Jacob Bell
New Member
Hypothesizing that marijuana smokers are at a significantly lower risk of carcinogenicity relative to tobacco-non-marijuana smokers: evidenced based on statistical reevaluation of current literature
The decision to use marijuana for medical purposes should be accurately informed by scientific findings of benefit versus risk. The effects of marijuana smoking on cancer remain hypothetical but warrant attention. Without advocating or opposing the use of marijuana for any purpose, this study reconsiders the carcinogenicity of marijuana smoke.
THE CONTROVERSIAL ISSUE
A hypothetical link between marijuana smoking and cancer has been established based on the following observations: (1) marijuana smoke contains carcinogenic hydrocarbons; (2) cannabinoid administration promotes cancer under certain laboratory conditions; (3) lesions similar to those caused by tobacco smoke are found in the bronchial epithelium of marijuana smokers; and (4) marijuana tar produces tumors when painted on the skin of animals. The best evidence to date on the link between marijuana and cancer, however, derives from epidemiologic case-control studies, especially those with a large number of cases and/or randomly selected controls. Such studies tend to suggest, if anything, an inverse association between marijuana use and cancers. To test the hypothesis that marijuana smoking significantly lowers the risk of developing cancer in humans, we analyzed published data from a prospective cohort study on cancer incidence among nonsmokers (NS), marijuana-only smokers (MS), tobacco-only smokers (TS), and marijuana and tobacco smokers (MTS).
Ads by GoogleThe Effects of Smoking
Get your questions answered by the experts. Free online resource.
Medical Information Websites and Medical Web Design for Doctors - HealthCommunities.com - HealthCommunities
Electric Vapor Free Offer
Sign Up Today For Your Free Trial. #1 Rated Product. Limited Supplies.
PrLog.org/News
The following brief review of the literature on some key issues concerning marijuana smoking and cancer will provide the reader with a conceptual framework to allow for intelligent comprehension of the controversial nature of this important medical issue.
Marijuana Smoke Contains Higher Levels of Carcinogenic Polycyclic Aromatic Hydrocarbons (PAH) Than Those Found in Tobacco Smoke
As a rule, the level of carcinogenic PAHs within a chemical mixture is less important than the overarching influence of the mixture on carcinogenic PAH activation (Mahadevan et al. 2005). Indeed, marijuana smoke is presumed carcinogenic not only because it contains carcinogenic PAHs, but also because cannabinoids, in their own right, influence the cytochrome enzymes (CYP1) that determine carcinogenic PAH activation (Roth et al. 2001). Perhaps due to their phenolic hydrocarbon structure, cannabinoids share the anticarcinogenic disposition of polyphenols to increase levels of CYP1A1 messenger RNA while competitively reducing CYP1A1 enzyme activity through aryl hydrocarbon receptor ligation (Ciolino, Wang & Yeh 1998). Not surprisingly, spiking tobacco tar with delta-9 THC markedly reduced carcinogenic activity (Roth et al. 2001).
Cannabinoid Administration Promotes Cancer in Mice
Intraperitoneal administration of immunosuppressive cannabinoids (> 5 mg/kg) promotes cannabinoid-resistant lines of cancer in immune-responsive strains of mice (e.g., BALB/cJ) (McKallip, Nagarkatti & Nagarkatti 2005; Gardner et al. 2003; Zhu et al. 2000). Most in vivo studies, however, have shown that cannabinoid administration, either locally or systemically over a wide range of doses, inhibits the growth of cancer (Duntsch et al. 2006; Grimaldi et al. 2006; Kogan et al. 2006, 2004; Ligresti et al. 2006; McKallip et al. 2006, 2002; Bifulco et al. 2004, 2001; Blazquez et al. 2004, 2003; Massi et al. 2004; Casanova et al. 2003; ; Portella et al. 2003; Recht et al. 2001; Sanchez et al. 2001; Chan et al. 1996; Munson et al. 1975). The neoplastic effects of cannabinoids vary according to strain/species, cell type, dosage, timing, route of administration, serum proteins, distribution of receptors (e.g., cannabinoid and vanilloid), enzymatic degradation, and degree of malignancy. Cannabinoids may control the growth of human cancers, in part, through the cannabinoid receptor 2 gene (CNR2), which emerged as the best-connected "super hub" in an inferred large-scale association network for breast cancer data (Schafer & Strimmer 2005).
Marijuana Smokers Develop Precancerous Lesions Similar to Those Caused by Tobacco Smoke
Marijuana smoking causes precancerous epithelial lesions (PEL) such as squamous metaplasia (SM) and other changes associated with respiratory exposure to smoke in general (Barsky et al. 1998). On the one hand, SM not only follows exposure to potent carcinogens in laboratory strains of mice, but it also precedes the development of squamous cell carcinoma of lungs (SCCL) in humans. On the other hand, the multiplicity of SM was higher among SCCL-resistant mice (e. g., C57BL/6J = 5. 0 - 6. 0) than it was among SCCL-susceptible mice (e.g., NIH Swiss = 2. 1 - 4. 9; Barsky et al. 1998); in humans, PEL such as SM are generally reversible and often regress spontaneously (Winterhaldler et al. 2004).
As it turns out, PEL may have little, if any, predictive value. In fact, according to recent findings, "Distribution and outcome of preneoplastic lesions have been found to be unrelated to various risk factors such as smoking ... The initial WHO classification of any preneoplastic lesion cannot be reliably used for accurate risk assessment of field carcinogenesis" (Breuer et al. 2005).
Marijuana Tar Produces Tumors When Painted on the Skin of Animals
Tar from marijuana smoke, like that from tobacco smoke, was shown to produce benign tumors (i.e., squamous-cell papilloma) when painted on the skin of animals (Cottrell, Sohn & Vogel 1973); however, tar from tobacco smoke caused frank malignancies (i.e., squamous-cell carcinoma), whereas tar from marijuana smoke did not. In subsequent research on monkeys, prolonged exposure to marijuana smoke failed to produce any carcinogenic effects (Talaska et al. 1992). Interestingly enough, exposure to marijuana smoke was shown to retard the growth of sarcoma in rats (Watson 1989). This inhibition was unrelated to the cannabinoid content of the smoke.
The appearance of papillomas on the skin indicates that tar from marijuana smoke, like that from tobacco smoke, is an effective initiator of benign tumors; however, the absence of squamous-cell carcinoma is consistent with the observation that cannabinoid administration induces the regression of squamous-cell carcinoma of skin by transforming the vascular hyperplasia of engorged tumors into pallid networks of small, differentiated capillaries (Casanova et al. 2003). Such transformation reflects the tendency of cannabinoids to thwart the neoplastic expression of angiogenic factors (Casanova et al. 2003; Portella et al. 2003).
While none of the animal studies has demonstrated the carcinogenicity of marijuana smoke, future studies should be rigorously conducted (Huff & Chan 2005). Most in vivo studies on the neoplastic properties of cannabinoids have, unfortunately, been conducted with BALB/cJ and C57BL/6J murine strains, which are not susceptible to chemically-induced SCCL. Even the most susceptible strains do not develop SCCL as the result of exposure to tobacco smoke, the definitive benchmark for human carcinogens (Wang et al. 2004).
In evaluating the carcinogenicity of any type of smoke, it might help to remember that it was epidemiology, rather than animal research, that first incriminated tobacco smoke as a carcinogen (Proctor 2004). There have been no convincing findings of an etiologic association between marijuana smoking and tobacco-related cancer (TRC): Small hospital-based case-control studies have inconsistently found an association (Chacko et al. 2006; Llewellyn et al. 2004; Sasco et al. 2002; Zhang et al. 1999; Hsairi et al. 1993), whereas population-based case-control studies, especially those with a large number of cases and/or randomly selected controls, tend to suggest, if anything, an inverse association between marijuana use and cancers (Tashkin et al. 2006; Sarafian et al. 2006; Rosenblatt et al. 2004; Zhu et al. 2002; Ford et al. 2001).
STATISTICAL METHODS
To test the hypothesis that marijuana smoking significantly lowers the risk of developing cancer in humans, we analyzed published data from a prospective cohort study on cancer incidence among nonsmokers (NS), marijuana-only smokers (MS), tobacco-only smokers (TS), and marijuana and tobacco smokers (MTS) (Sidney et al. 1997). This study represents the most complete systematic evaluation of the issue of MS and cancer risks. The log linear model was used to calculate the probability of developing each cancer form as a function of the interaction effect of marijuana and cigarette smoking, as well as functions of marijuana and tobacco smoking main effects. Log linear models are a generalization of ordinary r by c chi-square analyses and appropriate for count data when more variables than rows (r) and columns (c) are being studied as is the case with these data. Chi square statistics were calculated for the interaction and main effect estimates. The outcomes of the analyses for males and females are shown in Table 1. For each form of cancer, the probability of developing the cancer form as a function of the interaction between marijuana and tobacco smoking as well as functions of main marijuana and tobacco smoking effects are reported (see Stevens 2002 for information on the method of analysis). We used the technique of collapsing across an effect to average over levels of the effect (e.g., for tobacco smoking, to take the average of tobacco smokers and nontobacco smokers and average them together for one mean).
OUTCOMES FOR MALES
All Sites
The interaction and main effects for all cancer sites in males appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 10.35, p = 0.001. For all sites, males had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.36) than if they smoked marijuana and not tobacco (r = 0.09) For all sites, males had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.09) than if they smoked neither marijuana nor tobacco (r = 0.29). In addition, for all sites, males had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.36) than if they smoked neither marijuana nor tobacco (r = 0.29).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing cancer (r = 0.61) than did males who did not smoke tobacco (r = 0.38). This difference was statistically significant [chi square] = 44, p = 0.02. The correlation coefficient was -.23, Z = -4.53, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. For all sites collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing cancer (r = 0.35) than did males who did not smoke marijuana (r = 0.65) This difference was statistically significant: [chi square] = 31, p < .00001. The correlation coefficient was .31, Z = 5.81, p < .05.
Tobacco Related Cancer
The interaction and main effects for TRC in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 1.05, p = .31.
Males had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.03) than if they smoked marijuana and not tobacco (r = 0.56) although this interaction effect was not statistically significant. For TRC, males had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.03) than if they smoked neither marijuana nor tobacco (r = 0.12). In addition, males had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.56) than if they smoked neither marijuana nor tobacco (r = 0.12) although, as already noted, this interaction effect was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing TRC (r = 0.84) than did males who did not smoke tobacco (r = 0.75). This difference proved to be statistically significant: [chi square] = 14.73, p = .001, with a correlation coefficient of -.86, Z = -6.14, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing TRC (r = 0.31) than did males who did not smoke marijuana (r = 0.68). This difference was statistically significant: [chi square] = 48.63, p = 0.03, with a correlation coefficient of .39, Z = 3.63, p < .05.
Colorectal Cancer
The interaction and main effects for colorectal cancer in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 2.08, p = 0.15. Males had a greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.33) than if they smoked marijuana and not tobacco (r = 0.08) although, as noted above, the interaction was statistically non-significant. Males also had a lower probability of developing colorectal cancer if they smoked Marijuana and not tobacco (r = 0.08) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as noted above, the interaction was statistically non-significant. Males had no greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.33) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as noted above, the interaction was statistically nonsignificant.
Main effects of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing colorectal cancer (r = 0.56) than did males who did not smoke tobacco (r = 0.44). This difference proved to be statistically significant: [chi square] = 8.20, p = .02, with a correlation coefficient of -0.11, Z = -0.74, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing colorectal cancer (r = 0.31) than did males who did not smoke marijuana (r = 0.69), [chi square] = 2.60, p = 0.27. The correlation coefficient was 0.39, Z = 2.47, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 0.85, p = 0.36. Males had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.67) than if they smoked marijuana and not tobacco (r = 0) although, as noted above, the interaction was statistically nonsignificant. Males did not display a difference in the probabilities of developing lung cancer if they smoked marijuana and not tobacco (r = 0) and neither marijuana nor tobacco (r = 0.04). They had no greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.67) than if they smoked neither marijuana nor tobacco (r = 0.04) although, as noted above, the interaction was statistically nonsignificant.
