When ionizing radiation interacts with biological tissue, it can harm cells directly by damaging the deoxyribonucleic acid (DNA) and other macromolecules or it can harm cells indirectly by ionizing water molecules, which in turn form free radicals that oxidize cellular molecules and break chemical bonds. In the context of the exposure of astronauts, radiation can have both acute and long-term health effects depending on the dose and dose rate. Acute effects include acute radiation sickness and cutaneous effects, and long-term health effects include cataracts, cancer, cardiovascular diseases, and degenerative tissue effects.
This chapter examines the risks to astronauts due to radiation exposure during spaceflight. It includes a brief background on the space radiation environment, the cellular and molecular damage that space radiation can cause, and the potential health impacts. The chapter also includes an overview of both the cancer risk model used by the National Aeronautics and Space Administration (NASA), which provides the basis for the agency’s space radiation health standard, and sex, age, and genetic factors that affect radiation-induced cancer risk.
Beyond the protection of Earth’s magnetic field, astronauts are exposed to a complex radiation environment (Simonsen and Nealy, 1991) comprised of galactic cosmic rays (GCR) and solar particle events (SPEs) (Cucinotta et al., 2013; Kronenberg and Cucinotta, 2012). GCR has the same intensity
regardless of the direction of the measurement (i.e., isotropic) and is composed of mostly highly energetic protons (85 percent), helium ions (14 percent), and high atomic number, high-energy (HZE) particles, defined as having an electric charge greater than 2+ (1 percent) (Schimmerling, 2011; Zeitlin et al., 2013). GCR ions of primary concern have an atomic number up to 28 and energies from less than ~1 mega-electronvolt (MeV) to greater than ~10 giga-electronvolt (GeV) per nucleon. They are highly penetrating and cannot readily be attenuated or stopped by shielding. GCR fluence rate for ions of less than 2 GeV per nucleon varies about a factor of two with the 11-year solar cycle (higher at solar minimums and lower at solar maximums). SPEs include particles, primarily protons with energies from ~1 MeV to several hundred MeV and with fluences exceeding 109 protons cm–2. SPEs can include other nuclei such as helium ions and HZE ions. SPEs occur sporadically with frequency also varying with the solar cycle, although both their frequency and intensity are unpredictable. SPE protons of less than 30 MeV are unable to penetrate spacecraft or extravehicular activity suits while higher energy protons will also contribute to radiation exposure in space (Schimmerling, 2011).
When considering the health effects of space radiation, two quantities are relevant:
- Linear energy transfer (LET) is the amount of energy that is deposited in matter (such as biological tissue) per unit distance that the ionizing radiation travels; and
- Relative biological effectiveness (RBE) is used to compare how damaging radiation is, using X-rays or gamma rays as a reference. A radiation that is 10 times more effective per unit dose than X-rays would have an RBE of 10. RBE varies with dose, dose rate, and measured endpoint, among other factors.
The biological damage caused by ionizing radiation depends on both the dose and the type of radiation as defined, for example, by its LET. The division between high- and low-LET radiations is often difficult to define, but many NASA investigators consider low LET radiation to be <10 KeV/µm. Different types of ionizing radiation possess different energies, which affects both how these types interact with cells and tissues and the damage they cause. So absorbed dose is insufficient to fully account for the estimated risk. To properly account for this variation in damage (RBE), the absorbed dose is multiplied by a radiation quality factor, yielding what is described as the equivalent dose (ICRP, 2007).
Depending on their atomic number and energy, GCR particles are typically characterized as high-LET compared to the low-LET of sparsely ionizing gamma rays and X-rays. High-LET HZE particles traversing material
deposit a large fraction of their energy as secondary electrons produced along the particle track, resulting in high ionization density along the track (Blakely, 2012; Katz et al., 1971). Depending on their atomic number and energy, HZE particle interactions can produce clustered damage in DNA (Cucinotta et al., 2000; Goodhead, 1994; Hada and Sutherland, 2006; Nikjoo et al., 1999; Rydberg, 1996) that are harder for a cell to repair and likely account for the high RBE for cell death, mutation, chromosome aberrations, and carcinogenesis (Held, 2009). SPE radiations, largely lacking the HZE particles included among GCR, have a different distribution of RBE with energy than the HZE particles. None of these experiments are at dose-rates that one would expect in space exposures, which also contributes to uncertainties. NASA recognizes that the specific biological effects of these highly ionizing particles are poorly understood, leading to large uncertainties in risk estimation (Cucinotta et al., 2013; NCRP, 2012; Simonsen and Slaba, 2020).
Health Impacts of Exposure to Space Radiation
While spacecraft shielding and protected spaces within the spacecraft can protect against SPEs and therefore protect astronauts, SPEs could affect astronauts on an extravehicular activity in space or on a planetary surface. Exposure to high amounts of radiation (1 or 2 gray [Gy] with some variation among individuals) could cause astronauts to develop an acute radiation syndrome (ARS). Symptoms can include anemia, leukopenia, and hemorrhage; gastrointestinal distress, damage, and pain; and fever and shock. The lethal dose to 50 percent of the human population (LD50) is at approximately 4–4.5 Gy (varying depending on medical support) (DOE, 2017). Doses above 8 Gy are almost always fatal, and at doses above 30 Gy, neurovascular symptoms (seizure, tremor, ataxia) occur prior to death. Lower levels of radiation are insufficient to cause acute radiation sickness but can increase the risk of several long-term health effects, including cancer, cardiovascular disease, cataracts, and degeneration of central nervous system tissue.
Ionizing radiation increases the risk of cancer with increasing dose and the effects are cumulative (IARC, 2012).
Risk models have been developed to assess an individual’s risk of developing cancer in general or at a specific site due to radiation exposure. Typically these models provide an average value and a range of possible values that capture key uncertainties. A minimal dose–response model contains parameters that reflect the dose of radiation received, the sex of
the individual, the age at exposure to radiation, and the attained age of the individual.
