This document contains text automatically extracted from a PDF or image file.

Formatting may have

been lost and not all text may have been recognized.
To remove this note, right-click and select "Delete table".
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
EDITORIALS Does Marijuana or Crack Cocaine Cause Cancer?
Epidemiologie studies ( I ) have indicated a strong association between tobacco smoking and various human Cancers, especially
lung cancer. Nearly one third of human cancers are estimated to be related to tobacco use. Direct molecular evidence has been
provided in a study (2) demonstrating that carcinogens in to-
bacco smoke cause a specific type of genetic damage required for cancer development. Fortunately, national anti-cigarette-
smoking efforts are expected to reduce the number of teenaged smokers in the next decade and to decrease rates of tobacco-
related cancer mortality. However, the number of marijuana smokers is increasing. A national survey by the Department of
Health and Human Services (3) indicated that 1.9% of U.S. high school students smoked marijuana daily in 1992 but that this
frequency had increased to 4.6% in 1995. An independent na-
tional survey by the Partnership for a Drug-Free America ( 4 ) has found that the willingness to experiment with marijuana by
chil-
dren 9-12 years o1d has risen from 9% in 1993 to 14% in 1997. Marijuana use is becoming a serious social problem in the United
States and may become more widespread because of nationwide anti-tobacCo campaigns that may have the unin-
tended consequenee that teenaged smokers will replace tobacco with marijuana. The prevalence of crack cocaine use among
teenagers continues at 8%, and newer easier methods of cocaine consumption continue to be devised (4). Moreover, the public
cannot be informed of the potential danger of increased risk for cancer development because of the lack of scientific evidence,
yet many known carcinogens exist in marijuana smoke and crack cocaine smoke. In an elegant study reported recently by
Fligiel et al. (5), histopathologic abnormalities such as hyperplasia and squamous cell metaplasia were found substantially more
frequently in the lungs of marijuana smokers and crack cocaine smokers than in the lungs of nonsmokers. The frequency of
histopathologic ab-
normalities was almost identical between smokers of marijuana or cocaine and smokers of cigarettes, suggesting that smokers of
all of these substances experienced similar upper airway dam-
age. It is interesting that those individuals who smoked both marijuana and tobacco or both cocaine and tobacco had an even
higher frequency of histopathologic abnormalities in their bron-
chial epithelium (5). Thus, there may be synergistic effects of marijuana or crack cocaine with tobacco that promote carcino-
genesis. In this issue of the Journal, Barsky et al. (6) report on, to our knowledge, the first work to demonstrate that smokers of
marijuana and crack cocaine have an increased frequency of molecular abnormalities in bronchial epithelium that are similar to
those identified in cigarette smokers. The results of these
studies corroborate the hypothesis that smoking marijuana and crack cocaine may cause cancer in the upper airway. It is believed
that cancer develops in a multistep process through the accumulation of genetic alterations (7). Carcinogen exposure can
damage the human genome and accelerate this accumulation, leading to malignant transformation of cells in the exposed area.
In the upper airway, cigarette smoking has been strongly associated with severe histopathologic abnormalities (8,9), growth
disregulation (10.11), and genetic alterations (12-
14) in epithelium. Molecular or genetic alterations could even be identified in histologically normal epithelium of cigarette smok-
ers (13,14). These findings support the so-called "field defect” or “field cancerization” hypothesis, i.e., that epithelial surfaces
chronically exposed to carcinogens share common susceptibility to tumorigenesis. Barsky et al. (6) studied several molecular
markers, including a cell proliferation marker (Ki-67), a growth factor receptor (EGFR), a tumor suppressor protein (p53),
and DNA ploidy. Abnormality of these markers was previously associated with tumor progression and was also found in the
bronchial epithe-
lium and oral epithelium of current and former cigarette smokers Some of the abnormalities might be reversible upon cessation of
cigarette smoking (II). Similarly, many of the his-
topathologic changes in epithelium of the upper airway may be reversible after smoking cessation. Lee et al. (15) have shown that
the number of bronchial metaplastic lesions was substan-
tially lower in individuals who had not smoked cigarettes for 6 months. In an earlier study (9), severely dysplastic vocal cord
lesions were identified in a substantial proportion of current smokers but were not seen in former smokers, suggesting that some
of these lesions are reversible. However, more recent stud-
ies ( I 6,1 7) have reported that more than half of the patients with lung cancer in the United States are former smokers; former
smokers may remain at high risk for lung cancer development. This notion is further supported by the persistence of genetic
abnormalities (including loss of genomic material at Critical
*Aßîliation of authors: Molecular Biology Laboratory', Department of Tho-
racic/Head and Neck Medical Oncology. The University' of Texas M. D. Ander-
son Cancer Center. Houston. Correspondence I0.' Li Mao, M.D., Molecular Biology Laboratory, Depart-
ment of Thoracic/Head and Neck Medical Oncology, The University of Texan M. D. Anderson Cancer Center, Holcombe Blvd..
