Field of the invention The subject matter of the invention is a process for post-preparation radiolabelling of liposomal formulations by ¹⁸ F (fluorine) isotope. Further subjects of the invention are the labelled liposomes prepared by the process, and the liposomes for diagnostic application. The new intermediates used in the process of the invention form a subject of the invention as well.

The process of the invention is especially suitable for studying the distribution of liposomes within the living organism, i.e. evaluating the effectiveness of liposomal delivery systems.

State of the art

The utilisation of a number of drugs is significantly limited by the fact that the drug molecule reaches the target cells or tissues only in a very small amount, due to unwanted metabolic or other unfavourable pharmacokinetic processes. Therefore, recently, an intensively researched field of the pharmaceutical industry is the improvement of delivery systems that enable appropriate delivery of effective but sensitive active pharmaceutical ingredients. Up to now, a number of nanoparticle-based pharmaceutical formulations have been marketed, and several hundred are under different phases of clinical trials. Among these, the most widespread are the liposome-based formulations which are, due to their special structure (aqueous core surrounded by a phospholipid double layer), suitable for the targeted delivery of hydrophobic and hydrophilic active pharmaceutical ingredients. The first authorised medicinal product of this type was the liposomal doxorubicin, the Doxil (in Europe Caelyx), which was followed by a number of formulations. Having knowledge on the distribution of liposomes within the living organism, i.e. evaluating the effectiveness of such delivery systems, is of primary importance, if possible, by using non-invasive labelling technique. Positron emission tomography (PET) is the most modern molecular imaging technique known today for studying living organisms, as it can provide the most accurate information both from a qualitative and quantitative point of view, in case of using appropriate isotope and delivery system. The isotope of mass number 18 of fluorine ( ¹⁸ F) is a pure positron emitter, and the energy of the emitted positron is also the lowest, thereby offering an optimal imaging option for the PET technique, furthermore the half life of ¹⁸ F is long enough (nearly 2 hours) to trace the biochemical distribution of liposomes, even during hours, under in vivo conditions. ln the literature of the field, several different procedures have been described for labelling liposomes with F-18. Oku et al. (Biochemica et Biophysica Acta, 1995, 1238, 86-90) have labelled the liposomes with the widespread and widely used fluoro-deoxy-glucose (FDG) in such a way that they formed the liposomes in a solution of FDG, and in a separation step the liposomes were separated from the untrapped FDG molecules. As FDG, and the monosaccharides in general, interact weakly with the phospholipid bilayer, the efficiency of said labelling corresponds to the ratio of the volume confined by the liposomes to the total volume, which, in case of preparation by extrusion through polycarbonate filter (which is the most widely used in drug production), typically reaches only a few per cent. Moreover, the labelling has to be inserted in an intermediate step of the preparation process, which can be accomplished under laboratory conditions, but cannot be fitted in a high- volume manufacturing process.

Marik et al., in 2007, reported on the preparation of (F-18)fluoro-glyceriol-dipalmitate (FDP) and its use for in vivo tracking of liposomes by PET technique (Nuclear Medicine and Biology, 2007, 34, 165- 171). The authors have inserted the FDP into the liposomes during their preparation, due to its lipophilic character, but they have not investigated the possibility of post-preparation insertability of this or any other tracing material. asmussen et al., in 2012, reported on the preparation of an ¹⁸ F-labelled cholesterol derivative and on its insertion into liposomes (Andreas T. U. Jensen et al. : PET imaging of liposomes labeled with an [ ¹⁸ F]-fuorocholesteryl ether probe prepared by automated radiosynthesis. Journal of Liposome Research, 2012; 22(4): 295-305). In the method elaborated by Rasmussen et al., the isotope labelling takes place in the initial phase of the preparation process, too.

Reibel et al. have accomplished the ¹⁸ F labelling of different cholesterol derivatives and their insertion into the phospholipid bilayer (Achim T. Reibel et al. : Fate of Linear and Branched Polyether- Lipids In vivo in Comparison to Their Liposomal Form ulations by 18F-Radiolabeling and Positron Emission Tomography. Biomacromolecules, 2015, 16(3), 842-851). Here as well, the labelling of the cholesterol derivatives took place first, and this was followed by the construction of the liposome using the already radiolabeled lipids, so this method does not offer post-preparation labelling either.

Several publications report on the insertion of the lipid derivative l-[ ¹⁸ F]fluoro-3,6-dioxatetracosane into the membrane of liposomes. The labelling strategy, however, was different in this case as well : the labelling of the lipid with ¹⁸ F isotope took place anteriorly, and then the labelled lipid derivative was inserted into the membrane of the liposome by a so-called solid phase transfer method (Fukuta T et al.: Real-time trafficking of PEGylated liposomes in the rodent focal brain ischemia analyzed by positron emission tomography. Artif Organs, 2014; 38(8):662-6). In this case the insertion of the labelled lipid can be regarded as being post-preparative, but the lipid derivatives labelled by the authors, due to their unfavourable structure, detach from the liposomes to a considerable extent within only an hour, based on the reported stability data, therefore the effectiveness of the labelling by this method is questionable. (Urakami T et al.: Novel amphiphilic probes for [ ¹⁸ F]-radiolabeling preformed liposomes and determination of liposomal trafficking by positron emission tomography. J Med Chem. 2007 Dec 27; 50(26):6454-7.).

In the literature, the so-called Cu(l)-catalysed Huisgen cycloaddition (also known as "click-reaction") between alkynes and azides has become known recently, which can also be used for the selective labelling of biomolecules under mild conditions. M. Glaser et al., although for a different aim, have successfully applied the method based on the click-reaction in the field of ¹⁸ F-labelling (Bioconjugate Chem. 2007, 18, 989-993) in such a way that the reaction partner containing the azide group was labelled by ¹⁸ F isotope, thereby obtaining the 2-[ ¹⁸ F]fluoroethylazide ([ ¹⁸ F]FEA), and then the„click" binding of this compound has been examined in reactions with model compounds containing different alkyne groups. The application of the click reaction so far in the labelling of biomolecules by ¹⁸ F isotope is summarized in the 2014 publication of Kettenbach et al. (Kettenbach K. et al.: 18F- labeling using click cycloadditions. Biomed Res Int. 2014:361329.). According to our present knowledge, however, this process has not been examined for direct isotope-labelling of liposomes until now.

