1. home >
  2. Library (Papers/Data) >
  3. Scientific Papers >
  4. Archived Scientific Papers (until 2019) >
  5. Commentary and Review Series (1989-2019) >
  6. Commentary and Review Series, 1996 >
  7. Commentary and Review Series 2-96

Commentary and Review Series 2-96

Radiation Effects Research Foundation research program: Historical perspectives, 1946-1995

Abrahamson S
Selected by RERF’s funding agencies, a 9-member ad hoc international review committee, referred to as the Blue Ribbon Panel, convened at the Hiroshima Laboratory, 5-7 February 1996, and at the Nagasaki Laboratory on 8 February. Prepared for the panel, this document outlines the evolution of the research program at ABCC-RERF.
Contents
Summary

The investigations of the Atomic Bomb Casualty Commission-Radiation Effects Research Foundation (RERF) to ascertain the effects produced by radiation in the survivors of the atomic bombings of Hiroshima and Nagasaki have been in progress for 50 years. This multi-faceted research effort has afforded the only opportunity to determine the late biological effects from exposure of a large population to a broad range of single, acute doses of ionizing radiation. The results obtained from these studies provide reliable information on health risks for the survivors and are used as a basis in setting health and medical services extended to the survivors. Fundamental to our understanding of the effects of ionizing radiation on humans, the ongoing RERF studies are essential for estimating radiation risk and setting safe standards for occupational, medical, and general population exposures. It is appropriate to ask at this juncture what further RERF research can contribute and what changes, if any, might be warranted in the future for RERF.

It is important that RERF investigations of cancer be continued to (1) strengthen the evidence for the shape of the dose-response curve for cancer, particularly at low doses; (2) obtain information on the impact of exposure to both the carcinogenic and noncarcinogenic effects of radiation among the youngest, and possibly most susceptible, members of the Life Span Study and Adult Health Study cohorts; (3) derive information on the biological (including genetic) and environmental factors that can modify radiation-induced effects; and (4) clarify the nature of excess noncancer mortality.

Exciting developments in molecular biology have led to a number of new approaches that, with further development, will permit detection of mutations in DNA from human populations at the level of nucleotide bases. These techniques are now becoming available for use in monitoring human mutation rates, and effort are now in progress to collect cells from exposed and nonexposed parents and their children. These cells are being immortalized using the Epstein-Barr virus, cultured, and stored for use whenever the new molecular approaches to mutation detection can be used cost effectively.

This molecular technology is simultaneously being applied to the somatic cells of the survivors to establish mutation rates, the nature of genetic changes leading to cancer, and changes in immune response capacity.

Biodosimetry based on chromosome changes, mutational endpoints, and electron spin resonance measurements of teeth are providing important insights into “true” dose estimates of the survivors when compared to assigned Dosimety Sysytem 1986 (DS86) doses, suggesting that revision of some aspects of DS86 may be required.

Given the current funding levels, priorities must be established. To conduct the research deemed important, certain research activities must be redirected, and recruitment of additional scientific and technical expertise will be necessary.

(return to contents)
Background

In all probability, in 1945, the populations of Hiroshima and Nagasaki were representative of a typical, heterogeneous, wartime population in Japan. In August of that year, atomic bombs were detonated over Hiroshima and Nagasaki. Damage in the two cities was caused by a combination of heat, blast, and fire. The energy of the Nagasaki bomb exceeded that of the Hiroshima bomb, but the burned-out areas of Hiroshima were greater because of differences in topography and in the distribution of buildings.

Because of the chaotic conditions after the bombings, the precise number of casualties will never be known. Estimates of death range from 90,000 to 140,000 persons in Hiroshima and from 60,000 to 80,000 persons in Nagasaki. Because in wartime, people moved in and out of each city, and in both, especially Hiroshima, there were relatively large numbers of military personnel–the exact population of the two cities at the time of the bombings is uncertain. Various estimates of the number of casualties have been made; however, the errors associated with the estimates may be quite large.

In November 1946, US President Harry Truman approved a directive to the National Academy of Sciences-National Research Council (NAS-NRC) to initiate a long-term investigation of the health effects associated with exposure to radiation from the atomic bombs. With funding provided by the Atomic Energy Commission (AEC), now the Department of Energy (DOE), the NAS-NRC established the Atomic Bomb Casualty Commission (ABCC) in March 1947, and investigations began in 1948. The Government of Japan, through the Japanese National Institute of Health, became a partner in this endeavor.

In 1975, the Radiation Effects Research Foundation (RERF) was established and assumed the responsibilities of ABCC. RERF’s official charter is:
To conduct research and studies–for peaceful purposes–on the medical effects of radiation on humans with a view toward contributing to the maintenance of the health and welfare of atomic-bomb survivors and to the enhancement of the health of all mankind.
This private foundation is equally funded by the governments of Japan, through the Ministry of Health and Welfare (MHW), and the United States, through the NAS under contract with the DOE. Thus, long-term investigations of the human health effects among survivors continue.

(return to contents)

Management of RERF Activities


RERF is managed by a board of ten directors; five are Japanese citizens and five are United States citizens. The day-to-day operations of RERF are managed by four permanent members of the board of directors, two from Japan and two from the US; three reside in Hiroshima and one in Nagasaki, Japan. The permanent directors are a chairman, a vice chairman, a chief of research, and one other director, each of whom has responsibility for oversight of specific RERF activities.

Epidemiological, statistical, medical, and laboratory investigations are conducted at the Hiroshima and Nagasaki facilities. The management of both laboratories is the responsibility of the RERF Executive Committee, which comprises all permanent directors. The research activities conducted in each laboratory are coordinated by the chief of research. Review of the research program, recommendations on the relevance and scientific quality of ongoing investigations, and guidance on future research directions are provided annually to the board of directors by a scientific council that conducts a three-day review of research activities each year. The Scientific Council is composed of ten scientific experts in research areas relevant to RERF, five from Japan and five from the US. In addition, workshops are held occasionally in specific research areas to obtain guidance from outside experts on promising approaches for possible implementation in the research programs.

