Human/Clinical Studies of Cannabis and Cannabinoids

Delta 8 THC

Source: National Cancer Institute

Among patients included in the studies, cannabinoids were found to be more effective than the conventional antiemetics prochlorperazinemetoclopramide, chlorpromazine, thiethylperazinehaloperidol, domperidone, and alizapride. 

Cannabis Pharmacology

When oral Cannabis is ingested, there is a low (6%–20%) and variable oral bioavailability.[1,2] Peak plasma concentrations of delta-9-tetrahydrocannabinol (THC) occur after 1 to 6 hours and remain elevated with a terminal half-life of 20 to 30 hours. Taken by mouth, delta-9-THC is initially metabolized in the liver to 11-OH-THC, a potent psychoactive metabolite. Inhaled cannabinoids are rapidly absorbed into the bloodstream with a peak concentration in 2 to 10 minutes, declining rapidly for a period of 30 minutes and with less generation of the psychoactive 11-OH metabolite.

Cannabinoids are known to interact with the hepatic cytochrome P450 enzyme system.[3,4] In one study, 24 cancer patients were treated with intravenous irinotecan (600 mg, n = 12) or docetaxel (180 mg, n = 12), followed 3 weeks later by the same drugs concomitant with medicinal Cannabis taken in the form of an herbal tea for 15 consecutive days, starting 12 days before the second treatment.[4] The administration of Cannabis did not significantly influence exposure to and clearance of irinotecan or docetaxel, although the herbal tea route of administration may not reproduce the effects of inhalation or oral ingestion of fat-soluble cannabinoids.

Highly concentrated THC or cannabidiol (CBD) oil extracts are being illegally promoted as potential cancer cures.[5] These oils have not been evaluated in any clinical trials for anticancer activity or safety. Because CBD is a potential inhibitor of certain cytochrome P450 enzymes, highly concentrated CBD oils used concurrently with conventional therapies that are metabolized by these enzymes could potentially increase toxicity or decrease the effectiveness of these therapies.[6,7]

Cancer Risk

A number of studies have yielded conflicting evidence regarding the risks of various cancers associated with Cannabis smoking.

A pooled analysis of three case-cohort studies of men in northwestern Africa (430 cases and 778 controls) showed a significantly increased risk of lung cancer among tobacco smokers who also inhaled Cannabis.[8]

A large, retrospective cohort study of 64,855 men aged 15 to 49 years from the United States found that Cannabis use was not associated with tobacco-related cancers and a number of other common malignancies. However, the study did find that, among nonsmokers of tobacco, ever having used Cannabis was associated with an increased risk of prostate cancer.[9]

A population-based case-control study of 611 lung cancer patients revealed that chronic low Cannabis exposure was not associated with an increased risk of lung cancer or other upper aerodigestive tract cancers and found no positive associations with any cancer type (oral, pharyngeal, laryngeal, lung, or esophageal) when adjusting for several confounders, including cigarette smoking.[10]

A systematic review assessing 19 studies that evaluated premalignant or malignant lung lesions in persons 18 years or older who inhaled Cannabis concluded that observational studies failed to demonstrate statistically significant associations between Cannabis inhalation and lung cancer after adjusting for tobacco use.[11] In the review of the published meta-analyses, the National Academies of Sciences, Engineering, and Medicine (NASEM) report concluded that there was moderate evidence of no statistical association between Cannabis smoking and the incidence of lung cancer.[12]

Epidemiologic studies examining one association of Cannabis use with head and neck squamous cell carcinomas have also been inconsistent in their findings. A pooled analysis of nine case-control studies from the U.S./Latin American International Head and Neck Cancer Epidemiology (INHANCE) Consortium included information from 1,921 oropharyngeal cases, 356 tongue cases, and 7,639 controls. Compared with those who never smoked Cannabis, Cannabis smokers had an elevated risk of oropharyngeal cancers and a reduced risk of tongue cancer. These study results both reflect the inconsistent effects of cannabinoids on cancer incidence noted in previous studies and suggest that more work needs to be done to understand the potential role of human papillomavirus infection.[13] A systematic review and meta-analysis of nine case-control studies involving 13,931 participants also concluded that there was insufficient evidence to support or refute a positive or negative association between Cannabis smoking and the incidence of head and neck cancers.[14]

