REVIEW ARTICLE

Drugs 45 (I): 15-28. 1993 0012-6667/ 93/0001 -0015/$07.00/0 © Adis International Limited. All rights reserved. DRU1227

Liposomes in Drug Delivery

Clinical, Diagnostic and Ophthalmic Potential Gregory Gregoriadis and Alexander T. Florence Centre for Drug Delivery Research, School of Pharmacy, University of London, London, England

Contents /5 16 17 19 19 20

21 22 23 25

Summary

Summary I. Liposome Structure and Properties 2. Behaviour of Liposomes In Vivo 3. Clinical Applications of Liposomes 3. 1 Cancer Chemotherapy 3.2 Antimicrobial rherapy 3.3 Liposomes as Immunological Adjuvants in Vaccines 3.4 Diagnostic Imaging 3.5 Liposomes for Ocular Delivery of Drugs 4. Conclusions

Liposomes (phospholipid-based vesicles) have been investigated since 1970 as a system for the delivery or targeting of drugs to specific sites in the body. Because of their structural versatility in terms of size, composisition, surface charge, bilayer fluidity and ability to incorporate almost any drug regardless of solubility, or to carry on their surface cell-specific ligands, liposomes have the potential to be tailored in a variety of ways to ensure the production of formulations that are optimal for clinical use. This includes controlled retention of entrapped drugs in the presence of biological fluids, controlled vesicle residence in the blood circulation or other compartments in the body, and enhanced vesicle uptake by target cells. Accumulated in vivo evidence, particularly in areas such as cancer chemotherapy, antimicrobial therapy, vaccines, diagnostic imaging and the treatment of ophthalmic disorders has indicated clearly that some Iiposome-entrapped drugs and vaccines exhibit superior pharmacological properties to those observed with conventional formulations. Such work has encouraged the application of liposomes in the treatment of diseases in humans. A large number of trials in patients with cancer or infections suggest that certain Iiposomal drug formulations are likely to prove clinically useful.

16

Concerted efforts during the last few decades to improve drug effectiveness in therapeutic and preventive medicine have been greatly assisted by parallel developments in molecular and cell biology. These include hybridoma and recombinant DNA technology on the one hand and, on the other, the discovery of a number of cell membrane receptors and the understanding of their interaction with respective ligands. Such developments have given a new impetus to the concept, first articulated early this century, of conferring selectivity on drugs through targeted delivery. The concept entails the use of carriers which can bring drugs to, or facilitate their release, where they are needed. Carriers can do so either through an ability, inherent or acquired (through structural modifications), to interact selectively with biological targets, or because they are instructed to release drugs near or at the target in ways that are conducive to optimal pharmacological action (Florence & Gregoriadis 1991). Thus, ligands (which in the widest sense include antibodies) bind to their receptors on the surface of cells in a highly specific fashion, whilst liposomes and other colloidal microspheres can introduce their drug contents into the interior of cells through endocytosis or other pathways. Alternatively, microspheres may be induced to act extracellularly by releasing drug through the action of external stimuli (Roerdink & Kroon 1989). Introduction of drug-containing carriers into the body is often accompanied by changes in their structure and function which may be brought about by elements in the biological milieu with which the carrier comes into contact, by carrier sequestration by nontarget tissues, or by other events (e.g. development of immune responses against the carrier) that are detrimental to effective carrier function. Such anatomical and physiological considerations have resulted over the years in systematic studies on carrier behaviour in vivo and, when possible, its control. Much of the relevant knowledge obtained so far concerns liposomes and other lipid-based vesicles (Gregoriadis I 988a, 1992a).

Drugs 45 (1) 1993

1. Liposome Structure and Properties Confronted with excess water, phospholipids and other polar amphiphiles form closed concentric bilayer membranes, entrapping water and dissolved solutes (e.g. drugs) in the process (Bangham et al. 1965) [fig. I). Lipid-soluble or lipid-bound drugs can also be accommodated into liposomal membranes by a variety of techniques, for instance by mixing such drugs with the lipids prior to forming liposomes. A wide array of phospholipids and other lipids (e.g. non-ionic surfactants), as well as lipids extracted from biological membranes, can be used to prepare liposomes or other lipid-based vesicles. Depending on the gel-liquid crystalline transition temperature (Tc) of phospholipids (i.e. the temperature at which the acyl chains melt), liposomal membranes can attain various degrees of fluidity at ambient temperature. This, in fact, can be controlled quite accurately to achieve a wide range of Tc values by using appropriate mixtures of 2 or more phospholipids. In addition, liposomal surfaces can be charged negatively or positively by the incorporation of charged amphiphiles, or enriched with reactive groups to which ligands can be covalently linked. It is also possible to adjust the average vesicle size by sonication, detergent dialysis, microfluidisation, homogenisation and other techniques (Gregoriadis 1993a). The uniquely versatile nature of liposomes and other lipid-based vesicles as well as their biodegradable and inoccuous nature and similarity to biological membranes led to their use, in the early 1970s, in targeted drug delivery (Gregoriadis & Ryman 1972). Since then, many hundreds of drugs, including antitumour and antimicrobial agents, chelating agents, peptide hormones, enzymes, other proteins, vaccines and genetic material have been incorporated into the aqueous or lipid phase of liposomes of various sizes, compositions and other characteristics by an ever increasing number of technologies (Gregoriadis 1993a). Liposomes have been studied extensively in terms of both behaviour in vivo following administration enterally or parenterally, and pharmacological effect exerted by their drug contents either in experimental animals

17

Liposomes in Drug Delivery

Unilamellar vesicle

MultilBmeliar vesicle

Surface modifying/ protein/polymers Soluble drug

Fig. 1. Schematic representation ofliposome structure. The basic lipid bilayer structure forms a permeability barrier separating I or more concentric aqueous compartments from the exterior aqueous phase. Drugs can be entrapped in the aqueous compartment(s) or within the bilayers. Surface properties can be modified by covalent or hydrophobic attachment of charged or polymeric groups at the bilayer surface. [From Fielding (1991), with permission.]

or clinically (Gregoriadis 1988a; Lopez-Berestein & Fidler 1989; Ostro 1983). Following the founding a decade ago of several companies specifically investigating liposomes, work on liposome research and technology has been chanelled into more realistic goals. Some 17 years after liposomes were first injected into humans (Gregoriadis et al. 1974), the first injectable liposome-based product (Vestar's AmBisome® formulation of amphotericin B) has now been licenced in a number of countries. Other products are undergoing clinical trials and are expected to be licenced in the near future (table I).