Main effects of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing lung cancer (r = 0.95) than did males who did not smoke tobacco (r = .02). This difference proved to be statistically significant: [chi square] = 9.04, p = .01, with a correlation coefficient of 1.56, Z = 4.31, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing lung cancer (r = 0.15) than did males who did not smoke marijuana (r = 0.37) This difference proved to be statistically significant: [chi square] = 40.47, p < 0.00001 with a correlation coefficient of 0.44, Z = 2.79, p < .05.
Melanoma
The interaction and main effects for melanoma in males appear in Table 1.
Interaction Effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = .86, p = 0.35. Males had a greater probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.43) than if they smoked marijuana and not tobacco (r = 0.08), and displayed a difference between the probabilities of developing melanoma if they smoked marijuana and not tobacco (r = 0.08) and smoking neither marijuana nor tobacco (r = 0.31) although, as noted above, the interaction was statistically nonsignificant. Males displayed a greater probability of developing melanoma if they smoked tobacco but not marijuana (r = 0.67) than if they smoked neither marijuana nor tobacco (r = 0.31), although, as noted above, the interaction was statistically nonsignificant.
Main effect of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing melanoma (r = 0.61) than did males who did not smoke tobacco (r = 0.32). This difference proved to be statistically significant: [chi square] = 17, p < 0.00001 with a correlation coefficient of -0.23, Z = -1.95, p < .05.
Main effect of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing melanoma (r = 0.27) than did males who did not smoke marijuana (r = 0.73) although this difference proved to be only marginally statistically significant: [chi square] = 4.67, p = .09, with a correlation coefficient of 0.51, Z = 3.87, p < .05.
Prostate Cancer
The interaction and main effects for prostate cancer in males appear in Table 1.
Interaction Effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 0.62, p = 0.43. Males had a greater probability of developing prostate cancer if they smoked tobacco and not marijuana (r = 0.50) than if they smoked marijuana and not tobacco (r = 0.12). They also displayed a difference between the probabilities of developing prostate cancer if they smoked marijuana and not tobacco (r = 0.12) and neither marijuana nor tobacco (r = 0.21); and between the probabilities in developing this form of cancer if they smoked tobacco and not marijuana (r = 0.50) and neither tobacco nor marijuana (r = 0.21) although, as noted above, these interaction effects were statistically nonsignificant.
Main effect of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing prostate cancer (r = .67) than did males who did not smoke tobacco (r = .33). This difference proved to be statistically significant: [chi square] = 8.90, p = .01, with a correlation coefficient of -.34, Z = 2.12, p < .05.
Main effect of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing prostate cancer (r = 0.29) than did males who did not smoke marijuana (r = 0.71), [chi square] = 5.78, p = .05, with a correlation coefficient of 0.46, Z = 2.68, p < .05.
OUTCOMES FOR FEMALES
All Sites
The interaction and main effects for all cancer sites in females appear in Table 1.
Interaction Effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 34.60, p < 0.00001 For all sites, females had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.34) than if they smoked marijuana and not tobacco (r = 0.08). Also, for all sites, females had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.08) than if they smoked neither marijuana nor tobacco (r = 0.39). In addition, for all sites, females had no substantially greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.34) than if they smoked neither marijuana nor tobacco (r = 0.39).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing cancer (r = 0.51) than did females who did not smoke tobacco (r = 0.48). This difference was statistically significant: [chi square] = 248, p < 0.00001 with a correlation coefficient of -0.04, Z = -1.25, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for all sites collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing cancer (r = 0.27) than did females who did not smoke marijuana (r = 0.73) This difference was statistically significant: [chi square] = 36.09, p < 0.00001 with a correlation coefficient of .50, Z = 14.27, p < .05.
Tobacco Related Cancer
The interaction and main effects for TRC in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 3.14, p < 0.00001. Females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.66) than if they smoked marijuana and not tobacco (r = 0). Also, females had a lower probability of developing TRC if they smoked marijuana and not tobacco (r = 0) than if they smoked neither marijuana nor tobacco (r = 0.22). In addition, females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.66) than if they smoked neither marijuana nor tobacco (r = 0.22).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing TRC (r = 0.78) than did females who did not smoke tobacco (r = 0.22). This difference was statistically marginally significant: [chi square] = 3.14, p = 0.08, with a correlation coefficient of -0.64, Z = -4.82, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing TRC (r = .12) than did females who did not smoke marijuana (r = .88). This difference was statistically significant: [chi square] = 28, p = 0, with a correlation coefficient of .99, Z = 5.90, p < .05.
Colorectal Cancer
The interaction and main effects for colorectal cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant, or only marginally so: [chi square] = 3.03, p = 0.08. Females had a greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.45) than if they smoked marijuana and not tobacco (r = 0.01) although, as already noted, the difference was only marginally statistically significant. Also, females had a lower probability of developing colorectal cancer if they smoked marijuana and not tobacco (r = 0.01) than if they smoked neither marijuana nor tobacco (r = 0.44). Also as already noted, the difference was only marginally statistically significant. In addition, females had no greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.45) than if they smoked neither marijuana nor tobacco (r = 0.44).
Main effects of tobacco smoking collapsing across marijuana smoking. For colorectal cancer collapsing across marijuana smoking, females who smoked tobacco had no higher probability of developing cancer (r = 0.55) than did females who did not smoke tobacco (r = 0.45). The difference did not prove to be statistically marginally significant: [chi square] = 1.08, p = 0.29, with a correlation coefficient of -0.10, Z = -0.76, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing colorectal cancer (r = 0.11) than did females who did not smoke marijuana (r = 0.89), [chi square] = 3.58, p = 0.17, with a correlation coefficient of 1.0, Z = 5.14, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant: [chi square] = 1.08, p = 0.30. Females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.73) than if they smoked marijuana and not tobacco (r = 0) although, as already noted, the difference was not statistically significant. Also, females had a slightly lower probability of developing lung cancer if they smoked marijuana and not tobacco (r = 0) than if they smoked neither marijuana nor tobacco (r = 0.10) although, also as already noted, the difference was not statistically significant. Females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.73) than if they smoked neither marijuana nor tobacco (r = 0.44) although, as already noted, the difference was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a higher probability of developing lung cancer (r = 0.90) than did females who did not smoke tobacco (r = 0.10). The difference proved to be statistically significant: [chi square] = 38.38, p < 0.0001, with a correlation coefficient of -1.0, Z = -4.60, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, as shown in Table 1, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing lung cancer (r = 0.16) than did females who did not smoke marijuana (r = 0.84). This difference was statistically significant: [chi square] = 31.43, p < 0.00001 with a correlation coefficient of 0.82, Z = 4.22, p < .05.
Melanoma
The interaction and main effects for melanoma in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant, or only marginally so: [chi square] = 3.14, p = 0.08. Females had a slightly greater probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.28) than if they smoked marijuana and not tobacco (r = 0.12) although, as already noted, the interaction effect was not statistically significant. Also, females had a lower probability of developing melanoma if they smoked marijuana and not tobacco (r = 0.12) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as already noted, the interaction effect was not statistically significant. Females had a slightly lower probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.28) than if they smoked neither marijuana nor tobacco (r = 0.35) although, as already noted, the interaction effect was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. For melanoma collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing cancer melanoma (r = 0.52) than did females who did not smoke tobacco (r = 0.48). The difference proved to be statistically significant: [chi square] = 7.37, p = 0.03, with a correlation coefficient of -.05, Z = -.37, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for melanoma collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing melanoma (r = 0.37) than did females who did not smoke marijuana (r = 0.63), [chi square] = 3.27, p = 0.019, with a correlation coefficient of .27, Z = 2.68, p < .05.
Breast Cancer
The interaction and main effects for breast cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 23.56, p < 0. 00001. Females had a slightly greater probability of developing breast cancer if they smoked tobacco and not marijuana (r = 0.31) than if they smoked marijuana and not tobacco (r = 0.06). Also, females had a lower probability of developing breast cancer if they smoked marijuana and not tobacco (r = 0.06) than if they smoked neither marijuana nor tobacco (r = 0.48). They had a slightly lower probability of developing breast cancer if they smoked tobacco and not marijuana (r = 0.31) than if they smoked neither marijuana nor tobacco (r = 0.48).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a lower probability of developing breast cancer (r = 0.48) than did females who did not smoke tobacco (r = 0.55). The difference proved to be statistically significant: [chi square] = 135, p = 0.03, with a correlation coefficient of .07, Z = 1.37, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing breast cancer (r = 0.22) than did females who did not smoke marijuana (r = 0.81). This difference proved to be statistically significant: [chi square] = 25, p < 0.00001 with a correlation coefficient of .66, Z = 10.21, p < .05.
Cancer of the Cervix
The interaction and main effects for cervical cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 8.17, p = 0.004. Females had a greater probability of developing cancer of the cervix if they smoked tobacco and not marijuana (r = 0.27) than if they smoked marijuana and not tobacco (r = 0.16). Also, females had a lower probability of developing cancer of the cervix if they smoked marijuana and not tobacco (r = 0.16) than if they smoked neither marijuana nor tobacco (r = 0.30). They had no substantially different probability of developing cancer of the cervix if they smoked tobacco and not marijuana (r = 0.27) than if they smoked neither marijuana nor tobacco (r = 0.30).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a higher probability of developing cancer of the cervix (r = 0.54) than did females who did not smoke tobacco (r = 0.46). The difference proved to be statistically significant: [chi square] = 13.85, p = 0.001, with a correlation coefficient of -0.07, Z = -1.27, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing cancer of the cervix (r = 0.43) than did females who did not smoke marijuana (r = 0.57). This difference proved to be statistically significant: [chi square] = 9.72, p = 0.008, with a correlation coefficient of 0.14, Z = 2.41, p < .05.
MALES AND FEMALES COMBINED
All Sites
The interaction and main effects for all cancer sites for males and females combined appear in Table 1.
Interaction effect. The interaction effect shown in Table 1 proved to be statistically significant: [chi square] = 47.52, p < 0.00001. For all sites, males and females had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.39) than if they smoked marijuana and not tobacco (r = 0.13). Also, for all sites, males and females had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.13) than if they smoked neither marijuana nor tobacco (r = 0.32). In addition, for all sites, males and females had no substantially greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.39) than if they smoked neither marijuana nor tobacco (r = 0.32).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing cancer (r = 0.55) than did males and females who did not smoke tobacco (r = 0.45). This difference was statistically significant: [chi square] = 288, p < 0.00001 with a correlation coefficient of -0.09, Z = -3.42, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for all sites collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing cancer (r = 0.29) than did males and females who did not smoke marijuana (r = 0.71). This difference was statistically significant: [chi square] = 36.09, p < 0.00001 with a correlation coefficient of 0 .50, Z = 14.27, p < .05.
Tobacco-Related Sites
The interaction and main effects for TRC in males and females appear in Table 1.
Interaction Effect. The interaction effect shown in Table 1 proved to be statistically significant: [chi square] = 4.17, p = .04. Males and females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.63) than if they smoked marijuana and not tobacco (r = .04). Also, males and females had a lower probability of developing TRC if they smoked marijuana and not tobacco (r = .04) than if they smoked neither marijuana nor tobacco (r = 0.14). In addition, males and females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.63) than if they smoked neither marijuana nor tobacco (r = 0.14).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing TRC (r = 0.82) than did males and females who did not smoke tobacco (r = 0.18). This difference was statistically marginally significant: [chi square] = 57.86, p = .000003, with a correlation coefficient of -0.75, Z = -7.84, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing TRC (r = 0.23) than did males and females who did not smoke marijuana (r = 0.77). This difference was statistically significant: [chi square] = 76.43, p = .00003, with a correlation coefficient of 0.62, Z = 6.96, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in males and females appear in Table 1.
Interaction effect. The interaction effect shown in Table 1 did not prove to be statistically significant: [chi square] = 2.21, p = .14. Males and females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.72) than if they smoked marijuana and not tobacco (r = .02). In addition, males and females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = .72) than if they smoked neither marijuana nor tobacco (r = .05).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing lung cancer (r = 0.73) than did males and females who did not smoke tobacco (r = 0.07). This difference was statistically significant: [chi square] = 30.51, p = .00001, with a correlation coefficient of -.75, Z = -7.84, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing lung cancer (r = 0.23) than did males and females who did not smoke marijuana (r = 0.77) This difference was statistically significant: [chi square] = 71.61, p = .00003, with a correlation coefficient of 0.61, Z = 5.06, p < .05.