S. Cavalli et al. describe liposomes bearing alkyne groups on their surface, and they confirm by fluorescence resonance energy transfer (FRET) method that the click reaction takes place on the outer surface of the liposomes (S. Cavalli et al:„The chemical modification of liposome surfaces via a copper-mediated [3+2] azide-alkyne cycloaddition monitored by a colorimetric assay. Chem Commun., 2006, 3193-3195). As a conclusion of their work they foreshadow that by click reaction, peptides bearing azido function could be attached to the outer membrane of liposomes. P. Marques- Gallego et al., in turn, describe coupling of peptides to the surface of azido-functionalized liposomes by click reaction. (P. Marques-Gallego et al.: Ligation Strategies for Targeting Liposomal Nanocarriers. BioMed Research International, 2014, http://dx.doi.org/10.1155/2014/129458). None of these authors mention any possibility of an ¹⁸ F isotope labelling. ln the Patent Application No. US 20090162424 Al and in Kumar et al.: "Clickable", polymerized liposomes as a versatile and stable platform for rapid optimization of their peripheral compositions. Chem. Commun., 2010, 46, 5746-5748, post-preparative surface functionalization of polymerised liposomes is described by "click" chemical method. For the feasibility of their method, they describe the simultaneous fulfilment of two conditions: the lipids constituting the liposomes must be polymerisable and at the same time they must contain "click" -functional groups. They state that only liposomes stabilized by polymerisation can bear the reaction conditions of the "click" reaction used by them, without decomposition or a considerable structural change of the liposome.

The technical problem to be solved None of the hitherto known and applied methods for labelling drug-carrier liposomes by ¹⁸ F isotope make possible efficient labelling separately from their preparation (using a so called post-preparation method) of the utmost importance for its practical applicability.

The preparation of liposomes (especially the ones loaded with an active ingredient) is typically a complex procedure, into which the isotope labelling - especially due to radiation protection considerations - cannot or could only be included in a very restricted manner. Therefore such a procedure in which the labelling isotope is first bound to a suitable carrier molecule, and thereafter, during the preparation of the liposome, is encapsulated into the core of the liposome (Biochemica et Biophysica Acta, 1995, 1238, 86-90) or is inserted into the membrane of the liposome ( eibel AT et al.: Biomacromolecules. 2015; 16(3):842-51.), is obviously poorly applicable due to the mentioned radiation protection and technological drawbacks.

Therefore the aim of the invention is to elaborate a process for post-preparation labelling of liposomes by ¹⁸ F isotope. By post-preparation labelling those methods are meant, in which the isotope labelling phase takes place separately in time, after the process of the preparation of the liposome at any point in time (of course within the shelf life of the liposomes) without breaking the bilayer structure of the lipid constituting the membrane of the liposome or significantly altering it. This is an important aspect for liposomal formulations that carry an active therapeutic agent inside them.

Brief description of the invention

As described above, the following fundamental problems have arisen in connection with the methods previously used for efficient labelling: too high concentrations of reagents (copper ions/ascorbate/ligand system) in the click-coupling led to the breakdown of liposomes, or, providing stability during reaction could be ensured only by using polymerized liposomes, or the stability of the finished labelled liposomal formulations was not satisfactory.

In contrast, we have found that phospholipid derivatives modified by terminal alkyne functional group(s) may be produced which, when mixed in small amounts to the lipids constituting the liposomes, are incorporated into liposomal membranes, so that such lipids will only have a negligible influence on the structure and stability of the liposomes. Thereby, liposomes containing free terminal alkyne functions on their surface are obtained which can be labelled in aqueous solution, under mild reaction conditions with ¹⁸ F-labelled azides such as [ ¹⁸ F]flouroethylazide ([ ¹⁸ F]FEA). Surprisingly, we have found that liposomes of generally used compositions, consisting only of phospholipids with saturated fatty acids and cholesterol derivatives, are suitable for ¹⁸ F isotope labelling by click reaction at suitably low reagent concentrations. We have shown that by click-coupling the post-preparation ¹⁸ F-labelling of the liposomes can be performed in a short time and that the labelled liposome remains stable in the serum for the time period required for the contemplated studies (even for several hours) in such a way that the labelling isotope does not detach at all from the liposome (the isotope fraction not bound to the liposome remains in the negligible range).

The present invention thus provides a process for post-preparation labelling of liposomal formulations with an ¹⁸ F (fluorine) isotope, wherein a) during the preparation of the liposomal formulation, a phospholipid derivative modified by terminal alkyne functional group(s) is incorporated into the membrane of the liposome, then b) the alkyne groups of the liposomes obtained in step a) are reacted with an ¹⁸ F-labelled azide derivative.

This process is particularly advantageous when a novel liposomal formulation is under development, or in preclinical phase, when it is still possible to modify the composition of the liposome by adding the phospholipid derivatives having terminal alkyne functionality within a certain concentration range. The liposomes with the modified composition, after the subsequent labelling, will be suitable for in vivo biodistribution studies with PET technique.

The phospholipid derivative modified by terminal alkyne functional group(s) is preferably a compound of formula (A) and/or (B)

(A)

(B) wherein x and R ₂ are the same or different and each represent an acyl group derived from a saturated fatty acid; n represents 1, 2, 3, 4, 5, 6; and Y represents an alkynyl group.

The phospholipid derivative modified by terminal alkyne functional group(s) is more preferably (R)- 2,3-bis(palmitoyloxy)propyl-2-(3-(prop-2-ynyloxy)propaneamid o)ethyl phosphate of the following formula (1)

or (R)-2,3-bis(palmitoyloxy)propyl-2-(triprop-2-ynylammonio)eth yl phosphate of the following formula (2)

In step (a), preferably, a phospholipid derivative modified by terminal alkyne functional group(s) is mixed to the other liposome-forming lipids in an amount of from 3 to 25 mol% relative to the total lipid content. Then the liposome formulations are prepared using any of a regular method.

