Clin Pharmacokinet 2003; 42 (15): 1319-1330 0312-5963/03/0015-1319/$30.00/0

LEADING ARTICLE

© Adis Data Information BV 2003. All rights reserved.

Bone-Specific Drug Delivery Systems Approaches via Chemical Modification of Bone-Seeking Agents Hideki Hirabayashi1 and Jiro Fujisaki2 1 2

Biopharmaceutical and Pharmacokinetic Research Laboratories, Fujisawa Pharmaceutical Company, Osaka, Japan Manufacturing Technology, Fujisawa Pharmaceutical Company, Osaka, Japan

Abstract

Despite several decades of progress, bone-specific delivery is still limited by the unique anatomical features of bone, which mainly consists of inorganic hydroxyapatite. A practical approach to this problem is to produce targeted drugs that have a high affinity for hydroxyapatite. Bisphosphonates are a class of synthetic compounds structurally related to pyrophosphate. Bisphosphonates rapidly localise on the bone surface after being administered either intravenously or orally, since the P-C-P portion of the bisphosphonate structure has high affinity for hydroxyapatite. Therefore, bisphosphonate modification might be a promising method for targeting drugs selectively to the bone. Bisphosphonate-conjugated drugs are hydrophilic and highly water-soluble due to the acidic nature of the bisphosphonate moiety at physiological pH, and therefore they hardly permeate through the biological membrane of soft tissues. These physicochemical changes also reduce the intrinsic susceptibility of the drug to metabolism, promoting urinary or biliary excretion as unchanged drug. All these physicochemical and pharmacokinetic alterations contribute to the exceptional skeletal disposition of bisphosphonate-conjugated drugs. Bisphosphonate conjugation is based on chemical modification of the targeting molecule, and therapeutically optimised bisphosphonate derivatives have to be custom-developed on a case-by-case basis. The bisphosphonate moiety is usually coupled with the targeting drug through a specific linkage. The high affinity of bisphosphonate conjugates for the bone is not simply dependent on the bisphosphonate moiety but on the resultant molecule as a whole, including the linker and the linked drug. Lipophilicity (represented as log P) appears to be an appropriate index for predicting the osteotropic properties of bisphosphonate derivatives. Several strategies using bisphosphonate-conjugated drugs have been investigated at a laboratory level with the aim of obtaining therapeutically optimised treatments for conditions such as osteoporosis, osteoarthritis and bone cancer. In each case, the intention is to achieve prolonged local exposure to high concentrations of the targeting drug, thereby improving therapeutic index by enhancing pharmacological efficacy and minimising systemic adverse effects. Although most examples of bone-specific drug delivery via bone-seeking agents still remain in preclinical studies, several phosphonate-coupled radiopharmaceuticals, such as samarium-153 complexed to tetraphosphonate, are expected to be an effective pain palliation therapies for metastatic bone cancer and are currently being developed in clinical trials. Furthermore, recent reports on

1320

Hirabayashi & Fujisaki

bisphosphonate-modified proteins have illustrated the feasibility of bone-specific delivery of biologically active protein drugs, such as cytokines and growth factors.

Bone diseases such as osteoporosis, osteoarthritis and bone cancer are closely associated with the aging process. A rise in the number of patients who suffer from such bone metabolic diseases would be a serious problem in an aging society. For example, osteoporosis is a debilitating disease of bone mass and microarchitecture which causes nearly 1.3 million fractures each year.[1] Bone fractures, especially hip and spine fractures, are often associated with particularly high morbidity and mortality, as well as disruptions in the quality of life of patients. As is often the case, the chronic and progressive nature of bone diseases makes long-term pharmacological therapy inevitable. The economic burden of bone disease, therefore, has increased during recent years in line with the increasing number of patients. For instance, the annual cost for treatment of osteoporosis amounted to $US12.9 billion in the US in 1997.[2] These costs are estimated to double during the next 50 years.[1] Thus, development of a new strategy for drug therapy of bone diseases is of considerable importance, not only to improve patients’ quality of life but also to limit treatment costs. A number of therapeutic agents are beneficial for treatment of bone diseases. However, because bones are distributed throughout the body, the systemic drug concentration needs to be maintained at a level such that the drug can exhibit its pharmacological activity at the peripheral site. This often causes unfavourable systemic adverse effects, thus producing a very narrow toxic-therapeutic window for treatment of bone diseases. Therefore, new drug delivery systems are needed to overcome these problems. Drug delivery systems can be categorised into two general approaches: time-controlled (or sustained) drug delivery and spatially controlled (or targeted) drug delivery. For skeletal tissue, the former approach has included local implantation of ceramic, biocompatible or biodegradable materials containing bioactive compounds, which has been widely investigated in animal models in recent years.[3-6] The implantation may be performed © Adis Data Information BV 2003. All rights reserved.

simultaneously with a surgical bone resection or restoration of bony tissue. Because the drugs are continuously and directly released from the implant materials to the defective bone region, a high concentration of the drug at the target site can be maintained, and, consequently, systemic adverse effects are minimised. Many research groups have indeed obtained promising results on sustained drug delivery via local implantation, and some of these may not be far from clinical trials. Unfortunately, these strategies are not applicable for most patients, and management of the post-implantation drug concentration is difficult in practice. Considering the very wide diversity of patient conditions, a drug targeting technique to enhance and prolong the drug concentration locally in the microenvironment of bone after systemic administration would be more ideal and convenient. Although a number of drug targeting systems have shown promise in improving therapeutic index by increasing efficacy and minimising adverse effects, a true bone-specific delivery system still remains to be developed. To attain this goal, there are several obstacles to be overcome. First, because bones possess a membrane consisting of lining cells, which functions as a marrow-blood barrier, the accessibility of exogenous large substances to the bone surface is extremely limited.[7] Thus, drug carriers or vehicles with large diameters, such as liposomes, are not suitable for skeletal drug delivery. Secondly, bones are mainly composed of the mineral hydroxyapatite. Thirdly, the expression of biomolecules with biological affinity that can present specific targets (e.g. enzymes, receptors and antigens) may be low in mineralised tissues, which implies that utilising biological affinity for bone-specific drug targeting is difficult. All things considered, one of the most practical approaches would be to develop drugs possess hydroxyapatite affinity via chemical modification of bone-seeking molecules. The purpose of this article is to review the literature on bone-specific drug targeting systems via chemical modification with bone-seeking agents. To our knowledge, bisphosphonate, phosphonate, tetraClin Pharmacokinet 2003; 42 (15)

