Epidemiological findings in a brain-tumor cluster in Western Missouri

Transcription

1 J Neurosurg 62:856-86, 1985 Epidemiological findings in a brain-tumor cluster in Western Missouri ROBERT A. MORANTZ, M.D., JOHN S. NEUBERGER, DR.P.H., LARRY H. BAKER, M.D., GARY B. BERINGER, DR.P.H., ANDREW B. KAUFMAN, M.D., AND TOM D. Y. CHIN, M.D., M.P.H. Department of Neurosurgery, Research Medical Center, Kansas City, Missouri; and Department of Community Health, University of Kansas School of Medicine, Kansas City, Kansas ~" A cluster of seven primary brain neoplasms was identified in a town of population during the 1- year period from 197 to With six deaths, this represents an age-adjusted mortality rate 4.1 times greater than expected. No other neoplasms were found to be in excess of the anticipated incidence in this town. When brain-tumor mortality rates in 6 other towns of approximately the same population were calculated, only one other town was found to have an excessive rate. All seven tumors in this study were histologically verified: six were diagnosed as glioblastoma multiforme. Interviews were conducted with patients or next-of-kin to obtain the exposure histories of the patients. A number of respondents reported occupational or residential exposure either to a shoe factory or to one of the several chicken hatcheries in the town. Many of the patients ate fish from local ponds that had previously been used as coal mining strip pits. Two patients were siblings. None of the patients had a history of significant head trauma. This unique situation may provide an opportunity to learn more about environmental risk factors for brain neoplasia. Further epidemiologicai studies are planned. KEY WORDS " brain tumor 9 epidemiology 9 glioblastoma multiforme 9 environmental carcinogens p RIMARY brain tumors are generally considered to be relatively rare, accounting for approximately 1.2% of all primary neoplasmsfl 7 Mortality rates from malignant neoplasms of the brain have been increasing in the United States since the 19's. The extent to which this increase is a reflection of improved case detection, increased case fatality rates, or increased incidence is not clear. 4'5'~9 Risk factors for brain tumors of adult onset include phacomatosis, familial glioma, head injury, 9 exposure to ionizing radiation, ~6 toxoplasmosis, 2~ and exposure to specific chemical carcinogens, such as ethylnitrosoutea and vinyl chloride, j~ Living on a farm and/or exposure to farm animals has been noted as a risk factor. ~s There have also been reports of an excessive incidence of brain tumors among Swedish chemists and United States petrochemical, pharmaceutical, and rubber workers. ~ ,25 One possible approach by which hypotheses as to risk factors may be generated involves the examination of any temporal or spatial clustering of primary brain tumors. While there have been reports, for example, of geographic clusters of leukemia and Hodgkin's disease, TM there is only one previous study in which a clustering of brain-tumor patients has been reported. 2 Several years ago, we noticed that an unusually large number of patients living in a small town near Kansas City (Town W) were being referred for the surgical treatment of glioblastoma. This observation prompted us to carry out a descriptive epidemiological investigation, in which we identified all patients with primary brain tumors who resided in this town at the time of diagnosis. We then interviewed either the patients or their families. In this paper we report the results of this investigation. Clinical Material and Methods The standardized mortality ratio for primary-site brain cancer in Town W was calculated for the 197 to 1982 period using the 197 age-specific mortality rate for 1 age groups for Caucasian residents of Missouri and for the entire United States. The mortality rates for brain neoplasia and 1 other cancers were calculated 856 J. Neurosurg. / Volume 62~June, 1985

2 Epidemiology of brain-tumor cluster TABLE 1 Summary of brain-tumor cases in Town W Case Sex, Age Year of Year of Period of No. (yrs) at Diagnosis Death Residence in Diagnosis Town W (yrs) 1 M, * F, M, * M, F, F, F, * The patients in Cases 2 and 4 were siblings. TABLE 2 Mortality data from brain neoplasms in Town W residents, Factor No. actual no. of deaths 6 expected no. of deaths* 1.8 Standardized Mortality Ratiot. * Based on Missouri death rates. f Observed rate/expected rate for Missouri 1. The difference was significant (p <.1 ). for Town W using a direct age-adjustment procedure. Brain-tumor mortality rates were calculated for all towns in Missouri with a white population similar to Town W (25 to 5 population), using the indirect method of age