632 RADIATION PROTECTION occurrence of specific traits in progeny of irradiated animals. In studying irradiated males, the experimenter can determine the genetic manifestations in the progeny corresponding to the stages of development of spermatogonia and spermatozoa in the parent. This can be accomplished by selecting suitable time intervals between irradiation and mating. Experimentally one measures visible traits in the offspring (such as coat color changes in the mouse or failure of pupal development in the fruit fly). These traits are then attributed to specific gene mutations in the parent germ cell. The effect is therefore considered to be directly proportional to the number of genetic changes induced in the parental germ cell. It is well demonstrated that the curve showing effect against dose in experimental animals is linear within the range of 37 r to 1,000 r total acute dose, and geneticists believe that there is no threshold for the genetic effect. The finding of a dose-rate dependence effect (chronic exposure is approximately one-fourth as effective in inducing mutations as is acute exposure) probably represents partial recovery at low dose-rates and does not conflict with the no threshold concept. 2.24 For somatic effects, unlike genetic mutation effects, there is no general agreement among scientists on the dose-effect relationships. It is known, for example, that the nature of the dose-response curve can be altered drastically by changes in the external environment of the organism. In addition, although radiation may be the initiating event, there may be other promoting factors operating before the manifestations are evident. Such factors mentioned in the literature include cocarcinogens: hormones, chemicals, and viruses. 2.25 Because of the complexities of animals and man, there may be many mechanisms by which radiation produces effects. One of the mechanisms may be the induction of a primary effect by radiation which, after a sequence of secondary events over a period of time, leads to a clinical manifestation such as neoplasia. In this hypothesis, the induction of the primary effect could be consistent with a linear no threshold concept of dose-effect relationship, yet the successive manifestations of the damage could be nonlinear and not consistent with a threshold concept. Therefore, in the case of neoplasia, the demonstration of linearity or nonlinearity for the gross effect does not predict the presence or absence of a threshold dose for the primary insult. 2.26 There are some somatic effects in animals which do not support a linear no threshold concept (e.g., acute mortality; splenic, thymic and testicular atrophy, incidence of lens opacity, duration of depression of mitotic activity, and incidence of heterologous tumor implants). However, the experiments demonstrating these effects were not performed primarily to examine threshold theory and were done at high dose ranges above 100 r. Considering the diversity of results in different species of animals, extrapolations to man for these effects at low doses should be made with caution. 2.27 In man, the chief evidence for a linear dose-effect relationship for somatic effects comes from some of the leukemia studies (see Table 2.2). Data are available for acute exposures above 50 rads in adults. Predictions of the incidence of leukemia in the general population per rad of exposure have been made by extrapolations from these data. Certain of these predictions have involved the assumption that the occurrence of radiation-induced leukemia per rad will remain constant for the life of the population, the assumption of no difference among effects of irradiation of various parts of the body and the assumption of a constant probability of occurrence of leukemia per rad of acute and chronic exposure. There is no direct evidence that justifies extrapolation from the condition of acute exposure to one of a low dose chronic external exposure, or to the radiation from internal emitters. 2.28 In summary, the evidence is insufficient to prove either the hypothesis of a damage threshold or the hypothesis of no threshold in man at low doses. Depending on the assumptions used, forceful arguments can be made either way. It is therefore prudent to adopt the working principle that radiation exposure be kept to the lowest practical amount. Genetic Effects 2.29 The following working assumptions have been derived from the evidence considered in this staff report: (1) radiation induced mutations, at any given dose rate, increase in direct RADIATION PROTECTION 633 linear proportion to the genetically significant dose; 2 (2) mutations, once completed, are irreparable; (3) almost all the observed effects of mutations are harmful; (4) radiation-induced mutations are, in general, similar to naturally occurring mutations; and, (5) there is no known threshold dose below which some effect may not occur. 2.30 The linearity is established in fruit flies down to 25 r and is confirmed in mouse spermatogonia down to 37 r, but there is no direct evidence for linearity below these doses. Although the studies in animals do not involve a period comparable to the 30-year period of chronic irradiation in humans, the hypothesis used in this staff report is that the mutations induced by small dose rates of radiation to human reproductive cells are cumulative over long periods of time. Under this assumption, irradiation of the whole population from any source is expected to have genetic consequences. 2.31 In addition to genetic effects in the progeny of an exposed individual, attention must be given to the total genetic effect on the population. Within the working assumptions above, the total genetic load is independent of the distribution of the exposure within the population. Therefore, when radiation protection standards are established for large numbers of exposed persons, limitations may be imposed by considerations of population genetics (the effects on population as a whole). 