The University of MichiganNews Services
The University Record Online
Updated 10:00 AM October 16, 2006




view events

submit events

UM employment

police beat
regents round-up
research reporter


Advertise with Record

contact us
meet the staff
contact us
contact us

Study fails to find link between marijuana use and cancer

Although marijuana smoke is known to contain carcinogenic agents, and many studies in humans, animals and cell cultures have indicated that marijuana smoking may predispose a person to cancer, a large epidemiologic study of cancer patients and cancer-free controls in Los Angeles has found no clear association between pot-smoking and cancer.

"We didn't find any evidence for an increased risk of head and neck or lung cancers among heavy, long-term marijuana users," says study leader Hal Morgenstern, chair of epidemiology at the School of Public Health. "I wouldn't go so far as to say there is no increased cancer risk from smoking marijuana, but if there is an effect, it is not very large."

The study, which involved colleagues at the University of California, Los Angeles and the University of Southern California, zeroed in on head and neck cancers and lung cancer because they typically are associated with tobacco smoking.

The team interviewed more than 1,200 Los Angeles-area cancer patients and more than 1,000 cancer-free control subjects, matched by age, gender and neighborhood. Each subject's reported use of marijuana was measured carefully and quantified as "joint-years" of use, where one joint-year is equivalent to smoking one marijuana cigarette per day for one year.

Users who had accumulated more than 30 joint-years had a higher cancer incidence, but this association disappeared when the findings were adjusted for other risk factors, including tobacco use.

"The association of these cancers with marijuana use, even long-term or heavy use, is not strong and may be too small to detect," Morgenstern says.

As an offshoot of this study, UCLA epidemiologist Zuo-Feng Zhang also looked at possible genetic differences between the cancer patients and the controls and found no clear pattern that would indicate that smoking marijuana had more risk for some people than for others.

Though there is ample biological evidence that marijuana smoke should be harmful to humans, earlier epidemiologic studies of its cancer-causing potential had found mixed results, Morgenstern says.

What's been lacking is a scientifically rigorous study involving a large number of middle-aged or older adults with heavy, long-term exposures to marijuana. Because heavy marijuana use is a relatively new cultural phenomenon that started around 1970 among persons under age 25, this sort of study hasn't been possible until recently. Most of the subjects in this study were over 45, the time in life when the risk of lung and head and neck cancers becomes appreciable.

"The negative results reported here underscore the importance of conducting well-designed, population-based epidemiologic studies to assess health risks," says study co-author Dr. Donald Tashkin, a pulmonary medicine specialist at UCLA. "Several clinical reports have indicated an unusually high proportion of marijuana smokers among young individuals diagnosed with lung and head and neck cancer. But while this was interpreted as suggesting an association between marijuana and respiratory cancer, such findings are uncontrolled and require rigorously conducted epidemiologic studies for confirmation."

Though this study was more rigorous than those attempted previously, "we did have some concerns that the subjects might not accurately recall how much marijuana they smoked many years ago," Morgenstern says. "We expected some underreporting, but what we actually found was in line with national and California studies on drug use." Eleven percent of the control group had accumulated 10 or more joint-years (equivalent to 3,650 or more joints).

Another source of difficulty in the study is that marijuana potency is much less standardized than the tobacco in cigarettes, and it is smoked in many different ways, so quantities and dosages are much harder to pin down with a measure like joint-years. Tobacco studies long have relied on a measure called "pack years," in which scientists can rely on very specific content analysis of each brand the user smoked to calculate dosages.

Despite these limitations, Morgenstern says the results did not suggest any harmful effects of marijuana use, using different methods of analysis. Pot-smoking did not appear to increase the risk of cancer in this population, regardless of how much tobacco individuals had smoked in their lifetimes.

The study, "Marijuana Use and the Risk of Lung and Upper-aerodigestive-tract Cancers: Results of a Population-based Case-control Study," was funded by the National Institute on Drug Abuse and appears in the October issue of the journal Cancer Epidemiology, Biomarkers and Prevention.

More Stories