3
Radiation Effects

EFFECTS ON HUMANS

Some Elementary Concepts in Radiation Biology

In order to understand the mechanisms by which human beings are threatened by exposure to radiation, one must consider a living organism as any other system, in terms of its parts and how well they work together. An airplane, which is one of the more complicated achievements of engineering, could be described as 50,000 parts flying in formation. A human being is approximately 1014 parts, not only flying in formation but also engaged in parallel processing and generation of information. An airplane cannot safely miss any parts (bolts, wires, seats) or parts of systems (metal skin on wings, fuel tank hoses, computers); similarly, an organism cannot safely miss too many cells in its brain or its heart, nor survive the loss of essential tissues and organs. Furthermore, while understanding the effect of a given scenario on one airplane can be reasonably extended to other airplanes, the effects of radiation on human beings are different from person to person. Radiation biologists must take into consideration interindividual differences, as well as determine if effects on individual systems in isolation are consistent with the effects on an organism as a whole. However, an airplane cannot repair itself, reproduce itself, or decide where it wants to fly. A human being can do all those things, and many more.

The parts of a living organism are the cells and their aggregation into correlated groups, that is, tissues and organs. Two minimal conditions must be met for the organism to be healthy: all the critical parts must be there, and they all must function together properly. The number of cells required for proper tissue and organ function is determined by the ability of cells to divide and maintain their structure. The function of tissues and organs is determined by the ability of constituent cells to keep sending and receiving the required signaling molecules. In a healthy organism, all the required cells are there, and talk to each other in a timely and businesslike manner. All of these aspects of living systems can be perturbed, in many cases permanently, by exposure to radiation, as summarized in Figure 3-1.

When an organism is exposed to radiation, the energy from the radiation is deposited at the cellular level by interactions between the radiation and the electrons of molecules making up the cells. As a consequence, the carbon, oxygen, nitrogen, and other atoms that make up complex molecules may lose the electron bonds that tie them to the rest of the molecule. In many cases, they don’t move very far, and other electrons can be used to re-establish the chemical bonds. In some cases, however, the molecule does not recover. In the case of template molecules such as deoxyribonucleic acid (DNA), this changes the cell’s ability to manufacture the proper signals at the proper time.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 49
3 Radiation Effects EFFECTS ON HUMANS Some Elementary Concepts in Radiation Biology In order to understand the mechanisms by which human beings are threatened by exposure to radiation, one must consider a living organism as any other system, in terms of its parts and how well they work together. An airplane, which is one of the more complicated achievements of engineering, could be described as 50,000 parts flying in formation. A human being is approximately 1014 parts, not only flying in formation but also engaged in parallel processing and generation of information. An airplane cannot safely miss any parts (bolts, wires, seats) or parts of systems (metal skin on wings, fuel tank hoses, computers); similarly, an organism cannot safely miss too many cells in its brain or its heart, nor survive the loss of essential tissues and organs. Furthermore, while under- standing the effect of a given scenario on one airplane can be reasonably extended to other airplanes, the effects of radiation on human beings are different from person to person. Radiation biologists must take into consideration interindividual differences, as well as determine if effects on individual systems in isolation are consistent with the effects on an organism as a whole. However, an airplane cannot repair itself, reproduce itself, or decide where it wants to fly. A human being can do all those things, and many more. The parts of a living organism are the cells and their aggregation into correlated groups, that is, tissues and organs. Two minimal conditions must be met for the organism to be healthy: all the critical parts must be there, and they all must function together properly. The number of cells required for proper tissue and organ function is determined by the ability of cells to divide and maintain their structure. The function of tissues and organs is determined by the ability of constituent cells to keep sending and receiving the required signaling molecules. In a healthy organism, all the required cells are there, and talk to each other in a timely and businesslike manner. All of these aspects of living systems can be perturbed, in many cases permanently, by exposure to radiation, as summarized in Figure 3-1. When an organism is exposed to radiation, the energy from the radiation is deposited at the cellular level by interactions between the radiation and the electrons of molecules making up the cells. As a consequence, the carbon, oxygen, nitrogen, and other atoms that make up complex molecules may lose the electron bonds that tie them to the rest of the molecule. In many cases, they don’t move very far, and other electrons can be used to re-establish the chemical bonds. In some cases, however, the molecule does not recover. In the case of template molecules such as deoxyribonucleic acid (DNA), this changes the cell’s ability to manufacture the proper signals at the proper time. 49

