Lung cancer mortality in the European uranium miners cohorts analyzed with a biologically based model taking into account radon measurement error

The biologically based two-stage clonal expansion (TSCE) model is used to analyze lung cancer mortality of European miners from the Czech Republic, France, and Germany. All three cohorts indicate a highly significant action of exposure to radon and its progeny on promotion. The action on initiation is not significant in the French cohort. An action on transformation was tested but not found significant. In a pooled analysis, the results based on the French and German datasets do not differ significantly in any of the used parameters. For the Czech dataset, only lag time and two parameters that determine the clonal expansion without exposure and with low exposure rates (promotion) are consistent with the other studies. For low exposure rates, the resulting relative risks are quite similar. Exposure estimates for each calendar year are used. A model for random errors in each of these yearly exposures is presented. Depending on the used technique of exposure estimate, Berkson and classical errors are used. The consequences for the model parameters are calculated and found to be mostly of minor importance, except that the large difference in the exposure-induced initiation between the studies is decreased substantially.     

Impact of measurement error in radon exposure on the estimated excess relative risk of lung cancer death in a simulated study based on the French Uranium Miners’ Cohort

Measurement error (ME) can lead to bias in the analysis of epidemiologic studies. Here a simulation study is described that is based on data from the French Uranium Miners’ Cohort and that was conducted to assess the effect of ME on the estimated excess relative risk (ERR) of lung cancer death associated with radon exposure. Starting from a scenario without any ME, data were generated containing successively Berkson or classical ME depending on time periods, to reflect changes in the measurement of exposure to radon (²²²Rn) and its decay products over time in this cohort. Results indicate that ME attenuated the level of association with radon exposure, with a negative bias percentage on the order of 60% on the ERR estimate. Sensitivity analyses showed the consequences of specific ME characteristics (type, size, structure, and distribution) on the ERR estimates. In the future, it appears important to correct for ME upon analyzing cohorts such as this one to decrease bias in estimates of the ERR of adverse events associated with exposure to ionizing radiation.     

Semiparametric regression modeling with mixtures of Berkson and classical error, with application to fallout from the Nevada test site

We construct Bayesian methods for semiparametric modeling of a monotonic regression function when the predictors are measured with classical error. Berkson error, or a mixture of the two. Such methods require a distribution for the unobserved (latent) predictor, a distribution we also model semiparametrically. Such combinations of semiparametric methods for the dose response as well as the latent variable distribution have not been considered in the measurement error literature for any form of measurement error. In addition, our methods represent a new approach to those problems where the measurement error combines Berkson and classical components. While the methods are general, we develop them around a specific application, namely, the study of thyroid disease in relation to radiation fallout from the Nevada test site. We use this data to illustrate our methods, which suggest a point estimate (posterior mean) of relative risk at high doses nearly double that of previous analyses but that also suggest much greater uncertainty in the relative risk.     

Thyroid cancer following scalp irradiation: a reanalysis accounting for uncertainty in dosimetry

In the 1940s and 1950s, over 20,000 children in Israel were treated for tinea capitis (scalp ringworm) by irradiation to induce epilation. Follow-up studies showed that the radiation exposure was associated with the development of malignant thyroid neoplasms. Despite this clear evidence of an effect, the magnitude of the dose-response relationship is much less clear because of probable errors in individual estimates of dose to the thyroid gland. Such errors have the potential to bias dose-response estimation, a potential that was not widely appreciated at the time of the original analyses. We revisit this issue, describing in detail how errors in dosimetry might occur, and we develop a new dose-response model that takes the uncertainties of the dosimetry into account. Our model for the uncertainty in dosimetry is a complex and new variant of the classical multiplicative Berkson error model, having components of classical multiplicative measurement error as well as missing data. Analysis of the tinea capitis data suggests that measurement error in the dosimetry has only a negligible effect on dose-response estimation and inference as well as on the modifying effect of age at exposure.     