adjustment. For all calculations, statistical significance was ascertained using a two-tailed Poisson test. Death certificates were obtained for each brain-tumor case identified in Town W, and the patient's physician or hospital was contacted. We included only patients for whom there was a histologically confirmed diagnosis (based upon a specimen obtained either at surgery or at autopsy). All pathological slides were reviewed again in a blinded fashion by a board-certified neuropathologist who was not associated with the study. Consent for the release of information was obtained from either the patient or the next-of-kin. Subsequently, an interview was conducted either directly or by telephone with that person. A written questionnaire was also given to provide additional information. These interviews concerned the patient's social, residential, and occupational history. Data concerning exposure to risk factors were subsequently tabulated for each case. Sites of possible carcinogenic exposure that had been referred to in the interviews were visited. Spot mapping of the patient's residence at the time of diagnosis was carried out. Results Six patients died from primary-site brain cancer in the 1-year period from 197 to 1982 in Town W. All the patients were Caucasian. An additional case, which was diagnosed in 1981, was found after the investigation was initiated; this patient is still alive. This gives a total occurrence of seven cases for the 197 to 1982 period. There were three males and four females. All patients had lived in Town W for a minimum of 5 years prior to diagnosis (Table 1). During this period of time, the population of Town W remained constant at approximately (of which 99% were Caucasian). Table 2 indicates the actual and expected mortality from brain malignancies in Caucasians for Town W. The expected mortality (1.8) is based on Missouri death FIG. 1. Incidence of brain tumor by year of diagnosis in residents of Town W. rates for this disease. When the data are combined for both sexes for the decade from 197 to 1982, the town's Standardized Mortality Ratio (SMR) was significantly elevated. Although not shown, the expected number of deaths for that period of time based on death rates in the United States as a whole (rather than in Missouri alone) was 1.45, which would yield an SMR of 41.8 (p <.1). When the surviving patient, who was diagnosed in 1981, is included and comparison is made to incidence of brain cancer in Caucasians based on 197 to 1977 SEER data, 26 the expected number of cases is 2.1 and the Standardized Incidence Ratio is. (p <.1). No other cancer mortality rates were significantly elevated in Caucasian residents of Town W for the 197 to 1982 period. When examining brain-tumor mortality rates in 6 other Missouri towns with similar populations, one other town had a significantly (p <.5) elevated rate. Thus, the rates in two towns were excessive, although only one town would be expected, using a two-tailed test. Town W had the largest number of deaths observed (six); the other town with a significantly elevated rate had five deaths. One town had four deaths, four towns had three deaths, and all other towns had two or fewer deaths. Figure 1 shows the distribution of the seven braintumor cases by year of diagnosis. Two cases each were diagnosed in 1974 and No cases have occurred in 198, 1984, or 1985 to date. A blinded review of all the pathological specimens by an independent neuro- J. Neurosurg. / Volume 62 / June,

3 R. A. Morantz, et al. pathologist confirmed the histological diagnosis of glioblastoma multiforme in six cases and malignant ependymoma in one case. This is almost twice the expected incidence for glioblastoma. 19 A review of hospital records indicated that there was no evidence of the presence of any of the phacomatoses among these patients. Two of the patients were siblings, but there was no other consanguinity among the rest. Possibly relevant environmental exposures of the seven patients from Town W are indicated in Table. Six of the seven patients at one time or another either lived in close proximity to a chicken hatchery (within 2 yards), or worked or had a spouse who worked in a chicken hatchery. The chicken hatchery industry has been present continuously within the town since Four of the patients had been employed in the town's major industrial plant, a shoe factory which has been in operation since 194. Four of the patients fished in the abandoned coal mining strip pits that surround the town, while five of the patients frequently ate fish caught in these ponds. The one survivor was a 28-year-old woman who had lived for 5 years in Town W. She reported living near a chicken hatchery for 1 year, but did not