EP0364016A1 - Polyphosphazene polymers and their preparation - Google Patents

Abstract

This invention relates to new polyphosphazene polymers and a process for their preparation. The new polyphosphazenes have different side chains coupled to the phosphorus atoms of the -N=p- main chain via nitrogen atoms, said side chains including imidazolyl groups, aromatically substituted groups and, if required, non-aromatically substituted groups. The polymers may be used in several forms as implant for gradual drug release.

Description

This invention relates to new polyphosphazene polymers and to a process for their preparation.

Polyphosphazenes are high-molecular polymers consisting of a chain of alternately nitrogen and phosphorus atoms separated by formally alternating single and double bonds, two side chains being linked to each phosphorus atom. Polyphosphazene polymers therefore consist essentially of recurring units of formula 1, wherein R is a side chain consisting of an organic group. The polymers may be diagrammatically represented by the general formula N=PR₂]n wherein n is a large number in the order of 10³-10⁵, e.g. n = 15000.

For the preparation of polyphosphazene polymers the starting material is usually poly(dichlorophosphazene) polymers which, in turn, have been obtained by ring opening polymerization of hexachlorocyclotriphosphazene. The chlorine atoms linked to the phosphorus atoms in the poly(dichlorophosphazene) polymers may be replaced in different macromolecular substitution reactions by all types of different organic side chains. For general information concerning polyphosphazenes and their preparation reference is made to Allcock in "Phos­ phorus-Nitrogen Compounds", Academic Press, New York and London (1972).

For different reasons polyphosphazene polymers are a promising group of chemical materials, especially in view of biomedical uses. A favourable aspect of polyphosphazene polymer is, for instance, that polymers having a great variety of chemical, physical and biological properties can be prepared by simply varying the nature of the organic side chains R: see Allcock et al., Makro­ molecules 21, 323 (1988). Furthermore, various known polyphosphazene polymers have been found to be biodegradable or bioerodible into harmless products and to have good biocompatibility. Consequently, it has been suggested to use polyphosphazenes as a bioerodible matrix for pharmacologically active agents (hereinafter called drugs) in the form of implantable articles or microspheres, implants for short, which ensure a gradual release of the drug enclosed in the matrix. In addition to these so-called monolithic systems in which drug and polymer are present in physically mixed condition, so-called pending chain release systems have also been proposed, in which the drugs are chemically combined with the polymer in the form of side chains at the phosphorus atoms. In connection with these possible uses of polyphosphazenes reference is made to Allcock, Makromol.Chem., Suppl. 4, 3 (1981); Laurencin et al., J.Biomed.Mat.Res. 21, 1231 (1987); and Goedemoed and de Groot, Makromol.Chem., Macromol.Symp. 19, 341-365.

The main drawback of most of the known polyphosphazene polymers is, however, that they show disappointing degradation characteristics in a biological environment.

A known category of polyphosphazene polymers consists of, e.g., polyphosphazenes substituted by one or several amino acid ethyl esters (see Goedemoed and de Groot, loc.cit.). For this category of polyphosphazenes a hydrolysis mechanism has been proposed in which a free carboxyl group formed by hydrolysis of the carboxylic ester group attacks the polymer chain, resulting in the formation of phosphazenes and ultimately leading to chain scission. However, after flow-through eluation analysis, it appeared that the degradation rate caused by hydrolysis is unsatisfactorily low. For a polyphosphazene substituted by the relatively hydrophilic glycine ethyl ether a percentage of degradation of only 0.4% after one week eluation was found for an article of 100 mg at pH 7.0. If it is allowed to extrapolate this result, it will take more than 6 years before all glycine moieties are released and the polymer is degraded.

Polyphosphazenes which, in addition to glycine ethyl ester substituents, also contained phenylalanine ethyl ester or glutamic acid diethyl ester side chains at the phosphorus atoms were found to show an even lower rate of degradation. This is probably due to their relatively hydrophobic nature, as a result of which hydrolysis of the ethyl ester groups proceeds very slowly.

Laurencin et al., loc.cit., prepared and tested polyphosphazene polymers containing imidazolyl side chains and methyl phenoxy side chains. A polymer in which 20% of side chains R consisted of the imidazolyl group of formula 2 showed a low rate of degradation of 4% polymer in 25 days. In contrast, 30% of a polymer in which 45% of side chains R consisted of the imidazolyl group of formula 2 proved degraded after about 12.5 days. But then the rate of degradation decreased to such an extent that total degradation would require approximately 3 years.

In spite of the presence of imidazole groups which can easily be released from the phosphazene chain by means of hydrolysis these known polyphosphazene polymers do not show the degradation properties sought. This is probably due to the fact that the methyl phenoxy groups do not release easily.

The invention provides a new category of polyphosphazene polymers for realizing a wide range of degradation characteristics.

