Patterns of Use of Medical Cannabis Among Israeli

Vol. 49 No. 2 February 2015 Journal of Pain and Symptom Management 223

Original Article Patterns of Use of Medical Cannabis Among Israeli Cancer Patients: A Single Institution Experience Barliz Waissengrin, MD, Damien Urban, MD, Yasmin Leshem, MD, Meital Garty, BA, and Ido Wolf, MD Institute of Oncology (B.W., M.G., I.W.), Tel Aviv Sourasky Medical Center, Tel Aviv; Institute of Oncology (D.U., Y.L.), Chaim Sheba Medical Center, Ramat Gan; and Sackler Faculty of Medicine (I.W.), Tel Aviv University, Tel Aviv, Israel

Abstract Context. The use of the cannabis plant (Cannabis sativa L.) for the palliative treatment of cancer patients has been legalized in multiple jurisdictions including Israel. Yet, not much is currently known regarding the efficacy and patterns of use of cannabis in this setting. Objectives. To analyze the indications for the administration of cannabis among adult Israeli cancer patients and evaluate its efficacy. Methods. Efficacy and patterns of use of cannabis were evaluated using physician-completed application forms, medical files, and a detailed questionnaire in adult cancer patients treated at a single institution. Results. Of approximately 17,000 cancer patients seen, 279 (<1.7%) received a permit for cannabis from an authorized institutional oncologist. The median age of cannabis users was 60 years (range 19e93 years), 160 (57%) were female, and 234 (84%) had metastatic disease. Of 151 (54%) patients alive at six months, 70 (46%) renewed their cannabis permit. Renewal was more common among younger patients and those with metastatic disease. Of 113 patients alive and using cannabis at one month, 69 (61%) responded to the detailed questionnaire. Improvement in pain, general well-being, appetite, and nausea were reported by 70%, 70%, 60%, and 50%, respectively. Side effects were mild and consisted mostly of fatigue and dizziness. Conclusion. Cannabis use is perceived as highly effective by some patients with advanced cancer and its administration can be regulated, even by local authorities. Additional studies are required to evaluate the efficacy of cannabis as part of the palliative treatment of cancer patients. J Pain Symptom Manage 2015;49:223e230. Ó 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

Drs. Waissengrin and Urban contributed equally to Accepted for publication: May 28, 2014. this article. Address correspondence to: Ido Wolf, MD, Institute of Oncology, Tel Aviv Medical Center, Tel Aviv 52621, Israel. E-mail: [email protected]

Ó 2015 American Academy of Hospice and Palliative 0885-3924/$ - see front matter Medicine. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpainsymman.2014.05.018 224 Waissengrin et al. Vol. 49 No. 2 February 2015