1. Behaviour of Liposomes In Vivo Administration of liposomes into animals and humans has been carried out by every conceivable enteral and parenteral route (Gregoriadis I 988a). However, most of our knowledge concerning liposomal behaviour in vivo has been obtained by injecting a variety of formulations intravenously. This is probably because not only is it easier to monitor liposomes in the blood and their possible extra-

vasation and uptake by tissues, it is also the most important route for a range of therapeutic applications. The oral route, which is more convenient for the patient, is problematic since many liposome formulations are rapidly destabilised in the gut, following interaction with bile salts. Further, it is likely that absorption of orally administered liposomes and other microspheres will be low or modest and also variable. Intravenous injection of liposomes is normally followed by their interaction with at least two distinct groups of plasma proteins, probably simultaneously (Gregoriadis 1988b). These are (a) the so-called opsonin which, by adsorbing onto the surface of vesicles, mediates their endocytosis by the fixed macrophages of the reticuloendothelial system (RES) and circulating monocytes; and (b) high density lipoproteins (HDL). The latter remove phospholipid molecules from the vesicle bilayers, leading to varying degrees of vesicle disintegration and release of encapsulated solutes at rates dependent on the extent of bilayer damage. Destabilised liposomes and solutes still entrapped are

Drugs 45 (1) 1993

18

intercepted by the RES, presumably through the opsonin on the bilayer surface. Rates of uptake by the RES are influenced by a number of parameters including vesicle size, surface charge and bilayer fluidity. The influence of bilayer fluidity on the rate of liposomal clearance from the circulation first became apparent during attempts to reduce or abolish the detrimental effect of HDL on vesicle structural integrity. It was observed that packing the relatively loose bilayers made of egg phosphatidylcholine by the addition of cholesterol (Gregoriadis & Davis 1979; Kirbyet al. 1980), or by replacing some or all of the phosphatidylcholine with phospholipids that exhibit Tc values higher than 37"C, led to a drastic reduction of phospholipid removal by HDL and hence in ensuing solute leakage both in vitro and in vivo (Gregoriadis 1988b). It turned out that the more liposomes were stable in the presence of blood in vitro at 37°C, the greater was their residence time in the circulation, presumably because of reduced opsonin adsorption on the bilayer surface. Thus, for small unilamellar vesicles (SUV) composed of distearoyl-phosphatidylcholine (DSPC) and equimolar cholesterol, elimination half-lives of up to 20 hours were attained in mice (Senior & Gregoriadis 1982). However, the effect of lipid composition on vesicle half-life is not as substantial with large liposomes, since vesicle size overrides bilayer fluidity in determining uptake by the RES (Senior et aI. 1985).

Prolonged residence of liposomes in the circulation is required when these are designed to act on non-RES tissues within the vascular system, extravascularly through leaky capillaries, or as circulating drug reservoirs. All such functions would be optimal with long-lived small liposomes, especially if the lipid to drug mass ratio can be reduced by using lipid-drug conjugates, or by using newly developed techniques that ensure substantial passive drug entrapment in the aqueous phase (Gregoriadis et al. 1990). There are situations, however, where large liposomes are required for the transport of water-soluble, large molecular weight agents (e.g. haemoglobin in the case of haemosones). Recent work has shown that a hydrophilic vesicle surface achieved by the use of polyethyleneglycol and other hydrophilic molecules substantially prolongs the half-life of liposomes with an average diameter of up to 100nm (Blume & Cevc 1990; Gabizon & Papahadjopoulos 1992; Klibanov et al. 1990; Senior et al. 1991). Although such developments are encouraging, it is hoped that it will be possible to extend the half-lives of liposomes of much larger size. Regardless of the residence time of liposomes within the vascular system, much of an injected dose is taken up via endocytosis by RES cells to end up in the lysosomal apparatus. It has long been established that liposomes are disrupted within the lysosomes, and the released drugs act either locally or, after diffusion outside the organelles, in other

Table I. Some of the liposome-based drug formulations undergoing clinical trials Drug

Indication

Company

Nystatin Muramyltripeptide Salbutamol aerosol 'Stealth'doxorubicin Amphotericin B Doxorubicin Gentamicin Amphotericin Ba Doxorubicin 1111ndium

Fungal infections Activation of tumouricidal macrophages Asthma Kaposi's sarcoma Fungal infections Cancer Intracellular mycoplasma Fungal infections Kaposi's sarcoma Imaging

Argus Ciba-Geigy Liposome Technology Inc. Liposome Technology Inc. The Liposome Company The Liposome Company The Liposome Company Vestar Vestar Vestar

a

Already licensed.

19

Liposomes in Drug Delivery

cell compartments (Gregoriadis 1976). It has been shown that lysosomal localisation in vitro can be avoided with liposomal contents being introduced into the cytoplasm by employing vesicles that can fuse with the endocytic vacuoles following endocytosis (Connor & Huang 1985). It remains to be seen whether this alternative means of gaining access into the cell's interior can be applied in vivo. Because of their size (about 25nm minimum diameter), intravenously injected liposomes cannot undergo transcapillary passage except in areas where vessels become leaky (e.g. inflamed tissue) . Although larger liposomes localise mostly in the liver and spleen macrophages, smaller vesicles « 1OOnm diameter), can reach the hepatic parenchymal cells through the fenestrae by which they are endocytosed. There is also extensive uptake of small liposomes with long half-lives by bone marrow macrophages (Gregoriadis 1988b). Observations on stability, clearance and tissue distribution as discussed above, are also valid for liposomes given by other parenteral routes, for instance intraperitoneally, subcutaneously and intramuscularly (Gregoriadis 1988b). A proPQrtion of liposomes, determined by vesicle size, composition and route of injection, enters the lymphatic and eventually the blood circulation where vesicles behave as if given intravenously (Turner et al. 1983). Whereas, however, liver, spleen and bone marrow take up nearly all liposomes given intravenously, these tissues account for a smaller proportion of the dose given by other routes. The remainder (up to about 80% of liposomes injected subcutaneously or intramuscularly) is either detained at the site of injection and attacked by infiltrating macro phages and other factors, or intercepted by the lymph nodes draining the injection site. Indeed, relative to their size, lymph nodes take up a much greater (over 100-fold) proportion than any other RES tissue (Gregoriadis 1988b). Efforts to ascertain whether liposomes given orally facilitate absorption of drugs in the gastrointestinal tract have given mixed results. Although agents such as insulin, factor VIII, anticoagulants, tubocurarine and lipid-soluble vitamins administered via liposomes do reach the blood circula-

tion, their absorption is minimal and unpredictable. It is nevertheless apparent that liposomes of a lipid composition that renders them resistant to attack by bile salts or phospholipases survive long enough to facilitate absorption of some of their contents, probably through the lymphatics (Gregoriadis 1988b).

3. Clinical Applications of Liposomes An obvious major consideration in the use of liposomes in drug delivery is that a given liposome-drug formulation should exhibit clear benefits over and above those seen with the conventional formulation of the drug. Encouraging results with liposomal drugs in the treatment or prevention of a wide spectrum of diseases in experimental animals and in humans indicate that liposomebased products for clinical and veterinary applications may be forthcoming (Gregoriadis 1988a, 1991; Lopez-Berestein & Fidler 1989) [table I]. These could include treatment of skin and eye diseases, antimicrobial and anticancer therapy, metal chelation, enzyme and hormone replacement therapy, vaccines and diagnostic imaging. Major modes of liposomal action in mediating effective drug delivery are illustrated in table II. Below, we discuss some of the applications with realistic prospects of development for clinical use. 3.1 Cancer Chemotherapy There is substantial evidence that, in experimental animals at least, drug concentrations (administered intravenously via small liposomes) in certain tumours are higher than in neighbouring normal tissues (Gregoriadis 1991; Gregoriadis & Florence 1991). One or more of the following factors may account for this: (a) higher endocytic activity of some tumour cells combined with augmented local permeability of capillaries allowing the passage of small liposomes; (b) diffusion of drugs from liposomes either during circulation or after they have been lodged in tissues near tumours, followed by preferential drug entry into the tumour mass; (c) engulfment of liposomes by cir-

