Radiopharmaceuticals for Bone Metastasis Therapy and Beyond: A Voyage from the Past to the Present and a Look to the Future

Radionuclides and ligands as potential radiopharmaceuticals for therapy of bone metastasis

An important aspect in the development of effective radiopharmaceuticals is to ensure their selective uptake at the region of interest, for example, the tumor. Thus, the antitumor effect is enhanced while the risk of harming the surrounding healthy tissue is minimized. This is particularly important in the bone cancer and metastasis, where during radiotherapy, the sensitive bone marrow could be subjected to the harmful radiation emitted by the radionuclide used to the palliative treatment of the bone tumor. ^33,34

The therapeutic radiopharmaceuticals are commonly composed by two key components: a radionuclide and a targeting ligand with which it is complexed. The radionuclide produces a relief after selective uptake at the target, ideally with negligible damage to healthy tissue. The function of the ligand is to prevent dissociation of the complex and facilitate the transfer of the radionuclide to the target, as in the case of secondary bone metastasis, requiring that it should be bone seeking. So, the ligand of choice should selectively accumulate in regions of high Ca²⁺ concentration, which is characteristic of areas affected by secondary bone metastasis. ^14,33,34

There is a great deal of interest in geminal diphosphonates or diphosphonic acids. ³⁵ These versatile molecules are characterized by a P-C-P structural element, and have a variety of important uses such as metal chelation ligands, ³⁶ therapeutic use in patients with two main types of disorders, ectopic calcification and ossification, and increased bone resorption, ³⁵ and in nuclear medicine as ligands for radiometals in bone-seeking diagnostic and therapeutic agents. ³⁷

Phosphonates are known to have a particular affinity for calcium (Ca²⁺), so they accumulate selectively in bone. ¹⁴ Some diphosphonic acid derivates include ethylenediaminetetramethylene phosphonate (EDTMP), methylene diphosphonic acid (MDP), 1-hydroxyethylenediphosphonic acid (HEDP), 1-hydroxy-3-aminopropylidene-diphosphonic acid (APD), N,N,-dimethylene-phosphonate-1-hydroxy-4-aminopropylidene diphosphonate (APDDMP), and recently, polyethyleneiminomethyl phosphonic acid (PEI-MP). ³⁸ The last two improve the properties of ¹⁵³Sm-EDTMP, widely used in the pain palliation therapy of patients suffering from bone cancer, not only providing pain relief but also suppressing and decreasing bone metastasis and even osteosarcomas. ^14,33,34

APD had been applied in the inhibition of osteoclast activity by adsorption on the bone surface (hydroxyapatite). In addition to minimizing resorption, APD also regenerates bone tissue by mobilizing Ca²⁺ and Mg²⁺ from blood plasma and subsequent deposition onto bone. ²³ Being such a versatile ligand, attempts were made to capitalize its capabilities by complexing it to a radioactive metal ion, and use it as a bone cancer diagnostic agent, for example, ^99mTc-APD for bone scintigraphy. ³⁹ However, in studies with trivalent lanthanides such as [¹⁶⁶Ho]Ho³⁺ and [¹⁵³Sm]Sm³⁺, a neutral complex [MLH]⁰ is obtained, and hence a colloid is formed at pH 7.4, which results in excessive liver uptake. Furthermore, APD exhibited a high affinity for Ca²⁺, which inhibits the delivery of the radionuclide [¹⁶⁶Ho]Ho³⁺ to the bone. In an endeavor to avoid the formation of neutral species, APD was modified by adding two charged methylene phosphonate groups at the primary amine centre, resulting in the synthesis of N,N,-dimethylene-phosphonate-1-hydroxy-4-aminopropylidene-diphosphonate (APDDMP)—with a net charge of 7. ⁴⁰ The radiolabeling of APDDMP with the same trivalent lanthanides resulted in complexes with a negative charge. Subsequent studies in animal models, using the complexes ¹⁶⁶Ho-APDDMP and ¹⁵³Sm-APDDMP, demonstrated that the uptake by the liver was avoided to a large extent. However, only the complex ¹⁵³Sm-APDDMP showed a good bone uptake, although less than ¹⁵³Sm-EDTMP. ⁴⁰

MDP (also known as medronate) is use as a radioactive bone-imaging agent after labeling with ^99mTc (^99mTc-MDP). Hydroxymethylene diphosphonate (HMDP) is sometimes used to replace MDP to target bone. Comparing both agents, studies showed that the cancerous/compact bone uptake is greater for HMDP than for MDP. Another ligand used is HEDP (also known as etidronate). As both HEDP and HMDP have a hydroxy group that MDP lacks, the MDP and the HEDP were chosen for further studies as possible pain palliative radiopharmaceuticals after labeling with an appropriate radionuclide. Although the bone uptake of ^99mTc-HEDP is lower and its blood clearance slower than of ^99mTc-MDP, it gives a greater contrast between regions of higher and lower calcification rates. As the areas with higher rates of calcification should be the target of a possible palliative radiopharmaceutical, HEDP seems to be a promising candidate. ⁴¹

When planning radionuclide therapy, key factors to be considered in selecting an effective radionuclide are (1) the half-life; (2) the radiation characteristics; (3) the ability to produce the radionuclide with high specific activity (e.g., high amount of radioactivity per unit mass), and (4) the radionuclide purity. ^42,43

The physical half-life of a radionuclide determines the initial dose rate, and therefore the total amount of radioactivity to be administered. An higher initial dose rate may result in a more effective cell killing, but the therapeutic ratio between malignant cell destruction and normal cell recovery may be less. An overly long physical half-life increases the amount of radiation that is delivered to tumor cells to achieve the therapeutic level before excretion, but also can create problems related with environmental safety in case of a spill or early death of the patient. An extremely short physical half-life may not allow enough time for the tumor-targeting process to take place, resulting that the majority of the radioactive decays occurs in the vicinity of or even in the health tissues. Also, a short physical half-life requires a larger total activity administered for therapy; therefore, it increases the radiation dose to personnel and family members what will require hospitalization with increase costs. ^34,44 So, the physical half-life of the radionuclides should preferably be of the same order of magnitude as the biological half-life of the radionuclide or the radionuclide conjugate. ³⁴ It seems reasonable to assume that the most suitable physical half-lives range from a few hours up to a few days when the targeting of disseminated cells is desired. Longer physical half-lives (up to one or a few weeks) might be needed to achieve significant uptake in solid tumor masses. ³⁴

The main precondition for a successful radionuclide therapy is the delivery of a high local radiation dose to the tumor cells and a low dose to healthy tissues. This defines the main requirement for a radionuclide: the energy emitted during its decay should be mainly deposited locally, while whole-body irradiation must be as small as possible. ³⁴

The characteristics of the radiation emitted during radionuclide decay are also an important point to be considered. If the radionuclide during its decay emits γ-photons with an appropriate energy, it could be useful to monitor the distribution of the radiopharmaceutical in the patient for assessing dosimetry. They also present a source of radiation exposure to personnel and to the family. ⁴⁴ To meet the requirements, the general demands for the physical properties of radionuclides should be that

• the radionuclide should emit α- or β-particles, Auger and/or conversion electrons in adequate abundance to induce tumor cell death;

There are three major groups of radionuclides for therapy, the β-particle-emitting radionuclides ( ⁶⁷ Cu, ⁹⁰ Y, ¹³¹I, ¹⁷⁷Lu, ¹⁸⁶Re, and ¹⁸⁸Re), Auger electron cascades ( ¹¹¹ In and ¹²⁵I), and α-particle-emitting radionuclides (²¹¹At, ²¹²Bi, ²¹³Bi, ²²⁵Ac, ²²⁷Th, and ²²³Ra). ^34,45,46 There are several commercially available radionuclides for therapy (see Table 2).