Key Words Cannabis, cancer, palliative care, pain, chemotherapy

Introduction for any symptom attributed to the disease it- self or to treatment (e.g., pain, CINV, weight An increasing number of countries, loss, depression). Permits were granted including Israel, have legalized the use of the following a formal request by the treating cannabis plant (Cannabis sativa L.) for the oncologist, who was required to specify the treatment of a wide range of cancer-related clinical condition of the patient and the pre- symptoms.1 The cannabis plant contains cise indications for prescribing cannabis. more than 70 cannabinoids, with D 9-tetrahy- The cannabis was supplied to the patients by drocannabinol (THC), cannabinol, and can- alegaldistributor. nabidiol (CBD) being considered the most We aimed to analyze the indications for the active and clinically relevant.2 These com- administration of cannabis among adult Israeli pounds are primarily found in the dried flow- cancer patients and evaluate the efficacy of ering tops and leaves of the female plant. cannabis both directly, using a detailed question- These dried parts are known as marijuana, naire, and indirectly, by examining the prescrip- and the content of THC in these parts may tion renewal pattern among these patients. exceed 10%.3 The cannabinoids mediate their action through two kinds of G-protein-coupled receptors, CB1 and CB2, found in membranes of nerve cells. The CB1 receptor is found pre- Methods dominantly in the central and peripheral ner- Participants vous systems and suppresses neuronal The Sheba Medical Center is an academic excitability,4 whereas the CB2 receptor is medical center, affiliated with Tel Aviv Univer- found predominantly in immune tissue and sity in Israel, serving as a regional and a tertiary is not related to psychoactive effects.5,6 center for cancer patients. All consecutive Several studies have tested the efficacy of adult patients ($18 years old) who were either synthetic or purified cannabinoids in treated at the Oncology and Gynecology the treatment of cancer-related symptoms. Oncology Departments of Sheba Medical Cen- Thus, synthetic cannabinoids were found to ter and received a permit for medical cannabis reduce chemotherapy-induced nausea and between October 2010 and September 2011 vomiting (CINV) compared with either placebo were included in the study. Symptoms were or older generations of antiemetic medica- managed by the treating oncologist according tions6,7 but did not increase appetite compared to best practice, with a dedicated palliative care with placebo or megestrol.8,9 Two recent Phase team consisting of a trained nurse and physi- III trials noted a modest effect of either THC/ cian being available for consultation. Per na- CBD extract or the synthetic cannabinoid nabi- tional guidelines, all cancer patients receiving lone compared with placebo for the treatment either active oncology treatment or supportive e of intractable cancer-related pain.10 12 To our care, as well as patients receiving adjuvant knowledge, no controlled trials have tested chemotherapy, were eligible for a permit. the efficacy of the cannabis plant in cancer The use of cannabis in the adjuvant setting patients. Despite lack of evidence-based data, was limited, per institutional guidelines, only cannabis use has been legalized in several West- for the treatment of severe side effects ern countries or jurisdictions, including The refractory to conventional treatment. All re- Netherlands13 and the U.S.14 quests for eligible patients by the treating on- In 2010, the Israeli Ministry of Health cologists were approved by the institutional authorized five oncologists to issue permits authorizing oncologist (I. W.). Patients with for cannabis use in oncology patients being hematological malignancies and children treated at their institution. Permits were valid were excluded. The study was approved by for a period of six months and could be issued the local research ethics board. Vol. 49 No. 2 February 2015 Cannabis in Cancer Patients 225

Measures Results Data on demographics, clinical diagnosis, Study Population treatment, and indications for cannabis were During the one year study period, 279 of collected from the permit application forms. approximately 17,000 oncology patients being At the time of the study, permits for cannabis treated or observed at the Sheba Medical Cen- use were valid for six months, and their ter received a permit for medical cannabis. renewal required reapplication by the treating Thus, the estimated percentage of patients oncologists. We used the reapplication forms, who received a permit was 1.7%. The demo- together with survival data obtained from med- graphic and clinical characteristics of all ical charts, to classify patients into three patients who received a permit, as well as those groups: alive and requested permit renewal, who completed the detailed questionnaire, are alive but did not request renewal, and dead. presented in Table 1. The median age of To assess the perceptions of patients patients was 60 years (range 19e93 years); regarding the efficacy of cannabis, a detailed 57% were female; and the most common can- questionnaire was administered at least one cer diagnoses were lung (18%), ovarian month and up to six months after cannabis (12%), breast (10%), colon (9%), and pancre- approval to Hebrew-speaking patients able to atic (7.5%) cancer. The majority of the pa- answer the questionnaire. The questionnaire tients (84%) had metastatic disease. The was completed by the patients during their visit oncology treatment was defined by the refer- to the oncology clinic. We focused only on ring physicians as active palliative in 199 patients who were actively using cannabis. The (71%), supportive care in 36 (13%), and cura- questionnaire was based on the literature, as tive in 17 (6%) patients. In most patients (216; well as previous experience and expert opinion. 77%), cannabis was requested for multiple in- It comprised 39 questions: 15 questions dications. The most common indication for collected demographic and clinical data; 10 which cannabis was prescribed was pain questions about the use of cannabis, including (76%), with anorexia, generalized weakness, the start date, indications, and mode of use; 10 and nausea also being common indications questions about the efficacy of and attitudes to- (56%, 52%, and 41%, respectively). ward cannabis (scored on a five-point Likert scale, with 1 ¼ strongly disagree and 5 ¼ strongly agree); and four questions about Renewal Pattern adverse effects and their severity (scored on a Almost half of the patients (128; 46%) died five-point Likert scale). An improvement in within six months, and 36 (13%) died within symptom control was defined as a 4e5 on the one month of receiving the first cannabis Likert scale, and side effects were categorized permit. Of the remaining 151 patients, 70 as mild (1e3 on the Likert scale) or severe (46%) renewed their permit. The demo- (4e5 on the scale). A pilot study of five medical graphic and disease characteristics of patients personnel and five patients was conducted to who did and did not renew the license are pre- assure proper wording. No external validation sented in Table 2. Renewers were more likely was performed. to be younger (median age 53 vs. 61 years, respectively; P < 0.001), have metastatic dis- ¼ Statistical Analysis ease (P 0.02), and managed with supportive ¼ This was an observational exploratory study care only (12 vs. 7, respectively; P 0.03). of the use of cannabis in a single institution. Demographic and clinical data for patients Survey of Patients who renewed or did not renew the application Of the 279 patients, 113 (41%) were alive and for cannabis use were compared using the chi- using cannabis at least one month following square test for discrete variables and the t-test approval. Sixty-nine (61%) of these patients for continuous variables. All the statistical ana- completed the study questionnaire. Their demo- lyses were done with Statistical Analysis System graphic and clinical characteristics also are pre- software v. 9.1 (SAS Institute, Inc., Cary, NC). sented in Table 1 and are similar to those of the Statistical significance was set at P < 0.05, using whole study group, although the former group chi-square, Fisher, and t-tests. were less likely to use cannabis for weakness 226 Waissengrin et al. Vol. 49 No. 2 February 2015