Drugs 45 (1) 1993

20

Table II. Major modes of liposomal action and related applications Mode of action

Application

Intracellular uptake (Iysosomes, endosomes/ cytoplasm)

Microbial diseases Metal storage diseases Gene manipulation Uptake by some tumour cells Macrophage activation to a tumouricidal/microbiocidal state Efficient antigen presentation by antigen-presenting cells (vaccines) Tumours near fixed macrophages Cardiotoxicity of doxorubicin Blood surrogates Drug delivery to skin, lungs, eyes, mucosal tissues

Slow release of drugs near the target area Avoidance of tissue sensitive to drugs Circulating reservoirs Facilitation of drug uptake by certain routes

culating monocytes and migration of the latter to tumours. However, the case for using liposomes in cancer chemotherapy appears stronger when the aim is to reduce toxicity while maintaining a cytotoxic effect. Thus, experiments with anthracycline cytostatic drugs entrapped in liposomes have clearly shown reduced cardiotoxicity and dermal toxicity, and better survival of experimental animals compared with controls receiving the free drug (Gregoriadis 1991). Such beneficial effects of liposomal anthracyclines have been observed with a variety of liposomal formulations regardless of their lipid composition provided, however, that lipids used (e.g. increased cholesterol content or the use of phospholipids with a high Tc) are conducive to drug retention by the vesicles in the presence of blood (Mayer et al. 1990). An additional structural characteristic of liposomes that appears to favour increased efficacy of the encapsulated cytostatic drug doxorubicin is small vesicle size. For instance, liposomes with an average diameter of 100nm exhibited a greatly improved cytotoxic effect presumably because of augmented extravasation into tumours (Mayer et al. 1989). These and a variety of earlier data from animal work (for reviews see Gregoriadis 1988a, 1993b) by a number of laboratories (e.g. Forssen & Tokes 1981; Gabizon et al. 1982; Rahman et al. 1980) have culminated in liposome-entrapped anthracyclines being used in clinical trials (Janoff 1992; Rahman 1990; Rahman et al. 1990). Trials have

been facilitated by proprietary liposomal doxorubicin formulations that are stable and provide high drug to lipid ratios (Mayer et al. 1990). In phase II studies, for instance, doxorubicin-containing liposomes have exhibited substantial activity in metastatic breast cancer with no apparent cardiotoxicity (Rahman et al. 1993). Moreover, application of liposomal doxorubicin in other forms of cancer in which use of the free drug has been limited beqj,use of toxicity is now under consideration (Janoff 1992). 3.2 Antimicrobial Therapy Use of liposome-entrapped drugs against microbial diseases was necessitated, in part, because of the inability of otherwise potent agents to reach infected intracellular sites effectively (Gregoriadis 1991). As already discussed, liposomes are able to localise in the liver and spleen, especially the RES components, where many pathogenic microorganisms reside. There is much evidence (for review see Gregoriadis 1988a) that liposome formulations are superior to free antimicrobial agents both in terms of distribution to the relevant intracellular sites and therapeutic efficacy. For instance, current treatment of immunocompromised patients (e.g. those with AIDS) infected with microorganisms of the Mycobacterium avium complex (MAC) is ineffective probably because the organism resides intracellularly, mostly in monocytes. Entrapment of antibiotics into liposomes offers a means to in-

21

Liposomes in Drug Delivery

troduce drugs such as aminoglycosides to areas of infection. To this end, effective treatment of MAC in mice was achieved with liposomal amikacin .(Duzgunes et al. 1988). Similarly encouraging results were previously obtained with liposomal ampicillin in the treatment of Listeria monocytogenes in mice (Bakker-Woudenberg et al. 1985), liposomal cefalothin against Salmonella typhimurium (Desiderio & Campbell 1983), liposomal benzylpenicillin in Staphylococcus aureus infection in mice (Barsoum & Reich 1982), and, in earlier work, with liposomal dihydrostreptomycin against intracellular S. aureus (Bonventre & Gregoriadis 1978). More recent studies (Swenson et al. 1990) with liposomal gentamicin in rodents have extended previous work with the same drug (Dees et al. 1985) to demonstrate therapeutic advantages in disseminated infections. Indeed, liposomal gentamicin has now received orphan drug status from the Food and Drug Administration in the USA. Current phase II clinical trials for the treatment of MAC in AIDS patients indicate that liposomal gentamicin is both well-tolerated and active (Janoff 1992). Another group of microbial diseases, severe disseminated fungal infections, has benefited substantially from the use of liposomes as a drug delivery system. The drug of choice in these infections is amphotericin B, but its dose-dependent toxicities have limited its potential as a curative agent. Extensive in vivo work during the last decade has shown that a number ofliposomal formulations of amphotericin B retain the therapeutic advantage of the drug while at the same time reducing its toxicity (Lopez-Berestein et al. 1983; New et al. 1981; Taylor et al. 1982). These animal studies were subsequently extended to the clinic with results encouraging enough to attract the attention of industries specialising in liposomes (e.g. Lopez-Berestein et al. 1987). Work on the technology of amphotericin B liposomal formulations revealed that several structurally distinct entities were possible. For instance, high amphotericin B : lipid ratios produced unusual structures termed 'ribbons' (Janoff et al. 1988) arising from amphotericin B-dependent disruption of the bilayer (Janoff 1992). Release of the active drug

monomer from the structures in vivo was found to occur only in the presence of fungal lipases (Perkins et al. 1992). As a result, 'ribbons' were highly effective in alleviating the toxicity of the drug without compromising its efficacy (Janoff 1992; Kan et al. 1991). Conventionalliposomal amphotericin B preparations made of phospholipid and cholesterol, the latter serving to promote the stability of the bilayer, were also capable of reducing drug toxicity in clinical studies (Davidson et al. 1991; Vincent et al. 1992). Recent multicentre studies with AmBisome® (small unilamellar liposomes composed of phosphatidylcholine, cholesterol, distearoyl-phosphatidylglycerol and amphotericin B) [Proffitt et al. 1991] have shown that augmented concentrations of the drug in the blood and tissues of mice and rats could be achieved with nontoxic doses (Proffitt et al. 1991). Using the same formulation, Adler-Moore et al. (1991) were able to show in preclinical work superior efficacy of AmBisome® against murine candidosis and cryptococcosis when compared with conventional amphotericin B. A large number of clinical trials have now confirmed the effectiveness of AmBisome® as a safe alternative of the conventional formulations of amphotericin B in a majority of patients with invasive or superficial fungal infections (e.g. Chopra et al. 1991; Meunier et al. 1991; Ringden et al. 1991). 3.3 Liposomes as Immunological Adjuvants in Vaccines Aluminium salts are presently the only immunological adjuvant licensed for use in vaccines for humans. The emergence of new-generation recombinant subunit and synthetic peptide vaccines has renewed demands for novel adjuvants that are both acceptable and effective in potentiating humoral and cell-mediated immunity. It has been nearly 20 years since liposomes were first shown to improve immune responses to entrapped diphtheria toxoid (Allison & Gregoriadis 1974). Liposomes are now one of the few major nontoxic candidate adjuvants for human use currently under investigation with a wide range of