Bone-Specific Drug Delivery

1321

cycline and an oligopeptide have so far been investigated as bone-specific carriers. The majority of the article is concerned with chemical modification with bisphosphonates, which have been extensively investigated in bone-specific targeting of therapeutic drugs, radiopharmaceuticals and proteins. As human data on pharmacokinetics and therapeutic efficacy are limited, we use animal data as a background to an overview of recent findings, and discuss the future of skeletal drug targeting systems. 1. Bisphosphonates Bisphosphonates are well-characterised boneseeking agents. Their pharmacokinetic properties have been reviewed recently.[8] Bisphosphonates are a class of synthetic compounds structurally related to pyrophosphate, an endogenous regulator of calcium homeostasis. Bisphosphonates feature two geminal phosphates (P-C-P bond) in their structures instead of the P-O-P bond of pyrophosphate (figure 1). Because they can act as a physiological modulator of calcium metabolism, several bisphosphonates, including clodronate, pamidronate, alendronate and tiludronate, have been developed clinically for treatment of bone metabolic diseases such as Paget’s disease, hypercalcaemia of malignancy, bone metastasis and osteoporosis.[9-12] Because the P-C-P portion of the chemical structure has high affinity for calcium crystals, bisphosphonates are rapidly trapped selectively in calcified tissues after entering the systemic circulation via either intravenous or oral administration.[13,14] Thus, bisphosphonate therapy itself is a bone-specific drug delivery system, in that bisphosphonates seek their target site and exert their pharmacological effects preferentially on the bones. Additionally, bisphosphonates are highly water-soluble and acidic compounds at any physiological pH, because of which they hardly distribute to soft tissues and the major route of elimination from the body is urinary excretion. In addition, OH O

P OH

OH R1 OH

OH O

P HO

Pyrophosphate

O

O

P

O

P

O

OH R1 OH Bisphosphonate

Fig. 1. Chemical structures of pyrophosphate and bisphosphonate.

© Adis Data Information BV 2003. All rights reserved.

unlike the P-O-P bond of pyrophosphate, the P-C-P bond in bisphosphonate is resistant to chemical and enzymatic hydrolysis.[15] Therefore, once bound to bone, bisphosphonates remain there for a long period, with a half-life of several months. Desorption of bisphosphonates from bone is probably dependent on the process of bone turnover, in which bone calcification and resorption occur by the interplay of osteoblasts and osteoclasts. These characteristics suggest that if the linkage between the bisphosphonate moiety and the targeting drug is precisely designed to be cleaved in the bone compartment, sustained release of parent drug after skeletal uptake can be expected. Based on these pharmacokinetic properties, bisphosphonate-conjugated compounds could be a promising approach for bone-specific drug delivery systems. 2. Impact of Bisphosphonate Modification on Pharmacokinetic Properties In addition to imparting high affinity for hydroxyapatite, attachment of bisphosphonate moiety also alters the physicochemical properties of the drug. These changes, consequently, alter the intrinsic pharmacokinetic characteristics of the drug. In this section, the pharmacokinetic properties of bisphosphonate-conjugated compounds are reviewed. 2.1 Bone-Specific Distribution

The bisphosphonate moiety preserves its high affinity for calcium crystals and osteotropic potential even when coupled to a bulky drug. Skeletal distribution of bisphosphonate conjugates is very rapid, and the peak concentration in the bone usually occurs within several hours after the dose.[16-18] On the other hand, bisphosphonate conjugates are not considered as typical of molecules that distribute into soft tissues, except for the kidney where they are eliminated into the urine. Due to their acidic and highly water-soluble properties, bisphosphonate conjugates hardly penetrate biological membranes.[19] The interplay of high affinity for calcified tissues and less for soft tissues explains the exclusive distribution of bisphosphonate conjugates in the skeleton. Clin Pharmacokinet 2003; 42 (15)

1322

Intra-bone distribution of bisphosphonates is not homogeneous. Generally, bisphosphonates tend to preferentially distribute in the part of bones where mineral density or calcium turnover rates are higher.[20,21] Similar observations were made by Thushima et al. for the distribution of bisphosphonate-conjugated estrogenic agent in rat tibia.[17] The deposition of bisphosphonate-conjugated estrogenic agent was highest in the growth plate of the tibia, and the epiphysis and metaphysis contained higher concentrations than the diaphysis. For the same reason, bisphosphonates significantly accumulate in regions of cancer metastasis in comparison with normal bones. Using this property, the bisphosphonateconjugation approach can be applied to cancer chemotherapy and diagnostic imaging of metastases in bone cancer, where pinpoint targeting to the pathogenic site can be achieved (see section 7). 2.2 Long Retention in Bones

Another prominent feature of the pharmacokinetics of bisphosphonates is their long retention in the bone. As mentioned in section 1, this is due to the high stability of the P-C-P bond against chemical and enzymatic degradation. Based on this property, Fujisaki et al. have proposed and validated the concept of controlled bone-specific delivery, which they termed the ‘osteotropic drug delivery system’ (ODDS), using a model bisphosphonate prodrug of carboxyfluorescein (CF-bisphosphonate).[16,22,23] After intravenous injection into rats, CF-bisphosphonate was rapidly cleared from the circulation and over 60% of the dose was rapidly localised in body bones. Once taken up into the bones, CF-bisphosphonate was retained for a long period with a halflife of 3.2 days. As the linker was subjected to hydrolysis either chemically or enzymatically in the bone, parent compound CF was released to the bones, and, consequently, into the circulation. As a result, the bone and plasma concentration of regenerated CF was maintained for a prolonged period, although CF itself has high plasma clearance and less affinity for the bone. Fluorescence microscopy was used to delineate the post-distribution behaviour in the bone. CF-bisphosphonate, bound to the bone surface, was buried into the deeper compartment of the bone in accordance with the bone calcification process. Therefore, the parent com© Adis Data Information BV 2003. All rights reserved.