Table 1 Demographic and Clinical Characteristics of All Patients and Those Completing the Detailed Questionnaire Regarding Cannabis Use Patients Completing Characteristic All Patients, N ¼ 279, n (%) Questionnaire, N ¼ 69, n (%) P-value

Age Median (range) 60 (19e93) 57 (24e81) 0.17 Sex Male 119 (42.6) 35 (50.7) 0.28 Female 160 (57.4) 34 (49.3) Ethnic origin European/American/Russian 63 (22.6) 14 (20.3) 0.89 Africa/Asia 52 (18.6) 14 (20.3) Israel 164 (58.8) 41 (59.4) Cancer diagnosis Lung 51 (18.3) 8 (11.6) 0.78 Ovary 33 (11.8) 8 (11.6) Breast 29 (10.4) 6 (8.7) Colon 25 (9) 6 (8.7) Pancreas 21 (7.5) 6 (8.7) Other 120 (43) 35 (50.7) Indication for cannabis* Pain 211 (75.6) 54 (78.3) 0.04 Appetite 156 (55.9) 33 (47.8) Weakness 146 (52.3) 16 (23.1) Nausea 115 (41.2) 27 (39.1) Disease extent Local 45 (16.1) 17 (24.6) Metastatic 234 (83.9) 52 (75.4) Intent Cure 17 (6.1) 6 (8.7) 0.08 Active palliative 199 (71.3) 46 (66.7) Supportive care 36 (12.9) 15 (21.7) Unknown 27 (9.7) 2 (2.9) Principal physician responsible for symptom control Oncologist NA 44 (63.8) Family doctor 6 (8.7) Pain clinic 9 (13) Other 10 (14.5) Method of administration Smoking NA 63 (91.3) Other 6 (8.7) Who recommended you to use cannabis Oncologist NA 34 (42.0) Media 18 (22.2) Other patient 12 (14.8) Family member 7 (8.6) Pain clinic 6 (7.4) Family doctor 3 (3.7) Other 1 (1.2) NA ¼ not available. *Cannabis could be requested for multiple indications.

