The chemical and biological effects of irradiation depend not only on the amount of energy absorbed in an irradiated medium but also on the distribution of the absorbed energy within the medium. For equal absorbed doses, various types of ionizing radiation often differ in the efficiency with which they elicit a particular chemical or biological response. The relative biological effectiveness (RBE) describes the effectiveness or efficiency with which a particular type of radiation evokes a certain chemical or biological effect. The RBE is computed by comparing the results obtained with the radiation in question with those obtained with a reference radiation (e.g. medium-energy X-rays or ⁶⁰Co radiation):

(4.17)

The LET is defined as the amount of energy deposited in a unit length along the distance traveled (see Chapter 1). Therefore, for ionizing radiation, LET is a measure of the density of ionizing events in a material. Ionizing radiation that deposits more energy over shorter distances have a higher probability of damaging DNA within the cell compared with radiations in which ionization is spread over larger distances. Therefore, damage is increased for a given radiation dose as the LET is increased. Particles with mass and charge have higher LET than photons. In addition, higher LET radiation produces relatively more damage by direct effects than by indirect effects. This is why the oxygen effect, basically a process that enhances indirect effects, is diminished for higher LET radiation.

Chemical agents may be used to modify the response of cells to ionizing radiation. These agents generally influence the indirect effects of radiation, i.e. the effects of free radicals. Because of the speed with which free radical damage is produced, the agents must be present during irradiation. It is more difficult to modify the direct effects of ionization; therefore, high-LET radiation effects are usually not modified significantly by the administration of chemical agents.

Agents that reduce the effects of radiation on cells are termed radioprotectors. Substantial interest in radioprotectors has been investigated because of their potential applications in military and space programs as well as for medical uses such as protection of normal tissues during radiation therapy. One group of researchers has synthesized over 2000 potentially radioprotective compounds in the past 40 years [13].

Mechanisms of action for radioprotectors include inactivation of free radicals by preferential binding or scavenging (e.g. sulfhydril components such as cysteine and cysteamine are good free-radical scavengers), induction of hypoxia to reduce the sensitizing presence of oxygen (e.g. epinephrine and carbon monoxide are often used to induce hypoxia), and occupation of sensitive sites in biomolecules by forming covalent bonds that are less easily disrupted by free radicals. Some protection against both direct and indirect effects has also been achieved by blocking cells in the G₁ phase of the cell cycle, thereby allowing more time for the action of repair enzymes before the S phase where DNA is replicated.

Radiosensitizers are agents that increase the effects of radiation on cells. Radiosensitizers are most applicable in radiation therapy where distinction must be made between “apparent” sensitizers and “true” sensitizers. Apparent radiosensitizers are toxic agents that are particularly effective in cases where radiation alone is less effective. True radiosensitizers are agents that promote both direct and indirect effects of radiation by inhibiting repair enzymes and weakening the sugar–phosphate bond of the DNA molecule. These include analogs of pyrimidines (5-bromodeoxyuridine [5-BUDR] and 5-iododeoxyuridine [5-IUDR]) and purines (6-mercaptopurine).

The direct effects of radiation may also be enhanced through the use of compounds containing boron to increase the effects of neutron radiation therapy [14]. Low-energy neutrons (0.025 eV, or “thermal neutrons”) have an unusually high probability of interaction with boron compared with elements that are normally present in the body. When a neutron interacts with ¹⁰B, the unstable nuclide ¹¹B is formed. The ¹¹B fissions producing α-particles that deliver a high absorbed dose in the immediate vicinity of the boron-containing compound causing increased damage.

Cosmic rays have always bombarded the earth. A typical US resident receives 0.34 mSv from cosmic rays. The earth’s atmosphere provides some shielding from cosmic rays. This shielding is reduced at higher elevations, and the cosmic ray dose is increased in these regions, e.g. inhabitants of Denver, Colorado, at an elevation of 1600 m (1 mile), receive 0.50 mSv/yr from cosmic rays while those in Leadville, Colorado, at an elevation of 3200 m (2 miles), receive 1.25 mSv. The average value of 0.34 mSv reflects the distribution of the US population living at different elevations.

Terrestrial background originates from radioisotopes that are found everywhere in our surroundings. All elements found in nature have radioactive isotopes, many of which are present in the areas we live. The exact composition of soil influences the local terrestrial background because the minerals present determine which elements are most abundant. Terrestrial background sources are categorized as “primordial” if their half-lives are the same order of magnitude as the presumed lifetime of the earth (i.e. 4.5 × 10⁹ years). That is, these sources were present when the earth was formed, and there is no way to replenish them in nature. Two isotopes of uranium, ²³⁸U and ²³⁵U, and one of thorium ²³²T, are primordial and give rise to three different decay series. In each of these series, the radioactive nuclides decay to stable isotopes of bismuth or lead. Seventeen other nuclides are primordial but are not part of a decay series. Of these nondecay series, radionuclides ⁴⁰K and ⁸⁷Rb make the greatest contribution to background dose.