have any exposure to the shoe factory or strip mine pits. No patient reported a history of previous significant head trauma. Six of the patients had lived on a farm at some time in the past. There was no known previous exposure to ionizing radiation among these patients. Figure 2 indicates the homes of the patients at the time of tumor diagnosis. Three of the patients lived in close proximity to each other and to a chicken hatchery. Discussion In 1972, Brooks 2 studied all brain-tumor patients hospitalized in Kentucky from 196 to When these cases were analyzed according to county and adjusted for age, it was found that the majority came from six contiguous counties in which the incidence of gliomas was 4.4 times that expected. These counties were rural in character, with many of the men being employed in the coal mining industry. However, three other studies of patients with brain tumors in the United States, Denmark, and Newfoundland have failed to reveal any evidence of geographic clustering. 12,1.15 This study has revealed a significantly high incidence of malignant brain tumors in a town of approximately population during a 1-year period. Although the number of cases is small, and the elevated rate calculated could be due to chance, no excessive incidence was found for other malignancies in the town. Among 6 towns of similar size studied, an excess of brain tumors was found in only one other town. However, TABLE Possibly relevant exposures in seven brain-tumor cases fi~om Town W Exposure Males Females Total Cases related to a brain-cancer patient family history of brain cancer large no. of siblings (_> 4) 2 5 significant head trauma head x-ray film cigarette smoking alcohol use 1 2 used hair coloring drank coffee 2 5 drank tea lived on a farm I pesticide use shoe factory employee or spouse 2 5 worked in shoe factory chicken hatchery employee or 1 1 spouse worked in chicken hatchery lived near chicken hatchery 6 chicken hatchery employee or 6 spouse worked in chicken hatchery or lived near chicken hatchery fished in coal mining strip pits 1 4 ate fish caught in coal mining 2 5 strip pits used coal heat, or fished in coal 6 mining strip pits or spouse was a coal miner total cases 4 7 O HATCHERY C OCASES FIG. 2. Location of residence (black dots) in Town W of brain-tumor patients at time of diagnosis. 858 J. Neurosurg. / Volume 62/June, 1985

4 Epidemiology of brain-tumor cluster the pathological diagnoses of the reported cases in the other town have not been confirmed and, since death certificate data may very well be inaccurate in the case of brain tumors, whether or not this other town's elevated rate is accurate remains to be seen. The potential importance of this geographic clustering of patients with malignant brain tumors lies in the fact that it may offer us clues as to causative factors for the development of this type of neoplasm. Since this is a descriptive study, with no controls utilized, no definite statement can be made as to causation. However, certain hypotheses can be generated which can be tested in subsequent epidemiological studies. It is possible that the risk factors for the increased incidence of brain tumors in this town are related to fishing in the coal mine strip pits or working in the chicken hatcheries or shoe factory. It must be remembered, however, that these are common activities among the townspeople. Therefore, the use of analytical studies employing case-control methodology will be necessary to determine the significance of these factors. The rationale for believing that chemical carcinogens may be related to the induction of brain tumors in humans comes in part from experimental studies in animals, where brain tumors of various types can be induced by the systemic or intracerebral administration of carcinogens. For example, intracerebral implantation of methylcholanthrene 27 and many related agents into rodents has been shown to produce brain tumors. Also, when ethylnitrosourea is injected intravenously into pregnant animals in the latter half of pregnancy, many of the offspring subsequently develop brain tumors. ~J Therefore, it is possible that the pond water or certain chemicals used in the shoe factory or certain pesticides used in the chicken hatcheries were carcinogenic. Brain tumors may also be induced experimentally in a variety of species of animals by the inoculation of oncogenic viruses. At present eight deoxyribonucleic acid (DNA)-containing viruses and four ribonucleic acid (RNA)-containing viruses have been associated with the formation of central nervous system neoplasms. 2 For example, 75% of mongrel dogs injected with Rous sarcoma virus developed brain tumors, 6 while intracerebral injection of JC papovavirus into hamsters produced a variety of brain tumors including glioblastoma. 