Drugs 45 (1) 1993

22

bacterial, viral, protozoan, tumour and other antigens. In immunisation experiments, a variety of liposomal antigens administered parenterally or enterally were effective in protecting animals against disease (Gregoriadis 1990), and at least one clinical trial is now in progress (Fries et al. 1992). Extensive studies (reviewed by Gregoriadis 1990) on the adjuvant use of liposomes have shown that a prerequisite for effectiveness is physical association of the antigen with the vesicles. Further, available evidence (Gregoriadis 1990) indicates that liposomal adjuvanticity is an attribute of the vesicular structure of the system and, most likely, its lipoid nature. Information on the fate of liposomes in vivo (as briefly outlined above) suggests that humoral immunity to antigens incorporated in liposomes can be attributed to the function of the latter as an antigen depot, supplying antigen-presenting cells (APC) with free or liposome-associated antigen or both. Supply of antigen to the cells by formulations which have proven effective presumably occurs at rates or in a form leading to its efficient presentation. A number of structural variables of liposomes, for instance vesicle size, bilayer fluidity, surface charge as well as the mode of antigen localisation within the vesicles and the lipid to antigen mass ratio, all influence adjuvanticity, which however does not appear to depend crucially on any specific formulation. Indeed, it may be that optimal characteristics of the system in terms of immunoadjuvant action vary depending on the antigen used and its route of administration (e.g. Gregoriadis et al. 1992a). Interestingly, liposomal adjuvanticity can be improved by approaches which lead to its amplification. These include administration of liposomal antigens together with other adjuvants such as lipid A (Alving et al. 1986), muramyldipeptide and its lipophilic derivatives (Kersten et al. 1988), saponins (Manesis et al. 1979) and interleukin-2 (Tan & Gregoriadis 1989), and the receptor-mediated targeting of liposomal antigens to APC (Garcon et al. 1988). Considerable progress has been made recently in the study ofliposomal adjuvanticity in vitro. For instance, the role that a variety of APC play in the

processing of liposome-entrapped antigens (Szoka 1992), the significance of antigen localisation on the liposomal surface (Szoka 1992), the possible cooperation of APC in terms of antigen presentation (Harding et al. 1992), and class I major histocompatibility complex-mediated presentation of liposomal antigens (Huang et al. 1992). Such studies are expected to help in the elucidation of liposomal adjuvanticity but, as discussed elsewhere (Gregoriadis 1992a), translation of in vitro observations into ways of tailoring adjuvanticity in vivo is likely to be met with great difficulties. 3.4 Diagnostic Imaging Liposomes containing contrast agents have been used in diagnostic imaging, including computed tomography (CT), magnetic resonance imaging (MRI) and radio nuclide imaging. Early work by a number of laboratories (e.g. Caride et al. 1982; Seltzer et al. 1981) suggested that administration of contrast agents, such as sodium amidotrizoate and brominated lecithin, via liposomes of varying structural characteristics led to more effective imaging of target tissues. Further improvements in the formulations, in terms of entrapping maximum amounts of contrast material, were made by incorporating a positively charged lipid in the liposomal bilayer (Seltzer et al. 1984), by using large unilamellar vesicles (Ryan et al. 1984) or by entrapping agents by the dehydration-rehydration method (Seltzer et al. 1988). Contrast agents administered with lipososes increase the attenuation coefficient of the liver and spleen during CT imaging (White et al. 1990). The lack of tissue-specific biodistribution of paramagnetic agents used in MRI has prompted the testing of liposomes (among other systems) as agent carriers. Caride et al. (1984) for instance, employed chelates of manganese (e.g. Mn-DTPA) in positively charged multi lamellar liposomes. Following injection of this formulation, accumulation of the chelate in the liver and spleen was drastically improved. The same authors also observed that the structural integrity of liposomes and their cellular or subcellular localisation were important in

23

Liposomes in Drug Delivery

changing tissue relaxation rates. Furthermore, by forming amphipathic gadolinium-DTPA complexes, Kabalka and colleagues (1991) were able to incorporate gadolinium as a paramagnetic contrast agent into SUVs; similar approaches have allowed gadolinium-enhanced imaging of liver and spleen (Karlik et al. 1991; Schwendener et al. 1990). Distribution of radioactive agents within tissues forms the basis of radionuclide images. One of the advantages of this approach to imaging is the availability of radioactively labelled agents that can be tissue- or even cell-specific. At the tissue level, images of the RES obtained with radioactive tracercontaining colloids and liposomes have been found useful in this respect (e.g. Richardson et al. 1978). Other workers have succeeded in producing 1iposomes that accumulate in non-RES tissues or in lesions, for instance myocardial infarcts, lymph nodes, tumours and abscesses (for a review see Seltzer 1988). Thus, bone marrow imaging with 99mTc-liposomes in patients with malignant lymphoma has now been used to measure bone marrow suppression during chemotherapy and radiotherapy (Yu 1990). It has been suggested that liposomal Patent Blue V which produces contrast between retroperitoneal lymph nodes and surrounding fatty tissue after endolymphatic injection might provide the selectivity of excision in lymphnodectomy (Hirnle 1991). Kaposi's sarcoma and malignant lymphoma have also been successfully imaged with 111 In-labelled liposomes in two patients with AIDS (Presant et al. 1990). Discrete lesions along the left foot and calf were detected and neoplastic lymph nodes selectively ima~ed. Selective tissue imaging is likely to benefit from the use of small liposomes which can be designed to exhibit long half-lives in the circulation (Gregoriadis 1988b). This was demonstrated in early work (Gregoriadis 1980; Gregoriadis et al. 1977) in tumour-bearing mice injected with 111 In-bleomycin-containing liposomes and, more recently, with glycolipid-incorporating small liposomes possessing a long half-life in a variety of mouse tumours (Gabizon et al. 1990). Data from such studies give clues as to the future diagnostic uses of liposomes. However, caution has to be exercised since the ap-

parent selectivity is often only apparent because of the ability of imaging to reduce background readings and produce foci of high intensity. 3.5 Liposomes for Ocular Delivery of Drugs Drugs are normally applied to the eye as solutions (eyedrops) and more rarely as suspensions or as ointments. Much is now known about the pharmacokinetics and disposition of drugs applied to the cornea (Burnstein 1989). The loss of drug through the onset of lacrimation, and consequent reduction in the drug diffusion gradient and also the loss of excess fluid either by tear flow or by drainage through the punctae into the nasolacrimal duct must be overcome for optimal delivery. Reducing the size of the instilled droplet can improve response, as can increasing the viscosity of the formulation, for example by using hydrophillic polymers to augment the duration of action of applied drugs. There are, however, limits to this latter approach, since highly viscous materials cause blurring of vision and the eyelids to stick. The use of reservoirs such as ocular inserts of gelatin or other hydrogel materials, or polymeric devices (e.g. the Ocusert® device), is potentially useful but involves the insertion of a foreign object into the conjunctival sac. Liposomal formulations have some advantages in providing a reservoir for slow release of drug in a nonviscous medium (Mezei 1988). Nearly all reports to date refer to animal experiments. Smolin et al. (1981) demonstrated improved efficacy of liposomal versus conventional idoxuridine in the treatment of herpes simplex keratitis. While it has been reported (Schaeffer & Krohn 1982) that encapsulation of both lipophilic and hydrophilic drugs in liposomes increases corneal penetration, others have not always corroborated this finding. Singh and Mezei used liposomes to deliver both triamcinolone acetonide (Singh & Mezei 1983) and dihydrostreptomycin sulphate (Singh & Mezei 1984) in large multilamellar liposomes. They observed significantly higher concentrations (> 2-fold) of triamcinolone in ocular tissue compared with a suspension, for up to 5 hours after administration,