(P ¼ 0.04). More than 90% of patients used The efﬁcacy of cannabis use in patients, as cannabis by smoking. Most patients (44; 63.8%) perceived by those who completed the question- reported that their symptoms were principally naire, is presented in Figure 1. Improvement in managed by their treating oncologist. Yet, pain control was reported by 70% of the patients, whereas 42% of the responders stated that general well-being by 70%, improved appetite by cannabis was recommended to them by their 60%, reduced nausea and vomiting by 50%, and oncologist, more than 45% stated that they reduced anxiety by 44%. Importantly, 83% were recommended cannabis by nonmedical graded the overall efﬁcacy of cannabis as high. sources including the media (18; 22%), other Among responders to the questionnaire, 21 patients (12; 14.8%), and family members (30%) died within six months. Of the remaining (7; 8.6%). 48 patients, 32 (67%) renewed their permits. Vol. 49 No. 2 February 2015 Cannabis in Cancer Patients 227

Table 2 Demographic and Clinical Characteristics by Application to Renew Cannabis Permit at Six Months Applied for Renewal, Did Not Apply for Renewal, Characteristic N ¼ 70, n (%) N ¼ 81, n (%) P-value

Age Median (range) 52.5 (23e81) 61 (19e93) <0.001 Sex Male 32 (46) 33 (41) 0.62 Female 38 (54) 48 (59) Ethnic origin European/American/Russian 15 (31.4) 22 (27) 0.17 Africa/Asia 7 (10) 14 (17) Israel 48 (68.6) 45 (56) Cancer diagnosis Lung 7 (10) 17 (21) 0.08 Ovary 11 (15.7) 15 (18.5) Breast 8 (11.4) 6 (7.4) colon 4 (5.7) 7 (8.6) Pancreas 1 (1.4) 6 (7.4) Other 39 (55.7) 30 (37) Indication for cannabis Pain 54 (69) 56 (77) 0.83 Appetite 40 (57) 42 (52) Weakness 37 (53) 49 (61) Nausea 25 (36) 35 (43) Anxiety 26 (37) 32 (40) Disease extent Metastatic 51 (72.9) 72 (88.9) 0.02 Local 19 (27.1) 9 (11.1) Intent Cure 6 (8.6) 6 (7.4) 0.33 Active palliative 42 (60) 59 (72.8) Best supporting care 12 (17.1) 7 (8.6) Unknown 10 (1.4) 9 (11.1)

The majority of the responders (43; 62%) For example, the current prevalence of medi- reported no side effects attributable to cal cannabis use in The Netherlands is eight cannabis. The most common side effects per 100,000 inhabitants.13 Recently, a lively reported were fatigue (14; 20.3%) and dizzi- debate regarding the medicinal use of ness (13; 18.8%). A minority of patients cannabis was launched by the New England Jour- described other side effects including delu- nal of Medicine, followed by a poll among sions (4; 5.8%) and mood change (3; readers.15,16 Surprisingly, despite the lack of 4.35%). One patient reported requiring hospi- strong scientific evidence, 76% of 1446 voters talization because of dyspnea attributed to supported the use of cannabis. Obviously, cannabis. because of selection bias, these data should be interpreted with caution. The implementation of a new system in Discussion Israel for issuing permits for cannabis use Although several Phase III studies evaluated enabled us to collect data regarding prescrip- the efficacy of synthetic cannabinoids or puri- tion patterns, renewal requests, and efficacy fied extracts of THC/CBD in the treatment of cannabis in cancer patients. During the of cancer-associated symptoms, to our knowl- study period, 279 cancer patients treated at edge, no prospective clinical trials reporting the Sheba Medical Center received a permit. the efficacy of natural cannabis in cancer pa- This number represents a small fraction e tients have been published.5 8 Accordingly, (approximately 1.7%) of all patients. The vast its use is not approved by regulatory agents majority of these patients suffered from and is not recommended by formal guidelines. advanced disease, almost half died within six Despite the lack of evidence or formal recom- months and 13% died within one month of mendations, the use of cannabis is increasing issuing the cannabis permit. The use of in Israel and other Western countries.1,5,13,14 cannabis in the adjuvant setting was limited, 228 Waissengrin et al. Vol. 49 No. 2 February 2015

Fig. 1. Efficacy of cannabis use in patients, as perceived by patients who completed a detailed questionnaire (n ¼ 69). per institutional guidelines, only for the treat- was indicated for more than one symptom. ment of severe side effects refractory to con- This unique prescription pattern differs from ventional treatment. Thus, the use of medical that of most approved medications, which are cannabis remained relatively restricted. This often registered and prescribed for very spe- may be attributed to the reluctance of the on- cific indications, based on data obtained cologists or the patients to use an illicit drug or from clinical trials. Interestingly, the reasons may be related to the lack of evidence for cannabis use in cancer patients in our study regarding the efficacy of the treatment. differ from the general medical marijuana The uncertainty regarding the role of population in the U.S., where the most com- cannabis is reflected by the multiple indica- mon indications among more than 286,000 tions for its use. Treatment was recommended registered users were severe or chronic pain by the treating oncologists for a wide array of (92%), muscle spasms (21%), and nausea symptoms, and in most patients, the treatment and vomiting (12%).17 Vol. 49 No. 2 February 2015 Cannabis in Cancer Patients 229