(return to contents)

Program Goals and Objectives


Long-term follow-up of this unique radiation-exposed population has provided important information on the early and late health effects of radiation exposure. To carry out its responsibilities in an orderly, scientifically sound manner and to honor its charter, the goals of RERF are:
  • To determine the somatic and genetic late health effects produced in humans resulting from exposure to ionizing radiation.
  • To obtain information on the temporal pattern of cancer expression and on the role of biological and environmental factors that might modify the carcinogenic effects resulting from exposure to ionizing radiation.
The strategy for addressing these goals is identified in the following research objectives:
  • To conduct long-term epidemiological studies of a fixed sample of exposed and nonexposed individuals to determine the frequency-dose relationships of morbidity and mortality resulting from radiation exposure and to obtain information on differential sensitivity of various tissues.
  • To conduct organ-specific and case-control studies on cancer induced in specific tissues or organs in order to determine the cell types affected and the effects of modifying factors.
  • To conduct research in molecular and cellular biology to detect somatic mutation, cytogenetic changes, cell transformation, changes in immunological competence, and other biological changes for use in estimating radiation risk and serving as biodosimetric tools.
  • To use all available sensitive and cost-effective approaches for measuring mutation in the children of exposed and nonexposed individuals.
(return to contents)

Research Program Strategy


A series of genetics studies of newborns was initiated formally in 1948 at ABCC. These studies included observations on birth defects and other physical conditions, sex, birth weight, viability at birth, and neonatal survival of all newborns, a subsample of whom were re-examined at nine months of age. The genetic studies eventually encompassed more than 80,000 births. These major clinical studies were terminated in 1954, after it was determined that no dramatic findings could be expected. As described later in this historical review, the early genetics cohort thus established was the source of the original epidemiologic mortality sample of first-generation (F1) children born to the atomic-bomb (A-bomb) survivors.

During the first decade of ABCC operation, studies on survivors and those exposed in utero were diverse and designed on an ad hoc basis with each investigator establishing his own study population.

To conduct a successful long-term epidemiologic investigation to address the goals of ABCC-RERF, data must be collected on a continuing, prospective basis using specified samples of exposed individuals and matched comparison groups of nonexposed persons. These data must also be collected in a systematic and epidemiologically sound manner to reduce the chance of bias or ambiguity affecting interpretation of observed results.

Such concerns about the nature of ABCC research and the future viability of the organization itself resulted in the creation, in 1955, of the Francis Committee. As discussed below, the report of the Francis Committee established a solid foundation for the ABCC-RERF research program. In 1975, in conjunction with the creation of RERF, a second major ad hoc review of ABCC research programs was undertaken by the Crow Committee. The recommendations of these committees had a major, positive impact on the ongoing research agenda at ABCC and RERF. In response to these NAS-initiated reviews, research was drastically redirected and strengthened in accordance with the committee recommendations. In both instances, this had a positive impact on the morale of the ABCC-RERF scientific staff.

Francis Report (November 1955)

Following preparatory meetings in the US, the Francis Committee met in Hiroshima and Nagasaki from 20 October 1955 through 9 November 1955 to carefully review research activities. Their report was completed in November 1955.

The Francis Committee strongly recommended the development of a unified study program centered on three platform protocols that would require the establishment of fixed cohorts: the Life Span Study (LSS) for mortality follow-up, the Adult Health Study (AHS) for clinical examination, and the Pathology Study. To do this, the ABCC staff defined the radiation exposure status for individual survivors included in the major ABCC samples. This so-called Master Sample became the source of fixed cohorts for the long-term studies. As originally defined, the LSS included about 100,000 people (about 26,500 of whom were not in either city at the time of the bombings). The AHS and the Pathology Study, defined as subsets of the LSS, totaled about 20,000 and 70,000 members, respectively. The development of the in utero and F1 cohorts also was a direct consequence of the Francis Committee recommendations.

The implementation of a unified core of platform research protocols introduced a sound epidemiologic approach to the monitoring of mortality and the incidence of various diseases of A-bomb survivors. Thus, the Francis Committee recommendations ensured that the A-bomb survivor studies would evolve into a credible basis for radiation risk estimates and radiation protection standards.
　
Names and affiliations of the committee members
Thomas Francis Jr Chairman Department of Epidemiology, University of Michigan
Seymour Jablon Follow-up Agency, NAS-NRC
Felix E Moore National Heart Institute, National Institutes of Health


Crow Report (26 March 1975)

Before the reorganization of ABCC into RERF, the Crow Committee was convened to (a) review the ongoing ABCC research protocols, (b) recommend which components of the ongoing programs should be continued by the new foundation, and (c) assess the need for introducing new technology to monitor the possible late effects of A-bomb radiation on somatic and germline cells in the A-bomb survivors and their offspring.

The committee members were US authorities in the fields of cancer epidemiology, radiation carcinogenesis, radiation biology, genetics, pathology, and hematology. They visited Nagasaki from 13 to 15 February and Hiroshima from 16 to 21 February 1975.

After extensive review of the programs, they recommended:
  1. Continuation and enhancement of the LSS and F1 mortality study.
  2. Modification of the AHS program to increase emphasis on laboratory measurements and reduce emphasis on physical examinations.
  3. Addition of the in utero cohort to the AHS.
  4. Gradual termination of the autopsy program, but with increased emphasis on the analysis of tissue specimens by pathologists.
  5. Implementation of a biochemical genetics program and continuation of the somatic and F1 cytogenetic studies.
On 26 March 1975, the committee’s report was approved by NAS-NRC. The current research activities at RERF strongly reflect Crow Committee recommendations.

Names and affiliations of the committee members
James F Crow Chairman Department of Human Genetics, University of Wisconsin
Henry S Kaplan Department of Radiology, Stanford University School of Medicine
Paul A Marks Faculty of Medicine, Columbia University
Robert W Miller Epidemiology Branch, National Cancer Institute
John B Storer Pathology and Immunology, Oak Ridge National Laboratory
Arthur C Upton School of Basic Health Sciences, State University of New York at Stony Brook
Seymour Jablon ABCC staff officer Medical Follow-up Agency, NAS-NRC

(return to contents)

Major Atomic-bomb Survivor Study Populations

Master sample

In response to the recommendations of the Francis Committee, the ABCC staff developed a comprehensive roster of Japanese A-bomb survivors and others who might be eligible for inclusion in one of the fixed cohorts. Data from the A-bomb survivor survey conducted at the time of the 1950 Japanese National Census were used to develop this roster. This survey identified 284,000 survivors including about 159,000 Hiroshima survivors and 125,000 Nagasaki survivors. The Master Sample, which was restricted to those survivors who were resident in either city at the time of the census, includes 98,000 Hiroshima residents and 97,000 Nagasaki residents. The Master Sample was supplemented with a small number of persons who were residents of Hiroshima or Nagasaki at the time of the census but who were not in the cities (NIC) at the time of the bombings (ATB). Since the late 1950s, all survivor studies conducted at RERF have focused on subsamples of the Master Sample. The sampling frames for the in utero and F1 populations are less well defined than the Master Sample. In utero survivors and potential controls were identified from birth records and ABCC investigations, supplemented with data from the A-bomb survivor survey conducted in conjunction with the 1960 national census. The population from which the F1 cohort was defined is based on data from the early genetic studies (mentioned earlier) and data obtained from officially maintained family registration (koseki) records of LSS members.