Houston, TX 77030 (e-mail: lmao@notes.mdacc.tmc.edu). See "Note" following “References."
© Oxford University Press
1182 EDITORIALS Journal ofthe National Cancer Institute, Vol. 90, No. 16, August 19, 1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

chromosomal regions containing tumor suppressor genes) in bronchial epithelium after cessation of cigarette smoking ( 13 )_
Therefore, genetic analyses to identify alterations that are re-
quired for developing epithelial cancers are crucial to address further the issue of carcinogen-specific genetic damage in smok-
ers of marijuana and crack cocaine. A body of circumstantial evidence already exists to implicate marijuana in the etiology of
aerodigestive tract cancers. Mari-
juana contains many known carcinogens, such as benz[a]anthra-
cene and benzo[a]pyrene, at concentrations even higher than those found in cigarettes. Donald (I8) provided anecdotal evi-
dence of a group of six patients with head and neck carcinoma (19-38 years old) who were habitual marijuana users. Although
four of these six patients also smoked cigarettes, the remaining two never smoked cigarettes and did not drink alcohol. In an-
other study reported by Raub et al. (19), among 19 young (S45 years old) patients with lung cancer, 68% (13 of l9) were ha-
bitual marijuana smokers, whereas only 6% (six of 95) of older years old) patients with lung cancer were habitual mari-
juana smokers. These observations suggest that marijuana smok-
ing plays a role in carcinogenesis either alone or with other factors such as cigarette smoking, especially in the development of
cancers of the upper airway in younger adults. A recent epidemiologic study, however, does not suppoń an association berween
marijuana smoking and oral or lung can-
cers. Sidney et al. (20) undertook an epidemiologic study in-
cluding almost 65 O00 people in California to evaluate whether marijuana users are at an increased risk of cancer. This study did
not find an association between marijuana use and total inci-
dence of cancers at all sites or between marijuana use and to-
bacco-related cancers except for prostate and cervical cancers. However, the study’s design had two critical defects. First„ the
criterion for regular marijuana smoking was more than six uses, and no subgroup analysis was done on the basis of smoking
duration. Since cancer development requires that individuals be exposed to carcinogens for many years, the short-term or infre-
quent exposure measured in the study may not strongly intiu-
ence the risk of cancer development. Second, smoking history of the study participants was obtained only once, at an average of
8.6 years before cancer incidence was analyzed. Therefore, it was unknown whether the smoking status had changed for these
individuals during the rather long follow-up period. Therefore, the data reported in this study should not be deemed conclusive.
Carefully designed retrospective and prospective epidemiologic studies are absolutely necessary to determine the relationship
between marijuana and cocaine use and cancer risk. It has taken more than a half century for scientists to prove that tobacco
causes human cancers and to provide sufficient evidence to convince the public. lt may be even harder to study epidemiologic
associations between drug use and cancer risk because of the difficulty in obtaining reliable information and follow-up from
users of illicit drugs. It is precisely for this reason, however. that any potential role of marijuana or crack cocaine use in cancer
development must be analyzed through extensive scientific investigation and then publicized. These ef-
forts may not only help prevent cancers, such as those of the mouth and lung in a young adult population, but also may serve as
an added deterrent to the use of marijuana and cocaine. Ex-
trapolating from findings that former cigarette smokers are still
Journal ofthe National Cancer Institut@ Vul. 90, No. 16, August 19, 1998
at high risk to develop lung cancer (I 7,18), we can conjecture that former marijuana smokers and cocaine smokers might retain
a certain risk for cancer development even after they stop using the drugs. Privacy is the key in obtaining accurate information
about illicit drug use. It may be possible to develop comprehen-
sive database software with the use of encryption technology that would allow patients to enter such information privately and
anonymously. Clinical information can be added separately with or without allowing researchers to identify individual patients.
lf indeed marijuana and cocaine are found to cause cancer, the research community is challenged to provide definitive evidence
of this etiologic relationship. The efforts aimed at protecting the public from cancer may now also discourage the growing trends
toward marijuana and cocaine use in a teenaged and young adult population.
