This invention relates to improvements in and relating to methods of sonodynamic therapy and, in particular, to the treatment of diseases characterised by

hyperproliferative and/or abnormal cells. The compositions and methods herein described are especially suitable for the treatment of deep-sited tumours which are difficult to treat using conventional non-invasive methods.

Methods known for use in the treatment of cancer include photodynamic therapy (PDT). PDT involves the application of photosensitising agents to the affected area, followed by exposure to photoactivating light to convert these into cytotoxic form. This results in the destruction of cells and surrounding vasculature in a target tissue. Photosensitisers which are currently approved for use in PDT absorb light in the visible region (below 700 nm). However, light of this wavelength has limited ability to penetrate the skin; this penetrates to a surface depth of only a few mm. Whilst PDT may be used to treat deeper sited target cells, this generally involves the use of a device, such as a catheter-directed fibre optic, for activation of the photosensitiser. Not only is this a complicated procedure, but it precludes access to certain areas of the body, for example the brain. It also compromises the noninvasive nature of the treatment. Thus, although appropriate for treating superficial tumours, the use of PDT in treating deeply seated cells, such as tumour masses, is limited.

Sonodynamic therapy (SDT) is a relatively new concept and involves the combination of ultrasound and a sonosensitising drug (also referred to herein as a "sonosensitiser"). In a manner similar to PDT, activation of the sonosensitiser by acoustic energy results in the generation of reactive oxygen species (ROS), such as singlet oxygen, at the target site of interest. Such species are cytotoxic, thereby killing the target cells or at least diminishing their proliferative potential. Many known photosensitising agents can be activated by acoustic energy and are thus suitable for use in SDT.

Since ultrasound readily propagates through several cm of tissue, SDT provides a means by which tumours which are located deep within the tissues may be treated. As with light, ultrasound energy can also be focused on a tumour mass in order to activate the sonosenitiser thereby restricting its effects to the target site. However, problems still remain to be addressed in the development of clinical methods of SDT. A significant problem is that systemic administration of the sonosensitiser facilitates distribution throughout the body. The active drug eventually clears from normal tissues and is selectively retained by proliferating cells (e.g. cancer cells). In some cases, however, the time for clearance can be up to several days, during which period the sonosensitiser may be activated and become toxic by exposure of the patient to ambient light. This poses a significant risk.

The clinical application of PDT was initially driven by the non-invasive or minimally invasive nature of the approach. Since there is a general trend towards minimally invasive procedures for general surgery, further developments in PDT and/or SDT based approaches to the treatment of cancer, particularly with respect to broadening potential targets by inclusion of deep-sited targets, has obvious socioeconomic benefits, e.g. in terms of reduced patient trauma, reduced treatment expense and reduced costs associated with any hospital stay.

This invention addresses some of the challenges faced by existing PDT and/or SDT procedures and, in particular, addresses the need for a minimally invasive procedure to treat deep-sited, inaccessible tumours without adverse side effects.

The inventors have now found that attachment of sonosensitisers to a microbubble confers a number of advantages when used in methods of sonodynamic therapy. What they have found is that the formation of a microbubble-sonosensitiser complex permits effective delivery of the active sonosensitiser in a site-specific manner (e.g. to an internal tumour) by a controlled destruction of the bubble using ultrasound. Subsequent or simultaneous sono-activation of the targeted sonosensitiser results in cell destruction at the target site and regression of tumour tissues. Furthermore, the use of a microbubble leads to a reduction in toxic side effects due to the shielding of the sonosensitiser from potential light activation prior to reaching the desired target site. By using the microbubble as a carrier, nonspecific uptake by non-target tissues is reduced and this provides a significant advantage over systemic delivery of the sonosensitiser alone. This process is completely non-invasive. It is also more effective than PDT due to the greater penetration of ultrasound through tissues. Thus viewed from one aspect the invention provides a microbubble-sonosensitiser complex. The complex comprises a microbubble attached to or otherwise associated with at least one sonosensitiser, preferably a plurality of sonosensitisers. Where the microbubble is attached to more than one sonosensitiser, these may be the same or different. Generally, however, the sonosensitisers will be identical. To the extent that such a complex is intended for use in methods of SDT, it will be ultrasound-responsive. Specifically, it is intended that the microbubble component of the complex can be ruptured by application of ultrasound, thereby releasing the sonosensitiser at the desired target site.

The sonosensitiser (or sonosensitisers) may be linked to the microbubble through covalent or non-covalent means, e.g. via electrostatic interaction.

As used herein, the term "microbubble" is intended to refer to a microsphere comprising a shell having an approximately spherical shape and which surrounds an internal void which comprises a gas or mixture of gases. The "shell" refers to the membrane which surrounds the internal void of the microbubble.

Microbubbles are well known in the art, for example as ultrasound contrast agents. Their composition and methods for their preparation are thus well known to those skilled in the art. Examples of procedures for the preparation of microbubbles are described in, for example, Christiansen et al., Ultrasound Med. Biol., 29:

1759-1767, 2003; Farook et al., J. R. Soc. Interface, 6: 271 -277, 2009; and Stride & Edirisinghe, Med. Biol. Eng. Comput., 47: 883-892, 2009, the contents of which are hereby incorporated by reference.