OCR for page 49
50 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION ORGANISM Cell Function Cell Division Undisturbed cells: Normal Normal Yes Irradiated cells: Repair? Dead cells Tissue damage - Early effects Programmed - Late effects No cell death No Cancer Other late effects? Mutations FIGURE 3-1 Pathways of biological damage produced by exposure to radiation. R01155, Figure 3-1 An irradiated cell may be able to repair any damage that it has suffered, using one or more of several chemi- cal pathways available in nature. If it repairs itself correctly, then, by definition, it is indistinguishable from a normal cell and goes back to its undisturbed state. If the cell does not repair the damage, it may die. In that case, it is removed from the system. In principle, this is a good thing, because a dead cell cannot become a cancer cell. However, if too many cells of a tissue die, organ function will be compromised. If sensitive cells, for example, cells in the gut, die in large enough numbers, then the gut cannot absorb food or maintain electrolyte balance. This is why, after a dose of radiation killing a large enough number of cells, nausea and vomiting set in. However, if the radiation dose is delivered over a period of time that is long compared with the repair time constant of the cells, the cells can repair and maintain, or delete and replace, a sufficient number of cells for function to be undisturbed. These different radiation time courses are referred to as “early” and “late,” to describe how the body responds to different dose rates. Cellular repair mechanisms are not always fully successful. In some cases, repair can leave the cell in suf- ficiently good shape to limp through another few cell divisions, thus maintaining the number of cells for a while. However, the daughter cells will inherit some of the original, incompletely or poorly repaired damage and die off or lead to dying or aberrant cells in subsequent divisions. This unstable state of the cell and its progeny is often called genomic instability, but is not fully understood and is likely to encompass a much broader range of phenomena, including possible responses to molecular signals from nearby damaged cells. In any case, the even- tual death of cells is also good for the organism, again since dead cells cannot initiate cancer. However, just as in the case of acute effects, the loss of too many cells will compromise tissue and organ function, possibly leading to serious consequences and even death. While these effects are similar to acute effects, the fact that they do not

OCR for page 49
51 RADIATION EFFECTS occur more or less immediately during or after irradiation means that other biological processes can be signifi- cant, and thus the consequences can be different. Chronic effects of radiation appear to involve the induction of a cytokine cascade as a response to tissue damage. This often leads to the replacement of a functional cell type with fibrocytes, thus causing fibrosis and other dysfunctional processes. Other undefined mechanisms may also contribute to late effects. However, ionizing radiation injures normal cells through various molecular pathways. In general, the radia- tion sensitivity of a given tissue, and in turn of a given organ, depends on the radiation sensitivity of the key cells in the system, although there appears to be much interaction among the tissues, so that effects on one tissue can have impact on other organ systems in the body. Also important in understanding radiation sensitivity studies are several physical and biological variables: dose size, dose mode (internal or external), dose rate, fractionation (division of the total dose into small doses administered at intervals), the size of the irradiated field, the time of observation after exposure, and the condition of the stroma and vascular supply. If full repair of cells fails, but not to the point of leading to the death of subsequent generations of cells, damaged cells may survive and transform into cells that can become cancer precursors or change further and develop into tumors. Even if a damaged cell does not evolve into cancerous tissue, its communication with other cells may be impaired to the point that it cannot provide adequate limiting or stimulating signals for proper function within a given tissue. For example, most higher organisms carry a complement of reserve cells. These “stem” cells can differentiate into cells of any description, and can replace them. If stem cells are damaged, they may not properly differentiate and learn to function in a given tissue. For instance, cells in blood vessels in the brain may lose properties required to keep blood contained, possibly leading to stroke. The use of stem cell replacement therapy, especially for bone marrow, has been considered by NASA for many years, but the difficulties associated with performing such procedures in space have been considered too great to overcome. The delayed effects of radiation are not independent of the initial events, as any damage to cells may take time to manifest, after they have reproduced, interacted with other cells, and undergone evolution of whole-tissue responses. Figure 3-2 shows the time course of radiation injury for somatic cells, thus excluding genetic and heredi- tary aspects, which are less well understood but also can be assumed to follow their own, complex time course. FIGURE 3-2 Temporal relationships among somatic effects. SOURCE: Fajardo et al., 2001. Copyright 2001; reprinted with permission from Oxford University Press. R01155, Figure 3-2 Fixed image, not changeable

OCR for page 49
52 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION Finally, it is useful to distinguish between effects on the properties of cell systems, such as tissues and organs, and the consequences of modifying single cells. The function of tissues and organs persists to some extent, even when cells are lost. It is only when a substantial number of cells are missing that the tissue structure or the organ function cannot continue. Thus, tissue and organ effects are characterized by a threshold beyond which the loss of function is irreversible. The threshold will be characterized by a step function with a slope that depends on the number of cells required for minimal function, their replacement, the rate of loss of support structure and fluids, external intervention, and so on. The term for these effects is “deterministic,” to indicate that they will occur inevitably once a certain dose of radiation is delivered. As an example, Figure 3-3 shows the calculated probability of radiation exposure leading to death as a function of radiation dose and medical treatment. The acute response curves in the figure show that thresholds are generally quite sharp, so that any uncertainty in the dose (or in con- version of the dose into a biological common scale) is quickly amplified by the response probability, and a small probability may, in actual fact, be significantly larger. To the contrary, the degree of repair and the transformation of a single cell are stochastic variables, described by a probability distribution. Particularly at the lowest doses of radiation, which are of greatest interest for the consideration of occupational radiation exposure and where no deterministic effects are likely to be observed, radiation exposure may lead a cell to the initial stage of cancer, or may lead a cell to induce another cell to do so. These processes are modeled probabilistically, which indicates their vastly different scientific complexity. Figure 3-3 shows an estimated nonzero probability even for the smallest doses. Cancer risks are not measured but calculated using absorbed dose from radiation, following prescriptions such as are given in NCRP Report 132 (NCRP, 2000). Such calculations are a function of a multitude of factors, and the uncertainties in the knowledge of these factors propagate to result in uncertainties in the calculated risk. Thus, the risk is given by a probability distribution function (PDF) and not by a point estimate. The uncertainties in these calculations can be estimated and the PDF evaluated for various radiation environments (NCRP, 1997; Cucinotta et al., 2001) to obtain probability distributions for the risk incurred at a given dose. As an example, Figure 3-4 FIGURE 3-3 Risk versus dose for different levels of medical treatment. NOTE: Although this figure is based on work of the National Council on Radiation Protection and Measurements (NCRP), advances in supportive care have dramatically changed the shape of the supportive curve. It is much closer to that seen for intensive care. SOURCE: Adapted from NCRP, 1989. R01155, Figure 3-3 Fixed image, not changeable