On Method of Moments Estimation in Linear Mixed Effects Models with Measurement Error on Covariates and Response with Application to a Longitudinal Study of Gene-Environment Interaction

We study a linear mixed effects model for longitudinal data, where the response variable and covariates with fixed effects are subject to measurement error. We propose a method of moment estimation that does not require any assumption on the functional forms of the distributions of random effects and other random errors in the model. For a classical measurement error model we apply the instrumental variable approach to ensure identifiability of the parameters. Our methodology, without instrumental variables, can be applied to Berkson measurement errors. Using simulation studies, we investigate the finite sample performances of the estimators and show the impact of measurement error on the covariates and the response on the estimation procedure. The results show that our method performs quite satisfactory, especially for the fixed effects with measurement error (even under misspecification of measurement error model). This method is applied to a real data example of a large birth and child cohort study.     

A cautionary comment on the generation of Berkson error in epidemiological studies

Exposure measurement error can be seen as one of the most important sources of uncertainty in studies in epidemiology. When the aim is to assess the effects of measurement error on statistical inference or to compare the performance of several methods for measurement error correction, it is indispensable to be able to generate different types of measurement error. This paper compares two approaches for the generation of Berkson error, which have recently been applied in radiation epidemiology, in their ability to generate exposure data that satisfy the properties of the Berkson model. In particular, it is shown that the use of one of the methods produces results that are not in accordance with two important properties of Berkson error.     

Mixtures of Berkson and classical covariate measurement error in the linear mixed model: Bias analysis and application to a study on ultrafine particles

The ultrafine particle measurements in the Augsburger Umweltstudie, a panel study conducted in Augsburg, Germany, exhibit measurement error from various sources. Measurements of mobile devices show classical possibly individual–specific measurement error; Berkson–type error, which may also vary individually, occurs, if measurements of fixed monitoring stations are used. The combination of fixed site and individual exposure measurements results in a mixture of the two error types. We extended existing bias analysis approaches to linear mixed models with a complex error structure including individual–specific error components, autocorrelated errors, and a mixture of classical and Berkson error. Theoretical considerations and simulation results show, that autocorrelation may severely change the attenuation of the effect estimations. Furthermore, unbalanced designs and the inclusion of confounding variables influence the degree of attenuation. Bias correction with the method of moments using data with mixture measurement error partially yielded better results compared to the usage of incomplete data with classical error. Confidence intervals (CIs) based on the delta method achieved better coverage probabilities than those based on Bootstrap samples. Moreover, we present the application of these new methods to heart rate measurements within the Augsburger Umweltstudie: the corrected effect estimates were slightly higher than their naive equivalents. The substantial measurement error of ultrafine particle measurements has little impact on the results. The developed methodology is generally applicable to longitudinal data with measurement error.     

Shared uncertainty in measurement error problems, with application to Nevada Test Site fallout data

In radiation epidemiology, it is often necessary to use mathematical models in the absence of direct measurements of individual doses. When complex models are used as surrogates for direct measurements to estimate individual doses that occurred almost 50 years ago, dose estimates will be associated with considerable error, this error being a mixture of (a) classical measurement error due to individual data such as diet histories and (b) Berkson measurement error associated with various aspects of the dosimetry system. In the Nevada Test Site(NTS) Thyroid Disease Study, the Berkson measurement errors are correlated within strata. This article concerns the development of statistical methods for inference about risk of radiation dose on thyroid disease, methods that account for the complex error structure inherence in the problem. Bayesian methods using Markov chain Monte Carlo and Monte-Carlo expectation-maximization methods are described, with both sharing a key Metropolis-Hastings step. Regression calibration is also considered, but we show that regression calibration does not use the correlation structure of the Berkson errors. Our methods are applied to the NTS Study, where we find a strong dose-response relationship between dose and thyroiditis. We conclude that full consideration of mixtures of Berkson and classical uncertainties in reconstructed individual doses are important for quantifying the dose response and its credibility/confidence interval. Using regression calibration and expectation values for individual doses can lead to a substantial underestimation of the excess relative risk per gray and its 95% confidence intervals.     