In step b) of the process, the reaction between the alkyne groups of the liposome and the ¹⁸ F- labelled azide derivative is preferably a copper(l) ion-catalysed cycloaddition reaction. The ¹⁸ F-labelled azide derivative used in the process is preferably an [ ¹⁸ F]fluoroalkyl azide, more preferably [ ¹⁸ F]fluoroethyl azide.

The ¹⁸ F isotope-labelled liposomes are preferably recovered from the labelling reaction mixture by filtration on a gel column.

The present invention further provides liposomes prepared according to the process of the present invention.

The ¹⁸ F-labelled liposomes of the present invention can be used for diagnostic purposes. More specifically, they are suitable for studying their pharmacokinetic behaviour in vivo by PET imaging. Accordingly, the present invention also provides the above liposomes for diagnostic use.

The invention also provides novel compounds of general formula (A) for the production of liposomes of the invention

(A) wherein

Ri and R ₂ are the same or different and each represent an acyl group derived from a saturated fatty acid; and n represents 1, 2, 3, 4, 5, 6; and compounds of general formula (B)

(B) wherein

Ri and R ₂ are the same or different and each represent an acyl group derived from a saturated fatty acid; and

Y represents an alkynyl group.

Description of the drawings

Figure 1 shows the structure of liposomes suitable for ¹⁸ F-labelling, illustrated by a schematic diagram. On the figure, (A): phosphatidylcholine; (B): PEG-phospholipids; (C): alkynated phospholipids.

Figure 2 shows the labelling of DPPEP-containing (0.22 μιηοΙ/ιτιΙ DPPEP concentration in the reaction mixture) liposomes with [ ¹⁸ F]FEA, the labelling efficacy is shown on the figure in function of the concentration of CuS0 ₄ in the reaction mixture, using THPA or BPDS. Figure 3 shows the HPLC-SEC chromatograms (2AD1: radioactivity signal) of the following samples: (A) [ ¹⁸ F]FEA stock solution; (B) reaction mixture according to Example 2, i.e., the reaction mixture of sterically stabilized liposomes with a DPPEP concentration of 4 mg/ml according to the THPTA protocol; (C) reaction mixture according to Example 3, i.e., the reaction mixture of sterically stabilized liposomes with a DPPEP concentration of 4 mg/ml according to the BPDS protocol; (D) A ¹⁸ F-DPPEP- SSL-liposome sample purified on a PD-10 gel column.

Figure 4 illustrates the result of the purification of the ¹⁸ F-labelled DPPEP/SSL-liposomes of Example 2 on a PD-10 Sephadex G25 column. The figure shows the activity of the fractions obtained by the separation. T1-W6 1-1 ml, except T2 = 0.5 ml; T1-T3: eluate; L1-L3: ¹⁸ F-DPPEP/SSL liposome fractions; W1-W6: fractions of [ ¹⁸ F]FEA and other small molecules OM: residual activity on the column.

Figure 5 illustrates the kinetics of the cycloaddition reaction between [ ¹⁸ F] FEA and DPPEP-lipid at 60°C. The labelling efficiency is shown on the figure in function of the reaction time.

Figure 6 illustrates the biodistribution study of the ¹⁸ F-DPPEP-lipid. Dynamic in vivo distribution of ¹⁸ F-DPPEP-lipid between 0 and 60 minutes after i.v. administration in healthy mice. Figure 7 illustrates the biodistribution study of the ¹⁸ F-DPPEP-liposomes of Example 1. Dynamic in vivo distribution of ¹⁸ F-DPPEP-liposomes (sterically non-stabilized) between 0 and 60 minutes after i.v. administration in healthy mice.

Figure 8 illustrates the biodistribution study of the ¹⁸ F-DPPEP-SSL-liposomes of Example 2. Dynamic in vivo distribution of ¹⁸ F-DPPEP-SSL-liposomes (sterically stabilized) between 0 and 60 minutes after i.v. administration in healthy mice.

Figure 9 illustrates the biodistribution study of the ¹⁸ F-DPPETP-SSL-liposomes of Example 4. Dynamic in vivo distribution of ¹⁸ F-DPPETP-SSL-liposomes (sterically stabilized) between 0 and 60 minutes after i.v. administration in healthy mice.

Detailed description of the invention In the description, unless otherwise indicated, the expressions "liposome", "liposomal formulation", "liposomal delivery system", and the like, refer to liposomes suitable for carrying drugs, and here we mean both so-called model liposomes, which do not contain an active pharmaceutical ingredient, and liposomes containing an active pharmaceutical ingredient. The stable double layer membrane of liposomes can typically be formulated from a phosphatidylcholine and cholesterol. If a phospholipid containing polyethylene glycol chain (PEG) is also mixed to the other lipid components, a special liposome will be obtained, with said hydrophilic PEG chains located on its surface. It has been observed that such so-called sterically-stabilized liposomes (SSLs) containing PEG reside in the bloodstream for a longer time period in vivo, than their counterparts without PEG chain. Both types of liposomes may be suitable as carriers for certain therapeutic agents, therefore, we have prepared samples containing different concentrations of terminal alkyne groups for both types of liposomes.

According to the present invention, such phospholipid derivatives are built into the membrane of the liposomes that contain alkyne group(s) in the head of the molecule. Such derivatives are referred to herein as "phospholipid derivatives modified by terminal alkyne functional group(s)" or shortly "alkynated lipids", "alkyne-derivatives" or "docking lipids". Alkyne derivatives derived from phospholipids which are structurally close to or same as the other phospholipids that constitute the liposome, are preferred. Since some kind of phosphatidylcholine is typically used as a building block for the liposomes suitable to carry active substances, alkyne derivatives derived from phosphatidylcholines are particularly preferred.