24

but liposomal encapsulation reduced dihydrostreptomycin penetration, leading the authors to state that for hydrophilic drugs the procedure provided 'no advantages as far as drug delivery is concerned'. Both size and charge of the liposomes determine outcome. Larger multilamellar and unilamellar vesicles produced better results than small unilamellar Iiposomes (Singh & Mezei 1984), presumably because of slower loss via the punctae, although very large vesicles would be required to completely block exit by this route. The literature is not clear on the value of Iiposomal delivery to the corneal surface. Schaefer and Krohn (1982) reported that the ocular bioavailability of benzylpenicillin and indoxole were both enhanced by delivery in Iiposomes. However, Benita et al. (1984) applied pilocarpine 0.2% in liposomes and compared its effectiveness with 1 and 2% solutions of pilocarpine, but came to no clear conclusion as to the effectiveness of the approach. Stratford et al. (1983) observed a 50% reduction in adrenaline (epinephrine) absorption but a greatly enhanced absorption of the water soluble inulin. Lee et al. (1984, 1985), also using inulin as a model, concluded that liposomes were useful provided they had an affinity for corneal surfaces, binding for sufficient periods to optimise the release of the encapsulated drug. Liposome interaction with the corneal surface has been investigated in some detail (Guo et al. 1989/1990; Schaeffer & Krohn 1982), but a clear pattern has not emerged, as the outcome is dependent on the nature of the drug carried. The degree of interaction of vesicles with the corneal surface was in the following order: positively charged multi lamellar vesicles (ML V) > positively charged SUV > negatively charged SUV > MLV and SUV. However, the order of benzylpenicillin transport across the cornea was in the order of positively charged SUV > negatively charged MLV > positively charged MLV > negatively charged SUV > MLV. This discrepancy is perhaps not surprising because several processes are in operation. Diffusion of drug from the vesicles (whether bound or not), binding of the vesicle to the corneal surface, and stability of the vesicle all

Drugs 45 (l) 1993

play their part, so it is not yet possible to unravel the various effects. In addition, the noncorneal route of access for hydrophilic compounds can be significant, entry to the anterior segment being via the conjuctival-scleral laminate (Ahmed & Patton 1985). Guo et al. (1989/1990) confirmed the importance of positive charge on corneal retention of Iiposomes, presumably as a result of association with the polyanionic corneal and conjunctival mucoglycoproteins. The corneal surface is saturable with Iiposomes, the half-life of clearance being 2 hours. This was confirmed by McCalden and Levy (1990) with liposomes containing benzyl-dimethylstearyl ammonium chloride, but was extended to 4 hours with liposomes containing dimethyl-dioctyldecyl ammonium chloride. 3.5.1 Targeting Liposomes to the Corneal Surface This was first pursued by Megaw et al. (1981) using lectins, an approach used later by Schaefer and Krohn (1982). Immunoliposomes bearing antibody against cell surface viral glycoproteins have been suggested as targeting carriers in the treatment of ocular herpetic keratitis (Norley et al. 1986). Idoxuridine and aciclovir were entrapped in phosphatidyIcholine vesicles, which however did not sufficiently delay the release of idoxuridine, 80% being released within the first hour, therefore providing no biopharmaceutical advantage. In contrast, in vitro studies showed that there was therapeutic potential when aciclovir-containing immunoliposomes, applied around 5 hours postinfection, bound specifically to infected cells and inhibited viral development, whereas free drug and unmodified vesicles containing the drug were ineffective. 3.5.2 Subconjunctival and 1ntravitreal1njection Liposomes have been used as vehicles for subconjunctival and intravitreal injection of both cytotoxic drugs and antibiotics. Because of a substantial reduction in the retinal toxicity of cytarabine in liposomal form, it has been suggested that this combination offers promise in the treatment of ocular proliferative disorders (Liu et al. 1989)

Liposomes in Drug Delivery

as an alternative to fluorouracil, which is intrinsically less toxic. Liposomal fluorouracil produced significantly higher concentrations of drug in the vitreous humour after bilateral intravitreal injection in rabbit eyes and to a lesser extent after subconjunctival injection when compared to drug injected in phosphate-buffered saline (Fishnan et al. 1989). 48 hours after intravitreal injection, drug concentrations were approximately 580 mg/L compared with a little over 1 mgfL for the saline formulation. This illustrates the value of liposomal delivery into confined body spaces where the vesicles act as a reservoir, although a distinction should be made between free and total drug concentrations, not yet possible analytically. Intravitreal clindamycin-containing liposomes have been evaluated in staphylococcal endophthalmitis in the rabbit (Rao et al. 1989). However, the controls were drug-free liposomes and 'no treatment', so it is not possible from this study to evaluate the effect of encapsulation of the drug. A study of liposomal amphotericin B involved a comparison with free drug as well as drug-free liposomes (Liu et al. 1989). Reduced amphotericin B toxicity was associated with a reduced efficacy against Candida albicans. Infection was eradicated with Img free drug and 20ILg of liposomal drug. Ganciclovir and trifluridine have been administered in liposomal formulations intravitreally to rabbits (peyman et al. 1989). Prolonged effective concentrations of both drugs were found up to 14 days after injection, while there was no evidence of gross retinal toxicity. Liposomal 5-fluoroorotate has also been administered intravitreally in glaucoma filtration surgery to promote wound healing (Alvarado 1990). Use of a laser-pulse to induce release of encapsulated dye from liposomes to measure circulation time in the retinal vasculature provides an insight into possible uses of selective release of liposomal drugs for therapeutic indications (Khoobehi et al. 1989, 1990). Laser light was delivered through the pupil and the heat pulse lysed the liposomes in the target vessel. Such responsive systems, particularly if the process was repeatable, could have therapeutic implications (Khoobehi et al. 1988). Selective

25

uptake into locally heated eyes following systemic administration of cytarabine and 5-fluorouridine in vesicles was indicated by the fact that samples of aqueous humour and vitreous showed significantly higher concentrations of both drugs in the microwaved hyperthermic eye.

4. Conclusions 22 years of research into the use of liposomes in drug delivery have led to vastly improved technology in terms of drug capture, vesicle stability on storage, scaled-up production and the design of formulations for specialised tasks (Gregoriadis 1993b). In parallel, remarkable advances have been made in understanding and controlling liposomal behaviour in vivo. This has facilitated the application of a wide range of liposomal drugs in the treatment and prevention of disease in experimental animals and clinically. Severalliposomal preparations (one of them injectable) have already been licensed and a number of others are likely to follow soon. The future ofliposomes in drug delivery systems appears to be secure. They will no doubt continue to contribute significantly to more efficient use of 'old' drugs and also find applications with agents now produced by recombinant DNA technology.