There are several β-emitting radioisotopes currently being used for pain relief, including Phosphorus-32 ( ³² P), Strontium-89 ( ⁸⁹ Sr), Samarium-153 (¹⁵³Sm), Rhenium-186 (¹⁸⁶Re), and Rhenium-188 (¹⁸⁸Re). The individual nuclides differ in terms of efficacy, duration of pain relief, tumoricidal effects, repeatable treatments, toxicity, and expense. Despite these differences, all of the radionuclides or their attached ligands preferentially target osteoblastic surfaces, suggesting a greater benefit in metastasis associated with increased osteoblastic activity. ^34,45

It has been indicated that each radionuclide can only be used optimally for tumors of a specific size. ³⁴ By choosing a radionuclide with the most suitable β⁻ energy, it becomes possible to modulate the penetration range in different tissues. ⁶

High-energy β-particles, such as those emitted by Yttrium-90 ( ⁹⁰ Y) and ¹⁸⁸Re, are useful for treatment of bulky tumors. In this case, the long range can compensate the poor penetration of the targeting molecule into a tumor mass and overcome a possible heterogeneity of target expression. They are not suitable for killing single disseminated cells or small metastasis, since only a small fraction of the electron energy will be deposited in such small targets. Most of the energy will instead travel beyond the tumor target to be absorbed in surrounding, often healthy, tissues. High-energy β-particles might, on the other hand, be important for treatment in cases of nonuniform radioactivity distribution in large tumor areas. Irradiation from the targeted cells will then enable a more uniform dose distribution and potentially elicit therapeutic effects on nontargeted tumor cells. In addition, it might be advantageous to use radionuclide cocktails to minimize the impact of heterogeneity. ^34,45 On the other hand, radionuclides emitting low-energy β-particles such as Copper-67 ( ⁶⁷ Cu), Iodine-131 (¹³¹I), and Lutetium-177 (¹⁷⁷Lu) are options for treatment of small tumor deposits or even single disseminated tumor cells. However, they could be inefficient for destroying single cancer cells or small micrometastasis, because most of the energy associated with the radioactive decay is deposited outside the malignant cell. So, a comparatively high dose of radioactivity per cell is needed when low-energy β-particles are used, thereby requiring a well-developed targeting process. ³⁴

The average β-energy lies between 0.13 and 1.7 MeV. The physical half-lives of the mentioned nuclides differ considerably from each other (from 1 to 52 days) (see Table 3). Therefore, the energy per time unit transferred to the tissue varies from nuclide to nuclide, although the total amount of radiation energy may be the same for two radionuclides. The radiation dose can be applied over a very short period, needing a high dose rate, or over a longer time period administering a radionuclide with a low dose rate. Since the killing of tumor cells has become the important issue, new treatment schemes that use repeated radionuclide applications and that administer radionuclides combined with low-dose chemotherapy are increasingly being favored. ⁸

The application of radionuclides for treatment of painful metastasis has been investigated for several decades. In 1960s, the first nuclide administered for pain therapy of multiple osseous metastasis was ³² P. ⁴⁷ Initially, it was believed that its effect was mainly from incorporation into the tumor itself. However, the tumor-to-nontumor ratio was not very favorable, and the relief of pain is primarily because of its uptake by the bone mineral, and not by the tumor. In addition, uptake was high in a rapidly dividing tissue such as the bowel, but particularly in the red marrow itself. ⁴⁴ Today, we know that ³² P is incorporated into the DNA of rapidly proliferating cells of the bone marrow as well as in the trabecular and cortical structures of the bone. The ratio of normal bone to metastatic tissue was calculated as 1:2, and therefore is relatively low. ²⁸ This unfavorable ratio and the frequently observed strong myelosuppression were the reasons for abandoning of the ³² P. ⁸ Also, the absence of any γ-radiation emitted during its decay complicated the study of its biodistribution and biokinetics in humans. ⁴⁴ Many different radiopharmaceuticals such as Strontium-89 chloride, Yttrium-90 citrate, Rhenium-186-HEDP, Samarium-153-EDTMP, Tin-117m-diethylene-triamine-pentaacetic acid (DTPA), and Rhenium-188-HEDP have been investigated for use in therapy. ^47,48 Strontium-89 chloride ( ⁸⁹ Sr) was the nuclide most widely used in nuclear medicine for therapy. It has a long physical half-life, requiring a low administered activity, resulting in a rather low initial dose rate. In addition, it does not need repeated administrations for effect. ⁴⁷ Laing et al. treated 119 prostate cancer patients with painful metastatic bone disease, who did not respond to conventional therapy, by application of ⁸⁹ Sr. A total of 75% of the patients demonstrated a marked improvement of the pain status, and 20% of these patients were almost completely pain free. The effect of ⁸⁹ Sr treatment began 10–20 days postinjection and reached a maximum after 6 weeks. ^48,49 The pain improvement lasted for 6 months on average with a variation between 4 and 15 months. This group evaluated the efficacy of treatment by the decreasing of pain intensity, change of pain medication, the patient's mobility, and a score for the general patient's condition. The authors could not find a significant advantage of an activity of 3.0 MBq/kg body weight above that of 1.5 or 2.2 MBq/kg, resulting in a recommended activity of 150 MBq of ⁸⁹ Sr. This activity has been considered the standard ever since. ⁴⁹ Lewington et al. performed a randomized, placebo-controlled, double-blinded study in prostate cancer patients who were refractory to hormonal treatment and external radiation therapy. ⁵⁰ The patients treated with ⁸⁹ Sr showed a significant pain reduction compared to the patients in the placebo group. In this study, similarly to the Laing's, the evaluation of the efficacy included all the above-mentioned parameters. Considering that these patients were end-stage patients who had failed all conventional therapies, the effect of ⁸⁹ Sr treatment is impressive. Further studies confirmed the beneficial effect of ⁸⁹ Sr for pain treatment in prostate cancer patients. ⁵⁰ Quilty et al. demonstrated in 284 prostate cancer patients that one injection of ⁸⁹ Sr was as efficient as a hemibody irradiation that often showed intolerable side effects. ⁴⁷ Strontium-89 is excreted by the kidney to 70%–80% and is eliminated from the vascular compartment within the first few hours. ⁵¹ Except for the bone uptake and the excretion via the urinary system, there is no accumulation in any organ or system. Depending on the extension of the metastatic disease, the tracer uptake in the skeletal system ranges between 12% and 90% of the administered activity. With extensive presence of bone metastasis, the higher the uptake in to the skeleton is. The accumulation of Strontium-89 chloride ( ⁸⁹ SrCl₂) in the metastatic lesions is 5–20 times as high as the accumulation in normal bone tissue. Ninety days after the administration, 10%–88% of the injected Strontium-89 activity was found in metastatic bone lesions. ⁵² The effective half-life was calculated to be over than 50 days; thus, Strontium-89 chloride delivers a low dose rate radiation. ⁸