The following sections provide some background definitions and the current status of the major RERF study populations.

Life Span Study Sample

The LSS cohort was defined from the ABCC Master Sample. As originally defined, the LSS included each person in the Master Sample whose honseki (place of permanent family registration) was in Hiroshima or Nagasaki. This population was devided into four groups: a) those who were exposed within 2000 m of the hypocenter at the time of the bombing (ATB), b) persons situated 2000-2499 m from the hypocenter, c) an almost equal number who were 2500 to 10,000 m from the hypocenter who were matched with the first group by city, age, and sex, and d) a similar number who were not in the cities (NIC) and were also matched with the first group with respect to age ATB and sex. The original LSS cohort, which included 99,393 persons, was extended in the late 1960s and again in 1980. The first extension added all remaining members of the Master Sample exposed within 2500 m regardless of honseki. The second extension added all remaining Master Sample members within 10,000 m of the Nagasaki hypocenter ATB. With these changes, the size of the LSS was increased to 120,321 persons. However, over the past decade, most LSS analyses have focused primarily on the 93,741 LSS cohort members who were in the cities ATB.

Adult Health Study Sample

The original AHS cohort was defined based on the recommendations of the Francis Committee in 1955. The core group in the AHS includes all 4993 exposed persons in the original LSS known to be alive in 1950 who were within 2000 m of the hypocenter ATB and who reported symptoms of acute radiation exposure. The remainder of the AHS population includes three city-, age- and sex-matched samples drawn from the LSS. Each of these samples is approximately the same size as the core group. The three groups are 1) survivors without acute radiation symptoms who were within 2000 m of the hypocenters ATB; 2) distal survivors who were 3000 to 3499 m from the hypocenter in Hiroshima and 3000 to 3999 m in Nagasaki; and 3) persons who were not in the cities ATB.

In 1977, because of concerns about attrition of the high-dose survivors, the original AHS sample was enlarged by adding all surviving exposed LSS members with assigned tentative 1965 dosimetry dose estimates in excess of 1 Gy and an equal number of age- and sex-matched controls. This led to the addition of 2436 persons to the 22,397-person AHS cohort. Active follow-up of the not-in-city portion of the AHS cohort ended in the 1980s.

Pathology Study Sample

The Pathology Study was a major component of the Unified Study Program proposed by the Francis Committee. This subsample consisted of 70,000 original LSS members whose families resided in Hiroshima and Nagasaki and could be contacted to grant autopsy permission. The original objectives of this study were to perform as many postmortem examinations as possible on all persons in the LSS sample to determine specific causes of death and to search for other possible radiation-induced changes. The autopsy rates in the early 1960s were relatively high compared with those in the other parts of Japan; however, they declined substantially in subsequent years. As a result of the marked decrease in the number of autopsies, the Scientific Council recommended in 1978 that the autopsy program be modified to emphasize a more active relationship with physicians in both cities for notifying RERF of any deceased person known to have been exposed to A-bomb radiation. It was also recommended that research programs relevant to radiation biology in cytopathology, immunopathology, and tissue culture be developed and implemented. In addition to the autopsy program, the surgical pathology program was active from the late 1950s through the late 1960s. Under this program, tissue specimens, including tissue blocks, were submitted to ABCC, providing a useful source of information and specimens before the establishment of the tissue registries. A large number of tissue specimens have been collected over the years, including 7500 autopsy and 13,000 surgical specimens for RERF cohort members.

Our pilot studies have demonstrated the usefulness of these old archived tissues for molecular biological studies. However, it must be recognized that with the discontinuation of the RERF autopsy and surgical pathology programs, pathology materials for more recent cases will have to be obtained from outside sources.

In Utero Sample

Around 1960, two overlapping fixed cohorts of persons who were in utero ATB were assembled from records on about 10,000 births that took place in or near the two cities and occurred between the time of the bombings and the end of May 1946. The in utero clinical cohort included all Japanese survivors exposed in utero who were within 1500 m of the hypocenter ATB together with city-, sex-, and date-of-birth-matched samples from two comparison groups of comparable size. The comparison groups were chosen from persons whose mothers were 3000 to 5000 m from the hypocenter ATB or who were NIC ATB. The in utero clinical cohort includes 1608 persons of which a subset of 1021 is contacted to voluntarily participate in the AHS biennial examination program.

The in utero mortality cohort currently includes 2817 subjects. This cohort, which was established around 1964, was defined in terms of a core group consisting of all persons exposed in utero within 2000 m of the hypocenter and three similarly sized, matched comparison groups chosen from persons exposed in utero at distances of 1500 to 2000 m, 2000 to 3000 m, and 3000 to 10,000 m from the hypocenter. An additional comparison group of comparable size was selected from NIC children born during the nine months after the bombings.

Current analyses of mortality and cancer incidence among the persons exposed in utero make use of data from the combined clinical and mortality cohorts. This combined cohort now comprises 3654 persons exposed (within 10 km of the hypocenter ATB).

Cohort of First-generation Children Born to Survivors

Studies of genetic effects have been of central importance since the earliest days of ABCC. In the early period, major efforts were made to obtain data on all births in the two cities for a series of studies on congenital abnormalities, sex ratio, and other genetic effects of radiation exposure. By the mid-1950s, it had become apparent that there would be no dramatic findings. Nevertheless, the Francis Committee recognized the need for continued follow-up of the children of the survivors and suggested that a fixed cohort be defined. Shortly after, the original F1 mortality cohort was defined, although in several poorly documented stages that have led to some confusion about the size and structure of the so-called F1 cohort. Because of this confusion, which was heightened by the introduction of the dosimetry revision, called Dosimetry System 1986 (DS86), an internal ABCC Genetics Study Database Committee was organized to review and clarify the definition of the F1 cohort. The description that follows is based on the committee’s findings.

This cohort includes 54,243 persons born between 1 May 1946 and the end of 1958; follow-up data are available for only 53,518 of them. The core group included all eligible children with at least one parent within 2000 m of the hypocenter ATB. There were two sex- and age-matched comparison groups of the same size as the core group. The first comparison group included children having at least one parent exposed between 2500 m and 9999 m from the hypocenter. The second comparison group included children neither of whose parents were within 10,000 m of the hypocenter ATB.