Microbubbles comprise a shell which surrounds an internal void comprising a gas. Generally, these are approximately spherical in shape, although the shape of the microbubble is not essential in carrying out the invention and is therefore not to be considered limiting. The size of the microbubble should be such as to permit its passage through the pulmonary system following administration, e.g. by

intravenous injection. Microbubbles typically have a diameter of less than about 200 μηι, preferably in the range from about 0.5 to about 100 μηι. Particularly suitable for use in the invention are microbubbles having a diameter of less than about 10 μηι, more preferably 1 to 8 μηι, particularly preferably up to 5 μηι, e.g. about 2 μηι. The shell of the microbubble will vary in thickness and will typically range from about 10 to about 200 nm. The precise thickness is not essential provided that the shell performs the desired function of retaining the gas core.

Materials which may be used to form the microbubbles should be biocompatible and suitable materials are well known in the art. Typically, the shell of the microbubble will comprise a surfactant or a polymer. Surfactants which may be used include any material which is capable of forming and maintaining a

microbubble by forming a layer at the interface between the gas within the core and an external medium, e.g. an aqueous solution which contains the microbubble. A surfactant or combination of surfactants may be used. Those which are suitable include lipids, in particular phospholipids. Lipids which may be used include lecithins (i.e. phosphatidylcholines), e.g. natural lecithins such as egg yolk lecithin or soya bean lecithin and synthetic lecithins such as dimyristoylphosphatidylcholine, dipalmitoylphosphatidylcholine or distearoylphosphatidylcholine; phosphatidic acids; phosphatidylethanolamines; phosphatidylserines; phosphatidylglycerols;

phosphatidylinositols; and mixtures thereof. The use of phospholipids having a net overall charge (e.g. a negative charge) such as, for example, those derived from soya bean or egg yolk; phosphatidylserines; phosphatidylglycerols;

phosphatidylinositols; and phosphatidic acids, is advantageous for ionic linkage of the microbubble to the sonosensitiser.

Polymer materials which are suitable for use in forming the shell of the microbubble include proteins, in particular albumin, particularly human serum albumin. Other biocompatible polymers which may be used include polyvinyl alcohol) (PVA), poly(D,L-lactide-co-glycolide) (PLGA), cyanoacrylate, poloxamers (Pluronics) or combinations thereof.

The microbubble shells may comprise single or multiple layers of the same or different materials. Multiple layers may, for example, be formed in cases where the basic shell material (e.g. a lipid) bears one or more polymers or polysaccharides. Examples of such polymers include polyethylene glycol and polyvinylpyrrolidone. The microbubble shell may also be coated with polymers, such as poly-L-lysine and PLGA, and/or polysaccharides, such as alginate, dextran, diethylamino-ethyl- dextran hydrochloride (DEAE) or chitosan. Methods for attaching these coating materials may involve electrostatic or covalent interactions. Different coating materials (polymers, polysaccharides, proteins, etc.) may be used in order to improve the properties of the microbubble, for example by increasing the rigidity, stability in circulation and/or tissue permeation capability of the microbubble-based reagents, by manipulating the net surface charge of the microbubble and, perhaps most importantly, by increasing its payload capacity. One way of achieving an increase in payload capacity is by the application of the layer-by-layer (LBL) assembly technique. This involves the attachment of multiple layers of a sonosensitiser onto preformed microbubbles in order to increase the sonosensitiser loading capacity. The LBL technique is described by Borden et al. in DNA and polylysine adsorption and multilayer construction onto cationic lipid-coated microbubbles, Langmuir 23(18): 9401 -8, 2007.

In addition, coating of the microbubbles can increase stability of the payload, particularly when the coating material serves as an immobilisation matrix for the sonosensitiser (e.g. via cross-linking).

Lipids forming either a monolayer, bilayer or multilamellar structure may also be used. Examples of these include unilamellar or multilammellar liposomes and micelles.

The microbubble shells may comprise further components which aid delivery of the bubble to the target site. For example, these may be functionalised such that these incorporate or have bound thereto a ligand or targeting agent which is able to bind to a target cell or tissue. Microbubbles having targeting agents attached to their shell are particularly preferred for use in the invention. Examples of suitable targeting agents include antibodies and antibody fragments, cell adhesion molecules and their receptors, cytokines, growth factors and receptor ligands. Such agents can be attached to the microbubbles using methods known in the art, e.g. by covalent coupling, the use of molecular spacers (e.g. PEG) and/or the avidin-biotin complex method. For example, the incorporation of a lipid-PEG-biotin conjugate in lipid-based microbubbles followed by the addition of avidin enables functionalisation of the microbubble surface with a biotinylated targeting ligand. The gas within the core of the microbubble should be biocompatible. The term "gas" encompasses not only substances which are gaseous at ambient

temperature and pressure, but also those which are in liquid form under these conditions. Where the "gas" is liquid at ambient temperature this will generally undergo a phase change to a gas at a temperature of 30 'Ό or above, more preferably 35°C or above. For any gas which is a liquid at ambient temperature, it is generally preferred that this will undergo a phase change to a gas at a temperature between about 30 and 37°C, preferably at around normal body temperature. Any reference herein to "gas" should thus be considered to encompass not only gases and liquids, but also liquid vapours and any combination thereof, e.g. a mixture of a liquid vapour in a gas.

Gases which are suitable for incorporation within the microbubbles according to the invention include air, nitrogen, oxygen, carbon dioxide, hydrogen; inert gases such as helium, argon, xenon or krypton; sulphur fluorides such as sulphur hexafluoride, disulphur decafluoride; low molecular weight hydrocarbons such as alkanes (e.g. methane, ethane, propane, butane), cycloalkanes (e.g. cyclopropane, cyclobutane, cyclopentane), alkenes (e.g. ethylene, propene); and alkynes (e.g. acetylene or propyne); ethers; esters; halogenated low molecular weight hydrocarbons; and mixtures thereof.