DISCUSSION
It has been suggested that the cohort study by Sidney and colleagues (1997) did not follow its participants long enough to observe an excess number of lung and TRC cases among MS. Only three such cases were observed among MS (n = 10,710), yet, surprising enough, the follow-up period was sufficient to observe 179 cases of TRC (including lung) among TS (n = 14,592). If the incidence of TRC among MS equals that among TS, then 130 cases would be expected among MS; if the incidence of TRC among MS equals that among NS (n = 23,430), then 16 cases would be expected among MS.
If the observed effect of concomitant tobacco and marijuana smoking reflects the fact that MTS consume relatively less tobacco, then the "precancerous" abnormalities from smoking just a few marijuana cigarettes per day are not, in fact, the equal to those from smoking 20 or more tobacco cigarettes per day, and the "precancerous" abnormalities from concomitant tobacco and marijuana smoking are not, in fact, greater than those from smoking tobacco alone. The true effect of marijuana smoking on cancer may soon emerge from large-scale epidemiologic consortia, such as those under the auspices of the International Agency for Research on Cancer's International Association of Cancer Registries (q.v., the International Head and Neck Cancer Epidemiology Consortium). A major limitation in this literature analysis is that it was post hoc, and we encourage prospective studies utilizing rigorous controls.
The posological question remains as to whether smoking marijuana produces antineoplastic concentrations of cannabinoids in the exposed tissues of recreational or medicinal users. Perhaps the biggest obstacle to developing chemotherapy for lung cancer is the fact that optimal tissue concentrations cannot easily be reached in the bronchial epithelium, where anti-neoplastic cannabinoids from smoked marijuana naturally concentrate. Inhalation of marijuana smoke was shown to produce 800% to 1000% higher concentrations of THC in the lungs compared with those found in blood (75 38 ng/g wet wt tissue vs. 9.2 2.0 ng/ml) (Zhu et al. 2002). If the bronchial epithelium does, in fact, harbor malignant neovasculature prior to clinical detection, then smoking marijuana may treat lung cancer at its roots early on, while treatment still matters (Sarafian et al. 2006).
In support of our findings, Hashibe and colleagues (2006) reviewed the literature and concluded that despite several lines of evidence suggesting the biological plausibility of marijuana being carcinogenic, epidemiologic findings are inconsistent. They conducted a population-based case-control study of the association between marijuana use and the risk of lung and upper aerodigestive tract cancers in Los Angeles. The study included 1,212 incident cancer cases and 1,040 cancer-free controls matched to cases on age, gender, and neighborhood. Subjects were interviewed with a standardized questionnaire. The cumulative use of marijuana was expressed in joint-years, where one joint-year is equivalent to smoking one joint per day for one year. No association was consistently monotonic across exposure categories, and restriction to subjects who never smoked cigarettes yielded similar findings. The results may have been affected by selection bias or error in measuring lifetime exposure and confounder histories; but they suggest that the association of these cancers with marijuana, even with long-term or heavy use, is not strong and may be below practically detectable limits.
CONCLUSION
The illegality of marijuana hinders epidemiologic research. To collect data on the personal use of any illicit substance, invasive questions must be asked, often at the expense of low response rates, for cases and controls alike. For this reason, most epidemiologic studies have not included stratified data on the quantity, frequency, and duration of marijuana use and thus may have been affected by selection bias or errors in measuring lifetime exposure (Llewellyn et al. 2004). Stratified data is needed to ascertain dose-related effects: Without stratification, the full effects of chronic use, in either direction, are diluted by the negligible effects of occasional use. Nevertheless, our re-evaluation of a large cohort study indicates that marijuana smokers are at a significantly lower risk of cancer relative to those who smoke tobacco and not marijuana. While this may be true, it is not our intent to endorse the illegal use of marijuana or for that matter any illicit psychoactive drug. However, if bioactive compounds can be developed from certain constituents of cannabis this may be an important futuristic milestone in medicine.
REFERENCES
Barsky, S.H.; Roth, M.D.; Kleerup, E.C.; Simmons, M. & Tashkin, D.P. 1998. Histopathologic and molecular alterations in bronchial epithelium in habitual smokers of Marijuana; cocaine; and/or tobacco. Journal of the National Cancer Institute 90 (16): 1198-205.
Bifulco, M.; Laezza, C.; Valenti, M.; Ligresti, A.; Portella, G. & D.I. Marzo, V. 2004. A new strategy to block tumor growth by inhibiting endocannabinoid inactivation. Journal of the Federation of American Societies for Experimental Biology 18 (13): 1606-8.
Bifulco, M.; Laezza, C.; Portella, G.; Vitale, M.; Orlando, P.; De Petrocellis, L. & Di Marzo, V. 2001. Control by the endogenous cannabinoid 2001. Control by the endogenous cannabinoid system of ras oncogene-dependent tumor growth. Journal of the Federation of American Societies for Experimental Biology 15 (14): 2745-7.
Blazquez, C.; Gonzalez-Feria, L.; Alvarez, L.; Haro, A.; Casanova, M.L. & Guzman, M. 2004. Cannabinoids inhibit the vascular endothelial Cannabinoids inhibit the vascular endothelial growth factor pathway in gliomas. Cancer Research 64 (16): 5617-23.
Blazquez, C.; Casanova, M.L.; Planas, A.; Del Pulgar, T.G.; Villanueva, C.; Fernandez-Acenero, M.J.; Aragones, J.; Huffman, J.W.; Jorcano, J.L. & Guzman, M. 2003. Inhibition of tumor angiogenesis by cannabinoids. Journal of the Federation of American Societies for Experimental Biology 17 (3): 529-31.
Breuer, R.H.; Pasic, A.; Smit, E.F.; van Vliet, E.; Vonk Noordegraaf, A.; Risse, E.J.; Postmus, P.E. & Sutedja, T.G. 2005. The natural course of preneoplastic lesions in bronchial epithelium. Clinical Cancer Research 11 (2 Pt 1): 537-43.
Casanova, M.L.; Blazquez, C.; Martinez-Palacio, J.; Villanueva, C.; Fernandez-Acernero, M.J.; Huffman, J.W.; Jorcano, J.L. & Guzman, M. 2003. Inhibition of skin tumor growth and angiogenesis in vivo by activation of cannabinoid receptors. Journal of Clinical Investigation 111 (1): 43-50.
Chacko, J.A.; Heiner, J.G.; Siu, W.; Macy, M. & Terris, M.K. 2006. Association between marijuana use and transitional cell carcinoma. Urology 67 (1):100-4.
Chan, P.C.; Sills, R.C.; Braun, A.G.; Haseman, J.K. & Bucher, J.R. 1996. Toxicity and carcinogenicity of D9-tetrahydrocannabinol in Fischer rats and B6C3F1 mice. Fundamental & Applied Toxicology 30: 109-117
Ciolino, H.P.; Wang, T.T. & Yeh, G.C. 1998.Diosmin and diosmetin are agonists of the aryl hydrocarbon receptor that differentially affect cytochrome P450 1A1 activity. Cancer Research 58 (13): 2754-60.
Cottrell, J.C.; Sohn, S.S. & Vogel, W.H. 1973. Toxic effects of marihuana tar on mouse skin. Archives of Environmental Health 26 (5): 277-8.
Duntsch, C.; Divi, M.K.; Jones, T.; Zhou, Q.; Krishnamurthy, M.; Boehm, P.; Wood, G.; Sills, A. & Moor, B.M. 2nd. 2006. Safety and efficacy of a novel cannabinoid chemotherapeutic, KM-233, for the treatment of high-grade glioma. Journal of Neuro-oncology 77 (2): 143-52.
Ford, D.; Vu, H.; Hauer, C.; Helzlsouer, K. & Anthony, J. 2001. Marijuana use is not associated with head, neck or lung cancer in adults younger than 55 years: Results of a case cohort study. Paper presented at the National Institute on Drug Abuse Workshop on Clinical Consequences of Marijuana, Rockville, MD, August 13.
Gardner, B.; Zhu, L.X.; Sharma, S.; Tashkin, D.P. & Dubinett, S.M. 2003. 2003. Methanandamide increases COX-2 _expression and tumor growth in murine lung cancer. Journal of the Federation of American Societies for Experimental Biology 17 (14): 2157-9.
Grimaldi, C.; Pisanti, S.; Laezza, C.; Malfitano, A.M.; Santoro, A.; Vitale, M.; Caruso, M.G.; Notarnicola, M.; Iacuzzo, I.; Portella, G.; Di Marzo, V. & Bifulco, M. 2006. Anandamide inhibits adhesion and migration of breast cancer cells. Experimental Cell Research 312 (4): 363-73.
Hashibe, M.; Morgenstern, H.; Cui, Y.; Tashkin, D.P.; Zhang, Z.F.; Cozen, W.; Mack, T.M. & Greenland, S. 2006. Marijuana use and the risk of lung and upper aerodigestive tract cancers: Results of a population-based case-control study. Cancer Epidemiology Biomarkers & Prevention 15: 1829-34
Hsairi, M.; Achour, N.; Zouari, B.; Ben Romdhane, H.; Achour, A.; Maalej, M. & Nacef, T. 1993. Etiologic factors in primary bronchial carcinoma in Tunisia. Tunisie Medicale 71 (5): 265-8.
Huff, J. & Chan, P. 2005. Is marihuana [THC] anti-carcinogenic? Toxicon 46 (3): 357-9.
Kogan, N.M.; Blazquez, C.; Alvarez, L.; Gallily, R.; Schlesinger, M.; Guzman, M. & Mechoulam, R.A. 2006. Cannabinoid quinone inhibits A. 2006. Cannabinoid quinone inhibits angiogenesis by targeting vascular endothelial cells. Molecular Pharmacology 70 (1): 51-9.
Kogan, N.M.; Rabinowitz, R.; Levi, P.; Gibson, D.; Sandor, P.; Schlesinger, M. & Mechouam, R. 2004. Synthesis and antitumor activity of quinonoid derivatives of cannabinoids. Journal of Medical Chemistry 47 (15): 3800-6.
Ligresti, A.; Schiano, M.A.; Starowicz, K.; Matias, I.; Pisanti, S.; De Petrocellis, L.; Laezza, C.; Portella, G.; Bifulco, M. & Di Marzo, V. 2006. Anti-tumor activity of plant cannabinoids with emphasis on the effect of cannabidiol on human breast carcinoma. Journal of Pharmacology and Experimental Therapeutics. 318 (3):1375-87.
Llewellyn, C.D.; Johnson, N.W. & Warnakulasuriya, K.A. 2004. Risk factors for oral cancer in newly diagnosed patients aged 45 years and younger: A case-control study in Southern England. Journal of Oral Pathology and Medicine 33 (9): 525-32.
Llewellyn, C.D.; Linklater, K.; Bell, J.; Johnson, N.W. & Warnakulasuriya, S. 2004 An analysis of risk factors for oral cancer in young people: A case-control study. Oral Oncology 40 (3): 304-13.
Mahadevan, B.; Marston, C.P.; Dashwood, W.M.; Li, Y.; Pereira, C. & Baird, W.M. 2005. Effect of a standardized complex mixture derived from coal tar on the metabolic activation of carcinogenic polycyclic aromatic hydrocarbons in human cells in culture. Chemical Research in Toxicology 18 (2): 224-31.
Massi, P.; Vaccani, A.; Ceruti, S.; Colombo, A.; Abbracchio, M.P. & Parolaro, D. 2004. Antitumor effects of cannabidiol, a nonpsychoactive cannabinoid, on human glioma cell lines. Journal of Pharmacology and Experimental Therapeutics 308 (3): 838-45.
McKallip, R.J.; Nagarkatti, M. & Nagarkatti, P.S. 2005. Delta-9-tetrahydrocannabinol enhances breast cancer growth and metastasis by suppression of the antitumor immune response. Journal of Immunology 174 (6): 3281-89.
