WO2005116050A1 - Steroid-hormone conjugates with polyamines and their therapeutic use as anti-cancer and angiostatic agents - Google Patents

Abstract

Compounds of formula (I): {in which: one of R1 and R2 represents a group of formula (II): -NH-(CH2)N-NH-[(CH2)m-NH]x-[(CH2)p-NH]y (II) [where n, m and p each represents a number from 2 to 6; and x and y each represents zero or a number from 1 to 2]; and the other of R1 and R2 represents hydrogen, oxo, hydroxyl, a Cl - C6 alkanoyl group or said group of formula (II); R3 and R4 each represents oxo, hydroxy, mercapto, hydrogen, halogen, alkoxy or aryloxy; R5 represents hydrogen or methyl or is not present; the dotted line indicates a single or douse bond; and the dotted circle indicates none, one, two or three double bonds between pairs of adjacent carbon atoms}; and esters thereof are useful for the treatment or prevention of tumours and/or diabetic retinopathy.

Description

STEROID-HORMONE CONJUGATES WITH POLYAMINES AND THEIR THERAPEUTIC USE AS ANTI-CANCER AND ANGIOSTATIC AGENTS

The present invention relates to a series of novel compounds having a polyamine moiety and an androgen or oestrogen steroid moiety and to their use in the prophylaxis and treatment of tumours, especially those associated with the sex hormones, such as breast and prostate cancers.

Correlations between breast cancer and early menarche, late menopause, high endogenous oestrogen (E) levels, and long term administration of therapeutic oestrogens have all been established and indicate that breast cancer risk increases with increasing oestrogen exposure. Oestrogen exerts its effects by binding to intracellular oestrogen receptor subtypes, ERα and ERβ. Oestrogen-occupied receptors (ER) undergo structural and conformational alterations to recognize specific DNA sequences, oestrogen response elements (ERE), in responsive genes and thus trigger a series of events, culminating in the transcription of these genes. ER-activated gene transcription ultimately controls cell proliferation and differentiation.

ERα and ERβ display a range of transcriptional activities, depending on whether the receptor is occupied by the hormone or by a synthetic analogue, such as selective oestrogen receptor modulators (SERMs). SERMs, such as tamoxifen and raloxifen, bind to the oestrogen receptor (ER) and mimic oestrogen action in certain tissues while opposing it in others. Tamoxifen behaves as an anti-oestrogen in the breast and is thus an effective treatment for hormone-responsive breast cancer. In the uterus, however, tamoxifen is oestrogenic. Tamoxifen is efficacious against about 50% of breast cancers, which express ER. However, about 50% of breast cancers at diagnosis do not express appreciable levels of ER, and these ER-negative tumours have a poorer prognosis than ER-positive tumours, with shorter times to relapse and shorter survival times.

Progesterone is a risk factor for breast cancer as well. In most hormone replacement therapies (HRTs), a progestin is given along with the oestrogen to counteract oestrogenic side effects in the uterus. Such combined HRT results in a significantly larger increase in breast cancer risk than does administration of a pure oestrogen.

Testosterone is a known mitogen in the prostate, and prostate cancer is the most common cancer in ageing men. Like breast cancer, a large percentage of prostate cancers are at diagnosis hormone refractory, expressing very little androgen receptor (AR). Anti-androgens, such as cyproterone and bicalutamide, can be used in hormone positive prostate cancers, and can appreciably slow the progression of the disease in such cases. However, once again, by analogy with breast cancers, prostate cancers tend to become androgen and hormone refractory upon prolonged treatment. The presence of erbB2 in many prostate tumours raises the possibility that the same mechanisms are involved in many hormone resistant cases in both tumour types.

