Radiol Oncol 1997; 31; 279-85. Radioprotcction of salivary glands by amifostine in high-dose radioiodine therapy investigated in a new rabbit animal model Ralf-Harto Hühner,1 Karl H. Bohuslavizki,1 Winfried Brenner,1 Susanne Klutmann, Bernd Feyerabend,3 Jutta Lüttges,3 Stephan Tinnemeyer,1 Janos M ester,2 Malte Clausen,2 Eberhard Henze' Departments of 'Nuclear Medicine and Pathology, Christian-Albrechts-University of Kiel, Germany 2Depcirtment of Nuclear Medicine, University Hospital Eppendorf, Germany Salivary gland damage following high-dose radioiodine treatment (HD-RIT) is a well known side effect. Since differentiated thyroid cancer (DTC) has a very good prognosis, the reduction of long-term side effect is of major interest. Therefore, the radioprotective effect of amifostine was investigated in a rabbit animal model. Quantitative salivary gland scintigraphy was performed on 5 rabbits prior to and up to 3 months after HD-RIT applying I GBq 1-131. The uptake of Tc-99m-pertechnetate was calculated as a measure of parenchymal function. Three animals received 200 nig/kg amifostine prior to HD-RIT, and two seived as controls. Salivary glands were examined hislopathologically. In two control rabbits HD-RIT significantly (p<0.001) reduced pertechnetate uptake by 63 % and 46% in parotid and submandibular glands, respectively, and lipomatosis was found hislopathologically. In contrast, in three rabbits treated with amifostine parenchymal function was not decreased significantly (p = 0.953), and lipomatosis was negligible. In conclusion, salivary gland impairment induced by HD-RIT can be evaluated quantitatively by salivary gland scintigraphy in rabbits, and amifostine significantly reduced salivary gland damage induced by HD-RIT. These encouraging results need further evaluation in patients since it may help to increase the quality of life of patients with diffentiated thyroid cancer. Key words: Salivary glands - radiation effects; radiation-protective agents; - amifostine; radiotherapy-adverse effects; rabbits Introduction A standard therapy in differentiated thyroid cancer requires a total thyroidectomy and a high-dose radioiodine therapy in order to completely ablate thyroid remnants,' Apart from thyroid tissue the fi-emitting iodine isotope 1-131 used for radioiodine therapy is accumulated actively by an ATP depend- Correspondence to: Dr, Karl H, Buhoslavizki, Department of Nuclear Medicine, Christian-Albrechts-University of Kiel, Arnold-Heller-Str, 9, D-24105 Kiel, Germany, Phone: +49 4 31 5 97 - 30 76, Fax,: +49 4 31 5 97 - 30 65, UDC: 616.316-001,28-084 ent Na+/K+/2Cl~-cotransport due to its similar atomic diameter and its comparable electric charge." This causes an undesired accumulation of 1-131 in parietal cells of the stomach as well as in acinar cells of salivary glands.6"9 Consequently, well recognized side effects of high-dose radioiodine therapy are transient gastritis and long-lasting xerostomia.111"16 Therefore, a radioiodine therapy is performed under salivary gland stimulation in order to decrease the impairment of salivary gland function.17"2' However, even under salivary gland stimulating conditions, a parenchymal damage could be shown after high-dose radiodiodine therapy using quantitative salivary gland scintigraphy.13-24-27 Since 280 Hübner R-H ei ai differentiated thyroid cancer has very good prognosis, reduction of long-term side effects following high-dose radioiodine therapy is important for the patients' quality of life.1 In the last few years various reports dealt with radioprotective effects of amifostine,28-53 a phos-phorylated aminothiol chemically described as S-2-[3-aminopropylamino]-ethylphosphorotioic acid (Figure 1). Since amifostine accumulates markedly in salivary glands,94 it has been used successfully in external radiotherapy in patients with head and neck tumors in order to prevent xerostomia.'5"4" H2N-CH2-CH2-CHJ-NH-CH2-CH2-S-PO(OH)2 ^ [Alkaline Phosphatase] H2N-CH2-CH2-CH,-NH-CH2-CH2-SH Figure 1. Chemical structure of amifostin (above) and its active metabolite WR-1065 (below). Therefore, it looked worthwhile to transfer the radioprotection of salivary glands by amifostine to high-dose radioiodine therapy in order to prevent patients from xerostomia, and, thus, to increase the tolerance of high-dose radioiodine therapy. As a first step we established a rabbit animal model and report on first results. Materials and methods In order to investigate the cytoprotective effect of amifostine an animal model was established. Five male New Zealand white rabbits aged three months, weighing 2.5±0.1 kg, were treated with 1 GBq 1-131 intravenously in order to ablate the thyroid and to destruct salivary gland parenchyma. Prior to the application of radioiodine all animals received 4 nig Dexamethason (Fortecortin®, Merck, Darmstadt) and 