Blocking the way forward

Many drugs, while very effective in targeting disease, can be difficult to administer. They may not be orally available because they break down or perhaps it is not easy to get them to the active site. Sarah Houlton looks at a novel alternative - micelles

Numerous methods for targeting drugs to their active sites have been developed over the years. One way that is receiving a good deal of attention at the moment is the use of block copolymer micelles to facilitate delivery. A micelle is a sub-microscopic agglomeration of molecules. The colloidal particles are formed by the aggregation of molecules such as detergents, with the centre having one polarity, and the outside the opposite.

This means they can be used to solubilise molecules that would not otherwise be miscible, for example oils in water, and is the principle operating behind soaps.

A block copolymer is a polymer that consists of a chain of one monomer at one end, attached to a chain of a different monomer at the other. They can spontaneously form into micelles in aqueous solution, which have a hydrophilic outer shell, and a hydrophobic inner core. And incorporating a moiety such as polyethylene glycol at the end of the flexible tethered polymer strands renders them biocompatible. When mixed with a drug substrate at the right concentration, the block copolymers will form micelles with the drug encapsulated inside.

These micelles made up from block copolymers were first proposed as functional particles by the Japanese scientists Masayuki Yokoyama and Teruo Okano of Tokyo Women's Medical University, and Kazunori Kataoka of the University of Tokyo.

They showed that the particles could act as stable carrier systems within the bloodstream, and also that they could accumulate in cancerous tissues, indicating their potential for targeting anticancer agents at tumours.

huge potential

This sort of passive targeting, where drugs are encouraged to accumulate in the target regions, has huge potential. A small molecule drug will generally spread throughout the body in the bloodstream, reaching all organs in varying degrees through the capillaries.

Cancer tissues replicate themselves so rapidly that the structure of the new vessels is poor, and so they can be as much as 10 times more permeable than normal blood vessels. As a result, polymer molecules and nanoparticles such as micelles can easily permeate and accumulate in cancer tissues.

The cancerous tissues will also have immature lymph systems, if they have them at all, and so the polymers are more likely to remain within the cancer, unlike normal cells which discharge them to the lymph system. So if anticancer drugs could be stabilised within a micelle, they would accumulate within the cancer tissue, and if they remain active, this is important.

Active targeting is also possible using block copolymer systems. This happens when molecules with high affinity are used, such as antibody molecules that connect selectively with specific antigens on cancer tissues.

Chao-Pin Li of GlaxoSmithKline explained some of the factors that must be considered for using block copolymer systems in drug delivery.

He said they may help solubility, but that the amount of drug that is loaded into the micelle has to be carefully considered. In vivo distribution must also be considered: where does the micelle actually go? What are its pharmacokinetics and its efficacy? And then there is the potential for active drug targeting. Could a ligand be attached to direct it to a particular active site?

high efficacy

An example of a block copolymer based drug delivery system targeted at tumours was given by Kazunori Kataoka, of the department of materials science and engineering at the University of Tokyo. 'To hit a tumour successfully, you have to make sure the drug doesn't go into the reticular endothelial system, and isn't taken up by the kidney or liver it needs to stay in the blood for a long time,' he explained.

'The particle should be less than 100nm in size. If you can achieve this, then you can expect high efficacy in tumours. But for some drugs, you have to get selective uptake into tumour cells by exercising control over trafficking the drug into the cytoplasm.'

For example, the anticancer agent Adriamycin (doxorubicin) can be delivered using PEG-poly-aspartic acid. 'The drug is both chemically conjugated and physically trapped by the polymer,' said Kataoka. 'This formulation is now undergoing Phase I trials, and gives an increased accumulation of Adriamycin in the tumour cells - it is 10 to 20 times higher than is seen with Adriamycin alone. And this gives very much better antitumour activity.

'We hope it will enter Phase II trials in 2003, and we are still working on the system in order to stabilise it further. Instead of a covalent bond, we are looking at a linker mediated system. At pH 6, there is almost no release of adriamycin; as the pH drops, it is released.'

Kataoka explained that there are different drivers for metal-based drugs such as the platinum anticancer agent cisplatin. The copolymer PEG-b-poly([α,β]-aspartic acid), or PEG-P-Asp, works here, and is very stable. He has also tried the glutamic acid analogue.

The glutamic acid system is even more stable, dissociating in around 100 hours, compared with 20 hours for the aspartic acid version.

The PEG-P-Glu micelles exhibited lower toxicity and higher antitumour activity than the same dose of cisplatin on its own. Cisplatin is incorporated into the copolymer by chemical conjugation. In comparison, paclitaxel is physically encapsulated. But doxorubicin is incorporated by both chemical and physical means.

Paclitaxel is being worked on by Japanese company Nanocarrier along with Nippon Kayaku, and the encapsulated drug is currently in preclinical trials.

new treatments

Kataoka also reports that block copolymer systems have applications in photodynamic therapy.

This procedure was initially used for treating skin cancers, but now can be used on the much deeper-seated lung cancers. A photosensitiser is administered intravenously to the patient, which increases sensitivity to light. If it is encapsulated into a micelle, it can increase in activity. A strong photodynamic effect is seen with dendrimers.