Hirabayashi & Fujisaki

pounds regenerated from the conjugates by enzymatic or chemical cleavage of the linker have to diffuse through the bone matrix to reach the systemic circulation, which may explain the long duration of the parent compound in the skeletal and systemic compartments. In the ODDS, bones act not only as an affinity target but also as a drug reservoir. 2.3 Metabolism and Elimination

Bisphosphonate conjugates are polar compounds under physiological conditions. In general, polar compounds tend to be cleared from the body via either urinary or biliary excretion, or both, as unchanged drug. The intrinsic susceptibility of the conjugated drug to metabolism may be reduced by bisphosphonate modification.[17] For clinically used bisphosphonates, with low molecular weights, the relative contribution of biliary excretion to the overall elimination is considered to be insignificant in comparison to that of urinary excretion.[21] Such small and hydrophilic compounds with low molecular weights are not expected to undergo extensive biliary excretion.[24] On the other hand, some bisphosphonate-conjugated drugs are reported to be excreted significantly into the bile in rats.[17,18] This may be because conjugation of the bulky drug may increase the molecular size and the lipophilicity of the compound as a whole. Thus, in addition to urinary excretion, bisphosphonate-conjugated drugs can be subjected to biliary excretion, especially when a bulky drug is used. 3. Pharmacokinetic Drawbacks to the Bisphosphonate Conjugation Approach A major pharmacokinetic drawback in using bisphosphonate as a bone-seeking moiety is that the resulting conjugates can easily form colloids or precipitates of calcium salt/complex in the blood plasma. This physicochemical feature can also affect the pharmacokinetics and pharmacological properties, and cause adverse effects in other soft tissues. Due to the poor oral availability of bisphosphonates, parenteral routes are usually chosen in administration of bisphosphonate conjugates. However, special caution must be taken with high doses or rapid intravenous injection of bisphosphonate conClin Pharmacokinet 2003; 42 (15)

Bone-Specific Drug Delivery

jugates. It is empirically and experimentally known that, like bisphosphonates, bisphosphonate conjugates can be taken up in soft tissues, mainly in the liver and spleen, and retained there for a long period of time, especially after high doses and rapid administration.[18] Although the underlying mechanisms are still unclear, this may be caused by the formation of complexes of bisphosphonate conjugates with endogenous metal ions in the blood plasma. These complexes are recognised by hepatosplenic macrophages as foreign substances and then undergo phagocytosis.[25-27] Indeed, we have recently reported the significant role of macrophages in the hepatic uptake of a bisphosphonic prodrug of diclofenac (DIC-bisphosphonate).[28] This hepatosplenic disposition reduces the bone selectivity of bisphosphonate conjugates. Additionally, the possibility of tissue damage due to high amounts of the drug in liver and spleen cannot be ruled out. To avoid formation of colloids or precipitates of bisphosphonate conjugate and endogenous metals, dosage or mode of administration has to be optimised. Indeed, the in vivo disposition of DIC-bisphosphonate was affected by its maximal concentration in the plasma. Effective delivery of DIC-bisphosphonate to the skeleton was ensured by keeping the plasma concentration lower than that at which precipitation of complex between DIC-bisphosphonate and calcium occurs, via a slow infusion.[29] 4. Design of Bisphosphonate Conjugates The difficulties in chemical modification approaches are that the optimum conjugation has to be designed on a compound-to-compound basis. Attachment of the bisphosphonate moiety to drugs is usually achieved through a linkage between the bisphosphonate and a reactive group of the drug, such as an amino, hydroxyl or carboxyl group. Because in many cases these functional groups are, more or less, associated with the affinity to target receptors or enzymes, bisphosphonate conjugation can alter the intrinsic pharmacological activity of the drug. On the other hand, attachment to the two residual side chains available on the geminal carbon is considered to affect pharmacological and pharmacokinetic properties. Thus, the pharmacological and pharmacokinetic properties of the resultant conjugates are dependent on the targeting drug, the bisphospho© Adis Data Information BV 2003. All rights reserved.

1323

nate moiety and the linker, since they all interact with each other. Optimal design of bisphosphonate conjugates could be attained by a suitable combination of these three components of the conjugate. 4.1 Effects of the Bisphosphonate Moiety as a Bone-Seeking Agent

The mechanism of skeletal distribution of bisphosphonates is simple adsorption onto the bone surface.[30] Conjugation with a bisphosphonic moiety can provide many types of drugs with greater affinity for hydroxyapatite. Although the quantitative relationships of the binding property to bones have not been completely clarified, the in vitro affinity of bisphosphonates for hydroxyapatite has been long considered the most appropriate index for osteotropic ability. The two residual side-chains available on the geminal carbon (R1 and R2 in figure 1) are considered to be involved in binding to the calcium crystals. A hydroxyl or amino group on the geminal carbon provides higher affinity for hydroxyapatite via tridentate interaction of the two phosphates of the P-C-P bond and the hydroxyl or amino group.[31,32] Thus, high affinity for bone may be expected for bisphosphonates with hydroxyl or amino groups on the geminal carbon. In contrast, recent discussion on designing bisphosphonate derivatives often concentrates on the solubility of the bisphosphonate-calcium salt/complex, which may reduce the oral absorption of bisphosphonates and, as mentioned in section 3, stimulate the reticuloendothelial cell system, causing hepatosplenic distribution.[33-35] The susceptibility to reticuloendothelial cell recognition is correlated with the stability or solubility of the bisphosphonate-calcium complex/salt. The high affinity of bisphosphonates for calcium is partially associated with high precipitability of their calcium salt/complex, and thus to avoid hepatosplenic distribution bisphosphonates with low affinity for calcium are favoured. 4.2 Effects of the Linker Between the Bisphosphonate Moiety and the Drug