As permits are only valid in Israel for six interest in the use of medical cannabis in Israel months, we examined renewal patterns and surrounding its legalization. It is plausible that noted that less than half of the patients alive the exposure to a large amount of nonmedical at six months applied for a permit renewal. information regarding the beneficial use of These data are consistent with a recent survey cannabis encouraged patients to request of Israeli patients, in which only 50% of those cannabis for a wide range of symptoms. who received permits were actually using The reported side effects in this study were cannabis six to eight weeks after commencing similar to those described in the literature treatment.18 Although some patients (i.e., for this population of patients.21,22 Yet, the those receiving adjuvant treatment) may not magnitude of side effects of cannabis could need renewal, our data suggest that a signifi- not be accurately assessed in this study. cant percentage of patients and/or physicians Taken together, our data indicate that nearly do not perceive benefit from cannabis. As ex- half of all cancer patients who start treatment pected, patients with advanced disease and with cannabis continue using it for prolonged those receiving supportive care only were periods of time and some of the patients more likely to ask for renewal. This may reflect describe the treatment as effective for a wide difficulties in controlling symptoms in these pa- range of symptoms, including general well- tients using currently available measures. The being. Although these data cannot endorse long-term efficacy of cannabis in cancer pa- the use of cannabis for specific symptoms, tients should be addressed in a controlled trial. they support the view that its use may be justi- The vast majority of the participants who fied as part of palliative treatment in selected completed the detailed questionnaire stated cancer patients. Overuse or abuse could not that cannabis use was associated with an be evaluated, but our experience suggests improvement in all aspects that were surveyed, that administration of cannabis to cancer and 83% of them graded the overall efficacy of patients can be regulated by local medical cannabis as high. These efficacy data are strik- authorities. Further prospective studies are ing compared with reported Phase III tri- required to confirm the role of cannabis in als.6,8,19 The demographic and clinical the palliative care of cancer patients. characteristics of these patients are similar to those of the entire group (Table 1). As only patients actively using cannabis at one month Disclosures and Acknowledgments were asked to complete the questionnaire, This study was supported by the Israeli the results may be affected by a selection Cancer Association. The authors have no bias. It is also possible that those who disclosures. benefited from cannabis were more likely to respond to the questionnaire. Yet, it is still clear that a significant portion of the patients References perceive the treatment as highly effective. It is possible that the natural plant is more effec- 1. Shelef A, Mashiah M, Schumacher I, Shine O, Baruch Y. Medical grade cannabis (MGC): regula- tive than either the synthetic cannabinoids or tion mechanisms, the present situation around the the pure extracts that were tested in previous world and in Israel. Harefuah 2011;150:913e917. trials.20 Alternatively, the major beneficial ef- 935, 934. fect of cannabis may be its psychoactivity. 2. Turcotte D, Le Dorze JA, Esfahani F, et al. Exam- Indeed, a meta-analysis evaluating the efficacy ining the roles of cannabinoids in pain and other of cannabinoids on CINV noted a clear prefer- therapeutic indications: a review. Expert Opin Phar- e ence for treatment with cannabinoids, despite macother 2010;11:17 31. the marginal objective effect on CINV.6 3. Mehmedic Z, Chandra S, Slade D, et al. Potency D Although the request forms were completed trends of 9-THC and other cannabinoids in confis- cated cannabis preparations from 1993 to 2008. by the treating oncologist, almost half of the pa- J Forensic Sci 2010;55:1209e1217. tients stated that they requested cannabis 4. Rahn EJ, Hohmann AG. Cannabinoids as phar- following recommendations made by other pa- macotherapies for neuropathic pain: from the tients, family members, and even the media. bench to the bedside. Neurotherapeutics 2009;6: This may have resulted from widespread public 713e737. 230 Waissengrin et al. Vol. 49 No. 2 February 2015