The original F1 cohort has been extended on three occasions. These extensions, which involved the addition of all children born between 1 May 1946 and 31 December 1984 with at least one parent in the extended LSS cohort, increased the size of the F1 cohort to 88,485.

Overlapping subsets of the F1 cohort have been used in various ABCC-RERF study programs. The following table shows the size of the groups selected for or included in specific F1 study programs.

Program Number of persons
Mortality follow-up
76,816
Laboratory examination
25,695
Biochemical genetics study (BGS)
23,731*
Cytogenetic study
16,392*
Peripheral lymphocyte culture (PLC)
3,690**
*The figures for the BGS and cytogenetics programs indicate only the number of people examined and do not inidcate the total number of people from which these people were selected. There is considerable, but not complete,overlap of these two programs.

**The number of F1 cohort members in the sampling frame for the PLC studies. Plans call for establishing cultures for about 1500 F1 members and their parents (roughly 1000 families).

The Genetics Study Database Committee recommended that mortality follow-up be extended to include persons participating in the laboratory examination programs (about 4000 additional persons) and, if resources allowed, all persons in the F1 cohort. These recommendations were endorsed by the RERF Scientific Council in 1994, but due to budgetary restrictions, they have not yet been implemented.

(return to contents)

Atomic-bomb Dosimetry


Characterization of the radiation doses received by individual survivors has been central to the work of ABCC and RERF. During the first decade of ABCC operations, no individual dose estimates existed, and analyses were based on comparisons of proximally exposed survivors to those exposed at greater distances or to persons not in the city ATB. In the late 1950s, the US government declassified the so-called York air-dose curves, which allowed one to compute an estimated dose for a person given his or her distance from the hypocenter. The York curves were used as the basis of the tentative 1957 dosimetry (T57D). However, T57D was never officially adopted by ABCC, and the doses were used only on a limited basis. In the 1960s, a major effort led by scientists at Oak Ridge National Laboratory was undertaken to develop a comprehensive dosimetry system for use in the ABCC studies. This led to the development of the tentative 1965 dosimetry (T65D) which provided the basis for ABCC-RERF risk estimation after 1968. The T65D doses were the first generally accepted dose estimates for the survivors.

The T65D system includes equations for computing free-in-air gamma and neutron kinetic energy released in materials (kerma) estimates as a function of distance from the hypocenter. T65D also used various methods for computing the effect of shielding by external structures or terrain. Shielded kerma estimates for proximal survivors depended heavily on information contained in shielding histories that were obtained for virtually all LSS cohort members. Detailed shielding histories were obtained for more than 28,000 survivors, including most survivors who were within 1600 m of the hypocenter in Hiroshima and 2000 m in Nagasaki. Less detailed shielding information is available for all members of the Master Sample. In the late 1970s, organ transmission factors were made available to RERF, but virtually all T65D-based analyses made use of shielded kerma estimates.

By 1981, problems arose with T65D, and a major binational effort to develop a new dosimetry system was begun. This led to the development of the DS86 dosimetry, which has been used in RERF analyses since 1987. DS86 provides more detailed information on dose components than T65D and allows the computation of doses for 15 different organs. Under DS86, neutron dose estimates in Hiroshima were much lower than the corresponding T65D estimates, and unshielded DS86 gamma doses were generally higher than the corresponding T65D estimates, but external structures, particularly typical Japanese houses, were assumed to provide more shielding than had been assumed in T65D. While some questions have been raised about the DS86 estimates, they continue to be used as the basis of RERF risk estimates.

For the past year, RERF staff have been working to ensure that the DS86 system runs properly on the new RERF UNIX-based computer system. We also have been trying to design a modern data base to allow more efficient and effective use of the data on shielding and dosimetry.

(return to contents)

Follow-up Methods


Koseki Record Checks

Mortality follow-up in the RERF cohorts is carried out through a special arrangement that allows RERF access to the records of the mandatory family registration (koseki) system. Records for all members of a family are stored at the place of registration (honseki) of the head of the family. The koseki record includes information on current residence and vital status. Koseki records for all surviving members of the mortality cohorts are checked at least once every three years. Copies of death certificates are obtained for all deceased persons. Data on cause of death is coded at RERF following International Classification of Diseases rules and codes. The ability to access the koseki records guarantees virtually complete mortality ascertainment for persons who have not emigrated from Japan. If a person does emigrate, this fact is noted in the family record. Emigrants in the mortality cohorts constitute a small group and are treated as lost to follow-up as of the emigration date indicated in their koseki.

Tumor and Tissue Registries

The earliest program for monitoring cancer incidence in A-bomb survivors involved the registration of leukemia and other hematological disorders. This program, known as the Leukemia Registry, grew out of efforts initiated in the late 1940s by physicians at the Hiroshima Red Cross Hospital and Nagasaki University. Since the early 1950s, this registry has been operated by ABCC-RERF with the cooperation of hematologists in both cities. Leukemia Registry data has been used as the basis of numerous reports on leukemia in survivors. Recently, the Leukemia Registry has been relatively inactive, largely because cases of leukemia and other hematopoietic cancers are routinely ascertained by the Hiroshima and Nagasaki tumor registries.

With the cooperation of the local medical associations, population-based tumor registries were established in 1957 in Hiroshima and 1958 in Nagasaki. Since their inception, the registries have been operated by the staff of ABCC-RERF. Data from the registries is routinely linked to individuals in the LSS, in utero, and F1 cohorts. The registries ran smoothly for the first few years, but political problems arose that limited the usefulness of the Hiroshima registry data. However, during the 1980s, these problems were resolved, and a major effort was begun to update and standardize the registry data and to modernize the operation of the registries. By the early 1990s, tumor registry data were felt to be of sufficient quality and completeness to warrant the analyses that led to a series of comprehensive reports on cancer incidence in the LSS.

The Hiroshima and Nagasaki registries are unique among Japanese cancer registries because they use active case ascertainment methods rather than relying on voluntary reporting by physicians; that is, trained abstractors visit hospitals and clinics to review records. The Hiroshima and Nagasaki registries are generally regarded as the best tumor registries in Japan. Data from these registries are routinely reported in the International Agency for Research on Cancer publication series, Cancer Incidence in Five Continents.

Prefectural tissue registries, managed by RERF in cooperation with the local medical associations, began operation in 1973. The tissue registries are a valuable supplement to the tumor registries and are especially useful for conducting site-specific cancer studies which require standardized reviews of pathology material. They also are likely to become increasingly important as a source of material for molecular studies linked to the RERF cohorts.