Like Strontrium-89, the calcium analog (it follows the biochemical pathways of calcium in the body) Yttrium-90 ( ⁹⁰ Y) is taken up by the bone depending on the intensity of the osseous metabolism. ^8,51

Radionuclides that have also previously been studied include ¹⁵³Sm and ¹⁶⁶Ho with observed β⁻ ranges of 0.55 and 2.7 mm respectively. ¹⁵³Sm and ¹⁶⁶Ho emit medium- and high-energy β-radiation, respectively, which limits an increase in the administered radiation activity not to endanger bone marrow in terms of the radiotoxicity. Therefore, their therapeutic efficacy is compromised. Samarium-153 is one of the vital radionuclides among the lanthanide elements from the point of view of nuclear medicine. Due to its short half-life, ¹⁵³Sm replaced the comparatively long-lived ⁸⁹ Sr isotope [41]. The therapeutic activity is 30 times more economic than the ⁸⁹ SrCl₂. Clinical studies show that the toxicity is lower, but the palliative effect is of shorter duration than desired. ^53,54

In the group of new radiopharmaceuticals, ¹⁵³Sm-EDTMP and ¹⁸⁶Re-HEDP are well studied and also commercially available. ^55,56

Samarium-153 chelated with EDTMP is widely used in the clinic for the effective palliative treatment of widespread skeletal metastasis, as it can be concentrated in bone metastasis having an osteoblastic component. Samarium-153 has a short physical half-life that can be advantageous, because it can be administered repeatedly. However, because of its short physical half-life and its production by reactor, delivery is difficult. The range of its β-particles is short (average 0.55 mm), resulting in good bone-to-marrow ratios. ⁴⁴ One hundred eighteen patients with painful bone metastasis were randomly assigned to receive a single dose of 18.5 MBq/kg of ¹⁵³Sm-EDTMP. The results of a patient-rated scale revealed a progressive decrease in pain during the first 4 weeks of the study in the treatment groups. ¹⁵³Sm-EDTMP is rapidly cleared from the blood into the urine, and only 1% of injected activity remains in the blood 4 hours postadministration, whereas it is retained in the bone for a long time. ^53,54,57 Alberts et al. concluded from a trial with 35 patients that 0.04 MBq/kg activity is adequate for ¹⁵³Sm-EDTMP, often requiring multiple applications for safe palliation of pain associated with metastatic bone cancer. ⁵⁸ This radiopharmaceutical is also a useful agent for bone metastasis from cancer of the prostate. ^37,59 Palliative and even curative effects have been demonstrated in dogs using ¹⁵³Sm-EDTMP to treat a variety of skeletal neoplasias, both primary and metastatic. In a double-blind placebo-controlled study, Serafini et al. investigated the effect of ¹⁵³Sm in 80 prostate cancer patients. Four weeks after the injection of a single dose of 0.03 MBq/kg body weight, an improvement of the pain situation was observed in 72% of the patients. In 31% of the patients, an almost complete pain reduction could be found. Four months after the treatment, 43% of the patients showed a continuing improvement of pain symptoms. In this study, a visual analog scale for different regions of the body, the consumption of analgesics, and pain scoring performed by the physician served as criteria for treatment response. The response rate of the ¹⁵³Sm group was significantly better than that of the placebo group, showing response rates of 40% and 2% after 4 weeks and 4 months, respectively. Furthermore, the study provided evidence that a dose of 0.03 MBq/kg body weight resulted in more frequent and longer periods of pain reduction than a dose of 0.13 MBq/kg body weight. ⁶⁰ However, Tian et al. were not able to confirm in their multicenter trial that the two different activity groups of ¹⁵³Sm experienced different quality on pain palliation. ⁶¹ Collins et al. reported that the onset of pain relief can be expected after 7–14 days. ⁶² Also for ¹⁸⁶Re, Maxon et al. demonstrated in a group of 20 patients that a significant improvement of pain can be achieved in 80% of the cases after a single injection. ⁶³

The radiopharmaceuticals mentioned above were not exclusively used to treat prostate cancer patients, but also those other tumors such as breast cancer and lung cancer. The tumor most frequently studied after prostate cancer is carcinoma of the breast. Robinson RG reported a response rate of 81% in breast cancer patients with multiple bone metastases on investigating 500 patients with different tumors after injection of ⁸⁹ Sr at a standard activity. ⁶⁴ Baziotis et al. treated 64 breast cancer patients by a single injection with 2 MBq/kg body weight of ⁸⁹ Sr. They found an improvement of the pain situation in 80% of the cases, including 35% of the patients, demonstrating almost complete pain relief. The average time response was 3 months. ^{65 153}Sm-EDTMP was also studied by Serafini et al. and Tian et al. They reported effective pain relief in 72%–85% of metastatic bone disease with a mean duration of 1–2 months. After 4 months, the response rate was still 43%. ^60,61 Hauswirth et al. investigated 17 breast cancer patients who received 0.95 MBq ¹⁸⁶Re and found a response rate of 60% and a mean duration of response of 5 weeks. ⁶⁶ Han et al. investigated 24 breast cancer patients in a dose–escalation study administering doses between 0.95 and 2.16 MBq and assessed the therapeutic effect by a multimodality pain evaluation scoring system. They also found a response rate of 60% and a mean duration of 1 month. In summary, also in breast cancer patients, there is evidence that radionuclide therapy is effective in palliating painful bone metastasis. ⁶⁷

