Breast cancer imaging with radioiodinated oestradiol

Use of Radionuclide-Based Imaging Methods in Breast Cancer

Breast cancer is one of the most commonly occurring cancers in women globally and is the primary cause of cancer mortality in females. Thus, early and effective breast cancer diagnosis is crucial for enhancing the survival rate. Current standard diagnostic techniques to assess the hormone receptor status in biopsies include immunohistochemistry and fluorescence in situ hybridization. However, in recent years, there has been an increase in research on noninvasive techniques for molecular imaging of hormone receptors. These methods offer many advantages over conventional imaging, as repeated measurements can be used to capture heterogeneous tumor expression throughout the body, as well as transformations in receptor status during disease progression. Thus, the noninvasive method, as an adjunct to conventional imaging, offers the potential to improve patient selection, optimize dose and schedule, and streamline the assessment of response.

Receptor Imaging in Oncology by Means of Nuclear Medicine: Current Status

To date, our understanding of the role of receptors and their cognate ligands in cancer is being successfully translated into the design and development of an arsenal of new, less toxic, and more specific anticancer drugs. Because most of these novel drugs are cytostatic, objective response as measured by morphologic imaging modalities (eg, computed tomography or magnetic resonance imaging) cannot be used as a surrogate marker for drug development or for clinical decision making. Positron emission tomography (PET) can be used to image and quantify the in vivo distribution of positron-emitting radioisotopes such as oxygen-15, carbon-11, and fluorine-18 that can be substituted or added into biologically relevant and specific receptor radioligands. Similarly, single-photon emission computed tomography (SPECT) can be used to image and quantify the in vivo distribution of receptor targeting compounds labeled with indium-111, technetium-99m, and iodine-123. By virtue of their whole-body imaging capacity and the absence of errors of sampling and tissue manipulation as well as preparation, both techniques have the potential to address locoregional receptor status noninvasively and repetitively. This article reviews available data on the in vivo evaluation of receptor systems by means of PET or SPECT for identifying and monitoring patients with sufficient receptor overexpression for tailored therapeutic interventions, and also for depicting tumor tissue and determining the currently largely unknown heterogeneity in receptor expression among different tumor lesions within and between patients.

Estrogen receptor status in primary breast cancer: iodine 123-labeled cis-11beta-methoxy-17alpha-iodovinyl estradiol scintigraphy

PURPOSE

To determine the sensitivity of iodine 123 ((123)I)-labeled cis-11beta-methoxy-17alpha-iodovinyl estradiol (Z-MIVE) scintigraphy for the detection of estrogen receptors in patients with primary breast carcinoma.

MATERIALS AND METHODS

In 22 patients, estrogen receptor status was assessed with planar scintigraphy and single photon emission computed tomography (SPECT) 4 hours after the injection of 185 MBq (123)I-labeled Z-MIVE. For histologic and estrogen receptor immunohistochemical analysis, breast carcinoma tissue was obtained in all patients by means of biopsy or resection of the primary tumor. Two experienced physicians semiquantitatively scored the scintigraphic and immunohistochemical findings. The uptake ratio at scintigraphy and the immunohistologic staining intensity were scored as negative, weak, intermediate, or strong.

RESULTS

All patients had histologically proven breast cancer. Immunohistologic staining for estrogen receptors yielded negative findings in four patients and positive findings in 18 (weak staining, n = 2; intermediate staining, n = 6; strong staining, n = 10). Planar (123)I-labeled Z-MIVE scintigraphic findings were negative in five patients and positive in 17 (weak uptake, n = 2; intermediate uptake, n = 10; strong uptake, n = 5), resulting in one false-negative finding. Findings at (123)I-labeled Z-MIVE SPECT were negative in four patients and positive in 18. The sensitivities of (123)I-labeled Z-MIVE scintigraphy for estrogen receptors were 100% with SPECT and 94% with planar scintigraphy. The correlation between immunohistologic and planar scintigraphic scores of estrogen receptor status was 0.72 (P <.01).

CONCLUSION

(123)I-labeled Z-MIVE scintigraphy is a sensitive noninvasive tool for the detection of estrogen receptors in patients with breast cancer.

