"Low Level
Radiation Health Effects: Compiling  the Data"

Revision 2
March 19, 1999

by Radiation, Science, and Health, Inc.,
Edited by J. Muckerheide

1.3
Animal & Plant
Biology

1.3.1
Mammals

References

"This study was designed to measure the life shortening of C57BL/6J male mice as a result of exposure to five external doses from 60Co gamma radiation delivered at six different dose rates. Total doses ranged from 20 to 1620 rad at exposure rates ranging from 0.7 to 36000 R/day. The ages of the mice at exposure were newborn, 2, 6, or 15 months. Two replications were completed. Although death was the primary endpoint, we did perform gross necropsies. The life span findings are variable, but we found no consistent shortening compared to control life spans. Therefore, we cannot logically extrapolate life shortening to lower doses, from the data we have obtained. In general, the younger the animals were at the beginning of exposure, the longer their life spans were compared to those of controls. This relationship weakened at the higher doses and dose rates, as mice in these categories tended not to have significantly different life spans from controls. Using life span as a criterion, we find this study suggests that some threshold dosage may exist beyond which effects of external irradiation may be manifested. Up to this threshold, there is no shortening effect on life span compared to that of control mice. Our results are in general agreement with the results of other researchers investigating human and other animal life span effects of irradiation."

I. INTRODUCTION
"Many researchers have studied the radiation-induced effects on life span and considered injuries or diseases induced by exposure to ionizing radiation. In 1965 L. H. Strong referenced 96 publications on the subject, more recently (March 1978) life shortening and late biological effects of exposure to ionizing radiation were reviewed in the proceedings of an International Atomic Energy Agency Symposium in Vienna. Storer et al. (1979) also referenced much of the literature on radiation-induced life shortening. One fascinating treatise examines the effects of ionizing radiation and hormesis,’ defined as the stimulation of any system by sub-harmful amounts of any agent. This monograph4 documents the lengthening of life span by using sub-harmful doses of radiation. Thus, studies on the life span of mammals at low doses and dose rates may yield a variety of findings; our study examines the shortening of life as a function of these radiation variables."

"Life shortening, whether it stems from a specific malady such as cancer or an irreparable injury to many life-sustaining cell, tissue, and organ systems, has been generally accepted as one biological endpoint of exposure to ionizing radiation. Because no human or experimental animal data are sufficiently elaborate to quantitate the adverse effects from a very low dose or dose-rate exposure, we generally accept that any life-shortening effects from a large dose and/or dose-rate exposure could be extrapolated to represent the effects that could not be measured experimentally. Research data using life span as the measured endpoint have been compiled at different laboratories by different reearchers using several strains or species under a variety of environmental conditions. Our comprehensive study was designed to eliminate as many variables as feasible, to provide reliable life span data from which reasonable predictions and extrapola-tions could be made. The original study called for extensive histopathological examination of all tissues, but financial limitations precluded this most important end-point."

"II. MATERIALS AND METHODS
"Mice used in this study were strain C57BL/6J from the Jackson Laboratory at Bar Harbor, Maine. Over 10 years ago that laboratory reported the mean age at death of this strain was 693 + 16 days for females and 670 + 20 days for males (Yuhas 1969). Gross pathology showed the cause of death to be nonspecific, without predominant pattern. Thus, this strain was considered genetically stable."

"Control mice were not included in the experiment until the age at exposure had been obtained. For example, when a 6-month-old exposure regime began, the corresponding control mice were randomly selected from the shipment received for this group. Control mice were placed in our general housing conditions, whereas the experimental mice receiving chronic irradiation were placed in one of the two 60Co source rooms. After the irradiated animals received their prescribed doses, they were housed alongside the control mice until death. Two dummy exposure rooms creating exact housing conditions for true control were not feasible, but we might have considered the lowest total dose and dose-rate animal groups as the control groups for comparing life spans and tumor incidence."

"A basic 5 x 6 x 4 factorial design was used (Table I). Five doses (20, 60, 180, 540, and 1620 rad) were given at six exposure rates (0.7, 2.1,6.3, 18.9,56.7, and 36000 R/day) to four age groups (newborn, 2 month, 6 month, and 15 month). Doses and dose rates were both increased by multiples of 3 to be easily interpretable logarithmically. Approximately 25 mice were used per cell-point and, except for the newborn age group, the 180-rad dose group at 18.9 R/day with 2-month-old mice, and the 180-rad dose group at 0.7 R/day with 15-month-old mice, the experiment was replicated once. To minimize a possible seasonal environment effect, all dose-rate/age combination exposures were started on the same day for each replication"

"We recorded the findings of gross necropsies performed on all animals at death. All grossly appearing abnormal tissues were examined histopathologically, but the scope and cost of complete microscopic analyses were prohibitive. Our findings confirm those of the Jackson Laboratory. The cause of death in this strain (C57BL/6J) is rather nonspecific; no predominant diseases accompany death."

"III. RESULTS
"We designed our study to produce data feasible for curve fitting of dose, dose rate, and age at exposure. We anticipated, from earlier radiation effects studies with rodents, [17 refs] that life shortening resulted from whole-body exposure to ionizing radiation and, to some degree, was dose, dose-rate, and age dependent. Our study suggests that although the design assumptions may have been prudent, they were at best unfounded over the wide range of variables. In comparing life span data, the random variable X represents the lifetime of mice given a Table II(A-F) shows median life spans and Fig. l(a)-( v) [Ed. Note: Figs not shown] shows survival plots for various total doses given at fixed exposure rates. Table IIA and Fig. l(a)-(d) [Ed. Note: Figs not shown]reveal the pattern evident throughout most of this experiment".