Halogenated hydrocarbons are preferred for use in the invention. Those which contain one or more fluorine atoms are particularly preferred and include, for example, bromochlorodifluoromethane, chlorodifluoromethane,

dichlorodifluoromethane, bromotrifluoromethane, chlorotrifluoromethane, chloropentafluoroethane, dichlorotetrafluoroethane, chlorotrifluoroethylene, fluoroethylene, ethyl fluoride, 1 ,1 -difluoroethane and perfluorocarbons.

Preferred for use in the invention are fluorocarbon compounds which include perfluorocarbons. Perfluorocarbons include perfluoroalkanes such as

perfluoromethane, perfluoroethane, perfluoropropanes, perfluorobutanes, perfluoropentanes, perfluorohexanes and perfluoroheptanes; perfluoroalkenes such as perfluoropropene, perfluorobutenes; and perfluorocycloalkanes such as perfluorocyclobutane. Microbubbles containing perfluorinated gases, in particular, perfluorocarbons such as perfluoropropanes, perfluorobutanes, perfluoropentanes and perfluorohexanes are particularly preferred due to their stability in the bloodstream.

Microbubbles containing a perfluorocarbon, particularly a perfluoroalkane, and a shell comprising a phospholipid are particularly preferred for use in the invention and are described in, for example, Nomikou & McHale, Cancer Lett., 296: 133-143, 2010. One example of such a microbubble is Sonidel SDM202 (available from Sonidel Ltd.). The perfluorocarbon may either be present as a gas or in liquid form. Those containing a liquid core may be prepared from nanoemulsions which may subsequently be converted to a gas microbubble upon exposure to ultrasound, e.g. as described in Rapoport et al., Bubble Sci. Eng. Technol.1: 31 -39, 2009.

Sonosensitisers which may be used in the invention include compounds which render target cells or tissues hyper-sensitive to ultrasound. In some cases, a sonosensitiser may be capable of converting acoustic energy (e.g. ultrasound) into ROS that result in cell toxicity. Others may render the target cell or tissues hypersensitive to ultrasound by compromising the integrity of the cell membrane. It is well known that many known sonosensitisers can facilitate photodynamic activation and can also be used to render cells or tissues hypersensitive to light.

Examples of compounds suitable for use as sonosensitisers include phenothiazine dyes (e.g. methylene blue, toluidine blue), Rose Bengal, porphyrins (e.g.

Photofrin®), chlorins, benzochlorins, phthalocyanines, napthalocyanines, porphycenes, cyanines (e.g. Merocyanine 540 and indocyanine green),

azodipyromethines (e.g. BODIPY and halogenated derivatives thereof), acridine dyes, purpurins, pheophorbides, verdins, psoralens, hematoporphyrins,

protoporphyrins and curcumins.

Preferred for use as sonosensitisers in the invention are methylene blue, Rose Bengal and indocyanine green.

Methods for the formation of microbubbles are known in the art. Such methods include the formation of a suspension of the gas in an aqueous medium in the presence of the selected shell material. Techniques used to form the microbubble include sonication, high speed mixing (mechanical agitation), coaxial

electrohydrodynamic atomisation and microfluidic processing using a T-junction (see e.g. Stride & Edirisinghe, Med. Biol. Eng. Comput., 47: 883-892, 2009).

Sonication is widely used and generally preferred. This technique may be carried out using an ultrasound transmitting probe. More particularly, an aqueous suspension of the microbubble shell components is sonicated in the presence of the relevant microbubble component gas.

Other methods which may be used to form the microbubbles include vaporisation of a nanodroplet core in a nanoemulsion (see e.g. Rapoport et al., supra). The core of such nanodroplets will typically be formed by an organic perfluorocompound which is encased by walls of a biodegradable amphiphilic block copolymer such as poly(ethylene oxide)-co-poly(L-lactide) or poly(ethylene oxide)-co-caprolactone. Alternatively, nanoemulsions may be prepared by extrusion through sizing membranes, for example using albumin as the shell material. The droplet-to-bubble transition may be induced by physical and/or mechanical means which include heat, ultrasound and injection through a fine-gauge needle. In one embodiment of the invention, such microbubbles may be formed at the point of administration to the patient (e.g. during the step of administration using a fine-gauge needle) or in vivo at the desired target cells or tissues (e.g. by subjecting the nanoemulsion to ultrasound).

The microbubble-sonosensitiser complexes herein described may be prepared using methods and procedures known in the art. Methods which may be used for covalently attaching the sonosensitiser to the microbubble include known chemical coupling techniques. The exact method used will be dependent on the exact nature of the microbubble and sonosensitiser, specifically the nature of any pendant functional groups. If necessary, either the microbubble and/or sonosensitiser may be functionalised, e.g. to include reactive functional groups which may be used to couple the molecules. Suitable reactive groups include acid, hydroxy, carbonyl, acid halide, thiol and/or primary amine. Methods for the introduction of such functional groups are well known in the art.

Examples of methods which may be used to covalently link a microbubble to one or more sonosensitisers include, but are not limited to, the following: a) Carbodiimide based coupling methods. These may be used to couple microbubbles containing either an amine or carboxylic acid functionality and sonosensitisers having either a carboxylic acid or amine functionality. Such methods result in the formation of ester or amide bonds; b) "CLICK" reaction (i.e. 1 ,3-dipolar cycloaddition reaction). This may be used to react azide or acetylene functionalised microbubbles with a sonosensitiser having either acetylene or azide functionality; c) Schiff base formation (i.e. imine bond formation). This reaction may be used to bond aldehyde or amine functionalised microbubbles to a sonosensitiser containing amine or aldehyde functionality; and d) Michael addition reaction. In addition to coupling of the sonosensitiser to a pre-formed microbubble, the sonosensitiser may

alternatively be linked to a lipid (e.g. using any of the methods described above) and that lipid may subsequently be incorporated into the lipid shell of the

microbubble during its preparation. Methods for preparing a microbubble- sonosensitiser complex as herein defined using any of these techniques form a further aspect of the invention.