5. Bowles DW, O’Bryant CL, Camidge DR, 13. Hazekamp A, Heerdink ER. The prevalence Jimeno A. The intersection between cannabis and and incidence of medicinal cannabis on prescrip- cancer in the United States. Crit Rev Oncol Hema- tion in The Netherlands. Eur J Clin Pharmacol tol 2012;83:1e10. 2013;69:1575e1580. 6. Machado Rocha FC, Stefano SC, De Cassia 14. Room R. Legalizing a market for cannabis for Haiek R, Rosa Oliveira LM, Da Silveira DX. Thera- pleasure: Colorado, Washington, Uruguay and peutic use of Cannabis sativa on chemotherapy- beyond. Addiction 2014;109:345e351. induced nausea and vomiting among cancer 15. Bostwick JM, Reisﬁeld GM, DuPont RL. Clinical patients: systematic review and meta-analysis. Eur J e decisions. Medicinal use of marijuana. N Engl J Med Cancer Care 2008;17:431 443. 2013;368:866e868. 7. Zutt M, H€anssle H, Emmert S, Neumann C, 16. Adler JN, Colbert JA. Clinical decisions. Medic- Kretschmer L. Dronabinol for supportive therapy inal use of marijuanadpolling results. N Engl J Med in patients with malignant melanoma and liver me- 2013;368:e30. tastases. Hautarzt 2006;57:423e427. 17. Bowles DW. Persons registered for medical 8. Jatoi A, Windschitl HE, Loprinzi CL, et al. Dro- marijuana in the United States. J Palliat Med 2012; nabinol versus megestrol acetate versus combina- 15:9e11. tion therapy for cancer-associated anorexia: a North Central Cancer Treatment Group study. 18. Bar-Sela G, Vorobeichik M, Drawsheh S, et al. J Clin Oncol 2002;20:567e573. The medical necessity for medicinal cannabis: prospective, observational study evaluating the treat- 9. Strasser F, Luftner D, Possinger K, et al. Com- ment in cancer patients on supportive or palliative parison of orally administered cannabis extract care. Evid Based Complement Alternat Med 2013; and delta-9-tetrahydrocannabinol in treating pa- 2013:510392. tients with cancer-related anorexia-cachexia syn- drome: a multicenter, phase III, randomized, 19. Uritsky TJ, McPherson ML, Pradel F. Assess- double-blind, placebo-controlled clinical trial from ment of hospice health professionals’ knowledge, the Cannabis-In-Cachexia Study Group. J Clin On- views, and experience with medical marijuana. col 2006;24:3394e3400. J Palliat Med 2011;14:1291e1295. 10. Campbell FA, Tramer MR, Carroll D, et al. Are 20. Hall W, Solowij N. Adverse effects of cannabis. cannabinoids an effective and safe treatment option Lancet 1998;352:1611e1616. in the management of pain? A qualitative systematic 21. Johnson JR, Burnell-Nugent M, Lossignol D, review. BMJ 2001;323:13. et al. Multicenter, double-blind, randomized, 11. Frank B, Serpell MG, Hughes J, Matthews JN, placebo-controlled, parallel-group study of the efﬁ- Kapur D. Comparison of analgesic effects and pa- cacy, safety, and tolerability of THC: CBD extract tient tolerability of nabilone and dihydrocodeine and THC extract in patients with intractable for chronic neuropathic pain: randomised, cross- cancer-related pain. J Pain Symptom Manage 2010; over, double blind study. BMJ 2008;336:199. 39:167e179. 12. Maida V, Ennis M, Irani S, Corbo M, 22. Brisbois TD, de Kock IH, Watanabe SM, et al. Dolzhykov M. Adjunctive nabilone in cancer pain Delta-9-tetrahydrocannabinol may palliate altered and symptom management: a prospective observa- chemosensory perception in cancer patients: results tional study using propensity scoring. J Support On- of a randomized, double-blind, placebo-controlled col 2008;6:119e124. pilot trial. Ann Oncol 2011;22:2086e2093.