References Adler-Moore JP, Chiang S-M, Satorius A, Guerra D. McAndrews B, et al. Treatment of murine candidosis and cryptococcosis with a unilamellar liposomal amphotericin B formulation (AmBisome). Journal of Antimicrobial Therapy 28 (Suppl. B): 63-71, 1991 Ahmed I, Patton TF. Importance of the non-corneal absorption route in topical ophthalmic drug delivery. Investigative Ophthalmology and Visual Science 26: 584-587, 1985 Allison AC, Gregoriadis G. Liposomes as immunological adjuvants. Nature 252: 252, 1974 Alvarado JA. The use of a liposome encapsulated 5-fluoro-orotate for glaucoma surgery. I. Animal studies. Transactions of the American Ophthalmology Society 87: 489-514, 1990 Alving CR, Richards RL, Moss J, Alving LI, Clements JD, et al. Effectiveness ofliposomes as potential carriers of vaccines: applications to cholera toxin and human malaria antigen. Vaccine 4: 166-172, 1986 Bakker-Woudenberg IAJM, Lookerse AF, Roerdink, F, Regts D, Michel MF. Free versus liposome-entrapped ampicillin in treatment of infection due to Listeria monocytogenes in normal and athymic nude mice. Journal of Infectious Diseases 151: 917-924, 1985

26

Bangham AD, Standish MM, Watkins Jc. Diffusion of univalent ions across the lamellae of swollen phospholipids. Journal of Molecular Biology 13: 238, 1965 Barsoum IS, Reich M. The effect of liposome-entrapped penicillin G on Staph. aureus infection in mice. Pharmacology 10: 358, 1982 Benita S, Plenecassagne JD, Caves G, Drouin D, Dong PLH, et a!. Pilocarpine hydrochloride liposomes: characterization in vitro and preliminary evaluation in vivo in rabbit eye. Journal of Microencapsulation I: 203-216, 1984 Blume G, Cevc G. Liposomes for sustained drug release in vivo. Biochimica et Biophysica Acta 1029: 91-97, 1990 Bonventre P, Gregoriadis G. Killing of intraphagocytic Staph. aureus by dihydrostreptomycin entrapped in liposomes. Antimicrobial Agents and Chemotherapy 13: 1049-1051, 1978 Burnstein NL. Basic science of ocular pharmacology. In Bartlett & Joanus (Eds) Clinical ocular pharmacology, 2nd edn, Butterworths, Boston, 1989 Caride VJ, Sostman HD, Twickler J, Zacharis H, Orphanoudakis SC, et a!. Brominated radiopague liposomes: contrast agent for computed tomography of liver and spleen. A preliminary report. Investigative Radiology 17: 381-385, 1982 Caride VJ, Sostman RD, Winchell RJ, Gore Jc. Relaxation enhancement using Iiposomes carrying paramagnetic species. Magnetic Resonance Imaging 2: 107-112, 1984 Chopra R, Blair S, Strong J, Cervi P, Patterson KG, et a!. Liposomal amphotericin B ('AmBisome') in the treatment of fungal infections in neutropenic patients. Journal of Antimicrobial Chemotherapy 28 (Suppl. B): 93-104, 1991 Connor J, Huang L. Efficient cytoplasmic delivery of a fluorescent dye by pH-sensitive imnuno Iiposomes. Journal of Cell Biology 101: 582-589, 1985 Davidson RN, Croft SL, Scott A, Maini M, Moody AH, et a!. Liposomal amphotericin in drug resistant visceral leishmaniasis. Lancet 337: 1061-1062, 199-1 Dees C, Fountain MW, Taylor JR, Schultz RD. Enhanced intraphagocytic killing of Brucella abortus in bovine mononuclear cells by Iiposomes containing gentamycin. Vetinary Immunology Immunopathology 8: 171-182,1985 Desiderio JV, Campbell SG. Intraphagocytic killing of Salmonella typhimurium by liposome-encapsulated cephalothin. Journal of Infectious Disease 148: 563-570, 1983 Duzgunes N, Perumal VK, Kesavalu L, Goldstein JA, Debs RJ, et a!. Enhanced effect of liposome-encapsulated amikacin on Mycobacterium avium-M. intracellular complex infection in beige mice. Antimicrobial Agents and Chemotherapy 32: 14041411, 1988 Fielding RM. Liposomal drug delivery: advantages and limitations from a clinical pharmacokinetic and therapeutic perspective. Clinical Pharmacokinetics 21: 155-164, 1991 Fishman P, Peyman GA, Hendricks R, Hui SL. Liposome-encapsulated 3H-5-FU in rabbits. International Ophthalmology 13: 361-365, 1989 Aorence AT, Gregoriadis G. Targeting of peptides and proteins. Trends in Biotechnology 9: 295-297, 1991 Forssen E, Tokes Z. Use of amionic liposomes for the reduction of chronic doxorubicin-induced cardio-toxicity. Proceedings of the National Academy of Science 78: 1873-1877, 1981 Fries LF, Gordon DM, Richards RL, Egan JE, Hollingdale MR, et a!. Liposomal malaria vaccine in humans: a safe and potent adjuvant strategy. Proceedings of the National Academy of Science 89: 358-362, 1992 Gabizon A, Dagam A, Goren D, Barenholz Y, Fuks Z. Liposomes as in vivo carrier of adriamycin: reduced cardiac uptake and preserved antitumour activity in mice. Cancer Research 42: 4734-4739, 1982 Gabizon A, Papahadjopoulos D. The role of surface charge and hydrophilic groups on liposome clearance in vivo. Biochimica et Biophysica Acta 1103: 94-100, 1992

