Biokinetic model of americium in the beagle dog. Salt Lake City, UT (U.S.A.)

Transcription

1 Biokinetic model of americium in the beagle dog A. Luciani 1, E. Polig 2, R. D. Lloyd 3, S. C. Miller 3 1 ENEA Radiation Protection Institute via dei Colli, Bologna, (Italy) 2 Forschungszentrum Karlsruhe HS/ÜM Postfach Karlsruhe (Germany) 3 University of Utah Radiobiology Division, Dept. Radiology 2334 CAMT, 729 Arapeen Drive, Salt Lake City, UT (U.S.A.) Abstract. A biokinetic model of the systemic distribution of americium in the beagle dog is presented. The model is based on a previous biokinetic model of plutonium. The data sets used for the development of the model are the measurements of excreted activity (urine and feces) and organ burdens (skeleton, liver and other soft tissues). In developing the model, the part of the plutonium model representing the skeleton, which was based on histomorphometric and autoradiographic investigations, has not been modified with regard to both its structure (compartments for the trabecular/cortical volume, surface and marrow) and the remodelling rates. Other skeletal parameters such as the transfer rate from marrow to blood and the partitioning of transferred activity from the blood to trabecular/cortical surface and volume were optimized to describe the element-specific biokinetics in the skeleton. The model well describes the fractions of americium in the skeleton, liver and soft tissues and the total fraction excreted in urine and feces. Particularly it demonstrates the possibility of describing the behavior of americium in the skeleton with a model substantially analogous to the model of plutonium in humans. However it differs from the ICRP model of the skeleton with regard to the fundamental structure and the predictive power. This study will be the starting point for a future improvement of the currently used americium model for humans, particularly for the skeleton. 1. Introduction In the past decades a number of experiments were carried out to investigate the metabolism of plutonium and americium in beagle dogs. These experiments were designed to study the long term effects to be expected from internal burdens of these nuclides and to evaluate the possible risks to humans. The experimental data on plutonium metabolism were analysed previously to develop a biokinetic model of the systemic distribution and dosimetry of 239 Pu in the beagle dog [1]. The compartmental structure of the biokinetic model was based on the simple structure previously used to demonstrate the possibility of modelling the microdistribution of 239 Pu in the beagle skeleton [2]. The model parameters were determined from retention and excretion measurements by optimization procedures, whereas the portion representing the skeleton was defined on the basis of the histomorphometric and autoradiographic investigations. Up to now the results of the corresponding americium experiments have not yet been analysed for developing a biokinetic model, but only empirical retention functions in the main deposition organs were calculated for different injection levels [3]. By keeping the structure of the model for plutonium, the main purpose of the present work is the development of a compartmental model able to describe the available data for americium retention and excretion. 2. Materials and methods 2.1. Experimental data Since 195 s several experimental investigations on the metabolism of americium in beagle dogs were performed. Young adult beagles were given a single intravenous injection of 241 Am (III) citrate ranging from.67 to 167 kβq kg -1 body mass [3, 4, 5]. At the time of introduction into the experiment, the dogs were on the average about 18 months old, an age at which skeletal maturity is reached. For the purpose of the present work, only the data from dogs injected with the lowest levels (below 3.7 kbq kg -1, i.e..1 µci kg -1 ) were considered, in order to exclude any acute toxic effect (damage of tissue cells and interference with bone remodelling) that could significantly disturb the normal physiology of an organ. 1

