ENVIRONMENTAL RESEARCH56, 131--143 (1991) Epidemiologic Investigation of a Cancer Cluster in Professional Football Players ALLEN KRAtrr,* EVA CHAN,'~ PAUL J. LIov,$ FREDRICK B. COHEN,§ BERNARD O. GOLDSTEIN,:~ AND PHILIP J. LANDRIGANt *Departments of Internal Medicine and Community Health Sciences, University of Manitoba, NA--618 700 McDermot Avenue, Winnipeg, Canada R3E OW3; tDivision of Environmental and Occupational Medicine, Department of Community Medicine, Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1057, New York, New York 10029; ~:Department of Environmental and Community Medicine and the Environmental and Occupational Health Sciences Institute, UMDNJ--Robert Wood Johnson Medical School, 675 Hoes Lane, Piscataway, New Jersey 08854-5635; and §Department of Oncology, Beth Israel Medical Center, 201 Lyons Avenue, Newark, New Jersey 07112 Received June 19, 1991 In 1976, the New York Giants professional football team relocated to the newly con- structed Meadowlands Sports Complex (MSC) in East Rutherford, NJ. Between 1980 and 1987 four team members developed cancer: one case each of non-Hodgkin's lymphoma, glioblastoma, angiosarcoma, and Hodgkin's disease. Because the surrounding area contains three superfund sites, concern was widespread that the cancers were related to environ- mental contamination. To assess for a possible environmental etiology, we conducted clin- ical, environmental, and epidemiologic studies at the MSC. Measurements of volatile or- ganic compounds were all below occupational exposure limits and were similar to ambient levels in nearby Lyndhurst, NJ. Outdoor AM radio broadcast field strengths were in the uppermost 0.1% of field strengths measured in urban areas of the United States. Propor- tionate mortality ratio and proportional cancer incidence ratio studies of the MSC workforce found no excesses of cancer deaths or of incident cancer cases either for all sites combined or for any specific site. No significant differences in cancer incidence or mortality were found between indoor and nonindoor workers. Based on examination of all available data, the four cancer cases were judged most likely to have been clustered by chance and not to have been caused by environmental conditions at the MSC. © 1991 Academic Press,Inc. INTRODUCTION Cancer clusters, groups of cancer cases that occur together in place and time, pose a major dilemma in public health (Bender et al., 1990; Neutra, 1990). Clus- ters have in some instances provided the first warning of newly emerging prob- lems in environmental carcinogenesis; examples have included mesothelioma in asbestos-exposed individuals (Wagner et al., 1960), angiosarcoma of the liver in vinyl chloride polymerization workers (Creech and Johnston, 1974), and lung cancer in chemical workers producing bis-chloromethyl ether (Figueroa et al., 1973). Many other reported clusters have, however, had no discernible common etiology or have been shown on further study to contain no more cancer cases than would be expected in the general population (Greenberg et al., 1985). Cancer clusters require evaluation. They must be studied in sufficient detail to permit identification of those with a common etiology and thus to distinguish true clusters from statistically random events. Evaluation must additionally be con- 131 0013-9351/91 $3.00 Copyright© 1991 by Academic Press,Inc. All rightsof reproduction in any formreserved.