Biologically based analysis of the data for the Colorado uranium miners cohort: age, dose and dose-rate effects.

This study is a comprehensive analysis of the latest follow-up of the Colorado uranium miners cohort using the two-stage clonal expansion model with particular emphasis on effects related to age and exposure. The model provides a framework in which the hazard function for lung cancer mortality incorporates detailed information on exposure to radon and radon progeny from hard rock and uranium mining together with information on cigarette smoking. Even though the effect of smoking on lung cancer risk is explicitly modeled, a significant birth cohort effect is found which shows a linear increase in the baseline lung cancer risk with birth year of the miners in the cohort. The analysis based on the two-stage clonal expansion model suggests that exposure to radon affects both the rate of initiation of intermediate cells in the pathway to cancer and the rate of proliferation of intermediate cells. However, in contrast to the promotional effect of radon, which is highly significant, the effect of radon on the rate of initiation is found to be not significant. The model is also used to study the inverse dose-rate effect. This effect is evident for radon exposures typical for mines but is predicted to be attenuated, and for longer exposures even reversed, for the more protracted and lower radon exposures in homes. The model also predicts the drop in risk with time after exposure ceases. For residential exposures, lung cancer risks are compared with the estimates from the BEIR VI report. While the risk estimates are in agreement with those derived from residential studies, they are about two- to fourfold lower than those reported in the BEIR VI report.     

Residential magnetic fields predicted from wiring configurations: II. Relationships To childhood leukemia.

Case-control data on childhood leukemia in Los Angeles County were reanalyzed with residential magnetic fields predicted from the wiring configurations of nearby transmission and distribution lines. As described in a companion paper, the 24-h means of the magnetic field's magnitude in subjects' homes were predicted by a physically based regression model that had been fitted to 24-h measurements and wiring data. In addition, magnetic field exposures were adjusted for the most likely form of exposure assessment errors: classic errors for the 24-h measurements and Berkson errors for the predictions from wire configurations. Although the measured fields had no association with childhood leukemia (P for trend=.88), the risks were significant for predicted magnetic fields above 1.25 mG (odds ratio=2.00, 95% confidence interval=1.03-3.89), and a significant dose-response was seen (P for trend=.02). When exposures were determined by a combination of predictions and measurements that corrects for errors, the odds ratio (odd ratio=2.19, 95% confidence interval=1.12-4.31) and the trend (p =.007) showed somewhat greater significance. These findings support the hypothesis that magnetic fields from electrical lines are causally related to childhood leukemia but that this association has been inconsistent among epidemiologic studies due to different types of exposure assessment error. In these data, the leukemia risks from a child's residential magnetic field exposure appears to be better assessed by wire configurations than by 24-h area measurements. However, the predicted fields only partially account for the effect of the Wertheimer-Leeper wire code in a multivariate analysis and do not completely explain why these wire codes have been so often associated with childhood leukemia. The most plausible explanation for our findings is that the causal factor is another magnetic field exposure metric correlated to both wire code and the field's time-averaged magnitude.     

Breast cancer risk in atomic bomb survivors from multi-model inference with incidence data 1958–1998