(17alpha,20E/Z)-iodovinyl- and 16alpha-iodP618-homoestradiol derivatives: synthesis and evaluation for estrogen receptor imaging

Three new 125I-radioiodinated estrogens featuring a 13beta-ethyl instead of the natural 13beta-methyl group, i.e. 18-homoestradiols, were synthesized and evaluated as potential estrogen receptor imaging agents. The 16alpha-iodo-18-methylestradiol and the 125I-labeled analog were synthesized from the corresponding 16beta-bromo analog by the halogen-exchange method. The cis-bromohydrin precursor was obtained by bromination of an estrone enolacetate, followed by epimerization and reduction. The isomeric (17alpha,20E/Z)-iodovinyl-18-methylestradiols were prepared via the vinyltin intermediates. Treatment of 18-methyl-17alpha-ethynylestradiol with tri-n-butyltin hydride, in the presence of azobisisobutyronitrile as catalyst and heating at 90-100 degrees C afforded the (17alpha,20E)-tri-n-butylstannyl isomer as the major product. Changing the catalyst for triethyl borane, at room temperature, mainly gave the 20Z-isomer. The nca 125I-labeled analogs were obtained from their corresponding tin intermediates upon treatment with [125I]NaI in the presence of H2O2. The 16alpha-[125I]iodo- and isomeric (17alpha,20E/Z)-[125I]iodovinyl-18-methylestradiols were evaluated for estrogen receptor-mediated uterine uptake in immature female rats. Homologation of the C13-methyl group did improve the uterine uptake of the iodovinyl derivatives, but also increased blood retention, resulting in lower target uptake ratios. In the case of the 16alpha-iodo analog uterine retention decreased upon C13-homologation.

New iodinated progestins as potential ligands for progesterone receptor imaging in breast cancer. Part 1: Synthesis and in vitro pharmacological characterization

Five putative iodinated progesterone receptor (PR) binding ligands were synthesized and evaluated as potential imaging agents for PR-positive human breast tumours. Two compounds (E- and Z-17-hydroxy-21-iodo-19-nor-17alpha-pregna-4,20-dien-3-one; E- and Z-IPG1) were previously described, but are re-evaluated. The other three were novel compounds: two nortestosterone analogues derived from ORG 3236 (E- and Z-13-ethyl-17-hydroxy-21-iodo-11-methylene-18,19-dinor-17alpha-pre gna-4,20-diene-3-one; E- and Z-IPG2) and one norprogesterone analogue derived from ORG 2058 (21-[4-iodophenoxy]-16alpha-ethyl-19-norpregn-4-ene-3, 20-dione; IPG3). The E-iodovinyl nortestosterone compounds were obtained by a new route of synthesis. Competitive binding studies were performed to determine their binding affinities for the PR in three types of tissue (human MCF-7 breast tumour cells and rat uterine and mammary tumour tissue) and for the androgen receptor (AR) in human MCF-7 breast tumour cells, as well as for the sex hormone-binding globulin (SHBG) and corticosteroid-binding globulin (CBG) in human plasma. All four 17alpha-iodovinyl nortestosterone derivatives displayed high binding affinity for the human PR, that of Z-IPG1 and E- and Z-IPG2 being even higher than that of ORG2058. Their affinities for the rat PR were somewhat lower, especially those of both E-isomers. The affinity of IPG3 was lower for both the human and rat PR. The nortestosterone derivatives also showed AR binding, the relative binding affinities ranging from 4.3 to 17.0% as compared with 5alphaDHT. Additionally, neither of these steroids displayed any significant binding to either SHBG or CBG in human plasma. We conclude that the in vitro binding properties of all four 17alpha-iodovinyl nortestosterone derivatives warrant evaluation of the distribution characteristics of their 123I-labelled analogues to determine their usefulness as PR imaging agents.

Synthesis, estrogen receptor binding, and tissue distribution of a new iodovinylestradiol derivative (17alpha,20E)-21-[123I]Iodo-11beta-nitrato-19-norp regna-1,3,5 (10),20-tetraene-3,17-diol (E-[123I]NIVE)