Dissociation of bisphosphonates from the bone surface is very slow, so there is a concern that if biologically active drugs are retained for too long in Clin Pharmacokinet 2003; 42 (15)

1324

the bone they could cause unexpected long-term adverse effects. To avoid this, the spacer between the drug and the bisphosphonic moiety should be designed not only to be stable during the period in which its conjugate is distributed to bone, but also to be unstable after uptake into the bone to allow release of the parent drug. For this purpose, an appropriately bioreversible linkage such as an amide or ester bond is used. Unfortunately, there is neither a multi-applicable linkage nor a general understanding of the principles of selecting proper linkages. Gil et al. synthesised several bisphosphonate prodrugs of prostaglandin E2 with different types of linkages, and then sought on a trial and error basis an optimal linkage that has an adequate balance of high stability in the bloodstream and instability in the bone compartment.[36] Two different research groups, Bauss et al.[37] and Fujisaki et al.,[38,39] attempted to improve the therapeutic profile of 17β-estradiol via bisphosphonateconjugated prodrugs, based on different designs of spacers, and have reported different results on pharmacokinetics and pharmacological effects. This indicates that the choice of spacer is an important factor for determining the pharmacokinetics and pharmacological potential of the conjugates. 4.3 Effects of Lipophilicity

The osteotropic abilities of bisphosphonate conjugates are different on a compound-to-compound basis. For optimal design of bone-seeking bisphosphonate conjugates, it is crucial to know the variability of osteotropic ability among compounds. Recently, we have reported the relationship between compound lipophilicity, represented as calculated log P (the theoretically calculated water/octanol partition coefficient), and the osteotropic properties of various types of bisphosphonate conjugates.[28,29] The fractional dose delivered to the skeleton after intravenous injection into rats decreased linearly with an increase in calculated log P (r = –0.881). Several interpretations can be proposed that are consistent with the combined pharmacokinetic behaviours of bisphosphonate conjugates. One is that the fraction of peripheral distribution could be increased by an increase in compound lipophilicity, thus augmenting biliary excretion in proportion to lipophilicity. Another is that the higher lipophilicity may © Adis Data Information BV 2003. All rights reserved.

Hirabayashi & Fujisaki

accelerate the ease of formation or the stability of bisphosphonate-metal complexes, which could encourage phagocytosis by macrophages. Since information on quantitative structure-pharmacokinetic relationships for bisphosphonate conjugates is scanty, these findings provide better insight for design of bisphosphonate conjugates with desirable osteotropic ability. 5. Bisphosphonate Conjugates of Therapeutic Agents Under Investigation Several bone-specific drug delivery systems via bisphosphonate conjugates have been tested in preclinical and clinical studies (table I). They aim at the optimum use of therapeutic agents in various bone diseases, including osteoporosis (prostaglandin E2,[36] estradiol[37-39] and synthetic estrogenic agents[17]), osteoarthritis (nonsteroidal antiinflammatory drugs[18]), chronic infections (fluoroquinolones[40]) and bone metastatic cancer (cisplatin,[41,42] melphalan,[43] methotrexate[44,45] and radiopharmaceuticals[46-49]). Bone targeting by conjugation with bisphosphonates has shown promise in enhancing and prolonging the pharmacological effects in bone and reducing adverse effects in other tissues. Among bone-related diseases, osteoporosis has its highest rate of occurrence in postmenopausal women. This disease particularly increases the risk of hip and spine fractures, which are often associated with high morbidity and mortality. Drug delivery systems capable of delivering antiosteoporotic agents selectively to bone tissues are of great interest.[50] In the next sections, several strategies for estradiol delivery to bone are presented and compared. 6. Bone-Specific Delivery Systems for Estradiol Since postmenopausal osteoporosis is mainly caused by estrogen deficiency, hormone replacement therapy with estrogen is considered to be an effective treatment. However, since estrogen receptors are present in many tissues, the opportunity for treatment is limited by various systemic adverse effects such as oedema and the risk of cancer.[51,52] These systemic adverse effects are related to the Clin Pharmacokinet 2003; 42 (15)

© Adis Data Information BV 2003. All rights reserved.

Preclinical

Preclinical

Preclinical

Estradiol

Prostaglandin E2

SM-16896 (estrogenic

(153Sm,

186Re)

Clinical

Radiopharmaceuticals

Complex

Hybrid (analogue)

Bone targeting

Bone targeting

Bone selective effects

Enhance effects, reduce toxicity

Enhance effects

Preclinical

Bone targeting

Methotrexate

Hybrid

Preclinical

Melphalan

Bone targeting

Enhance effects, long activity

Hybrid

Enhance effects, bone selective

Preclinical

Bone targeting

less medication

sustained release

Cisplatin

Hybrid (analogue)

Enhance effects, reduce toxicity,

Enhance effects, reduce toxicity

Bone targeting,

Bone targeting

resorption inhibitor and bone

sustained release formation enhancer)

Bifunctional effects (bone

(palliation, imaging)

Bone radiotherapy

(metastasis)

Bone cancer

Bone cancer

(metastasis)

Bone cancer

(osteomyelitis)

Chronic infection

Osteoarthritis

replacement)

(hormone

Osteoporosis

Osteoporosis

replacement)

medication Bone targeting,

(hormone

effects, reduce toxicity, less

sustained release

Osteoporosis

Indication

Injectable estradiol, enhance

Expectation

Bone targeting,

Pharmacokinetic properties

effects

Preclinical

Fluoroquinolone

Prodrug

Hybrid (analogue)

Hybrid (prodrug)

Prodrug

Conjugate type

antibacterials

Preclinical

Diclofenac

agent)

Development stage

Conjugated compound

Table I. Bisphosphonate and phosphonate conjugates of therapeutic agents in preclinical and clinical studies

46

44,45

43

41,42

40

18

17

36

37-39

References

Bone-Specific Drug Delivery 1325

Clin Pharmacokinet 2003; 42 (15)

1326

Hirabayashi & Fujisaki

high lipophilicity of estrogens, allowing wide distribution in the whole body.[53] To overcome this problem, several approaches via chemical modification using bisphosphonate, tetracycline and an oligopeptide as bone-seeking agents have been investigated in preclinical studies (figure 2).