When designing or selecting an appropriate cancer risk model, experts need to carefully consider and evaluate the model’s performance in describing currently available human and/or animal radiation carcinogenesis data, as well as the mechanistic and biological plausibility of the model’s assumptions.
In general, cancer risk models from low-LET radiations are often fitted with data from acute exposures and relatively large doses (typically the Japanese atomic bomb survivor data and specifically the Life Span Study [LSS]). It is therefore necessary to incorporate factors that account for potential differences in extrapolating from high to low dose and from high to low dose rates. The approach uses the dose and dose-rate effectiveness factor (DDREF). The values of the DDREF recommended by national and international committees range from about 1.5 to about 3. Whenever it is applied, the DDREF functions to lower the slope of the linear-nonthreshold (LNT) function.1 Because only the dose is used in most determinations of both the dose and the dose-rate components of the DDREF, many groups have suggested separating the dose-rate effect (DREF) from the low dose effect (LDEF) (NCRP, 2020).
Although most radiation risk models are based primarily on the Japanese atomic bomb survivors, there are a large number of other epidemiological studies of radiation and cancer. Most relevant to NASA’s risk models are the growing number of studies of occupational radiation exposures including the multicenter international nuclear worker cohort study INWORKs, the U.S. Radiologic Technologists Study, and the Million Person Study. Results from these studies, particularly for total cancer risk, are broadly consistent with the LSS (Ozasa et al., 2018); this provides some level of assurance for the radiation risk models. There are some exceptions though, such as the sex differences in lung cancer risk (see below for more details).
Transfer of Risks Across Populations
Other parameters built into the models that affect the output values of estimated risk relate to the transfer of risks across populations. These
1 The appropriateness of using the LNT model to extrapolate risks at low (less than 100 mSv) doses has been strongly debated for some time. NCRP Commentary 27 (NCRP, 2018a) concluded that recent epidemiological studies are compatible with the continued use of the LNT model for radiation protection. The committee was not tasked with assessing the appropriateness of the LNT model in setting radiation protection standards for astronauts, which are set at doses above the typical level of dispute of the LNT model.
parameters account for the differences in background incidence rate for specific cancers between two populations, as well as accounting for the synergistic effects between radiation and other risk factors such as smoking.
Studies have shown that the background rates for various cancers differ across populations. For example, the LSS cohort and contemporary Japanese populations have a greater background incidence rate of stomach cancer and a lower rate of breast cancer compared with a contemporary U.S. population (IARC, 2017; Thompson et al., 1994). These differences manifest when considering whether to use excess relative risk (ERR) or the excess absolute risk (EAR) to transfer the risk model from the Japanese population to a contemporary U.S. population. The increase in cancer risk attributable to a particular absorbed dose of radiation is given as excess risk relative to the background (ERR) or in addition to the background (EAR) risk of developing a particular cancer type. The decision to use ERR indicates that the radiation interacts, in some fashion, with other risk factors that comprise the background risk for a particular cancer. Perhaps the most studied example is for the interactions between radiation exposure and tobacco consumption and their effects on lung cancer risk (ICRP, 2010; NRC, 1999; Pierce et al., 2003).
Relative Biological Effectiveness
RBE values derived from irradiation of animals provide a better model for the human space situation than RBE values from irradiated cellular systems. RBE studies have been performed with mice and other animal models irradiated to mimic HZE radiation and SPE proton storms in space. The RBE value changes significantly depending on the survival level of cells (and animals), and thus all RBE calculations present uncertainty in predicting the human situation, since exposure conditions, cellular endpoints, and even numbers of cells exposed in a person would vary from one exposure to another. Despite these concerns about uncertainty, space-irradiation experiments in animal systems have provided useful RBE values to include in cancer risk models.
For example, LD50/302 values for C57BL/6J mice, demonstrated RBEs of 1.4 for Si-28 and 0.99 for C-12 (Suman et al., 2012). Tumor induction using C ions in C3H female mice revealed an RBE of 1–2.1 when lifetime tumorigenesis was examined (Ando et al., 2014). The RBE for GCR/HZE ions is likely to be higher than for protons, demonstrated in tumor induction using the plateau region of those GCR ions that were present (Shuryak et al., 2017; Suman et al., 2016). Using a jejunal crypt microcolony assay
to examine the RBE for single cell survival in vivo, Mason et al. (2007) identified an RBE of 1.1 for protons, and a similar study examining the RBE at the Spread Out Bragg Peak of the proton beam revealed an RBE of 1.1–1.2 (Gueulette et al., 2005). These studies allow one to conclude that RBEs range higher (2–5-fold) for GCR exposure than for protons, whose RBEs are very close to those of gamma rays when tested in animals for single-cell survival assays or for carcinogenesis. Studies on the induction of Harderian gland tumors (a tumor found uniquely in rodents) have shown higher RBE values of space radiation exposures than other cancer-related endpoints; these RBE values have also been shown to vary considerably with dose rate (Chang et al., 2016).
There remain a number of uncertainties associated with the development of radiation risk estimates from epidemiology studies, including the LSS data. These are, in general, dosimetric uncertainties, epidemiological and methodological uncertainties, uncertainties in modeling epidemiological data, and when considering the potential effects of other radiation-induced outcomes, uncertainties in assessment of non-cancer data and uncertainties in assessment of heritable effects (NCRP, 2012).
Sex Differences and Radiation-Induced Cancer Risk
There are several important considerations regarding the incorporation of sex differences into the calculation of risk of exposure-induced death (REID), and the application of these on occupational exposure limits for female and male astronauts.
Based on reviews of the literature on lifetime risks associated with radiation (NCRP, 2000, 2014), it was concluded that women had a higher excess risk of cancer than men from the same level of radiation exposure. NASA subsequently incorporated the difference in sex-specific response to radiation into their protection guidance for astronauts (NASA, 2014), noting that planned career exposure for radiation shall not exceed 3 percent REID for cancer mortality, adjusted for age and sex, as estimated under the current NASA computational model for space radiation cancer risk projections (Cucinotta et al., 2013). The operational outcome was that female astronauts were allowed to have less time in space than their male counterparts.