The polyamines, putrescine

spermidine [H2N(CH2)3NH(CH2)4NH₂], and spermine

are organic cations with multiple functions in cell proliferation and differentiation. Polyamine synthesis in mammalian cells begins with the production of putrescine by the decarboxylation of the amino acid, ornithine, by ornithine decarboxylase (ODC). Subsequent addition of an aminopropyl group to putrescine leads to spermidine, and further addition of an aminopropyl group leads to the formation of spermine. The aminopropyl group is derived by the decarboxylation of S-adenosylmethionine through the action of S-adenosylmethionine decarboxylase (SAMDC). ODC and SAMDC are rate-limiting enzymes in polyamine biosynthesis. Addition of aminopropyl groups to putrescine and spermidine is catalysed by the enzyme spermidine synthase and spermine synthase respectively.

Polyamines are positively charged at the primary and secondary amino groups at physiological pH. Therefore, electrostatic interactions through the cationic amino groups, and hydrophobic interactions through the methylene bridging groups are dominant. Thus, polyamines may act as ligands at multiple sites on DNA, RNA, proteins, phospholipids and nucleotide triphosphates.

Studies have shown that a large part of the biological function of polyamines is in the regulation of gene expression both by altering DNA structure and by modulating signal transduction pathways. Thus, DNA conformational transitions induced by polyamines can serve as a basis for regulation of gene expression by unravelling DNA- protein interactions and chromatin structure.

Another aspect of polyamine-DNA interactions involves modulation of DNA- protein interactions. Polyamines enhance the binding of several gene-regulatory proteins to the specific regulatory sequences, called response elements. For example, spermine facilitates the binding of the oestrogen receptor to its response element and nuclear factor KB (NF-KB) to its response element at 100-500 μM concentrations, indicating a possible mechanism for the action of polyamines in breast cancer cell proliferation.

Ligand-receptor interactions offer another target for polyamines. Indeed, the best-characterised system is the effect of polyamines on the function of N-methyl-D- aspartate (NMD A) receptors. Polyamines potentiate of inhibit glutamate-mediated responses of this channel complex in a concentration-dependent manner. Polyamines also modulate other ligand-receptor interactions, including oestradiol binding to oestrogen receptors (ER). Recent studies indicate that protein- protein interactions are also altered by polyamines. Inhibition of polyamine biosynthesis was shown to drastically reduce the number of proteins associated with ER, and addition of spermidine enhanced ER-association of numerous proteins, again suggesting a mechanism for the action of polyamines in breast cancer cell proliferation.

In addition to the direct effects of polyamines on gene regulation through DNA- protein and protein-protein interactions, there is evidence for polyamine involvement in different stages of signal transduction. For example, the activity of purified casein kinase LT increased by 2-to 20-fold in the presence of polyamines. Another example of polyamines acting as signalling molecules is observed in using rat smooth muscle cells, where the cytostatic effect of difluoromemylornithine (DFMO), an inhibitor of ODC, could be prevented by MAP kinase 1/2 inhibitor. In these cells, addition of DFMO (which blunts polyamine synthesis) leads to activation of p42/p44 MAPK and induction of p21WaflCTPl. This protein is an inhibitor of cell cycle progression as it inhibits the action of cyclin/cdk complexes required for cell cycle progression. Ornithine decarboxylase (ODC) has long been known as a marker of carcino genesis and tumour progression. Studies demonstrating that over-expression of the ODC gene in NTH 3T3 cells leads to transformation of these cells support a role for ODC and polyamines in the origin and progression of neoplastic diseases. Blockade of polyamine synthesis can be achieved by the inhibition of ODC. Cancer cells undergo cytostasis in the presence of DMFO, an irreversible inhibitor of ODC, and this growth arrest can be prevented by the treatment of cells with putrescine.

Polyamine analogues are unable to carry out the functions of natural polyamines in the cell, therefore causing collapse of cellular functions. Since cancer cells have higher levels of natural polyamines and increased requirement and/or defective regulatory mechanisms, they can be expected to be more sensitive to the action of polyamine analogues than normal cells.