6.2 Bisphosphonate-Conjugated Estradiol

Fujisaki et al. have introduced a bisphosphonate moiety to estradiol via an ester bond to synthesise disodium [17-(3′-hydroxy-1′,3′,5′estratrienyloxy)carbonylpropylcarboxamidomethylene]bisphosphonate(E2-bisphosphonate), and have demonstrated its usefulness from pharmacological and pharmacokinetic viewpoints in rats (figure 2).[38,39] After intravenous injection into rats at 1 mg/kg, E2-bisphosphonate was bound to the bone and retained there for a long period with a half-life of 13.5 days, whereas estradiol alone administered intravenously at 0.48 mg/kg (equimolar to 1 mg/kg of E2-bisphosphonate) was rapidly cleared from the bone compartment. E2-bisphosphonate in the bone seemed to be releasing the parent drug at a constant rate, since the bone concentration of regenerated

6.1 Tetracycline-Conjugated Estradiol

The concept of bone-specific drug delivery of estradiol via modification with a bone-seeking agent was first tested with a tetracycline conjugate (figure 2).[54] The resultant conjugate of estradiol and tetracycline showed high affinity for hydroxyapatite. However, there is no information on its pharmacokinetics or pharmacological effects on the bone. HO

OH O

O

O

HO O O O

NH+ O−

O

CF3

OH NH2

O O R N H

O

OH O

OH P

O

P N H

3

OH OH

P R=

O 1

CH3

OH

P OH

2

O(H)

ONa ONa

O

O H H

H 4

(H)O Estradiol

5 COOH CH2 CO

CH2 CH2

CO

NH

C H

CO

OH 6

Fig. 2. Strategies for bone-specific drug delivery of estradiol via chemical modification with bone-seeking moieties: (1) tetracycline;[54] (2) bisphosphonate (by Fujisaki et al.[38,39]); (3) bisphosphonate (by Bauss et al.[37]); (4 and 5) hexa-L-aspartic acid peptide.[55,56]

© Adis Data Information BV 2003. All rights reserved.

Clin Pharmacokinet 2003; 42 (15)

Bone-Specific Drug Delivery

estradiol was apparently constant throughout the experimental period of 28 days. As a result, the area under the concentration-time curve of regenerated estradiol in the bone when E2-bisphosphonate was injected intravenously was 64.6 times higher than when estradiol alone was administered. The pharmacological effects of estradiol and E2bisphosphonate on bone mineral density and uterine weight were compared in ovariectomised rats. A single treatment with E2-bisphosphonate 1 mg/kg (equimolar to 0.48 mg/kg of estradiol) exhibited significant inhibitory effects against the loss of bone mineral density (76.7% of control) without a significant increase in the uterine weight. On the other hand, oral treatment with estradiol 1 mg/kg five times per week significantly improved bone mineral density and also increased the uterine weight. Considering that E2-bisphosphonate or the bisphosphonate moiety itself had much less in vitro activity against bone resorption than typical clinically used bisphosphonates, the results demonstrate the prodrug property of E2-bisphosphonate. Thus, E2bisphosphonate has the potential to exert sustained therapeutic effects of estradiol selectively on the bone, which may improve patient compliance because of minimal adverse effects or less frequent medication. 6.3 Oligopeptide-Conjugated Estradiol

Recently, acidic oligopeptides have been proposed as novel bone-specific drug carriers.[55,56] This concept is based on the high affinity of small acidic peptides for hydroxyapatite crystals. After intravenous injection into mice, estradiol conjugated with hexa-L-aspartic acid peptide, E2-(L-Asp)6 (figure 2), was rapidly cleared from the plasma and exhibited preferential distribution into bone, where it gradually regenerated the parent drug. As a result, once weekly treatment with E2-(L-Asp)6 showed pharmacological activity comparable with that obtained with estradiol treatment every 3 days. Moreover, E2-(L-Asp)6 did not exhibit any systemic adverse effects typical of estradiol over the observed range of doses. Unlike the P-C-P bond of bisphosphonates, Lpeptides are biologically labile and undergo 1

1327

enzymatic degradation in the bone compartment. This property is advantageous for efficient drug release into the target site, and there is no concern about unexpected long-term effects of the boneseeking moiety. Further, E2-(L-Asp)6 seems very attractive as an alternative bone-seeking carrier to bisphosphonates, in that it cannot form colloids or precipitates with endogenous metals. 7. Radiopharmaceuticals Coupled with Bone-Seeking Phosphonates Based on a similar concept of bisphosphonate conjugation, several phosphonate-coupled radiopharmaceuticals are being aggressively investigated in clinical trials as agents for pain palliation in cancer metastasised to bone.[46-49] Among them, samarium Sm-153 lexidronam (Quadramet®1), which consists of the beta-emitter 153Sm coupled to the tetraphosphonate ethylenediamine-tetramethylenephosphonic acid (EDTMP), has been approved for routine use by the US FDA. In this product, 153Sm is coupled to the phosphonate part of EDTMP through formation of a stable chelate in a 1 : 1 ratio. 153Sm-EDTMP has a high affinity for bone minerals and selectively distributes into skeletal tissues, especially in lesions of enhanced metabolic activity. The accumulation of 153Sm-EDTMP in the metastatic region was about four times higher than that in normal bone. This is probably because calcium is the major target for chelation of 153Sm-EDTMP, and the metastatic region where metabolic turnover of calcium is very high may encourage the adsorption of EDTMP. Therefore, pinpoint targeting of radiopharmaceuticals to pathogenic sites in bone can be achieved. On the other hand, because the duration of therapeutic activity depends on the radiation half-life (46 hours for 153Sm), it is not necessary for 153Sm to be disengaged from the complex, which simplifies this system in comparison with the other strategies discussed earlier in this review. In addition, 153Sm is also a weak gamma emitter, suggesting the feasibility of medical imaging for delineating the progression of bone metastases.