Phospholipid derivatives suitable for modifying liposomes by alkyne group(s) are prepared by coupling one or more terminal alkyne groups to the ethanolamine moiety of the phospholipid through spacers having various structures and chain lengths. Suitable compounds are known to those skilled in the art. Examples for compounds as sources of alkyne moiety are 3-(prop-2- ynyloxy)propanoic acid succinimidyl ester or propargyl bromide.

Particularly preferred phospholipid derivatives modified by terminal alkyne functional group(s) are compounds of general formulae (A) and (B):

(A)

(B) wherein x and R ₂ are the same or different and each represent an acyl group derived from a saturated fatty acid; n represents 1, 2, 3, 4, 5, 6; and Y represents an alkynyl group.

In the context of the present invention, the "acyl group derived from a saturated fatty acid" can have a straight chain with 10 to 18 carbon atoms, preferred are decanoyl, lauroyl, myristoyl, pentadecanoyl, palmitoyl, heptadecanoyl, stearoyl groups.

The alkynyl group may be a straight chain or branched chain group with 2 to 6 carbon atoms, having a triple bond at the end of the chain, preferably ethynyl, propargyl or butynyl group.

A particularly preferred compound of general formula (A) is (R)-2,3-bis(palmitoyloxy)propyl-2-(3- (prop-2-ynyloxy)propaneamido)ethyl phosphate (hereinafter referred to as DPPEP) having the following structural formula (1):

A particularly preferred compound of general formula (B) is (R)-2,3-bis(palmitoyloxy)propyl-2- (triprop-2-ynylammonio)ethyl phosphate (hereinafter referred to as DPPETP) having the following structural formula (2):

(2)

Compounds of formula (A) and (B) are novel and are part of the invention.

Compounds of formula (A) can be prepared according to the following Scheme 1 (the preparation of said compounds is further illustrated in Reference Example la):

Scheme 1:

Compounds of formula (B) can be prepared according to the following Scheme 2 (the preparation of these compounds is further illustrated in Reference Example lb): Scheme 2:

By using the alkynated lipids described above in a certain concentration range such liposomes can be prepared whose membrane composition and size differ only to a negligible extent from the corresponding parameters of commercially available liposomal formulations. This process is carried out by adding alkynated lipids to the other lipids constituting the liposome, expediently in an amount of between 3 and 25 mol% relative to the total lipid content, then liposomes are prepared by one of the methods known in the art, preferably by a thin layer hydration method and subsequent extrusion. It is not advisable to use alkynated lipids in larger quantities because in too high concentration they may have a perceptible influence on the structure of the liposome. On the other hand, using smaller molar ratio of them, too few alkyn functional groups would be available on the surface of the liposomes for the implementation of the ¹⁸ F-labelling.

The schematic drawing of PEG-containing (i.e., sterically stabilized) liposomes containing phospholipid derivatives modified by terminal alkyne functional group(s) is shown in Fig 1.

It has been found that the thus obtained liposomes with alkyne binding sites (so-called docking groups) on their surface react with fluorinated azide derivatives containing ¹⁸ F isotope under mild reaction conditions in an aqueous solution such that the structure of the isotopically labelled liposomes does not differ significantly from the original liposome. The radioisotope labelling can thus be separated in time from the liposome formation process.

Preferred azide derivatives for the isotopic labelling are [ ¹⁸ F]fluoroalkyl azides having a straight or branched short alkyl chain (C ₂ -C ₆ ). Particularly preferred is [ ¹⁸ F]fluoroethyl azide (hereinafter referred to as [ ¹⁸ F]FEA). The preparation of [ ¹⁸ F]fluoroethyl azide is described in Bioconjugate Chem. 2007, 18, 989-993. A specific preparation process is also shown in the following Reference Example 2.

The cycloaddition between the liposomes containing alkyne group and the fluorinated azide derivative is carried out in the presence of copper(l) ions formed in the reaction mixture.

The formation of copper(l) ions in the reaction mixture is preferably achieved by in situ reduction of Cu(ll) ions (e.g. by ascorbic acid). However, the concentration of copper ions should not be too high because it would decrease the stability of the liposome, and therefore, the concentration of Cu(l) required is reduced in the reaction mixture to as low level as possible, preferably to the range of 10 to 40 mM, in such a way that the alkyne azide cycloaddition still take place with a good efficiency. This can be achieved by using a suitable complexing ligand during the reaction. Examples of such ligands include tris(triazolyl)methylamine and bipyridine/fenatroline derivatives. A preferred ligand is tris(3-hydroxypropyltriazolyl)methylamine (THPTA) or bathophenanthroline disulphonate (BPDS).

The reaction medium is an aqueous solution. Preferably carbonate, 4-(2-hydroxyethyl) -1-piperazine- ethanesulfonic acid (HEPES) or phosphate buffer, more preferably phosphate buffered saline (PBS) is used, at a pH ranging from 6.5 to 8.0, up to a concentration of 200 mM. Ideally, the same buffer is the most suitable, in which the liposome has been prepared. However, the use of Tris(hydroxymethyl)aminomethane (TRIS) buffer should be avoided as it reduces the catalytic action of copper ions. While maintaining the stability of the liposomes, a few percent (<10%) of organic solvents may be present in the reaction mixture (e.g. due to the acetonitrile content of the [ ¹⁸ F]fluoroalkyl azide solution). The reaction temperature is about 20-60°C, preferably about 60°C.

The reaction time is about 30 to 80 minutes, preferably about 45 minutes. The optimal reaction time for reacting specific reaction partners can be determined by a person skilled in the art by standard kinetic studies (an example of such a study is shown below). The lower limit of the reaction time is determined by the efficiency of the reaction; as to the upper limit, the half-life of the ¹⁸ F isotope (109.8 minutes) should be considered. The ¹⁸ F-labelled I iposomes can be recovered in a pure form from the labelling reaction mixture by simple filtration on a gel column, and may be isolated in any physiological solution (buffer) suitable for intravenous injection. According to our experience, the labelled liposomes remain stable in vitro during several hours (in accordance with the half-life of the ¹⁸ F isotope) as examined in calf serum. We have shown that the ¹⁸ F-labelled liposomes are suitable for measuring the biodistribution of such formulations in vivo, in suitable systems for small animal PET studies alone or combined with MRI or CT.