In an attempt to improve on the success of ¹⁵³Sm-EDTMP, two possible strategies may be deployed. The first alternative is to use higher-energy β-emitting radionuclides, as was attempted with Holmium-166-EDTMP (¹⁶⁶Ho-EDTMP). ¹⁶⁶Ho is a radionuclide that emits higher-energy β-particles, which are thought to improve the therapeutic efficacy of bone-seeking radiopharmaceuticals due to the deeper soft tissue penetration. However, in the past few years, others have argued that minimizing the dose to bone marrow by using low-energy particle emitters such as ^117mSn will spare the bone marrow, and is therefore more likely to deliver a therapeutic dose to the cortical bone. However, the chemistry of a substitute radionuclide (which generally belongs to a different chemical element) will differ from that of Sm. Even when the radionuclides are chemically similar and occur in the same oxidation state (3+ for Sm), there may still be a significant difference in their in vivo behavior, as was proven for Ho(III). ¹⁴ A different radionuclide normally requires a different ligand, so that the complex radiopharmaceutical shows a bone uptake that would resemble that of ¹⁵³Sm-EDTMP. The high natural abundance of ¹⁶⁵Ho (100%), from which ¹⁶⁶Ho is easily produced by neutron activation, lowers the cost of production compared with that of ¹⁵³Sm. ¹⁴ The attempts with the above-mentioned ligand together with ¹⁶⁶Ho proved to be unsuccessful, not because of the radiation characteristics of ¹⁶⁶Ho, but rather due to the chemical properties of Ho(III) in combination of the ligands investigated. With ¹⁶⁶Ho-EDTMP, the in vivo stability of the complex proved to be inadequate [owing to changes in the Ho(III)-EDTMP formation constants, compared with Sm(III)-EDTMP], resulting in incomplete bone uptake. The MDP, HEDP, and APD ligands all form neutral species with Ho(III) and with Sm(III) at physiological pH resulting in colloid formation, which results in a reticuloendothelial uptake, which is undesirable. APDDMP (designed as a highly charged diphosphonate to overcome the neutral species scenario) proved to have a high affinity for Ca²⁺. The latter competes for APDDMP in blood plasma, resulting in less Ho(III)-APDDMP species being available in blood plasma and low bone uptake of ¹⁶⁶Ho. ¹⁴

¹⁸⁶Re and ¹⁸⁸Re are excellent examples of β-emitting radionuclides that could be used for pain palliation of bone metastasis. ^{68,69 186}Re-HEDP is the diphosphonate complex most frequently studied. ^{70 –72} This radiopharmaceutical can also be prepared using ¹⁸⁸Re to form ¹⁸⁸Re-HEDP, requiring both carrier Re to ensure a good yield of the bone-seeking agent. ^59,73 The physical half-life of ¹⁸⁶Re allows frequent repeat administrations in a short period of time. However, the average β energy is considerably higher than of the ¹⁵³Sm, and, consequently, the range is higher what is undesirable for the bone marrow. ⁴⁴ Reaction conditions for the synthesis of ¹⁸⁶Re-MDP must be acidic (pH 1.4–1.6) to facilitate reduction of the perrhenate with stannous ion and to keep the reduced species from being oxidized. ⁷⁴ The synthesis and stabilization of ¹⁸⁶Re-HEDP are different and can be accomplished at pH 5–8. Both ¹⁸⁶Re-HEDP and ¹⁸⁸Re-HEDP have been used quite successfully ⁴⁴ in alleviating pain and for treatment of multiple metastatic foci of bone in bone cancer patients. It has also been reported that the comparative instability of the ¹⁸⁶Re-HEDP radiopharmaceutical to in vivo oxidation in to perrhenate is a possible advantage. ³² The radiopharmaceutical ¹⁸⁶Re-HEDP washes off from normal bone faster than it does from cancerous bone, and consequently, the abnormal/normal bone uptake ratio increases with time. ^37,59

¹⁸⁸Re is an excellent candidate for β-particle therapy. ^{68,69 188}Re is easily obtained carrier free as perrhenate ion (ReO₄ ⁻) in a physiological saline solution ⁷⁵ or perrhenic acid (HReO₄) in aqueous HCl or HNO₃ ³⁵ from a rugged and economical alumina-based Tungsten-188/Rhenium-188 generator system. The alumina-based Tungsten-188/Rhenium-188 generator is completely analogous to the Molybdenum-99/Technetium-99m generator; however, the Tungsten-188/Rhenium-188 generator system has a greater shelf life, since Tungsten-188 (¹⁸⁸W) (t_1/2=69 days) has a much longer half-life than ⁹⁹ Mo (Molybdenum-99) (t_1/2=66 hours). ^{35,76 –79} Its energetic β-particles (2.1 MeV) have a maximum penetration in tissue of 10–11 mm, making this radionuclide a suitable option for large tumor masses. ⁸⁰ High-energy β-particles, such as ¹⁸⁸Re, are not effective for killing single disseminated cells or small metastasis, since only a small fraction of the electron energy will be deposited in such small targets. Most of the energy will instead travel beyond the tumor target to be absorbed in surrounding, often healthy, tissues. High-energy β-particles might on the other hand be important in cases where nonuniform radioactivity distribution in large tumor areas is needed. ^34,45 Its γ-ray emission (0.155 MeV) can be exploited for dosimetry purposes and to monitor biological distribution during therapy. ^{45,69,78,81,82} It also has a relatively short physical half-life of 16.9 hours that allows the use of high doses and reduces the problem of radioactive waste handling and storage. ^31,81 Targeting of bone by ¹⁸⁸Re has been accomplished effectively by using the complex ¹⁸⁸Re-HEDP. This agent has shown therapeutic efficacy in the treatment of metastatic bone pain associated with cancer of the breast, prostate, lung, and others. ^31,68,83

Hsieh et al. compared various Rhenium-188-labeled diphosphonates for the treatment of bone metastasis. In this study, they labeled MDP, HEDP, and HDP with ¹⁸⁸Re, and they analyzed the biodistribution and bone uptakes after an intravenous injection in rabbits. Their results showed that ¹⁸⁸Re-MDP and ¹⁸⁸Re-HDP tended to accumulate in the soft tissue and the liver. They believe that reactions between rhenium and diphosphonates are not the same as those between technetium and diphosphonates, irrespective of being in the same group on the periodic table. In fact, technetium is inherently a better oxidant than rhenium. More SnCl₂ needs to be added in the labeling of the rhenium-diphosphonate complex. In this study, they concluded that the presence of a carrier significantly affects the biodistribution of ¹⁸⁸Re-HEDP in rabbits. The bone-to-soft-tissue ratio of ¹⁸⁸Re-HEDP significantly increased after adding the carrier to the preparation. However, the carrier did not affect the biodistribution of ¹⁸⁸Re-MDP or ¹⁸⁸Re-HDP. The mechanism of the carrier effect is still not clear, requiring further study. Thus, the authors concluded that the HEDP was better than MDP and HDP as a bone-seeking tracer together with ¹⁸⁸Re. ⁵⁹

The commercially available radiopharmaceuticals for bone palliation ( ⁸⁹ Sr chloride, ¹⁸⁶Re-HEDP, and ¹⁵³Sm-EDTMP) are deposited in close vicinity to the bone, emitting the radiation to tumor and pain mediator cells. The higher the β-energy of the radionuclide the longer is the range of the electrons into the tissue. Since the bone is characterized by high-energy absorption, the range of therapeutic electrons emitted by osteotropic radionuclides does not reach more than a few millimeters at a maximum. ⁸ These agents are excreted mainly by the kidneys, and they disappear rapidly from the vascular compartment. ^45,84 Twelve hours after administration, 50% of the administered activity of ¹⁸⁶Re-HEDP and ¹⁵³Sm-EDTMP has been eliminated renally. The uptake of the injected activity for ¹⁸⁶Re-HEDP and ¹⁵³Sm-EDTMP in the skeleton was, respectively, 20%–30% and 30%–50%. Depending on the intensity of bone metabolism, the radiolabeled phosphonates are accumulated via adhesion to bone and bone metastases. The accumulation in the metastatic lesions is between 3 and 20 times as high as normal bone. The effective half-life of ¹⁸⁶Re-HEDP and ¹⁵³Sm-EDTMP lies in the range of 2–3 days. ⁸