0.5 mg Tropisetron (Navoban®, Sandoz, Nürnberg) as antiemetic treatment. In addition, three out of five rabbits received 200 mg/kg amifostine (Ethyol®, Essex, München), and two rabbits served as controls, receiving physiological saline solution. To quantify parenchymal function, salivary gland scintigraphy was performed prior to as well as four weeks, eight weeks and twelve weeks after the application of 1-131. Rabbits were put in prone position directly onto a low energy high resolution collimator of a large field - of - view gamma camera (Bodys-can, Siemens, Erlangen). After injection of 100-140 MBq Tc-99m-pertechnetate sequential images of one minute each were acquired up to 25 minutes. Images were stored digitally in a 256 x 256 matrix. For quantification one rectangular background ROI was positioned caudally to the left parotid gland, and five oval ROIs were drawn over both parotid and subma-nidbular glands and the thyroid gland, respectively. ROIs were copied from the study performed prior to radioiodine treatment to the studies obtained after radioiodine treatment. As a measure for parenchymal function the uptake of Tc-99m-pertechnetate was calculated in percent of the injected activity. For compensation of noise and, thus for stabilisation of data, uptake was averaged from 21.-23. minute post injection. Whole body distribution of Tc-99m-pertechnetate in a rabbit is shown in Figure 2A and ROIs used for quantification are depicted in Figure 2B. Twelve weeks after radioiodine therapy all animals were sacrificed to remove salivary glands for histopathological examination. Salivary glands were stained with Hematoxilin/Eosin in conventional manner. Animal studies were approved by the local government (XI 330a 72241.11-17). Data are given as mean ± one standard deviation. Two-tailed U-test according to Wilcoxon, Mann and Whitney was used to evaluate statistical differences between animal subsets.41 For p<0.05 data were considered to be statistically significant. Results Controls Details of Tc-99m-pertechnetate uptake in salivary glands of controls and amifostine rabbits are given in Table 1. Salivary gland scintigrams of a control rabbit are given in Figure 3 (upper row). In controls thyroid uptake declined to almost zero as early as four weeks after radioiodine treatment, thus documenting a thyroid ablative dose of radioiodine. In parallel, parenchymal function of salivary glands decreased. Twelve weeks after the injection of I-131 Tc-99m-pertechnetate uptake was reduced by 63 % and 46 % in parotid and submandibular glands, respectively, (Figure 4, open symbols). Amifostine group Rabbits treated with amifostine exhibited complete ablation of the thyroid four weeks after the application of 1-131 as well. This is shown in Figure 3 (lower row). In contrast, in these animals parenchy- Radioprotection of salivary glands 281 Tabic 1. Uptake of Tc-99m-pertechnetate in percent of injected activity prior to, 4, 8, and 12 weeks after the application of 1 GBq Iod-131 in control rabbits and in rabbits treated with amifostine 200 ing/kg body weight. Numbers represent mean of right and left parotid and submandibular glands, respectively. Controls Amifostine Parotid glands Submandibular glands Parotid glands Submandibular glands prior to 1-131 0.226 ± 0.042 0.295 ± 0.070 0.241 ± 0.030 0.230 ± 0.074 4 weeks after 0.140 ± 0.018 0.199 ±0.046 0.215 ±0.038 0.215 ±0.060 8 weeks after 0.106 ±0.019 0.187 ±0.067 0.209 ± 0.032 0.210 ±0.065 12 weeks after 0.080 ± 0.011 0.154 ±0.057 0.208 ± 0.023 0.212 ±0.057 ii 0,347'; i 0.362% 0.343 ' < 0.338 % Figure 2. Whole body distribution of Tc-99m-pertechnetate (A) and the magnification of the head (B) visualizing the ROIs used for quantification. Numbers represent uptake of Tc-99m-pertechnetate in percent of the injected activity in parotid, submandibular glands, and thyroid gland, respectively. „100 £ !80 Cu 3 "g 60 N "c3 1 40-j Z 20- 0 Prior to 1-131 4wk. 8 wk. p < 0.05 p<0.01 p< 0.001 12 wk. Figure 4. Normalized uptake of Tc-99m-pertechnetate in parotid (circles) and submandibular (squares) glands of control rabbits (open symbols) and of trabbits treated with amifostine (filled symbols) prior to, 4, 8 and 12 weeks after application of 1 GBq 1-131. 0.229 % 0.209 % 0.XS8 % 0.176% ,). 130 % .00S % 0.147% ,0.120% 0.001 % ■v ... o.' mm 0.114% 0.080 % 0.001 % 0.309 % 0.246% 0.779 % 0.277' 0.215 1 0.009' V C 0.279 % 0.219% 0.002 % 0.280' 0.219 1 .002 % Figure 3. Salivary gland scintigraphy in the control group (upper row) and in the amifostine group (lower row) prior to (A), 4 (B), 8 (C), and 12 weeks (D) after the application of 1 GBq 1-131. Numbers represent uptake of Tc-99m-pertechnetate in percent of the injected activity in parotid, submandibular glands, and thyroid gland, respectively. 282 Hühner R-H et al. r % ¡»SUf. t^jp '