"Most of the irradiated animals lived longer or no differently than did the non-irradiated controls; however, in several cases differences were significant. For newborn mice exposed to 180 rad at 0.7 R/day, the life span was significantly longer than it was for controls. At all dose levels the 2-month age group lived significantly longer than did the median controls. Although there were no differences among 6-month-old mice, the 15-month group with the 20-rad dose lived significantly longer than did their controls."

"The 2-month age group had the only mice with significantly altered life span at a dose rate of 2.1 R/day [Table IIB and Fig. l(e)-(g)]. [Ed. Note: Figs not shown] For the newborns exposed at 6.3 R/day [Table IIC and Fig. l(h)], the survival function for the controls was below that of every exposed group; therefore, a greater proportion of the exposed animals survived for any given length of time. For the 2- month group, each dose group is entirely different from the control curve [Fig. l(i)]. The 20-rad dose for the 2- month group shows strikingly different results from all others. Many deaths between 320 and 400 days in this group were probably due to a competing cause of death, and perhaps these deaths should have been censored (not counted as suitable for determining average life span). For the 6-month group, the controls differed significantly only from the 20-rad dose group, and the same is true for the 15-month age group. With a dose rate of 18.9 R/day [Table IID and Fig. 1(1)], the survival curve for the controls lies beneath that of every treated group of 2- month-old mice, with the exception of some points at the 1620-rad dose. For the 6- and 15-month age groups, there were no significant differences between control and irradiated groups [TableIID and Fig. 1(m) and (n)]."

"IV. DISCUSSION
"Storer et al. (1979) present life-shortening data for strain RFM mice (female and male) given external whole-body gamma-ray exposures of doses ranging from 0 to 400 rad at 45 R/minute. Although female mice given 10 rad showed no significant life shortening, with doses of 25 rad and above life-shortening effects were observed. Male mice showed significant life shortening only when given doses of 50 rad or more, and mean life span was lengthened at a total dose of 10 rad. At a lower rate (8.3 R/day) life shortening was reported at 50 rad and greater doses, but the life-shortening slope was below that of the higher rate. These data are not inconsistent with data from our earlier study with this same strain of mice in which we gave 12 relatively high dose/dose-rate combinations. (Spalding 1967) The results of Storer et al. (Storer 1979) also are not inconsistent with the data obtained from the present study when we consider the mouse strain, dose, exposure-rate, exposure-age, and condition differences between experiments. Some irradiation studies of human populations may help us evaluate our own study. Research in life shortening from nonspecific causes among 82000 Japanese A-Bomb survivors is ongoing. Any life shortening that has been observed is attributed to carcinogenic effects."

"The relatively few benign and malignant tumors that we observed in our study did not appear to be influenced by dose, exposure rate, or age at exposure. When the number of benign and malignant tumors was expressed as a percentage of the number of mice observed, control mice had fewer than 0.03 benign and 0.05 malignant tumors; treated mice, fewer than 0.01 benign and 0.02 malignant tumors. For each age group, we determined the mean age at death for control and treated mice with grossly obvious tumors, and we found no significant influence on the median life spans reported in the tables."

"The results of our study are in general agreement with the results of other researchers in human and animal life span irradiation studies. The largest discrepancy occurs in the abundant lengthening of life spans we saw in younger animals at most dose rates, not allowing an extrapolation to zero dose. Storer et al. (1979) used a linear extrapolation for the dose/life-shortening relationships in RFM male mice with doses ranging from at least 0 to 50 rad. This linear relationship at low doses and dose rates was also reported by Sacher and Grahn (1964). The linear-quadratic relationship between life shortening and dose obviously will fit these circumstances because the squared term does not dominate until higher doses are reached. Our data obtained over widely ranging dose exposure-rate, and exposure-age conditons fail to consistently support any mathematical function that may predict radiation-induced life shortening from radiation exposures approaching backgrourid levels. In fact, our data suggest beneficial effects from low dose and low-dose- rate gamma-ray exposure. Radiation-induced hormesis invertebrates appeared in the literature in the early 1900s.4 Low-1evel external irradiation was shown to increase resistance to various infections and to favorably alter physiological function. Luckey’s monograph’ tabulates specific instances and three recent publications (Luckey 1981, Hickey 1981, Hickey et al 1982) discuss low-level ionizing radiation as a stimulating agent. Luckey reports’ the results of Lorenz et al. (1954) in which g-dose rates of 0.11, 1.1, and 2.2 R/day resulted in some instances in increased life spans over the controls (mice, guinea pigs, and rabbits)."

"The BEIR report (1980) deals with life shortening in terms of cancer induction, and a recent NCRP Publication (1980) states that 'studies on the effects on life span contribute appreciably to quantitative knowledge on the influence of dose-rate on tumorigenesis.' This reasonable approach recognizes the uncertainties in life shortening at near-background levels. Because the research data are insufficient to statistically separate the life spans of experimental subjects from those of controls, to use cancer risk in terms of prorated individual exposures (man-rems) is logical. This is appealing because it enables establishment of exposure guidelines based upon cancer risk without using terms of life shortening. In conclusion, our data and earlier studies suggest that life shortening from exposure to ionizing radiation is not a feasible endpoint on which to base any harmful biological effects that may result at very low doses and dose rates. The lengthening of some of the mean life spans observed in this study indicates the possible existence of a threshold dose for the onset of measurable life span shortening. Cancer incidence, however, may not be directly related to population life span at these dosage levels. From the many total dose, dose-rate, and exposure-age combinations that compose our data, we conclude that life spans among irradiated groups were not shortened compared with life spans of corresponding control mice. We suggest that these dosage regions be considered important in selecting external radiation parameters that avoid obvious deleterious effects, including life-shortening cancer."