Yttrium-90 ( ⁹⁰ Y) is a pure high-energy β-emitter. Its higher-energy β-particles and hence longer particle range in tissues provide the ability to treat larger tumors. Yttrium-90 is generator produced from its parent Strontium-90 ( ⁹⁰ Sr) and is available with high specific activity (no-carrier-added). Its trivalent ion is commonly used for chelation purposes. Used as the citrate salt, it shows 80% uptake in the bones. ^85,86 In this form, it has been used for pain palliation of metastatic disease. It has the chemical properties suitable for chelation to several commonly used compounds or macrocyclic ligands such as 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA). ⁸⁶ Although ⁹⁰ Y-citrate indicates high accumulation in the bone, a part of the Y³⁺ released by the dissociation of the citrate complex binds to serum, which results in delayed blood clearance and accumulation in the liver. ⁸⁵ Ogawa et al. hypothesized that a bone-specific ⁹⁰ Y-labeled radiopharmaceutical could be developed. So, they chose DOTA as the chelating site, and DOTA was conjugated with 4-amino-1-hydroxybutylidene-1,1-bisphosphonate (HBP). They studied the biodistribution of ⁹⁰ Y-DOTA-HBP and compared it with ⁹⁰ Y-citrate. Their results showed that ⁹⁰ Y-DOTA-HBP had superior biodistribution characteristics as a bone-seeking agent and led to a decrease in the level of unnecessary radiation exposure compared to ⁹⁰ Y-citrate. Even so, the plasma stability of ⁹⁰ Y-DOTA-HBP was not as high as expected. In addition, in the case of the absorbed dose to red marrow, which is the dose-limiting factor of radiopharmaceuticals for palliation of metastatic bone pain, the ratios of the absorbed dose in red marrow to that in osteogenic cells were almost the same for ⁹⁰ Y-DOTA-HBP and ⁹⁰ Y-citrate. Both compounds might show similar degrees of myelosuppression, which is the most important side effect. Because the radiation dose to bone marrow is highly influenced by the accumulation of radioactivity in the bone, improvement in the clearances from the blood and other tissues does not contribute much to the radiation dose to red marrow. Meanwhile, although the ratios of the absorbed dose in soft tissues to that in osteogenic cells of ⁹⁰ Y-DOTA-HBP were also lower than those of ¹⁵³Sm-EDTMP, ¹⁵³Sm-EDTMP has the advantage over ⁹⁰ Y-DOTA-HBP in terms of the effective dose equivalent (0.387 mGy/MBq of ¹⁵³Sm-EDTMP compared with 0.840 mGy/MBq of ⁹⁰ Y-DOTA-HBP) and effective dose (0.232 mGy/MBq of ¹⁵³Sm-EDTMP compared with 0.652 mGy/MBq of ⁹⁰ Y-DOTA-HBP). It is attributed to the difference in the radiation of red marrow. Therefore, ¹⁵³Sm could be preferred to ⁹⁰ Y as a radionuclide used in palliation therapy, because the energy of the ⁹⁰ Y β particles could be too high. ⁸⁵

¹⁷⁷Lu can be produced at adequate specific activities by irradiation of the natural lutetium target in moderate-flux reactors, and its long half-life allows for shipment over long distances. Additionally, its 208-keV γ-emission (11% abundance) allows imaging of its distribution to facilitate dose calculations. Recently, the usefulness of ¹⁷⁷Lu-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetramethylene phosphonate (¹⁷⁷Lu-DOTMP) was demonstrated in a study performed in a mouse model. ⁸⁷ The polyazamacrocyclic ligand framework may offer a complex that is kinetically more inert than ¹⁵³Sm-EDTMP. ⁸⁸ Bryan et al. evaluated in normal dogs the toxicity of ¹⁷⁷Lu-DOTMP used as a potential therapeutic radiopharmaceutical. ⁸⁷ Dogs with osteosarcoma were previously demonstrated to be important models of naturally occurring disease in humans. ⁸⁹ The results showed that the dogs receiving ¹⁷⁷Lu-DOTMP tolerated the administration and the effects of the compound without apparent clinical toxicity, supporting the further evaluation in tumor-bearing dogs of ¹⁷⁷Lu-DOTMP as a potential therapy for metastatic bone cancer and primary bone tumors in humans and dogs. ⁸⁶

In contrast to the β-emitters, the α-particle emitters deliver a much more energetic and localized radiation, defined as high-linear energy-transfer (LET) radiation. ^89,90 α-particles are heavy, charged (two positive charges), and produce densely ionizing tracks through tissue that induce predominantly nonreparable double DNA-strand breaks. Patients with skeletal metastases often have chemoresistant disease and/or micrometastases with dormant clonogenic tumor cells residing in the cell cycle growth phase G₀. High-LET irradiation from α-emitters will kill such cells at a lower dose/dose-rate than low-LET irradiation. ²⁸ Despite the fact that α-emitters are more toxic and mutagenic than β-emitters, these adverse properties can be compensated in the targeted therapy due to irradiation of smaller volumes of normal cells when α-emitters are targeting tumor cell clusters. This feature promotes the treatment of skeletal metastases, because the short α-tracks cause a lower dose delivered from the bone surfaces to the clonogenic bone marrow cells located within the bone cavities. Also, the spatial distribution of the hydroxyapatite target within an osteoblastic tumor would facilitate a volume distribution of the radionuclide where the tumor cells would be easier reached by α-particles despite the limited track lengths. ²⁸

The progress in the biomedical application of α-emitters have been delayed by the low availability of radionuclides with proper physical and chemical characteristics, supply limitations, as well as the costs for the most popular α-emitters, Astatine-211 (²¹¹At; t½=7.2 hours), bismuth 213 (²¹³Bi; t½=45.6 minutes), and Actinium-225 (²²⁵Ac; t½=10 days). ^28,34 Recent research on α-emitters led to the development of long-term operating generators that can provide them in large quantities. Examples of such α-emitters are Radium-223 (²²³Ra; t½=11.4 days), Radium-224 (²²⁴Ra; t½=3.7 days), Thorium-227 (²²⁷Th; t½=18.7 days), and the α-emitter generator Lead-212 (²¹²Pb; t½=10.6 hours). In the absence of suitable complexing agents for radium isotopes, investigation of ²²³Ra in radioimmunotherapy could not occur, but methods have recently been developed to encapsulate ²²³Ra and ²²⁵Ac into liposomes, ensuring adequate stability. ²⁸

Like strontium, radium is a natural bone seeker that has previously been used for targeting nonmalignant skeletal diseases, such as in the use of Radium-224 (²²⁴Ra) for treating ankylosing spondylitis, characterized by elevated bone synthesis. ⁵⁰

