The health risks of low-dose ionizing radiation Abel Russ Community-Based Hazard Management Program George Perkins Marsh Institute Clark University Radiation Basics Ionizing radiation is an emission of energy than can create ions Radiation Basics Radiation Basics A cell that is “hit” by radiation reacts to the event in several ways: The cell has defense mechanisms that are triggered- Cell-cycle checkpoints Cell death Repair machinery The hit cell interacts with nearby cells (bystander effect), The hit cell may also produce unstable daughter cells as it divides (genomic instability) Ultimately, affected cells may be altered and may begin to behave differently. Radiation Basics • Cells that grow out of control become cancers • leukemia/lymphoma and “solid cancer” • Noncancer effects of radiation include blood disorders, circulatory problems, thyroid problems, liver disease, others • The magnitude of these effects depends on the radiation dose and other variables Radiation Basics Dose units are confusing! • The mean dose received by the atomic bomb survivors was 0.2 Sv • A “low dose” of radiation is generally thought of as less than 0.1 Sv • The average nuclear worker is exposed to 0.02 Sv • Average background exposure to radiation is ~ 0.003 Sv per year Atomic bomb survivors • The main source of data used to estimate radiation risk • ~100,000 people in Hiroshima and Nagasaki • 440 solid cancer deaths 1950-1997 • 87 leukemia deaths 1950-1990 • 472 noncancer deaths 1950-1997 Atomic bomb survivors/ Epidemiology basics Between 1950 and 1997 there were 9,335 deaths from solid cancer. Based on background rates only 8,895 were expected. Relative Risk = 9,335/8,895 = 1.05 Excess Relative Risk = 0.05 Atomic bomb survivors/ Epidemiology basics Preston DL et al. 2003. Studies of mortality of atomic bomb survivors. Report 13. Radiat Res 381-407. Low-dose risks Effective dose from 1 CT scan (12 mSv) Brenner DJ et al. 2003. Cancer risks attributable to low doses of ionizing radiation: Assessing what we really know. Proc Natl Acad Sci USA 100(24):13761-6. Low-dose risks 2 Nuclear workers Techa River cohort 1.6 ERR/Sv (SE) 1.2 0-50 mSv 0.8 0-150 mSv 0-200 mSv 0-100 mSv 0-125 mSv 0.4 0-500 mSv 0 0 20 40 60 80 100 mean dose (mSv) Estimated solid cancer mortality risk coefficient over increasing ranges of dose (data of Preston et al. 2003). Low-dose risks Recent data from nuclear workers and Techa River cohort Low-dose atomic bomb Nuclear workers Techa River survivors (Preston et al. 2003) (Cardis et al. 2005) (Krestinina et al. 2005) Mean dose 0.02 Sv1 (colon) 0.02 Sv (colon) 0.03 Gy (stomach) ERR for 0.93/Sv 0.87/Sv 0.92/Gy solid cancer (SE 0.85) (0.0-1.9) (0.2-1.7) 1This group received colon doses of 0-0.05 Sv. Groups with higher risks Women and children 3 2 ERR per Sv 1 0 0-9 10-19 20-39 40+ age at exposure (years) male female Other sources of epidemiological information •Nuclear workers •Techa River residents •People exposed to medical radiation •Nuclear test site downwinders Children’s risk • Many examples of childhood sensitivity to radiation •Exposures to x-rays in the womb: • Significant childhood cancer risks at doses as low as 0.006 Sv • ERR 30 to 50/Sv • Childhood leukemia near the Nevada Test Site and Chernobyl Women’s risk Solid cancer mortality in Techa River residents (Krestinina 2005): • ERR 0.6/Gy (men), 1.2/Gy (women) Solid cancer mortality among nuclear workers from three countries (Cardis et al. 1995): • ERR for men- -0.07/Sv (90%CI -0.4-0.3) • ERR for women- 0.97/Sv (90%CI <0.9-8.2) Leukemia risks 25 20 15 ERR/Sv 10 5 0 Techa River Nuclear workers Atomic bomb survivors Noncancer risks Circulatory disease mortality: Atomic bomb survivors: ~165 deaths associated with exposure 1.04 1.03 Relative Risk at 20 mSv 1.02 1.01 1.00 atomic bomb atomic bomb 3-country workers Chernobyl workers survivors (heart survivors (stroke) disease) Noncancer risks: exposure age and disease incidence 2 thyroid disease RR at 1 SV liver disease uterine myoma cataract hypertension 1.5 myocardial infarction 1 10 25 40 age at exposure National Academy of Sciences: BEIR VII Reaffirms that the linear, no threshold model for radiation and cancer is the most appropriate available model. ATSDR Public Health Assessments are inconsistent with this idea by supporting the use of a threshold. Rejects the idea of hormesis for purposes of assessing radiation risks Uses a DDREF (dose- and dose-rate effectiveness factor) to reduce risk estimates below the linear model predictions for low levels of exposure; risk estimates are lower than those produced by the UN Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), which does not use this factor. Health Effects of Ionizing Radiation: Half-time review • Radiation increases the risk of cancer and other disease • There is no “safe” dose of radiation • Risk is related to dose, although simple extrapolations from high doses may underestimate risk • Women and children are relatively sensitive to radiation Communities near nuclear facilities Maximum allowable doses are less than average background radiation exposures Community members are routinely exposed to many carcinogens Individual exposure information is not available If exposures are kept within allowable limits, we should not expect to see clear evidence of a health impact Areas at risk Studies often count cancer cases within a certain distance from a facility or in a convenient administrative unit (county) Black RJ et al. 1994. Leukemia and non-Hodgkin’s lymphoma: incidence in children and young adults resident in the Dounreay area of Caithness, Scotland in 1968-91. J Epidemiol Community Health 48:232-6. Rocky Flats Rocky Flats Significant correlations between soil plutonium and total cancer and radiosensitive cancer in males for both periods Significant correlations between soil plutonium and total cancer and radiosensitive cancer in females for the 1979-1981 period Cancer incidence significantly elevated closer to Denver; the association between soil plutonium and cancer incidence diminishes when controlling for this effect Crump et al. 1987. Childhood leukemia: US The National Cancer Institute conducted a nationwide survey of cancer data at the county level near nuclear facilities. • 52 nuclear power plants • 10 DOE sites Incidence data were only available for counties in Connecticut and Iowa. In the four eligible counties there were 81 cases of leukemia under age 10 diagnosed between the date of plant start-up and 1984. The incidence ratio, 1.36, was significantly elevated (p<0.01). There were 2,572 deaths from leukemia under age 10 in counties with a nuclear facility. This represents a relative risk of 1.03 compared to control counties (confidence information not provided). Jablon et al. 1991. LLNL workers and melanoma Excess melanoma between 1969-1980 related to: •Occupational exposures (chemicals and ionizing radiation) •Other factors (sun exposure, skin tone, etc.) Odds Ratio 2.3 (95%CI 1.0-7.6) Atomic bomb survivors Relative Risk 3.1 (95%CI 0.7-13.5) Melanoma excess declining over time Data from Los Alamos National Lab are not consistent with such a large excess (SIR 105, 95%CI 51-198) Precautionary principle The burden of proof should fall on the proponent of an activity and not the public “Proof” vs. “reasonable cause of concern” The decision-making process should be open and democratic and should involve affected parties The full range of alternatives should be considered, including no action Transgenerational effects of radiation • A 1984 government survey found 6 childhood leukemia deaths in Seascale between 1968 and 1978 (1.4 expected). A formal case-control study was published in 1990. • LNHL before age 25, 1950-1985 • 52 cases of leukemia, 22 cases of non-Hodgkin’s