The NASA computational model for space radiation cancer risk projections (Cucinotta et al., 2013) incorporates information on the background
rates of lung cancer. Lung cancer is the leading cause of cancer death in the United States, excluding skin cancer (Howlader et al., 2021); however, the rates have been declining for men and women (particularly over the past two decades), even in never-smokers (Thun et al., 2013), with higher decline rates in men (Siegel et al., 2021). Although lung cancer has the largest contribution of all cancers to the calculation of REID for fatal cancers, it is unclear how much taking into account sex-specific differences in radiation risks of lung cancer would affect the overall calculations of REID.
Evidence suggesting the potential for significant sex differences in radiation risks of lung cancer (as well as esophagus and stomach) continues to derive primarily from one study—the study of the Japanese atomic bomb survivors. The latest cancer risk updates from this population continues to show that the risk of death from radiation-induced lung cancer in nonsmokers was nearly three times higher for women than for men (Furukawa et al., 2010; Ozasa et al., 2012). A summary of relevant studies (organized by high- and low-LET radiation) is presented in Table A-1. The committee conducted a systematic search of publications in English from PubMed using MeSH terms. Studies were selected that presented either estimates of external radiation doses, frequencies of lung cancer deaths, or estimates of radiation risks of lung cancer separately by sex. Studies of internal radiation exposures were included only if study participants were also exposed to external radiation and risks of these exposures were analyzed separately. Studies that presented radiation risks estimates for males and females together were included if the authors stated that there were no differences in risks between them.
The majority of studies of occupationally exposed populations (Boice et al., 2011, 2014, 2019; Cardis et al., 2007; Golden et al., 2019; Haylock et al., 2018; Muirhead et al., 2009; Richardson et al., 2018; Silver et al., 2013; Velazquez-Kronen et al., 2020) did not find significant differences in risks of death from lung cancer caused by radiation between men and women, but not all studies adequately assessed smoking. Studies of occupationally exposed workers frequently include very few women and many of them tend to have doses of radiation exposures which are lower than doses for men, which complicates analyses of sex-specific risks due to low statistical power. Differences in radiation risks of lung cancer have been reported for Mayak workers from Russia (Gilbert et al., 2013) but the differences in risk were only observed in analyses with plutonium and there were no differences in risks due to external exposures (Gilbert et al., 2013; Gillies et al., 2017). Similarly, studies of long-term effects of exposures to significant fluoroscopy doses to tuberculosis patients (Boice et al., 2019; Davis et al., 1989; Howe, 1995) or to high-dose radiation treatment for primary cancer or peptic ulcer (Carr et al., 2002; Gilbert et al., 2003; Little et al., 2013; NCRP, 2011) did not find significant differences in risks of death from lung
cancer caused by radiation between men and women. The available epidemiological evidence is currently being evaluated by the National Council on Radiation Protection and Measurements (NCRP) Scientific Committee SC 1-27, charged with assessing radiation-induced lung cancer in populations exposed to chronic or fractionated radiation and developing methods for analyzing these data for sex-specific differences (NCRP, 2019). The reasons for the differences between these studies and the LSS are uncertain, but they could be due to low statistical power to assess sex-specific differences in studies of occupationally exposed workers or some limitations inherent to the LSS (e.g., underestimation of smoking among Japanese women, including passive smoking exposure).
Current ground-based systems of radiation protection do not differentiate between sex in either their limitation or numeric protection criteria structures (ICRP, 2007; NCRP, 2018b). The dosimetric quantity recommended for radiological protection, effective dose, is computed by averaging age and sex (ICRP, 2007; NCRP, 2018b).
Under the International Commission on Radiological Protection (ICRP) system for adults, equivalent doses for specific organs are calculated by the sex averaging of values obtained using male and female phantoms. Effective doses are then calculated using age- and sex-averaged tissue weighting factors, based on risk data and are “intended to apply as rounded values to a population of both sexes and all ages” (ICRP, 2007, p. 13). ICRP specifically noted the following with respect to the application of the system of radiation protection for both sexes for ground-based applications:
In view of the uncertainties surrounding the values of tissue weighting factors and the estimate of detriment, the Commission considers it appropriate for radiological protection purposes to use age- and sex-averaged tissue weighting factors and numerical risk estimates. The system of protection is sufficiently robust to achieve adequate protection for both sexes. Moreover, this obviates the requirement for sex- and age-specific radiological protection criteria which could prove unnecessarily discriminatory. (ICRP, 2007, p. 42)
Similarly, NCRP addresses this point as follows:
While recognizing that there are variations in cancer risk between organs, between males and females, and at different ages, the system of protection needs to be applied consistently to all individuals in the population. Thus, numeric protection criteria for stochastic effects are specified as a single value based on a population average over both sexes and all ages. (NCRP, 2018b, p. 41)
Age and Radiation-Induced Cancer Risk
Population mortality rates of lung cancer increase with age in both men and women. The highest rates have been reported for men aged 75 years and over (439 per 100,000 per year in 2017; Howlader et al., 2021). Evidence of the modifying effects of age at exposure on the association between radiation exposure and lung cancer comes primarily from the study of atomic bomb survivors from Japan (Cahoon et al., 2017). Radiation risks increased with increasing age at exposure, but declined with increasing attained age. There were no indications of sex dependence in effect modification by attained age or age at exposure. Analyses of nuclear workers (Cardis et al., 2007; Haylock et al., 2018; Richardson et al., 2018) either did not examine (Haylock et al., 2018) or did not find (Cardis et al., 2007; Muirhead et al., 2009) variations in radiation risks of lung cancer with attained age or age at exposure, most likely owing to low statistical power for such analyses. Pooled analysis of uranium miners showed a decrease in radiation risks of lung cancer with attained age (both in exposure-age-concentration and exposure-age-duration models) (NRC, 1999).