We have synthesized and evaluated E-11beta-nitrato-17alpha-iodovinylestradiol (E-NIVE; E-3c) and its 123I-labelled form, as a new potential radioligand for imaging of estrogen receptor (ER)-positive human breast tumors. E-[123I]NIVE was prepared by stereospecific iododestannylation of the E-tri-n-butylstannylvinyl precursor (E-2c), obtained from reaction of 11beta-nitrato-estrone (8) with E-tributylstannylvinyllithium. In competitive binding studies, E-NIVE proved to have high binding affinity for both the rat and the human ER (Ki 280-730 pM), without significant binding to human sex hormone binding globulin. Distribution studies in normal and mammary tumor-bearing rats showed specific ER-mediated uptake of E-[123I]NIVE in the estrogen target tissues, i.e., uterus, ovaries, pituitary, and hypothalamus, but not in the mammary tumors. Selective retention in these target tissues, including tumor tissue, resulted in significant increases over time for the target tissue-to-muscle uptake ratios, but not for the target tissue-to-fat uptake ratios. The tumor-to-fat uptake ratio even appeared constantly below 1. In the primary estrogen target tissues, E-[123I]NIVE displayed high specific ER-mediated uptake and retention, which resulted in moderate target-to-nontarget tissue uptake ratios. In contrast, in tumor tissue, E-[123I]NIVE uptake appeared to be rather low and not ER-specific. As a consequence, E-[123I]NIVE appears to be a less favorable radioligand for ER imaging in breast cancer than the previously studied stereoisomers of 11beta-methoxy-17alpha-[123I]iodovinylestradiol (E- and Z-[123I]MIVE; [123I]E- and [123I]Z-3b).

The Z-isomer of 11 beta-methoxy-17 alpha-[123I]iodovinylestradiol is a promising radioligand for estrogen receptor imaging in human breast cancer

The potential of both stereoisomers of 11 beta-methoxy-17 alpha-[123I] iodovinylestradiol (E- and Z-[123I]MIVE) as suitable radioligands for imaging of estrogen receptor (ER)-positive human breast tumours was studied. The 17 alpha-[123I]iodovinylestradiol derivatives were prepared stereospecifically by oxidative radioiododestannylation of the corresponding 17 alpha-tri-n-butylstannylvinylestradiol precursors. Both isomers of MIVE showed high in vitro affinity for dimethylbenzanthracene-induced rat and fresh human mammary tumour ER, that of Z-MIVE however being manyfold higher than that of E-MIVE. In vivo distribution studies with E- and Z-[123I]MIVE in normal and tumour-bearing female rats showed ER-mediated uptake and retention in uterus, ovaries, pituitary, hypothalamus and mammary tumours, again the highest for Z-[123I]MIVE. The uterus- and tumour-to-nontarget tissue (far, muscle) uptake ratios were also highest for Z-[123I]MIVE. Additionally, planar whole body imaging of two breast cancer patients 1-2 h after injection of Z-[123I]MIVE showed increased focal uptake at known tumour sites. Therefore, we conclude that Z-[123I]MIVE is a promising radioligand for the diagnostic imaging of ER in human breast cancer.

The stereoisomers of 17alpha-[123I]iodovinyloestradiol and its 11beta-methoxy derivative evaluated for their oestrogen receptor binding in human MCF-7 cells and rat uterus, and their distribution in immature rats

We studied the potential of both stereoisomers of 17alpha-[123I]iodovinyloestradiol (E- and Z-[123I]IVE) and of 11beta-methoxy-17alpha-[123I]iodovinyloestradiol (E- and Z-[123I]MIVE) as suitable radioligands for the imaging of oestrogen receptor(ER)-positive human breast tumours. The 17alpha-[123I]iodovinyloestradiols were prepared stereospecifically by oxidative radio-iododestannylation of the corresponding 17alpha-tri-n-butylstannylvinyloestradiol precursors. Competitive binding studies were performed in order to determine the relative binding affinity (RBA) of the unlabelled 17alpha-iodovinyloestradiols for the ER in both human MCF-7 breast tumour cells and rat uterine tissue, compared with that of diethylstilboestrol (DES). Target tissue uptake, retention and uptake selectivity of their 123I-labelled analogues were studied in immature female rats. All four 17alpha-iodovinyloestradiols showed high affinity for the ER in human MCF-7 cells, as well as rat uterus. Their RBA for the ER showed the following order of decreasing potency: RBA of DES >Z-IVE >Z-MIVE >E-MIVE > or =E-IVE. Neither of these 17alpha-iodovinyloestradiols showed any significant binding to the sex hormone binding globulin in human plasma. The biodistribution studies showed ER-mediated uptake in the uterus, ovaries and pituitary, that of E- and Z-[123I]MIVE being higher than that of E- and Z-[123I]IVE. High target-to-non-target tissue uptake ratios, especially at longer periods after injection (up to 24h), were exhibited by both isomers of [123I]MIVE. The uterus-to-blood uptake ratio was higher for E-[123I]MIVE. However, the uterus-to-fat uptake ratio appeared to be higher for the Z-isomer of [123I]MIVE, especially at 24h after injection. Metabolic properties and temperature effects, which play a more important role in vivo, probably cause the discrepancies seen between in vitro and in vivo binding results. On the basis of their in vitro binding properties and in vivo distribution characteristics we conclude that E- and Z-[123I]MIVE could be suitable radioligands for the diagnostic imaging of ER in human breast cancer. Therefore, further studies with these radioligands in mature normal and tumour-bearing rats are warranted.