In the concrete, the invention provides a polyphos­ phazene polymer consisting essentially of recurring units of formula 1, wherein a part of side chains R consists of the imidazolyl group of formula 2, another part of side chains R consists of an aromatically substituted group of formula 3, wherein R₁ stands for H, F, C₁₋₄ alkyl or C₁₋₄ alkoxy, and X stands for a group according to any of formulae 5-7, wherein a is an integer of 0-2 and R³ stands for
-OC_bH_2b+1, -OCH₂CON(C_bH_2b+1)_cH_d or -N(C_bH_2b+1)_cH_d,
wherein b is an integer of 1-4, c and d are each an integer of 0-2 and c+d=2, and, if required, yet another part of side chains R consists of a non-aromatically substituted group of formula 4, wherein R² stands for a group according to any of formulae 8-9, wherein c is an integer of 1-2 and R⁴ has the same meanings as R³.

A preferred embodiment of the polyphosphazene polymer according to the invention is characterized in that 40-60% of side chains R consists of an aromatically substituted group of formula 3, 0-20% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

Most preferred is a polyphosphazene polymer according to the invention characterized in that 45-55% of side chains R consists of an aromatically substituted group of formula 3, 5-15% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

A polyphosphazene polymer according to the invention can be prepared in different ways but is preferably prepared by replacing in a poly(dichlorophosphazene) polymer first a part of the chlorine atoms by an aromatically substituted group of formula 3, then another part of the chlorine atoms by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms, if any, by a non-aromatically substituted group of formula 4.

The preferred procedure is that first 45-55% of the chlorine atoms in a poly(dichlorophosphazene) polymer is replaced by an aromatically substituted group of formula 3, then 80-99% of the remaining chlorine atoms is replaced by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms are replaced by a non-aromatically substituted group of formula 4.

By first replacing about half of the chlorine atoms in the poly(dichlorophosphazene) polymer selected as starting material by an aromatically substituted group of formula 3 it is ensured that in the following substitution with imidazole substantially no geminal imidazole substitutions occur. Geminal imidazole substitutions, i.e. the occurrence of 2 imidazolyl side chains at one phosphorus atom, are undesirable, because they lead to cross-linkings in which there are formed uninterest­ ing cross-linked products insoluble in organic solvents, such a tetrahydrofuran. In contrast to the non-aromatically substituted groups of formula 4, the aromatically substituted groups of formula 3 have a marked preference for non-geminal substitution, apparently because of their "bulky" character, as a result of which two of these voluminous groups will not fit to the same phosphorus atom.

The total replacement of the remaining chlorine atoms by imidazolyl groups requires a very intensive treatment, e.g., 48 hours refluxing in tetrahydrofuran. Such an intensive treatment is undesirable, partly because of the deterioration of the polymer resulting in lower molecular weights, lower intrinsic viscosities, etc. According to the invention this problem is avoided when, after a substantial part of the chlorine atoms has been replaced by imidazolyl groups, the still remaining chlorine atoms are replaced by a non-aromatically substituted group of formula 4. Although in this reaction imidazolyl groups that are already bound to the polymer chain are dislodged too, it has been found in practice that only a minor proportion of the imidazolyl groups is dislodged.

Examples of compounds that can be used to supply an aromatically substituted group of formula 3 are:
- L, D, and DL phenylalanine alkyl esters (obtainable in the form of the hydrochloride salt), especially the ethyl esters. For antimicrobial and anticancer applications the use of D-phenylalanine ethyl ester is advantageous over L-phenylalanine ethyl ester which may serve as nutrient source for microorganisms or tumour cells. The use of D,L-phenylalanine ethyl ester may be advantageous to obtain very regularly substituted phosphazene chains, thereby increasing the glass transition temperature (Tg) and obtaining improvements in the water penetration, the swelling behaviour, the degradation and drug release processes, and the production of microspheres. A weak point of the alkyl esters is, however, that they can hardly be hydrolyzed by water. A 55% phenylalanine ethyl ester -45% imidazole substituted polyphosphazene did not show any release of phenylalanine or phenylalanine ethyl ester during 10 days through-flow eluation.
-L, D, and DL p-fluorophenylalanine alkyl esters (obtainable in the form of the hydrochloride salt), especially the ethyl esters. These compounds are very suitable for antimicrobial or anticancer applications, because p-fluorophenylalanine as phenylalanine antagonist inhibits the protein synthesis.
- L, D, and DL phenylalaninamide (obtainable as free base). During the substitution reaction phenylalanin-amide will serve as a HCl acceptor and, in a tetrahydrofuran medium, lead to a white precipitate. When compared to the phenylalanine alkyl esters which need to be released from their hydrochloride salts by first refluxing in tetrahydrofuran with triethylamine, the above-mentioned compounds have the advantage that a very close adjustment of the degree of substitution is possible. Moreover, polyphosphazenes containing phenylalaninamide substituents are more hydrophilic than corresponding polyphosphazenes substituted by phenylalanine alkyl esters.
-L, D, and DL phenylglycine alkyl esters (available in the form of the hydrochloride salt). Phenylglycine is an unnatural amino acid and for this reason attractive for antimicrobial and anticancer applications. The absence of a -CH₂- bridge between the chiral carbon atom and the phenyl group means that the side chains are stiffer, which causes a higher Tg.
- L, D, and DL phenylglycinamide (available as a free base). The absence of an alkyl ester group leads again to a lower Tg. The availability as a free base is a clear advantage for the above reasons.
- L, D, and DL homophenylalanine alkyl esters (available in the form of the hydrochloride salt). The extra -CH₂- bridge between the chiral carbon atom and the phenyl group leads to a lower Tg.
- L, D, and DL homophenylalaninamide (available as a free base or in the form of the hydrochloride salt). The polymers have a Tg even lower than that with strenghened side chains.
- Dimethylglycolamide esters of L, D, and DL phenylalanine, phenylglycine and homophenylalanine (available in the form of the hydrochloride salt). These compounds have the important advantage that they are susceptible to enzymatic hydrolysis by (cholin)esterases, such as pseudocholin esterease, in a biological environment. After aqueous hydrolysis has led to the release of imidazole units, this enzymatic hydrolysis of the dimethyl­ glycolamide esters can stimulate the further degradation of the polyphosphazene polymer.
- α-amino-β - orγ-phenyl- γ -butyrolactone (available in the form of the hydrochloride salt). The hydrolyzable lactone ring increases the rate of degradation.