Mail Surveys and Interview Studies

Since 1960, interviews and mail surveys have been conducted periodically among the LSS sample members. These surveys provide cross-sectional and longitudinal data on a broad range of disease risk factors, such as tobacco smoking, alcohol intake, dietary and nutritional factors, reproductive history as well as occupations and other socioeconomic variables. Although these data have been available, past analyses were restricted largely because of the difficulty in managing and analyzing large data sets from various files. Recent progress in accessing and merging data from various sources facilitated analysis of these data. Some aspects of the roles of nonradiation factors have recently been published. Much more work is needed in this important area of research. The case-control interview approach has also been useful in assessing confounding and interactive effects of factors other than radiation exposure. It is especially useful when combined with studies of biological materials, as exemplified in the nested case-control study now underway on liver cancer in relation to hepatitis-B and C infection and conventional risk factors.

Adult Health Study Biennial Examinations

Regular biennial clinical examinations of the AHS sample began in 1958. The examination schedule was set so that in any one month a more or less representative cross section of the entire population would visit the clinic. These examinations are now in the 19th cycle and serve as the only point of direct contact between RERF and the survivors. The survivors benefit from these examinations and consider them a useful service provided continuously by ABCC-RERF. RERF research benefits from the acquisition of blood and other biological specimens, which have been and continue to be used in the clinical laboratory and in genetics and radiobiology programs.

Over time, attrition has reduced the AHS population to about 50% of the original sample; however, among exposed cohort members who still live in Hiroshima or Nagasaki, participation rates remain in excess of 70%. This high participation rate reflects the spirit of cooperation of the survivors, an effective patient contacting group within RERF, and a program of night clinic and home visits for those who cannot come to the RERF clinic for the regular examination. Recently, efforts are being made to develop procedures for more frequent contact by letter or phone with members of the AHS, and some thought is being given to developing epidemiological disease surveillance systems for the AHS.

The RERF medical examination consists of a complete physical examination and includes anthropometric measurements, blood pressures, electrocardiography, radiography, and abdominal and thyroid ultrasonography. In addition to routine laboratory tests, specific biochemical and immunological assays for cancer screening are recorded.

In addition to routine data collection for longitudinal analysis spanning the entire study period, specific clinical studies have been conducted over more limited examination cycles for cross-sectional analysis.

(return to contents)

Data Management and Computing at ABCC-RERF


Early Processing Methods

Methods and resources for data management and computing at ABCC up through about 1965 were equivalent to the best available elsewhere in Japan. From 1949 to 1963, a keypunch-card-based system was used for data processing. Tasks were performed on three main machines: a card sorter/counter; a collator that combined two sets of punch card data according to a designated matching field; and a machine that performed basic calculations. To carry out tasks, technicians hand-wired entire sets of instructions into these machines.

Mainframe Computing

A major system transformation took place in 1961, when an IBM mainframe computer was introduced for data management, statistical analysis, and business computing. With the volume of accumulated data growing rapidly, large reel magnetic tapes were introduced in 1963 and eventually overtook punch cards as the primary permanent storage media. The use of hard disks began in 1965 to provide more immediate access to the most commonly used data. A succession of different IBM models was in use until 1976, when a major transition effort to an NEC mainframe system was made. Several NEC model upgrades were made in the ensuing years.

The change to the NEC system largely reflected the 1975 transition from ABCC, a US-funded organization, to RERF, a binationally-funded foundation. Although the NEC mainframe systems proved extremely reliable in operation, the lack of commercial software using up-to-date technology to perform comprehensive database management and scientific data analysis was a major drawback. The incompatibility of NEC systems with IBM mainframes and the smaller mainframe market share held by NEC contributed to lack of interest by software vendors in providing software for NEC mainframes. Proposals to stay with IBM or move to Japanese-manufactured IBM-compatible mainframes were unsuccessful. To help cope with the situation, efforts were made in the early 1980s to convert IBM mainframe software when source codes could be obtained, and new statistical software was developed in-house to meet the specialized analysis needs at RERF.

Transition to Distributed Computing

To further address software needs not met by the mainframe computer and to introduce new technology to the RERF computing system, efforts were initiated to partially decentralize computing resources. The first significant move in this direction came with the advent of personal computers (PCs), which were introduced to RERF beginning around 1983. During the mid-to-late 1980s, the number of computer users grew along with the scope and complexity of their tasks, and it became clear that measures would be necessary to further address unmet software needs and to deal with mainframe inefficiencies and exorbitant rental costs as well as other issues.

By the late 1980s, modern technology, increased power, and decreasing prices of PCs and UNIX reduced instruction set computing (RISC) workstations, provided newer and more relevant platform alternatives to improve the needs for data management, analysis, and communication that had not been met by the mainframe system. The newer hardware alternatives, the need to address current and future RERF computing needs, and the gaining momentum behind computer system downsizing both outside and within Japan provided the impetus for change. So, in fall 1991 efforts were launched to build a distributed, networked computing environment linked to the Internet that would replace the antiquated mainframe system. A foundation-wide effort unprecedented in size was undertaken in both cities to replace equipment and address other issues. At the end of 1994, the mainframe system was removed, 15 months ahead of schedule. This ended an era of computing centered around mainframe systems that had spanned almost 34 years.

Current Computer Environment

The increased power and capabilities of present-day PCs have brought powerful tools directly to the desktops of research and support staff, increasing the accessibility and integration of computing resources into daily activities. PC software now handles many of the analysis and automation needs in each department. RISC workstations provide the additional analysis power needed for more computation-intensive jobs, serve as the engines that run RERF database operations, and support Internet services. The new internal network allows staff in both cities to communicate with each other; to exchange messages, documents, and files more efficiently; to share common resources such as printers, disk drives, and hard drives; and to access and run software residing on the workstations. The link to the Internet enables access to researchers and resources worldwide through use of email, access to newsgroups and other reference information, electronic file transfers to upload/download information, and login capabilities to/from other computing facilities (when authorized). The Internet also increases the visibility of RERF in Japan and to international communities. In turn, RERF is working to use Internet resources, eg, the World Wide Web, to give the outside community easier access to, and better awareness of, the information that RERF has available for the public.