[Immunoscintigraphy in oncology]

In vivo immunotargeting of specific antigens with radiolabeled monoclonal antibodies is a new method of diagnosis which has largely improved over the past 15 years. It has been especially developed in the oncologic field. Although it is not yet a daily procedure, immunoscintigraphy has been proven a useful diagnostic tool in many situations. It gives information about the presence of lesions and their neoplastic nature. For this reason, immunoscintigraphy appears as very complementary to morphological imaging and is mainly indicated when lesions have insufficient anatomical definition. Early detection of recurrences is actually the best indication of immunoscintigraphy. However, some problems still exist, concerning the tumor/tissues uptake ratio which is not sufficient for an optimal imaging contrast, and the apparition of human anti-mouse antibodies (HAMA) in the patients. At present, many works are in progress which could resolve these problems. Ultimately, well-performed clinical trials are necessary to establish the right place for immunoscintigraphy in the diagnostic and therapeutic strategies in oncology.

Comparisons of radioiodoestradiol blood-tissue exchange after intravenous or intraarterial injection

PURPOSE

Our study determines and compares how the major organs of large animals handle exogenous halogenated bioactive sex steroids within the first minutes after their i.v. or i.a. injection. The rationale is that an understanding is needed of the acute physiological events because they affect decisions for how to optimize delivery of radiohalogenated sex steroid receptor ligands for purposes of medical imaging and modes of radiotherapy.

METHODS AND MATERIALS

We used an indicator dilution technique that allows monitoring of blood-tissue exchange of radioactivity in a continuous manner in anesthetized surgically prepared swine.

RESULTS

In swine, with by-passed liver circulation, the lungs allow the vast majority of [I-125]-16 alpha-iodo-17 beta-estradiol ([I-125]E) to be extracted from the blood perfusing the lung in the initial transit after i.v. injection in vivo. Similar outcome was observed for most major organs, including the CNS, intestines, spleen, peripheral appendages, and kidneys after i.a. injection of [I-125]E in vivo. However, within minutes the organs released the [I-125]E in its original chemical form back into the vascular system, with the exception of estrogen receptor (ER) rich tissues and the kidneys that retained the [I-125]E in its original form, although in the kidneys a nonpolar metabolite also accumulated.

CONCLUSION

Our experiments confirm in a large animal model that radioiodoestradiol can be sequestered or concentrated in ER-rich sites. The liver and sex steroid receptor-rich organs modify considerably, by metabolism and sequestration, respectively, the acute distribution of bioactive steroids. Our data indicate potential for detection of ER in vivo in hormone-sensitive tumors, that is, in breast and endometrial cancers, and offer improved understanding of the recent studies in subjects with breast cancer that demonstrated that receptor imaging in vivo of steroid receptors with high-affinity radiolabeled ligands is possible in a clinical setting.

Radiolabeled steroidal estrogens in cancer research

It has now been more than 30 years since the laboratory verification of the localization of estrogen in certain animal tissues. Much has been learned since that time regarding the details of this process, including the presence of specific receptors for these hormones in target tissues, the mechanism of ligand binding, the association of the ligand-receptor complex with unique chromatin sequences, and the activation of transcription. A concrete use of this knowledge has been the exploitation of these receptors as a targeting mechanisms for radiopharmaceuticals. This is an exciting area that encompasses both diagnosis and therapy. This review will summarize the in vitro and in vivo data obtained from evaluation of the many compounds that have been examined as radiolabeled receptor ligands, and will also discuss the chemistry necessary for their preparation. In particular, relative binding affinity values for relevant compounds will be tabulated, grouped according to molecular class. For those materials for which biodistributions have been performed, uterine (target), liver (nontarget, clearance), and, when available, tumor tissue uptake values are presented. These data should provide a reminder of what has been accomplished, and should serve as a working reference for those engaged in the pursuit of new candidates for these applications.

Rapid vascular escape of arterially injected 16 alpha-radioiodo,17 beta-estradiol

PURPOSE

We undertook this study because confirmation of a rapid vascular escape and slow release back into the circulatory system suggests that arterial injection of radiohalogenated steroid receptor ligands might provide an efficacious route of administration for imaging or treatment of receptor-rich malignant tumors in peripheral tissues.