The invention will be further illustrated by the following reference examples and examples. The reference examples relate to the preparation of the reagents used for the process of the invention, such as alkynated phospholipids and [ ¹⁸ F]fluorethyl azide. The examples show procedures for the preparation and purification of ¹⁸ F-labelled liposomes as well as the biodistribution study of the prepared liposomes.

Preparation of reagents necessary for the ¹⁸ F labelling Reference Example 1: Synthesis of alkynated phospholipids.

Reference Example 1. A.: Synthesis of DPPEP [compound of formula (1)]

Scheme 3:

500 mg of l,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine (hereinafter: DPPE, 0.72 mmol) was mixed in 10 ml of chloroform/methanol 65:35 (v/v) mixture. To this was added 250 mg of 3-(prop-2- ynyloxy)propanoic acid succinimidyl ester (hereinafter: propargyl-NHS, 1.11 mmol) dissolved in 2.5 ml of chloroform/methanol 65:35 (v/v) mixture. While stirring at room temperature, further 1000 μΙ of 8.4% NaHC0 ₃ aqueous solution was added to the reaction mixture. After stirring briefly the solution cleared up, and a homogeneous solution was obtained. Stirring was continued for 6 hours at room temperature and the solvent was evaporated nearly to dryness under reduced pressure at 40°C. To the oily yellowish-white residue 20 ml of 0.5 M hydrochloric acid solution and 50 ml of diethyl ether were added. After mild stirring, the oily residue dissolved, both phases remained clear. After separation the ether phase was extracted with additional 3 x 20 ml of 0.5 M hydrochloric acid. The ether phase was then dried over sodium sulphate, was filtered, and was evaporated to dryness at 35°C under reduced pressure. The white fluffy product thus obtained (521 mg, 93% DPPEP) proved to be sufficiently pure based on HPLC-MS analysis.

For assessing the purity of DPPEP, Alltima HP C-18-Amide-3 μιη (150x2.1 mm) column was used. The flow rate of the mobile phase was 1 ml/min, while its composition was A: 10 mM NH ₄ OAc (pH 9.5), B: Acetonitrile, from 85% (B) to 98% (B) with linear gradient elution over 8 minutes. The elution time of DPPEP was between 4 and 4.5 minutes under these conditions. The ^X H and ¹³ C NM measurement of DPPEP confirmed the structure of the molecule: ^X H NMR (CDCI3), d _1H (ppm): 0.88 (6H, t, J = 6.7 Hz); 1.27 (48H, m); 1.60 (4H, m); 2.31 (2H, t, J = 7.6 Hz); 2.33 (2H, t, J = 7.6 Hz); 2.49 (1H, t, J = 2.6 Hz); 2.55 (2H, t, J = 6.4 Hz); 3.54 (2H, t, J = 5.0 Hz); 3.82 (2H, t, J = 6.4 Hz); 4.05 - 4.20 (7H, m); 4.35 (1H, dd, J = 12.0, 4.1 Hz); 5.24 (1H, m); 6.13 (2H, br.s); 7.02 (1H, br.s). ¹³ C NMR (CDCI3), d _13C (ppm): 14.1; 22.7; 24.8; 29.0; 29.1; 29.3; 29.5; 29.6; 29.7; 31.9; 34.0; 34.2; 36.4; 39.8 (d, ³ J _C , _P = 6.5 Hz); 58.3; 61.9; 65.1 (d, ² J _C , _P = 5.6 Hz); 65.8; 66.3 (d, ² J _C , _P = 5.9 Hz); 69.4 (d, ³ J _C , _P = 7.5 Hz); 75.0; 79.2; 172.2; 173.1; 173.4.

Reference Example 1. B.: Synthesis of DPPETP [compound of formula (2)] Scheme 4:

To 300 mg (0.434 mmol) of l,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine (DPPE) 40 mL of THF was added in a two-necked round bottom flask. The solution was heated to 60°C with stirring for approx. 15 minutes, and a slightly opaque solution was obtained. To the reaction mixture further 0.2 ml of a solution of 80% propargyl bromide (1.9 mmol) in toluene and 1 g of anhydrous potassium carbonate (7.2 mmol) were added. The temperature of the reaction mixture was raised to about 70°C (reflux) and kept at this temperature for one day. The next day, the solution was filtered by washing with hexane and then concentrated under reduced pressure. 150 mg of the resulting brownish oily crude product was dissolved in 400 μΙ of hexane, which was purified on a silica gel column by washing with hexanes followed by chloroform/methanol (65:35) elution. The product obtained after evaporation of the solvent (70 mg, mixture of quaternary and tertiary amines 10:1) was used without further purification to prepare DPPETP-containing liposomes.

DPPETP was identified by chromatography on Alltima HP C18-Amide 3μ (150x2. lmm) column A: 10 mM NH ₄ OAc (pH 8)/B: MeCN at a flow rate of 0.5 ml/min, from 93% (B) to 98% (B) with linear gradient elution over 17 minutes, t _DPPETP = 9.0-9.2 minutes

Reference example 2.: Synthesis of [ ¹⁸ F]fluoro-ethylazide by SN ₂ -type nucleophil fluorination

The ¹⁸ F isotope required for the labelling was prepared with Eclipse-HP (Siemens) cyclotron, by irradiating water enriched in ¹⁸ 0 (> 95%) with protons of HMeV energy, by ¹⁸ 0(p,n) ¹⁸ F nuclear reaction. During the experiments, typically during a 30 minute irradiation period, with a 60 μΑ beam intensity, 35 to 40 GBq ¹⁸ F isotope was obtained. The resulting ¹⁸ F ions were separated from the enriched water on a Chromafix 30-PS-HCO ₃ ^~ anion exchange column, using a mixture of 11.6 mg/ml K ₂ C0 ₃ and 36.6 mg/ml Kryptofix 2.2.2 in 50% water-acetonitrile for the elution. The obtained ¹⁸ F ions were dried in three evaporation steps by adding a small amount (3 x 100 μί) of dry acetonitrile. The toluenesulfonic acid 2-azidoethyl ester (Ts-AEE) precursor compound required for the radiochemical synthesis and the inactive version of flouroethylazide, which served as reference compound in the HPLC assay, were prepared based on the method described by Kubus et al. (Applied Radiation and Isotopes, 2009; 67: 1977-1984).