OCR for page 49
53 RADIATION EFFECTS (H2) (H2) (H2) FIGURE 3-4 Probability density functions (PDFs) for 40-year-old males on a solar minimum Mars swing-by mission behind 20-g/cm2 shields of aluminum, polyethylene, or liquid hydrogen. Effective doses, point estimates, and 95 percent confidence interval for risk of exposure-induced death (REID) are shown in the inset. SOURCE: Cucinotta et al., 2005. R01155, Figure 3-4 shows the PDFs representing uncertainties in projecting fatal cancer risk (in the form of risk of exposure induced Fixed image, not changeable death; REID) associated with a Mars swing-by mission (600 days), for three different kinds of shielding. The area under the curve represents the cumulative probability of that value for REID. In the case of aluminum, half the area under the curve lies to the left of REID = 3.2% (the other half lies to the right); only 5% of the area under the curve lies to the right of REID = 10.5%, making that the 95% confidence-level upper limit. Similarly, only 5% of the area lies to the left of REID = 1.0%, making that the 95% confidence-level lower limit. These PDFs are gener- ated by using Monte Carlo methods to propagate risk from the component biological and physical uncertainties through the risk estimate process. Similar calculations have been made for GCR at various dose levels (Cucinotta et al., 2001), using quality factors and extrapolation from low-linear-energy-transfer (LET) epidemiology studies, and on mechanistic insights from ground-based studies conducted at facilities such as the NASA Space Radiation Laboratory (NSRL). The use of a linear extrapolation without a threshold, as is customary in the interpretation of the atomic bomb survivor data on which epidemiological dose response is based, means that, at any given dose, there is a definite probability of incurring a fatal cancer, even if the mean (or any other point estimate) may correspond to a minor risk. At higher values of LET, the mean risk of fatal cancer increases, following the rise of the quality factor. Only at the highest values of LET, or for high doses at any LET, the mean risk begins to decrease, as a consequence of the decreasing survival of highly irradiated cells. However, very high values of LET for high atomic number and energy (HZE) particles do not necessarily translate to decreased cell survival. Because of the track structure at large velocities (i.e., β = 0.88 for Fe with a LET of 150 keV/μm), many cells may only receive small doses of electrons (delta rays) that are not necessarily fatal. Despite recent progress made in the area of space radiobiology, especially in studies relevant to carcinogenesis, many uncertainties remain about the specific biological effects of extended GCR and solar particle event (SPE)

OCR for page 49
54 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION exposure. An example of estimates of the probability of REID, presented by NASA, is found in Table 1-4, where the uncertainty ratio (the ratio between the upper end of the confidence interval and the mean risk prediction) is generally around 3. Much of what is known of the biological effects of space radiation has come out of NASA’s Space Radiation Biology Research program. The strategy of this research program was based on a rational calculation that break- throughs in biology have occurred at a rapid rate, reflecting the status of biology as the cutting edge of science. Early in the program’s conception, Curtis et al. (1995) estimated the radiation risk uncertainty ratio to be 4-15×. A National Research Council (NRC) report (NRC, 1996) estimated that the likely time for reducing these uncertainties would require a timescale extending beyond the working lifetime of program scientists. Accordingly, the program budget and level of effort were increased so that, based solely on work within this program, the uncertainty could be halved within 10 to 15 years. However, the calculation that breakthroughs would occur inevitably in a fast- moving, revolutionized science field like biology led to the expectation that even greater reductions are possible within that time frame. In the year 2000, the uncertainty ratio was recalculated to be as high as 6× for a 1,000-day mission to Mars (Cucinotta et al., 2001). Several uncertainty ratio reduction curves are shown in Figure 3-5. The curve corresponding to the uncertainty ratio dropping by half over 30 years (representing the progress before the program was accelerated) is shown together with a curve assuming that the uncertainty ratio can be halved in 15 years under continuous progress, as estimated for the increase in funded research. However, using an assump- tion, for illustrative purposes, that breakthroughs that can help reduce the uncertainty by a factor of 2 occur about once every 5 years, it may be seen that a reduction in the uncertainty ratio to 1.5× would be achievable around 2015; 1.5× (meaning that risk is estimated to within ±50 percent) is an estimate of the best that one could expect to do about the risk uncertainty; once this level is attained, research can be more productively focused elsewhere. Some examples of breakthroughs, available in 2005, were as follows: • Cancer susceptibility genes for which genetic testing is available have been discovered; examples include the gene for hereditary retinoblastoma, the breast cancer genes BRCA1 and BRCA2, genes associated with some types of colon cancer, and a gene that is involved in susceptibility to cancer development and is associated with radiation sensitivitythe gene for ataxia telangiectasia. • Signal transduction pathways have been defined that link cellular communication systems and result in altered gene expression and altered cellular phenotypes; new information on such pathways is being obtained on an almost daily basis. Uncertainty ratio FIGURE 3-5 Evolution of risk uncertainty for Mars exploration. SOURCE: Adapted from NASA, 1998. R01155, Figure 3-5 Fixed image, not changeable, except “uncertainty ratio”