Examples of compounds that can be used to supply a non-aromatically substituted group of formula 4 are:
- glycine alkyl esters (available in the form of the hydrochloride salt). With, e.g., glycine ethyl ester a 100% substitution can easily be obtained, but an accurate quantitative substitution is difficult, because the alkyl esters are not available as free base.
- glycinamide.
Because this substance requires even less space than the glycine esters, it has a still higher substitution potential. Unfortunately, this substance, too, is not available as a free base.
-β-alanine alkyl esters and β-alaninamide (available in the form of the hydrochloride salts). These compounds are attractive again for antimicrobial and anticancer applications. The β-alaninamide is a more effective substituent than, e.g., β-alanine ethyl ester.
- glycine dimethylglycolamide ester and β-alanine dimethyl­ glycolamide ester (both available in the form of the hydrochloride salt). These compounds are preferred, because they are susceptible to enzymatic hydrolysis.
-α-amino-γ-butyrolactone (available in the form of the hydrobromide salt).
This compound gives a substituent occupying little space, which substituent is easily degraded by hydrolysis of the lactone ring to form a free α-carboxyl group.

Of course, one polyphosphazene polymer may simultaneously comprise different aromatically substituted groups of formula 3. The same applies to the non-aromatically substituted groups of formula 4. By way of example, side chains R consisting of an aromatically substituted group of formula 3 may derive from phenylglycine dimethyl­ glycolamide ester, homophenylalanine dimethylglycolamide ester, p-fluorophenylalanine dimethylglycolamide ester and p-fluorophenylalaninamide.

The above described polyphosphazene polymers according to the invention may be used for various purposes, but will be of special value to biomedical applications. In this connection they particularly seem appropriate for realizing biodegradable products showing a gradual drug release.

Starting from the polyphosphazene polymers and drugs, articles (tablets, discs, rods, etc.) or microspheres comprising a drug-containing matrix of polyphosphazene polymer can be manufactured and used as a biodegradable implant.

The invention, however, also provides the possibility of polyphosphazene polymers with chemically bound drugs, namely in the form of a polyphosphazene polymer according to the above definition, which is characterized in that yet another part of side chains R consists of a drug-carrying imidazolyl group of formula 10, wherein R⁵ is a drug bound to the imidazolyl ring via a spacer.

A preferred embodiment of such a polyphosphazene polymer according to the invention is characterized in that 15-30% of side chains R consists of an aromatically substituted group of formula 3, 20-35% of side chains R consists of a drug-carrying imidazolyl group of formula 10, wherein groups of formulae 3 and 10 together form 40-60% of side chains R, 10-40% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

Most preferred is a drug-carrying polyphosphazene polymer according to the invention, which is characterized in that 20-25% of side chains R consists of an aromatically substituted group of formula 3, 25-30% of side chains R consists of a drug-carrying imidazolyl group of formula 10, 15-35% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

Preferably, groups of formulae 2 and 10 together form 45-65% of side chains R, because in that case an optimum rate of hydrolysis and degradation of the polymer can be realized.

The drug-carrying polyphosphazene polymers according to the invention are preferably prepared by replacing in a poly(dichlorophosphazene) polymer first a part of the chlorine atoms by an aromatically substituted group of formula 3, then another part of the chlorine atoms by a drug-carrying imidazolyl group of formula 10, then yet another part of the chlorine atoms by the imidazolyl group of formula 2, and finally still remaining chlorine atoms, if any, by a non-aromatically substituted group of formula 4.

The preferred procedure is that 45-55% of the chlorine atoms in a poly(dichlorophosphazene) polymer is replaced by the groups of formulae 3 and 10 together, then 80-99% of the remaining chlorine atoms is replaced by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms are replaced by a non-­ aromatically substituted group of formula 4.

The drug-carrying imidazolyl group of formula 10 is like the aromatically substituted group of formula 3 a "bulky" group, so that a later geminal substitution of imidazole can be inhibited by replacing about half of the chlorine atoms in the starting material by these groups of formulae 3 and 10. In the last substitution reaction, in which the still remaining chlorine atoms are replaced by non-aromatically substituted groups of formula 4, a part of the drug-carrying imidazolyl groups is dislodged by those groups of formula 4, but it has been found in practice that the number of dislodged drug-carrying imidazolyl groups is limited.

