ISEE-116 Introduction: A dramatic increase in childhood leukaemia has recently occurred in Churchill County, a rural area of about 25,000 people in western Nevada, USA. Aim: The purpose of this paper is to describe the unique nature of the cluster and the likelihood that the cluster could be due to chance. Methods: From 1999 to 2001, ten cases of acute lymphocytic leukaemia (ALL) and one case of acute myelocytic leukaemia were diagnosed in this area. An additional five cases of ALL were identified from the years 1997 to 2002 among children who were not residents at the time of diagnosis, but lived in Churchill County at some point prior to diagnosis. In the preceding twenty years, only one case of childhood leukemia was reported to the Nevada Central Cancer Registry, among Churchill County residents. We have very recently learned that one new case of ALL was diagnosed in Churchill County in 2004. The primary economic sources in this area are agriculture and the nearby Fallon Naval Air Station. Despite intensive efforts by local and national investigators, a clear cause of the cluster has not been identified. Results: Although elevated levels of arsenic, tungsten, and DDE were found community wide, no differences were found between case subjects or family members and randomly selected controls. The possibility that the cluster may be due to chance has been hypothesized. However, using national rates of leukaemia for comparison, we calculated an age standardized rate ratio (RR) for Churchill County for the years 1999 to 2001 of 12.0 [95% confidence interval (CI), 6.0-21.4; n = 11 cases]. This corresponds to a probability of 4.3 x 10-9, or about 1 in 232 million. Based on the population of Churchill County relative to the US population, a cluster of this magnitude would be expected to occur in the US by chance about once every 22,000 years. In comparison, the p -value for the cancer cluster in Woburn, Massachusetts, was 8.4 x 10-3, or about one in 120. Routine tests of statistical significance may not be appropriate in the post-hoc analysis of clusters since the specific location and time frame being assessed is defined by the cluster, not set a priori. However, an important feature of the Churchill County cluster was that the cluster continued after state authorities began their investigation and the rate ratio remained high after this time (RR = 11.2; 95% CI, 3.6-26.3). Conclusion: The dramatic rise in the number of cases, the short time frame in which the cases were diagnosed, and the small population of the source area, highlight the highly unusual nature of this cancer cluster and suggest that this group of cases is one of the most remarkable childhood cancer clusters ever reported.
Read full abstract