Radium-223 (¹²³Ra) is the most promising radium isotope, with favorable properties for use in targeted radiotherapy. ²²³Ra decays via a chain of short-lived daughter radionuclides into stable Lead, producing four α-particles. ²⁸ Radium-223 can be effectively produced in large amounts from sources of the precursor ²²⁷Ac (t½=21.7 years) in a long-term operating generator. Moreover, ²²³Ra half-life provides enough time for its preparation, distribution (including long-distance shipment), and administration to patients. Its low γ-irradiation facilitates handling, radiation protection, and treatment on an outpatient basis. ²⁸

Based on the excellent physical characteristics of ²²³Ra, were performed a series of studies, aiming to analyse the potential of this radioisotope in the treatment of bone metastases. In a study with ²²³Ra in mice, biodistribution was measured at 1, 6, 24 hours, 3 days, and 14 days after injection. A rapid uptake and prolonged retention were demonstrated in the skeleton, whereas soft tissue radioactivity cleared relatively rapidly. ^46,91

Animal data and dosimetric studies have indicated that bone-targeting α-emitters can deliver therapeutically useful radiation doses to bone surfaces and skeletal metastases, at activity levels that are acceptable for sparing bone marrow. ⁹² In a comparative study of ²²³Ra and the β-emitter ⁸⁹ Sr, it was found out that ²²³Ra and ⁸⁹ Sr had similar bone uptake, and estimates of dose deposition in bone marrow demonstrated a clear advantage of α-particle emitters being bone marrow sparing. ⁹³

The therapeutic efficacy of ²²³Ra was studied in a nude mice model. In this study, the animals were injected with 10 million MT-1 human breast cancer cells into the left ventricle. Seven days later, they were treated with ²²³Ra dosage, ranging 6 to 30 kBq per animal. All untreated control animals had to be scarified due to tumor-induced paralysis 20 to 30 days after injection of tumor cells, whereas the mice treated with an activity superior to 10 kBq of ²²³Ra showed a significantly increased symptom-free survival. ⁹¹

Based on the encouraging preclinical results, a phase I study has been conducted. This study involved 25 patients with bone metastases (10 females and 15 males). Each of the patients received a single injection of ²²³Ra and were monitored closely at the injection day, days 1, 2 and 7, and thereafter weekly to 8 week after the injection of ²²³Ra. Five patients were enrolled at each dosage level, starting at 46 kBq/kg and then increasing to 93, 163, 213, and 250 kBq/kg of body weight. ²²³Ra was well tolerated at therapeutically relevant dosages. The mild myelotoxicity, the generally weak side effects, and the encouraging pain scores found in this study encouraged the authors to conduct a phase II study. ⁹¹

In a randomized, double-blind, placebo-controlled, multicentre phase II study, the aim was to investigate the effect of repeated ²²³Ra doses in men with symptomatic, hormone-refractory prostate cancer. Sixty-four patients due to receive local-field, external-beam radiotherapy to relieve pain from bone metastases were assigned to receive either four repeated monthly injections of 50 kBq/kg ²²³Ra (33 patients) or repeated injections of saline (31 patients). Treatment lasted for 12 weeks, during which four injections were given at 4-week intervals, with the first injection given at the time of external-beam radiotherapy and no later than 7 days afterward. The results showed that ²²³Ra was well tolerated with little or no myelotoxic effect, and showed promising evidence of efficacy. ⁹⁴

A phase III, randomized, placebo-controlled, double-blind international human study had been conducted, comparing Alpharadin against placebo in men with symptomatic castration-resistant prostate cancer (CRPC) that has metastasized to the bone (Alpharadin in Symptomatic Prostate Cancer [ALSYMPCA]). It involved 921 patients, in over 100 centers in 19 different nations. They all were ineligible for docetaxel, could not tolerate it, or had not responded to the therapy with docetaxel. Patients were given either Alpharadin or a placebo intravenously up to six times, 4 weeks apart. In this study, 809 patients of the 921 patients were included in a planned interim analysis. In June 3, 2011, the Independent Data Monitoring Committee recommended stopping the trial early due to evidence of a significant treatment benefit. The more recent data of the 921 patients showed that Alpharadin improved overall survival by 44%, resulting in a 30.5% reduction in the risk of death compared to placebo. The median overall survival benefit with Alpharadin was 3.6 months (14.9 months in patients given Alpharadin vs. 11.3 months with placebo). In addition to improving overall survival, ²²³Ra dichloride led to a statistically significant delay in the time to skeletal-related events. Alpharadin has been granted Fast Track designation by the US Food and Drug Administration (FDA). Bayer plans to file Alpharadin seeking marketing approval for CRPC with regulatory authorities in the United States and Europe based on the ALSYMPCA data in the second half of 2012. ^95,96

Low-energy Auger electrons, which are emitted during electron capture or isomeric transition decay, are also considered as suitable particles for inactivation of single-spread malignant cells. These particles, due to their high yield per decay, are extremely radiotoxic if their tracks meet DNA, leading to a very high probability to induce a severe double-strand break and hence inactivate the cell. The major challenge in the use of Auger electrons for therapy is that their short path makes them effective only if the radioactive decay occurs in close proximity of the DNA. For this reason, a targeting agent, labeled with an Auger emitter, should be internalized by a malignant cell, translocated to the nucleus, and, ideally, incorporated into DNA. ³⁴

A radionuclide that could prove promising is radioactive ^177mSn. It emits monoenergetic conversion electrons (energies of 126–158 keV) with a discrete range (0.2–0.3 mm) in bone tissue, which allows for larger bone radiation doses without excessive radiation to the bone marrow. ²⁰ Being an Auger-emitting radionuclide, ^117mSn will introduce a highly localized distribution of the electrons once inside or close to the cell. Furthermore, ^117mSn possesses a favorable half-life of 13.6 days that, depending on in vivo pharmacokinetics, is long enough to deliver more Auger electrons for the treatment. In addition to these favorable radiation characteristics, tin ions exhibit an inherent affinity for bone as is observed in biodistribution experiments with rats and adsorption studies of hydroxyapatite. ²⁰ Although difficult to produce with the required specific activity, the interest in ^117mSn arises from its favorable half-life and discrete range in bone tissue, as compared with ¹⁵³Sm, ³² P, and ¹⁶⁶Ho. ⁵⁹