2 In vivo measurements, based on total and partial body counting, allowed the estimation of the retention in liver and non-liver tissues (mainly skeleton). For several dogs that were subject to autopsy, the distribution of 241 Am in individual bones, liver and other soft tissues are also available. It was found that the ratio of the skeletal 241 Am to total non-liver 241 Am averaged.885 with no significant dependence upon either injection level or time after injection [3]. Therefore in vivo measurements of non-liver tissues, multiplied by.885, were used to evaluate retention values in skeleton. This provided additional data for the skeletal retention other than the values obtained by autopsy analyses. Analogously, in the present work, a similar approach was used to extend the retention data for soft tissue other than liver, for which only a limited number of measurements is available. Thus, in vivo measurements of non-liver tissues were multiplied by.115 in order to obtain retention values for the other soft tissues. This additional source of data will enhance the accuracy in model fitting. Measurements of the cumulative urine and fecal excretion of americium are available only for four beagle dogs and during a short time after injection (up to 21 days). These dogs were given injections significantly above the quoted limit for low levels of 3.7 kbq kg -1 (about 14 kbq kg -1 for two dogs that died because of radiation effects, and 33.3 kbq kg -1 for the other two dogs) Biokinetic model As a starting point, the structure of the biokinetic model for plutonium metabolism in the beagle dog was adopted (Fig. 1). It closely resembles the human model for actinides, as recommended in Publication 67 [6] of the International Commission on Radiological Protection (ICRP). The ICRP skeletal model was modified to achieve a better agreement with the physiology of bone remodelling and autoradiographic analyses [7]. As in ICRP model the liver and the remaining soft tissues are considered separately. The three compartments of the ICRP model for the remaining soft tissues were reduced to a single compartment. Owing to the lack of excretion data for low injection levels and the limited number of data even for the high level cases, no attempt was made to model accurately the excretion pathways for americium. The ICRP compartment kidney was lumped into the soft tissue compartment. The remaining excretion compartments of the ICRP model have no feedback to the main system and, therefore, can be dropped without affecting the computations for the remaining compartments. Trab. marrow r m F st F vt Trab. surface Trab. volume f rt f ft r l f l Liver r m Cort. marrow F sc F vc Cort. surface Cort. volume f rc f fc Blood r s f s Other soft tissues f e FIG. 1. Compartmental model of the systemic distribution of 241 Am in the beagle dog; f-parameters are partitioning factors, r(f)-parameters are transfer rates. 2

3 All transfer rates from the blood to other compartments were determined as the product of the general blood clearance (r) times the fraction (f) of the clearance to the respective compartment. The fraction for the whole skeleton, f skel, is partitioned into the four pathways to the respective surface and volume compartments of trabecular and cortical bone (f skel = f rt + f ft + f rc + f fc ). Once f skel is given, the four partitioning factors are calculated using the expressions available in the literature [7], depending on the affinity ratio of resting to forming bone surfaces (q rf ) and cortical to trabecular bone surfaces (q ct ), the remodelling rates of trabecular and cortical bones (F vt, F vc ), the mean wall thickness (MWT) of the trabecular bone units, the osteonal radius (R o ) and the Haversian canal radius (R h ), the surface to volume ratios (S vt, S vc ) and the trabecular and cortical surface (S t, S c ). With exception of the affinity ratios, all these parameters depend on the morphometry and physiology of the beagle skeleton and, therefore, are not related to specific radionuclide characteristics. On the other hand the affinity ratios are typical for the radionuclide considered. For americium q rf = 1. was chosen because no experimental values are available. Parameter q ct will be determined by model optimization. Parameters values and derived quantities are listed in Table I and have been presented previously [1]. Table I. Histomorphometric parameters of the skeleton. Parameter Symbol Value Volume turnover trabecular bone (% y -1 ) F vt 1 Volume turnover cortical bone (% y -1 ) F vc 5 Mean wall thickness (µm) MWT 4 Osteonal radius (µm) R o 75 Haversian canal radius (µm) R h 15 Trabecular surface to volume ratio (cm -1 ) S vt 288 Cortical surface to volume ratio (cm -1 ) S vc 54 Trabecular surface (cm 2 ) S t, 15,56 Cortical surface (cm 2 ) S c 13,44 Affinity ratio resting/forming q rf Optimization procedure The model calculations and optimization procedures were performed using a commercially available software 1 package. The partitioning factors f skel, f l, f s for the skeleton, liver and soft tissue compartments, respectively, and the clearance rates r, r m, r l and r s from blood, bone marrow, liver and soft tissue are determined by minimizing a suitable objective function, which is a measure of the deviation between observations and model predictions. The total excreted fraction of americium in urine and feces is derived from the other partitioning factors (f e = 1 f skel f l f s ). For the optimization all the measurements are equally weighted. 3. Results The optimum parameter values were determined by fitting the model predictions to the 241 Am retention in the skeleton, liver, non-liver soft tissues (other soft tissues) and the total cumulative excretion. A few blood concentration values were available up to about 25 hrs after injection. But because of the short interval of observation they were disregarded for the optimization. The results of the optimization procedure are given in Table II. The fitted values (second column) are listed together with the uncertainty as percentage standard deviation (third column). For derived parameters (f rt, f ft, f rc, f fc and f e ), calculated as functions of other fitted parameters, the uncertainties are estimated by using the standard formula for error propagation. 1 SAAM User Guide. SAAM Institute, Inc. 453 Union Bay Place, NE, Seattle, WA 9815, USA. 3