Charged sonosensitisers may be electrostatically linked to a charged microbubble. For example, an anionic bubble may be linked to a cationic sonosensitiser and vice versa. One example of a charged sonosensitiser is methylene blue which may be electrostatically attached to an anionic microbubble.

Solutions containing the complexes may be stabilised, for example by the addition of agents such as viscosity modifiers, emulsifiers, solubilising agents, etc.

The complexes of the invention have properties which render these useful in methods of sonodynamic therapy or sonodynamic diagnosis.

Viewed from a further aspect the invention provides a microbubble-sonosensitiser complex as herein described for use in a method of sonodynamic therapy or sonodynamic diagnosis. Use of a complex according to the invention in a method of sonodynamic therapy and, simultaneously, a method of diagnostic imaging forms a preferred aspect of the invention. In such methods, diagnostic imaging may be used to monitor payload deposition and/or accumulation of the complex at the target site of interest. The complexes are suitable for the treatment of disorders or abnormalities of cells or tissues within the body which are responsive to sonodynamic therapy. These include malignant and pre-malignant cancer conditions, such as cancerous growths or tumours, and their metastases; tumours such as sarcomas and carcinomas, in particular solid tumours. The invention is particularly suitable for the treatment of tumours, especially those which are located below the surface of the skin.

Examples of tumours that may be treated using the invention are sarcomas, including osteogenic and soft tissue sarcomas; carcinomas, e.g. breast, lung, cerebral, bladder, thyroid, prostate, colon, rectum, pancreas, stomach, liver, uterine, hepatic, renal, prostate, cervical and ovarian carcinomas; lymphomas, including Hodgkin and non-Hodgkin lymphomas; neuroblastoma, melanoma, myeloma, Wilm's tumour; leukemias, including acute lymphoblastic leukaemia and acute myeloblasts leukaemia; astrocytomas, gliomas and retinoblastomas.

In addition to providing a means of targeting a sonosensitiser to a particular site in vivo, the methods herein described may further be exploited ex vivo. For example, in autologous bone marrow transplantation in the treatment of leukaemia, bone marrow from the patient may be treated ex vivo by molecular targeting of the microbubble-sonosensitiser complex to cancerous cells. These mixtures may then be treated with ultrasound to destroy the cancerous cells and the treated marrow may then be used to re-establish haematopoiesis in the patient following radiation treatment. Alternatively, the methods of the invention may be carried out ex vivo to remove unwanted tissues from organs harvested for conventional transplant. Such tissues may be targeted and destroyed prior to re-transplantation.

For use in any of the methods herein described, the complexes will generally be provided in a pharmaceutical composition together with at least one

pharmaceutically acceptable carrier or excipient. Such compositions form a further aspect of the invention.

The compositions according to the invention may be formulated using techniques well known in the art. The route of administration will depend on the intended use. Typically, these will be administered systemically and may thus be provided in a form adapted for parenteral administration, e.g. by intradermal, subcutaneous, intraperitoneal or intravenous injection. Suitable pharmaceutical forms include suspensions and solutions which contain the active complex together with one or more inert carriers or excipients. Suitable carriers include saline, sterile water, phosphate buffered saline and mixtures thereof.

The compositions may additionally include other agents such as emulsifiers, suspending agents, dispersing agents, solubilisers, stabilisers, buffering agents, preserving agents, etc. The compositions may be sterilised by conventional sterilisation techniques.

Preferably, the compositions for use in the invention will be provided in the form of an aqueous suspension of the complex in water or a saline solution, e.g.

phosphate-buffered saline.

The methods herein described involve administration of a therapeutic or diagnostic amount of the composition. The complex is then allowed to distribute to the desired portion or target area of the body prior to activation. Once administered to the body, the target area is exposed to ultrasound at a frequency and intensity to achieve the desired therapeutic or diagnostic effect. A typical procedure is shown schematically in attached Figure 1 . This shows a two-step process in which the microbubbles (MB) are first ruptured by focused ultrasound thereby releasing the sonosensitiser (SS) which is then able to penetrate the desired target tissue (e.g. tumour). Subsequent sono-activation of the sonosensitiser within the target cells results in production of singlet oxygen which can oxidise various cell components such as proteins, lipids, amino acids and nucleotides thereby destroying the target cells. Whilst it is envisaged that activation of the sonosensitiser will typically take place subsequent to its delivery (i.e. following burst of the microbubbles to release the sonosensitiser), delivery of the complex and activation of the sonosensitiser may nevertheless be simultaneous.

The effective dose of the compositions herein described will depend on the nature of the complex, the mode of administration, the condition to be treated, the patient, etc. and may be adjusted accordingly. The frequency and intensity of the ultrasound which may be used can be selected based on the need to achieve selective destruction of the microbubble at the target site and may, for example, be matched to the resonant frequency of the

microbubble. Ultrasound frequencies will typically be in the range 20 kHz to 10 MHz, preferably 0.1 to 2 MHz. Intensity (i.e. power density) of the ultrasound may range from about 0.1 W/cm ² to about 1 kW/cm ² , preferably from about 1 to about 50 W/cm ² . Treatment times will typically be in the range of 1 ms to 20 minutes and this will be dependent on the intensity chosen, i.e. for a low ultrasound intensity the treatment time will be prolonged and for a higher ultrasound intensity the treatment time will be lower. Ultrasound may be applied in continuous or pulsed mode and may be either focused or delivered as a columnar beam.