OCR for page 49
55 RADIATION EFFECTS • Mechanisms associated with the tumor suppressor gene TP53 have been discovered, including apoptosis or programmed cell death. Modulation of the cell cycle in mammalian cells by cyclin-dependent kinases and a general understanding of DNA damage checkpoints at the phase boundaries of the cell cycle have drastically improved our understanding of the response of cells to radiation. It is not possible to predict when breakthroughs such as those listed above will occur, but it is possible to predict that they will happen, based on current experience in the biological sciences. NASA can ensure that these breakthroughs are applied in a timely and cost-effective manner to the space radiation problem by ensuring the existence of a scientific community with a critical mass of investigators who are active at the cutting edge of the field, who understand the nuances of space radiation biology, and who know NASA needs well enough to leverage discoveries. Considerable progress has been made by NASA in several areas: • Substantial increments in GCR data and the development of accurate models have reduced the uncertainty in predictions of the interplanetary GCR environment to an estimated 10 to 15 percent (Cucinotta et al., 2005). • The acquisition of data on the physics of nuclear interactions, in combination with refinements in transport codes, has led to an uncertainty in shielding calculations estimated to be 30 to 50 percent (NRC, 1996; based on Schimmerling et al., 1987, 1989; and Shavers et al., 1993). • Data obtained in laboratory radiation research, leading to a better understanding of quality factors; continu- ing analyses of health effects on cohorts exposed to radiation; and refinements in the probabilistic analysis of space radiation risk seem to indicate that NASA is roughly on course in its strategy to reduce uncertainty along the lines of the curve labeled “T1/2 = 15 years with breakthroughs every 5 years” in Figure 3-5 (Cucinotta et al., 2005). However, at the present time it is premature to make numerical estimates of decreased risk uncertainty with time, given the knowledge gaps listed below. Recent radiation biology research has demonstrated that there are still fundamental responses of cells and tissues to radiation that are not clearly understood and that might affect risk assessments. Rather than uniformly decreasing the uncertainties associated with risk estimates, some discoveries may reduce the uncertainties and may even reveal that the risks are currently overestimated; conversely, some new discoveries may reveal increased risks, or increased uncertainty about risks thought to be well understood at present. For example, in the past several years investigators have identified a bystander effect to radiation, where unirradiated cells in the neighborhood of irradiated cells in cell cultures show responses including genetic instabil- ity, chromosomal abnormalities, and the induction of radiation response genes. The mechanism(s) responsible for this bystander effect have not yet been elucidated, but they clearly involve damaging consequences to unirradiated cells that could impact cell survival, mutation induction, and thus tissue responses. While the bystander effect has been detected following exposure to both alpha particles and low-LET radia- tions, it is also likely to be induced by GCR and SPEs. For HZE exposure, the notion of bystander cells needs to be modified slightly, to refer to cells that are either irradiated by the extended distribution of delta rays (i.e., cells not directly traversed) as well as cells participating in subsequent biochemical signaling pathways. In the situation of low-fluence particle exposures, bystander responses could contribute significantly to overall risk by effectively increasing the number of cells subject to damage far above the number of cells directly traversed by a particle. Nearly all bystander experimental research has been done in cell culture. Biological responses such as these have yet to be demonstrated to occur in normal tissues, and conclusions drawn from cell culture research should be validated in a normal tissue context where possible. Further research in this area and other poorly understood bio- logical responses (such as delayed mutation induction) may continue to complicate our understanding of radiation responses, possibly leading to further increases in uncertainties before these uncertainties can be decreased. Knowledge Gaps As noted above, the risks incurred as a consequence of exposure to space radiation are not measured; instead, they are calculated using available information. This information may consist of measured or archived data on the