Genetics and Radiation-Induced Cancer Risk
It is well recognized that certain rare genetic mutations significantly increase an individual’s risk of developing cancer. For example, inheritance of mutations in genes that regulate genome integrity, such as the TP53 or BRCA1 tumor suppressors, lead to increased risk of cancers by 50–85 percent over an individual’s life span. TP53 mutation carriers develop multiple types of cancer, such as sarcomas, breast, brain, leukemias, and adrenocorticoid carcinomas, and BRCA1 patients develop breast, ovarian, and prostate cancers. Radiation treatment of individuals with TP53-mutant cancers increases the number of second cancers (Heymann, 2010). Other less penetrant variants, known as single nucleotide polymorphisms, present in the genome also contribute to increased risk of various diseases (Wand et al., 2021).
Among individuals with certain genetic polymorphisms, radiation has been found to further increase cancer risk. For example, in a recent study of long-term childhood cancer survivors, radiation therapy increased the occurrences of subsequent cancers among individuals with genetic polymorphisms in genes that regulate DNA double strand break repair (Morton et al., 2020). Efforts are under way to quantify the contributions of these genetic polymorphisms to provide reliable estimates of disease risk, including cancer (Wand et al., 2021).
NASA requested that this study committee consider the agency’s space radiation risk management process, including the management of uncertainties related to cancer risk from exposure to space radiation. This committee was not tasked with performing an evaluation of NASA’s cancer risk model. The following overview of NASA’s cancer risk model serves to describe the foundation on which NASA’s risk management process and Permissible Exposure Limit for Spaceflight Radiation Exposure Standard are based.
Overview of the NSCR 2012 Model
The current NASA cancer risk model (NSCR 2012) is based on the model developed by Cucinotta et al. (2013) and incorporates input from the National Research Council (2012) review committee. The general formulation closely follows those developed by other national and international committees (NRC, 2006; UNSCEAR, 2008). The risk for each of 20 radiosensitive tissues is estimated separately according to age at exposure using risk models developed primarily from the LSS of the Japanese atomic bomb survivors. A DDREF is applied to scale from an acute exposure (received from the atomic bomb) to the chronic exposure received by space travel. Radiation dose estimates include quality factors that account for the increased RBE of particle radiation encountered in space compared to gamma rays.
To transfer the risk models from the Japanese population to a contemporary U.S. population, NSCR 2012 uses a mixture model in the risk projection that randomly combines EAR, in which the risk of radiation exposure is calculated as a separate number from the baseline risk of cancer for unexposed individuals in the atomic bomb survivor cohort, and ERR, in which the risk of radiation exposure is calculated as a multiple of the baseline risk.
Background cancer rates for U.S. never-smokers are used to approximate the likely cancer risks in the astronaut population, which results in lower risk estimates for smoking-related cancers. Radiation-induced cancer incidence estimates for each tissue are converted to cancer mortality using cancer-specific, incidence-based mortality factors. The REID from cancer is then estimated by summing risks across cancer sites and attained age, with adjustment for competing causes of death using a life table approach.
Uncertainties in the NSCR 2012 Model
The NSCR 2012 model includes key uncertainties in inputs currently used in other radiation risk models for the U.S. population:
- The Poisson regression uncertainty in the risk coefficients from the atomic bomb survivors, which include the sex of the survivor, the age at exposure, and the attained age for which the risk is being calculated;
- Uncertainty estimates for the doses assigned to the atomic bomb survivors;
- Uncertainty in the DDREF for protracted exposures in contrast to the acute exposures received by the atomic bomb survivors; and
- Uncertainty in the risk transfer from the Japanese atomic bomb survivors to a contemporary U.S. population.
The NSCR 2012 model has incorporated a set of uncertainty estimates for the biggest uncertainty contributor, which is the radiation quality of the HZE particulate radiations that form the great majority of the astronauts’ exposure, which have very high LET, in contrast to the low-LET gamma radiation that was the predominant exposure of the Japanese atomic bomb survivors. The NSCR 2012 model also has incorporated uncertainty estimates for the radiological physics aspects of the dose estimates for the astronauts.
NSCR 2012 appears to lack a few types of uncertainty estimates in the atomic bomb survivor data that are considered in the radio-epidemiological model used by the U.S. Environmental Protection Agency (Pawel, 2013). These include uncertainty owing to incomplete follow-up with the atomic bomb survivors, the uncertainty caused by errors in disease detection and diagnosis in the atomic bomb survivors, uncertainty caused by failure to capture diagnostic exposures, and the uncertainty caused by potential selection bias in the atomic bomb survivor cohort. Note, however, that these uncertainties, while absent from the NSCR 2012 model, are relatively small compared to the uncertainties that govern in the NSCR 2012 calculations. NASA may consider utilizing an influence diagram as a visual tool to explain the relationship between the various uncertainties.
The NSCR 2012 model also does not consider a number of other known possible sources of uncertainty related to the effects of radiation, one of which is the possibility of non-targeted effects in which the risk of cancer in a given organ results from a dose to a different organ (Desouky et al., 2015). A number of other uncertainties not currently considered in NSCR 2012 are listed elsewhere (see Figure 5 of Simonsen and Slaba, 2020). These include
- The uncertainty in scaling space radiation carcinogenesis from that due to gamma rays, including possibly both different latency and different lethality of tumors;
- Shape of dose response at low doses—linear-non-threshold as assumed or linear quadratic (the 2012 UNSCEAR report emphasizes the uncertainty in the dose–response model and suggests such approaches as multi-model inference);
- Mixed field additivity for high- and low-LET radiations and DREFs;
- Translation of animal experimental data to humans;
- Individual radiation sensitivity; and
- The effect of other combined stressors of spaceflight on cancer risk.