Breast cancer risk from radiation exposure has been analyzed in the cohort of Japanese a-bomb survivors using empirical models and mechanistic two-step clonal expansion (TSCE) models with incidence data from 1958 to 1998. TSCE models rely on a phenomenological representation of cell transition processes on the path to cancer. They describe the data as good as empirical models and this fact has been exploited for risk assessment. Adequate models of both types have been selected with a statistical protocol based on parsimonious parameter deployment and their risk estimates have been combined using multi-model inference techniques. TSCE models relate the radiation risk to cell processes which are controlled by age-increasing rates of initiating mutations and by changes in hormone levels due to menopause. For exposure at young age, they predict an enhanced excess relative risk (ERR) whereas the preferred empirical model shows no dependence on age at exposure. At attained age 70, the multi-model median of the ERR at 1 Gy decreases moderately from 1.2 Gy⁻¹ (90% CI 0.72; 2.1) for exposure at age 25 to a 30% lower value for exposure at age 55. For cohort strata with few cases, where model predictions diverge, uncertainty intervals from multi-model inference are enhanced by up to a factor of 1.6 compared to the preferred empirical model. Multi-model inference provides a joint risk estimate from several plausible models rather than relying on a single model of choice. It produces more reliable point estimates and improves the characterization of uncertainties. The method is recommended for risk assessment in practical radiation protection.     

Lung cancer incidence attributable to residential radon exposure in Finland

This study aimed to estimate (1) the number of avoidable lung cancer cases attributable to residential radon in Finland in 2017, separately by age, sex, dwelling type and smoking status, (2) the impact of residential radon alone and the joint effect of residential radon and smoking on the number of lung cancers and (3) the potential decrease in the number of radon-attributable lung cancers if radon concentrations exceeding specified action levels (100, 200 and 300 Bq m^?3) would have been mitigated to those levels. Population-based surveys of radon concentrations and smoking patterns were used. Observed radon levels were contrasted with 25 Bq m^?3 representing a realistic minimum level of exposure. Lung cancer risk estimates for radon and smoking were derived from literature. Lastly, the uncertainty due to the estimation of exposure and risk was quantified using a computationally derived uncertainty interval. At least 3% and at most 8% of all lung cancers were estimated as being attributable to residential radon. For small cell carcinoma, the proportion of cases attributable to radon was 8–13%. Among smokers, the majority of the radon-related cases were attributable to the joint effect of radon and smoking. Reduction of radon exposure to 100 Bq m^?3 action level would eliminate approximately 30% of radon-attributable cases. Estimates were low compared with the literature, given the (relatively high) radon levels in Finland. This was mainly due to the lower radon levels and higher smoking prevalence in flats than in houses and a more realistic point of comparison, factors which have been ignored in previous studies. The results can guide actions in radon protection and in prevention of lung cancers.

     

A biologically based model for liver cancer risk in the Swedish thorotrast patients

Data on liver tumors among 416 Swedish patients who were exposed to Thorotrast between 1930 and 1950 were analyzed with the biologically based two-step clonal expansion (TSCE) model. For background data, the Swedish Cancer Register for the follow-up period 1958 to 1997 was used. Effects of radiation on the initiating mutation and on the clonal expansion rate explained the observed patterns well. The TSCE model permits the deduction of several kinetic parameters of the postulated tumorigenesis process. Dose rates of 5 mGy/year double the spontaneous initiation rate. The clonal expansion rate is doubled by 80 mGy/year, and for females it reaches a plateau at dose rates beyond 240 mGy/year. For males the plateau is not significant. The magnitude of the estimated promoting effect of radiation can be explained with a moderate increase in the cell replacement probability for the intermediate cells in the liver, which is strikingly similar to the situation in lung tumorigenesis.     

Lung Cancer Mortality (1950-1999) among Eldorado Uranium Workers: A Comparison of Models of Carcinogenesis and Empirical Excess Risk Models

Lung cancer mortality after exposure to radon decay products (RDP) among 16,236 male Eldorado uranium workers was analyzed. Male workers from the Beaverlodge and Port Radium uranium mines and the Port Hope radium and uranium refinery and processing facility who were first employed between 1932 and 1980 were followed up from 1950 to 1999. A total of 618 lung cancer deaths were observed. The analysis compared the results of the biologically-based two-stage clonal expansion (TSCE) model to the empirical excess risk model. The spontaneous clonal expansion rate of pre-malignant cells was reduced at older ages under the assumptions of the TSCE model. Exposure to RDP was associated with increase in the clonal expansion rate during exposure but not afterwards. The increase was stronger for lower exposure rates. A radiation-induced bystander effect could be a possible explanation for such an exposure response. Results on excess risks were compared to a linear dose-response parametric excess risk model with attained age, time since exposure and dose rate as effect modifiers. In all models the excess relative risk decreased with increasing attained age, increasing time since exposure and increasing exposure rate. Large model uncertainties were found in particular for small exposure rates.     