Any radiation source capable of producing acoustic energy (e.g. ultrasound) may be used in the methods herein described. The source should be capable of directing the energy to the target site and may include, for example, a probe or device capable of directing energy to the target tissue from the surface of the body.

In cases where the ultrasound frequencies and/or intensities that are needed to achieve cavitation (or microbubble destruction) and those required to cause sonosensitiser activation are different, these different sets of ultrasound parameters (frequency/intensity) may be applied simultaneously or in a two (or multiple)-step procedure.

A further aspect of the invention relates to a method of sonodynamic treatment of cells or tissues of a patient, which method comprises:

(a) administering to the affected cells or tissues an effective amount of a complex or composition as herein described; and

(b) subjecting said cells or tissues to ultrasound.

The complexes may be formulated or administered with other agents in order to enhance the sonodynamic effects. Alternatively, or in addition, other agents known for their chemotherapeutic effects may be used to improve the SDT. Such agents may be administered according to known methods for their use, e.g. administration route, dosage, formulation, etc. Depending on their intended function, these may be administered to the patient prior to, during, or subsequent to any SDT procedure as herein described. A synergistic benefit may be achieved by combining SDT with PDT for the treatment of non-internal tumours.

In some cases, other agents such as chemotherapeutic agents may be coadministered with the complexes according to the invention. Where these are coadministered, these may be administered in a single pharmaceutical preparation or may be administered separately.

In a further aspect the invention thus provides a pharmaceutical composition comprising a microbubble-sonosensitiser complex as herein described in combination with one or more anti-cancer agents. Examples of suitable anti-cancer agents may include, but are not limited to, chemotherapeutics, antibiotics, antivirals, anti-inflammatories, cytokines, immunomodulators, immunotoxins, anti-tumour antibodies, anti-angiogenic agents and combinations thereof.

In a still further aspect the invention provides a kit comprising a microbubble- sonosensitiser complex as herein described and, separately, an anti-cancer agent for use in treating a disorder or abnormality of any cells or tissues within the body which are responsive to sonodynamic therapy. When used, the active components of the kit may be administered simultaneously, separately or sequentially.

Whilst the various methods and uses according to the invention are primarily described herein in the context of administration of a "ready-to-use" microbubble- sononsensitiser complex, in an alternative embodiment a precursor of the complex may be administered. The term "precursor" as used herein is intended to refer to a precursor for the microbubble-sonosensitiser complex which is converted in vivo to it and is thus essentially equivalent thereto. Thus, for example, the term "precursor" encompasses nanoemulsions or nanodroplet formulations which are capable of conversion to the desired microbubble-sonosensitiser complex either in vivo or during administration. In one embodiment, such precursors are capable of conversion to the desired complex upon accumulation in the target tissue (e.g. tumour tissue). Following distribution to the target tissue or cells, the droplet-to- bubble transition may be triggered by methods which include ultrasound.

Alternatively, the step of administration of a precursor of the complex may itself induce formation of a microbubble-sonosensitiser complex according to the invention. For example, where the precursor takes the form of a nanoemulsion, droplet-to-bubble transition may be induced by injection through a fine gauge needle. Direct injection of suitable nanoemulsions into target cells or tissues, for example into tumours, forms a preferred aspect of the invention.

As will be appreciated, in any of the compositions, methods or uses herein described, any reference to a microbubble-sonosensitiser complex according to the invention may be replaced by a suitable "precursor" as defined herein.

Nanoemulsions or nanodroplet formulations for use as microbubble-sonosensitiser precursors according to the invention may be produced by appropriate modification of methods and procedures known in the art, for example those disclosed by Rapoport et al. (supra). In such formulations, the cores of nanoemulsion droplets, which may be formed by a liquid perfluorocarbon (e.g. a perfluoroalkane), are encased by walls of suitable polymeric shell materials (e.g. any of the polymers described herein in relation to the microbubble-sonosensitiser complexes). Linkage of the polymeric shells of the nanodroplets to a sonosensitiser may be achieved using conventional methods and include any of those described above for covalently attaching the sonosensitiser to a pre-formed microbubble. The exact method used will be dependent on the exact nature of the shell material and sonosensitiser, specifically the nature of any pendant functional groups. If necessary, either the polymeric shell and/or the sonosensitiser may be

functionalised, e.g. to include reactive functional groups which may be used to couple the moieties. Suitable reactive groups include acid, hydroxy, carbonyl, acid halide, thiol and/or primary amine.

The invention will now be described further with reference to the following non- limiting Examples and the accompanying drawings, in which:

Figure 1 is a schematic representation of ultrasound-activated sonosensitisation of a bubble-complex according to an embodiment of the invention.

Figure 2 is a graph showing singlet oxygen production of (a) Rose Bengal and (b) Methylene Blue upon photo- and sono-activation. Open squares and crosses in each graph represent control experiments for PDT and SDT, respectively (i.e. where no light or ultrasound activation was used, only drug). The open circles and open diamonds in each graph represent photo- and sono-activation of drug- containing wells, respectively.

Figure 3 is a graph showing the effect of ultrasound on RIF-1 tumour cells treated with indocyanine green.

Figure 4 is a graph showing the effect of ultrasound on RIF-1 tumour cells treated with the cationic microbubble SDM202.

Figure 5 is a graph showing the effect of ultrasound on RIF-1 tumour cells treated with a combination of indocyanine green and the cationic microbubble SDM202.