OCR for page 49
56 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION radiation environment and the physical properties of its components, as well as knowledge about the consequences of particular levels and microscopic patterns of energy deposition in human tissues and organs. As a consequence, the expression of the uncertainties in risk prediction depend on and incorporate uncertainties in the particular theoretical frameworkthe modelused in the calculation. The model generally used for establishing radiation limits is based on the detailed epidemiological and medi- cal observations of atomic bomb survivors, supplemented, where appropriate, with data from occupational radia- tion, medical, and accidental radiation exposures. Dose-response curves from atomic bomb survivor data, mainly survivors exposed to relatively high doses of instantaneously delivered gamma rays, are used in conjunction with other information to extrapolate the probability of health effects to lower doses, delivered over long periods of time, to peacetime populations other than Japanese survivors, and to other kinds of radiation. Significant improve- ments in the accuracy of risk prediction at low doses are not likely in the context of this phenomenological and epidemiological modeling alone. Compilations of data for radiation workers or radiation accidents have not yet provided a level of quality comparable to the data on Japanese survivors. Radiation workers generally receive very low exposures, so that epidemiological studies are hampered by poor statistics even more than studies of Japanese survivors. Dosimetry for victims of radiation accidents, especially individuals with low doses, is often extremely uncertain. Much of the low-dose data is from laboratory experiments and not for cancer epidemiology. However, various learned bodies, such as the NCRP and the Biological Effects of Ionizing Radiation (BEIR) study groups follow the development of databases by the U.S. Department of Energy and other agencies and may eventually be able to use such data to improve epidemiological estimates based on Japanese survivors. However, this is not yet the case. Improvements in risk prediction, especially for space radiation, could be made by linking mechanisms of cellular and molecular processing of radiation damage to macroscopic processes at the tissue, organ, and organism levels. There are a finite number of biological functions in cells, and nature is economical—it tends to use the same chemical reactions in many different contexts rather than devise a new chemical process for every occasion. Completion of the human genome sequence and the rise of methods for massively parallel expression experi- ments have enabled the development of systems biology, a new approach to biological problems that seeks to understand the whole organism through integrating information about all of its biochemical processes and gene and protein network interactions. This systems biology approach is being developed as an alternative to current molecular models and may provide the next major breakthrough in biological science. These studies are likely to complement ongoing experimental studies by providing additional models for understanding molecular events in the context of whole organisms. In the short term, the same insights from cellular and tissue radiobiology that are laying the basis for systems biology also impact the existing models of radiation injury and are required to address existing knowledge gaps. In the long term, integrating the explosive development of systems biology into epidemiological studies and space radiation risk management will depend on the support of a science community of investigators at institutions where this work is proceeding. Results of this research will be important in securing the sustainability of a space research program beyond the first few missions. The knowledge gaps fit into five categories: 1. Carcinogenesis: Excess lifetime morbidity and mortality risk from radiation-induced cancers. Radiation quality and susceptibility are likely to be factors in the epidemiological modeling of cancer incidence. Population- averaged values do not account for dispersion due to genetic factors (familial, high-and low-penetrance genes, single-nucleotide polymorphisms). However, genetic susceptibility dependence of radiation risks may be invariant to radiation quality. Neutron carcinogenesis studies show relative biological effectiveness (RBE) variations across mouse strains for the same tissue; however, similar studies do not exist for HZE. Radiation susceptibility, by which certain individuals are more likely than others to develop cancer in response to radiation exposure, is unlikely to be due to the same mechanism in all individuals concerned. Different factors may lead to different injury end points. It is well known that some individuals have genetic predispositions for breast cancer, others for colon cancer, and so on. An understanding of the various cellular signal pathways leading to cancer in different tissues and organs might enable the selection of diagnostic techniques for earlier detection or for more successful intervention, in both cases reducing the career risk.

OCR for page 49
57 RADIATION EFFECTS 2. Neurological damage (central nerous system, CNS): acute or late modifications to neurological perfor- mance, fatigue, or increased Alzheimer’s or other late effects. One aspect of CNS exposure that has been suggested is a possible link between a loss of stem cells and compromised recovery from trauma. In addition, several studies have shown that there are CNS effects for which the relationship between LET and RBE is complex or nonexistent. Therefore, current risk assessments for such effects may be invalid (Rabin et al., 2007; F. Cucinotta, NASA. “Radiation Risk Assessments for Lunar Missions—Shielding Evaluation Criteria,” presented to the committee on December 12, 2006). Recent studies suggest that contrary to previous assumptions, the CNS may be a relatively radiosensitive organ, on the same order as the gastrointestinal tract. The CNS may even be affected in the absence of direct irradiation when other tissues have been irradiated. 3. Degeneratie tissue diseases related to accelerated aging: including death from heart, circulatory, and digestie diseases, or morbidity from these diseases, cataracts, and others diseases. Very few of these degenera- tive diseases are being actively examined from a research perspective, although increasingly more information is pointing to the importance of radiation in affecting these organ systems. The question of a dose-threshold for degenerative tissue effects has not been resolved. Although some earlier studies by Yang and Ainsworth (1982) showed no threshold for HZE ions, the atomic bomb studies have shown cardiovascular disease effects at doses at 50 cSv as well as dose-response effects on various immunological markers of inflammation. In addition, there remain serious questions about modeling noncancer risks. Using epidemiological models, it is not as clear as with cancer how risks transfer across populations, whether there is a dose and dose- rate dependence, and what the quality factors are for specific risks. Even if these characteristics can be modeled satisfactorily, there remains the vexing question of how cancer and noncancer mortality effects should be weighted if they lead to different amounts of life-loss. A number of studies on cataract induction have raised concerns about the accelerated onset of cataracts and the possible consequences for very long term missions. While dose thresholds have been established based on medical exposures, there is some evidence that cataracts have features that are similar to stochastic effects. 4. Acute radiation syndromes leading to nausea, omiting, or death that could occur following a solar particle eent. Acute radiation effects generally have a threshold, corresponding to relatively high radiation dose, as may be seen in Table 3-1. Note that Table 3-1 refers to absorbed doses and not to dose equivalent, showing the lack of information required to convert such acute-effect data from low-LET x-rays and gamma rays to charged par- ticles. The NCRP has attempted to deal with this problem in NCRP Report No. 132 (NCRP, 2000) by prescribing a tabulated set of LET-dependent “RBE” factors, to be multiplied by absorbed dose in acute irradiation, leading to a “Gy-equivalent” quantity to be used for limiting acute radiation risk. These radiation levels are likely to be exceeded only during activities involving minimal shielding, as would be expected during an extravehicular activity or in a rover. The dose rates likely to lead to acute effects are expected only for SPEs. The cumulative effect of exposure to penetrating, hard-to-shield GCR is most likely to lead to cancer, CNS, and degenerative risks. Acute radiation effects and cancer risks are expected to be the most important consequences of SPEs. TABLE 3-1 Thresholds for Radiation Effects Effect Absorbed Dose (cGy) Blood-count changes 20-50 Vomiting or nausea 100 Death Minimal care 320-360 Medical treatment 480-540 Autologous bone marrow transplant 1,100 Permanent sterility Males 350 Females 250 Cataracts 200-500 SOURCE: NCRP, 1989, Table 5-1.