Internal background is the dose imposed by the isotopes contained in our bodies. A small percentage of the potassium in the human body is ⁴⁰K. This radioactive nuclide emits both locally absorbed radiation and more penetrating gamma radiation. Similarly, ¹⁴C comprises a small percentage of the carbon atoms found in organic molecules throughout our bodies and contributes to a total dose of 0.28 mSv/yr to the average US resident from internal radiation.

As part of the ²³⁸U decay series, radon (specifically, ²²²Rn, referred to hereafter as simply radon) is a source of significant exposure because it is an alpha emitter that exists as an inert gas. When radon is generated by the decay of ²²⁶Ra at some depth in the soil, it does not bind chemically with other elements. Instead, it percolates up to the surface to escape into the atmosphere. Being heavier than most constituents of the atmosphere, it tends to remain at lower elevations. Although minable deposits of uranium ore are primarily associated with granite rock formations, uranium is found everywhere in the earth’s crust. Ninety-nine percent of the uranium found in nature is ²³⁸U. Thus, the air we breathe anywhere on earth contains some amount of radon.

Radon itself is not particularly hazardous when inhaled because it does not react and in most cases is simply exhaled. A more significant concern is that two of the decay products of radon (²¹⁸Po and ²¹⁴Po) are not inert. These products adhere to dust particles in the air, which may then be inhaled into the lung. The two alpha emitting isotopes ²¹⁸Po and ²¹⁴Po account for 85% of the dose to the lung from radon, two other beta and gamma emitting isotopes in the series, ²¹⁴Pb and ²¹⁴Bi, account for most of the remaining 15% of the dose. A typical US resident receives an annual dose equivalent of 2.3 mSv from radon and its decay products [27].

The exact value of radon dose equivalent received by an individual is most heavily influenced by the location where the person spends his or her time, i.e. inside or outside of buildings. With a combination of an unusually strong source (e.g. an unfinished crawl space below a house in a region with a high concentration of uranium in the soil) and air flow (e.g. radon buildup because of the lack of air exchange in an energy efficient house, or radon buildup because of a “chimney effect” in a drafty house), some homes can have radon levels that are significantly elevated. Although the risk to any one individual is very small, many people (i.e. the entire population) are exposed to some level of radon. By applying the linear, no-threshold model of radiation-induced cancer (see Section 4.4.5.3), one can estimate that a significant number of lung cancers may be caused each year by long-term exposure to high radon levels. Therefore, the US Environmental Protection Agency (EPA) recommends that radon concentrations be measured in buildings and that remedial action be taken if the radon concentrations exceed a certain level (e.g. 4 pCi/l of air). This level of radon has been estimated to yield a lung dose in a typical adult of 0.05 mGy.

Various human activities add to the radiation background we receive. Foremost among these are diagnostic radiology and nuclear medicine [27]. Radiation received by patients in radiation treatment is not counted in man-made background because this is a therapeutic use. This omission has little impact on the average level of man-made background because the number of persons that receive radiation treatment is a very small fraction of the US population.

Various consumer products may emit small amounts of radiation [28]. Some examples include smoke detectors that contain ²⁴¹Am, an alpha emitter. Luminous markings on dials, clocks, and instruments contained ²²⁶Ra at one time but since 1968 have been completely replaced by ³H and ¹⁴⁷Pm, both of which are low-energy beta emitters. The low-energy beta particles emitted by these substances are absorbed in the instrument components and provide negligible amounts of radiation dose to their owners. Collectively, consumer products are estimated to contribute approximately 0.1 Sv to the yearly dose from man-made background radiation.

Other contributions to man-made radiation include the aspects of the nuclear fuel cycle, ranging from mining to transportation and waste storage. These components are thought to account for approximately 0.02 mSv/yr.

The United States and other countries detonated over 500 nuclear weapons above ground during the time period from 1945 to 1963. This activity had a measurable effect on natural background radiation levels. In a nuclear explosion, new elements are created through the processes of nuclear fission and fusion. These elements are injected into the upper atmosphere by the force of the blast, where they drift over large distances before eventually returning to earth. The United States and the former Soviet Union ceased above-ground testing in 1963. A few countries conducted above-ground tests after that time, but by 1980, all above-ground testing has stopped. Individuals who were alive during the time period 1945–1980 received radiation doses because of inhalation and ingestion of some atoms of radioactive materials as a result of these tests. Calculation of this dose involves complicated functions of rate of uptake, rate of decay of the nuclides in individuals, rate of insertion of new isotopes into the environment, and other factors. The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) has estimated that an individual who was born before 1945 has received a total of 4.5 mSv because of nuclear testing. The current dose rate is estimated to be 0.01 mSv/yr [3].