Figure 6 is a graph showing the effect of ultrasound on RIF-1 tumours established in C3H/HeN mice and treated via intratumoural injection with indocyanine green.

Figure 7a schematically illustrates preparation of the Rose Bengal derivative RB1 and Figure 7b shows a schematic representation of covalent coupling of RB1 to a microbubble.

Figures 8a and 8b are photomicrographs showing microbubbles (a) before, and (b) after conjugation with RB1 in accordance with Example 6.

Figure 9 is a plot of relative absorbance of DPBF at 410 nm against time for:

microbubble-Rose Bengal (MB-RB) conjugate (triangles) and control MBs that were subjected to same treatment as MB-RB but with no coupling agents present (diamonds).

Figure 10 shows the effects of the MB-RB conjugate and ultrasound on tumour cells (RIF-1 ) in vitro.

Figure 1 1 shows the effect of ultrasound on tumours in an in vivo model treated with the microbubble-Rose Bengal (MB-RB) conjugate of Example 6. Example 1 - Ultrasound-mediated generation of free radicals

1 ,3-diphenylisobenzofuran (DPBF) is an organic dye that is converted to the corresponding diketone upon reaction with singlet oxygen. The loss of the characteristic DPBF absorbance at 410 nm upon reaction with singlet oxygen is used as a method to determine the single oxygen production capability of certain sonosensitisers.

To determine the singlet oxygen production capability upon photo- and sono- activation, 5 μΜ of sonosensitiser (either Methylene Blue or Rose Bengal) was exposed to (a) light at 75 W/cm ² (at 650 nm for methylene blue; at 560 nm for Rose Bengal) and (b) ultrasound at 0.5 W/cm ² , 50% duty cycle and a pulse repetition rate of 100 Hz in the presence of 1 ,3-diphenylisobenzofuran (DPBF). In each case, irradiation time was 60 minutes.

The results are shown in Figure 1 . It can be seen that singlet oxygen production (expressed as loss of DPBF absorbance at 410 nm) is observed upon both photo- and sono-activation of Rose Bengal and Methylene Blue. However, the amount of singlet oxygen produced is significantly greater for sono-relative to photo-activation under the experimental conditions used.

Example 2 - Effect of indocyanine green (ICG) on RIF-1 tumour cells in an ultrasonic field

In order to demonstrate that the photosensitiser indocyanine green (ICG) can serve as a sonosensitiser, the mouse tumour cell line, RIF-1 was treated with ICG and the cells were subsequently exposed to ultrasound.

RIF-1 cells were incubated in the wells of 96-well plates at a concentration of 2 x 10 ⁴ cells in a total volume of 200 μΙ per well. Plates were incubated at 37 ^Q C in a 5% C0 ₂ humidified atmosphere overnight. 8 μΙ aliquots of a 5 mg/ml solution of ICG were then added to each well and incubated for 1 hour prior to ultrasound treatment. Target wells were then treated with ultrasound using a Sonidel SP100 sonoporator at various power densities and at a frequency of 1 MHz for 30 s, using a 50% duty cycle and a pulse repetition rate of 100 Hz. Following treatment, plates were placed in the 37 ^Q C incubator overnight. The following day, cell viability was determined by removing the medium and incubating each well in 30 μΙ of trypsin- EDTA together with an equal volume of trypan blue (1 mg/ml in PBS). Cells were subsequently counted directly using a haemocytometer.

The results obtained are shown in Figure 2. It can be seen that cell viability remained relatively intact when cells were treated with ultrasound in the absence of ICG. At the highest ultrasound power density, cell viability had only decreased to approximately 80%. However, when cells were treated with ultrasound in the presence of ICG, it was found that cell viability decreased significantly at ultrasound power densities above 1 .3 W/cm ² . At an ultrasound power density of 2.0 W/cm ² , cell viability had decreased to approximately 30%.

It is apparent from Figure 2 that treating the cell population with ICG renders these cells hypersensitive to ultrasound, i.e. sonoactivated ICG will kill target tumour cells in vitro.

Example 3 - Effect of cationic microbubbles on RIF-1 cells in an ultrasonic field

In order to determine the effect of microbubbles on an RIF-1 mouse tumour cell line in an ultrasonic field, cells were incubated in 96-well plates in a manner similar to that described above for Example 2. 3 μΙ aliquots of a PEGylated lipid-shelled microbubble (Sonidel SDM202: 7 x 10 ⁸ microbubbles per ml) containing a perfluorocarbon gas centre, with an average diameter of 2 μηι and having a slight cationic surface charge were added to each well immediately prior to treatment. Mixtures were then exposed to ultrasound for 30 s using a Sonidel SP100 sonoporator at various power densities and at a frequency of 1 MHz and using a 50% duty cycle with a pulse repetition rate of 100 Hz. Following treatment, cells were incubated overnight at 37 ^Q C in a 5% C0 ₂ humidified atmosphere. Cell counts were determined by counting using a trypan blue dye exclusion assay.

The data obtained are shown in Figure 3. It can be seen that in the presence of microbubbles, at ultrasound power densities above 1 .0 W/cm ² , cell viability was significantly reduced. At 2.0 W/cm ² cell viability was reduced to approximately 45% clearly demonstrating that a combination of microbubbles and ultrasound can facilitate cell destruction.