The NASA model includes assumptions for radiation-induced excess risk in relation to time since exposure, for the DDREF, and for radiation quality. Note that various combinations of models for these three factors substantially increase the upper tail of the uncertainty distribution for the ensemble compared to the NSCR 2012 model, which does not include the other uncertainties just mentioned (Desouky et al., 2015; Pawel, 2013).
The UNSCEAR 2012 report suggests that for estimates of lifetime risk for all solid cancers combined, “Addressing known sources of uncertainty, the 95 percent credible [confidence]3 intervals span about an order of magnitude,” and that “estimates of site-specific cancer risks have larger uncertainties still” (UNSCEAR, 2015, p. 22). NASA has provided a plot (see Figure 3 of Simonsen and Slaba, 2020) that suggests that in NSCR 2012 the ratio is only about 3/0.83 = 3.6 between the 50th and 95th percentile (Simonsen and Slaba, 2020). This is close to an order of magnitude for a 95 percent credible interval, as shown in Figure 4-1. The NSCR 2012 estimate includes the particularly large uncertainty of radiation quality for HZE radiations that is not included in the interval suggested in the UNSCEAR report.
The resulting risk distribution from NSCR 2012 is unimodal, with a peak, but it is asymmetrical and has a long “upper tail,” where risk magnitudes several fold higher than the mean or median have non-negligible probabilities. Such a complicated shape for the risk predictions poses challenges for (1) determining which part(s) of the distribution (e.g., mean, median, 97.5 percent or some other percentile of the cumulative distribution) are most useful for setting a corresponding radiation dose limit for astronaut exposures; and (2) effectively communicating the radiation-induced risk
3 UNSCEAR defines a credible interval as “an interval defined from the distribution of the degree of belief of the value of the quantity of interest within which a certain probability is assigned (e.g. 95%) representing the assessor’s degree of belief that the true value of a quantity of interest falls within the interval” (UNSCEAR, 2019, p. 8).
from space missions to astronauts. Using a high percentile of the cumulative predicted risk distribution for these tasks appears to be conservative in terms of protecting astronaut health because, by definition, the model predicts that the risk will be lower than the communicated value with very high probability. In other words, an astronaut could interpret such a number as a plausible upper bound on his or her risk at the given accumulated dose. However, the accuracy of the high percentiles of the risk distribution is likely to be relatively low because of the inherent limitations of currently available input data for the risk model and assumptions that are unavoidable in the modeling process (described above), for example.
It is likely that improvements in data quality and amount over time, as well as evolution and improvement of risk modeling strategies, can substantially alter the predicted upper percentiles of the risk distribution. In other words, the shape and length of the “upper tail” of this distribution are not very well known. In contrast, it is likely that improved data and modeling methodologies will have less effect on altering the central portions of the risk distribution such as the mean or median, which are generally more stable than the extremes, as shown in NASA’s report on ensemble models (Simonsen and Slaba, 2020).
Ando, K., S. Koike, Y. Ohmachi, Y. Ando, and G. Kobashi. 2014. Tumor induction in mice after local irradiation with single doses of either carbon-ion beams or gamma rays. International Journal of Radiation Biology 90(12):1119–1124.
Blakely, E. A. 2012. Lauriston S. Taylor lecture on radiation protection and measurements: What makes particle radiation so effective? Health Physics 103(5):508–528.
Boice, J. D., S. S. Cohen, M. T. Mumma, E. D. Ellis, K. F. Eckerman, R. W. Leggett, B. B. Boecker, A. B. Brill, and B. E. Henderson. 2011. Updated mortality analysis of radiation workers at Rocketdyne (Atomics International), 1948–2008. Radiation Research 176(2):244–258.
Boice, J. D., Jr., S. S. Cohen, M. T. Mumma, E. D. Ellis, D. L. Cragle, K. F. Eckerman, P. W. Wallace, B. Chadda, J. S. Sonderman, L. D. Wiggs, B. S. Richter, and R. W. Leggett. 2014. Mortality among mound workers exposed to polonium-210 and other sources of radiation, 1944–1979. Radiation Research 181(2):208–228.
Boice, J. D., E. D. Ellis, A. P. Golden, L. B. Zablotska, M. T. Mumma, and S. S. Cohen. 2019. Sex-specific lung cancer risk among radiation workers in the Million Person Study and patients TB-Fluoroscopy. International Journal of Radiation Biology 7:1–12.
Cahoon, E. K., D. L. Preston, D. A. Pierce, E. Grant, A. V. Brenner, K. Mabuchi, M. Utada, and K. Ozasa. 2017. Lung, laryngeal and other respiratory cancer incidence among Japanese atomic bomb survivors: An updated analysis from 1958 through 2009. Radiation Research 187(5):538–548.
Cardis, E., M. Vrijheid, M. Blettner, E. Gilbert, M. Hakama, C. Hill, G. Howe, J. Kaldor, C. R. Muirhead, M. Schubauer-Berigan, T. Yoshimura, F. Bermann, G. Cowper, J. Fix, C. Hacker, B. Heinmiller, M. Marshall, I. Thierry-Chef, D. Utterback, Y. O. Ahn, E. Amoros, P. Ashmore, A. Auvinen, J. M. Bae, J. Bernar, A. Biau, E. Combalot, P. Deboodt, A. Diez Sacristan, M. Eklöf, H. Engels, G. Engholm, G. Gulis, R. R. Habib, K. Holan, H. Hyvonen, A. Kerekes, J. Kurtinaitis, H. Malker, M. Martuzzi, A. Mastauskas, A. Monnet, M. Moser, M. S. Pearce, D. B. Richardson, F. Rodriguez-Artalejo, A. Rogel, H. Tardy, M. Telle-Lamberton, I. Turai, M. Usel, and K. Veress. 2007. The 15-country collaborative study of cancer risk among radiation workers in the nuclear industry: Estimates of radiation-related cancer risks. Radiation Research 167(4):396–416.