McKallip, R.J.; Jia, W.; Schlomer, J.; Warren, J.W.; Nagarkatti, P.S. & Nagarkatti, M. 2006. Cannabidiol-induced apoptosis in human leukemia cells: A novel role of cannabidiol in the regulation of p22phox and Nox4 expression. Molecular Pharmacology. 70(3):897-908.
McKallip, R.J.; Lombard, C.; Fisher, M.; Martin, B.R.; Ryu, S.; Grant, S.; Nagarkatti, P.S. & Nagarkatti, M. 2002. Targeting CB2 cannabinoid receptors as a novel therapy to treat malignant lymphoblastic disease. Blood 100 (2): 627-34.
Munson, A.E.; Harris, L.S.; Friedman, M.A.; Dewey, W.L. & Carchman, R.A. 1975. Antineoplastic activity of cannabinoids. Journal of the National Cancer Institute 55 (3): 597-602.
Portella, G.; Laezza, C.; Laccetti, P.; De Petrocellis, L.; Di Marzo, V. & Bifulco, M. 2003. Inhibitory effects of cannabinoid CB1 receptor stimulation on tumor growth and metastatic spreading: Actions on signals involved in angiogenesis and metastasis. Journal of the Federation of American Societies for Experimental Biology 17 (12): 1771-3.
Proctor, R.N. 2004. The global smoking epidemic: A history and status report. Clinical Lung Cancer 5 (06): 371-76
Recht, L.D.; Salmonsen, R.; Rosetti, R.; Jang, T.; Pipia, G.; Kubiatowski, T.; Karim, P.; Ross, A.H.; Zurier, R.; Lintofsky, N.S. & Burstein, S. 2001. Antitumor effects of ajulemic acid (CT3); A synthetic non-psychoactive cannabinoid. Biochemical Pharmacology 62 (6): 755-63.
Rosenblatt, K.A.; Daling, J.R.; Chen, C.; Sherman, K.J. & Schwartz, S.M. 2004. Marijuana use and risk of oral squamous cell carcinoma. Cancer Research 64 (11): 4049-54.
Roth, M.D.; Marques-Magallanes, J.A.; Yuan, M.; Sun, W.; Tashkin, D.P. & Hankinson, O. 2001. Induction and regulation of the carcinogen-metabolizing enzyme CYP1A1 by marijuana smoke and delta (9)-tetrahydrocannabinol. American Journal of Respiratory Cell and Molecular Biology 24 (3): 339-44.
Sanchez, C; de Ceballos, M.L.; del Pulgar, T.G.; Rueda, D.; Corbacho, C.; Velasco, G.; Galve-Roperh, I.; Huffman, J.W.; Ramon y Cajal, S. & Guzman, M. 2001. Inhibition of glioma growth in vivo by selective activation of the CB (2) cannabinoid receptor. Cancer Research 61 (15): 5784-9.
Sarafian, T.A.; Habib, N.; Oldham, M.; Seeram, N.; Lee, R.P.; Lin, L.; Tashkin, D.P. & Roth, M.D. 2006. Inhaled marijuana smoke disrupts mitochondrial energetics in pulmonary epithelial cells in vivo. American Journal of Physiology--Lung Cellular and Molecular Physiology 290 (6): L1202-9.
Sasco, A.J.; Merrill, R.M.; Dari, I.; Benhaim-Luzon, V.; Carriot, F.; Cann, C.I. & Bartal, M.A. 2002. Case-control study of lung cancer in Casablanca, Morocco. Cancer Causes and Control 13 (7): 609-16.
Schafer, J. & Strimmer, K. 2005. An empirical Bayes approach to inferring large-scale gene association networks. Bioinformatics 21 (6): 75464.
Sidney, S.; Quesenberry, C.P. Jr; Friedman, G.D. & Tekawa, I.S. 1997. Marijuana use and cancer incidence (California; United States). Cancer Causes and Control 8 (5): 722-8.
Stevens, J.P. 2002. Applied Multivariate Statistics for the Social Sciences. Second Ed. Mawah, NJ: Lawrence Erlbaum Associates.
Talaska, G.; Schamer, M.; Bailey, J.R.; Ali, S.F.; Scallet, A.C.; Slikker, W. Jr & Paule, M.G. 1992. No increase in carcinogen-DNA adducts in the lungs of monkeys exposed chronically to marijuana smoke. Toxicology Letters 63 (3): 321-32.
Tashkin, D.; Morgenstern, H.; Zhang, Z.; Greenland, S.; Cozen, W. & Mack, T. 2006. Marijuana use and cancers of the lung and upper aerodigestive tract: Results of a case-control study. Paper presented at the American Thoracic Society 102nd International Conference, San Diego, CA, May 23.
Wang, Y.; Zhang, Z.; Yan, Y.; Lemon, W.J.; LaRegina, M.; Morrison, C.; Lubet, R. & You, M. 2004. A chemically induced model for squamous cell carcinoma of the lung in mice: Histopathology and strain susceptibility. Cancer Research 64 (5): 1647-54.
Watson, E.S. 1989. The effect of marijuana smoke exposure on murine sarcoma 180 survival in Fisher rats. Immunopharmacology and Immunotoxicology 11 (2-3): 211-22.
Winterhalder, R.C.; Hirsch, F.R.; Kotantoulas, G.K.; Franklin, W.A. & Bunn, P.A. Jr. 2004. Chemoprevention of lung cancer: From biology to clinical reality. Annals of Oncology 15 (2): 185-96.
Zhang, Z.F.;Morgenstern, H.; Spitz, M.R.; Tashkin, D.P.; Yu, G.P.; Marshall, J.R.; Hsu, T.C. & Schantz, S.P. 1999. Marijuana use and increased risk of squamous cell carcinoma of the head and neck. Cancer Epidemiology Biomarkers and Prevention 8 (12): 1071-8.
Zhu, K.; Levine, R.S.; Brann, E.A.; Hall, H.I.; Caplan, L.S. & Gnepp, D.R. 2002. Case-control study evaluating the homogeneity and heterogeneity of risk factors between sinonasal and nasopharyngeal cancers. International Journal of Cancer Research 99 (1): 119-23.
Zhu, L.X.; Sharma, S.; Stolina, M.; Gardner, B.; Roth, M.D.; Tashkin, D.P. & Dubinett, S.M. 2000. Delta-9-tetrahydrocannabinol inhibits antitumor immunity by a CB2 receptor-mediated; cytokine-dependent pathway. Journal of Immunology 165 (1): 373-80.
Amanda L.C. Chen, Ph.D. *; Thomas J.H. Chen, Ph.D. **; Eric R. Braverman, M.D. ***; Vanessa Acuri, B.S. ****; Mallory Kerner ****; Michael Varshavskiy ****; Dasha Braverman, P.A. *****; William B. Downs, B.Sc. *****; Seth H. Blum, B.A. ****** Kimberly Cassel ****** & Kenneth Blum, Ph.D. *******
([dagger]) The authors are appreciative of the work, contribution and original concept of Cory A. Gordon, especially for the intensive literature search. The authors are also thankful for the statistical analysis by Manuel Martinez-Pons of the Department of Education, Brooklyn College CUNY, New York, NY.
* Associate Professor Department of Engineering and management of advanced Technology, Chang Jung Christian University, Tainan, Taiwan.
** Professor Department of Occupational Safety and Health, Chang Jung Christian University, Tainan, Taiwan; Changhua Christian Hospital, Changhua, Taiwan.
*** Clinical Assistant Professor, Department of Neurosurgery, Weill Cornell College of Medicine; Medical Director, Path Medical Foundation, New York, NY.
**** Research Associate, Department of Neurological Research, Path Medical Foundation, New York, NY.
***** Senior Research Scientist, Department of Nutrigenomics, LifeGen Inc. La Jolla, CA.
****** Research Associate, Department of Personalized Medicine, Synaptamine, Inc., San Antonio, TX.
******* Scientific Director, Path Medical Foundation, New York, NY; President & CEO, Synapatmine, Inc., San Antonio, TX; Research Professor(Adjunct), Department of Physiology & Pharmacology, Wake Forest University School Of Medicine, Winston-Salem, NC.
Please address correspondence and reprint requests to Kenneth Blum, Wake Forest University School Of Medicine, Department of Physiology & Pharmacology, Medical Center Boulevard, Winston- Salem, North Carolina 27157; email: drd2gene@aol.com
TABLE 1
Cancer Incident by Site and Gender
Marijuana x
Cigarettes
Interaction
Effect
[chi
square] p
Males, All Sites 10.35 .001
Males, Tobacco (TRC) 1.05 .31
Males, Colorectal 2.08 .15
Males, Lung .85 .36
Males, Melanoma .86 .35
Males, Prostate .62 .43
Females, All Sites 34.60 .00001
Females, Tobacco (TRC) 3.14 .00001
Females, Colorectal 3.03 .08
Females, Lung 1.08 .30
Females, Melanoma 3.14 .08
Females, Breast 23.56 .00001
Females, Cervix 8.17 .004
Males and Females, All Sites 47.52 .00001
Males and Females, Tobacco (TRC) 4.17 .04
Males and Females, Lung 2.21 .14
Main Effect for
Marijuana Only
[chi
square] p r Z
Males, All Sites 31 .00001 .31 5.81
Males, Tobacco (TRC) 48.63 .03 .39 3.63
Males, Colorectal 2.60 .27 .39 2.47
Males, Lung 40.47 .00001 .44 2.79
Males, Melanoma 4.67 .09 .51 3.87
Males, Prostate 5.78 .05 .46 2.68
Females, All Sites 36.09 .00001 .50 14.27
Females, Tobacco (TRC) 28 .00001 .99 5.90
Females, Colorectal 3.58 .17 1.0 5.14
Females, Lung 31.43 .00001 .82 4.22
Females, Melanoma 3.27 .019 .27 2.68
Females, Breast 25 .00001 .66 10.21
Females, Cervix 9.72 .008 .14 2.41
Males and Females, All Sites 36.09 .00001 .50 14.27
Males and Females, Tobacco (TRC) 76.43 .00003 .62 6.96
Males and Females, Lung 71.61 .00003 .61 5.06
Main Effect for
Cigarettes Only
[chi
square] p r Z
Males, All Sites 44 .02 -.23 -4.53
Males, Tobacco (TRC) 14.73 .001 -.86 -6.14
Males, Colorectal 8.20 .02 .11 -.74
Males, Lung 9.04 .01 1.56 4.31
Males, Melanoma 17.00 .00001 -.23 -1.95
Males, Prostate 8.90 .01 -.34 -2.12
Females, All Sites 248 .00001 -.04 -1.25
Females, Tobacco (TRC) 3.14 .08 -.64 -4.82
Females, Colorectal 1.08 .29 -.10 -.76
Females, Lung 38.38 .0001 -4.0 -4.60
Females, Melanoma 7.37 .03 -.05 -.37
Females, Breast 135 .03 .07 1.37
Females, Cervix 13.85 .001 -.07 -1.27
Males and Females, All Sites 288 .00001 -.09 -3.42
Males and Females, Tobacco (TRC) 57.86 .00003 -.75 -7.84
Males and Females, Lung 30.51 .00001 -.75 -7.84
Note: This is a reanalysis of data presented in Sidney et al. 1997.
Source: Hypothesizing that marijuana smokers are at a significantly lower risk of carcinogenicity relative to tobacco-non-marijuana smokers: evidenced based on statistical reevaluation of current literature
The decision to use marijuana for medical purposes should be accurately informed by scientific findings of benefit versus risk. The effects of marijuana smoking on cancer remain hypothetical but warrant attention. Without advocating or opposing the use of marijuana for any purpose, this study reconsiders the carcinogenicity of marijuana smoke.