Drug-carrying imidazole compounds that can be used to supply a drug-carrying imidazole group of formula can be obtained by reacting a drug with a suitable imidazole derivative. Examples of imidazole derivatives suitable for this purpose are:
- imidazole-4-acetic acid, which is a degradation product in the metabolism of histamine and is therefore presumed to be devoid of any pharmacological activity. Drugs with monofunctional groups can be coupled to the 4-acetic acid group, namely amino group-containing drugs via an amide bond, alcohol group-containing drugs via an ester bond, and carboxylic acid group-containing drugs via an anhydride bond.
-imidazole-4-acrylic acid (urocanic acid), which is also a metabolic degradation product (it is obtained by desamination of histidine by the enzyme histidine ammonia lyase) and is therefore presumed to be free of any pharmacological activity. Drugs can be coupled to the 4-acrylic acid group again, depending on the nature of the monofunctional group in the drug, via an amide bond, an ester bond or an anhydride bond. When compared with the -CH₂- spacer of the imidazole 4 acetic acid, the -CH=CH- spacer has the advantage of a larger distance between drug and imidazole ring, so that the imidazole group is less sterically hindered during its reaction with the phosphazene chain. Furthermore, the double bond has a favourable effect on the susceptibility to hydrolysis of the amide, ester or anhydride bond.
- 4-hydroxymethyl imidazole, which is known to be devoid of any pharmacological activity. Drugs with a single carboxylic acid group can be coupled to the 4-hydroxymethyl group via an ester bond.
-β-carnosine methyl ester of formula 11, a commercially available compound having promising properties as a spacer between the polyphosphazene chain and a drug with a carboxylic acid group, with the drug being coupled via an amide bond to the side chain in the 4 position of the imidazole ring. Interposed between the imidazole ring and the drug is a relatively long chain, so that the imidazole-phosphazene substitution reaction is hardly sterically hindered by the drug. It is further known that β-carnosine enhances antineoplastic activity of anticancer agents.
-β-carnosine dimethylglycolamide ester of formula 12. The advantage of dimethylglycolamide esters over alkyl esters has been mentioned before. When compared with -carnosine methyl ester, this spacer is more hydrophilic, which is very favourable to lipophilic drugs, such as Brequinar sodium.
- imidazole-4-acetic acid glycolic acid ester of formula 13, to which amino group-containing drugs (amide bond), alcohol group-containing drugs (ester bond) and carboxylic acid group-containing drugs (anhydride bond) may be coupled. By means of the additional glycolic acid ester group the distance between drug and imidazole ring is lengthened with respect to imidazole-4-acetic acid, while in the case of an amino group-containing drug (amide bond) a structure sensitive to enzymatic hydrolysis by esterases is present.
- urocanic acid glycolic acid ester with an even longer distance between drug and imidazole ring and a double bond which influences both the chemical and the enzymatic stability of the glycolic acid ester bond. - 4-hydroxymethylimidazole glycolic acid ester of formula 14, which ensures a longer distance between drug and imidazole ring with respect to 4-hydroxymethylimidazole and shows an increased sensitivity to esterases due to the extra ester bond.

Examples of drugs with monofunctional groups, appropriate for coupling to a polyphosphazene polymer via an imidazole spacer comprise:
drugs containing an amino group
- melphalan methyl ester of formula 15, a drug used in the treatment of multiple myeloma and melanoma; 3-cyclohexyl analogon of aminoglutethimide (formula 16 ) used for the treatment of estrogen-dependent breast cancer;
- levodopa methyl ester of formula 17 for the treatment of Parkinson's disease;
- various small peptides such as
Enkephalines
Tyr - Gly - Gly - Phe - Lau - NH₂ Leucine enkephalinamide
Gly - Gly - Phe - Leu - NH₂ Des-Tyr¹- enkephalinamide
Growth Hormone Raleasing Peptide
Tyr - Gly - D-Trp - Phe - D-Phe - NH₂
Opioid Peptide
Tyr - Pro - Phe - Pro - NH₂ Morphiceptin
Oxytocin fragment
Tyr - Pro - Leu - Gly - NH₂
Molluscan cardioexcitatory neuropeptide
Phe - Leu - Arg - Phe - NH₂
MSH - Release Inhibitor
Pro - Leu - Gly - NH₂
- diamino acids, such as 2,4-diamino butyric acid (treatment of malignant glioma cells),α-trifluoromethyl ornithine (treatment of small cell lung carcinoma, potentiation of other anticancer agents against antineoplastic diseases) and δ-hydroxy lysine (inhibitor of glutamine synthetase), in which the carboxyl group is esterified (e.g., methyl ester) and one of the amino groups is blocked by, e.g., an acetoxyethoxycarbonyl group which is easily hydrolyzed by water;
drugs containing an alcohol group
- metronidazole (formula 18) for the treatment of periodontal diseases;
- bambuterol of formula 19 (a prodrug of terbutaline) for the treatment of asthma;
- propranolol (formula 20) for the treatment of anxiety and stress;
- lorazepam (formula 21) for the treatment of anxiety;
- phenprocoumon (formula 22) for use as an anticoagulant agent;
- acenocoumarol (formula 23), also an anticoagulant agent;
- trans-4-hydroxy tamoxifen (formula 24) for the treatment of estrogen-dependent breast cancer;
- steroids, such as the progestins ethisterone (formula 25), norethindrone (formula 26), norethynodrel (formula 27), dimethisterone (formula 28), ethinylestrenol (formula 29) and norgestrel (formula 30), the estrogens -estradiol 17-acetate (formula 31) and mestranol (formula 32), the androgens testosterone (formula 33), methyltestosterone (formula 34) and danazol (formula 35) and the anabolic steroids calusterone (formula 36), ethylestrenol (formula 37), methandrostenolone (formula 38) and stanozolol (formula 39);
drugs containing a carboxylic acid group
- flavone-8-acetic acid (formula 40) for the treatment of solid tumours of colon, mamma, ovarium and pancreas;
- Brequinar sodium (formula 41), an antipyrimidine agent for the treatment of solid tumours;
- indomethacin (formula 42) for the treatment of rheumatic diseases;
- mefenamic acid (formula 43) for the treatment of rheumatic diseases;
- flufenamic acid (formula 44) for the treatment of rheumatic diseases;
- chlorambucil (formula 45) for the treatment of chronic lymphocytic leukemia and primary macroglobulinemia;
- pefloxacine (formula 46) for the treatment of periodontal diseases and other infectious diseases;
- ofloxacine (formula 47) for the treatment of periodontal diseases and other infectious diseases;
- and various penicillin and cephalosporin derivatives.