References
19961274430-2. (3) National survey results on drug use from the monitoring the future study, 1975-1995, U.S. Department
of Health and Human Services, vol. 1. Bethesda (MD): National Institutes of Health; 1996. DHHS Publ N0. (NIH)96-4139.
(4 ) The partnership attitude tracking study (PATS)~by the Partnership for a Drug-Free America. Washington (DC): US
Information Agency, 1998 [electmnic publicationl. (5) Fligiel SE, Roth MD, Kleerup EC. Barsky SH, Simmons MS, Tashkin DP.
Tracheobronchia1 histnpathology in habitual smokers of cocaine. mari-
juana. and/or tobacco. Chest 199711 12:319-261. (6) Barsky SH, Roth MD, Kleerup EC, Simmons M, Tashkin DP. Hislopath-
ologic and molecular alterations in bronchial epithelium in habitua1 smok-
ers Of marijuana, cocaine, and/or tobacco. I Natl Cancer Inst 1998;90: ll98-1205. (7) Vogelstein B, Fearon ER, Hamilton
SR, Kern SE, Preisinger AC. Leppert M, et al. Genetic alterations during colorectal-tumor development. N Engl J Med
198813191525-32. (8) Auerbach O, Stout AP, Hammond EC. Garfinkel L, Changes in bronchial epithelium in relation to
cigarette smoking and in relation to lung cancer. N Engl J Med l96l;265:253-67, (9) Auerbach O. Hammond EC. Garfinkel
L. Histologie changes in the larynx in relation to smoking habits. Cancer l970;25:92~l04. (I0) Shin DM. Voravud N, R0 JY,
Lee JS, Hong WK, Hittelman WN. Sequen» tial increases in proliferating cel] nuclear antigen expression in head and neck
tumorigenesis: a potential biomarker. J Natl Cancer Inst l993;85: 971-8. (Il) Kurie JM, Shin HJ, Lee JS, Morice RC, Ro JY.
Lippman SM. et al. Increased epidermal growth factor receptor expression in metaplastic bron» chìal epithelium, Clin Cancer
Res 199612: l787»93. (I2) Mao I., Lee JS, Fan YH, R0 IY. Batsakis JG. Lippman S. et al, Frequent mìcrosatellile alterations at
chromosomes 9p21 and 3p14 in oral premalig-
nant lesions and their value in cancer risk assessment. Nal Med l996;2: 682-5. (l5) Mao L. Lee JS, Kuríe IM. Fan YH. Líppman
SM, Lee JJ, et al. Clonal genetic alterations in the lungs of current and former smokers. I Natl Cancer Inst l997;89:857-62. (I4)
Wistuba ll, Lam S, Behrens C, Virmzmi AR, Fong KM, LeRich€ J, et al. Molecular damage in the bronchial epithelium of
current and former smok-
ers. J Natl Cancer Inst (I5) Lee JS, Lippman SM, Benner SE, Lee JJ, Ro JY. Lukeman JM, el al. Randomized placebo-controlled
trial of isotretinoìn in chemnpreven-
tion of bronchial squamous metaplasìa. J Clin Oncol l994;l2: 937415.
EDITORIALS 1 1 83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

(I6) Strauss GM, Gleason RE. Sugarbaker DJ. Screening for lung cancer re-
examíned. A reinterpretatiori of the Mayo Lung Project random-
ized trial on lung cancer screening. Chest l993;l03(Suppl): 33'/S-41S. (I7) Tong L, Spitz MR, Fueger JJ, Amos CA. Lung
carcinoma in former smok-
ers. Cancer l996;78:lO04-10. (I8) Donald PJ. Marijuana smoking-possible cause of head and neck carci-
noma in young patients. Otolaryngol Head Neck Surg 1986;94: 517-21. (I9) Raub WA Jr, Sridhar KS, Duncan RC, McCoy C. Is
marijuana smoking a
cause of lung cancer at a young age in tobacco smokers? Proc Annu Meet Am Assoc Cancer Res l993;34:A1577. (20) Sidney S,
Quesenberry CP Jr. Friedman GD, Tekawa IS. Marijuana use and cancer incidence (California, United States). Cancer Causes
Control 1997; 81722-8.