THE CONTROVERSIAL ISSUE
A hypothetical link between marijuana smoking and cancer has been established based on the following observations: (1) marijuana smoke contains carcinogenic hydrocarbons; (2) cannabinoid administration promotes cancer under certain laboratory conditions; (3) lesions similar to those caused by tobacco smoke are found in the bronchial epithelium of marijuana smokers; and (4) marijuana tar produces tumors when painted on the skin of animals. The best evidence to date on the link between marijuana and cancer, however, derives from epidemiologic case-control studies, especially those with a large number of cases and/or randomly selected controls. Such studies tend to suggest, if anything, an inverse association between marijuana use and cancers. To test the hypothesis that marijuana smoking significantly lowers the risk of developing cancer in humans, we analyzed published data from a prospective cohort study on cancer incidence among nonsmokers (NS), marijuana-only smokers (MS), tobacco-only smokers (TS), and marijuana and tobacco smokers (MTS).
Ads by GoogleThe Effects of Smoking
Get your questions answered by the experts. Free online resource.
Medical Information Websites and Medical Web Design for Doctors - HealthCommunities.com - HealthCommunities
Electric Vapor Free Offer
Sign Up Today For Your Free Trial. #1 Rated Product. Limited Supplies.
PrLog.org/News
The following brief review of the literature on some key issues concerning marijuana smoking and cancer will provide the reader with a conceptual framework to allow for intelligent comprehension of the controversial nature of this important medical issue.
Marijuana Smoke Contains Higher Levels of Carcinogenic Polycyclic Aromatic Hydrocarbons (PAH) Than Those Found in Tobacco Smoke
As a rule, the level of carcinogenic PAHs within a chemical mixture is less important than the overarching influence of the mixture on carcinogenic PAH activation (Mahadevan et al. 2005). Indeed, marijuana smoke is presumed carcinogenic not only because it contains carcinogenic PAHs, but also because cannabinoids, in their own right, influence the cytochrome enzymes (CYP1) that determine carcinogenic PAH activation (Roth et al. 2001). Perhaps due to their phenolic hydrocarbon structure, cannabinoids share the anticarcinogenic disposition of polyphenols to increase levels of CYP1A1 messenger RNA while competitively reducing CYP1A1 enzyme activity through aryl hydrocarbon receptor ligation (Ciolino, Wang & Yeh 1998). Not surprisingly, spiking tobacco tar with delta-9 THC markedly reduced carcinogenic activity (Roth et al. 2001).
Cannabinoid Administration Promotes Cancer in Mice
Intraperitoneal administration of immunosuppressive cannabinoids (> 5 mg/kg) promotes cannabinoid-resistant lines of cancer in immune-responsive strains of mice (e.g., BALB/cJ) (McKallip, Nagarkatti & Nagarkatti 2005; Gardner et al. 2003; Zhu et al. 2000). Most in vivo studies, however, have shown that cannabinoid administration, either locally or systemically over a wide range of doses, inhibits the growth of cancer (Duntsch et al. 2006; Grimaldi et al. 2006; Kogan et al. 2006, 2004; Ligresti et al. 2006; McKallip et al. 2006, 2002; Bifulco et al. 2004, 2001; Blazquez et al. 2004, 2003; Massi et al. 2004; Casanova et al. 2003; ; Portella et al. 2003; Recht et al. 2001; Sanchez et al. 2001; Chan et al. 1996; Munson et al. 1975). The neoplastic effects of cannabinoids vary according to strain/species, cell type, dosage, timing, route of administration, serum proteins, distribution of receptors (e.g., cannabinoid and vanilloid), enzymatic degradation, and degree of malignancy. Cannabinoids may control the growth of human cancers, in part, through the cannabinoid receptor 2 gene (CNR2), which emerged as the best-connected "super hub" in an inferred large-scale association network for breast cancer data (Schafer & Strimmer 2005).
Marijuana Smokers Develop Precancerous Lesions Similar to Those Caused by Tobacco Smoke
Marijuana smoking causes precancerous epithelial lesions (PEL) such as squamous metaplasia (SM) and other changes associated with respiratory exposure to smoke in general (Barsky et al. 1998). On the one hand, SM not only follows exposure to potent carcinogens in laboratory strains of mice, but it also precedes the development of squamous cell carcinoma of lungs (SCCL) in humans. On the other hand, the multiplicity of SM was higher among SCCL-resistant mice (e. g., C57BL/6J = 5. 0 - 6. 0) than it was among SCCL-susceptible mice (e.g., NIH Swiss = 2. 1 - 4. 9; Barsky et al. 1998); in humans, PEL such as SM are generally reversible and often regress spontaneously (Winterhaldler et al. 2004).
As it turns out, PEL may have little, if any, predictive value. In fact, according to recent findings, "Distribution and outcome of preneoplastic lesions have been found to be unrelated to various risk factors such as smoking ... The initial WHO classification of any preneoplastic lesion cannot be reliably used for accurate risk assessment of field carcinogenesis" (Breuer et al. 2005).
Marijuana Tar Produces Tumors When Painted on the Skin of Animals
Tar from marijuana smoke, like that from tobacco smoke, was shown to produce benign tumors (i.e., squamous-cell papilloma) when painted on the skin of animals (Cottrell, Sohn & Vogel 1973); however, tar from tobacco smoke caused frank malignancies (i.e., squamous-cell carcinoma), whereas tar from marijuana smoke did not. In subsequent research on monkeys, prolonged exposure to marijuana smoke failed to produce any carcinogenic effects (Talaska et al. 1992). Interestingly enough, exposure to marijuana smoke was shown to retard the growth of sarcoma in rats (Watson 1989). This inhibition was unrelated to the cannabinoid content of the smoke.
The appearance of papillomas on the skin indicates that tar from marijuana smoke, like that from tobacco smoke, is an effective initiator of benign tumors; however, the absence of squamous-cell carcinoma is consistent with the observation that cannabinoid administration induces the regression of squamous-cell carcinoma of skin by transforming the vascular hyperplasia of engorged tumors into pallid networks of small, differentiated capillaries (Casanova et al. 2003). Such transformation reflects the tendency of cannabinoids to thwart the neoplastic expression of angiogenic factors (Casanova et al. 2003; Portella et al. 2003).
While none of the animal studies has demonstrated the carcinogenicity of marijuana smoke, future studies should be rigorously conducted (Huff & Chan 2005). Most in vivo studies on the neoplastic properties of cannabinoids have, unfortunately, been conducted with BALB/cJ and C57BL/6J murine strains, which are not susceptible to chemically-induced SCCL. Even the most susceptible strains do not develop SCCL as the result of exposure to tobacco smoke, the definitive benchmark for human carcinogens (Wang et al. 2004).
In evaluating the carcinogenicity of any type of smoke, it might help to remember that it was epidemiology, rather than animal research, that first incriminated tobacco smoke as a carcinogen (Proctor 2004). There have been no convincing findings of an etiologic association between marijuana smoking and tobacco-related cancer (TRC): Small hospital-based case-control studies have inconsistently found an association (Chacko et al. 2006; Llewellyn et al. 2004; Sasco et al. 2002; Zhang et al. 1999; Hsairi et al. 1993), whereas population-based case-control studies, especially those with a large number of cases and/or randomly selected controls, tend to suggest, if anything, an inverse association between marijuana use and cancers (Tashkin et al. 2006; Sarafian et al. 2006; Rosenblatt et al. 2004; Zhu et al. 2002; Ford et al. 2001).
STATISTICAL METHODS
To test the hypothesis that marijuana smoking significantly lowers the risk of developing cancer in humans, we analyzed published data from a prospective cohort study on cancer incidence among nonsmokers (NS), marijuana-only smokers (MS), tobacco-only smokers (TS), and marijuana and tobacco smokers (MTS) (Sidney et al. 1997). This study represents the most complete systematic evaluation of the issue of MS and cancer risks. The log linear model was used to calculate the probability of developing each cancer form as a function of the interaction effect of marijuana and cigarette smoking, as well as functions of marijuana and tobacco smoking main effects. Log linear models are a generalization of ordinary r by c chi-square analyses and appropriate for count data when more variables than rows (r) and columns (c) are being studied as is the case with these data. Chi square statistics were calculated for the interaction and main effect estimates. The outcomes of the analyses for males and females are shown in Table 1. For each form of cancer, the probability of developing the cancer form as a function of the interaction between marijuana and tobacco smoking as well as functions of main marijuana and tobacco smoking effects are reported (see Stevens 2002 for information on the method of analysis). We used the technique of collapsing across an effect to average over levels of the effect (e.g., for tobacco smoking, to take the average of tobacco smokers and nontobacco smokers and average them together for one mean).
OUTCOMES FOR MALES
All Sites
The interaction and main effects for all cancer sites in males appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 10.35, p = 0.001. For all sites, males had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.36) than if they smoked marijuana and not tobacco (r = 0.09) For all sites, males had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.09) than if they smoked neither marijuana nor tobacco (r = 0.29). In addition, for all sites, males had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.36) than if they smoked neither marijuana nor tobacco (r = 0.29).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing cancer (r = 0.61) than did males who did not smoke tobacco (r = 0.38). This difference was statistically significant [chi square] = 44, p = 0.02. The correlation coefficient was -.23, Z = -4.53, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. For all sites collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing cancer (r = 0.35) than did males who did not smoke marijuana (r = 0.65) This difference was statistically significant: [chi square] = 31, p < .00001. The correlation coefficient was .31, Z = 5.81, p < .05.
Tobacco Related Cancer
The interaction and main effects for TRC in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 1.05, p = .31.
Males had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.03) than if they smoked marijuana and not tobacco (r = 0.56) although this interaction effect was not statistically significant. For TRC, males had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.03) than if they smoked neither marijuana nor tobacco (r = 0.12). In addition, males had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.56) than if they smoked neither marijuana nor tobacco (r = 0.12) although, as already noted, this interaction effect was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing TRC (r = 0.84) than did males who did not smoke tobacco (r = 0.75). This difference proved to be statistically significant: [chi square] = 14.73, p = .001, with a correlation coefficient of -.86, Z = -6.14, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing TRC (r = 0.31) than did males who did not smoke marijuana (r = 0.68). This difference was statistically significant: [chi square] = 48.63, p = 0.03, with a correlation coefficient of .39, Z = 3.63, p < .05.
Colorectal Cancer
The interaction and main effects for colorectal cancer in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 2.08, p = 0.15. Males had a greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.33) than if they smoked marijuana and not tobacco (r = 0.08) although, as noted above, the interaction was statistically non-significant. Males also had a lower probability of developing colorectal cancer if they smoked Marijuana and not tobacco (r = 0.08) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as noted above, the interaction was statistically non-significant. Males had no greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.33) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as noted above, the interaction was statistically nonsignificant.
Main effects of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing colorectal cancer (r = 0.56) than did males who did not smoke tobacco (r = 0.44). This difference proved to be statistically significant: [chi square] = 8.20, p = .02, with a correlation coefficient of -0.11, Z = -0.74, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing colorectal cancer (r = 0.31) than did males who did not smoke marijuana (r = 0.69), [chi square] = 2.60, p = 0.27. The correlation coefficient was 0.39, Z = 2.47, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in males appear in Table 1.
Interaction effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 0.85, p = 0.36. Males had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.67) than if they smoked marijuana and not tobacco (r = 0) although, as noted above, the interaction was statistically nonsignificant. Males did not display a difference in the probabilities of developing lung cancer if they smoked marijuana and not tobacco (r = 0) and neither marijuana nor tobacco (r = 0.04). They had no greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.67) than if they smoked neither marijuana nor tobacco (r = 0.04) although, as noted above, the interaction was statistically nonsignificant.
Main effects of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing lung cancer (r = 0.95) than did males who did not smoke tobacco (r = .02). This difference proved to be statistically significant: [chi square] = 9.04, p = .01, with a correlation coefficient of 1.56, Z = 4.31, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing lung cancer (r = 0.15) than did males who did not smoke marijuana (r = 0.37) This difference proved to be statistically significant: [chi square] = 40.47, p < 0.00001 with a correlation coefficient of 0.44, Z = 2.79, p < .05.
Melanoma
The interaction and main effects for melanoma in males appear in Table 1.
Interaction Effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = .86, p = 0.35. Males had a greater probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.43) than if they smoked marijuana and not tobacco (r = 0.08), and displayed a difference between the probabilities of developing melanoma if they smoked marijuana and not tobacco (r = 0.08) and smoking neither marijuana nor tobacco (r = 0.31) although, as noted above, the interaction was statistically nonsignificant. Males displayed a greater probability of developing melanoma if they smoked tobacco but not marijuana (r = 0.67) than if they smoked neither marijuana nor tobacco (r = 0.31), although, as noted above, the interaction was statistically nonsignificant.