From basic experiments involving cell cultures and laboratory animals, it is known that the dose rate plays a role in determining biological effects. In the range of total doses over which an effect is observed, the effect often decreases if the dose rate is lowered. Data in human populations support this concept [4,31]. Indeed, dose rate and dose fractionation are two commonly used principles of radiation therapy. Tissue repair mechanisms are not as robust in tumors as in normal tissue; therefore, fractionated radiation therapy treatments are designed to provide normal tissue a chance for repair while maintaining damage to tumor tissue.

If the radiation dose to any tissue is large enough, tissue damage will occur. In addition, in most tissues, the damage is greater, and the morbidity is greater if the amount of tissue irradiated is larger, e.g. whole organ irradiation produces a greater morbidity. This is due to the structure of tissues and organs into functional subunits (FSUs). A FSU is a set of tissues or organs whose ultimate function is dependent on the overall workings of each subunit, e.g. proper digestion of food requires the entire digestive tract to function properly. When the tissue or organ has parallel functional structure, rather than in series, damage may be compensated for more easily and some function can be maintained. Therefore, whole organ irradiation or irradiation of an entire FSU reduces or eliminates the ability of a tissue or organ to be repaired. Examples of whole organ irradiation effects can be seen following radiation therapy with the production of pericarditis and pneumonitis. In both of these processes, the dose threshold to the thorax to produce these effects is very large, e.g. 6–20 Gy.

The cells associated with the GI tract, particularly epithelial cells that line the intestinal surfaces, are particularly susceptible to radiation damage because they have a rapid reproduction rate. Following an exposure of 6 Gy or more diarrhea, hemorrhage, electrolyte imbalance, dehydration, etc., occur as a consequence of cell damage.

At doses above 50 Gy, damage to the relatively radiation-insensitive neurologic and cardiovascular tissue is severe enough to produce an almost immediate life-threatening syndrome. Death in this syndrome is brought about by destruction of blood vessels in the brain, fluid buildup, and neuronal damage. Death may occur in a few days or, at higher doses (greater than 100 Gy), within hours [32]. Because of the involvement of brain and nervous system tissues, the cerebrovascular syndrome is sometimes referred to as central nervous system syndrome. The hematopoietic and GI systems are also severely damaged at this dose level but would not contribute to the cause of death in such a short time.

After whole body exposure to a high dose of radiation, the body will experience a combination of all three syndromes. The International Atomic Energy Association (IAEA) has developed tables of symptoms that can be used to help determine how critical the exposure is and potential medical responses [32]. These tables divide the symptoms seen into early, i.e. prodromal, and critical or late phases. The prodromal phase is defined as “soon after” exposure. The effects that occur are classified from mild to severe to lethal, based on the total body dose. Whole body doses below a few gray are considered mild to moderate, and survival can be considered good; above 8 Gy, the dose is assumed to be lethal. These analyses should be used as guides only because accurate dosimetry and distribution of radiation dose to the whole body is highly variable. In addition, individual response to radiation is also highly variable.

The exposure of a population to radiation can cause damage in germ cells that ultimately result in mutations in the population’s decedents. However, these mutations are characteristic to the cell line exposed so that the radiation damage only produces DNA sequencing errors that might have occurred naturally. Therefore, instead of producing unique mutations, damage from radiation exposure only results in a higher frequency of spontaneous mutations. Consequently, radiation does not cause the production of monsters as seen in the movies. In addition, a large number of animal studies and the Life Span Study in humans have shown that the adverse effects from radiation exposure in subsequent generations are negligible, i.e. there is no direct evidence at any dose level that exposure of parents to radiation leads to excess genetic disease in offspring.

Cancer induction is arguably the most important and the most feared radiation effect. From the discovery of ionizing radiation, there has been documented evidence of radiation-induced cancer in both animal and human studies. The early human experiences were all at high absorbed dose levels from people working in radiation areas or using radiation without the knowledge of its potential harm. More recent analysis has shown increased cancer incidence from the Life Span Study and the early medical use of radiation in treatment and diagnostic studies at high exposure rates.