Use of tradenames is for product identification only and does not imply endorsement.

© Adis Data Information BV 2003. All rights reserved.

Clin Pharmacokinet 2003; 42 (15)

1328

8. Bone-Specific Protein Delivery In treatment of established osteoporosis, a number of growth regulatory factors, such as fibroblast growth factor-I (FGF-I), bone morphogenetic proteins (BMPs) and transforming growth factor β (TGFβ), are being considered as potential agents to stimulate osteoblast accumulation, proliferation and differentiation. However, the optimal use of therapeutic proteins is often limited by their instability under in vivo conditions, with very short half-lives. Repeated administration of proteins is therefore essential in obtaining satisfactory effects, but may cause adverse effects on other tissues, or increase cost and hinder compliance. Recently, Uludag et al. have published the first reports on the feasibility of bone-specific protein delivery based on the concept of bisphosphonate modification.[57-59] They synthesised bisphosphonate-conjugated bovine serum albumin (BSA) as a model protein and investigated its affinity for bone in vitro and in vivo. Bisphosphonate-modified BSA showed a nonspecific adsorption on mineral hydroxyapatite and its affinity increased in accordance with the extent of bisphosphonate modification. After injection into the medullary cavity of the tibia of rats, 40–50% of the administered dose was recovered in the tibia, which is the nearest osseous tissue to the injection site. This suggests that bisphosphonate-conjugated BSA can rapidly permeate through capillary walls and diffuse thorough the perivascular space to reach binding sites in bone. Moreover, the retention of bisphosphonate-conjugated BSA entrapped in tibia was about 10-fold longer than that of unmodified BSA. Although further studies are required, these reports represent a step toward bone-specific delivery of biologically active proteins via bisphosphonate modification. 9. Conclusions In recent years, there has been growing interest in utilising the osteotropic properties of so-called ‘bone-seeking agents’ for bone-specific drug delivery. In particular, chemical modification approaches with bisphosphonates, typical bone-seeking agents, have been advanced as potential methods to enhance and prolong pharmacological effects in bone, reduce adverse effects in other tissues, and improve com© Adis Data Information BV 2003. All rights reserved.

Hirabayashi & Fujisaki

pliance because of less frequent need for medication. The difficulty is that bisphosphonate derivatives have to be designed and optimised on a compoundto-compound basis so that they can be efficiently delivered to osseous tissues and released ideally to target sites. To address these difficulties, comprehensive understanding of the physicochemical, pharmacological and pharmacokinetic properties of bisphosphonates is required. A major pharmacokinetic drawback is that bisphosphonate conjugates can form colloids or precipitates of calcium salt/complex in plasma after rapid injection of high doses. This sometimes leads to hepatosplenic or renal accumulation, and may cause adverse effects in such tissues. Thus, optimisation of the dosage regimen to avoid rapid increases in plasma concentrations of bisphosphonates is of great importance for the most efficient delivery of bisphosphonate conjugates to the bone. Several methods for bone-specific drug delivery via bisphosphonate conjugates have been tested in preclinical and clinical studies. Although most of these still remain at the preclinical level, they show promise in enhancing and prolonging pharmacological efficacy and minimising systemic adverse effects. Phosphonate-conjugated radiopharmaceuticals are in clinical use, and can be considered as proof of the concept that pinpoint targeting of drugs to pathogenic sites in bone can be achieved with the use of bone-seeking agents. The recent exploitation of bisphosphonate conjugation for bone-specific protein delivery is worthy of note. Acknowledgements Financial support for this manuscript was provided by Fujisawa Pharmaceutical Co. Ltd. The authors wish to thank Bindu Gadani, M.S., for critical reading and reviewing of the manuscript. The authors have no conflicts of interest that are directly relevant to the content of this manuscript.

References 1. Abbott III TA, Lawrence BJ, Wallach S. Osteoporosis: the need for comprehensive treatment guidelines. Clin Ther 1996; 18 (1): 127-49 2. Hoerger TJ, Downs KE, Lakshmanan MC, et al. Healthcare use among US women aged 45 and older: total costs and costs for selected postmenopausal health risks. J Womens Health Gend Based Med 1999; 8 (8): 1077-89 3. Tahara Y, Ishii Y. Apatite cement containing cis-diamminedichloroplatinum implanted in rabbit femur for sustained

Clin Pharmacokinet 2003; 42 (15)

Bone-Specific Drug Delivery

4. 5. 6.

7. 8. 9. 10. 11.

12.

13. 14. 15.

16.

17.

18.

19.

20.