With a hypothesis that chronic marijuana use produces adverse effects on the human endocrine and reproductive systems, the association between Cannabis use and incidence of testicular germ cell tumors (TGCTs) has been examined.[15-17] Three population-based case-control studies reported an association between Cannabis use and elevated risk of TGCTs, especially nonseminoma or mixed-histology tumors.[15-17] However, the sample sizes in these studies were inadequate to address Cannabis dose by addressing associations with respect to recency, frequency, and duration of use. In a study of 49,343 Swedish men aged 19 to 21 years enrolled in the military between 1969 and 1970, participants were asked once at the time of conscription about their use of Cannabis and were followed up for 42 years.[18] This study found no evidence of a significant relation between “ever” Cannabis use and the development of testicular cancer, but did find that “heavy” Cannabis use (more than 50 times in a lifetime) was associated with a 2.5-fold increased risk. Limitations of the study were that it relied on indirect assessment of Cannabis use; and no information was collected on Cannabis use after the conscription-assessment period or on whether the testicular cancers were seminoma or nonseminoma subtypes. These reports established the need for larger, well-powered, prospective studies, especially studies evaluating the role of endocannabinoid signaling and cannabinoid receptors in TGCTs.

An analysis of 84,170 participants in the California Men’s Health Study was performed to investigate the association between Cannabis use and the incidence of bladder cancer. During 16 years of follow-up, 89 Cannabis users (0.3%) developed bladder cancer compared with 190 (0.4%) of the men who did not report Cannabis use (P < .001). After adjusting for age, race, ethnicity, and body mass index, Cannabis use was associated with a 45% reduction in bladder cancer incidence (hazard ratio, 0.55; 95% confidence interval (CI), 0.33–1.00).[19]

A comprehensive Health Canada monograph on marijuana concluded that while there are many cellular and molecular studies that provide strong evidence that inhaled marijuana is carcinogenic, the epidemiologic evidence of a link between marijuana use and cancer is still inconclusive.[20]



Patterns of Cannabis Use Among Cancer Patients

A cross-sectional survey of cancer patients seen at the Seattle Cancer Care Alliance was conducted over a 6-week period between 2015 and 2016.[21] In Washington State, Cannabis was legalized for medicinal use in 1998 and for recreational use in 2012. Of the 2,737 possible participants, 936 (34%) completed the anonymous questionnaire. Twenty-four percent of patients considered themselves active Cannabis users. Similar numbers of patients inhaled (70%) or used edibles (70%), with dual use (40%) being common. Non–mutually exclusive reasons for Cannabis use were physical symptoms (75%), neuropsychiatric symptoms (63%), recreational use/enjoyment (35%), and treatment of cancer (26%). The physical symptoms most commonly cited were pain, nausea, and loss of appetite. The majority of patients (74%) stated that they would prefer to obtain information about Cannabis from their cancer team, but less than 15% reported receiving information from their cancer physician or nurse.

Data from 2,970 Israeli cancer patients who used government-issued Cannabis were collected over a 6-month period to assess for improvement in baseline symptoms.[22] The most improved symptoms from baseline include the following:

Before treatment initiation, 52.9% of patients reported pain scores in the 8 to 10 range, while only 4.6% of patients reported this intensity at the 6-month assessment time point. It is difficult to assess from the observational data if the improvements were caused by the Cannabis or the cancer treatment.[22] Similarly, a study of a subset of cancer patients in the Minnesota medical Cannabis program explored changes in the severity of eight symptoms (i.e., anxiety, appetite loss, depression, disturbed sleep, fatigue, nausea, pain, and vomiting) experienced by these patients.[23]. Significant symptomatic improvements were noted (38.4%–56.2%) in patients with each symptom. Because of the observational and uncontrolled nature of this study, the findings are not generalizable, but as the authors suggested, may be useful in designing more rigorous research studies in the future.