'Long circulation gives high accumulation in tumour cells,' Kataoka said. The micelles enhance the effect of photoirradiation, giving around 50 times higher efficacy.

Another drug that is being looked at in conjunction with block copolymer delivery systems is amphotericin B.

Dubbed 'amphoterrible' by AIDS patients because of its ghastly side-effects, it is a potent membrane acting drug that is used to combat life-threatening systemic fungal infections. A liposomal formulation of amphotericin B is much less toxic than the standard treatments, but its antifungal activity is also much lower, quite apart from being significantly more expensive.

Glen Kwon of the School of Pharmacy at the University of Wisconsin - Madison said the problem with the drug is that it likes to aggregate and in this form it punches holes in cell membranes, as well as those of the fungus.

When deaggregated, it is less toxic. but this leads to administration problems. While trying to tailor-make micelles to improve its delivery properties, Kwon found that using Pluronic F127, a nonionic block copolymer surfactant from BASF, works very well.

It is easy to deaggregate completely, and the system is now being tested in vivo in animals. In theory, the combination should be non-toxic and a strong antifungal agent.

Kwon is also working on fine tuning the properties of the micelles by tweaking the polymer structures using simple chemical reactions. The aim is to get a long circulating depot of monomeric amphotericin B, while keeping the polymers sufficiently small so that they can be excreted by the kidneys.

increased activity

One way of loading the drug into the carrier, Kwon said, is by solvent evaporation: simply mixing the amphotericin B and the polymer in methanol, with the methanol being evaporated prior to freeze drying.

In tests of the micellar amphotericin B formulation against the fungal agents Candida albicans, C. neoformans and Aspergillus fumigatus the novel formulation has been shown to kill them all, with an antifungal activity very similar to Fungizone, and five times more active than the liposomal formulation. 'The self-aggregated state can be modulated,' said Kwon.

'Encapsulated amphotericin B gives reduced haemolysis and retains its potent antifungal activity,' he said.

Sasha Kabanov of the University of Nebraska Medical Center is also working with BASF's Pluronic polymers.

'Pluronics are common, safe, and often used as excipients,' he said at the meeting. 'BASF makes many different ones, and just one methyl group difference between the blocks can make the middle of the micelle hydrophobic.'

potential carriers

They spontaneously assemble into micelles, and are ideal drug delivery system carriers, making small micelles from 10 to 100nm in diameter, with the size being changed by altering the lengths of the blocks.

The corona is formed by a nonionic moiety such as polyethylene glycol, and the core is ideal for incorporating hydrophobic compounds; the relative sizes of core and corona can also be changed by varying the block length. Interactions between the core and the guest molecule can also be fine tuned to change loading and release properties, and affect the overall stability of the system. Kabanov explained that it is possible to target micelles by conjugating their external chains with an antibody.

If the Pluronics are too hydrophobic, they can have a dramatic effect on the membrane, and get stuck. A Pluronic doxorubicin formulation, for example, has had Phase I/IIA trials completed in the UK, he said, and they have also been investigated as potential carriers for gene therapy.

Following earlier work on rhodamine 123, Kabanov has looked at a range of structurally diverse drug molecules in combination with Pluronic P85, including fluorescein, doxorubicin, etoposide, Taxol, AZT, valproic acid and loperamide in two different model systems.

Drug permeability studies on models of the blood-brain and intestinal barriers showed that exposure of the cells to Pluronic P85 significantly enhanced permeability coefficients for fluorescein, Taxol, doxorubicin and AZT in both the model systems studied, and in one system for etoposide, valproic acid and loperamide.

All these compounds had previously been shown to be affected by P-glycoprotein and/or multidrug resistant associated protein efflux systems. The block copolymer clearly has a broad specificity with respect to drugs and efflux systems, and this could well be a valuable property in developing pharmaceutical formulations that would increase drug accumulation in selected organs, and overcome both acquired and intrinsic drug resistance that can limit the effectiveness of numerous drug compounds.

A variety of other structures can be created that have uses in drug delivery, Kabanov said. The unimer is also present in the formulation, and this seems to have an effect on multidrug resistant cells, and the company Supratek has been set up to look at this.

Vesicles are being investigated by Adi Eisenberg at McGill University. Vesicles are microscopic hollow spheres that consist of a thin skin of hydrophobic material, with hair-like structures of a hydrophilic material around them.

They are structurally analogous to liposomes, which are small molecule phospholipids, but vesicles are much more robust. In contrast to micelles, they can be used to deliver hydrophilic materials to the body.

Eisenberg has been experimenting with attaching one type of material to the exterior surface of the structure, and a different one to the interior. This means that a three way delivery system is a possibility, with the exterior attached to a targeting moiety, and with a water soluble drug suspended in the interior cavity and water repellent materials attached to the wall.

This would result in vesicles that could be used to deliver both water soluble and water repellent drugs, with different species of drugs or targeting moieties associated with the interior or exterior surfaces of a vesicle, and where each type of drug can be released in the right proportion.

This is currently not an easy problem for the formulator to solve. The modified vesicles may result in a single delivery system that could deliver active agents to specific parts of the body, conceivably minimising toxic side-effects.