Traditional polyamine analogues have been designed by altering the chain lengths of methylene groups bridging primary and secondary amino groups, introducing substituents at various positions, and by alkylating the amino end groups [Casero RA and Pegg AE (1993) FASEB J. 7: 653-61 & Porter CW and Bergeron RJ (1988) Adv. Exp. Med. Biol. 250: 677-90]. Porter et al. [(1987) Cancer Res. 47: 2821-25], prepared bis(ethyl) derivatives of putrescine, spermidine and spermine. Among these compounds, bis(ethyl)spermine analogues were more effective anti-proliferative agents than putrescine or spermine derivatives. Subsequent developments included asymmetrically alkylated polyamine analogues, dimethylsilane compounds or conformationally constrained polyamine analogues.

Direct action of polyamine analogues with DNA remains to be a major pathway in the mechanism of action of polyamine analogues. Polyamine analogues might interfere with DNA-polyamine interactions and cause problems in condensation, packaging of DNA and its unravelling during transcription and replication. As a result, analogues may alter cell cycle regulatory mechanisms, including the induction of cyclin Bl, cyclin Dl, p53, and cyclin dependent kinase inhibitors. These changes in gene expression may occur due to interactions of the analogues with DNA, interference with transcription factor-DNA interactions or by altering the stability of mRNAs. In the case of certain asymmetrically alkylated polyamine analogues, anti-mitotic effects of these compounds were attributed to their ability to interfere with tubulin polymerisation.

Polyamines are involved in cellular proliferation and differentiation, and increased ODC activity and accumulation of intracellular polyamines has been shown to play an important role in the development and growth of many cancers including breast cancer. Polyamines appear to support breast cancer cell proliferation as well as tumour progression to a hormone-independent, aggressive phenotype. The role of polyamines in breast cancer is also supported by findings showing that increased ODC activity in primary breast cancer specimens was an independent negative prognostic factor and was superior to lymph node status in predicting disease-free and overall survival.

Irreversible inhibition of polyamine synthesis has therefore been an important target of intervention therapy, even in the case of cancer incidence due to alterations in cancer susceptibility genes. Indeed, irreversible inhibition of ODC by DFMO has been shown to thwart proliferation of both hormone-dependent and -independent breast cancers in vitro and in vivo, and tamoxifen and DMFO have been shown to have additive effects on inhibition of MCF-7 growth. However, clinical trials using DMFO as a single drug have given disappointing results.

One reason for inefficacy of selective enzyme inhibitors such as DMFO is probably that selective blockade of a single enzyme triggers compensatory changes in PA (polyamine) metabolism and transport. Moreover, the ubiquitous occurrence of PAs, their importance in normal functions of nearly all mammalian cells, and the capacity of the organism to utilize exogenous gastrointestinal PAs seriously limit the efficacy of PA synthesis inhibitors in cancer chemotherapy. Effective inhibition of tumour growth might therefore require that several reactions or regulatory functions of PAs are impaired at the same time. Recent efforts have therefore been focusing on the design and study of structural analogues of polyamines (PAA). One such compound, SL-11093, was recently shown to effectively inhibit growth of human prostate tumour xenograft in nude mice.

As indicated above, direct action of PAAs with DNA remains to be a major pathway in the mechanism of action of polyamine analogues. However, no single mechanism explains all of the effects of PAAs (see above), and because of the multiple functions of polyamines in the cell and in the organism, the non-toxic treatment window of PAAs is probably narrow. However, increased efficacy and reduced toxicity might be expected if PAAs were designed to target specific cancer cell types. Breast and prostate cancer cells have higher levels of oestrogen and androgen receptors respectively, when compared to most other cell types. PAAs designed to bind with selective affinity to these receptors might therefore preferentially target breast cancer cells, and prostate cancer cells through binding to ER and AR respectively.

We have now surprisingly found that certain compounds having a polyamine portion and a steroid portion can inhibit the growth of cancer cells, especially those of the sex hormone-related cancers, and can inhibit angiogenesis. Although we do not wish to be limited by any theory, it is thought that steroid hormone-substituted PAs, designed to bind to the corresponding hormone receptors, could effectively block ER- and AR-activated gene transcription and ultimately oestrogen-and andro gen-mediated control of cell proliferation and differentiation. Moreover, direct interaction of these novel PAAs with DNA should cause collapse of cellular functions, and death of breast and prostate cancer cells that do not express appreciable levels of either ER or AR, i.e. tamoxifen-resistant cells and anti-androgen refractory cells.