In the radiochemical synthesis of [ ¹⁸ F]FEA, 1 ml of a 6,5 mg/ml solution of Ts-AEE precursor in acetonitrile was used, the labelling was continued at 105°C for 10 minutes in a closed reaction vessel. The resulting [ ¹⁸ F]FEA was then distilled at 85°C in 8 minutes into a collecting vessel containing 4 ml of water, by means of a nitrogen gas stream. During the distillation, we sought the lowest distillation temperature and the shortest time, wherein the majority of the more volatile [ ¹⁸ F]FEA was already distilled (typically 10 to 15 GBq, about 60% decay corrected yield) but the concentration of acetonitrile in the distillate remained still acceptably low (<20% v/v). (For more sensitive click reactions, the [ ¹⁸ F]FEA solution can be further purified on a C-18 SepPak Plus cartridge, so the acetonitrile content can be reduced below 1 v/v%.)

The identification and radiochemical purity of [ ¹⁸ F]FEA was investigated by HPLC on a reverse phase column (Accucore P-MS 150x2.1 mm, 2.6 μιη) with a linear gradient elution of 10 min. The eluent was a mixture of acetonitrile and (pH 9.3) 10 mM ammonium acetate, in which the ratio of acetonitrile increased from the initial 40% to 70% in 10 minutes. The mobile phase had a flow rate of 0.4 ml/min. Detection was performed with UV detector (226 nm) and flow-cell gamma detector (R _t, = 1-9 min). In the [ ¹⁸ F]FEA solution, due to the distillation step, neither the Ts-AEE precursor (R = 5.1 min) nor the [ ¹⁸ F]fluoride ion (R _t, 1-1 rnin) could be detected by radio-HPLC. The amount of acetonitrile that distilled together with [ ¹⁸ F]FEA was measured by gas chromatography.

Examples for the preparation of the ¹⁸ F-labelled liposomes

Step 1: Preparation of liposomes containing alkyne groups on their surface

Basically, two different type of liposomal solutions were prepared in order to model both sterically non-stabilized and sterically stabilized liposomal formulations. Composition of sterically non-stabilized liposomal formulations:

The first type of sample contained HSPC (hydrogenated soy lecithin) and cholesterol in a 3:1 mass ratio and a total concentration of 12.8 mg/ml, besides which the concentration of "docking" lipids (DPPEP or DPPETP) were varied between 0.5 and 4 mg/ml. This lipid composition modelled the sterically non-stabilized liposomal formulations. It has been found that incorporating the "docking" lipids into the liposomes, even in 4 mg/ml concentration, does not prevent the formation of spherical liposomes.

Composition of sterically stabilized liposomal formulations:

The second type of sample, besides HSPC and cholesterol, contained polyethylene glycol-containing phospholipid DSPE-PEG2000, which modelled the so-called sterically stabilized liposomal formulations (in a mass ratio of 1.75:1:1, and a total concentration of 12 mg/ml). The concentration of "docking" lipids (DPPEP or DPPETP) was varied between 0.3 and 4 mg/ml.

Preparation of the liposomes: When preparing both types of liposomes, lipids of the above composition were dissolved in chloroform, and after mixing (homogenization), the solvent was evaporated under reduced pressure. The residual solvent was removed by storing in vacuum oven for 12 hours. The thus-dried lipid film layer was hydrated in 10 mM PBS at 60°C, after which the samples were formed by freezing and reheating treatment. The size and shape of the thus-formed liposomes were then standardized by extrusion, by repeatedly passing through a 100-nm polycarbonate filter.

The size of the spherical liposomes produced by extrusion was 100-120 nm based on dynamic light scattering (DLS) measurements for both types of liposomes.

The sterically stabilized liposomes (containing PEG-lipid) containing sufficiently low concentrations (preferably < 4 mg/ml) of alkynated "docking" lipids can be made well before use (in the preceding days, weeks or months) as they can be stored aseptically for several months at 5°C without any significant structural changes. Therefore, by using this method, radioactive labelling can be separated in time from the preparation of liposomes. In the case of sterically non-stabilized liposomes with higher DPPEP-content (above 2 mg/ml concentration), the radioactive labelling should be started within one week after their preparation, as slow decomposition of the prepared "docking" lipid- containing liposome has been observed even with storage at 5°C for more than one week.

Step 2: Labelling liposomes containing alkyne groups on their surface with [ ¹⁸ F]fluorethyl azide

Determination of the click reaction conditions

It is an empirical fact that the liposomes will remain stable over a period of time (several hours) at about 60°C in aqueous solution. In order to determine how long DPPEP-containing liposomes should be reacted with ¹⁸ FEA at this temperature, reaction kinetic studies were performed with DPPEP lipid.