A retrospective study from Israel of 50 pediatric oncology patients who were prescribed medicinal Cannabis over an 8-year period reported that the most common indications include the following:[24]

  • Nausea and vomiting.
  • Depressed mood.
  • Sleep disturbances.
  • Poor appetite and weight loss.
  • Pain.

Most of the patients (n = 30) received Cannabis in the form of oral oil drops, with some of the older children inhaling vaporized Cannabis or combining inhalation with oral oils. Structured interviews with the parents, and their child when appropriate, revealed that 40 participants (80%) reported a high level of general satisfaction with the use of Cannabis with infrequent short-term side effects.[24]

Cancer Treatment

No ongoing clinical trials of Cannabis as a treatment for cancer in humans were identified in a PubMed search. The only published trial of any cannabinoid in patients with cancer is a small pilot study of intratumoral injection of delta-9-THC in patients with recurrent glioblastoma multiforme, which demonstrated no significant clinical benefit.[25,26] In a trial (NCT02255292) conducted in Israel, oral CBD was investigated as a single salvage agent for recurrent solid tumors. The study was projected to be completed in 2015; however, no results have been published. A small exploratory phase IB study was conducted in the United Kingdom that used nabiximols, a 1:1 ratio of THC:CBD in a Cannabis-based medicinal extract oromucosal spray, in conjunction with dose-dense temozolomide in treating patients with recurrent glioblastoma multiforme.[27][Level of evidence: 1iA] Of the 27 patients enrolled, 6 were part of an open-label group and 21 were part of a randomized group (12 treated with nabiximols and 9 treated with placebo). Progression-free survival at 6 months was seen in 33% of patients in both arms of the study. However, 83.3% of the the patients who received nabiximols were alive at 1 year compared with 44.4% of the patients who recieved placebo (P = .042). The investigators cautioned that this early-phase study was not powered for a survival endpoint. Overall survival rates at 2 years continued to favor the nabiximols arm (50%) compared with the placebo arm (22%) (these rates included results for the 6 patients in the open-label group who received nabiximols).[27]

In a 2016 consecutive case series study, nine patients with varying stages of brain tumors, including six with glioblastoma multiforme, received CBD 200 mg twice daily in addition to surgical excision and chemoradiation.[28][Level of evidence: 3iiiA] The authors reported that all but one of the cohort remained alive at the time of publication. However, the heterogeneity of the brain tumor patients probably contributed to the findings.

Another Israeli group postulated that the anti-inflammatory and immunosuppressive effects of CBD might make it a valuable adjunct in the treatment of acute graft-versus-host disease (GVHD) in patients who have undergone allogeneic hematopoietic stem cell transplantation. The authors investigated CBD 300 mg/d in addition to standard GVHD prophylaxis in 48 adult patients who had undergone transplantation predominantly for acute leukemia or myelodysplastic syndrome (NCT01385124 and NCT01596075).[29] The combination of CBD with standard GVHD prophylaxis was found to be safe. Compared with 101 historical controls treated with standard prophylaxis, patients who received CBD appeared to have a lower incidence of grade II to grade IV GVHD, suggesting that a randomized controlled trial (RCT) is warranted.

