Stochastic models for radiation carcinogenesis: temporal factors and dose rate effects.
Moolgavkar SH; Curtis S; Hazelton W; Luebeck EG
Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, R01-OH-007864, 2005 Sep; :1-4
Current radiation protection standards are based largely on the experience of the cohort of A- bomb survivors. An important question, however, is whether the risks estimated in a Japanese war-time population exposed to instantaneous radiation can be transported to contemporary western populations typically exposed to protracted radiation in the workplace or elsewhere. Additional complications arise when age- and time-related factors in radiation exposure must be considered. A recent analysis, using conventional epidemiologic methods, of a large Canadian cohort of workers occupationally exposed to low-LET radiation yielded estimates of excess relative risk that were an order of magnitude higher than those estimated from the A-bomb survivors' data. Can such inconsistencies be resolved? In this research we developed and used methods based on the biological principles of multistage carcinogenesis to analyze substantial data sets and to explore the consequences of measurement error on inferences regarding radiation carcinogenesis. These methods, which complement the traditional epidemiologic approaches to data analyses, can incorporate age- and time-dependent factors, including age at start, age at stop, and protraction of exposure in a transparent way. Analyses of lung cancer incidence in the Canadian cohort referred to above using these methods shows that discrepancy between the Canadian and A-bomb data disappears when protraction is properly addressed within the framework of multistage models. In epidemiologic studies, exposures are often measured with error. These errors in measurement of exposure often bias the estimates of risk. Broadly speaking, two distinct types of measurement error are recognized: classical error and Berkson error. In this research we developed methods for correction of biases resulting from both types of error and illustrated the methods by application to an epidemiologic data set on radiation-induced lung cancer. Finally, we investigated the consequences of gestational mutations on carcinogenesis. Specifically we examined the consequences of radiation-induced mutations during gestation on subsequent cancer risk, and concluded that radiation exposure to the fetus confers the largest risk of cancer when it occurs late during pregnancy.
Carcinogens; Carcinogenicity; Carcinogenesis; Cancer; Models; Mutagens; Exposure-levels; Exposure-assessment; Radiation; Radiation-exposure; Occupational-exposure; Dosimetry; Models; Dose-response; Risk-factors; Risk-analysis; Radiation-effects; Radiation-hazards; Atomic-absorption-spectrometry; Prenatal-exposure
Suresh H. Moolgavkar, MD, PhD, Fred Hutchinson Cancer Research Center, Seattle, Washington, 98109-1024
Final Grant Report
NTIS Accession No.
National Institute for Occupational Safety and Health
Fred Hutchinson Cancer Research Center