METHODS AND MATERIALS

We injected radiolabeled 16 alpha-iodo,17 beta-estradiol ([I]-E), an estrogen receptor ligand, into the femoral artery of swine in a solution that contained [125I]-E in a known ratio to [99mTc]-labeled red blood cells. Fractions of femoral venous blood were collected at short intervals during 10 min. We looked for changes in the ratio of the radiolabels. [99mTc]-labeled red blood cells are known to remain in the vascular system for an hour or more.

RESULTS

After passage of the injectate through the capillary bed of the swine leg, a dramatic decrease of the initial 125I:99mTc ratio to only 10% was observed in the femoral venous blood. This ratio increased gradually during the next 10 min to approximately 30% of that in the injectate, indicating that a significant portion (approximately 90%) of the [125I]-E was initially trapped in the limb and then slowly re-entered the vascular system. To obtain visual confirmation of the rapid vascular escape of iodo-estrogen, we injected either an imageable form of [I]-E (123I]-E) or [99mTc]-labeled red blood cells into the dorsal aorta of superovulated rabbits, whose smaller size allowed whole-body imaging. The biodistributions of these radiopharmaceuticals were surveyed continuously by real-time planar gamma imaging. Within 2 min after the injection of [123I]-E, the outlines of the circulatory system were entirely lost; however, some estrogen receptor-rich tissues (the ovaries) as well as some non-target tissues, for example, the lower leg extremities, yielded well-defined images. In contrast, after intra-arterial injection of [99mTc]-labeled red blood cells, the circulatory system remained sharply defined for the duration of the study (40 min).

CONCLUSION

A large fraction of [I]-E escapes from the vascular system during the first pass through an organ or limb, without regard to the estrogen receptor content of the tissue.

Breaks in DNA accompany estrogen-receptor-mediated cytotoxicity from 16 alpha[125I]iodo-17 beta-estradiol

Strategies for diagnosis and therapy in which sex steroid receptor ligands serve as carriers for radionuclides are attractive because a high incidence of carcinomas of the female genital tract and the breast that are seen clinically have an abundant expression of one or more of the receptor proteins. A radiohalogenated estrogen receptor (ER) ligand, 16 alpha-[123I]iodo-17 beta-estradiol [123I]E, has met clinical criteria for receptor-mediated diagnostic imaging. Its [125]I-labeled sister nuclide derivative [125I]E decays by orbital electron capture with emission of very-low-energy (Auger) electrons, which gives this latter radiohalogen the potential to serve in pharmaceuticals for radiotherapy; as examples, [125I]deoxyuridine, when incorporated into the DNA molecule, or [125I]E, when bound to the receptor within ER-rich tumor cells, are both cytotoxic in vitro. Whereas the mechanisms and subcellular changes that accompany the cytotoxicity from [125I]deoxyuridine are well documented in the form of aberrations and breaks in the cellular DNA, the effects at the subcellular level causing the cytotoxicity of the sex steroid receptor ligand [125I]E have not been characterized and are the focus of our study. We found that in a standard colony-forming assay the addition of [125I]E to the cultures decreased the survival rate of ER-positive MCF-7 cells in a dose-dependent manner. The decreased survival rate was prevented by the addition of competing excess radioinert ER ligand (diethylstilbestrol); [125I]E did not reduce survival in ER-negative MCF-7 cells. The [125I]E-induced and ER-mediated cytotoxicity was accompanied by aberrations in the DNA components of the nuclei of the cells. These included chromatid and chromosome breaks, gaps, and tri-radial chromosome formation. Our findings add plausibility and credence to the notion that the cytotoxicity imparted by Auger-electron-emitting radioligands for sex steroid receptors is in part attributable to radiodecay that causes double-stranded breakage of DNA.

17 alpha-Z-[125I]iodovinyloestradiol and its 3-acetate: chemical synthesis in vivo distribution studies in the rat. Comparison of tissue accumulation and metabolic stability with 17 alpha-E-[125I]iodovinyl and 16 alpha-[125I]iodo oestradiols