Concepts of association between cancer and ionising radiation: accounting for specific biological mechanisms

The probability that an observed cancer was caused by radiation exposure is usually estimated using cancer rates and risk models from radioepidemiological cohorts and is called assigned share (AS). This definition implicitly assumes that an ongoing carcinogenic process is unaffected by the studied radiation exposure. However, there is strong evidence that radiation can also accelerate an existing clonal development towards cancer. In this work, we define different association measures that an observed cancer was newly induced, accelerated, or retarded. The measures were quantified exemplarily by Monte Carlo simulations that track the development of individual cells. Three biologically based two-stage clonal expansion (TSCE) models were applied. In the first model, radiation initiates cancer development, while in the other two, radiation has a promoting effect, i.e. radiation accelerates the clonal expansion of pre-cancerous cells. The parameters of the TSCE models were derived from breast cancer data from the atomic bomb survivors of Hiroshima and Nagasaki. For exposure at age 30, all three models resulted in similar estimates of AS at age 60. For the initiation model, estimates of association were nearly identical to AS. However, for the promotion models, the cancerous clonal development was frequently accelerated towards younger ages, resulting in associations substantially higher than AS. This work shows that the association between a given cancer and exposure in an affected person depends on the underlying biological mechanism and can be substantially larger than the AS derived from classic radioepidemiology.

     

Lung cancer risk in mice: analysis of fractionation effects and neutron RBE with a biologically motivated model

Data from Argonne National Laboratory on lung cancer in 15,975 mice with acute and fractionated exposures to gamma rays and neutrons are analyzed with a biologically motivated model with two rate-limiting steps and clonal expansion. Fractionation effects and effects of radiation quality can be explained well by the estimated kinetic parameters. Both an initiating and a promoting action of neutrons and gamma rays are suggested. While for gamma rays the initiating event is described well with a linear dose-rate dependence, for neutrons a nonlinear term is needed, with less effectiveness at higher dose rates. For the initiating event, the neutron RBE compared to gamma rays is about 10 when the dose rate during each fraction is low. For higher dose rates this RBE decreases strongly. The estimated lifetime relative risk for radiation-induced lung cancers from 1 Gy of acute gamma-ray exposure at an age of 110 days is 1.27 for male mice and 1.53 for female mice. For doses less than 1 Gy, the effectiveness of fractionated exposure to gamma rays compared to acute exposure is between 0.4 and 0.7 in both sexes. For lifetime relative risk, the RBE from acute neutrons at low doses is estimated at about 10 relative to acute gamma-ray exposure. It decreases strongly with dose. For fractionated neutrons, it is lower, down to about 4 for male mice.     

Lung cancer risk of Mayak workers: modelling of carcinogenesis and bystander effect

Lung cancer mortality in the period of 1948-2002 has been analysed for 6,293 male workers of the Mayak Production Association, for whose information on smoking, annual external doses and annual lung doses due to plutonium exposures was available. Individual likelihoods were maximized for the two-stage clonal expansion (TSCE) model of carcinogenesis and for an empirical risk model. Possible detrimental and protective bystander effects on mutation and malignant transformation rates were taken into account in the TSCE model. Criteria for non-nested models were used to evaluate the quality of fit. Data were found to be incompatible with the model including a detrimental bystander effect. The model with a protective bystander effect did not improve the quality of fit over models without a bystander effect. The preferred TSCE model was sub-multiplicative in the risks due to smoking and internal radiation, and more than additive. Smoking contributed 57% to the lung cancer deaths, the interaction of smoking and radiation 27%, radiation 10%, and others cause 6%. An assessment of the relative biological effectiveness of plutonium was consistent with the ICRP recommended value of 20. At age 60 years, the excess relative risk (ERR) per lung dose was 0.20 (95% CI: 0.13; 0.40) Sv(-1), while the excess absolute risk (EAR) per lung dose was 3.2 (2.0; 6.2) per 10(4) PY Sv. With increasing age attained the ERR decreased and the EAR increased. In contrast to the atomic bomb survivors, a significant elevated lung cancer risk was also found for age attained younger than 55 years. For cumulative lung doses below 5 Sv, the excess risk depended linearly on dose. The excess relative risk was significantly lower in the TSCE model for ages attained younger than 55 than that in the empirical model. This reflects a model uncertainty in the results, which is not expressed by the standard statistical uncertainty bands.     