Current Data Management Activities

In data management, a new relational data base has been developed on the RISC workstations to process and manage RERF scientific data. Data from the master file, tumor registry, tissue registries, LSS, and DS86 dosimetry are included. Master file and tumor registry personnel use the system to record new information on individuals in the different RERF mortality and incidence studies. The new data base has eliminated data redundancy and greatly improved methods used by research staff to retrieve data, check data quality, and prepare working files for data analysis. Security built into the data base prevents access to individual identification information, such as name and address, except by authorized staff responsible for maintaining and updating that information. The shift to the new computing environment has sparked interest to carefully clarify the definitions and data sources of the major cohorts prior to inclusion in the database; work for the F1 and in utero cohorts is nearing completion. Much data, however, remains stored on disk or tape as independent flat files. More of this data needs to be integrated into the relational data base to expand ways that the data can be combined to provide more complete information on each study cohort member and to save time in performing data linkage tasks.

(return to contents)

Results of ABCC-RERF Studies


Life Span Study Reports

The series of comprehensive reports on mortality in the LSS cohort are generally considered to be the most important summaries of ABCC-RERF findings. As the mortality data have accumulated, these reports have evolved from relatively straightforward presentation of simple tests for radiation effects to detailed analyses and characterization of factors that affect the risk of death from cancer or other causes associated with the exposure to radiation from the atomic bombs. The most recent report, which will be published in 1996 [Radiat Res 146:1-27, 1996], discusses cancer mortality for the period 1950 through 1990. Particular attention is given to the nature of variations in the excess risks with sex, age at exposure, and time. The report also includes estimates of lifetime risks associated with radiation exposure in the LSS cohort.

Results given in the previous LSS report indicate that noncancer mortality increases with increasing radiation dose. A new report on noncancer mortality is being prepared. It is hoped that information obtained from the additional years of follow-up will lead to a clearer description and a better understanding of this important finding.

Because of the larger number of cases (despite restrictions on the follow-up period for solid cancers), more accurate diagnoses, and the ability to consider nonfatal cancers, the recent publication of the tumor-registry-based analyses of cancer incidence in the LSS was a landmark in ABCC-RERF history. Work on cancer incidence is continuing with a series of detailed site-specific incidence studies, including cancers of the breast, liver, lung, skin, salivary gland, brain and central nervous system, thyroid, ovaries, and lymphoma. An important issue that must be addressed in the near future is the use of both the mortality and morbidity data in developing a comprehensive description of cancer risks in the survivors.

Since the publication of LSS Report 10 in 1989, it has been the policy of RERF to release the data used in major LSS reports in a form that allows researchers outside of RERF to reproduce and even extend the results of our analyses. At this time, data sets for LSS Report 10 and LSS Report 11 in 1989 (including data on noncancer mortality) as well as data from the recent analyses of incidence of solid cancer and hematopoietic cancers in the LSS and F1 mortality and cancer incidence data are available to anyone who requests them. These data sets have been used as the basis of many publications by researchers outside RERF.

Summary of Cancer Findings

The LSS provides the most comprehensive data on the risk of cancer associated with whole-body exposure to low linear energy transfer (LET) radiation. Statistically significant excess risks have been demonstrated for leukemia, multiple myeloma and many of the most common types of solid cancer, including stomach, lung, liver, colon, esophagus, female breast, ovary, and bladder as well as a group of cancers at sites not considered in site-specific analyses. Statistically significant excess risks have not been detected in the LSS for cancers of the rectum, gallbladder, pancreas, uterus, and brain and central nervous system or for lymphoma. However, in view of the general evidence for a radiation effect at so many other sites, the failure to find significant effects at these sites should be interpreted cautiously. It may be due to small numbers of cases for some sites, even in a large cohort such as the LSS. Indeed, for few, if any, of the solid cancer types do the site-specific relative risks differ from the overall relative risk for all solid cancers as a group.

For leukemia, the dose response is nonlinear with an upward curvature, whereas it is remarkably linear for solid cancers. While the excess risks for leukemia have tended to decrease with time since exposure, the (absolute) excess rate for solid cancers has increased throughout the follow-up in a manner that is roughly proportional to the increase in background rates with increasing age. For most nonsex-specific cancer types, excess relative risks for women are about twice those for men. However, for many types of cancer, age-specific background rates for women are about half those in men, so differences in excess rates for men and women are not large.

Since the ABCC-RERF fixed cohorts do not include persons who died between the time of the bombings and 1950, the results may reflect a resistant subpopulation of survivors who are not representative of overall human risk. However, study after study has failed to show any difference in the radiation sensitivity of the survivor’s cells as a function of the survivor’s radiation dose.

In Utero Studies

The earliest studies on the in utero-exposed group demonstrated significant reduction in head size, ie, microcephaly, in 62 of the exposed children. Subsequently, 30 cases of severe mental retardation were found among this group. A recently published analysis of those cases, which were due to exposure and not Down’s syndrome, demonstrated that the 8- to 15-week period of gestation is the most sensitive period for induced mental retardation; weeks 16 to 25 are the second most sensitive period. Different threshold dose levels can be suggested for these two periods, but a linear response can not be excluded. IQ and school performance records for 1544 cases demonstrate significant decrease with increasing doses down to about 0.1 Gy.

Published results on cancer incidence have indicated a possible increase in the exposed cohort, but this was based on a small total number of cases.

Adult Health Studies

A number of specific publications from the Department of Clinical Studies have addressed a variety of radiation-related health issues.

In the area of cancer, the increase of breast and thyroid cancers was first demonstrated in the AHS subjects and subsequently confirmed by the LSS programs. Additionally, the biological material provided by the participants was analyzed to determine if viral infection was acting as a confounder for the dose-related increase in hepatoma frequency. Hepatitis B infection was shown not to affect this result. The AHS program also serves as an important resource for establishing cancer incidence data in the LSS, which depends on an accurate estimate of the number of survivors living in the catchment areas. The program determines who has left the regions through frequent personal contact, mailings, and telephone contacts.

Studies on nonmalignant neoplasms have been reported in our technical report (TR) series. Benign tumors of the parathyroid, thyroid, and uterus have shown significant dose-related increases, initially in the cross-sectional prevalence studies and subsequently confirmed in incidence analysis.

