Page 414 - 2014 Printable Abstract Book
P. 414
(PS7-63) Methodologic innovations in radiation dose and risk assessment: Case study of thyroid disease
1
2
following radiation exposure in Kazakhstan. Steven L. Simon, PhD ; Deukwoo Kwon, PhD ; Owen
1
3
1
Hoffman, PhD ; Brian E. Moroz, M.S. ; and Charles E. Land, PhD ; National Cancer Institute, Bethesda,
2
1
MD ; Sylvester Comprehensive Cancer Center University of Miami, Miami, FL ; and Oak Ridge Center for
Risk Analysis, Oak Ridge, TN
3
Dosimetic uncertainties that are shared among subgroups of a cohort can bias, distort, or reduce
the slope or significance of a dose response. Studies of radiation dose from environmental exposures are
often highly uncertain and, thus, susceptible to those methodologic limitations. In 2008, an analysis was
reported of radiation-related thyroid disease in a cohort of 2,994 villagers under the age of 21 years
between 1949 and 1962 and who lived downwind from the Semipalatinsk Nuclear Test Site in Kazakhstan.
The 2008 analysis found a statistically significant association between thyroid nodule prevalence and
reconstructed doses of fallout-related internal and external radiation to the thyroid gland; however, the
effects of dosimetric uncertainty were not evaluated since the doses were simple point ‘best estimates’.
In this work, we revised the 2008 study by a comprehensive treatment of dosimetric uncertainties using
a two-dimensional Monte Carlo (2DMC) dose estimation method to generate multiple sets of possibly
true doses to the entire cohort that properly account for shared and unshared errors. We analyzed the
multiple sets of doses for radiation risk by a Bayesian Model Averaging (BMA) approach. In our study, the
BMA method used 5,000 independent sets of cohort doses, each set composed of conditional individual
median internal and external dose estimates. Our findings indicated an increase in the risk per unit dose
from internal exposure by a factor of six compared to the 2008 findings and the dependency of risk on sex
increased from a previously reported factor of 11 to nearly 30. In contrast to earlier findings, we found
that the biological effectiveness for internal and external dose was similar. We believe these new results
to be more reliable because of the use of multiple sets of doses that distinguish between shared and
unshared uncertainties and the capabilities of the BMA method which was shown through extensive
simulation tests to be superior to traditional regression when complex uncertainty is present. Together,
the 2DMC and the BMA methods represent considerable innovation in dose and risk assessment methods
and the findings presented here are the first known use of the combined methods for analysis of the dose-
response.
(PS7-64) Reproductive outcomes in a cohort exposed in utero to Chernobyl fallout in Ukraine. Maureen
Hatch, NCI/NIH, Bethesda, MD
More than 15,000 pregnant women were exposed to radioactive releases from
Fukushima Daiichi, yet reproductive effects of such nuclear accidents are largely unknown. Iodine-131
131
( ), the principal component of fallout, concentrates in the thyroid gland, with potential risks to fetal
development. We investigated reproductive outcomes after the Chernobyl accident through a
retrospective review of prenatal, delivery and newborn records in a well-defined cohort of 2,532 in utero
exposed individuals from northern Ukraine, for whom individual estimates of fetal thyroid 131 dose are
available. Retrieval of archived records more than 25 years post-accident was challenging. We did,
however, obtain complete sets for 1,167 cohort members. Using registration delivery logs, we were able
to confirm that subjects without medical records were similar to those with records in respect to mean
131
gestational length and birth weight, and potential confounding factors. The relationships of fetal dose
as a categorical (zero, dose quartiles) and continuous measure with neonatal anthropometrics and length
1
2
following radiation exposure in Kazakhstan. Steven L. Simon, PhD ; Deukwoo Kwon, PhD ; Owen
1
3
1
Hoffman, PhD ; Brian E. Moroz, M.S. ; and Charles E. Land, PhD ; National Cancer Institute, Bethesda,
2
1
MD ; Sylvester Comprehensive Cancer Center University of Miami, Miami, FL ; and Oak Ridge Center for
Risk Analysis, Oak Ridge, TN
3
Dosimetic uncertainties that are shared among subgroups of a cohort can bias, distort, or reduce
the slope or significance of a dose response. Studies of radiation dose from environmental exposures are
often highly uncertain and, thus, susceptible to those methodologic limitations. In 2008, an analysis was
reported of radiation-related thyroid disease in a cohort of 2,994 villagers under the age of 21 years
between 1949 and 1962 and who lived downwind from the Semipalatinsk Nuclear Test Site in Kazakhstan.
The 2008 analysis found a statistically significant association between thyroid nodule prevalence and
reconstructed doses of fallout-related internal and external radiation to the thyroid gland; however, the
effects of dosimetric uncertainty were not evaluated since the doses were simple point ‘best estimates’.
In this work, we revised the 2008 study by a comprehensive treatment of dosimetric uncertainties using
a two-dimensional Monte Carlo (2DMC) dose estimation method to generate multiple sets of possibly
true doses to the entire cohort that properly account for shared and unshared errors. We analyzed the
multiple sets of doses for radiation risk by a Bayesian Model Averaging (BMA) approach. In our study, the
BMA method used 5,000 independent sets of cohort doses, each set composed of conditional individual
median internal and external dose estimates. Our findings indicated an increase in the risk per unit dose
from internal exposure by a factor of six compared to the 2008 findings and the dependency of risk on sex
increased from a previously reported factor of 11 to nearly 30. In contrast to earlier findings, we found
that the biological effectiveness for internal and external dose was similar. We believe these new results
to be more reliable because of the use of multiple sets of doses that distinguish between shared and
unshared uncertainties and the capabilities of the BMA method which was shown through extensive
simulation tests to be superior to traditional regression when complex uncertainty is present. Together,
the 2DMC and the BMA methods represent considerable innovation in dose and risk assessment methods
and the findings presented here are the first known use of the combined methods for analysis of the dose-
response.
(PS7-64) Reproductive outcomes in a cohort exposed in utero to Chernobyl fallout in Ukraine. Maureen
Hatch, NCI/NIH, Bethesda, MD
More than 15,000 pregnant women were exposed to radioactive releases from
Fukushima Daiichi, yet reproductive effects of such nuclear accidents are largely unknown. Iodine-131
131
( ), the principal component of fallout, concentrates in the thyroid gland, with potential risks to fetal
development. We investigated reproductive outcomes after the Chernobyl accident through a
retrospective review of prenatal, delivery and newborn records in a well-defined cohort of 2,532 in utero
exposed individuals from northern Ukraine, for whom individual estimates of fetal thyroid 131 dose are
available. Retrieval of archived records more than 25 years post-accident was challenging. We did,
however, obtain complete sets for 1,167 cohort members. Using registration delivery logs, we were able
to confirm that subjects without medical records were similar to those with records in respect to mean
131
gestational length and birth weight, and potential confounding factors. The relationships of fetal dose
as a categorical (zero, dose quartiles) and continuous measure with neonatal anthropometrics and length