The preparation of polyphosphazene polymers according to the invention is preferably carried out according to the following prescriptions.

Prescription 1: Preparation of the starting polymer.

The starting material, poly(dichlorophosphazene)polymer, is prepared by thermal polymerization of hexachlorotri­ phosphazene. The melt polymerization is carried out in sealed glass tubes under low pressure (10⁻⁴^Torr) at a temperature of about 240°C. During the polymerization the glass tube tumbles continuously (60 rev./h), until the originally liquid content of the tube ceases to flow. Depending on the presence of impurity traces, polymerization times of 24-30 hours are used. The glass tube content is 20 g trimer. The polymer is separated from the remaining amount of trimer by stirring the crude polymerization mixture in 400 ml dry n-hexane and separating it from the insoluble polymer by decanting the solution. The n-hexane is evaporated and the polymer yield is calculated on the basis of the remaining trimer content. The remaining solid (polydichlorophosphazene) is dissolved in 500 ml dry tetrahydrofuran with stirring for several hours. The resulting liquid is filtered in order to remove glass particles and the insoluble cross-linked material, if present. The solution of linear poly(dichlorophosphazene) polymer is directly used for substitution reactions, while utmost care is taken to prevent reaction with moisture.

Prescription 2: Preparation of polyphosphazene matrix systems according to the invention. a) Substitution with the aromatically substituted "bulky" group.

The substitution reaction takes place by a nucleophilic attack of the free amino group in the starting material supplying the aromatically substituted group of formula 3. (This starting material is hereinafter called the aromatic amino acid derivative.)

Starting materials available as free bases, such as phenylalaninamide, phenylglycinamide and homophenyl­ alaninamide, are dissolved in 200 ml tetrahydrofuran. For each gram poly(dichlorophosphazene) polymer, calculated by measuring the remaining trimer content, 17.3 mmol aromatic amino acid derivative is taken (100% of the starting amount of chloride) together with 43.6 mmol triethylamine.

In case of aromatic amino acid derivatives in the form of hydrochloride salts, the free bases are first liberated by refluxing with 150% molar amount of triethylamine for 4 hours in 500 ml dry tetrahydrofuran. The liquid is filtered after cooling to room temperature. Either this solution or the above-mentioned solution obtained in case of aromatic amino acid derivatives available as free bases is added dropwise to the poly(di­ chlorophosphazene) solution, together with the triethylamine.

The substitution reaction is carried out under dry nitrogen atmosphere with stirring for 18 hours at room temperature. The resulting liquid is filtered, limiting exposure to the surrounding atmosphere as much as possible. The white triethylamine hydrochloride precipitate formed is dried and weighed. The substitution reaction is such that the degree of substitution is 50-55% of the amount of chlorine atoms available in the starting polymer.

b) Substitution with imidazole.

Based on the amount of chloride in the starting polymer, 100 mol% imidazole is taken and dissolved in 50 ml dry tetrahydrofuran with 200% triethylamine. The solution is added dropwise and stirred again for about 18 hours. The triethylamine hydrochloride precipitate is weighed again. The substitution reaction is such that the degree of imidazole substitution is above 40%.

c) Substitution with a starting bond supplying a non-­ aromatically substituted group of formula 4 (this compound is hereinafter called non-aromatic amino acid derivative).

Based on the amount of the starting polymer chloride, 5 mol% of the non-aromatic amino acid derivative is taken, together with 20% triethylamine. The free bases are first released from the hydrochloride salts by 4 hours refluxing in tetrahydrofuran with 150% triethylamine.