Example 4 - Effect of a combination of cationic microbubbles and sonosensitiser (ICG) in an ultrasonic field

Target cells (RIF-1 ) were cultured in 96-well plates as described in Examples 2 and 3. Combinations of ICG (8 μΙ/well) and the cationic microbubble SDM202 (3 μΙ/well) were added to each target well. Cells were then treated with ultrasound as described in Examples 2 and 3. Following treatment, plates were incubated at 37 ^Q C in a 5% C0 ₂ humidified atmosphere overnight and cell viability was then determined by direct counting using a trypan blue assay as described in Examples 2 and 3.

The results are shown in Figure 4. It can be seen that even at the lowest ultrasound power density (1 .0 W/cm ² ) a significant decrease in cell viability was observed. Since the decrease in cell viability is greater than the combined effects observed using either the ICG or the SDM202 alone, the results suggest that synergistic effects are observed.

These data clearly demonstrate that a combination of ICG and microbubbles in the presence of an ultrasonic field affords a significant advantage over treatments with each agent separately. This leads to the conclusion that both the sonosensitiser and the microbubble, when used in combination, provide a synergistic effect.

Example 5 - Effect of a combination of ICG and ultrasound on tumour growth in vivo

In order to determine that ICG, in serving as a sonosensitiser, could be employed together with ultrasound to facilitate a reduction in tumour growth in vivo, RIF-1 tumours were treated in the syngenic host.

RIF-1 cells were grown to 90% confluence and harvested to yield a suspension of 2-3 x 10 ⁷ cells/ml. 0.1 ml aliquots were injected intradermal^ to the flank of each animal. In all cases animals were treated humanely and in accordance with licensed procedures under the UK Animals (Scientific Procedures) Act, 1986. When tumours reached the appropriate size, tumours were injected (via

intratumoural injection) with 20 μΙ of ICG (5 mg/ml). Immediately following addition of ICG, tumours were treated with ultrasound using a Sonidel SP100 sonoporator at a frequency of 1 MHz, using a power density of 4 W/cm ² for 4 min at a duty cycle of 40%. Following treatment, tumour size was determined at the indicated times using a 3-leg measurement method and using the formula: tumour volume = 4/3 (TTR ³ ) where R = tumour diameter determined using an average of the 3-leg

measurements. The % tumour volume was determined using the starting tumour volume in each group.

The data obtained are shown in Figure 5 and clearly demonstrate that tumour growth was severely retarded when tumours were treated with ICG together with ultrasound. The data suggest that ICG can serve as a sonosensitiser in order to render tumours hypersensitive to ultrasound treatment.

Example 6 - Preparation of microbubble-Rose Bengal conjugate

The sonosensitiser Rose Bengal was first reacted with 8-bromooctanic acid to form a carboxylic acid functionalised derivative (RB1 ) (Figure 7a) suitable for

incorporation onto the surface of amino-functionalised microbubbles (Figure 7b). Amino-functionalised microbubbles were prepared by sonication of an aqueous dispersion of the lipid-based reagents in the presence of a perfluorobutane gas stream (Nomikou et al., Acta Biomaterialia 8: 1273-1280, 2012). The microbubbles were stabilised by the inclusion of a polyethylene glycol-lipid conjugate in the shells. The molar ratio of each lipid-based reagent in the microbubble shells was 51 % DSPC (distearoylphosphatidyl choline), 44% PEG (polyethylene glycol)-40- steararte and 5% DSPE-PEG (disteroylphosphadityl ethanolamine-polyethylene glycol)-amino (AVANTI, USA). The preparation was adjusted to a concentration of 1 x 10 ⁹ microbubbles/ml using PBS (phosphate buffered-saline). The mean diameter of the microbubbles was 1 .7 μηι.

The Rose Bengal derivative (RB1 ) was covalently bonded to the amino- functionalised microbubbles using dimethylaminopropyl-3-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysulfosuccinimide (sulfo-NHS) as coupling agents in PBS buffer (pH = 7.4 ± 0.1 ) (Figure 7b). The resulting microbubble: RB1 conjugates (MB-RB) were purified by centrifugation followed by washing with PBS buffer.

Example 7 - Ultrasound-mediated generation of free radicals

To determine the singlet oxygen generating potential of the conjugate prepared in Example 6 upon ultrasound irradiation the 1 ,3-diphenylisobenzofuran (DPBF) based assay was used (see Example 1 ).

Specifically, 2ml of the MB-RB conjugate of Example 6 containing a MB

concentration of 0.69 mg/ml and a RB1 concentration of 0.47 mg/ml (as determined by UV-Vis spectroscopy) was added to an aerated solution of DPBF (10 μΜ) in an EtOH:H ₂ 0 (50:50) solvent system. The solution was then irradiated with ultrasound delivered by a Sonidel SP100 sonoporator (emitting at a frequency of 1 MHz, using an energy density of 1 .5 W/cm ² for 60 min and a 50% duty cycle at a pulse repetition frequency of 100 Hz).

Two controls were also employed - the first involved using exactly the same conditions as explained above but with no ultrasound irradiation. The second control involved using a 2 ml sample of microbubbles and RB1 that were stirred together under the same experimental conditions outlined above without the presence of the coupling agents EDC and sulfo-NHS. This second control was used to determine the extent of non-covalent interactions between the MB and RB1 entities.