Three key papers published as TRs have been devoted to radiation-induced lenticular opacities (cataract) covering a 30-year period beginning in 1951. Recent reanalysis of this data using the DS86 dose system suggests that these induced opacities do have threshold doses relatively high for gamma irradiation (>0.7 Gy) and quite low for neutron exposure (<0.1 gy=”” what=”” remains=”” to=”” be=”” determined=”” is=”” whether=”” such=”” opacities=”” are=”” progressive=”” in=”” later=”” years=”” as=”” has=”” been=”” suggested=”” in=”” animal=”” radiation=”” studies=”” and=”” whether=”” new=”” cases=”” continue=”” to=”” appear=”” br=””>
In a series of papers involving cardiovascular disorders, heavily exposed women were first demonstrated to have increased incidences of myocardial infarction and stroke. The excess relative risk overall is about 10% to 15% that of cancer, but it is a significant risk, and again, this data served as an alert to what was later seen in the lifespan noncancer mortality study. By 1986, it was also apparent from X-ray examinations that the prevalence of aortic arch calcification was considerably greater in high- versus low-dose groups and markedly greater in the 20 to 29 age ATB group from the age of 45 onwards.

Longitudinal analysis over a 28-year period of AHS examinations revealed that the incidence of both chronic hepatitis and liver cirrhosis were increased significantly by increased exposure levels and served again as a forerunner of the evidence of a significant liver cancer incidence in the life span study.

Because the AHS examinations on a two-year cycle include height and weight measurements, it was possible to track the sequential growth pattern of those exposed prenatally as well as those exposed before the age of ten. Statistical analysis of these large data sets have also demonstrated that there were significant reductions in stature, the high-dose groups showing the most marked reduction. It is not clear whether these effects result from hormonal changes, such as thyroid dysfunction as noted in the Marshall Island study, or other direct effects on bone growth. What was clear was that there was no apparent socioeconomic, eg, nutritional, differences that could account for these growth differences.

It is obvious that a wealth of untapped information exists in the health data accumulated to date, over 38 years, not only on direct effects of radiation but also on normal aging processes in a population followed over its lifetime.

Genetics Studies

Before the ABCC genetics program was initiated, a distinguished panel of geneticists was convened in 1946 by the NAS to evaluate the proposed study. Concerned with the uncontrollable variables that might influence genetic studies, the panel nevertheless supported the work being done at RERF. Their conclusion was that “this unique possibility for demonstrating genetic effects caused by atomic radiation should not be lost.”

By March 1948, ABCC began an extensive program of clinic examinations and home visits of newborns in both cities. From 1948 to 1954, more than 76,000 children were examined by ABCC physicians; 30% of these children were re-examined 8 to 10 months later. Information was obtained on abortions, stillbirths, and neonatal deaths by means of the Japanese registration system that required women to register after the fifth month of pregnancy in order to receive extra rice rations. Congenital malformations were screened by the physicians. Analysis through 1953 revealed no significant genetic differences between the offspring of parents from the exposed and nonexposed groups, and the clinical studies were discontinued by 1954. A major publication summarizing these findings was produced in 1958. From 1955 to 1968, data collection dealt with sex ratio and neonatal death.

From 1968 to the present, cytogenetic studies have been undertaken involving more than 16,000 children, about one-half of whom are the progeny of exposed parents. Sex chromosome aneuploidies and balanced translocations, which are not selected against after birth, were the endpoints studied.

In 1976, a study on changes in protein structure was begun. For each individual, 30 blood proteins were examined by one-dimensional electrophoretic techniques. This massive study involved more than 23,000 individuals.

Simultaneously during these decades, ABCC also collected mortality and morbidity data on certain parts of the F1 cohort.

A series of papers was published in 1990 reporting results of mortality analyses that used DS86 dose estimates and incorporated additional years of follow-up. None of the endpoints studied to date have shown significant dose-related increases. A major publication in 1990 attempted to assess the doubling dose, a measure of the radiation sensitivity of parental germ cells. Given the uncertainties in such data, the following conclusion was reached: the human data indicates that doubling doses are in the range of 3 to 4 times greater than estimates developed from laboratory mouse experiments. Thus, humans are possibly one-third to one-fourth less radiosensitive than the mouse. The most important ABCC-RERF genetics study publications have been compiled in one volume titled The Children of the Atomic Bomb Survivors. A Genetic Study (edited by JV Neel and WJ Schull. Washington, DC, National Academy Press, 518 pp, 1991.)

Even though the results on genetic effects on the children of the survivors have demonstrated no significant increases, it is well established that high-energy radiation produces mutation in every species ever tested in laboratory studies. It would be indeed unexpected if the most sensitive tests that were becoming available by the mid 1980s, namely studies at the molecular level of the gene itself, DNA, would be refractory to analysis. Therefore, in 1985, the biochemical genetics laboratory began to develop the necessary technologies to enable detection of all possible changes in the organization of gene structure. Simultaneously, plans were formulated to establish cell lines from 1000 (500 exposed and 500 control) families, father-mother-and-child trios. Cell lines, approximately 109 cells, untransformed lymphocytes, and granulocytes are preserved in liquid nitrogen.

A workshop on human germline mutagenesis was held at RERF in 1991 to critique the status of newly emerging DNA technologies. The workshop recommended that a pilot study be started on 100 families, 50 with exposed parents and 50 unexposed. This study was to test the efficiency of three major DNA technologies, denaturing gradient gel electrophoresis, Southern blotting, and microsatellite screening. The results of these pilot studies have recently been published. Meanwhile, our researchers have developed new techniques that score 2000 spots of DNA fragments separated by two dimensional gel electrophoresis. This method primarily detects insertions, deletions and rearrangements of DNA and single DNA base substitutions at the recognition sequences of restriction enzymes that cut DNA to fragments. Pilot studies indicate that the method has the accuracy to detect a 50% increase or decrease in spot intensity caused by the changes in DNA.

Major studies on the 1000 families will begin when the most efficient technologies in cost, time, and manpower can be utilized and the most appropriate sets of target genes can be analyzed.

Cytogenetic Study on Survivors

Cytogenetic studies have been conducted routinely since 1968 using the peripheral blood lymphocytes of A-bomb survivors selected from among the AHS participants. Emphasis has been place on establishing a dose-response relationship for the purpose of biodosimetric analysis. By scoring cells with stable chromosome aberrations (mainly translocations), which should not be under negative selection from the time of induction, each survivor’s response to irradiation is determined. So far, a conventional study on aberration frequencies has been performed for 2500 AHS participants in the two cities. The results showed that the aberration frequencies increase with increasing dose assigned to each survivor. This has been confirmed by the new molecular technique termed fluorescence in situ hybridization (FISH) that provides unbiased estimates of translocations induced in somatic cells by exposure to atomic radiation in 1945. Further, based on repeated tests, the results suggested that the cells with stable chromosome aberrations persist with constant frequency for decades in the irradiated individuals. Both conventional and FISH data indicated that physically assigned DS86 doses for those exposed in the open or in factories are often overestimated as compared to those exposed inside Japanese houses. This trend is more pronounced for the Nagasaki survivors.