OCR for page 49
58 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION 5. Immune system responses: Exposure to the space enironment is known to alter cellular-mediated immune system functions such as cytokine production and lymphocyte proliferation and trafficking. While such responses during spaceflight are generally thought to be due to microgravity effects, the complex processes involved, including any direct contribution by radiation, are not yet fully understood. The potential interactions of the altered immune system with radiation responses have also not been thoroughly studied. For instance, decreased immune surveil- lance may allow more efficient radiation-induced carcinogenesis. Such potential interactions between radiation and other aspects of the space environment require further research, and are likely to become increasingly important to astronaut health as mission durations increase. In addition, hematopoietic stem cells are among the most radiosensitive cells in the body, and acute radiation effects include a marked depression of lymphocytes at doses near 100 cSv within the first 6 hours following radia- tion exposure. Effects of radiation exposure at much lower doses have been observed, suggesting a possibility for radiogenic immune dysfunction in astronauts during the time frame of their spaceflight. This could hamper astronaut function and lead to poor performance during missions. Further studies are needed to elucidate the nature and mechanisms of immune responses following exposure to high-LET radiation. Finding 3-1. Uncertainty in radiation biology. Lack of knowledge about the biological effects of and responses to space radiation is the single most important factor limiting the prediction of radiation risk associated with human space exploration. Finding 3-2. Funding cuts to radiation biology research. NASA’s space radiation biology research has been compromised by the recent cuts in funding, particularly in research addressing noncancer effects. EFFECTS ON MATERIALS AND DEVICES Space is a harsh environment for hardware as well as for humans. The radiation hazard for humans is the focus of this report. Accordingly, only a brief review of radiation effects on materials and devices is offered here, for the sake of completeness. In general, NASA has broad experience in coping with space radiation effects on materials and devices, and the committee expects that that expertise will be brought to bear on the design and operation of Exploration vehicles. Radiation effects limit the lifetime of spacecraft, limit the regions of space in which spacecraft can operate, and increase the risk of spacecraft failure. At this time there are four general classes of radiation-induced mechanisms that affect space systems: (1) permanent degradation, (2) transient damage, (3) single-event effects (SEEs), and (4) spacecraft charging, including both surface charging and internal (or so-called deep dielectric) charging. Permanent degradation of materials and microelectronics results from prolonged exposure to particle radiation. This degradation can result from ionization and displacement of atoms in a crystalline lattice, often referred to as a displacement kerma or as a non-ionizing energy loss (NIEL) (Leroy and Rancoita, 2007). Ionizing dose affects a wide class of materials and devices, whereas NIEL is typically only relevant in minority carrier semiconductors, such as bipolar technology devices, and in optical materials, such as detectors, solar cells, and optocouplers where changes in carrier lifetime affect device performance. For solar panels on interplanetary spacecraft—or on the Moon—the cumulative effect of displacement damage during large SPEs can be the limiting factor in their lifetime.1 Total ionizing dose is the dominant concern for majority carrier devices such as complementary metal oxide semiconductor technology devices and is often the limiting degradation mechanism for spacecraft electronic performance (van Lint et al., 1980; Messenger and Ash, 1986). At very high doses, greater than 100 kGy, a range of other non-electronic materials can begin to show serious degradation such as a color change, polymer cross-linking, cracking, material flaking, and embrittlement. Teflon is an example of a material that is very sensitive to radiation damage and can change its mechanical and electrical insulating properties when irradiated to high levels. In the very large SPE of July 14, 2000, the current output from the solar panels on the SOHO spacecraft suffered a decrease in current 1 output that was the equivalent of a year’s normal exposure (Brekke et al., 2005).