Drugs 45 (1) 1993

Gabizon A, Price DC, Huberty J, Breaslier RS, Papahadjopoulos D. Effect of liposome composition and other factors on the targeting of Iiposomes to experimental tumours: biodistribution and imaging studies. Cancer Research 50: 6371-6378, 1990 Garcon N, Gregoriadis G, Taylor M, Summerfield J. Targeted immunoadjuvant action of tetanus toxoid-containing liposomes coated with mannosylated albumin. Immunology 64: 743-745, 1988 Gregoriadis G. The carrier potential of liposomes in biology and medicine. New England Journal of Medicine 295: 704-710,1976 Gregoriadis G. The liposome drug carrier concept: its development and future. In Gregoriadis & Allison (Eds) Liposomes in biological systems, Academic Press, London, 1980 Gregoriadis G (Ed.). Liposomes as drug carriers: recent trends and progress, John Wiley, Chichester, 1988a Gregoriadis G. Fate of injected liposomes: observations on entrapped solute retention, vesicle clearance and tissue distribution in vivo. In Gregoriadis (Ed.) Liposomes as drug carriers: recent trends and progress, John Wiley, Chichester, 1988b Gregoriadis G. Immunological adjuvants: a role for liposomes. Immunology Today II: 89-97,1990 Gregoriadis G. Overview of liposomes. Journal of Antimicrobial Chemotherapy 28 (Suppl. B): 39-48, 1991 Gregoriadis G (Ed). Liposome Technology, 2nd ed., CRC Press Inc, Boca Raton, 1993a Gregoriadis G. Liposomes as immunological adjuvants: approaches to immunopotentiation including ligand-mediated targeting to macrophages. Research in Immunology 143: 178185, 245-246, 1992a Gregoriadis G, Davis C. Stability of liposomes in vivo and in vitro is promoted by their cholesterol content and the presence of blood cells. Biochemical and Biophysical Research Communications 89: 1287-1293, 1979 Gregoriadis G, Ryman B. Lysosomal localization of J3-fructofuranosidase-containing Iiposomes injected into rats. Some implications in the treatment of genetic disorders. Biochemical Journal 129: 123-133, 1972 Gregoriadis G, Aorence AT. Liposomes and cancer therapy. Cancer Cells 4: 144-146, 1991 Gregoriadis G, Swain CP, Wills EJ, Tavill AS. Drug carrier potential of Iiposomes in cancer chemotherapy. Lancet I: 13131316, 1974 Gregoriadis G, Neerunjun DE, Hunt R. Fate of Iiposome-associated agent injected into normal and tumour-bearing rodents. Attempts to improve localization in tissues. Life Sciences 21: 357-370, 1977 Gregoriadis G, da Silva H, Aorence AT. A procedure for the efficient entrapment of drugs in dehydration-rehydration Iiposomes (DRY). International Journal of Pharmaceutics 65: 235242, 1990 Gregoriadis G, Tan L, Ahmeida ETCS, Jennings R. Liposomes enhance the immunogenicity of reconstituted influenza virus A/PR/8 envelopes and the formation of protective antibody by influenza virus A/Sichuan/87 surface antiges. Vaccine 10: 747-753, 1992b Gregoriadis G, Aorence AT, Patel HM (Eds). Liposomes in drug delivery, Harwood, Chur, 1993b Guo LSS, Radhakrishnan R, Redmann CT. Adhesion of positively charged liposomes to mucosal tissues. Journal of Liposome Research 3: 319-337, 1989/1990 Harding CV, Collins D, Unanne ER. Processing of liposome-encapsulated antigens targeted to specific subcellular compartments. Research in Immunology 143: 188-191, 1992 Hirnle P. Histological findings in rabbit lymph nodes after endolymphatic injection ofliposomes containing blue dye. Journal of Pharmacy and Pharmacology 43: 217-218, 1991 Huang L, Reddy R, Nair SK, Zhou F, Rouse B. Liposomal delivery of soluble protein antigens for class I MHC-mediated

Liposomes in Drug Delivery

antigen presentation. Research in Immunology 143: 192-196, 1992 Janoff AS. Lipids, liposomes, rational drug design. Journal of Laboratory Investigation 66: 655-658, 1992 Janoff AS, Boni LT, Popescu Me, Minchey SR, Cullis PR, et al. Unusual lipid structures selectively reduce the toxicity of Amphotericin B. Proceedings of the National Academy of Science 85: 6122-6126, 1988 Kabalka GW, Davis MA, Moss TH, Buonocore E, Hubner K, et al. Gadolinium-labelled liposomes containing various amphiphilic Gd-DTPA derivatives: targeted MRI contrast enhancement agents for the liver. Magnetic Resonance in Medicine 19: 406-415,1991 Kan VL, Bennett JE, Amantea MA, Smolskis MC, McManus E, et al. Comparative safety, tolerance and pharmacokinetics of amphotericin B lipid complex and amphotericin B desoxycholate in healthy male volunteers. Journal of Infectious Disease 164: 418-421,1991 Karlik S, Florio E, Grant CWo Comparative evaluation of two membrane-based liposomal MRI contrast agents. Magnetic Resonance in Medicine 19: 56-66, 1991 Kersten GFA, van de Put AM, Teerhlink T, Beuvery EC, Crommelin DJA. Infection and Immunity 56: 1661-1664, 1988 Khoobehi B, Peyman GA, McTurnan WG, Niesman MR, Megin RL, et al. Externally triggered release of dyes and drugs from liposomes into the eye. An in vitro and in vivo study. Ophthalmology 95: 950-955, 1988 Khoobehi B, Peyman GA, Niesman MR, Oncel M. Hyperthermia and temperature sensitive liposomes: selective delivery of drugs into the eye. Japanese Journal of Ophthalmology 33: 405-412, 1989 Khoobehi B, Schuele KM, Ali OM, Peyman GA. Measurement of circulation time in the retinal vasculature using selective angiography. Ophthalmology 97: 1061-1070, 1990 Kirby C, Clarke J, Gregoriadis G. Effect ofthe cholesterol content of small unilamellar liposomes on their stability in vitro and in vivo. Biochemical Journal 186: 591-598, 1980 Klibanov AL, Maruyama K, Torchilin VP, Huang L. Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Letters 268: 235-237, 1990 Lee VHL, Takemoto, KA, Timoto DA, Precorneal factors influencing the ocular distribution of topically applied liposomal inulin. Current Eye Research 3: 585-591, 1984 Liu KR, Peyman GA, Khoobehi B. Efficacy of liposome-bound amphotericin B for the treatment of experimental fungal endophthalmitis in rabbits. Investigative Ophthalmology and Visual Science 30: 1527-1534, 1989 Lee VHL, Urrea PT, Smith, RE, Schanzlin DJ. Ocular drug bioavailability from topically applied liposomes. Survey of Opthalmology 29: 335-348, 1985 Lopez-Berestein G, Bodey GP, Fronkel LS, Mehta K. Treatment ofhepatosplenic candidiasis with liposomal amphotericin British Journal of Clinical Oncology 5: 310-317, 1987 Lopez-Berestein G, Fidler I (Eds). Liposomes in the therapy of infectious diseases and cancer, Alan Liss, New York, 1989 Lopez-Berestein G, Mehta R, Hopfer RL, Mill K, Kasi L, et al. Treatment and prophylaxis of disseminated infection due to Candida albicans in mice with liposomeencapsulated amphotericin. British Journal of Infectious Disease 147: 939-944, 1983 Manesis EK, Cameron CH, Gregoriadis G. Hepatitis B surface antigen-containing liposomes enhance humoural and cell-mediated immunity to the antigen. FEBS Letters 102: 107-111, 1979 Mayer LD, Tai LC, Ko DS, Masin D, Ginsberg RS, et al. Influence of vesicle size, lipid composition and drug to lipid ratio on the biological activity of liposomal doxorubicin in mice. Cancer Research 49: 5922-5930, 1989 Mayer LD, Tai LC, Bally MB, Mitilenes GN, Ginsberg RS, et al. Characterization of liposomal systems containing doxorubicin