lymphoma, 1001 controls • Children of radiation workers employed at conception showed a RR of 1.48 (0.59-3.75) for LNHL MJ Gardner et al. 1990. Results of case-control study of leukaemia and lymphoma among young people near Sellafield nuclear plant in West Cumbria. BMJ 300:423-9. Challenges to the Gardner hypothesis • Unlikely biological mechanism; childhood cancer is not known to have a strong heritable component • No leukemia in the children of atomic bomb survivors • Other explanations for the West Cumbria cluster including a virus spread during population mixing • No evidence of similar leukemia clusters in other communities R Doll et al. 1994. Paternal exposure not to blame. Nature 367:678-80. MP Little et al. 1995. A review of the risks of leukemia in relation to parental pre-conception exposure to radiation. Health Phys 68(3):299-310. JD Boice Jr. et al. 2003. Genetic effects of radiotherapy for childhood cancer. Health Phys 85(1):65-80. Spermatogenesis Spermatocyte stage is proficient in the repair of DNA damage; spermatozoa stage is deficient K Shiraishi et al. 2002. Persistent induction of somatic reversions of the pink-eyed unstable mutation in F1 mice born to fathers irradiated at the spermatozoa stage. Radiat Res 157:661-7. Pink-eyed Jackson mice • Induced reversion of unstable allele is visible as black coat spots or retinal spots • Male mice were exposed to x-rays and mated immediately or 15 weeks later • After irradiation of spermatozoa (1-6 Gy): • 1.8-fold increase in mutations at paternal allele • 1.5-fold increase in mutations at maternal allele • Eye spots of all sizes were found; irradiation did not determine timing of mutation K Shiraishi et al. 2002. Persistent induction of somatic reversions of the pink-eyed unstable mutation in F1 mice born to fathers irradiated at the spermatozoa stage. Radiat Res 157:661-7. Atomic bomb survivors • 31,150 children of exposed parents (16 cases) • 41,066 controls (17 cases) • Rate Ratio for childhood leukemia 1.2 (0.6-2.5) • Fewer than 2% of the F1 cohort were conceived within 6 months of the bombings and roughly half of these parents were men Y Yoshimoto et al. 1990. Malignant tumors during the first 2 decades of life in the offspring of the atomic bomb survivors. American Journal of Human Genetics 46:1041-52. Paternal preconceptional x-ray exams Childhood leukemia in Northeastern US 1959-62; any x-rays before conception OR 1.3 (0.9-1.9) Childhood leukemia in urban Shanghai 1974-86; any x-rays before conception OR 2.2 (1.5-3.3) Infant leukemia in Children’s Cancer Group (US and Canada) 1983-88; any OR 1.3 (0.5-3.5) x-rays within 1 year of conception Childhood LNHL in German Childhood Cancer Registry 1992-94; any x-rays OR 1.3 (1.1-1.6) within 2 years of conception Combined estimate OR 1.4 (1.2-1.7) S Graham et al. 1966. Preconception, intrauterine, and postnatal irradiation as related to leukemia. National Cancer Institute Monographs 19:347-71. XO Shu et al. 1988. A population-based case-control study of childhood leukemia in Shanghai. Cancer 62:635-44. XO Shu et al. 1994. Association of paternal diagnostic x-ray exposure with risk of infant leukemia. Cancer Epidemiology, Biomarkers and Prevention 3:645-53. R Meinert et al. 1999. Associations between childhood cancer and ionizing radiation: results of a population-based case-control study in Germany. Cancer Epidemiology, Biomarkers and Prevention 8:793-9. Conclusions Epidemiologic studies of preconceptional exposure tend to support the association with childhood leukemia risk Animal studies suggest new mechanisms that might explain this association: • Genomic instability originating in late-stage spermatids or spermatozoa • Radiation-induced sensitivity to second mutagenic exposure Preconceptional radiation exposure and other etiologies are not mutually exclusive
"The health risks of low-dose ionizing radiation"