Methods for estimation of radiation risk in epidemiological studies accounting for classical and Berkson errors in doses

With a binary response Y, the dose-response model under consideration is logistic in flavor with pr(Y=1 | D) = R (1+R)(-1), R = λ(0) + EAR D, where λ(0) is the baseline incidence rate and EAR is the excess absolute risk per gray. The calculated thyroid dose of a person i is expressed as Dimes=fiQi(mes)/Mi(mes). Here, Qi(mes) is the measured content of radioiodine in the thyroid gland of person i at time t(mes), Mi(mes) is the estimate of the thyroid mass, and f(i) is the normalizing multiplier. The Q(i) and M(i) are measured with multiplicative errors Vi(Q) and ViM, so that Qi(mes)=Qi(tr)Vi(Q) (this is classical measurement error model) and Mi(tr)=Mi(mes)Vi(M) (this is Berkson measurement error model). Here, Qi(tr) is the true content of radioactivity in the thyroid gland, and Mi(tr) is the true value of the thyroid mass. The error in f(i) is much smaller than the errors in ( Qi(mes), Mi(mes)) and ignored in the analysis. By means of Parametric Full Maximum Likelihood and Regression Calibration (under the assumption that the data set of true doses has lognormal distribution), Nonparametric Full Maximum Likelihood, Nonparametric Regression Calibration, and by properly tuned SIMEX method we study the influence of measurement errors in thyroid dose on the estimates of λ(0) and EAR. The simulation study is presented based on a real sample from the epidemiological studies. The doses were reconstructed in the framework of the Ukrainian-American project on the investigation of Post-Chernobyl thyroid cancers in Ukraine, and the underlying subpolulation was artificially enlarged in order to increase the statistical power. The true risk parameters were given by the values to earlier epidemiological studies, and then the binary response was simulated according to the dose-response model.     

Bone cancer risk in mice exposed to 224Ra: protraction effects from promotion

This paper analyzes data for the osteosarcoma incidence in life-time experiments of (224)Ra injected mice with respect to the importance of initiating and promoting action of ionizing high LET-radiation. This was done with the biologically motivated two step clonal expansion (TSCE) model of tumor induction. Experimentally derived osteosarcoma incidence in 1,194 mice following exposure to (224)Ra with different total radiation doses and different fractionation patterns were analyzed together with incidence data from 1,710 unirradiated control animals. Effects of radiation on the initiating event and on the clonal expansion rate, i.e. on promotion were found to be necessary to explain the observed patterns with this model. The data show a distinct inverse protraction effect at high doses, whereas at lower doses this effect becomes insignificant. Such a behavior is well reproduced in the proposed model: At dose rates above 6 mGy/day a longer exposure produces higher ERR per dose, while for lower rates the reverse is the case. The TSCE model permits the deduction of several kinetic parameters of a postulated two-step bone tumorigenesis process. Mean exposure rates of 0.13 mGy/day are found to double the baseline initiation rate. At rates above 100 mGy/day, the initiation rate decreases. The clonal expansion rate is doubled at 8 mGy/day, and it levels out at rates beyond 100 mGy/day.