In a study of Atkins HL, a bone-to-marrow ratio of 11 has been recorded for ^117mSn-DTPA, which is far better than its closest rival, ¹⁵³Sm-EDTMP. As the energies of the β-particles emitted by ¹⁶⁶Ho and ³² P are similar, a bone-to-marrow ratio of about 1.3 is expected for a radiopharmaceutical containing ¹⁶⁶Ho. From this comparison, it is understandable that marrow depression is recorded, especially for ⁸⁹ Sr, and even for ¹⁵³Sm. ⁵³ Since the radionuclide is effectively fixed to the bone matrix, leading to a long biological half-life, a longer rather than shorter physical half-life is preferred, delivering as high a radiation dose as possible. Considering ¹⁶⁶Ho, its short half-life (26.7 hours) might prove to be inadequate, while ^117mSn (13.6 days half-life) would be ideal with a long half-life, which is not too long to require radiation safety precautions. The oxidation state of the Sn is also important. ¹⁴ Zeevaart et al. sought an improved bone-seeking radiopharmaceutical, so they used ^117mSn(II)–APDDMP. The results, using ECCLES (a blood plasma model based on thermodynamic equilibrium) clearly showed that the target organs were the kidneys and bladder. ECCLES could in this case accurately predict that ^117mSn(II)–APDDMP would have some bone as well as liver uptake. It furthermore could explain the reasons for the high kidney uptake, namely ^117mSn(II) radiopharmaceuticals are also dependant on the weakness of the complex between Sn(II) and the ligand carrier. This was verified by animal experiments with ^117mSn(II)–APDDMP. ^7,97

A recent approach to develop an effective radiopharmaceutical for therapy of bone cancer, ensuring the selective uptake of the radiopharmaceutical, was to design a water-soluble polymer that is bone-seeking, and that would exploit the disrupted vasculature in tumors according to the Enhanced Permeability and Retention (EPR) effect, as discussed by Maeda et al. ⁹⁸ and Seymour ⁹⁹ —the process in which macromolecules accumulate within tumor tissue due to leaky blood vessels and poor lymphatic clearance. The principal behind this scenario is that water-soluble macromolecules accumulate passively in solid tumors. ^{14,72,100,101}

The EPR effect is commonly observed in most solid tumors, either primary or metastatic in nature. In tumor biology, little is known about selective or tumor-specific characteristics compared with those of normal tissues or organs. The concept of the EPR effect in solid tumors is one of the few tumor-specific characteristics that are becoming a gold standard in antitumor drug delivery. ⁹⁸

Lyer et al. explained this phenomenon by analyzing the anatomy of the tumor vasculature. ¹⁰² The blood vessels in the tumor are irregular in shape, dilated, leaky, or defective, and the endothelial cells are poorly aligned or disorganized with large fenestration. Also, the perivascular cells and the basement membrane, or the smooth-muscle layer, are frequently absent or abnormal in the vascular wall. The tumor vessel has a wide lumen, whereas tumor tissues have poor lymphatic drainage. This anatomical defect, along with functional abnormalities, results in extensive leakage of blood plasma components, such as macromolecules, nanoparticles, and lipid particles, into the tumor tissue. Moreover, the slow venous return to tumor tissue and the poor lymphatic clearance mean that macromolecules are retained in the tumor, whereas extravasation into the tumor interstitium continues. ^97,100,102 The EPR effect is also modulated or mediated by various factors produced by tumor cells, infiltrating leukocytes or even tumor-surrounding normal cells. Blood vessels near tumor tissue are affected by vascular mediators, such as vascular permeability factor, bradykinin and prostaglandins, nitric oxide (NO), peroxynitrite, and matrix metalloproteinases, which increase the vascular permeability of the tumor tissue. ^97,100,102

Through this phenomenon, very high local concentrations of polymeric drugs at the tumor site can be achieved, for instance, 10–50-fold higher than in normal tissue within 1–2 days. Interestingly, the EPR effect does not apply to low-molecular-weight drugs because of their rapid diffusion into the circulating blood, followed by renal clearance. ¹⁰² The polymer must, therefore, be large enough not to be taken up by healthy tissue, but not so large as to trapped in organs such as the liver or kidneys. ⁵ Because accumulation of macromolecules by the EPR effect is a progressive phenomenon, it is essential that the drugs are stable in the plasma for long periods. In addition to prolonging the half-life in plasma of low-molecular-weight drugs or proteins, polymer conjugation also guides the drugs or radionuclides to their target by the EPR effect. The alteration in conformation of some proteins or molecules constituting a drug gave a stealth character and the ability to suppress the antigenicity, as well as diminishes uptake by the reticuloendothelial system or macrophages. For example, succinylation of proteins or conjugation of poly(d-Glu-d-Ala-d-Lys) or poly(d-Glu-d-Lys) was all found to decrease substantially the immunogenicity of albumin, lactoglobulin, myoglobin, γ-globulin, and lysozyme. Therefore, the half-life of polymeric drugs in the blood circulation can be extended greatly. ¹⁰²

Owing to the prolonged retention of the polymeric complexes by the EPR effect and the enhanced plasma half-life, polymer-conjugated drugs or radionuclides require less-frequent administration compared with free drugs, which is a great benefit to patients. ^102,103 So, tumor-selective properties combined with a radionuclide with a short tissue penetration (with the resulting higher possible administered dose) could enable a very effective way of producing a radiopharmaceutical that would have not only palliative, but also therapeutic properties. ¹⁴

Dormehl et al., aiming for a molecule for use in palliative therapy for bone metastases after suitable radiolabeling and considering the EPR effect, developed polyethyleneiminomethyl phosphonic acid (PEI-MP), a water-soluble polymer polyethyleneimine, functionalized with methyl phosphonate groups and synthesized by condensation of polyethyleneimine, phosphonic acid, and formaldehyde. ⁵ In addition to being bone seeking, PEI-MP would accumulate in solid tumors due to the EPR effect. Studies followed to establish the biodistribution and pharmacokinetic properties of different complexes of PEI-MP/metal (^99mTc, ^117mSn, and ¹⁸⁶Re). ^38,104,105 They used various molecular-sized PEI-MP radiolabeled with ^99mTc, taking into account the EPR effect, choosing three different sizes of the PEI-MP, namely 30–300 kDa, 100–300 kDa, and 10–30 kDa, to compare differences in their biodistribution and pharmacokinetics, using a normal primate model and scintigraphy. From the results, macromolecules with sizes ranging 30–300 kDa were characterized by excessive liver (21%–57%) and kidney (40%) uptake and accompanying long residence times (t_1/2 up to 24 hours). The percentage bone uptake averaged at 8% for these particles, excluding sizes 100–300 kDa, where very little bone uptake was seen (<1%). In this case, the blood clearance was also slow (t_1/2 ∼2 hours). The fraction size 10–30 kDa had comparatively low accumulation and short residence times in the liver (20%; 22 minutes) and kidneys (17.5%; 20 minutes), and although the bone uptake of 18% in this case was high, it is still low for a bone-seeking agent. These particles cleared from the blood with t_1/2 of 25 minutes, and seemed suitable for labeling with a therapeutic radioisotopic agent. Anionic species in the fraction of 10 to 30 kDa achieve good tumor uptake with minimum uptake in healthy bone, kidneys, or liver. The polymer clearly demonstrates the potential to deliver a therapeutic radionuclide selectively to tumors. ⁵