In this model, it is assumed that the incidence of induced cancer at low dose levels increases linearly with radiation dose up to a certain dose level. Then, at high dose levels, the incidence increases much more rapidly, i.e. changes from a linear relationship at low dose levels to a quadratic relationship at high dose levels. The linear-quadratic model is derived from a large number of animal studies that demonstrate this fit in specific cancers, e.g. leukemia and nonmelanoma skin cancer. However, the shape of the linear-quadratic fit has large variation in the curve fitting parameters for different cancers/cell types, making the extrapolation to low dose levels variable. Therefore, utilizing these fits to predict cancer requires different models for each specific cancer type.

In these models, the incidence of induced cancer is linearly proportional to the dose at all dose levels. The difference between the two linear models is whether there is a threshold below which too few cells are affected or cellular repair mechanisms compensate for any damage that occurs, i.e. is there a threshold dose for stochastic effects similar to deterministic effects. If a threshold level for stochastic effects exists, it has been proposed that this would be near 0.01–0.5 Sv. This value is based on studies that suggest stochastic effects have not been demonstrated at or below this dose range [33–36]. There are also studies that have investigated the application of a linear-quadratic “threshold” model to low-dose situations. These studies suggest that if there is a threshold, it is also at a very low value, e.g. 0.07–0.11 Sv. It needs to be pointed out here that these “very low dose” thresholds are at the level where exposures from diagnostic radiology studies occur and at doses that are encountered by workers in radiation occupations.

The issues with low-dose extrapolation models are the main reason the DDREF concept was developed. Low-LET radiation at low doses and low dose rate studies in animals have shown a lesser carcinogenic effect than the same doses delivered acutely [5]. In addition, the analyses of the dose response for leukemia in the Life Span Study, which could be represented by a linear-quadratic model, support the use of DDREF at low dose rates. Therefore, when discussing cancer risks from exposure to low-LET radiation and at low doses/low dose rates, DDREF is often used.

Correct modeling of radiation effects at low doses is both complex and a controversial topic. The decisions on which model to use, if DDREF are used, as well as many other factors need to be considered, as well as the known biases and uncertainties in the data (see Section 4.4.8). In many situations, the linear no-threshold model is the most conservative model, and therefore, this model is used by all advisory and regulatory agencies when setting limits for radiation exposure to the public and radiation workers (see Section 4.6.3). However, in epidemiology studies when comparing risk to low-dose exposures to individuals, the use of the linear no-threshold model may not be appropriate.

Hormesis is the concept that suggests that low levels of a potentially damaging substance, e.g. ionizing radiation, can stimulate an organism to produce protective biological processes in its cells, molecules, or organ systems [37–43]. It is known that the body will initiate repair mechanisms whenever damage occurs. However, it is unknown if at low radiation exposure, there is sufficient damage to stimulate this repair process. Hormesis proposes that if an organism is in an environment where it is continuously exposed to low levels of radiation, it will develop an inherent repair process such that the organism can respond to minor insults of low levels of radiation to reduce any damage.

There are a few studies that suggest a reduction in the effect of radiation or repair of DNA in cellular cultures and animal models [44–48]. Yet, similar to verifying a link between a substance and the induction of cancer, the ability to measure hormesis is extremely difficult, and the uncertainty in these studies is very large. However, if hormesis does occur, it would not eliminate the potential of cancer induction at low dose levels, whereas it may modify which low dose extrapolation model is appropriate and the parameters used in the extrapolation.

Radiation-induced cancer is assumed to be primarily by the radiation-ionizing molecules that alter their structure (e.g. creation of free radicals), and it is these “altered” molecules that cause DNA damage (see Section 4.3). This damage is assumed to be occurring over small distances, i.e. within cells, because of the highly reactive nature of free radicals. Because of this, it is assumed only cells that have been irradiated would receive damage. Based on this assumption, there should be no effect seen in cells that have not been irradiated. Yet, there have been several studies that have shown that radiation effects can be seen in cells that have not been irradiated [13,49]. This damage to nonirradiated cells is termed the “bystander effect.” The bystander effect has been shown in single-particle microbeam experiments with both high- and low-LET radiations, but it is more apparent with high LET participles, e.g. alpha particles. If the bystander effect is validated, it would imply that radiation-induced changes will occur beyond the irradiated cells, and again, it would change which low-dose extrapolation model and the parameters within the model that may be appropriate to use.

Absolute risk is defined as the probability that a person who is disease free at a specific age will develop the disease at a later time following exposure to a risk factor, e.g. the probability of cancer induction 10 years following exposure to radiation. The absolute risk model assumes that the risk factor produces the disease in addition to the natural incidence of the disease. In this model, the absolute risk for radiation-induced carcinogenesis is defined as the number of cancers in a population exposed to radiation, divided by the number of people exposed. This is defined for a specific time frame of interest and the average dose the population received. In an equation form, this would be

(4.21)