Carr, Z. A., R. A. Kleinerman, M. Stovall, R. M. Weinstock, M. L. Griem, and C. E. Land. 2002. Malignant neoplasms after radiation therapy for peptic ulcer. Radiation Research 157(6):668–677.
Chang, P. Y., F. A. Cucinotta, K. A. Bjornstad, J. Bakke, C. J. Rosen, N. Du, D. G. Fairchild, E. Cacao, and E. A. Blakely. 2016. Harderian gland tumorigenesis: Low-dose and LET response. Radiation Research 185(5):449–460.
Cucinotta, F. A., H. Nikjoo, and D. T. Goodhead. 2000. Model for radial dependence of frequency distributions for energy imparted in nanometer volumes from HZE particles. Radiation Research 153(4):459–468.
Cucinotta, F. A., M.-H. Y. Kim, and L. J. Chappell. 2013. Space radiation cancer risk projections and uncertainties—2012. https://spaceradiation.jsc.nasa.gov/irModels/TP-2013-217375.pdf (accessed April 16, 2021).
Davis, F. G., J. D. Boice, Jr., Z. Hrubec, and R. R. Monson. 1989. Cancer mortality in a radiation-exposed cohort of Massachusetts tuberculosis patients. Cancer Research 49(21):6130–6136.
Desouky, O., N. Ding, and G. Zhou. 2015. Targeted and non-targeted effects of ionizing radiation. Journal of Radiation Research and Applied Sciences 8(2):247–254.
DOE (U.S. Department of Energy). 2017. The DOE ionizing radiation dose ranges chart. https://www.energy.gov/sites/prod/files/2018/01/f46/doe-ionizing-radiation-dose-ranges-jan-2018.pdf (accessed May 26, 2021).
Furukawa, K., D. L. Preston, S. Lönn, S. Funamoto, S. Yonehara, T. Matsuo, H. Egawa, S. Tokuoka, K. Ozasa, F. Kasagi, K. Kodama, and K. Mabuchi. 2010. Radiation and smoking effects on lung cancer incidence among atomic bomb survivors. Radiation Research 174(1):72–82.
Gilbert, E. S., M. Stovall, M. Gospodarowicz, F. E. Van Leeuwen, M. Andersson, B. Glimelius, T. Joensuu, C. F. Lynch, R. E. Curtis, E. Holowaty, H. Storm, E. Pukkala, M. B. van’t Veer, J. F. Fraumeni, J. D. Boice, Jr., E. A. Clarke, and L. B. Travis. 2003. Lung cancer after treatment for Hodgkin’s disease: Focus on radiation effects. Radiation Research 159(2):161–173.
Gilbert, E. S., M. E. Sokolnikov, D. L. Preston, S. J. Schonfeld, A. E. Schadilov, E. K. Vasilenko, and N. A. Koshurnikova. 2013. Lung cancer risks from plutonium: An updated analysis of data from the Mayak worker cohort. Radiation Research 179(3):332–342.
Gillies, M., I. Kuznetsova, M. Sokolnikov, R. Haylock, J. O’Hagan, Y. Tsareva, and E. Labutina. 2017. Lung cancer risk from plutonium: A pooled analysis of the Mayak and Sellafield worker cohorts. Radiation Research 188(6):645–660.
Golden, A. P., E. D. Ellis, S. S. Cohen, M. T. Mumma, R. W. Leggett, P. W. Wallace, D. Girardi, J. P. Watkins, R. E. Shore, and J. D. Boice. 2019. Updated mortality analysis of the Mallinckrodt uranium processing workers, 1942–2012. International Journal of Radiation Biology 17:1–21.
Goodhead, D. T. 1994. Initial events in the cellular effects of ionizing radiations: Clustered damage in DNA. International Journal of Radiation Biology 65(1):7–17.
Gueulette, J., H. Blattmann, E. Pedroni, A. Coray, B. M. De Coster, P. Mahy, A. Wambersie, and G. Goitein. 2005. Relative biologic effectiveness determination in mouse intestine for scanning proton beam at Paul Scherrer Institute, Switzerland. Influence of motion. International Journal of Radiation Oncology, Biology, Physics 62(3):838–845.
Hada, M., and B. M. Sutherland. 2006. Spectrum of complex DNA damages depends on the incident radiation. Radiation Research 165(2):223–230.
Haylock, R. G. E., M. Gillies, N. Hunter, W. Zhang, and M. Phillipson. 2018. Cancer mortality and incidence following external occupational radiation exposure: An update of the 3rd analysis of the U. K. national registry for radiation workers. British Journal of Cancer 119(5):631–637.
Held, K. D. 2009. Effects of low fluences of radiations found in space on cellular systems. International Journal of Radiation Biology 85(5):379–390.
Heymann, S., S. Delaloge, A. Rahal, O. Caron, T. Frebourg, L. Barreau, C. Pachet, M. C. Mathieu, H. Marsiglia, and C. Bourgier. 2010. Radio-induced malignancies after breast cancer postoperative radiotherapy in patients with Li-Fraumeni syndrome. Radiation Oncology 5:104.
Howe, G. R. 1995. Lung cancer mortality between 1950 and 1987 after exposure to fractionated moderate-dose-rate ionizing radiation in the Canadian fluoroscopy cohort study and a comparison with lung cancer mortality in the atomic bomb survivors study. Radiation Research 142(3):295–304.
Howlader, N., A. Noone, M. Krapcho, D. Miller, A. Brest, M. Yu, J. Ruhl, Z. Tatalovich, A. Mariotto, D. Lewis, H. Chen, E. Feuer, and K. Cronin. 2021. SEER cancer statistics review, 1975–2018. Bethesda, MD: National Cancer Institute.
IARC (International Agency for Research on Cancer). 2012. Radiation: IARC monographs on the evaluation of carcinogenic risks to humans, Vol. 100D. https://publications.iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-The-Identification-Of-CarcinogenicHazards-To-Humans/Radiation-2012 (accessed April 28, 2021).