The substitution reaction is regulated in such a manner that the degree of substitution is 5-10% of the original amount of chlorine atoms. For the purpose of this regulation the final triethylamine hydrochloride precipitate is also weighed.

The solution in tetrahydrofuran resulting in this substitution reaction is partially evaporated, until a concentrated solution is obtained (100-150 ml). The polymer solution is precipitated dropwise with stirring in n-hexane. The solid polymer is dissolved in tetrahydro­ furan and the precipitation procedure is carried out again. Based on the original amount of starting polymer, the yield of the final product is at about 50%.

Prescription 3: Preparation of drug-carrying poly­ phosphazene polymers according to the invention. a) Substitution with an aromatic amino acid derivative.

This substitution is carried out according to the same procedure as described under prescription 2. The required amounts, however, are adjusted as follows: In case of an aromatic amino acid derivative available as a free base use is made of 25 mol% thereof, based on the starting polymer chloride content, and of 50 mol% triethylamine (also based on the starting polymer chloride content). In case of aromatic amino acid derivatives used in the form of a hydrochloride salt 30 mol% thereof is taken, while 45 mol% triethylamine is necessary for liberating the free base and another 50 mol% triethyl­ amine is required for the substitution reaction.

The substitution reaction is regulated in such a manner that the degree of substitution is at 20-25% of the original starting polymer chloride content.

b) Substitution with a compound supplying a drug-carrying imidazolyl group of formula 10 (hereinafter called prodrug).

As starting compound supplying a drug-carrying imidazolyl group of formula 10 (prodrug) use is made of amides, esters or anhydrides of drugs with respectively a single amino group, a single alcohol group or a single carboxylic acid group and imidazole derivatives such as imidazole-4-acetic acid, urocanic acid, 4-hydroxymethyl imidazole, β-carnosine methyl ester, β-carnosine dimethyl glycolamide ester, imidazole-4-acetic acid glycolic acid ester, urocanic acid glycolic acid ester or 4-hydroxy­ methyl imidazole glycolic acid ester. Based on the amount of starting polymer chloride, 50 mol% prodrug is taken and dissolved in 100 ml dry tetrahydrofuran. A 500 ml three-necked flask provided with a stirrer, a nitrogen gas inlet and a water cooled condenser is charged with 300 ml dry tetrahydrofuran. On a molar basis, 45% NaH is added in the form of 60% NaH in a mineral oil. The flask is cooled by means of an ice bath. The prodrug solution is added dropwise to the suspension. When the addition is complete, the ice bath is removed and the solution is allowed to warm to room temperature. The solution is heated to 40°C by means of a heating mantle. The heating is continued for 12 hours. The resulting solution with the sodium salt present therein is allowed to cool to room temperature and added slowly and dropwise to the polymer solution. A heavy white precipitate of sodium chloride is slowly formed. The substitution reaction is continued for 18 hours. The sodium chloride precipitate is removed by filtering, dried and weighed. It is ensured that the resulting degree of substitution is at 25-30% of the original starting polymer chloride content.

c) Substitution with imidazole.

The same procedure is followed as described under prescription 2. However, the amounts are adapted, namely 40 mol% imidazole and 50 mol% triethylamine, both based on the chloride content of the starting polymer. This substitution reaction, too, is continued for 18 hours.

Based on the precipitated amount of triethylamine hydrochloride the degree of substitution is adjusted to 20-30%.

d) Substitution with a non-aromatic amino acid derivative.

Here, too, the same procedure is performed as described under prescription 2. However, the required amounts are adapted, namely 10 mol% non-aromatic amino acid derivative and 20 mol% triethylamine, based on the chloride content of the starting polymer, in case of a non-aromatic amino acid derivative used as a free base, while in case of a starting material in the form of the hydrochloride salt 12 mol% is used, in combination with 18 mol% triethylamine for releasing the free base and 20% triethylamine for the substitution reaction.

The substitution reaction is regulated in such a manner that the degree of substitution is at 15-35%.

The resulting solution in tetrahydrofuran is partially evaporated, until a concentrated solution is obtained. The polymer solution is precipitated dropwise with stirring in n-hexane. The solid polymer is dissolved in tetrahydrofuran and the precipitation procedure is repeated. Based on the original amount of starting polymer, the yield of the final product is about 40%.

Prescription 4: Preparation of dimethylglycolamide esters of phenylalanine, phenylglycine, homophenylalanine, glycine, β-alanine and β-carnosine.

For the preparation of the dimethylglycolamide esters of amino acids the starting material is N-tert.butyl­ oxycarbonyl(tBOC) derivatives, which are commercially available. The dimethylglycolamide esters are prepared by reacting these derivatives with N,N-dimethyl-2-chloro­ acetamide in N,N-dimethylformamide (DMF).