OCR for page 49
59 RADIATION EFFECTS Transient damage refers to the radiation damage component that is only present during the irradiation or that anneals out during normal environmental conditions. Transient effects include dose-rate effects, for example, photocurrents, as well as the annealing of transistor gain due to displacement damage effects. Dose-rate effects track the incident radiation rate, typically a gamma background level, and refer to a continuum effect rather than to ionization associated with a single heavy particle. Displacement damage can inject a severe degradation in the gain of a semiconductor that anneals out in the period as early as 1 microsecond and extending out to several months after the irradiation with, approximately, an exponential time constant (Messenger and Ash, 1986). SEEs refer to a class of effects in which the damage results from a single ionizing particle traversing a microelectronic device rather than the accumulated impact of a large number of particles. SEEs are disruptions of normal circuit response that occur as a result of electron hole pairs being generated along the path of the incident ionizing particle (Messenger and Ash, 1997). Both the total collected charge and the rate of charge col- lection can be important in triggering the disruptive effect. The particle source can be “primary” from the GCR or SPE particle populations or a secondary spallation product, generated from the nuclear interaction between a primary particle and nuclei of microelectronics materials. Even thermal neutrons can affect unhardened parts that include boron2 as a doping material in the semiconductor or in the glass passivation layer. SEEs range from the relatively benign bit flips (single-event upsets) that can be circumvented by engineering, to temporarily disruptive latchups (where the power must be cycled to reset the circuit), to catastrophic burnouts (where the induced parasitic current flows lead to permanent and irreversible circuit damage). A particular concern in the space-microelectronics community is emergent technologies with smaller feature sizes, for which single protons and neutrons can deposit enough charge directly to cause SEEs in radiation-hardened semiconductors. With older technologies, SEEs were caused by either heavy ions or by heavy spallation fragments. 3 This increased vulnerability of the emerging technologies violates some of the assumptions that underlie commonly-used methodologies for estimating SEE rates in space. Spacecraft surface charging and deep dielectric charging result from exposure to low-energy electrons (typically from a few electron volts to a few kiloelectronvolts) in plasmas and high-energy electrons (typically >100 keV), respectively. In hot plasmas, spacecraft surfaces cannot maintain a current balance, and large potentials can build up causing a discharge. Higher-energy electrons can penetrate spacecraft structures and bury themselves in dielectric materials until the dielectric breakdown level is reached and a discharge is released into circuits. Discharges result in background interference on instruments and detectors, biasing of instrument readings, physi- cal damage to materials, upsets, physical damage to electronics, increased current collection, the re-attraction of contaminants, and ion sputtering that leads to an acceleration of the erosion of materials. Deep-dielectric charging requires very large electron fluences, on the order of 1010-1011/cm2, collected over the timescale set by the dielectric leakage rate. Such electron intensities are not found in interplanetary space, and deep-dielectric effects have gener- ally been a concern for satellites exposed to transient increases in radiation belt electrons while operating in low Earth and geostationary orbits (Barth, 2002). As already mentioned, NASA has broad experience in dealing with space radiation effects on materials and devices. However, there is one area of potential concern where NASA has little experience: radiation effects on materials near ground nuclear power stations or near nuclear space propulsion systems. The radiation degradation considerations for materials near a reactor core go beyond the normal concerns over the degradation of electronic equipment and can involve changes in the mechanical and thermal properties of the material. The nuclear industry has experience in this area that can be applied to the NASA designs. A combination of distance and shielding are generally applied to lower the radiation levels at locations of sensitive equipment. Careful material selection and material testing are used for materials near the reactor. The reactor itself is generally designed with intrinsic feedback mechanisms, for example, thermal expansion, that ensure safe operation under all reasonable scenarios. The reactor control systems generally have some electronic components that are located several meters from the core itself in order to use distance and shielding to reduce the radiation environment. The material degradation of The relevant reaction is 10B(n,α)7Li. 2 The larger feature size and critical charge required for an upset made them immune to the charge delivered by the interaction of a single 3 proton or neutron.

OCR for page 49
60 MANAGING SPACE RADIATION RISK IN THE NEW ERA OF SPACE EXPLORATION the reactor materials themselves balances the radiation damage against the high-temperature material properties and corrosion and oxidation considerations for materials in the working environment. The ground station power conversion system, such as a Brayton engine, may be located close to the reactor and must take radiation degrada- tion mechanisms into accountfor example, degradation of seal materials and material embrittlement. Shielding of the reactor neutron/gamma radiation often uses high-atomic-number materials to attenuate the gammas and a combination of material rich in hydrogen together with boron and lithium-containing materials to attenuate the neutron flux. The hydrogen in the materials thermalizes the neutrons and the 10B or 6Li absorbs the neutron. An important consideration in the shield design is the generation of neutron-induced secondary gammas within the shield material itself. For thick shields, the neutron-induced secondary gammas are typically the most important radiation consideration on items inside of the shield. EFFECTS ON MISSION It is important that NASA carefully consider the potential for mission failure, in addition to health impacts, while establishing a radiation shielding strategy. While there is a very small possibility that an acute dose could render an astronaut unable to perform his or her duties, missions may also fail because they are restricted by flight rules. Radiation limits are chosen to reduce the long-term health effects on astronauts, effects that may not manifest until years after the mission has been completed. The principle of As Low As Reasonably Achievable (ALARA) is implemented through flight rules to ensure that astronauts do not approach these limits and to reduce the risk of exposing the astronauts to acute radiation effects. Among these flight rules will be a requirement to return to the most heavily shielded shelter if the radiation flux exceeds a predetermined dose-rate threshold. Because SPEs are unpredictable, the threshold will necessarily be conservative and well below a life-threatening dose rate. While the peak flux of severe SPEs may be limited to a few hours, the flux above an action level may persist for several days. The initial missions to the Moon will last only 1 or 2 weeks, and the ability to extend the mission to account for time lost to an SPE will be limited by consumables and orbital mechanics. Therefore, it is possible that the astronauts will be unable to accomplish prime mission objectives because they were not permitted to leave the outpost or, depending on the event timing, even land on the lunar surface in the first place. There are several courses of action available to NASA at this early stage in operational planning to reduce the risk of mission failure due to SPEs that in due course stay below a fluence that approaches radiation limits. First and foremost, when creating the flight rules, NASA could consider the probability that an event will occur that exceeds a certain threshold flux level, in addition to a threshold fluence level. The radiation shielding strategy could be made robust against long-duration, low-flux events. The simplest solution may be to include additional consumables and operational flexibility to extend the mission stay. This has the dual benefit of supporting extended stay for any number of additional contingencies that may delay a return to Earth (more time needed to complete objectives, unanticipated discoveries, broken or faulty equipment, and so on). However, NASA could also consider providing additional shielding for the astronauts on surface excursions and limiting the time exposed on the surface in a space suit. Emergency procedures that can be executed away from the outpost for limited periods should be explored and evaluated against returning to base. These could include provisions for prepositioned shelters or the ability to construct adequate shelter from available resources. Improved forecasting or nowcasting may also have a substantial value. If forecasters are able to reliably ensure all-clear periods of up to 8 hours, or if the projected fluence of ongoing events can be reliably forecast for 8 or more hours, then the dose-rate action level could be adjusted. A cost-benefit analysis or identification of cost-benefit metrics would help to quantitatively measure the value of enhancement to SPE or all-clear forecasts. REFERENCES Barth, J.M. 2002. Natural Radiation Environment Definition for Electronic System Design. Available at http://radhome.gsfc. nasa.gov/radhome/papers/GOMAC02_Barth.pdf. Brekke, P., M. Chaloupy, B. Fleck, S.V. Haugan, T. van Overbeek, and H. Schweitzer. 2005. Space weather effects on SOHO and its space weather warning capabilities. Pp. 109-122 in Effects of Space Weather on Technology Infrastructure. NATO Science Series Volume 176. Springer, The Netherlands.

