Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
9 Alternative Approaches Data on the occurrence of adverse reproductive outcomes following expo- sure to ionizing radiation could be derived from a variety of cohorts, in addition to the atomic bomb survivors, such as the children of (1) people residing in areas where the background of naturally occurring radiation is substantially higher than usual, (2) individuals, other than Atomic Veterans, exposed to fallout from atmospheric weapons testing, (3) people living near nuclear installations, (4) individuals exposed occupationally, (5) patients undergoing medical diagnostic procedures, and (6) patients undergoing medical therapy for benign or malignant disease. Each of these cohorts has strengths and limitations. Usually, these are related to sample size, population composition, certainty of dose, presence of concurrent disease, and other confounding factors. The study of the atomic bomb survivors is the largest, longest, and most comprehensive epidemiologic study of radiation-induced carcinogenesis and mutagenesis that has been undertaken. Its strengths are that it includes a large population of all ages and both sexes who were not selected because of occupa- tion or disease. Other strengths are that it includes a wide range of doses, has included follow-up for more than 45 years, has comprehensive individual do- simetrv, and can use internal comparisons. Weaknesses include the following: although the clinical examinations of the children of the survivors began in the spring of 1948 and the surveillance of mortality among these children covers the time since May 1946, the cohort on which the studies of cancer among the survi 74
ALTERNA TI BE APPROA CHES 75 vors themselves rests was defined on the basis of the 1950 national census and thus does not include the years 1945 through 1949. Moreover, of importance in the present context is the fact that the exposure was at a high rather than a low dose rate, and the possible contribution of neutrons is somewhat uncertain. The fact that the population is Japanese raises some question about the transfer of risk factors derived from this population to other populations that may have dif- ferent baseline rates of health outcomes. There have been a number of studies of populations (other than Atomic Vet- erans) exposed to radioactive fallout from weapons testing, weapons use, and nuclear plant accidents. These studies may involve relatively few people (as in the Marshall Islands), but most involve thousands or hundreds of thousands of people. The advantages of such studies is that they involve populations of both sexes and all ages. In addition, they may yield information on the effects of chronic exposure. The difficulties in risk assessment arise from the fact that the doses are usually quite low and are rarely available on an individual, specific basis. Typically, dose estimates are derived from computer modeling of the source, meteorology, environmental pathways, and assumptions about ingestion patterns and amounts. Often, the dose estimates can be made only collectively. Although fallout patterns can be modeled by computer, experience from the Chernobyl nuclear power plant accident in 1986 has shown that individual doses may vary by a factor of 10 or more from the estimated average. Studies of fallout within the United States as a result of weapons testing at the Nevada Test Site have been performed as part of a significant scientific effort to reconstruct doses. The advantages of that study were comprehensive expo- ~sure evaluation and protracted exposures at a low rate. In spite of the dosimetry estimates, there still remains considerable uncertainty about individual doses. In addition, the estimated cumulative doses are much lower than those experienced from natural background radiation. Studies have been performed on localized fallout from the 1954 BRAVO weapons tests in the Marshall Islands. Fewer than 8,000 people on these islands were affected. The advantages of that study are that the population was unse- lected, there has been an attempt at determining individual dosimetry, and there has been long-term comprehensive medical follow-up. The small sample size remains a problem for conducting studies in this population, as does the uncer- tainty about the dose due to short-lived radioiodines. There have also been studies on the populations exposed to fallout as a re- sult of the Chernobyl accident in 1986. The advantages of this group are that it is large, the population is unselected, and dosimetry has been done for highly contaminated villages. The limitations are that the length of follow-up is limited and the iodine dosimetry remains somewhat uncertain. Data on the environmental contamination of the Techa River in the eastern part of Russia and Semipalatinsk, the weapons testing site of the former Soviet Union, have also recently become available. Difficulties in both instances in
76 ADVERSE REPRODUCTIVE OUTCOMES elude accurate estimation of the absorbed doses or inadequate ascertainment of exposed individuals. Strengths include a wide range of doses, long follow-up, an unselected population, and a large population. Studies of populations living around nuclear power plants have the advan- tage that they can be well defined assuming that there has been little population mobility. Unfortunately, in the United States mobility is common and so an ef- fort must be made to guarantee that the patients with disease, in fact, lived near the plant at the appropriate time before the latent period for cancer induction. The very low doses from emissions of most normally functioning nuclear power plants makes the required sample sizes for statistical significance almost impos- sible to achieve in circumstances where the exposure (dose) is less than that at Chernobyl. Occupational studies are a major source of epidemiologic information. They have the advantage that work records and times of employment are known. Knowledge of the dosimetry in these studies ranges from good for workers in nuclear facilities to very poor for groups such as the early uranium miners. An- other advantage is that there is usually a large number of people who can be studied. One problem with occupational studies is that the workforce is pre- dominantlv voting or middle-aaed healthy males, and the applicability of these risk factors to other populations requires some assumptions. The so-called healthy worker effect needs to be considered. This effect tends to give standard- ized mortality rates (SMR) or standardized incidence rates (SIR) that are less than unity. Confounding factors, such as exposures to chemicals and other sub- stances in the workplace, also need to be considered. Smoking is another com- mon confounding factor that needs to be