2.32 Major areas of uncertainty in genetic information for man, with regard to both population and individual genetics, are the estimations of: the spontaneous and induced mutation rates; the genetic load of mutations; the influence of man-made factors (mortality reduction brought about by health protection, for example) operative in natural selection; and the influence of synergism of gene interaction. 2,33 Formulation of radiation protection standards has been based in part on estimates of genetic hazards to man. These in turn have been based chiefly on data from mice and from acute rather than chronic irradiation. Results of recent experiments considered pertinent to the evaluation of genetic effects are: (1) The genetic effects under some radiation conditions may not be as great as those estimated from the mutation rates obtained with acute irradiation. It has been shown in mice that fewer specific locus mutations are produced in spermatogonia and oocytes by a low dose rate (chronic gamma radiation at 90 r per week) than by a high dose rate (acute irradiation at 90 r per minute) for the same total accumulated dose above 100 r. A similar effect has been reported for sex-linked lethal mutations in the oogonia of fruit flies. The number of mutations induced in spermatogonia by chronic irradiation is smaller (about one-fourth) than that induced by acute irradiation. (2) Studies being planned may define quantitatively the dose-effect relationship with fractionated, low doses delivered at high dose rates. These data may be of direct significance to medical practice using fluoroscopy and radiography. (3) Life shortening has been demonstrated in the offspring of male mice irradiated at high doses. (4) Radiation doses of 25 r appear to produce chromosomal breakage in human cells grown in tissue culture. Items (1) and (2) above indicate that in the preparation of radiation protection standards based on the genetic effects, consideration should be given to dose rate as well as total dose. *The genetically significant dose to the individual is considered to be the accumulated dose to the gonads weighted by a factor for the future number of children to be conceived by the irradiated individual. The genetically significant dose for the population is defined as the dose which, if received by every member of the population, would be expected to produce the same total genetic injury to the population as do the actual gonad doses received by the various individuals. 634 RADIATION PROTECTION Leukemia 2.34 Information useful for study of the risk of leukemia among exposed persons is based on experimental data on animals, some observations on humans, and the rise in crude leukemia mortality rates observed in many countries. There is more information available on the correlation between radiation exposure and leukemia incidence in man than there is for other radiation effects. 2.35 Most of the reported investigations indicate that the incidence of leukemia among irradiated persons increases with the exposure dose. A definitely increased incidence of leukemia occurs after one large whole body dose or a large accumulated dose. The available evidence applicable to the general population under the assumptions listed in paragraph 2.27 indicates a linear correlation of dose to incidence down to about 50 rads of whole body acute exposure. The specific findings in other studies vary with the type of exposure and are speculative at lower doses. There have been reports that, during prenatal life, fetal doses as low as 2-10 r may double the incidence of leukemia, although other studies have not confirmed this finding. Prenatal exposure may be quite different from exposure of adults and there is no evidence that these low dose levels may be effective later in life. There is also no satisfactory evidence that chronic lymphatic leukemia is produced by radiation although this is the form of leukemia primarily responsible for the rising crude leukemia rate in the general population. 2.36 Past studies of leukemia-radiation correlations in human populations have limitations imposed by retrospective epidemiological techniques as well as factors inherent in the nature of leukemia. Epidemiological techniques which are retrospective in type are limited by the: (1) difficulty of determination of the radiation dose; (2) absence of uniform radiation recording methods; (3) difficulty of associating medical and vital statistical records: i.e., such studies introduce biases inherent in the techniques of interview, questionnaire, or manual searching; (4) statistical selection of cases which may be weighted with those cases having a disease related in some way to leukemia; and small. (5) the fact that the numbers of persons in the population groups studied are usually 2.37 The following factors produce difficulties in the evaluation of the findings on possible radiation produced leukemia; (1) Although leukemia has the advantage of the use of simpler procedures for the diagnosis of the disease than are available for other neoplastic diseases, it has the disadvantage that the classification of various types of leukemia is subject to debate. It is thus difficult to compare statistics of different origins. (2) The hematological effects such as are seen in leukemia can also be ascribed to sources other than radiation, (3) Leukemia ascribed to radiation cannot be distinguished from leukemia due to other causes. (4) Leukemia is a rare disease in humans whose crude annual incidence in the population-at-large is about 3 per 100,000 persons. (5) The various forms of leukemia have different clinical courses and the relative incidence of cytologic types varies with age. Not all the various forms of leukemia can be placed in one category since it does not appear that the chronic lymphatic form may be induced by radiation. 