27

entrapped in response to pH gradients. Biochimica et Biophysica Acta 1025: 143-151, 1990 McCalden T A, Levy M. Retention of topical liposomal formulations on the cornea. Experientia 46: 713-715, 1990 Megaw JM, Takei Y, Lerman S. Lectin-mediated binding of Ii posomes to the ocular lens. Experimental Eye Research 32: 395405, /981 Meunier F, Prentice HG, Ringden O. Liposomal amphotericin B (AmBisome): safety data form and phase II/III clinical trial. Journal of Antimicrobial Chemotherapy 28 (Suppl. B): 83-91, /991 Mezei M. Liposomes in topical application of drugs: a review. In Gregoriadis (Ed.) Liposomes as drug carriers: recent trends and progress, Wiley, Chichester, 1988 Ostro M (Ed.). Liposomes, Marcel Dekker Inc., New York, 1983 New RRC, Chance ML, Heath J. Antileishmanial activity ofamphotericin and other antifungal agents entrapped in liposomes. Journal of Antimicrobial Chemotherapy 8: 371-381, 1981 Norley SG, Huang L, Rouse BT. Targeting of drug-loaded immunoliposomes to herpes simplex virus-infected corneal cells: an effective means of inhibiting virus replication in vitro. Journal of Immunology 136: 681-685, 1986 Perkins R, Minchey SR, Boni LT, Swenson CE, Popescu MC, et al. Amphotericin B phospholipid interaction responsible for reduced mammalian cell toxicity, Biochimica et Biophysica Acta, in press, 1992 Peyman GA, Schulman JA, Khoobehi B, Alkan HM, Tawakol ME, et al. Toxicity and clearance of a combination of liposome-encapsulated ganciclovir and trifluridine. Retina 9: 232236, 1989 Presant CA, Blayney D, Proffitt RT, Turner AF, Williams LE, et al. Preliminary report: imaging of Kaposi sarcoma and lymphoma in AIDS with indium-III-labelled liposomes. Lancet 335: 1307-1309, 1990 Proffitt RT, Satorius A, Chiang S-M, Sullivan L, Adler-Moore JP. Pharmacology and toxicology ofa liposomal formulation of amphotericin B (AmBisome) in rodents. Journal of Antimicrobial Chemotherapy 28 (Suppl. B): 49-61, 1991 Rahman A. Antitumour activity of liposomal encapsulated doxorubicin in advanced breast cancer: phase II study. Journal of the National Cancer Institute 82: 1706, 1990a Rahman A, Kessler A, More N, Sikic R, Rowden G, et al. Liposomal protection of adriamycin-induced cardiomyopathy in mice. Cancer Research 40: 1532-1537, 1980 Rahman A, Treat J, Roe JK, Potkul LA, Alvord WG, et al. Phase I clinical trial and pharmacokinetic evaluation of liposomal encapsulated doxorubicin. Journal of Clinical Oncology 8: 10931100, 1990b Rahman A, Woolley PV, Treat J. A phase II trial of liposomeencapsulated doxorubicin in advanced measurable breast cancer. In Gregoriadis et al. (Eds) Liposomes in drug delivery, Harwood, Chur, 1993 Rao VS, Peyman GA, Khoobehi B, Vangipuram S. Evaluation of liposome-encapsulated clindamycin in Staphylococcus aureus endophthalmitis. International Ophthalmology 13: 181-185, 1989 Richardson VJ, Jeyasingh K, Jewkes RF, Ryman BR. Possible tumour localization of Tc 99m-Iabelled liposomes: effects of lipid composition, charge and liposomal size. Journal of Nuclear Medicine 19: 1049-1054, 1978 Ringden 0, Meunier F, Tollemar J, Ricci P, Tura S, et al. Efficacy of amphotericin B encapsulated in liposomes (AmBisome) in the treatment of invasive fungal infections in immunocompromised patients. Journal of Antimicrobial Chemotherapy 28 (Suppl. B): 83-91, 1991 Roerdink FH, Kroon AM (Eds). Drug carrier systems, Wiley, Chichester, 1989 Ryan PJ, Davis MA, Degaeta LR, Wanda B, Me1choir D. Lipo-

28

somes loaded with contrast material for image enhancement in computed tomography. Radiology 152; 759-762. 1984 Schaeffer HE, Krohn DL. Liposomes in topical drug delivery. Investigative Ophthalmology and Visual Science 21: 220-227, 1982 Schwendener RA, Wuthrich R, Duewell S, Wehrli E, von Schulthess GK. A pharmacokinetic and MRI study of unilamellar gadolinium-, manganese- and iron-DTPA-stearate liposomes as organ-specific contrast agents. Investigative Radiology 25: 922-932, 1990 Seltzer SE, Gregoriadis G, Dick R. Evaluation of the dehydrationrehydration method for production of contrast-carrying liposomes. Investigative Radiology 23: 131-138, 1988 Seltzer SE, Adams DF, Davis MA, Hessel SJ, Havron A. et al. Hepatic contrast agents for computed tomography: high atomic number particulate material. Journal of Computer Assisted Tomography 5: 370-374, 1981 Seltzer SE, Davis MAo Adams DF, Shulkin PM, Landis WJ, et al. Liposomes carrying diatrazoate characterization of biophysical properties and imaging application. Investigative Radiology 19: 142-151, 1984 Senior J, Gregoriadis G. Is half-life of circulating small unilamellar liposomes determined by changes in their permeability? FEBS Letters 145: 109-114, 1982 Senior J, Crawley JCW, Gregoriadis G. Tissue distribution of liposomes exhibiting long half-lives in the circulation after intravenous injection. Biochimica et Biophysica Acta 839: 18, 1985 Senior J, Delgado C, Fisher D, Tikock C. Gregoriadis G. Influence of surface hydrophilicity ofliposomes on their interaction with plasma protein and clearance from the circulation: studies with poly(ethyleneglycol)-coated vesicles. Biochimica et Biophysica Acta 1092: 77-82, 1991 Singh K, Mezei M. Liposomal ophthalmic drug delivery system. I. Ttrinocinolone acetamide. International Journal of Pharmaceutics 16: 339-344, 1983 Singh K, Mezei M. Liposomal ophthalmic drug delivery system. II. Dihydrostreptomycin sulphate. International Journal of Pharmaceutics 19: P263-269. 1984

Drugs 45 (1) 1993

Smolin G, Ikumoto M, Feiler S. Concon D. Idoxuridine-liposomes therapy for herpes-simplex keratitis. American Journal of Ophthalmology 91: 220-225, 1981 Stratford RE, Yang DC, Redell MA, Lee VHL. Effects of topically applied liposomes on disposition of epinephrine and inulin in the albino rabbit eye. International Journal of Pharmaceutics 13: 263-272, 1983 Swenson CE, Stewart KA, Hammett JL. Fitzsimmons WE, Ginsberg RS. Pharmacokinetics and in vivo activity of liposomeencapsulated gentamycin. Antimicrobial Agents and Chemotherapy 34: 235-240. 1990 Szoka Fe. The macrophage as the principle antigen-presenting cell for liposome-encapsulated antigens. Research in Immunology 143: 186-187, 1992 Tan L, Gregoriadis G. The effect of interleukin-2 on the immunoadjuvant action of liposomes. Biochemical Society Transactions 17: 693-694, 1989 Taylor RL, Williams PM, Draven Pc. Graybile JR. Drutz OJ, et al. Amphotericin B in liposomes: a novel therapy for histoplasmosis. American Reviews of Respiratory Disease 125: 610611, 1982 Turner A, Kirby C, Senior J, Gregoriadis G. Fate of cholesterolrich unilamellar liposomes containing III In-labelled bleomycin after subcutaneous injection into rats. Biochimica et Biophysica Acta 760: 119-125. 1983 Vincent M, Webster MH, Pether JVS. Liposomal amphotericin B. Lancet 339: 374-375, 1992 White C, Slifkin M, Seltzer SE, Blau M, Adzamli IK, et al. Biodistribution and clearance of contrast-carrying MREV liposomes. Investigative Radiology 25: 1125-1129. 1990 Yu BF. Preliminary clinical application of new bone marrow imaging agent 99mTc-polyphase liposomes in malignant lymphoma. Chung-Hua-Chung-Liu-Tsa-Chih 12: 297-300, 1990

Correspondence and reprints: Prof Gregory Gregoriadis, Centre for Drug Delivery Research, School of Pharmacy, University of London, 29-39 Brunswick Square, London WClN lAX, England.