IARC. 2017. Cancer incidence in five continents, Vol. XI (electronic version), edited by F. Bray, M. Colombet, L. Mery, M. Piñeros, A. Znaor, R. Zanetti, and J. Ferlay. Lyon, France: International Agency for Research on Cancer. https://ci5.iarc.fr (accessed March 17, 2021).
ICRP (International Commission on Radiological Protection). 2007. The 2007 recommendations of the International Commission on Radiological Protection. ICRP publication 103. Annals of the ICRP 37(2–4):1–332.
ICRP. 2010. ICRP publication 115. Lung cancer risk from radon and progeny and statement on radon. Annals of the ICRP 40(1):1–64.
Katz, R., B. Ackerson, M. Homayoonfar, and S. C. Sharma 1971. Inactivation of cells by heavy ion bombardment. Radiation Research 47(2):402–425.
Kronenberg, A., and F. A. Cucinotta. 2012. Space radiation protection issues. Health Physics 103(5):556–567.
Little, M. P., M. Stovall, S. A. Smith, and R. A. Kleinerman. 2013. A reanalysis of curvature in the dose response for cancer and modifications by age at exposure following radiation therapy for benign disease. International Journal of Radiation Oncology, Biology, Physics 85(2):451–459.
Mason, K. A., M. T. Gillin, R. Mohan, and J. D. Cox. 2007. Preclinical biologic assessment of proton beam relative biologic effectiveness at Proton Therapy Center Houston. International Journal of Radiation Oncology, Biology, Physics 68(4):968–970.
Morton, L. M., D. M. Karyadi, S. W. Hartley, M. N. Frone, J. N. Sampson, R. M. Howell, J. P. Neglia, M. A. Arnold, B. D. Hicks, K. Jones, B. Zhu, C. L. Dagnall, E. Karlins, M. S. Yeager, W. M. Leisenring, Y. Yasui, L. M. Turcotte, S. A. Smith, R. E. Weathers, J. Miller, B. S. Sigel, D. M. Merino, A. Berrington de González, S. Bhatia, L. L. Robison, M. A. Tucker, G. T. Armstrong, and S. J. Chanock. 2020. Subsequent neoplasm risk associated with rare variants in DNA damage response and clinical radiation sensitivity syndrome genes in the childhood cancer survivor study. JCO Precision Oncology 4:926–936.
Muirhead, C. R., J. A. O’Hagan, R. G. Haylock, M. A. Phillipson, T. Willcock, G. L. C. Berridge, and W. Zhang. 2009. Mortality and cancer incidence following occupational radiation exposure: Third analysis of the National Registry for Radiation Workers. British Journal of Cancer 100(1):206–212.
NASA (National Aeronautics and Space Administration). 2014. NASA spaceflight human system standard, Vol. 1, revision A: Crew health. NASA-STD-3001. https://standards.nasa.gov/standard/nasa/nasa-std-3001-vol-1 (accessed April 16, 2021).
NCRP (National Council on Radiation Protection and Measurements). 2000. Report 132: Radiation protection guidance for activities in low-Earth orbit. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2011. NCRP report no. 170: Second primary cancers and cardiovascular disease after radiation therapy. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2012. Report no. 171: Uncertainties in the estimation of radiation risks and probability of disease causation. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2014. Commentary 23: Radiation protection for space activities: Supplement to previous recommendations. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2018a. Implications of recent epidemiologic studies for the linear-nonthreshold model and radiation protection. NCRP Commentary 27. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2018b. Report 180: Management of exposure to ionizing radiation: Radiation protection guidance for the United States. Bethesda, MD: National Council on Radiation Protection and Measurements.
NCRP. 2019. SC 1-27: Evaluation of sex-specific differences in lung cancer radiation risks and recommendations for use in transfer and projection models. https://ncrponline.org/program-areas/sc-1-27-evaluation-of-sex-specific-differences-in-lung-cancer-radiation-risks-and-recommendations-for-use-in-transfer-and-projection-models (accessed April 16, 2021).
NCRP. 2020. Report no. 186: Approaches for integrating information from radiation biology and epidemiology to enhance low-dose health risk assessment. Bethesda, MD: National Council on Radiation Protection and Measurements.
Nikjoo, H., P. O’Neill, M. Terrissol, and D. T. Goodhead., 1999. Quantitative modeling of DNA damage using Monte Carlo track structure method. Radiation and Environmental Biophysics 38(1):31–38.
NRC (National Research Council). 1999. Health effects of exposure to radon: BEIR VI. Washington, DC: National Academy Press.
NRC. 2006. Health risks from exposure to low levels of ionizing radiation: BEIR VII phase 2. Washington, DC: The National Academies Press.
NRC. 2012. Technical evaluation of the NASA model for cancer risk to astronauts due to space radiation. Washington, DC: The National Academies Press.
Ozasa, K., Y. Shimizu, A. Suyama, F. Kasagi, M. Soda, E. J. Grant, R. Sakata, H. Sugiyama, and K. Kodama. 2012. Studies of the mortality of atomic bomb survivors, report 14, 1950–2003: An overview of cancer and noncancer diseases. Radiation Research 177(3):229–243.
Ozasa, K., E. J. Grant, and K. Kodama. 2018. Japanese legacy cohorts: The Life Span Study atomic bomb survivor cohort and survivors’ offspring. Journal of Epidemiology 28(4):162–169.
Pawel, D. J. 2013. U.S. Environmental Protection Agency radiogenic risk projections: Uncertainty analysis. Health Physics 104(1):26–40.
Pierce, D. A., G. B. Sharp, and K. Mabuchi. 2003. Joint effects of radiation and smoking on lung cancer risk among atomic bomb survivors. Radiation Research 159(4):511–520.
Richardson, D. B., E. Cardis, R. D. Daniels, M. Gillies, R. Haylock, K. Leuraud, D. Laurier, M. Moissonnier, M. K. Schubauer-Berigan, and I. Thierry-Chef. 2018. Site-specific solid cancer mortality after exposure to ionizing radiation: A cohort study of workers (inworks). Epidemiology 29(1):31.