2.38 Considerations of the above factors require that epidemiological studies include large samples of exposed subjects, provide mechanisms for follow-up over long periods of time, provide adequate control groups, and provide ascertainable exposure and outcome. ! RADIATION PROTECTION 635 2.39 Conclusions drawn from the studies listed in Table 2.2, indicate that: (1) Under certain conditions, there is a clear association between leukemia and prior radiation exposure. This association has been demonstrated only where the exposures are high. The effect may be discerned at doses of the order of several thousand r for prolonged intermittent exposure over many years in normal adults; or, doses of the order of 500 r for bone marrow exposure in adult males with pre-existing disease; or, doses of the order of 50-100 r for acute whole body exposure in a general population of all ages; or at acute dose possibly as low as 2-10 r to the fetus; radiation. (2) Long follow-up periods are required to assess cancer experience following ir (3) Little data exist on leukemia incidence among women exposed to therapeutic doses of radiation from radium or x-rays; (4) It is unlikely that retrospective studies will definitely solve the question of the shape of the dose-response curve at low levels of exposure or the existence of a threshold. Additional retrospective studies on population groups receiving high doses of radiation may provide refined quantitative knowledge. There are only a few prospective studies reported that can provide information on both the quantitative and qualitative effects of chronic low doses received over many years; (5) The risk of any one individual developing leukemia is small even with relatively large doses. However, when large populations are exposed, the absolute number of people affected may be considerable. 2.40 The leukemogenic effect of internally deposited isotopes requires special mention. Strontium: We have no documented evidence that bone depositions of strontium in humans have produced leukemia. Statements that radiostrontium is leukemogenic are based solely upon studies in mice. Since leukemia is a common disease spontaneously occurring in certain strains of mice, one cannot accept this observation as necessarily applicable to man. Thorium: Only a few cases of leukemia following thorium injections for medical diagnosis have been reported in the literature. The leukemias have occurred with latent periods up to 20 years. However, the dose calculations for irradiation of the bone are complicated by the presence of thorium daughters. Radium: No cases of leukemia have been reported in those persons who have had radium deposited in their bones, even though some persons developed bone cancers. This is not unexpected in view of the fact that radium deposited in bones results in a relatively small dose to the bone marrow. Iodine: Only a few cases of leukemia have been reported in patients receiving iodine-131 for the medical treatment of hyperthyroidism and cancer of the thyroid. It would seem that well planned large population studies on persons who received radioiodine medically would contribute to the knowledge of the leukemogenic and carcinogenic effect at the levels used. 2.41 The possibility of the detection of low doses of radiation by hematological techniques is deserving of high priority. The most sensitive indicator available at present may be the counting of binucleated lymphocytes, but the technique is not now practical for wide applications because of the need to examine large numbers of cells on hematology slides. The development of practical electronic devices to screen these cytologic blood specimens should be encouraged. The prognostic significance of the observations of morphological changes in the lymphocytes will be elucidated by long term follow-up of selected study and control groups. Other Neoplasms and Premalignant Changes 2.42 Clinical evidence indicates that irradiation in a sufficient amount to most parts of the body may produce cancer as a delayed effect although no inference of an incidence-dose relationship should be drawn. Some of the evidence in humans is based on: (1) skin cancers among radiologists in the early history of the use of x-ray; (2) thyroid cancers in children irradiated in the neck region; 636 RADIATION PROTECTION (3) Leukemia among children who were exposed in utero to x-ray for pelvimetry of the mother; (4) Bone sarcomas in radium dial painters and other persons exposed to radium-226; (5) Liver sarcomas in medical patients given thorotrast; and (6) Bronchogenic cancer in miners occupationally exposed to radon and its daughters. 2.43 The bulk of the evidence lies in the work done on animals with external whole and partial body doses, as well as with internally absorbed radionuclides. Both benign and malignant lesions have been produced, although the evidence is incomplete and there is no simple relationship between carcinogenesis and dose. Mice are more sensitive to all modalities of radiation exposure than man for the induction of skin and ovarian tumors and leukemia. TABLE 2.2 TYPES OF STUDIES THAT HAVE BEEN DONE IN HUMANS ON LEUKEMIA I. Occupational 1. Cases not reported in the literature. 2. Scattered reports in the literature. 3. Radiologists. 4. Uranium miners. II. Therapeutic and Diagnostic 1. Children receiving partial body exposure to x-rays. a. Infants treated for thymus gland enlargement. b. Infants similarly treated who had normal size thymus glands. c. Children treated for pertussis and lymphoid hyperplasia. d. Children treated for other benign conditions of many different types. e. Children treated for neuroblastoma. 2. Adults a. Patients with ankylosing spondylitis given x-ray treatment to the spine. b. Radiologists receiving partial body x-ray radiation over many years. c. Patients treated for hyperthyroidism with x-ray; and radioiodine. d. Patients treated for polycythemia with radiophosphorus. 3. Prenatal Maternal prenatal exposure to diagnostic doses of x-rays. III. General Population Japanese people who received whole body irradiation from A-bomb explosion. IV. Internal Emitters 1. Thorotrast 2. Radium 3. Iodine 4. Phosphorus |