10 mg DPPEP was dissolved in 100 μΙ dichloromethane/methanol (2: 1 v/v) mixture, then 25 μΙ of this solution was added to 100 μΙ of a copper sulphate and THPTA solution (which was prepared prior as follows: two volume parts of an aqueous solution of 22 mg/ml (50 μιηοΙ/ml) of THPTA, one volume part of an aqueous solution of 3.2 mg/ml (20 μιτιοΙ/ml) of copper sulphate was mixed), then 500 μΙ of ¹⁸ FEA aqueous solution and finally 200 μΙ of a 20 mg/ml sodium ascorbate solution was added to the mixture. The homogeneous reaction mixture was distributed in 80-μΙ aliquots, and the samples were incubated at 60°C for 5-60 minutes. In the samples, the ratio of ¹⁸ FEA to the resulting ¹⁸ F-DPPEP was determined by radio-HPLC separation. The chromatographic column was Alltima C18-Amide (150x2. lmm, 3 μιη). At a flow rate of 1 ml/min, a gradient elution was performed from 85:15 A:B to 90:10 A:B over 4 minutes (A: 10 mM NH ₄ OAc (pH 9.18), B: 100% acetonitrile). With these parameters, the retention time of ¹⁸ FEA was 0.5 minutes and the retention time of ¹⁸ F-DPPEP was about 3.0 minutes. Figure 5 shows the percentage of ¹⁸ F-DPPEP, relative to the cumulative area under the peaks of ¹⁸ FEA and ¹⁸ F-DPPEP. In the cycloaddition reaction between the azide group of ¹⁸ FEA and the alkyne group of DPPEP, based on the above reaction kinetic experiments, approximately 45 minutes of reaction time was required at 60°C for a sufficiently high labelling efficiency. Longer reaction time was not chosen due to the short half-life of the ¹⁸ F isotope (109.8 minutes).

Realisation of the radioactive labelling

The labelling of liposomes containing alkaline groups with [ ¹⁸ F]FEA ( ¹⁸ F isotope) was accomplished by copper(l) catalysed cycloaddition. Copper(l) ions were obtained in situ in the reaction mixture by reduction of copper(ll) ions with ascorbic acid. THPA and BPDS were used to accelerate the reaction. For both accelerating compounds, an optimized procedure was developed, as described below.

THPTA protocol

Scheme 5:

PEP

According to the literature, THPTA is recommended to be used in a five-fold molar excess over the amount of copper ions (Hong V et al.: Bioconjug Chem. 2010; 21(10):1912-6), this condition was followed in each experiment, but the number of alkyne groups on the outer surface of the liposome, which are available for the click reaction, is proportional to the amount of DPPEP incorporated into the liposome. Therefore, the optimal Cu/DPPEP ratio and the minimum amount of DPPEP in the liposome were determined so that the ¹⁸ FEA could be coupled to the liposomes in an acceptable ratio.

Figure 2 shows that, when using THPTA, one should remain within a relatively narrow copper concentration range to achieve the highest labelling efficiency, which was 0.04 to 0.08 μιηοΙ/ml in the reaction mixture.

By increasing the amount of DPPEP in the liposome membrane, the highest labelling efficiency (>90%) was obtained when the DPPEP concentration in the reaction mixture was close to 2 μιηοΙ/ml. For this higher DPPEP content higher concentration of copper ions (0.467 μιηοΙ/ml) was required, while the THPTA concentration was 2.33 μιτιοΙ/ml and the Na-ascorbate concentration was 13 μιτιοΙ/ml. (The Na ascorbate was always used in significant molar excess.)

After optimization, the click reaction between DPPEP-containing liposomes and ¹⁸ FEA was carried out with THPTA as follows (hereinafter referred to as the THPTA protocol):

To two volume parts of an aqueous solution of 22 mg/ml (50 μιτιοΙ/ml) of THPTA, one volume part of an aqueous solution of 3.2 mg/ml (20 μιτιοΙ/ml) of copper sulphate was added and, after shaking, 175 μΙ of the pure deep blue solution was added to 1 ml of a solution of DPPEP liposomes (e.g. of a concentration of 4 mg/ml) taken up in 10 mM PBS. To the liposomal solution thus prepared, 1.0 ml of ¹⁸ FEA aqueous solution containing also 10-20% (v/v) acetonitrile was added. The radioactive concentration of the ¹⁸ FEA solution used in the reaction ranged from 1.85 to 3.70 GBq/ml. Next, 325 μΙ of a sodium ascorbate aqueous solution with a concentration of 20 mg/ml (100 μιτιοΙ/ml) was added to the reaction mixture, then the reaction mixture was kept in a sealed vessel at 60°C for 45 minutes.

The above labelling protocol proved to be suitable for labelling both sterically-not-stabilized and sterically-stabilized, PEG-chain-containing liposomes (DPPEP-SSL liposomes) with ¹⁸ FEA. BPDS protocol Scheme 6:

^* N=N ⁺ =N ^"

60 °C, 45 mm.

Cu(I)

BPDS

To two volume parts of 10 mM PBS, one volume part of an aqueous solution of BPDS with a concentration of 11.9 mg/ml (20 μιηοΙ/ml) and one volume part of an aqueous solution of copper sulfate with a concentration of 3.2 mg/ml (20 μιηοΙ/ml) was added, and then, after shaking, 400 μΙ of the slightly opalescent solution was added to 1.2 ml of a solution of SSL liposomes (e.g. with a concentration of 4 mg/ml DPPEP or 0.5 mg/ml DPPETP) taken up in 10 mM PBS. To the liposomal solution thus prepared, 0.8 ml of an ¹⁸ FEA aqueous solution containing also 10-20% (v/v) acetonitrile was added. The radioactive concentration of the ¹⁸ FEA solution used in the reaction ranged from 1.85 to 3.70 GBq/ml. Next, 500 μΙ of a sodium ascorbate aqueous solution with a concentration of 20 mg/ml (100 μιηοΙ/ml) was added to the reaction mixture, then the reaction mixture was kept in a sealed vessel at 60°C for 45 minutes. The above labelling protocol did not result in a high efficiency labelling when labelling sterically non- stabilized liposomes with ¹⁸ FEA, but it was advantageously applicable for labelling sterically stabilized liposomes containing PEG chains and DPPETP lipids (DPPETP-SSL liposomes).

By using the above procedures, liposomes having, among others, the following compositions have been prepared:

The reaction mixtures obtained according to Examples 2 and 3 have been used without purification in assessing the efficiency of the ¹⁸ F-labell ing as described below and shown in Figure 3 (chromatograms B and C). The products according to Examples 1, 2 and 4 have been recovered from the reaction mixture as described below, and were thus used in the following biodistribution studies.