21. 22.

release of the anticancer drug and bone formation. J Orthop Sci 2001; 6 (6): 556-65 Kato H, Neo M, Tamura J, et al. Bone bonding in bioactive glass ceramics combined with a new synthesized agent TAK-778. J Biomed Mater Res 2001; 57 (2): 291-9 S´anchez E, Baro M, Soriano I, et al. In vivo-in vitro study of biodegradable and osteointegrable gentamicin bone implants. Eur J Pharm Biopharm 2001; 52 (2): 151-8 Woo BH, Fink BF, Page R, et al. Enhancement of bone growth by sustained delivery of recombinant human bone morphogenetic protein-2 in a polymeric matrix. Pharm Res 2001; 18 (12): 1747-53 Talmage RV. Morphological and physiological consideration in a new concept of calcium transport in bone. Am J Anat 1970; 129 (4): 467-76 Lin JH. Bisphosphonates: a review of their pharmacokinetic properties. Bone 1996; 18 (2): 75-85 Cantrill JA, Anderson DC. Treatment of Paget’s disease of bone. Clin Endocrinol (Oxf) 1990; 32 (4): 507-18 Heath D. The treatment of hypercalcaemia of malignancy. Clin Endocrinol (Oxf) 1991; 34 (2): 155-7 van Holten-Verzantvoort AT, Bijvoet OL, Cleton FJ, et al. Reduced morbidity from skeletal metastases in breast cancer patients during long-term bisphosphonate (APD) treatment. Lancet 1987; II (8566): 983-5 Storm T, Thamsborg G, Steiniche T, et al. Effect of intermittent cyclical etidronate therapy on bone mass and fracture rate in women with postmenopausal osteoporosis. N Engl J Med 1990; 322 (18): 1265-71 Myers HM. Structure-activity relationships (SAR) of hydroxyapatite-binding molecules. Calcif Tissue Int 1987; 40 (6): 344-8 Bisaz S, Jung A, Fleisch H. Uptake by bone of pyrophosphate, diphosphonates and their technetium derivatives. Clin Sci Mol Med 1978; 54 (3): 265-72 Fleisch H. Bisphosphonates: a new class of drugs in diseases of bone and calcium metabolism. In: Baker PF, editor. Handbook of experimental pharmacology. Vol. 83. Berlin-Heidelberg: Springer, 1988: 441-66 Fujisaki J, Tokunaga Y, Takahashi T, et al. Osteotropic drug delivery system (ODDS) based on bisphosphonic prodrug. I: synthesis and in vivo characterization of osteotropic carboxyfluorescein. J Drug Target 1995; 3 (4): 273-82 Tsushima N, Yabuki M, Harada H, et al. Tissue distribution and pharmacological potential of SM-16896, a novel oestrogenbisphosphonate hybrid compound. J Pharm Pharmacol 2000; 52 (1): 27-37 Hirabayashi H, Takahashi T, Fujisaki J, et al. Bone-specific delivery and sustained release of diclofenac, a non-steroidal anti-inflammatory drug, via bisphosphonic prodrug based on the Osteotropic Drug Delivery System (ODDS). J Control Release 2001; 70 (1-2): 183-91 Boulenc X, Marti E, Joyeux H, et al. Importance of the paracellular pathway for the transport of a new bisphosphonate using the human CACO-2 monolayers model. Biochem Pharmacol 1993; 46 (9): 1591-600 Lin JH, Chen I-W, deLuna FA, et al. Effects of dose, sex, and age on the disposition of alendronate, a potent antiosteolytic bisphosphonate, in rats. Drug Metab Dispos 1992; 20 (4): 473-8 Lin JH, Duggan DE, Chen I-W, et al. Physiological disposition of alendronate, a potent anti-osteolytic bisphosphonate, in laboratory animals. Drug Metab Dispos 1991; 19 (5): 926-32 Fujisaki J, Tokunaga Y, Takahashi T, et al. Physicochemical characterization of bisphosphonic carboxyfluorescein for osteotropic drug delivery. J Pharm Pharmacol 1996; 48 (8): 798-800

© Adis Data Information BV 2003. All rights reserved.

1329

23. Fujisaki J, Tokunaga Y, Sawamoto T, et al. Osteotropic drug delivery system (ODDS) based on bisphosphonic prodrug. III: pharmacokinetics and targeting characteristics of osteotropic carboxyfluorescein. J Drug Target 1996; 4 (2): 117-23 24. Smith RL. Excretion of drugs in bile. In: Brodie BB, Gillette JR, editors. Handbook of experimental pharmacology: concepts in biochemical pharmacology. Vol. 28. Berlin: Springer, 1971: 354-89 25. M¨onkk¨onen J, Urtti A, Paronen P, et al. The uptake of clodronate (dichloromethylene bisphosphonate) by macrophages in vivo and in vitro. Drug Metab Dispos 1989; 17 (6): 690-3 26. M¨onkk¨onen J, Ylitalo P. The tissue distribution of clodronate (dichloromethylene bisphosphonate) in mice: the effects of vehicle and the route of administration. Eur J Drug Metab Pharmacokinet 1990; 15 (3): 239-43 27. M¨onkk¨onen J, Rooijen N, Ylitalo P. Effects of clodronate and pamidronate on splenic and hepatic phagocytic cells of mice. Pharmacol Toxicol 1991; 68 (4): 284-6 28. Hirabayashi H, Sawamoto T, Fujisaki J, et al. Dose dependent pharmacokinetics and disposition of bisphosphonic prodrug of diclofenac based on Osteotropic Drug Delivery System (ODDS). Biopharm Drug Dispos 2002; 23 (8): 307-15 29. Hirabayashi H, Sawamoto T, Fujisaki J, et al. Relationship between physicochemical and osteotropic properties of bisphosphonic derivatives: rational design for osteotropic drug delivery system (ODDS). Pharm Res 2001; 18 (5): 646-51 30. Fleisch H. Diphosphonates: history and mechanisms of action. Metab Bone Dis Relat Res 1981; 3 (4-5): 279-87 31. van Beek E, Hoekstra M, van de Ruit M, et al. Structure requirement of bisphosphonate actions in vitro. J Bone Miner Res 1994; 9 (12): 1875-82 32. Sunberg R, Ebetino FH, Mosher CT, et al. Designing drugs for stronger bones. Chemtech 1991; 21: 304-9 33. Cohen H, Solomon V, Alferiev IS, et al. Bisphosphonates and tetracycline: experimental models for their evaluation in calcium-related disorders. Pharm Res 1998; 15 (4): 606-13 34. Cohen H, Alferiev IS, M¨onkk¨onen J, et al. Synthesis and preclinical pharmacology of 2-(2-aminopyrimidinio) ethylene-1,1-bisphosphonic acid betaine (ISA-13-1), a novel bisphosphonate. Pharm Res 1999; 16 (9): 1399-406 35. Hoffman A, Stepensky D, Ezra A, et al. Mode of administration-dependent pharmacokinetics of bisphosphonate and bioavailability determination. Int J Pharm 2001; 220 (1-2): 1-11 36. Gil L, Han Y, Opas EE, et al. Prostaglandin E2-bisphosphonate conjugates: potential agents for treatment of osteoporosis. Bioorg Med Chem 1999; 7 (5): 901-19 37. Bauss F, Esswein A, Reiff K, et al. Effect of 17β-estradiolbisphosphonate conjugates, potential bone-seeking estrogen pro-drugs, on 17β-estradiol serum kinetics and bone mass in rats. Calcif Tissue Int 1996; 59 (3): 168-73 38. Fujisaki J, Tokunaga Y, Takahashi T, et al. Osteotropic drug delivery system (ODDS) based on bisphosphonic prodrug. IV: effects of osteotropic estradiol on bone mineral density and uterine weight in ovariectomized rats. J Drug Target 1998; 5 (2): 129-38 39. Fujisaki J, Tokunaga Y, Takahashi T, et al. Osteotropic drug delivery system (ODDS) based on bisphosphonic prodrug. V: biological disposition and targeting characteristics of osteotropic estradiol. Biol Pharm Bull 1997; 20 (11): 1183-7 40. Herczegh P, Buxton TB, McPherson III JC, et al. Osteoadsorptive bisphosphonate derivatives of fluoroquinolone antibacterials. J Med Chem 2002; 45 (11): 2338-41 41. Klenner T, Valenzuela-Paz P, Keppler BK, et al. Cisplatinlinked phosphonates in the treatment of the transplantable osteosarcoma in vitro and in vivo. Cancer Treat Rev 1990; 17 (2-3): 253-9 42. Klenner T, Wingen F, Keppler BK, et al. Anticancer-agentlinked phosphonates with antiosteolytic and antineoplastic