OCR for page 49
61 RADIATION EFFECTS Cucinotta, F.A., W. Schimmerling, J.W. Wilson, L.E., Peterson, G. Badhwar, P. Saganti, and J. Dicello. 2001. Space radiation cancer risks and uncertainties for Mars missions. Radiation Research 156:(156):682-688. Cucinotta, F. A., M-H.Y. Kim, and L. Ren. 2005. Managing Lunar and Mars Mission Radiation Risks. NASA/TP-2005-213164. Available at http://marsjournal.org/contents/2006/0004/files/Cucinotta2005.pdf. Curtis, S.B., J.E. Nealy, and J.W. Wilson. 1995. Risk cross sections and their application to risk estimation in the galactic cosmic-ray environment. Radiation Research 141:57-65. Fajardo, L.P, M. Berthrong, and R.E. Anderson. 2001. Radiation Pathology. Oxford University Press, Oxford, U.K. Leroy, C., and P. Rancoita. 2007. Particle interactions and displacement damage in silicon devices operated in radiation envi- ronments. Reports on Progress in Physics 70:493-625. Messenger, G.A., and M.S. Ash. 1986. The Effect of Radiation on Electronic Systems. Van Nostrand Reinhold, New York. Messenger, G.A., and M.S. Ash. 1997. Single Eent Phenomena. International Thomson Publishing, New York. NASA (National Aeronautics and Space Administration). 1998. Strategic Program Plan for Space Radiation Health Research. NASA, Washington, D.C. Available at http://spaceresearch.nasa.gov/common/docs/1998_radiation_strat_plan.pdf. NCRP (National Council on Radiation Protection). 1989. Guidance on Radiation Receied in Space Actiities. NCRP Report No. 98. Bethesda, Md. NCRP. 1997. Uncertainties in Fatal Cancer Risk Estimates Used in Radiation Protection. NCRP Report No. 126. Bethesda, Md. NCRP. 2000. Radiation Protection Guidance for Activities in Low-Earth Orbit. NCRP Report No. 132. Bethesda, Md. NRC (National Research Council). 1996. Radiation Hazards to Crews of Interplanetary Missions: Biological Issues and Research Strategies. National Academy Press, Washington, D.C. Rabin, B.M., B. Shukitt-Hale, J.A. Joseph, K.L. Carrihill-Knoll, A.N. Carey, and V. Cheng. 2007. Relative effectiveness of different particles and energies in disrupting behavioral performance. Radiation and Enironmental Biophysics 46(2):173-177. Schimmerling, W., E.L. Alpen, P. Powers-Risius, M. Wong, R.J. DeGuzman, and M. Rapkin. 1987. The relative biological effectiveness of 670 MeV/A neon as a function of depth in water for a tissue model. Radiation Research 112:436-448. Schimmerling, W., J. Miller, M. Wong, M. Rapkin, J. Howard, H.G. Spieler, and B.V. Jarret. 1989. The fragmentation of 670A MeV Neon-20 as a function of depth in water. I. Experiment. Radiation Research 120:36-71. Shavers, M.R., J. Miller, L.W. Townsend, J.W. Wilson, and W. Schimmerling. 1993. The fragmentation of 670A MeV Neon-20 as a function of depth in water. III. Analytic multigeneration transport theory. Radiation Research 136:1-14. van Lint, V.A.J., T.M. Flanagan, R.E. Leadon, J.A. Nabor, and V.C. Rogers. 1980. Mechanisms of Radiation Effects in Electronic Materials: Volume . Wiley-Interscience, New York. Yang, V.V., and E.J. Ainsworth. 1982. Late effects of heavy charged particles on the fine structure of the mouse coronary artery. Radiation Research 91(1):135-144.