To a solution of 0.1 mol of the N-tBOC derivative in 100 ml DMF are added 0.11 mol triethylamine, 0,01 mol sodium iodide and 0.11 mol N,N-dimethyl-2-chloroacetamide. The mixture is stirred at room temperature overnight, poured into 500 ml water and then extracted with twice 500 ml ethyl acetate. The combined extracts are washed with a 2% aqueous solution of sodium thiosulphate, 2% sodium bicarbonate and water. After drying over anhydrous sodium sulphate, the ethyl acetate is removed under reduced pressure to give the N-t-BOC amino acid dimethylglycolamide esters.

The t-BOC group is subsequently removed with 25% trifluoroacetic acid (TFA) in dry methylene chloride. This reaction is carried out for 60 minutes at room temperature.

After the solvent and excess TFA have been removed under reduced pressure, the resulting TFA salts of amino acid dimethylglycolamide esters are converted to the hydrochloride salts by means of anionic exchange column chromatography. If required, the hydrochloride salts are further purified by recrystallization from ethanol-water, isopropanol-water or pure isopropanol.

Prescription 5: Preparation of glycolic acid esters of imidazole-4-acetic acid and urocanic acid.

The glycolic acid esters are prepared by reacting the carboxylic acids with 2-chloro-acetic acid anhydride in dimethylformamide.

To a solution of 0.1 mol imidazole-4-acetic acid (or urocanic acid) in 100 ml DMF are added 0.11 mol triethylamine, 0.01 mol sodium iodide and 0.055 mol 2-chloro-acetic acid anhydride. The mixture is stirred at room temperature overnight, poured into 500 ml water and extracted with twice 500 ml ethyl acetate. The combined extracts are washed with a 2% aqueous solution of sodium thiosulphate, 2% sodium bicarbonate and water. After drying over anhydrous sodium sulphate, the ethyl acetate is removed under reduced pressure to give the glycolic acid esters of imidazole-4-acetic acid (or urocanic acid).

Prescription 6: Preparation of the glycolic acid ester of 4-hydroxymethyl imidazole.

The glycolic acid ester of 4-hydroxymethyl imidazole is prepared by firts obtaining 4-bromomethyl imidazole and subsequently reacting this substance with glycolic acid.

0.2 Mol 4-hydroxymethyl imidazole hydrochloride is added to 270 ml thionyl bromide, which has been cooled to 0°C. The yellow-coloured solution is allowed to return to room temperature. Excess thionyl bromide is distillated for 45 minutes under reduced pressure. The remainder solidifies upon cooling. The yield of 4-bromomethyl imidazole hydrobromide salt is about 50%. The melting point is 167.5°C (from acetonitrile).

The hydrobromide salt is heated for 2 hours with 0.11 mol triethylamine in DMF. The solution is allowed to cool to room temperature and the precipitate is filtered off. To the resulting liquid are added 0.1 mol glycolic acid, 0.11 mol triethylamine, and 0.01 mol sodium iodide. The mixture is stirred at room temperature overnight, poured into 500 ml water and extracted with twice 500 ml ethyl acetate. The combined extracts are washed with a 2% aqueous solution of sodium thiosulphate, 2% sodium bicarbonate and water. After drying over anhydrous sodium sulphate, the ethyl acetate is removed under reduced pressure to give the glycolic acid ester of 4-hydroxy imidazole.

Prescription 7: Preparation of prodrug esters from carboxylic acid group-containing drugs and alcoholic imidazole derivatives.

0.1 Mol of a carboxylic acid group-containing drug as well as 0.1 mol 4-hydroxymethyl imidazole or 4-hydroxymethyl imidazole glycolic acid ester and 0.8 mol 4-dimethylaminopyrridine (DMAP) are dissolved or suspended in 500 ml dry methylene chloride at 0°C. 0.11 Mol dicyclohexyl carbodiimide (DCC) in 400 ml methylene chloride is dropwise added, maintaining the temperature at 0°C. After completion, the solution is stirred for 12-16 hours at room temperature. The resulting precipitate, dicyclohexylurea, is filtered off. The organic solvent is extracted with 1 N HCl and water, until a neutral pH value is obtained. After drying over anhydrous sodium sulphate, the methylene chloride is concentrated to 1/3 of the original amount and the remaining precipitated dicyclohexylurea is filtered off. The methylene chloride is removed under reduced pressure to give the required prodrug ester.

Prescription 8: Preparation of prodrug esters from alcoholic group-containing drugs and of imidazole-derived carboxylic acids.

By imidazole-derived carboxylic acids are meant, e.g., imidazole-4-acetic acid, urocanic acid, imidazole-­ 4-acetic acid glycolic acid ester and urocanic acid glycolic acid ester.

0.3 Mol imidazole-derived carboxylic acid and 0.3 mol DCC are dissolved in 400 ml dry pyridine and allowed to stand at room temperature for 30 minutes. The dicyclohexylurea is removed by filtration and then the filtrate is added dropwise to 400 ml of a pyridine solution containing 0.1 mol of an alcohol group-containing drug and 0.005 mol DMAP. The mixture is allowed to stand for 60 minutes at room temperature. The solvent is then removed under reduced pressure at 50°C and the residue is taken up in ethyl acetate. The resulting solution is successively washed with 1000 ml of 0.1 M formate buffer at pH 4.0, 1000 ml of 0.1 M phosphate buffer at pH 7.6, and water. The organic phase is dried over anhydrous sodium sulphate and the solvent is removed under reduced pressure at 50°C to give the required prodrug ester.