Clin Pharmacokinet 2003; 42 (15)

1330

43.

44.

45.

46. 47. 48. 49. 50. 51. 52.

Hirabayashi & Fujisaki

properties: a promising perspective in the treatment of bonerelated malignancies? J Cancer Res Clin Oncol 1990; 116 (4): 341-50 Klenner T, Wingen F, Keppler B, et al. Therapeutic efficacy of two different cytostatic-linked phosphonates in combination with razoxane in the transplantable osteosarcoma of the rat. Clin Exp Metastasis 1990; 8 (4): 345-59 Sturtz G, Couthon H, Fabulet O, et al. Synthesis of gembisphosphonic methotrexate conjugates and their biological response towards Walker’s osteosarcoma. Eur J Med Chem 1993; 28: 899-903 Hosain F, Spencer RP, Couthon HM, et al. Targeted delivery of antineoplastic agent to bone: biodistribution studies of technetium-99m-labeled gem-bisphosphonate conjugate of methotrexate. J Nucl Med 1996; 37 (1): 105-7 Serafini AN. Systemic metabolic radiotherapy with samarium-153 EDTMP for the treatment of painful bone metastasis. Q J Nucl Med 2001; 45 (1): 91-9 Lamb HM, Faulds D. Samarium 153Sm lexidronam. Drugs Aging 1997; 11 (5): 413-8 Eary JF, Collins C, Stabin M, et al. Samarium-153-EDTMP biodistribution and dosimetry estimation. J Nucl Med 1993; 34 (7): 1031-6 Lewington VJ. Cancer therapy using bone-seeking isotopes. Phys Med Biol 1996; 41 (10): 2027-42 Mundy GR. Pathogenesis of osteoporosis and challenges for drug delivery. Adv Drug Deliv Rev 2000; 42 (3): 165-73 Gambrell Jr RD. The menopause: benefits and risks of estrogenprogestogen replacement therapy. Fertil Steril 1982; 37 (4): 457-74 Colditz GA, Hankinson SE, Hunter DJ, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. N Engl J Med 1995; 332 (24): 1589-93

© Adis Data Information BV 2003. All rights reserved.

53. Pardridge WM, Mietus LJ. Transport of steroid hormones through the rat blood-brain barrier: primary role of albuminbound hormone. J Clin Invest 1979; 64 (1): 145-54 54. Orme MW, Labroo VM. Synthesis of β-estradiol-3-benzoate17-(succinyl-12A-tetracycline): a potential bone-seeking estrogen. Bioorg Med Chem Lett 1994; 4 (11): 1375-80 55. Yokogawa K, Miya K, Sekido T, et al. Selective delivery of estradiol to bone by aspartic acid oligopeptide and its effects on ovariectomized mice. Endocrinology 2001; 142 (3): 122833 56. Sekido T, Sakura N, Higashi Y, et al. Novel drug delivery system to bone using acidic oligopeptide: pharmacokinetic characteristics and pharmacological potential. J Drug Target 2001; 9 (2): 111-21 57. Uludag H, Kousinioris N, Gao T, et al. Bisphosphonate conjugation to proteins as a means to impart bone affinity. Biotechnol Prog 2000; 16 (2): 258-67 58. Uludag H, Gao T, Wohl GR, et al. Bone affinity of a bisphosphonate-conjugated protein in vivo. Biotechnol Prog 2000; 16 (6): 1115-8 59. Uludag H, Yang J. Targeting systemically administered proteins to bone by bisphosphonate conjugation. Biotechnol Prog 2002; 18 (3): 604-11

Correspondence and offprints: Dr Hideki Hirabayashi, Biopharmaceutical and Pharmacokinetic Research Laboratories, Fujisawa Pharmaceutical Company, Kashima 2chome, Yodogawa-ku, Osaka, 1-6, 532-8514, Japan. E-mail: [email protected]

Clin Pharmacokinet 2003; 42 (15)