Main effect of tobacco smoking collapsing across marijuana smoking. Males who smoked tobacco had a higher probability of developing melanoma (r = 0.61) than did males who did not smoke tobacco (r = 0.32). This difference proved to be statistically significant: [chi square] = 17, p < 0.00001 with a correlation coefficient of -0.23, Z = -1.95, p < .05.
Main effect of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing melanoma (r = 0.27) than did males who did not smoke marijuana (r = 0.73) although this difference proved to be only marginally statistically significant: [chi square] = 4.67, p = .09, with a correlation coefficient of 0.51, Z = 3.87, p < .05.
Prostate Cancer
The interaction and main effects for prostate cancer in males appear in Table 1.
Interaction Effects. The interaction effects appearing in Table 1 did not prove to be statistically significant: [chi square] = 0.62, p = 0.43. Males had a greater probability of developing prostate cancer if they smoked tobacco and not marijuana (r = 0.50) than if they smoked marijuana and not tobacco (r = 0.12). They also displayed a difference between the probabilities of developing prostate cancer if they smoked marijuana and not tobacco (r = 0.12) and neither marijuana nor tobacco (r = 0.21); and between the probabilities in developing this form of cancer if they smoked tobacco and not marijuana (r = 0.50) and neither tobacco nor marijuana (r = 0.21) although, as noted above, these interaction effects were statistically nonsignificant.
Main effect of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males who smoked tobacco had a higher probability of developing prostate cancer (r = .67) than did males who did not smoke tobacco (r = .33). This difference proved to be statistically significant: [chi square] = 8.90, p = .01, with a correlation coefficient of -.34, Z = 2.12, p < .05.
Main effect of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males who smoked marijuana had a lower probability of developing prostate cancer (r = 0.29) than did males who did not smoke marijuana (r = 0.71), [chi square] = 5.78, p = .05, with a correlation coefficient of 0.46, Z = 2.68, p < .05.
OUTCOMES FOR FEMALES
All Sites
The interaction and main effects for all cancer sites in females appear in Table 1.
Interaction Effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 34.60, p < 0.00001 For all sites, females had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.34) than if they smoked marijuana and not tobacco (r = 0.08). Also, for all sites, females had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.08) than if they smoked neither marijuana nor tobacco (r = 0.39). In addition, for all sites, females had no substantially greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.34) than if they smoked neither marijuana nor tobacco (r = 0.39).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing cancer (r = 0.51) than did females who did not smoke tobacco (r = 0.48). This difference was statistically significant: [chi square] = 248, p < 0.00001 with a correlation coefficient of -0.04, Z = -1.25, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for all sites collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing cancer (r = 0.27) than did females who did not smoke marijuana (r = 0.73) This difference was statistically significant: [chi square] = 36.09, p < 0.00001 with a correlation coefficient of .50, Z = 14.27, p < .05.
Tobacco Related Cancer
The interaction and main effects for TRC in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 3.14, p < 0.00001. Females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.66) than if they smoked marijuana and not tobacco (r = 0). Also, females had a lower probability of developing TRC if they smoked marijuana and not tobacco (r = 0) than if they smoked neither marijuana nor tobacco (r = 0.22). In addition, females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.66) than if they smoked neither marijuana nor tobacco (r = 0.22).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing TRC (r = 0.78) than did females who did not smoke tobacco (r = 0.22). This difference was statistically marginally significant: [chi square] = 3.14, p = 0.08, with a correlation coefficient of -0.64, Z = -4.82, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing TRC (r = .12) than did females who did not smoke marijuana (r = .88). This difference was statistically significant: [chi square] = 28, p = 0, with a correlation coefficient of .99, Z = 5.90, p < .05.
Colorectal Cancer
The interaction and main effects for colorectal cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant, or only marginally so: [chi square] = 3.03, p = 0.08. Females had a greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.45) than if they smoked marijuana and not tobacco (r = 0.01) although, as already noted, the difference was only marginally statistically significant. Also, females had a lower probability of developing colorectal cancer if they smoked marijuana and not tobacco (r = 0.01) than if they smoked neither marijuana nor tobacco (r = 0.44). Also as already noted, the difference was only marginally statistically significant. In addition, females had no greater probability of developing colorectal cancer if they smoked tobacco and not marijuana (r = 0.45) than if they smoked neither marijuana nor tobacco (r = 0.44).
Main effects of tobacco smoking collapsing across marijuana smoking. For colorectal cancer collapsing across marijuana smoking, females who smoked tobacco had no higher probability of developing cancer (r = 0.55) than did females who did not smoke tobacco (r = 0.45). The difference did not prove to be statistically marginally significant: [chi square] = 1.08, p = 0.29, with a correlation coefficient of -0.10, Z = -0.76, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing colorectal cancer (r = 0.11) than did females who did not smoke marijuana (r = 0.89), [chi square] = 3.58, p = 0.17, with a correlation coefficient of 1.0, Z = 5.14, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant: [chi square] = 1.08, p = 0.30. Females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.73) than if they smoked marijuana and not tobacco (r = 0) although, as already noted, the difference was not statistically significant. Also, females had a slightly lower probability of developing lung cancer if they smoked marijuana and not tobacco (r = 0) than if they smoked neither marijuana nor tobacco (r = 0.10) although, also as already noted, the difference was not statistically significant. Females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.73) than if they smoked neither marijuana nor tobacco (r = 0.44) although, as already noted, the difference was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a higher probability of developing lung cancer (r = 0.90) than did females who did not smoke tobacco (r = 0.10). The difference proved to be statistically significant: [chi square] = 38.38, p < 0.0001, with a correlation coefficient of -1.0, Z = -4.60, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, as shown in Table 1, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing lung cancer (r = 0.16) than did females who did not smoke marijuana (r = 0.84). This difference was statistically significant: [chi square] = 31.43, p < 0.00001 with a correlation coefficient of 0.82, Z = 4.22, p < .05.
Melanoma
The interaction and main effects for melanoma in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 did not prove to be statistically significant, or only marginally so: [chi square] = 3.14, p = 0.08. Females had a slightly greater probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.28) than if they smoked marijuana and not tobacco (r = 0.12) although, as already noted, the interaction effect was not statistically significant. Also, females had a lower probability of developing melanoma if they smoked marijuana and not tobacco (r = 0.12) than if they smoked neither marijuana nor tobacco (r = 0.35) although, also as already noted, the interaction effect was not statistically significant. Females had a slightly lower probability of developing melanoma if they smoked tobacco and not marijuana (r = 0.28) than if they smoked neither marijuana nor tobacco (r = 0.35) although, as already noted, the interaction effect was not statistically significant.
Main effects of tobacco smoking collapsing across marijuana smoking. For melanoma collapsing across marijuana smoking, females who smoked tobacco had a slightly higher probability of developing cancer melanoma (r = 0.52) than did females who did not smoke tobacco (r = 0.48). The difference proved to be statistically significant: [chi square] = 7.37, p = 0.03, with a correlation coefficient of -.05, Z = -.37, p > .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for melanoma collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing melanoma (r = 0.37) than did females who did not smoke marijuana (r = 0.63), [chi square] = 3.27, p = 0.019, with a correlation coefficient of .27, Z = 2.68, p < .05.
Breast Cancer
The interaction and main effects for breast cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 23.56, p < 0. 00001. Females had a slightly greater probability of developing breast cancer if they smoked tobacco and not marijuana (r = 0.31) than if they smoked marijuana and not tobacco (r = 0.06). Also, females had a lower probability of developing breast cancer if they smoked marijuana and not tobacco (r = 0.06) than if they smoked neither marijuana nor tobacco (r = 0.48). They had a slightly lower probability of developing breast cancer if they smoked tobacco and not marijuana (r = 0.31) than if they smoked neither marijuana nor tobacco (r = 0.48).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a lower probability of developing breast cancer (r = 0.48) than did females who did not smoke tobacco (r = 0.55). The difference proved to be statistically significant: [chi square] = 135, p = 0.03, with a correlation coefficient of .07, Z = 1.37, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing breast cancer (r = 0.22) than did females who did not smoke marijuana (r = 0.81). This difference proved to be statistically significant: [chi square] = 25, p < 0.00001 with a correlation coefficient of .66, Z = 10.21, p < .05.
Cancer of the Cervix
The interaction and main effects for cervical cancer in females appear in Table 1.
Interaction effects. The interaction effects shown in Table 1 proved to be statistically significant: [chi square] = 8.17, p = 0.004. Females had a greater probability of developing cancer of the cervix if they smoked tobacco and not marijuana (r = 0.27) than if they smoked marijuana and not tobacco (r = 0.16). Also, females had a lower probability of developing cancer of the cervix if they smoked marijuana and not tobacco (r = 0.16) than if they smoked neither marijuana nor tobacco (r = 0.30). They had no substantially different probability of developing cancer of the cervix if they smoked tobacco and not marijuana (r = 0.27) than if they smoked neither marijuana nor tobacco (r = 0.30).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, females who smoked tobacco had a higher probability of developing cancer of the cervix (r = 0.54) than did females who did not smoke tobacco (r = 0.46). The difference proved to be statistically significant: [chi square] = 13.85, p = 0.001, with a correlation coefficient of -0.07, Z = -1.27, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, females who smoked marijuana had a lower probability of developing cancer of the cervix (r = 0.43) than did females who did not smoke marijuana (r = 0.57). This difference proved to be statistically significant: [chi square] = 9.72, p = 0.008, with a correlation coefficient of 0.14, Z = 2.41, p < .05.
MALES AND FEMALES COMBINED
All Sites
The interaction and main effects for all cancer sites for males and females combined appear in Table 1.
Interaction effect. The interaction effect shown in Table 1 proved to be statistically significant: [chi square] = 47.52, p < 0.00001. For all sites, males and females had a greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.39) than if they smoked marijuana and not tobacco (r = 0.13). Also, for all sites, males and females had a lower probability of developing cancer if they smoked marijuana and not tobacco (r = 0.13) than if they smoked neither marijuana nor tobacco (r = 0.32). In addition, for all sites, males and females had no substantially greater probability of developing cancer if they smoked tobacco and not marijuana (r = 0.39) than if they smoked neither marijuana nor tobacco (r = 0.32).
Main effects of tobacco smoking collapsing across marijuana smoking. For all sites collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing cancer (r = 0.55) than did males and females who did not smoke tobacco (r = 0.45). This difference was statistically significant: [chi square] = 288, p < 0.00001 with a correlation coefficient of -0.09, Z = -3.42, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, for all sites collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing cancer (r = 0.29) than did males and females who did not smoke marijuana (r = 0.71). This difference was statistically significant: [chi square] = 36.09, p < 0.00001 with a correlation coefficient of 0 .50, Z = 14.27, p < .05.
Tobacco-Related Sites
The interaction and main effects for TRC in males and females appear in Table 1.
Interaction Effect. The interaction effect shown in Table 1 proved to be statistically significant: [chi square] = 4.17, p = .04. Males and females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.63) than if they smoked marijuana and not tobacco (r = .04). Also, males and females had a lower probability of developing TRC if they smoked marijuana and not tobacco (r = .04) than if they smoked neither marijuana nor tobacco (r = 0.14). In addition, males and females had a greater probability of developing TRC if they smoked tobacco and not marijuana (r = 0.63) than if they smoked neither marijuana nor tobacco (r = 0.14).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing TRC (r = 0.82) than did males and females who did not smoke tobacco (r = 0.18). This difference was statistically marginally significant: [chi square] = 57.86, p = .000003, with a correlation coefficient of -0.75, Z = -7.84, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing TRC (r = 0.23) than did males and females who did not smoke marijuana (r = 0.77). This difference was statistically significant: [chi square] = 76.43, p = .00003, with a correlation coefficient of 0.62, Z = 6.96, p < .05.
Lung Cancer
The interaction and main effects for lung cancer in males and females appear in Table 1.