22 The avian sarcoma virus was used to induce brain tumors in at least seven animal species including chickens? Therefore, the role of oncogenic viruses as possible etiological factors should be considered. Acknowledgments We express our thanks to Drs. F. Holmes, J. C. Chang, G. Land, B. Gathright, and W. Schramm for assistance in obtaining the Missouri mortality data and for calculating the ageadjusted rates. We also thank Dr. J. Kepes for pathological review, Drs. Schoolman, Smith, Windsor, and Hanson for the inclusion of their patients in the study, and Ms. G. Hix and Ms. M. Kelly for assistance in compiling the data and preparing the manuscript. References 1. Bigner DD, Pegram CN: Virus-induced experimental brain tumors and putative associations of viruses with human brain tumors: a review, in Thompson RA, Green JR (eds): Neoplasia in the Central Nervous System. Advances in Neurology, Vol 15. New York: Raven Press, 1976, pp Brooks WH: Geographic clustering of brain tumors in Kentucky. Cancer :92-926, Chadduck WM, Netsky MG: Familial gliomas: report of four families, with chromosome studies. Neurosurgery 1: , Garfinkel L, Sarokhan B: Trends in brain cancer tumor mortality and morbidity in the United States. Ann NY Acad Sci 81:1-5, Gold EB: Epidemiology of brain tumors, in Lilienfeld AM (ed): Reviews in Cancer Epidemiology. New York: Elsevier, 198, Vol 1, pp Grove AS Jr, Di Chiro G, Rabotti GF: Experimental brain tumors, with a report of those induced in dogs by Rous sarcoma virus. J Neurosurg 26: , Halperin W, Altman R, Stemhagen A, et al: Epidemiologic investigation of clusters of leukemia and Hodgkin's disease in Rutherford, New Jersey. J Med Soe NJ 77: , Heath CW Jr, Hasterlik R J: Leukemia among children in a suburban community. Am J Med 4: , Hochberg F, Toniolo P, Cole P: Head trauma and seizures as risk factors of glioblastoma. Neurology 4: , International Agency for Research on Cancer: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Volumes I to 2, and Suppl 1. Lyon: International Agency for Research on Cancer, Koestner A, Swenberg JA, Wechsler W: Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague-Dawley rats. Am J Pathol 6:7-56, Kurtzke JF: Geographic pathology of brain tumors. Acta Neurol Scand 45:54-555, Kurtzke JF, Stazio A: Geographic distribution of brain tumors. Trans Am Neurol Assoc 92:25-254, Mancuso TF, Ciocco A, El-Attar AA: An epidemiological approach to the rubber industry. A study based on departmental experience, d Occup Med 1:21-22, Maroun FB, Jacob JC: The frequency of intracranial neoplasms in Newfoundland. Can J Public Health 64: 5-57, Modan B, Baidatz D, Mart H, et al: Radiation-induced head and neck tumours. Lancet 1: , Rubinstein L J: Tumors of the Central Nervous System. Atlas of Tumor Pathology, Series 2, Fascicle 6. Washington, DC: Armed Forces Institute of Pathology, 1972, p Schoenberg BS: Nervous system, in Schottenfeld D, Fraumeni JF Jr (eds): Cancer Epidemiology and Prevention. Philadelphia: WB Saunders, 1982, pp Schoenberg BS, Christine BW, Whisnant JP: The descriptive epidemiology of primary intracranial neoplasms: the Connecticut experience. Am J Epidemiol 14:499-51, Schuman LM, Choi NW, Gullen WH: Relationship of central nervous system neoplasms to Toxoplasma gondii infection. Am J Public Health 57: , 1967 J. Neurosurg. / Volume 62/June,

5 R. A. Morantz, et al. 21. Selikoff I J, Hammond EC (eds): Brain Tumors in the Chemical Industry. Annals of the New York Academy of Science, Vol 81. New York: New York Academy of Science, Walker DL, Padgett BL, ZuRhein GM, et al: Human papovavirus (JC): induction of brain tumors in hamsters. Science 181: , Walsh JW, Zimmer SG, Perdue ML: Role of viruses in the induction of primary intracranial tumors. Neurosurgery 1:64-662, Waxweiler R J, Alexander V, Leffingwell SS, et ah Mortality from brain tumor and other causes in a cohort of petrochemical workers. JNCI 7:75-81, Waxweiler R J, Stringer W, Wagoner JK, et al: Neoplastic risk among workers exposed to vinyl chloride. Ann NY Acad Sci 271:4-48, Young JL, Percy CL, Asire A J, et ah Cancer incidence and mortality in the United States, Natl Cancer Inst Monogr 57:1-188, Zimmerman HM, Arnold H: Experimental brain tumors. I. Tumors produced with methylcholanthrene. Cancer Res 1:919-98, 1941 Manuscript received April 12, Accepted in final form December 19, This paper was presented in part at the Annual Meeting of the American Association of Neurological Surgeons, San Francisco, California, on April 1, Address reprint requests to: Robert A. Morantz, M.D., 642 Prospect, Suite T411, Kansas City, Missouri J. Neurosurg. / Volume 62/June, 1985