Rydberg, B. 1996. Clusters of DNA damage induced by ionizing radiation: Formation of short DNA fragments. II. Experimental detection. Radiation Research 145(2):200–209.
Schimmerling, W. 2011. The space radiation environment: An introduction. https://three.jsc.nasa.gov/concepts/SpaceRadiationEnviron.pdf (accessed May 26, 2021).
Shuryak, I., A. J. Fornace, Jr., K. Datta, S. Suman, S. Kumar, R. K. Sachs, and D. J. Brenner. 2017. Scaling human cancer risks from low LET to high LET when dose-effect relationships are complex. Radiation Research 187(4):486–492.
Siegel, R. L., K. D. Miller, H. E. Fuchs, and A. Jemal. 2021. Cancer statistics, 2021. CA: A Cancer Journal for Clinicians 71(1):7–33.
Silver, S. R., S. J. Bertke, M. J. Hein, R. D. Daniels, D. A. Fleming, J. L. Anderson, S. M. Pinney, R. W. Hornung, and C. Y. Tseng. 2013. Mortality and ionising radiation exposures among workers employed at the Fernald Feed Materials Production Center (1951–1985). Occupational and Environmental Medicine 70(7):453–463.
Simonsen, L. C., and J. E. Nealy. 1991. Radiation protection for human missions to the Moon and Mars. https://www.hq.nasa.gov/pao/History/alsj/WOTM/NASA_TP_3079.pdf (accessed May 14, 2021).
Simonsen, L. C., and T. C. Slaba. 2020. Ensemble methodologies for astronaut cancer risk assessment in the face of large uncertainties. https://ntrs.nasa.gov/api/citations/20205008710/downloads/NASA-TP-20205008710.pdf (accessed April 28, 2021).
Suman, S., K. Datta, D. Trani, E. C. Laiakis, S. J. Strawn, and A. J. Fornace. 2012. Relative biological effectiveness of 12C and 28Si radiation in C57BL/6J mice. Radiation and Environmental Biophysics 51(3):303–309.
Suman, S., S. Kumar, B. H. Moon, S. J. Strawn, H. Thakor, Z. Fan, J. W. Shay, A. J. Fornace, Jr., and K. Datta. 2016. Relative biological effectiveness of energetic heavy ions for intestinal tumorigenesis shows male preponderance and radiation type and energy dependence in APC1638N/+ mice. International Journal of Radiation Oncology, Biology, Physics 95(1):131–138.
Thompson, D. E., K. Mabuchi, E. Ron, M. Soda, M. Tokunaga, S. Ochikubo, S. Sugimoto, T. Ikeda, M. Terasaki, S. Izumi, and D. L. Preston. 1994. Cancer incidence in atomic bomb survivors. Part II: Solid tumors, 1958–1987. Radiation Research 137(2):S17–S67.
Thun, M. J., B. D. Carter, D. Feskanich, N. D. Freedman, R. Prentice, A. D. Lopez, P. Hartge, and S. M. Gapstur. 2013. 50-year trends in smoking-related mortality in the United States. New England Journal of Medicine 368(4):351–364.
UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). 2008. Effects of ionizing radiation, 2006 report, volume I: Report to the General Assembly, scientific annexes A and B. https://www.unscear.org/unscear/en/publications/2006_1.html (accessed April 15, 2021).
UNSCEAR. 2015. Report of the United Nations Scientific Committee on the Effects of Atomic Radiation to the General Assembly. https://www.unscear.org/docs/reports/2012/UNSCEAR2012Report_AnnexB_Uncertainty_AdvanceCopy.pdf (accessed May 14, 2021).
UNSCEAR. 2019. Sources, effects and risks of ionizing radiation. https://www.unscear.org/docs/publications/2019/UNSCEAR_2019_Annex-A.pdf (accessed May 26, 2021).
USNRC (United States Nuclear Regulatory Commission). 2021. Lethal dose (LD). https://www.nrc.gov/reading-rm/basic-ref/glossary/lethal-dose-ld.html (accessed May 26, 2021).
Velazquez-Kronen, R., E. S. Gilbert, M. S. Linet, K. B. Moysich, J. L. Freudenheim, J. Wactawski-Wende, S. L. Simon, E. K. Cahoon, B. H. Alexander, M. M. Doody, and C. M. Kitahara. 2020. Lung cancer mortality associated with protracted low-dose occupational radiation exposures and smoking behaviors in U. S. radiologic technologists, 1983–2012. International Journal of Cancer 147(11):3130–3138.
Wand, H., S. A. Lambert, C. Tamburro, M. A. Iacocca, J. W. O’Sullivan, C. Sillari, I. J. Kullo, R. Rowley, J. S. Dron, D. Brockman, E. Venner, M. I. McCarthy, A. C. Antoniou, D. F. Easton, R. A. Hegele, A. V. Khera, N. Chatterjee, C. Kooperberg, K. Edwards, K. Vlessis, K. Kinnear, J. N. Danesh, H. Parkinson, E. M. Ramos, M. C. Roberts, K. E. Ormond, M. J. Khoury, A. C. J. W. Janssens, K. A. B. Goddard, P. Kraft, J. A. L. MacArthur, M. Inouye, and G. Wojcik. 2021. Improving reporting standards for polygenic scores in risk prediction studies. Nature 591:211–219.
Zeitlin, C., D. M. Hassler, F. A. Cucinotta, B. Ehresmann, R. F. Wimmer-Schweingruber, D. E. Brinza, S. Kang, G. Weigle, S. Böttcher, E. Böhm, S. Burmeister, J. Guo, J. Köhler, C. Martin, A. Posner, S. Rafkin, and G. Reitz. 2013. Measurements of energetic particle radiation in transit to mars on the Mars Science Laboratory. Science 340(6136):1080–1084.