The results from these experiments are shown in Figure 9 and reveal a significant loss in the DPBF absorbance at 410 nm for the MB-RB conjugate indicating significant singlet oxygen production. However, no noticeable reduction in DPBF absorbance was observed for either of the controls. In addition to demonstrating ultrasound-mediated cytotoxic potential of the MB-RB conjugate, this further suggests the benefit of a covalent interaction between the sonosensitiser and the microbubble to enable significant quantities of singlet oxygen to be produced upon ultrasound irradiation. Example 8 - The effect of ultrasound and the microbubble-Rose Bengal conjugate (MB-RB) on a tumour cell line (RIF-1 ) in vitro

In order to confirm that the MB-RB conjugate of Example 6 responds to ultrasound by eliciting a toxic effect as suggested in Example 7, a mouse radiation-induced fibrosarcoma cell line (RIF-1 ) was used as a target (Nomikou et al., supra). The cells were maintained in RPMI 1640 medium supplemented with 10% (v/v) foetal bovine serum at 37 ^< C in a humidified 5% (v/v) C0 ₂ atmosphere. These cells were plated into the wells of a 96-well tissue culture plate at a concentration of 2 x 10 ⁴ cells per well and incubated overnight at 37°C in a humidified 5% C0 ₂ atmosphere. Medium was then removed from each well and replaced with 50 μΙ of the conjugate MB-RB preparation at a concentration of 2 x 10 ⁷ microbubbles/ml. These were then treated with ultrasound (MB-RB + U/S) for 30 s, using a frequency of 1 MHz, an ultrasound power density of 1 .5 W/cm ^"2 and a duty cycle of 50% (pulse frequency = 100 Hz). Control preparations consisted of (i) cells treated with ultrasound in PBS (U/S alone); (ii) cells treated with either RB1 (RB1 ) at a concentration equivalent to that bound to the microbubbles in RB-MB; (iii) microbubbles at a concentration equivalent to that used in RB-MB (MB) in the absence of ultrasound; and (iv) cells treated with ultrasound in the presence of an unconjugated microbubble and RB1 mixture (RB1 + MB + U/S).

Following treatment, samples were incubated for 30 min at 37°C in a humidified 5% C0 ₂ atmosphere. Wells were then emptied, washed with PBS and 200 μΙ aliquots of serum-containing medium were dispensed into each well. Following an overnight incubation at 37 ^< Ό, cell viability was determined using an MTT assay (Nomikou et at., supra).

The results from these experiments are shown in Figure 10 and demonstrate that when ultrasound was applied to the cells in the absence of any active reagent, no decrease in cell viability was noted (U/S only). The results also demonstrate that at the concentrations used, the sonosensitiser (RB1 ) or the microbubbles alone (MB) were non-toxic to the target cell population in the absence of an ultrasonic field. When unconjugated microbubbles and RB1 were incubated with cells and an ultrasonic field was applied, an approximate 65% decrease in cell viability was noted (RB1 + MB + U/S). However when cells were treated with the MB-RB conjugate a 75% decrease in cell viability was noted. These data demonstrate that the MB-RB conjugate elicits a cytotoxic effect on this target tumour cell line in the presence of an ultrasonic field.

Of all of the tests, the sample containing the MB-RB complex in the presence of ultrasound exhibited the highest degree of compromise to cell viability and these results confirm those observed in Example 8 where the conjugate was found to result in the highest degree of reactive oxygen species production when treated with ultrasound. Therefore in addition to demonstrating that the invention may elicit an ultrasound-induced toxic effect on tumour cells, the data also show that chemical coupling between the microbubble and sonosensitiser affords an advantage in terms of eliciting an enhanced toxic effect.

Example 9 - The effect of ultrasound on tumours treated with the MB-RB conjugate

Since Example 8 demonstrated an ultrasound-mediated toxic effect of the MB-RB conjugate it was of interest to examine the effects of ultrasound on conjugate- treated tumours using an in vivo model. To this end a human prostate tumour model in SCID mice was employed as a target.

Tumours were generated using a modified form of the LNCaP human prostate cell line LNCaP-Luc (Ming, Ph.D. Thesis, University of Ulster, 2009) and this was cultured in RPMI 1640 supplemented with 10% (v/v) foetal bovine serum, 100mM HEPES and 5 mM glucose at 37°C in a humidified 5% (v/v) C0 ₂ atmosphere.

Every second change of medium contained geneticin at a concentration of 300 μg/ml to maintain selective pressure. In order to induce tumour formation, 5 x 10 ⁶ cells in 100 μΙ aliquots of Matrigel ^® were injected subcutaneously on the dorsum of BALB/c SCID mice (8 weeks old). In all experiments animals were treated humanely and in accordance with licensed procedures under the UK Animals (Scientific Procedures) Act, 1986. When the tumours had reached an average size of 1 .24 cm ³ , a 30 μΙ aliquot of the MB-RB conjugate of Example 6 (2 x 10 ⁸ microbubbles/ml) was injected into each tumour. Animals were subsequently treated with ultrasound for 3 min using a frequency of 1 MHz, an ultrasound power density of 3.5 W/cm ² and 30% duty cycle (100 Hz pulse frequency). Control animals were injected with the conjugate but did not receive ultrasound. At various time intervals post treatment, tumour volumes were calculated using volume = 4/3( ^" πτ ³ ) where r was the tumour radius determined from 3-leg measurements taken at the indicated times.

The results from these experiments are shown in Figure 1 1 . When tumours were treated with the MB-RB conjugate in the absence of ultrasound those tumours continued to increase in size (Figure 1 1 a). However when tumours were treated with the MB-RB conjugate and subsequently treated with ultrasound a significant decrease in tumour growth was observed (Figure 1 1 a). Indeed when data from day 4 were analysed it was found that the tumour size actually regressed and this is shown more clearly in Figure 1 1 b.

The data clearly demonstrate that in the presence of ultrasound, the MB-RB conjugate elicits a cytotoxic effect on tumours in vivo. In demonstrating an ultrasound-specific effect, the data establish that the invention can facilitate site- specific treatment of tumours because in the absence of ultrasound, no effect on tumour growth was observed.