Zeevaart et al. proposed that PEI-MP could be combined with radioactive ^117mSn. For the EPR effect to apply, the macromolecules of ^117mSn-PEI-MP should be larger than 40 kDa, that is, large enough to avoid renal clearance. ¹⁴ Choosing two different sizes of the PEI-MP, namely 30–50 kDa and 10–30 kDa, Jansen et al. compared differences in their absorption characteristics when radiolabeled with ^117mSn in two different oxidation states (Sn²⁺ and Sn⁴⁺). ¹⁰⁶

In addition to the size of the polymer, the oxidation state of the tin had a significant effect on the adsorption behaviour. The affinity of the tin in both valence forms was governed by the size of the PEI-MP ligand, with a fourfold increase in the affinity constants, accompanied by a slight improvement in the maximum absorption capacities when in presence of the smaller fraction, namely PEI-MP (10–30 kDa). In general, the optimum results were with Sn²⁺ in the presence of PEI-MP (10–30 kDa), where the metal-ion exhibited a higher affinity than the ligand while the adsorption capacity of the two were essentially equivalent. However this may not necessarily be an optimal combination when considering the EPR effect, in which the larger PEI-MP fraction could predominate, while the adsorption characteristics serve merely to complement the EPR accumulation. Furthermore, the Tin-PEI-MP complexes were not effectively desorbed and became immobilized on the hydroxyapatite surface, which may be advantageous for therapy, thereby facilitating passive accumulation of the radiopharmaceutical. ¹⁰⁶

Dosimetry

The establishment of bone tissue occurs around conglomerations of tumor cells in primitive tumor bone or osteoma osteoid. ¹⁰⁷ If osteolytic metastases are present, a broad resorption line in lacunes is found at a distance 80–100 μm from the tumor borders. In the case of osteoblastic metastases, resorption lacunes are rarely found. Typically, the trabeculae are covered by freshly produced bone tissue, and the agent is integrated deeply into the bone structure. Therefore, the accumulation of radiopharmaceuticals is much higher in osteoblastic than in osteolytic metastases, with ratios of 1:15 and 1:3, respectively. The uptake of the radionuclide determines the therapeutic dose in the bone metastases, and thus the predictive value of bone scintigraphy previous to treatment is indispensable even if osseous metastases have already been diagnosed by other imaging modalities. ⁸

Taking standard activities of 0.1 MBq of ⁸⁹ Sr chloride, 1.89 MBq of ¹⁵³Sm-EDTMP, and 0.95 MBq of ¹⁸⁶Re-HEDP, the radiation dose to bone metastases lies between 8 and 90 Gy, 10 and 70 Gy, and 14 and 140 Gy, respectively. ⁸ The organ doses demonstrate that the kidneys and bladder receive noncritical doses. Normal bone tissue receives a dose between 1 and 2.5 Gy that is significantly below that of the metastatic lesions (see Table 4). ^63,108 Variation in radiation dose in osseous metastases can be explained by the different levels of radionuclide accumulation in the metastases. This stresses the importance of pretherapeutic scintigraphy to have at least a visual estimation of the tumor uptake and to predict therapy response. ⁸

New strategies to enhance the effect of radionuclide therapy

To enhance the effect of radionuclide therapy on the cancer cells, the following new strategies have recently been investigated: combined radionuclide and chemotherapy, high-dose radionuclide therapy, and repeated radionuclide therapy. ¹⁰⁹

It is known that special cytotoxic agents like cisplatin work as a radiosensitizer. This means that the addition of chemotherapy and radiation therapy not only has an accumulative effect on tumor cells, but this should result in an increased cell killing. If both treatment modalities are applied simultaneously, the side effects will decrease, because the administration of a reduced-dosage protocol for chemotherapy means that the side effects could be kept at a stable level, but the efficacy could increase due to a higher tumoricidal effect of combination therapy. ¹⁰⁹

However, in 1992, Mertens et al. investigated 18 hormone-refractory prostate cancer patients who received a combination of 0.1 MBq ⁸⁹ Sr and a low-dose cisplatin infusion (35 mg/m²). They observed good pain palliation and an improvement in hemoglobin, tumor markers, and bone scans in some patients. ¹⁰⁹ Sciuto et al. randomized 70 patients with painful bone metastases either to a group A receiving 148 MBq ⁸⁹ Sr and 50 mg/m² cisplatin or to a group B receiving 148 MBq ⁸⁹ Sr plus placebo. The follow-up was until death to evaluate outcome. Overall pain relief occurred in 91% of patients in group A and 63% of patients in group B, with a median duration of 120 days in group A and 60 days in group B. New painful sites on previously asymptomatic bone metastases appeared in 14% of patients in group A and in 30% of patients in group B. The median survival without new painful sites was 4 months in group A and 2 months in group B. Sciuto et al. observed a progression of bone disease in 27% and 64% of patients in group A and B, respectively. This shows quite clearly that the progression of bone metastases is slowed down by combined therapy. New painful sites are significantly less frequent, and this improves significantly the quality of remaining life, especially because hematologic toxicity is moderate. Between the two groups of the cited study, there is no significant difference regarding the side effects. The median global survival time was better (9 months) for combined therapy (only 6 months for ⁸⁹ Sr alone), but the difference was not statistically significant. ¹¹⁰

Concurrently, administered chemotherapy has been shown to be effective in enhance pain relief and delaying the onset of new painful metastases. However, the combined therapy does not necessarily yield a higher survival. Cisplatin and 5-fluorouracil are the most commonly used drugs. ¹⁷ The most effective radiotherapy or chemotherapy has not been established.

Another new approach to enhance efficacy of radionuclide therapy is by repeated administration, aiming at a higher radiation dose. Palmedo et al. demonstrated that the application of ¹⁸⁸Re-HEDP in humans is safe, and that pain palliation can be achieved in about 70% of patients. ¹¹¹ A major advantage is that ¹⁸⁸Re is inexpensively available from a ¹⁸⁸W/¹⁸⁸Re generator, and a kit is available for easy radiolabeling of the bone-seeking HEDP. It offers the possibility of repeated therapy without additional costs compared to those of a single injection. ⁸

A third new approach is the application of a high-dose radionuclide therapy necessitating bone marrow support. Anderson et al. administered different doses of ¹⁵³Sm-EDTMP (1, 3, 4.5, 6, 12, 19, and 30 mCi/kg body weight) in 30 patients with locally recurrent or metastatic osteosarcoma or skeletal metastases. The patients received peripheral blood progenitor cell CD34 (PBPC) or bone marrow support. The authors found that marrow radiation doses were linear with the injected amount of ¹⁵³Sm-EDTMP. Also, the grade of cytopenia was dose related. After PBPC or marrow infusion on day 14, and after ¹⁵³Sm-EDTMP injection, recovery of hematopoiesis was problematic in two patients who had received 0.81 MBq/kg body weight and were infused with <2×10⁶ CD34/kg. However, in the remaining patients, no complications were observed. Reduction or elimination of opiates for pain was seen in all patients, and there was no adverse change in appetite or performance status. The authors conclude that high-dose irradiation (39–241 Gy) by bone-targeted therapy with ¹⁵³Sm-EDTMP is feasible, and that nonhematologic side effects are minimal. ¹¹²