Clinical data regarding Cannabis as an anticancer agent in pediatric use is limited to a few case reports.[30,31]

Antiemetic Effect


Despite advances in pharmacologic and nonpharmacologic management, nausea and vomiting (N/V) remain distressing side effects for cancer patients and their families. Dronabinol, a synthetically produced delta-9-THC, was approved in the United States in 1986 as an antiemetic to be used in cancer chemotherapy. Nabilone, a synthetic derivative of delta-9-THC, was first approved in Canada in 1982 and is now also available in the United States.[32] Both dronabinol and nabilone have been approved by the U.S. Food and Drug Administration (FDA) for the treatment of N/V associated with cancer chemotherapy in patients who have failed to respond to conventional antiemetic therapy. Numerous clinical trials and meta-analyses have shown that dronabinol and nabilone are effective in the treatment of N/V induced by chemotherapy.[33-36] The National Comprehensive Cancer Network Guidelines recommend cannabinoids as breakthrough treatment for chemotherapy-related N/V.[37] The American Society for Clinical Oncology (ASCO) antiemetic guidelines updated in 2017 recommends that the FDA-approved cannabinoids, dronabinol or nabilone, be used to treat N/V that is resistant to standard antiemetic therapies.[38]

One systematic review studied 30 randomized comparisons of delta-9-THC preparations with placebo or other antiemetics from which data on efficacy and harm were available.[39] Oral nabilone, oral dronabinol, and intramuscular levonantradol (a synthetic analog of dronabinol) were tested. Inhaled Cannabis trials were not included. Among all 1,366 patients included in the review, cannabinoids were found to be more effective than the conventional antiemetics prochlorperazine, metoclopramide, chlorpromazine, thiethylperazine, haloperidol, domperidone, and alizapride. Cannabinoids, however, were not more effective for patients receiving very low or very high emetogenic chemotherapy. Side effects included a feeling of being high, euphoria, sedation or drowsiness, dizziness, dysphoria or depression, hallucinations, paranoia, and hypotension.[39]

Another analysis of 15 controlled studies compared nabilone with placebo or available antiemetic drugs.[40] Among 600 cancer patients, nabilone was found to be superior to prochlorperazine, domperidone, and alizapride, with nabilone favored for continuous use.

A Cochrane meta-analysis of 23 RCTs reviewed studies conducted between 1975 and 1991 that investigated dronabinol or nabilone, either as monotherapy or as an adjunct to the conventional dopamine antagonists that were the standard antiemetics at that time.[41] The chemotherapy regimens involved drugs with low, moderate, or high emetic potential. The meta-analysis graded the quality of evidence as low for most outcomes. The review concluded that individuals were more likely to report complete absence of N/V when they received cannabinoids compared with placebo, although they were more likely to withdraw from the study because of an adverse event. Individuals reported a higher preference for cannabinoids than placebo or prochlorperazine. There was no difference in the antiemetic effect of cannabinoids when compared with prochlorperazine. The authors concluded that Cannabis-based medications may be useful for treating refractory chemotherapy-induced N/V; however, they cautioned that their assessment may change with the availability of newer antiemetic regimens.

At least 50% of patients who receive moderately emetogenic chemotherapy may experience delayed chemotherapy-induced N/V. Although selective neurokinin 1 antagonists that inhibit substance P have been approved for delayed N/V, a study was conducted before their availability to assess dronabinol, ondansetron, or their combination in preventing delayed-onset chemotherapy-induced N/V.[42] Ondansetron, a serotonin 5-hydroxytryptamine 3 (5-HT3) receptor antagonist, is one of the mainstay agents in the current antiemetic armamentarium. In this trial, the primary objective was to assess the response 2 to 5 days after moderately to severely emetogenic chemotherapy. Sixty-one patients were analyzed for efficacy. The total response—a composite endpoint—including nausea intensity, vomiting/retching, and use of rescue medications, was similar with dronabinol (54%), ondansetron (58%), and combination therapy (47%) when compared with placebo (20%). Nausea absence was greater in the active treatment groups (dronabinol 71%, ondansetron 64%, combination therapy 53%) when compared with placebo (15%; P < .05 vs. placebo for all). Occurrence rates for nausea intensity and vomiting/retching episodes were the lowest in patients treated with dronabinol, suggesting that dronabinol compares favorably with ondansetron in this situation where a substance P inhibitor would currently be the drug of choice.

(Refer to the Cannabis section in the PDQ summary on Nausea and Vomiting Related to Cancer Treatment for more information.)