Growth of new blood vessels (angiogenesis) is essential for tumour growth and tumour proliferation (metastasis) and occurs throughout a tumour's lifecycle. Indeed, angiogenesis is an important prognostic factor in breast cancer. Blockade of the expression of the vascular endothelial growth factor (VEGF) gene, and hence the secretion of vascular endothelial growth factor, the main tumour angiogenesis mediator, is therefore an obvious target for drug treatment. Over-expression of VEGF by MCF-7 breast cancer cells promotes oestrogen independent tumour growth. Inhibition of VEGF gene expression together with blockade of nuclear hormone receptor activation would therefore be expected to inhibit both angiogenesis and the proliferation of hormone-dependent and hormone- independent cancers. PAAs have been repeatedly shown to interact with DNA, but inhibition of the expression of the VEGF gene and hence reduction of VEGF secretion through interaction of PAAs with DNA has not been demonstrated. However, squalamine, a cholestane substituted spermidine, was shown to block angiogenesis and human ovarian cancer when given together with cisplatin. Although VEGF secretion was not determined, this "non-specific" steroid-substituted PA significantly blocked VEGF-induced activation of MAP kinase, another important member of tumour promoting growth factors, and inhibited the proliferation of human vascular endothelial cells. It is possible therefore, that specific steroid-substituted PAs, such as oestrogen-, progestin-, and androgen-substituted PAs could effectively inhibit angiogenesis, both through inhibition of VEGF gene expression and blockade of VEGF-activated MAP kinase, and the growth and proliferation of hormone-dependent and hormone- independent cancers.

Retinal neovascularisation and macular oedema are central features of diabetic retinopathy, the major cause of blindness in the developed world. Since VEGF plays a pivotal role in retinal microvascular complications in diabetes, inhibition of angiogenesis represents an exciting target for intervention in diabetic retinopathy. Vitreous spermidine and putrescine levels have been shown to be increased in proliferative diabetic retinopathy and to correlate with vitreous VEGF content. Erythrocyte levels of spermidine are also significantly increased in diabetic patients with retinopathy, and spermine and spermidine levels have been shown to be associated with albuminuria and macroangiopathy in non insulin-dependent diabetes (TNTTDDM). Altogether, these findings suggest that increased production of spermidine is associated with diabetic retinopathy. Since polyamines are involved in cellular proliferation and differentiation, increased polyamine synthesis may drive retinal neovascularisation in diabetes. Polyamine analogues that block cellular growth and angiogenesis could therefore constitute an effective treatment and prevention of diabetic retinopathy. Thus, the present invention provides compounds of formula (I) :

in which:

one of R! and R^ represents a group of formula (II):

-NH-(CH2)_n-NH-[(CH₂)_m-NH]_χ-[(CH₂)p-NH]_y-H (IT) where n, m and p are the same as or different from each other and each represents a number from 2 to 6; and x and y are the same as or different from each other and each represents zero or a number from 1 to 2;

and the other of R and R^ represents a hydrogen atom, an oxo group, a hydroxyl group, a Ci - Cg alkanoyl group or said group of formula (II);

R^ and R^ are the same as or different from each other and each represents an oxo group, a hydroxy group, a mercapto group, a hydrogen atom, a halogen atom, an alkoxy group or an aryloxy group;

R-> represents a hydrogen atom or a methyl group or is not present;

the dotted line indicates that there may be a single or double bond between the respective carbon atoms; and

the dotted circle indicates that there may be none, one, two or three double bonds between pairs of adjacent carbon atoms; and esters thereof.