4 Generally the uncertainties amount to a few percent and even in the worst cases, as for the blood clearance (r), the affinity ratio of cortical to trabecular bone surfaces (q ct ), the clearance rate of bone marrow (r m ), and the excretion fraction (f e ), are below 2 %. The blood clearance rate for americium is greater by a factor of 1.7 than for plutonium (1.57 d -1 for americium and.9 d -1 for plutonium). The calculated blood retention decreases more slowly than the few measured blood concentrations [8]. This would suggest an even a higher blood clearance rate. As for plutonium, the liver and the skeleton are the main deposition organs. The total fraction of the clearance to the skeleton and the liver (f skel + f l ) is.85, very close to the value of.81 for plutonium [1], but the partitioning of americium between these organs (.35 for skeleton and.5 for liver) is reversed with respect to the partitioning of plutonium (.51 for skeleton and.3 for liver). The deposition of americium in the liver is generally higher than for plutonium. This is because the partitioning of americium in blood favors the liver, as compared to plutonium. The clearance rates (r l ) of the two nuclides are nearly identical (Am:.135d -1 ; Pu:.114d -1 ). A comparison with the ICRP biokinetic models for humans should be done with care, as they are based on a model for the skeleton [9] which differs with regard to structure and parameter values from the model for beagle dogs adopted here. However it could be worth pointing out that, as for beagles, the ICRP models for humans have the same total fraction f skel + f l for americium and plutonium (.56), and the partitioning of americium among the skeleton and liver (.21 and.35, respectively) is exactly reversed with respect to plutonium [6]. For the other soft tissues the clearance rate is not significantly different to the clearance from liver, even considering the small uncertainties associated to their values (Table II). This allows simplifying the model by combining the liver and the other soft tissue compartments, at least in case of very low intake levels, for which toxic effects with damage or alteration of hepatic tissues cannot be expected. The clearance rate of americium from bone marrow (r m ) is about 5% larger than for plutonium. The affinity of americium to cortical bone surfaces is nearly three times lower than for trabecular surfaces. For plutonium the cortical affinity was assumed to be 1 times lower than the trabecular affinity. However q ct, as it was defined in a previous publication [7] for both types of surfaces, is not independent on q rf. One may calculate an affinity ratio cortical/trabecular for resting surfaces only. Because q rf = 1 was assumed, the ratio for americium does not change but for plutonium it would increase to.137, i.e. the affinity is about seven times higher for trabecular than for cortical bone. This means that nuclide-specific differences remain for the ratios of americium and plutonium, even after this correction. Also the relatively lower affinity to trabecular surfaces and the lower affinity to forming surfaces combine to reduce the total fraction to trabecular bone (.26) compared to plutonium (.47). The excreted fraction f e =.1 is practically equal to the value for plutonium. The biokinetic model for americium in beagles presented here provides a realistic description of the experimental findings. The measured retention of americium in skeleton, liver and the other soft tissues and the total cumulative excretion are shown in Fig. 2 together with the model calculations. The calculations match the experimental data closely for each set of measurements over the whole interval of observation. 4