Interaction effect. The interaction effect shown in Table 1 did not prove to be statistically significant: [chi square] = 2.21, p = .14. Males and females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = 0.72) than if they smoked marijuana and not tobacco (r = .02). In addition, males and females had a greater probability of developing lung cancer if they smoked tobacco and not marijuana (r = .72) than if they smoked neither marijuana nor tobacco (r = .05).
Main effects of tobacco smoking collapsing across marijuana smoking. Collapsing across marijuana smoking, males and females who smoked tobacco had a slightly higher probability of developing lung cancer (r = 0.73) than did males and females who did not smoke tobacco (r = 0.07). This difference was statistically significant: [chi square] = 30.51, p = .00001, with a correlation coefficient of -.75, Z = -7.84, p < .05.
Main effects of marijuana smoking collapsing across tobacco smoking. Finally, collapsing across tobacco smoking, males and females who smoked marijuana had a lower probability of developing lung cancer (r = 0.23) than did males and females who did not smoke marijuana (r = 0.77) This difference was statistically significant: [chi square] = 71.61, p = .00003, with a correlation coefficient of 0.61, Z = 5.06, p < .05.
DISCUSSION
It has been suggested that the cohort study by Sidney and colleagues (1997) did not follow its participants long enough to observe an excess number of lung and TRC cases among MS. Only three such cases were observed among MS (n = 10,710), yet, surprising enough, the follow-up period was sufficient to observe 179 cases of TRC (including lung) among TS (n = 14,592). If the incidence of TRC among MS equals that among TS, then 130 cases would be expected among MS; if the incidence of TRC among MS equals that among NS (n = 23,430), then 16 cases would be expected among MS.
If the observed effect of concomitant tobacco and marijuana smoking reflects the fact that MTS consume relatively less tobacco, then the "precancerous" abnormalities from smoking just a few marijuana cigarettes per day are not, in fact, the equal to those from smoking 20 or more tobacco cigarettes per day, and the "precancerous" abnormalities from concomitant tobacco and marijuana smoking are not, in fact, greater than those from smoking tobacco alone. The true effect of marijuana smoking on cancer may soon emerge from large-scale epidemiologic consortia, such as those under the auspices of the International Agency for Research on Cancer's International Association of Cancer Registries (q.v., the International Head and Neck Cancer Epidemiology Consortium). A major limitation in this literature analysis is that it was post hoc, and we encourage prospective studies utilizing rigorous controls.
The posological question remains as to whether smoking marijuana produces antineoplastic concentrations of cannabinoids in the exposed tissues of recreational or medicinal users. Perhaps the biggest obstacle to developing chemotherapy for lung cancer is the fact that optimal tissue concentrations cannot easily be reached in the bronchial epithelium, where anti-neoplastic cannabinoids from smoked marijuana naturally concentrate. Inhalation of marijuana smoke was shown to produce 800% to 1000% higher concentrations of THC in the lungs compared with those found in blood (75 38 ng/g wet wt tissue vs. 9.2 2.0 ng/ml) (Zhu et al. 2002). If the bronchial epithelium does, in fact, harbor malignant neovasculature prior to clinical detection, then smoking marijuana may treat lung cancer at its roots early on, while treatment still matters (Sarafian et al. 2006).
In support of our findings, Hashibe and colleagues (2006) reviewed the literature and concluded that despite several lines of evidence suggesting the biological plausibility of marijuana being carcinogenic, epidemiologic findings are inconsistent. They conducted a population-based case-control study of the association between marijuana use and the risk of lung and upper aerodigestive tract cancers in Los Angeles. The study included 1,212 incident cancer cases and 1,040 cancer-free controls matched to cases on age, gender, and neighborhood. Subjects were interviewed with a standardized questionnaire. The cumulative use of marijuana was expressed in joint-years, where one joint-year is equivalent to smoking one joint per day for one year. No association was consistently monotonic across exposure categories, and restriction to subjects who never smoked cigarettes yielded similar findings. The results may have been affected by selection bias or error in measuring lifetime exposure and confounder histories; but they suggest that the association of these cancers with marijuana, even with long-term or heavy use, is not strong and may be below practically detectable limits.
CONCLUSION
The illegality of marijuana hinders epidemiologic research. To collect data on the personal use of any illicit substance, invasive questions must be asked, often at the expense of low response rates, for cases and controls alike. For this reason, most epidemiologic studies have not included stratified data on the quantity, frequency, and duration of marijuana use and thus may have been affected by selection bias or errors in measuring lifetime exposure (Llewellyn et al. 2004). Stratified data is needed to ascertain dose-related effects: Without stratification, the full effects of chronic use, in either direction, are diluted by the negligible effects of occasional use. Nevertheless, our re-evaluation of a large cohort study indicates that marijuana smokers are at a significantly lower risk of cancer relative to those who smoke tobacco and not marijuana. While this may be true, it is not our intent to endorse the illegal use of marijuana or for that matter any illicit psychoactive drug. However, if bioactive compounds can be developed from certain constituents of cannabis this may be an important futuristic milestone in medicine.
REFERENCES
Barsky, S.H.; Roth, M.D.; Kleerup, E.C.; Simmons, M. & Tashkin, D.P. 1998. Histopathologic and molecular alterations in bronchial epithelium in habitual smokers of Marijuana; cocaine; and/or tobacco. Journal of the National Cancer Institute 90 (16): 1198-205.
Bifulco, M.; Laezza, C.; Valenti, M.; Ligresti, A.; Portella, G. & D.I. Marzo, V. 2004. A new strategy to block tumor growth by inhibiting endocannabinoid inactivation. Journal of the Federation of American Societies for Experimental Biology 18 (13): 1606-8.
Bifulco, M.; Laezza, C.; Portella, G.; Vitale, M.; Orlando, P.; De Petrocellis, L. & Di Marzo, V. 2001. Control by the endogenous cannabinoid 2001. Control by the endogenous cannabinoid system of ras oncogene-dependent tumor growth. Journal of the Federation of American Societies for Experimental Biology 15 (14): 2745-7.
Blazquez, C.; Gonzalez-Feria, L.; Alvarez, L.; Haro, A.; Casanova, M.L. & Guzman, M. 2004. Cannabinoids inhibit the vascular endothelial Cannabinoids inhibit the vascular endothelial growth factor pathway in gliomas. Cancer Research 64 (16): 5617-23.
Blazquez, C.; Casanova, M.L.; Planas, A.; Del Pulgar, T.G.; Villanueva, C.; Fernandez-Acenero, M.J.; Aragones, J.; Huffman, J.W.; Jorcano, J.L. & Guzman, M. 2003. Inhibition of tumor angiogenesis by cannabinoids. Journal of the Federation of American Societies for Experimental Biology 17 (3): 529-31.
Breuer, R.H.; Pasic, A.; Smit, E.F.; van Vliet, E.; Vonk Noordegraaf, A.; Risse, E.J.; Postmus, P.E. & Sutedja, T.G. 2005. The natural course of preneoplastic lesions in bronchial epithelium. Clinical Cancer Research 11 (2 Pt 1): 537-43.
Casanova, M.L.; Blazquez, C.; Martinez-Palacio, J.; Villanueva, C.; Fernandez-Acernero, M.J.; Huffman, J.W.; Jorcano, J.L. & Guzman, M. 2003. Inhibition of skin tumor growth and angiogenesis in vivo by activation of cannabinoid receptors. Journal of Clinical Investigation 111 (1): 43-50.
Chacko, J.A.; Heiner, J.G.; Siu, W.; Macy, M. & Terris, M.K. 2006. Association between marijuana use and transitional cell carcinoma. Urology 67 (1):100-4.
Chan, P.C.; Sills, R.C.; Braun, A.G.; Haseman, J.K. & Bucher, J.R. 1996. Toxicity and carcinogenicity of D9-tetrahydrocannabinol in Fischer rats and B6C3F1 mice. Fundamental & Applied Toxicology 30: 109-117
Ciolino, H.P.; Wang, T.T. & Yeh, G.C. 1998.Diosmin and diosmetin are agonists of the aryl hydrocarbon receptor that differentially affect cytochrome P450 1A1 activity. Cancer Research 58 (13): 2754-60.
Cottrell, J.C.; Sohn, S.S. & Vogel, W.H. 1973. Toxic effects of marihuana tar on mouse skin. Archives of Environmental Health 26 (5): 277-8.
Duntsch, C.; Divi, M.K.; Jones, T.; Zhou, Q.; Krishnamurthy, M.; Boehm, P.; Wood, G.; Sills, A. & Moor, B.M. 2nd. 2006. Safety and efficacy of a novel cannabinoid chemotherapeutic, KM-233, for the treatment of high-grade glioma. Journal of Neuro-oncology 77 (2): 143-52.
Ford, D.; Vu, H.; Hauer, C.; Helzlsouer, K. & Anthony, J. 2001. Marijuana use is not associated with head, neck or lung cancer in adults younger than 55 years: Results of a case cohort study. Paper presented at the National Institute on Drug Abuse Workshop on Clinical Consequences of Marijuana, Rockville, MD, August 13.
Gardner, B.; Zhu, L.X.; Sharma, S.; Tashkin, D.P. & Dubinett, S.M. 2003. 2003. Methanandamide increases COX-2 _expression and tumor growth in murine lung cancer. Journal of the Federation of American Societies for Experimental Biology 17 (14): 2157-9.
Grimaldi, C.; Pisanti, S.; Laezza, C.; Malfitano, A.M.; Santoro, A.; Vitale, M.; Caruso, M.G.; Notarnicola, M.; Iacuzzo, I.; Portella, G.; Di Marzo, V. & Bifulco, M. 2006. Anandamide inhibits adhesion and migration of breast cancer cells. Experimental Cell Research 312 (4): 363-73.
Hashibe, M.; Morgenstern, H.; Cui, Y.; Tashkin, D.P.; Zhang, Z.F.; Cozen, W.; Mack, T.M. & Greenland, S. 2006. Marijuana use and the risk of lung and upper aerodigestive tract cancers: Results of a population-based case-control study. Cancer Epidemiology Biomarkers & Prevention 15: 1829-34
Hsairi, M.; Achour, N.; Zouari, B.; Ben Romdhane, H.; Achour, A.; Maalej, M. & Nacef, T. 1993. Etiologic factors in primary bronchial carcinoma in Tunisia. Tunisie Medicale 71 (5): 265-8.
Huff, J. & Chan, P. 2005. Is marihuana [THC] anti-carcinogenic? Toxicon 46 (3): 357-9.
Kogan, N.M.; Blazquez, C.; Alvarez, L.; Gallily, R.; Schlesinger, M.; Guzman, M. & Mechoulam, R.A. 2006. Cannabinoid quinone inhibits A. 2006. Cannabinoid quinone inhibits angiogenesis by targeting vascular endothelial cells. Molecular Pharmacology 70 (1): 51-9.
Kogan, N.M.; Rabinowitz, R.; Levi, P.; Gibson, D.; Sandor, P.; Schlesinger, M. & Mechouam, R. 2004. Synthesis and antitumor activity of quinonoid derivatives of cannabinoids. Journal of Medical Chemistry 47 (15): 3800-6.
Ligresti, A.; Schiano, M.A.; Starowicz, K.; Matias, I.; Pisanti, S.; De Petrocellis, L.; Laezza, C.; Portella, G.; Bifulco, M. & Di Marzo, V. 2006. Anti-tumor activity of plant cannabinoids with emphasis on the effect of cannabidiol on human breast carcinoma. Journal of Pharmacology and Experimental Therapeutics. 318 (3):1375-87.
Llewellyn, C.D.; Johnson, N.W. & Warnakulasuriya, K.A. 2004. Risk factors for oral cancer in newly diagnosed patients aged 45 years and younger: A case-control study in Southern England. Journal of Oral Pathology and Medicine 33 (9): 525-32.
Llewellyn, C.D.; Linklater, K.; Bell, J.; Johnson, N.W. & Warnakulasuriya, S. 2004 An analysis of risk factors for oral cancer in young people: A case-control study. Oral Oncology 40 (3): 304-13.