Assessing the efficiency of the ¹⁸ F-labelling

In the variously labelled liposome samples the efficiency of the labelling, i.e. the effectiveness of the reaction between the alkyne group of DPPEP (or DPPETP) and the azide group of ¹⁸ FEA, was measured by size exclusion chromatography (HPLC-SEC) of the samples taken from the reaction mixture cooled back to room temperature after the incubation at 60°C. The gel column employed was Zorbax GF-250 (4.6x250mm 4 μηη), the mobile phase was 100% 10 mM PBS (pH 7.2) and the flow rate was 0.6 ml/min (Shimadzu LC20ADX HPLC). The separated components were detected by a photodiode array detector (PDA) and a flow cell radioactivity detector connected in series. Liposomes with a diameter of about 100 nm were first eluted from the gel column near the exclusion volume (with an apparent volume partition coefficient of K _av = 0.1). The F-18 labelled liposomes are detected on the radiochromatogram with the same K _av as liposomes are detected on UV- chromatogram, while the ¹⁸ FEA (along with other small molecules) is eluted closer to the total volume of the column (with an apparent volume partition coefficient of K _av = 0.65). Figure 3 illustrates such a separation. The chromatograms belong to the following samples: (A) ¹⁸ FEA stock solution; (B) reaction mixture of sterically stabilized liposomes with a DPPEP concentration of 4 mg/ml according to the THPTA protocol (Example 2); (C) reaction mixture of sterically stabilized liposomes with a DPPEP concentration of 4 mg/ml according to the BPDS protocol (Example 3); (D) ¹⁸ F-DPPEP-SSL liposome sample purified on a PD-10 gel column as described below. The chromatograms confirm that the labelling takes place with a high efficiency (chromatograms (A) and

(B) ); in case of DPPEP-SSL liposomes, the THPTA protocol is more efficient (chromatograms (B) and

(C) ); and gel separation results in pure ¹⁸ F-DPPEP-SSL liposomes (chromatograms (B) and (D)).

Recovering the liposomes from the reaction mixture The liposomes were recovered from the reaction mixture as follows: the total volume of the 2.5-ml liposomal reaction mixture was, after cooling down to 30°C, applied to a PD-10 Sepadex G25 column, which was previously conditioned with sterile 10 mM PBS (pH 7.2). The reaction mixture was allowed to completely drip through the gel column (gravitationally), then, by washing the column with 3.0 ml of 10 mM PBS, the purified ¹⁸ F-labelled liposomal fraction was obtained. In order to determine the effectiveness of the separation, the column was further washed with 6.0 ml of 10 mM PBS to give the fraction containing mainly small molecules, in which a significant fraction of the unreacted ¹⁸ FEA was also present. As a result of the purification step, the labelled liposomal fraction had a radiochemical purity of >98% (chromatogram (D) of Figure 3) in 10 mM PBS. The separation of the activity of ¹⁸ F bound to liposomes during the purification steps can be seenon Figure 4 as well as it is obvious that, as a result of the click reaction carried out with THPTA and the subsequent separation on a PD-10 column, nearly 80% of the activity of ¹⁸ F could be recovered in liposome-bound form.

Biodistribution studies

In order to evaluate the in vivo applicability of the liposomes labelled as described above, the labelled liposomes were subjected to bio-distribution studies in small animal model experiments. Healthy house mice (mus musculus) grown for laboratory purposes, from Balb/C and Nu/Nu ('nude') strains, were included in biodistribution studies. The liposomes included in the studies were: (not SSL), 4 mg/ml DPPEP liposome (Example 1); 4 mg/ml DPPEP-SSL liposome (Example 2) and 0.5 mg/ml DPPETP-containing SSL liposome (Example 4). The preparation of these formulations was performed on the day of the studies by the cycloaddition of the alkyne group formed on the surface of the corresponding liposomes with ¹⁸ FEA, as described above. The labelled liposomes were always isolated from the reaction mixture by preparative gel chromatography (PD-10 Sephadex G25 column) so that the tracer was present in a radiochemical purity of >98% in the sterile isotonic saline solution for injection. The liposomes containing the DPPEP and DPPETP alkynated "docking" lipids at the desired concentrations were prepared in days, weeks or even months prior to the studies. The animals (n = 2-3 per group) received the radioactively labelled formulations (6 ± 3 M Bq/animal) after intravenous cannulation, in an anesthesia with a gas mixture of 2% isoflurane and 98% pure oxygen.

PET imaging of the animals was performed using NanoScan ^® PET/M I (Mediso) apparatus in dynamic mode. In all cases, the tests were performed with continuous PET data recording, initiated at the moment just before administration. The PET equipment collected the (raw) data continuously for 60 minutes after the start of administration, then the reconstruction algorithm formed the radioactivity data collected in the protocol timeslots into a three-dimensional spatial distribution frame ("PET frame").

From the results obtained, the following conclusions can be made. (i) The F-18 label remains stable in vivo for each formulation. So there was no visible bone accumulation that would indicate the presence of unbound ¹⁸ F isotope. The ¹⁸ F-DPPEP lipid is eliminated from the blood stream within a short period of time and is mainly accumulated in the liver (Figure 6).

(ii) Sterically non-stabilized liposomes, although with relatively slow kinetics, are transferred to the liver and spleen from the bloodstream, and after 30 minutes the majority of the activity can be found in these two organs (Figure 7).

(iii) The model liposomes containing polyethylene glycol chain (PEG), however, typically remain in the bloodstream and after 20 minutes only a slow decrease (largely due to the physical decomposition of the isotope) is observed (Figures 8 and 9). The new labelling method can be widely used for different liposome formulations. Thus the pharmacological behaviour, even of liposomes modified on their outer surface (e.g., with an antibody or peptide, suitable for targeted tumour therapy), can also be studied in the early stages of the development of the optimal drug formulation. In addition, the investigation of modified liposome structures with the most varied biodistribution can be realized owing to the fact that it is not necessary to incorporate the steps of radioactive labelling directly into the preparation process of the planned carrier liposome. Due to the post-preparative realisation of the radioactive labelling process used, this method can be applied in practice in an incomparably easier way than the hitherto known direct technique, that is, labelling during the preparation of the liposome.