Certain oestrogens on substitution with halogens, have been shown to retain receptor binding affinity and accumulate in target tissues and therefore could, when labelled with gamma-emitting halogen isotope, act as radiopharmaceuticals for the imaging of receptor positive breast tumours. In order to evaluate putative radiopharmaceuticals, 17 alpha-Z-iodovinyloestradiol ((17 alpha,20Z)-21-iodo-19-norpregna-1,3,5(10),20-tetraene-3,17 beta-diol) and its 3-acetate have been chemically synthesized and labelled with 125I. Tissue distribution studies in female rats have been compared to those obtained using 17 alpha-E-[125I]iodovinyloestradiol ((17 alpha,20E)-21-[125I]iodo-19-norpregna-1,3,5(10),20-tetraene-3,17 beta-diol) and its 3-acetate and to 16 alpha-[125I]iodo oestradiol (1,3,5(10)-estratien-16 alpha-[125I]iodo-3,17 beta-diol). All compounds showed a similar tissue distribution with the highest concentrations (cpm/g tissue) in the thyroid greater than liver greater than uterus greater than kidney greater than lung greater than blood greater than heart, spleen. The concentration in the uterus was highest after injection of 17 alpha-Z-iodovinyloestradiols or 16 alpha-iodo oestradiol. Target (uterus) to background (blood) tissue ratios were highest after injection of 17 alpha-Z-iodovinyloestradiol-3-acetate and 16 alpha-iodo oestradiol (6:1). Deiodination in vivo took place to a small degree, amounting to 1.6%, 0.9% and 0.35% of the injected dose after 4 hr with the 17 alpha-Z, 17 alpha-E and 16 alpha compounds, respectively. For reasons of chemical expediency and consideration of the biological results 17 alpha-Z-iodovinyloestradiol-3-acetate is the compound of choice for the detection of oestrogen receptor positive tissues.

Intracellular actions of steroid hormones and their therapeutic value, including the potential of radiohalosteroids against ovarian cancer

The biological activities of steroid hormones are effected via intracellular receptors. The receptors are part of a ligand-activated family of transcription regulator proteins that are critical for steroid-regulated cell differentiation. With recombinant cDNA technology, yeast and cultured animal cells can be made to express mammalian cDNA steroid receptors from cDNA clones that contain deletions and substitutions. Among the leading problems addressed in these models is the characterization of sequences that promote association or interaction with other transcription regulating molecules, including oncogene products. Recently it has been found that heat shock proteins may serve not only to stabilize the receptor proteins but also to precondition the activation imparted by ligand binding. Aberrant receptor proteins can be found in ovarian cancer. Whether aberrant receptor proteins are associated with transformation in general or with a variable clinical response to steroidal or anti-steroidal therapy is not known. Even after chemotherapy, steroid receptors are expressed in the metastases of ovarian cancers seen clinically, and they may have potential uses for localization and treatment of receptor-rich cancers. Radioligand pharmaceuticals appropriate for imaging or for site-directed radiocytotoxicity can be sequestered to the nuclei of receptor-rich cancers. Initial clinical imaging and therapy trials with such pharmaceuticals have been approved and begun. In the use of halogenated estrogen radiopharmaceuticals, liver metabolism and enterohepatic recirculation are important considerations. Ascites prolongs retention of a radiohalogenated estrogen in the abdominal cavity. Distant metastases have been localized with [123I]-estrogen in breast cancer patients in pre-operative procedures. Receptor-mediated cytotoxicity occurs when estrogen receptor radioligand pharmaceuticals that are Auger electron emitters are used in vitro.

Rapid liver metabolism, urinary and biliary excretion, and enterohepatic circulation of 16 alpha-radioiodo-17 beta-estradiol

The radiohalogenated estrogen 16 alpha-[123I]iodo-17 beta-estradiol ([123I]E2) is emerging as a diagnostic tool for imaging of ER-rich malignant tumors, with potential application for site-directed radiotherapy. Clinical use requires an accurate accounting for the biodistribution of the radioactivity, including an assessment of its enterohepatic circulation. We investigated the metabolism and circulation of [125I]E2 in the enterohepatic system in swine, a pharmacokinetic model that resembles humans. With indicator dilution methods, we found that, after its injection into the portal vein, more than 99% of [125I]E2 was cleared from the blood by the liver during the first pass. Water-soluble metabolites were then partly released into the blood and partly excreted into bile. After injection of [125I]E2 into the external jugular vein, one-third of the radioactivity was excreted in bile and two-thirds in the urine. More than 90% of the radioactivity in urine and bile was that of [125I]E2-glucuronide or [125I]E2-sulfate; only a very small fraction of the excreted radioactivity was from free 125I. Radioactivity in bile collected from one swine after i.v. injection of [125I]E2, and then infused into the proximal duodenum of a second swine, was almost totally absorbed during passage through the intestine at 5-7 hr after infusion. The reabsorbed radioactivity was cleared in the urine.