5 Table II. Biokinetic parameters. Parameter Symbol a Value Stand. Dev. (%) Source Clearance from blood r Fit Trab. surface to trab. marrow r st.362 Ref. (1) Trab. volume to trab. marrow r vt.274 Ref. (1) Cort. surface to cort. marrow r sc.14 Ref. (1) Cort. volume to cort. marrow r vc.137 Ref. (1) Trab./cort. marrow to blood r m Fit Liver to blood r l Fit Soft tissue to blood r s Fit Fraction to skeleton f skel Fit Fraction to liver f l.5 2. Fit Fraction to soft tissue f s Fit Fraction to excretion f e.1 14 calculated b Fraction to trab. surface f rt calculated b Fraction to trab. volume f ft calculated b Fraction to cort. surface f rc calculated b Fraction to cort. volume f fc calculated b Affinity ratio cort./trab. bone q ct Fit a Rate constants are designated by letter r and have dimension d -1. b f e calculated as 1 f skel f l f s. The fractions to bone volume and surface from expressions in [7]. The uncertainty is estimated as quadratic combination of the single parameters uncertainties.,6.6,4.4,2.2 Skeleton,6.6,4.4,2.2 Liver ,1.1,2.2,5.5 Other soft tissue,15.15,1.1,5.5 Total cumulative excretion FIG. 2. Fractional retention in the skeleton, liver and other soft tissues and fractional total cumulative excretion of 241 Am. Circles: Measurements; Solid lines: Model calculations. 5

6 4. Conclusions It has been demonstrated that this model of the biokinetics for 241 Am in beagle dogs is in agreement with measured retention data of a number of experimental studies. Significant analogies between the biokinetic models for americium and plutonium in beagles and humans (ICRP Publication 67) were discussed both with regard to the total deposition and the partitioning between the main deposition organs skeleton and liver. The set of measurements used in the fitting procedures was limited to the injection cases below to 3.7 kbq kg -1, (.1 µci kg -1 ). In previous investigations [3] it was pointed out that a more pronounced decrease of the retention in liver occurs at higher injection levels, presumably due to release from damaged cells. Correspondingly an increase of americium in the skeleton was observed, as some of the americium lost from the liver redeposits in the skeleton. In terms of compartmental modelling this could be taken into account by making the liver clearance dependent on the injection level. Acknowledgements- This work was supported by ENEA/Italy, Forschungszentrum Karlsruhe/Germany and NCI grant R1 CA66759/U.S.A. References 1) Polig, E., Bruenger, F. W., Lloyd, R. D. and Miller, S. C., Biokinetic and dosimetric model of plutonium in the dog. Health Phys. 78(2): , (2). 2) Polig, E., Kinetic model of the distribution of 239Pu in the beagle skeleton. Health Phys. 57(3): , (1989). 3) Lloyd, R. D., Mays, C. W., Jones, C. W., Atherton, D. R., Bruenger, F. W., Shabestari, L. R. and Wrenn, M. E., Retention and dosimetry of injected 241 Am in beagles. Radiat. Res. 1: , (1984). 4) Lloyd, R. D., Mays, C. W., Taylor, G. N. and Atherton, D. R., Americium-241 retention in beagles. Health Phys. 18: , (197). 5) Lloyd, R. D., Mays, C. W., Taylor, G. N. and Atherton, D. R., Americium-241 retention in beagles. In Report COO , Retention and dosimetry of some injected radionuclides in beagles, Lloyd, R. D. (Ed.), Radiobiology Division of the department of Anatomy, University of Utah - College of Medicine, (197). 6) International Commission on Radiological Protection, Age-dependent doses to members of the public from intake of radionuclides: Part 2 ingestion dose coefficients. Publication 67. Annals of ICRP, 23, No. 3/4, Pergamon Press, Oxford and New York, (1994). 7) Polig, E., Labels of surface seeking radionuclides in the human skeleton. Health Phys. 72(1): 19-33, (2). 8) Bruenger, F.W., Stevens, W. and Stover, B. J., 241 Am in the blood; in vivo and in vitro observations. Research in Radiobiology, Report COO : (1968). 9) International Commission on Radiological Protection, Age-dependent doses to members of the public from intake of radionuclides: Part 1. Publication 56. Annals of ICRP, 2, No. 2, Pergamon Press, Oxford and New York, ICRP, (1989). 6