Note We thank Margaret Spitz for helpful comments and J ulia Starr for editing.
Selenium, Vitamin E, and Prostate Cancer-

Ready for Prime Time? Philip R. Taylor, Demetrius Albcmes*

ln this issue of the Journal, Yoshizawa et al. (I) present findings relevant to a potential protective role for selenium in advanced
prostate cancer. This report is particularly timely in light of the dramatic rise in the incidence of prostate cancer and because We
have such limited understanding of risk factors for this disease beyond increasing age and being Afn`can«American. The study
by Yoshizawa et al. suggests a one-half to two-thirds reduc-
tion in the risk of advanced prostate cancer for men with the highest (as compared with the lowest) selenium status. This study is
dis-
tinguished by a number of strengths, including its prospective de-
sign, nearly complete follow-up, objective and time-integrated as-
sessment of status through selenium measurement in toenails, careful appraisal and control of other potential influencing factors,
and the largest number of cases studied to date. Yoshizawa et al. conclude that further prospective studies and randomized trials
of selenium and prostate cancer should be conducted. So where does this study leave us with regard to selenium and prostate
cancer? The accumulated evidence is limited and in-
consistent, but it is extremely provocative. Two earlier prospec-
tive studies (2,5) conducted in the United States among persons with selenium values in the normal range hinted at a benefit for
higher selenium levels, but these studies were based on numbers far too small (i.e., l l and 13 cases) to be considered informative.
A third prospective study (4 ), conducted in Finland, where se-
rum selenium concentrations in individuals were very low and a benefit for higher levels might be more likely, found no asso-
ciation between selenium levels and cancer risk based on 51 cases. The real enthusiasm for selenium in the prevention of prostate
Cancer, however, comes from the results of the clinical trial conducted by Clark et al. (5) in the United States among persons
with low-to-normal selenium status. In that trial, persons with a history of nonmelanoma skin cancer were ran-
domly assigned to receive selenium (200 png/day) or placebo and were followed for an average of years for the development of
basal cell or squamous cell carcinoma of the skin. Secondary analysis of the data (5) revealed that prostate cancer incidence was
reduced by two thirds among those in the selenium-
supplemented group compared with the placebo group (13 ver-
sus 35 cases, respectively).
The usual order of scientific investigation in humans involves exploring hypotheses through observational studies, with only the
more promising treatment or prevention leads meriting test-
ing in Clinical trials. As serendipity would have it, however, the most credible leads for the primary prevention of prostate
Cancer have emerged as secondary findings from randomized, con-
trolled trials, with Corroborative evidence subsequently being sought from observational studies such as the one by Yoshizawa et
al. ( I ). The other exciting recent lead in prostate cancer pre-
vention was reported by the Alpha-Toeopherol, Beta Carotene (ATBC) Cancer Prevention Study, a randomized trial conducted
in more than 29 O00 male smokers in Finland. Among the 246 new cases of prostate Cancer identified during the 5- to 8-
year intervention period, only 99 cases occurred among participants who received 50 mg of vitamin E daily compared with
147 cases among those who did not receive vitamin E. a striking one-third reduction (6,7). How might selenium and vitamin
E supplementation inhibit the development of prostate cancer? In the ATBC Study, the effect of vitamin E was limited to the
prevention of clinically evident Cancers of stages Il through IV, suggesting inhibition of the transformation of latent tumors to
more invasive disease (7). Both selenium and vitamin E can function as antioxidants (8), but whether the prevention of DNA
oxidative damage is as relevant to tumor progression as is the inhibition of cell prolif-
eration (9) or the promotion of apoptosis (IO), for example, is not known. Research aimed at identifying the cellular and mo-
lecular mechanisms operational in prostate Cancer and determin-
ing how these pathways may be modified by selenium or vita-
min E is needed. Beyond gaining a clearer understanding of underlying mecha-
nisms, we should strike While the iron is hot to resolve expedi-
tiously and definitively whether selenium or vitamin E (or both)
*Afïîliation ofauthors.' Cancer Prevention Studies Branch. Division of Clini-
cal Sciences, National Cancer Institute, Bethesda, MD. Correspondence fo.' Philip R. Taylor, MD.. Sc.D., National Institutes of
Health, 6006 Executive Blvd., Rm. 321. Bethesda. MD 208927058.
H84 EDITORIALS Journal ofthe National Cancer Institute, Vol. 90. N0. I6. August 19, 1998