Prescription 9: Preparation of prodrug amides from amino group-containing drugs and of imidazole-derived carboxylic acids.

To a solution of 0.1 mol imidazole-derived carboxylic acid in 100 ml dry tetrahydrofuran is added 0.11 mol carboxydiimidazole (CDI), resulting in a CO₂ development. The solution is stirred for 15 minutes at room temperature, then a solution of 0.1 mol of an amino group-containing drug in 250 ml dry tetrahydrofuran is added dropwise. The solution is stirred overnight at room temperature. The solvent is then removed under reduced pressure at 50°C and the residue is taken up in ether. The latter is successively washed with 200 ml diluted ammonia (2x) and with 300 ml water (3x) until neutral reaction of the solution. The organic phase is dried over anhydrous sodium sulphate and the solvent is removed under reduced pressure at 50°C to give the required prodrug amide.

Claims (14)

1. A polyphosphazene polymer consisting essentially of recurring units of formula 1, wherein a part of side chains R consists of the imidazolyl group of formula 2, another part of side chains R consists of an aromatically substituted group of formula 3, wherein R¹ stands for H, F, C₁₋₄ alkyl or C₁₋₄ alkoxy, and X stands for a group according to any of formulae 5-7, wherein a is an integer of 0-2 and R³ stands for -OC_bH_2b+1, -OCH₂CON(C_bH_2b+1)_cH_d or -N(C_bH_2b+1)_cH_d, wherein b is an integer of 1-4, c and d are each an integer of 0-2 and c+d=2, and, if required, yet another part of side chains R consists of a non-aromatically substituted group of formula 4, wherein R² stands for a group according to any of formulae 8-9, wherein c is an integer of 1-2 and R⁴ has the same meanings as R³.

2. A polyphosphazene polymer according to claim 1, characterized in that R³ and R⁴ are each independently -OC₂H₅, -OCH₂CON(CH₃)₂ or -NH₂, and R¹ stands for H or 4-F.

3. A polyphosphazene polymer according to claim 1 or 2, characterized in that 40-60% of side chains R consists of an aromatically substituted group of formula 3, 0-20% of side chains R consists of a non-­ aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

4. A polyphosphazene polymer according to claim 1 or 2, characterized in that 45-55% of side chains R consists of an aromatically substituted group of formula 3, 5-15% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

5. A biodegradable implant for gradually releasing a drug present therein, comprising a drug-containing matrix of a polyphosphazene polymer according to any of claims 1-4.

6. A polyphosphazene polymer according to claim 1 or 2, characterized in that yet another part of side chains R consists of a drug-carrying imidazolyl group of formula 10, wherein R⁵ is a drug bound to the imidazolyl ring via a spacer.

7. A polyphosphazene polymer according to claim 6, characterized in that 15-30% of side chains R consists of an aromatically substituted group of formula 3, 20-35% of side chains R consists of a drug-carrying imidazolyl group of formula 10, wherein groups of formulae 3 and 10 together form 40-60% of side chains R, 10-40% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

8. A polyphosphazene polymer according to claim 6, characterized in that 20-25% of side chains R consists of an aromatically substituted group of formula 3, 25-30% of side chains R consists of a drug-carrying imidazolyl group of formula 10, 15-35% of side chains R consists of a non-aromatically substituted group of formula 4, and the rest of side chains R consists essentially of the imidazolyl group of formula 2.

9. A polyphosphazene polymer according to claim 7 or 8, characterized in that groups of formulae 2 and 10 together form 45-65% of side chains R.

10. A biodegradable implant for gradually releasing a drug present therein, comprising a drug-carrying polyphosphazene polymer according to any of claims 6-9.

11. A process for preparing a polyphosphazene polymer according to claim 1, which comprises replacing in a poly(dichlorophosphazene) polymer first a part of the chlorine atoms by an aromatically substituted group of formula 3, then another part of the chlorine atoms by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms, if any, by a non-­ aromatically substituted group of formula 4.

12. A process according to claim 11, characterized by replacing in a poly(dichlorophosphazene) polymer first 45-55% of the chlorine atoms by an aromatically substituted group of formula 3, then 80-99% of the remaining chlorine atoms by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms by a non-aromatically substituted group of formula 4.

13. A process for preparing a polyphosphazene polymer according to claim 6, which comprises replacing in a poly(dichlorophosphazene) polymer first a part of the chlorine atoms by an aromatically substituted group of formula 3, then another part of the chlorine atoms by a drug-carrying imidazolyl group of formula 10, then yet another part of the chlorine atoms by the imidazolyl group of formula 2, and finally still remaining chlorine atoms, if any, by a non-aromatically substituted group of formula 4.

14. A process according to claim 13, characterized by replacing in a poly(dichlorophosphazene) polymer 45-55% of the chlorine atoms by groups of formulae 3 and 10 together, then 80-99% of the remaining chlorine atoms by the imidazolyl group of formula 2, and finally the still remaining chlorine atoms by a non-aromatically substituted group of formula 4.