considered. Moreover, in these studies, and indeed in all studies in which the dose is low, other sources of exposure to ionizing radiation, such as diagnostic or therapeutic irradiation, loom large as possible sources of confounding. Exposures to ionizing radiation as a result of medical diagnostic procedures are another potential source for information about the effects of ionizing radia- tion. The advantages of these studies are that the doses are reasonably well known, as is the field irradiated. The doses may not be as precise on a percent- age basis as those from radiation therapy since the doses are known to be low and the technical factors are not usually recorded. ~ ~ ~ ~ ~ A ~ ~ ~ ~ ~ J ~ ---I, ~ - ~ A ~ 1 1 The exposed populations consist of people of both sexes as well as most age ran Yes. The generally low ~1 , _t doses used In medical diagnostic procedures require extremely large sample sizes for statistical significance to be achieved. The purpose of the diagnostic study may also be a confounding factor, available for long-term follow-up. although generally the patients are Studies of patients who have received radiation therapy for benign diseases have the advantage of a known exposure field, a known type of radiation, and usually, good dosimetry. However, the dosimetry can be a problem in some cases in which children were irradiated and the organ of interest was near the
ALTERNA TINE APPROACHES 77 primary radiation field. In those cases, patient motion could cause substantial uncertainty in the actual doses to the organ of interest. The high doses of radio- therapy allow risk estimates to be derived with relatively small population groups, but the high doses also introduce the confounding factor of potential cell killing. The disadvantages typically are patient selection bias or confounding by disease, confounding by other therapies or economic status, and potential loss to follow-up if the disease did not require further treatment. Studies of the treatment of malignant diseases with ionizing radiation have the advantages of a well-defined dose and a known type of radiation in a local- ized field, confounders can be documented, and good follow-up is possible be- cause patients who receive these treatments are often followed closely for life. The disadvantages are the possibility of concurrent or other therapies that may confound the analysis, selection bias due to the disease, the possibility of cell killing rather than cancer or mutation induction, and a shortened lifespan for follow-up as a result of the malignant disease. A group including individuals who have been treated for malignant disease with ionizing radiation are survivors of childhood cancer. Current data do not indicate an increased risk for adverse reproductive and developmental effects in the offspring of male cancer survivors (Hawkins, 1991~. A study of more than 20,000 survivors and their offspring currently being conducted in the United States and Canada should provide im- portant new data on this particular population. Other exposed populations, however, have not been studied or have been studied inadequately. These include the children of (1) medical personnel who are occupationally exposed, (2) workers in nuclear facilities, and (3) members of The armed forces whose service functions involve exposure to ionizing radiation such as nuclear submariners and the crews of aircraft of the U.S. Strategic Air Command (SAC). Some of these groups offer opportunities pertinent to the concerns that have previously been discussed in this report. For example, studies of the reproduc- tive outcomes of SAC crews might be informative. These crewmen, officers, and enlisted personnel have the same age distributions as the Atomic Veterans at the time of their exposure to ionizing radiation, a similar dose distribution, and access to uniform health care. Also, hospital records are available for most, if not all, births to the wives of these SAC crews. Several tens of thousands of individuals are involved. l?resumably, much of the relevant information could be obtained by computer. Some of the potentially informative data are already available in a machine-retrievable format at the Armstrong Laboratory of Brooks Air Force Base in San Antonio, Texas. Moreover, a computer program, known as CARI, for estimating exposure to aircraft crews was developed at the request of the Federal Aviation Administration for commercial crews, but it should be applicable to Air Force crews as well. It requires information on flight pattern, duration, and altitude.
78 AD VERSE REPRODUCTI HE OUTCOMES Such a study would not be without its limitations. These include the possi- ble exposure to other potential mutagens such as chemicals, as well as heavy ion particles originating in space. Some of the information needed to estimate doses may be classified, but if the doses are computed by individuals in the Air Force with the appropriate security clearance, this might not be a major difficulty. It warrants noting, perhaps, that this same problem arose in the estimation of the doses received by the survivors of the atomic bombings of Hiroshima and Na- gasaki and was overcome by using suitably cleared technical personnel. Still other strategies, such as the applicability of recent developments in molecular and cellular biology, could be explored. These developments have given rise to a series of biological markers or biomarkers that can be used as estimators of exposure or dose, of biologic effects, and susceptibility. They in- clude such chromosomal techniques as fluorescent in situ hybridization (FISH), and biochemical measures of damage to the genetic material (DNA) at specific loci, for example, the X-linked hprt locus or the autosomal glycophorin A locus. At present, these techniques have limited applicability as measures of exposure or dose to doses of less than 0.10 Gy (10 red), and thus would seem to be of marginal use in the case of the Atomic Veterans, but this may change through the use of combined biomarker assays. The limitations of biomarkers as estimators of dose at doses of less than 0.10 Gy (10 red) reflect to a substantial degree the relatively large, normally occurring interindividual response to a given dose. However, even now these markers could be useful as measures of effect or sus- ceptibility. The strengths and current limitations of these methods have been addressed in recent workshops (NRC, 1995~. It must be noted, however, that these techniques are often time-consuming, expensive, and difficult to implement with large numbers of individuals and cannot be routinely applied to all people. For example, the glycophorin A assay can be used only on individuals who are of the MN blood type, and these individuals constitute only half of most popula- tions.