The dose of THC was dependent on body surface area (7.5mg if less than 1.4m2, 10mg for 1.4 to 1.8m2, and 12.5mg for greater than 1.8m2). Separate analyses were conducted on three groups of patients: patients who received their cancer chemotherapy on a single day some weeks apart (N=98); patients who received their chemotherapy on a daily basis over several successive days (N=41); and patients who discontinued the trial after a single episode of either THC or prochlorperazine. Outcomes were patient self-ratings of nausea and vomiting, and a variety of mood states and behaviours.
The results showed that there "were no statistically significant differences in the anti-nausea/anti-emetic effect of THC and prochlorperazine" (p640) in any of the three patient groups, even though there were differences between patients in the single- and multiple-day chemotherapy regimens in the time course of the nausea. There were differences in mood and behaviour between the THC and prochlorperazine trials, with patients reporting greater impairment of concentration and less social interaction after receiving THC. There were also more side effects from THC than prochlorperazine, with sedation, sleepiness and mental clouding being the most common. There was no difference in the frequency of panic attacks between the two drugs. Despite these differences in side effects there was a small patient preference in favour of THC as an anti-emetic, with 41 per cent experiencing less nausea on THC, 31 per cent experiencing less nausea on prochlorperazine, and 29 per cent reporting no difference in effectiveness. The effectiveness of THC was not related to age or prior experience with marijuana, but it was related to the experience of side effects, with patients experiencing them reporting less nausea.
Given the wide variety of patients who have been studied in terms of age and type of cancer, the wide variety of chemotherapeutic agents that have been used to treat their cancers, and the variety of different anti-emetics with which THC has been compared, the fact that findings of these studies are generally positive for THC is more impressive than the apparent differences in outcome. The positive results from the controlled trials also seem to be borne out by clinical experience with cannabinoids in managing cancer patients. A recent survey of a large sample of American oncologists, for example,found that 44 per cent of oncologists had recommended marijuana to at least one cancer patient, and 64 per cent of these physicians reported that it was successful controlling nausea in at least half of their patients. Overall, just under half of the oncologists in the sample (44 per cent) believed that cannabinoids could be safely used in the treatment of nausea caused by chemotherapy and radiotherapy (Dobin and Kleiman, 1991). A similar proportion (48 per cent) reported that they would prescribe marijuana for their patients if it was legal.
The general conclusion on the available literature is that THC is superior to placebo, and equivalent in effectiveness to other widely-used anti-emetic drugs, in its capacity to reduce the nausea and vomiting caused by some chemotherapy regimens in some cancer patients. There are a number of issues that remain to be resolved in deciding upon the clinical role of cannabinoids as anti-emetic agents in cancer chemotherapy. These issues include: the types of nausea against which it may be most effective, and hence the types of patients for which they are most appropriately prescribed; the degree of patient tolerance of the psychotropic side effects of THC and other cannabinoids; the potential seriousness of possible THC induced immunosuppression in patients who are already immunologically compromised; the most effective dosing schedules for THC as an anti-emetic; the potential use of THC in combination with other anti-emetic drugs; and the extent to which the motivation for the use of THC may have been reduced by the availability of newer anti-emetic drugs that are more effective than prochlorperazine (the main anti-emetic drug in the 1980s).
8.2.2 Which patients?
Which patients with what types of nausea are the most suitable for treatment with cannabinoids as anti-emetics? Patients with various forms of cancer have been the most extensively investigated patient group, but the numbers of different types of cancer have been too small to allow convincing analyses of differences in patient response. The same point can be made about types of chemotherapy regimens; they have varied widely in these studies, and have often not been reported, but there has been no systematic analysis of the effectiveness of cannabinoids in controlling emesis produced by different agents. It is uncertain to what extent the cannabinoids may be effective against nausea from other causes. The mechanisms that produce nausea are not well understood but there are believed to be one or more protective mechanisms located in the brain stem that can be triggered by a variety of emetic agents. This raises the possibility that cannabinoids may be therapeutically useful against nausea from a variety of causes.
8.2.3 Side effects
The psychoactive effects of cannabis which are prized by recreational users - euphoria, relaxation, drowsiness - are not always welcomed by older patients, most of whom are cannabis-naive. In some studies a substantial minority of such patients have discontinued the use of THC because of the unwelcome dysphoria and somnolence (Levitt et al, 1986). This has not been a universal experience, so further research is required to discover to what extent this has been the result of unnecessarily large doses, or poor patient preparation for these effects, and failure to adequately manage them by reassurance. What does seem to be the case is that the experience of some psychological effects of THC, including the "high", is necessary for the occurrence of a clinically significant anti-emetic effect. This fact has led to the search, so far unsuccessful, for cannabinoid derivatives of THC which possess its anti-emetic properties but not its psychoactive ones. The recent discovery of the cannabinoid ligand and receptor, and receptor subtypes (see pp7-8) has encouraged researchers to believe that this may be an achievable goal (Iversen, 1993).
A potentially more serious side effect of therapeutic THC is its possible immunosuppressive effect. Any such effect would limit its use as an anti-emetic in the treatment of cancer, since cancer patients experience immune suppression as a side effect of their treatment. There are several reasons why this may be less serious an issue that it seems at first glance. First, there are doubts about the existence of any immunosuppressive effect of cannabinoids (see Section 6.2 on the immune system, pp62-68), and the effect is small in those studies which report one. Second, the clinical significance of any such effects is doubtful in the use of THC in cancer chemotherapy. Such use would be intermittent, and relatively short-term, and the possible gain in increased life expectancy from being able to complete a course of cancer chemotherapy is such that most patients would be prepared to take the risk, in the same way that they chose to undergo the highly toxic chemotherapy in the first place.
8.2.4 Unresolved clinical issues
If THC has a place in the management of nausea from cancer treatment (Poster et al, 1981), and perhaps other causes, a number of clinical issues remain to be resolved (Levitt, 1986). Foremost among these is the best way in which to administer the drug. Should it be given well in advance of treatment at low doses to ensure a stable blood level, or should it be given in larger doses shortly before chemotherapy or radiotherapy? This issue has not been systematically studied (Levitt, 1986).
An additional question is whether there is any clinical benefit to be derived from combining THC with existing anti-emetic agents. There is suggestive evidence that there might be, since the mechanisms of action, while not well understood, appear to be different, raising the possibility that there may be positive synergistic effects from the combination of THC and other anti-emetics. One single-blind study of the combination of dronabinol and prochlorperazine, for example, suggested that the combination of these drugs may have a superior anti-nausea effect to either drug used alone (Plasse et al, 1991). Clearly, more research is warranted on this issue, especially as it may enable cannabinoids to be used as anti-emetics at lower doses with fewer unwanted psychotropic effects.
It seems surprising that the desirability of undertaking research on dosing and combined use of cannabinoids was highlighted by Poster et al in 1981 and by the Institute of Medicine in 1982. Yet very little research has been done, and THC has not been routinely incorporated into the management of nausea caused by cancer chemotherapy. One of the likely reasons has been the American controversy about the rescheduling of marijuana under the Controlled Substances Act, which some argue has discouraged clinical research on cannabinoids (see below). Another reason has been that the motivation for further research on the anti-emetic properties of THC has been removed by the recent development of newer anti-emetic drugs which are superior to prochlorperazine (Iversen, 1993), the "gold standard" drug when the major controlled trials were conducted on cannabis in the 1970s and 1980s. In the absence of trials comparing THC with these newer drugs, its comparative efficacy is unknown, although given its approximate equivalence to prochlorperazine it is likely to be inferior to the newer drugs.