Mahadevan, B.; Marston, C.P.; Dashwood, W.M.; Li, Y.; Pereira, C. & Baird, W.M. 2005. Effect of a standardized complex mixture derived from coal tar on the metabolic activation of carcinogenic polycyclic aromatic hydrocarbons in human cells in culture. Chemical Research in Toxicology 18 (2): 224-31.
Massi, P.; Vaccani, A.; Ceruti, S.; Colombo, A.; Abbracchio, M.P. & Parolaro, D. 2004. Antitumor effects of cannabidiol, a nonpsychoactive cannabinoid, on human glioma cell lines. Journal of Pharmacology and Experimental Therapeutics 308 (3): 838-45.
McKallip, R.J.; Nagarkatti, M. & Nagarkatti, P.S. 2005. Delta-9-tetrahydrocannabinol enhances breast cancer growth and metastasis by suppression of the antitumor immune response. Journal of Immunology 174 (6): 3281-89.
McKallip, R.J.; Jia, W.; Schlomer, J.; Warren, J.W.; Nagarkatti, P.S. & Nagarkatti, M. 2006. Cannabidiol-induced apoptosis in human leukemia cells: A novel role of cannabidiol in the regulation of p22phox and Nox4 expression. Molecular Pharmacology. 70(3):897-908.
McKallip, R.J.; Lombard, C.; Fisher, M.; Martin, B.R.; Ryu, S.; Grant, S.; Nagarkatti, P.S. & Nagarkatti, M. 2002. Targeting CB2 cannabinoid receptors as a novel therapy to treat malignant lymphoblastic disease. Blood 100 (2): 627-34.
Munson, A.E.; Harris, L.S.; Friedman, M.A.; Dewey, W.L. & Carchman, R.A. 1975. Antineoplastic activity of cannabinoids. Journal of the National Cancer Institute 55 (3): 597-602.
Portella, G.; Laezza, C.; Laccetti, P.; De Petrocellis, L.; Di Marzo, V. & Bifulco, M. 2003. Inhibitory effects of cannabinoid CB1 receptor stimulation on tumor growth and metastatic spreading: Actions on signals involved in angiogenesis and metastasis. Journal of the Federation of American Societies for Experimental Biology 17 (12): 1771-3.
Proctor, R.N. 2004. The global smoking epidemic: A history and status report. Clinical Lung Cancer 5 (06): 371-76
Recht, L.D.; Salmonsen, R.; Rosetti, R.; Jang, T.; Pipia, G.; Kubiatowski, T.; Karim, P.; Ross, A.H.; Zurier, R.; Lintofsky, N.S. & Burstein, S. 2001. Antitumor effects of ajulemic acid (CT3); A synthetic non-psychoactive cannabinoid. Biochemical Pharmacology 62 (6): 755-63.
Rosenblatt, K.A.; Daling, J.R.; Chen, C.; Sherman, K.J. & Schwartz, S.M. 2004. Marijuana use and risk of oral squamous cell carcinoma. Cancer Research 64 (11): 4049-54.
Roth, M.D.; Marques-Magallanes, J.A.; Yuan, M.; Sun, W.; Tashkin, D.P. & Hankinson, O. 2001. Induction and regulation of the carcinogen-metabolizing enzyme CYP1A1 by marijuana smoke and delta (9)-tetrahydrocannabinol. American Journal of Respiratory Cell and Molecular Biology 24 (3): 339-44.
Sanchez, C; de Ceballos, M.L.; del Pulgar, T.G.; Rueda, D.; Corbacho, C.; Velasco, G.; Galve-Roperh, I.; Huffman, J.W.; Ramon y Cajal, S. & Guzman, M. 2001. Inhibition of glioma growth in vivo by selective activation of the CB (2) cannabinoid receptor. Cancer Research 61 (15): 5784-9.
Sarafian, T.A.; Habib, N.; Oldham, M.; Seeram, N.; Lee, R.P.; Lin, L.; Tashkin, D.P. & Roth, M.D. 2006. Inhaled marijuana smoke disrupts mitochondrial energetics in pulmonary epithelial cells in vivo. American Journal of Physiology--Lung Cellular and Molecular Physiology 290 (6): L1202-9.
Sasco, A.J.; Merrill, R.M.; Dari, I.; Benhaim-Luzon, V.; Carriot, F.; Cann, C.I. & Bartal, M.A. 2002. Case-control study of lung cancer in Casablanca, Morocco. Cancer Causes and Control 13 (7): 609-16.
Schafer, J. & Strimmer, K. 2005. An empirical Bayes approach to inferring large-scale gene association networks. Bioinformatics 21 (6): 75464.
Sidney, S.; Quesenberry, C.P. Jr; Friedman, G.D. & Tekawa, I.S. 1997. Marijuana use and cancer incidence (California; United States). Cancer Causes and Control 8 (5): 722-8.
Stevens, J.P. 2002. Applied Multivariate Statistics for the Social Sciences. Second Ed. Mawah, NJ: Lawrence Erlbaum Associates.
Talaska, G.; Schamer, M.; Bailey, J.R.; Ali, S.F.; Scallet, A.C.; Slikker, W. Jr & Paule, M.G. 1992. No increase in carcinogen-DNA adducts in the lungs of monkeys exposed chronically to marijuana smoke. Toxicology Letters 63 (3): 321-32.
Tashkin, D.; Morgenstern, H.; Zhang, Z.; Greenland, S.; Cozen, W. & Mack, T. 2006. Marijuana use and cancers of the lung and upper aerodigestive tract: Results of a case-control study. Paper presented at the American Thoracic Society 102nd International Conference, San Diego, CA, May 23.
Wang, Y.; Zhang, Z.; Yan, Y.; Lemon, W.J.; LaRegina, M.; Morrison, C.; Lubet, R. & You, M. 2004. A chemically induced model for squamous cell carcinoma of the lung in mice: Histopathology and strain susceptibility. Cancer Research 64 (5): 1647-54.
Watson, E.S. 1989. The effect of marijuana smoke exposure on murine sarcoma 180 survival in Fisher rats. Immunopharmacology and Immunotoxicology 11 (2-3): 211-22.
Winterhalder, R.C.; Hirsch, F.R.; Kotantoulas, G.K.; Franklin, W.A. & Bunn, P.A. Jr. 2004. Chemoprevention of lung cancer: From biology to clinical reality. Annals of Oncology 15 (2): 185-96.
Zhang, Z.F.;Morgenstern, H.; Spitz, M.R.; Tashkin, D.P.; Yu, G.P.; Marshall, J.R.; Hsu, T.C. & Schantz, S.P. 1999. Marijuana use and increased risk of squamous cell carcinoma of the head and neck. Cancer Epidemiology Biomarkers and Prevention 8 (12): 1071-8.
Zhu, K.; Levine, R.S.; Brann, E.A.; Hall, H.I.; Caplan, L.S. & Gnepp, D.R. 2002. Case-control study evaluating the homogeneity and heterogeneity of risk factors between sinonasal and nasopharyngeal cancers. International Journal of Cancer Research 99 (1): 119-23.
Zhu, L.X.; Sharma, S.; Stolina, M.; Gardner, B.; Roth, M.D.; Tashkin, D.P. & Dubinett, S.M. 2000. Delta-9-tetrahydrocannabinol inhibits antitumor immunity by a CB2 receptor-mediated; cytokine-dependent pathway. Journal of Immunology 165 (1): 373-80.
Amanda L.C. Chen, Ph.D. *; Thomas J.H. Chen, Ph.D. **; Eric R. Braverman, M.D. ***; Vanessa Acuri, B.S. ****; Mallory Kerner ****; Michael Varshavskiy ****; Dasha Braverman, P.A. *****; William B. Downs, B.Sc. *****; Seth H. Blum, B.A. ****** Kimberly Cassel ****** & Kenneth Blum, Ph.D. *******
([dagger]) The authors are appreciative of the work, contribution and original concept of Cory A. Gordon, especially for the intensive literature search. The authors are also thankful for the statistical analysis by Manuel Martinez-Pons of the Department of Education, Brooklyn College CUNY, New York, NY.
* Associate Professor Department of Engineering and management of advanced Technology, Chang Jung Christian University, Tainan, Taiwan.
** Professor Department of Occupational Safety and Health, Chang Jung Christian University, Tainan, Taiwan; Changhua Christian Hospital, Changhua, Taiwan.
*** Clinical Assistant Professor, Department of Neurosurgery, Weill Cornell College of Medicine; Medical Director, Path Medical Foundation, New York, NY.
**** Research Associate, Department of Neurological Research, Path Medical Foundation, New York, NY.
***** Senior Research Scientist, Department of Nutrigenomics, LifeGen Inc. La Jolla, CA.
****** Research Associate, Department of Personalized Medicine, Synaptamine, Inc., San Antonio, TX.
******* Scientific Director, Path Medical Foundation, New York, NY; President & CEO, Synapatmine, Inc., San Antonio, TX; Research Professor(Adjunct), Department of Physiology & Pharmacology, Wake Forest University School Of Medicine, Winston-Salem, NC.
Please address correspondence and reprint requests to Kenneth Blum, Wake Forest University School Of Medicine, Department of Physiology & Pharmacology, Medical Center Boulevard, Winston- Salem, North Carolina 27157; email: drd2gene@aol.com
TABLE 1
Cancer Incident by Site and Gender
Marijuana x
Cigarettes
Interaction
Effect
[chi
square] p
Males, All Sites 10.35 .001
Males, Tobacco (TRC) 1.05 .31
Males, Colorectal 2.08 .15
Males, Lung .85 .36
Males, Melanoma .86 .35
Males, Prostate .62 .43
Females, All Sites 34.60 .00001
Females, Tobacco (TRC) 3.14 .00001
Females, Colorectal 3.03 .08
Females, Lung 1.08 .30
Females, Melanoma 3.14 .08
Females, Breast 23.56 .00001
Females, Cervix 8.17 .004
Males and Females, All Sites 47.52 .00001
Males and Females, Tobacco (TRC) 4.17 .04
Males and Females, Lung 2.21 .14
Main Effect for
Marijuana Only
[chi
square] p r Z
Males, All Sites 31 .00001 .31 5.81
Males, Tobacco (TRC) 48.63 .03 .39 3.63
Males, Colorectal 2.60 .27 .39 2.47
Males, Lung 40.47 .00001 .44 2.79
Males, Melanoma 4.67 .09 .51 3.87
Males, Prostate 5.78 .05 .46 2.68
Females, All Sites 36.09 .00001 .50 14.27
Females, Tobacco (TRC) 28 .00001 .99 5.90
Females, Colorectal 3.58 .17 1.0 5.14
Females, Lung 31.43 .00001 .82 4.22
Females, Melanoma 3.27 .019 .27 2.68
Females, Breast 25 .00001 .66 10.21
Females, Cervix 9.72 .008 .14 2.41
Males and Females, All Sites 36.09 .00001 .50 14.27
Males and Females, Tobacco (TRC) 76.43 .00003 .62 6.96
Males and Females, Lung 71.61 .00003 .61 5.06
Main Effect for
Cigarettes Only
[chi
square] p r Z
Males, All Sites 44 .02 -.23 -4.53
Males, Tobacco (TRC) 14.73 .001 -.86 -6.14
Males, Colorectal 8.20 .02 .11 -.74
Males, Lung 9.04 .01 1.56 4.31
Males, Melanoma 17.00 .00001 -.23 -1.95
Males, Prostate 8.90 .01 -.34 -2.12
Females, All Sites 248 .00001 -.04 -1.25
Females, Tobacco (TRC) 3.14 .08 -.64 -4.82
Females, Colorectal 1.08 .29 -.10 -.76
Females, Lung 38.38 .0001 -4.0 -4.60
Females, Melanoma 7.37 .03 -.05 -.37
Females, Breast 135 .03 .07 1.37
Females, Cervix 13.85 .001 -.07 -1.27
Males and Females, All Sites 288 .00001 -.09 -3.42
Males and Females, Tobacco (TRC) 57.86 .00003 -.75 -7.84
Males and Females, Lung 30.51 .00001 -.75 -7.84
Note: This is a reanalysis of data presented in Sidney et al. 1997.
Source: Hypothesizing that marijuana smokers are at a significantly lower risk of carcinogenicity relative to tobacco-non-marijuana smokers: evidenced based on statistical reevaluation of current literature