Preliminary analyses based on electron spin resonance (ESR) measurement, another sensitive dose-verification method, using tooth enamel of survivors indicate that dose estimates obtained by this method correlate more closely with the chromosome aberration frequency of an individual than with the DS86 dose assignments. This ESR approach provides key information for some unusual responses, such as decline of severe epilation frequency at high doses, which is most likely due to biased dose estimates rather than differential radiosensitivity.

Radiobiology Program


During the late 1970s, visiting scientists introduced modern immunological techniques into the AHS clinical program. Almost simultaneously, experimental radiation studies using live tissues were undertaken within the pathology program, the forerunner of the cell biology unit. A new immunology laboratory was completed in 1980, and an immunology research staff was recruited. Recommendations by the Scientific Council to strengthen and modify aspects of these research units led to major structural reorganization of RERF departments in 1985, at which time the immunology and cell biology groups were combined to form the Department of Radiobiology.

The important issues to be addressed by this research team included:
  1. Determining by means of cellular studies whether radiosensitive subsets in the AHS cohort exist, which might explain different early and late responses to A-bomb radiation exposure for survivors assigned similar doses. Critics of RERF’s cancer studies claim that highly radiosensitive individuals died soon after the bombings, resulting in cancer risk estimates biased toward a radioresistant and therefore underestimated population. Epidemiologic analysis could not refute these criticisms. However, experimental radiation studies on fibroblast and lymphocyte cells from high-dose individuals, low-dose or nonexposed individuals, including those who did and did not experience early radiation sickness symptoms and/or subsequent cancers, were shown to respond in similar fashion to cell killing and other measurable endpoints.
  2. Determining whether changes in immune competence and specific changes in lymphocyte subsets are induced by the A-bomb exposure. These studies have shown dose-related changes in immune system response, and continuing studies are addressing further aspects of the problem.
  3. Determining whether the somatic cells of survivors can be used to assess, at specific gene loci, any changes in rates of mutation that resulted from A-bomb exposure decades earlier. A number of studies done at RERF have demonstrated that lymphocyte somatic mutation systems do not have “long-term memory,” ie, selection has intervened to reduce or eliminate mutant-bearing cells, probably because deletional changes encompassing the loci under study were deleterious to the survival of such cells. Conversely, erythrocyte cells (derived from mutated erythroblasts) lack nuclei, and mutations resulting in loss of M or N antigen specificity among MN heterozygous individuals would not be at selective disadvantage. A test system using glycophorin A has been developed successfully by scientists at Lawrence Livermore National Laboratory, and further technical simplifications by our staff were introduced. Rapid cell sorting of millions of cells can determine the somatic mutation rate, either naturally occurring or induced for an individual. Based on population studies of AHS members, it has been shown that aging increases the natural mutation rate and that the mutation frequency is dependent on radiation dose.
    In the early 1990s, the RERF Scientific Council concluded that this program was a world leader in somatic mutation studies.
  4. Determining whether molecular changes in tissue obtained from survivors underlie carcinogenesis. Will radiation-induced cancers have unique molecular alterations distinguishing them from cancers of a different origin? These studies are in the early stages of experimentation.
  5. Continuing the introduction of new technologies to assess radiation damage or changes in cells of the A-bomb survivors. To this end, lymphocyte cells of all AHS participants are cryogenically preserved for future analysis. This will provide researchers with adequate samples of biological material even when the survivors are deceased.
(return to contents)

Dissemination of Research Results


Technical Report Series, 1959-1992
The official publication of ABCC technical reports (TRs) began in 1959. At that time, selected earlier research results, some of which already had been published in the scientific literature in English, were published again as TRs with Japanese and English texts appearing side-by-side. This method of bilingual in-house publishing and distribution to a mailing list of approximately 800 persons and institutions worldwide continued until 1989, when only selected TRs were thereafter produced in both languages. Between 1959 and 1992, 830 TRs were published by ABCC-RERF.

In 1993, in-house TR production and distribution ceased primarily because of the long production time required and escalating costs. TRs were replaced by RERF Reports, which begin as manuscripts that are officially approved by the RERF permanent directors for submission to journals. Upon acceptance by a peer-reviewed scientific journal, a manuscript is designated as an RERF Report, and journal article reprints are purchased from the publishers and bound into RERF Report covers with Japanese summaries. Although distribution of RERF Reports varies depending on the subject matter, at least 100 are sent to Japanese and American governmental ministries and agencies, local hospitals and A-bomb survivor groups, libraries, and RERF directors and consultants. Thus, RERF now relies mainly on journal articles to report scientific findings to the world.

Periodicals Produced by RERF

RERF produces the following periodicals to provide information for the local scientific community and for members of the national and international radiation-research community, and for RERF employees–past and present.

Journal of the Hiroshima Medical Association
Since 1960, the monthly Journal of the Hiroshima Medical Association has included a section devoted to ABCC-RERF research findings. Page charges have been contributed by ABCC-RERF. About 6,300 copies of this journal are sent to the members of the medical associations in the community, and medical libraries and other medical-related institutions throughout the country. In addition, reprints are purchased and distributed by RERF to the local media and medical institutions.

RERF Update
Since 1989, RERF has prepared and distributed this English-language publication, with institutional news and articles regarding present and past RERF investigations. This has been distributed to more than 1,000 individuals and institutions interested in RERF. Starting in Summer 2001, this is published on the RERF Homepage in both English and Japanese twice each year and in somewhat different form.

RERF Newsletter
Since the days of ABCC (1963), the monthly Newsletter (RERF Newsletter from 1975) has been published in Japanese to disseminate institutional, scientific, and employee news and distributed to individuals and institutions free of charge. It was redesigned in July 1996 as an in-house newsletter for the RERF employees, past and present.

RERF Annual Report
Bilingual documentation of ABCC’s research activities, administrative transactions, and miscellaneous events began in 1957 during the chairmanship of George B. Darling. Originally published under one cover, separate annual reports in English and Japanese have been produced since 1982 for distribution free of charge to individuals, institutions, and libraries. Limited copies are on file in the RERF Archives Unit, Library and Archives Section.

From 1950 to 1957, ABCC semiannual reports were published, comprising separate documents for research and administration. In 1950 and 1951, ABCC quarterly reports also were issued. A few copies of these publications are kept in the RERF Archives Unit, Library and Archives Section, and in the RERF Hiroshima Library.

(return to contents)

戻る