Marijuana use is legal in two states and additional states are considering legalization. Approximately 18 million Americans are current marijuana users. There is currently no consensus on whether marijuana use is associated with cancer risk. Our objective is to review the epidemiologic studies on this possible association. We identified 34 epidemiologic studies on upper aerodigestive tract cancers (n = 11), lung cancer (n = 6), testicular cancer (n = 3), childhood cancers (n = 6), all cancers (n = 1), anal cancer (n = 1), penile cancer (n = 1), non-Hodgkin lymphoma (n = 2), malignant primary gliomas (n = 1), bladder cancer (n = 1), and Kaposi sarcoma (n = 1). Studies on head and neck cancer reported increased and decreased risks, possibly because there is no association, or because risks differ by human papillomavirus status or geographic differences. The lung cancer studies largely appear not to support an association with marijuana use, possibly because of the smaller amounts of marijuana regularly smoked compared with tobacco. Three testicular cancer case-control studies reported increased risks with marijuana use [summary ORs, 1.56; 95% confidence interval (CI), 1.09-2.23 for higher frequency and 1.50 (95% CI, 1.08-2.09) for ≥10 years]. For other cancer sites, there is still insufficient data to make any conclusions. Considering that marijuana use may change due to legalization, well-designed studies on marijuana use and cancer are warranted.