The invention also provides the use of compounds of formula (I) and esters thereof for the manufacture of a medicament for the treatment or prevention of tumours and/or diabetic retinopathy. The invention also provides a method for the treatment or prevention of tumours and/or diabetic retinopathy in a mammal by the administration to a mammal suffering from or susceptible to tumours and or diabetic retinopathy of an effective amount of at least one compound of formula (I) or an ester thereof.

The invention still further provides the use of compounds of formula (I) and esters thereof for the manufacture of a medicament for the inhibition of angiogenesis.

In the compounds of the present invention, we prefer that one of R* and R² should represent said compound of formula (II) and the other should represent a hydrogen atom, an oxo group, a hydroxyl group, or a Ci - Cg alkanoyl group.

Where one of R and R² represents a Ci - Cg alkanoyl group, this may be a straight or branched chain group having from 1 to 6, preferably from 2 to 6 and more preferably from 2 to 4, carbon atoms. Examples of such groups include the formyl, acetyl, propionyl, butyryl, isobutyryl, t-butyryl, valeryl and hexanoyl groups, of which the acetyl group is most preferred.

Where one or both of R* and ² represents a group of formula (LT): -NH-(CH2)_n-NH-[(CH₂)_m-NH]_x-[(CH₂)_p-NH]_y-H (II)

we prefer that n, m and p, which may be the same as or different from each other, should be 3 or 4. We also prefer that x and y, which may be the same as or different from each other, should be 0 or 1. The most preferred groups of formula (II) are those derived from putrescine, i.e. H2N(CH₂)4NH- (x=0, y=0, n=4), spermidine, i.e. H2N(CH2)3NH(CH₂)4NH- or -HN(CH₂)₃NH(CH₂)₄NH₂ (x=l , y=0, n=3, m=4) or spermine, i.e. H₂ (CH₂)₃NH(CH2)4NH(CH₂)₃NH- (x=l, y=l, n=3, m=4, p=3). Of these, we especially prefer the groups derived from spermidine, especially the group of formula -HN(CH₂)₃NH(CH₂)₄NH₂.

Where R^ or R^ represents a halogen atom, this may be any of the halogens, for example, fluorine, chlorine, bromine or iodine. Where R^ or R^ represents an alkoxy group, this may be a straight or branched chain alkoxy group, preferably containing from 1 to 6 carbon atoms, for example the methoxy, ethoxy, propoxy, isopropoxy, butoxy, isobutoxy, sec-butoxy, t-butoxy, pentyloxy or hexyloxy group.

Where R-* or R^ represents an aryloxy group, this preferably has from 6 to 10 carbon atoms in one or more aromatic carbocyclic rings. Examples of such groups include the phenoxy, 1-naphthyloxy and 2-naphthyloxy groups, which may be substituted or unsubstittited. If substituted, there are preferably from 1 to 3 substituents. Such substituents may be selected from any groups or atoms commonly used in pharmaceutical chemistry, provided that they do not adversely affect the activity or do not affect it adversely to a substantial extent. Examples include alkyl groups (preferably Ci - Cg, such as methyl, ethyl, propyl, isopropyl and t-butyl groups), alkoxy groups

(preferably Ci - Cg, such as those exemplified above in relation to R^ and R^), aryl groups (such as the phenyl group), carboxy groups and carbamoyl groups.

Particularly preferred compounds of the present invention are those oestrogen derivatives of formula (la):

(in which R^, R^, R^ and R^ are as defined above); or those testosterone derivatives of formula (lb):

(in which R¹, R², R^ and R^ are as defined above, and R^^a represents a hydrogen atom or a methyl group, preferably a methyl group); or those dihydrotestosterone derivatives of formula (Ic):

(in which R R , R^ and R^ are as defined above, and R^^a represents a hydrogen atom or a methyl group, preferably a methyl group); or those progesterone derivatives of formula (Id):

(in which R¹, R³ and R⁴ are as defined above, and R^5a represents a hydrogen atom or a methyl group, preferably a methyl group).

We especially prefer compounds of formula (la) in which R¹ represents a hydroxyl group, R² represents a group of formula (II), and R³ and R⁴ both represent hydrogen atoms; those compounds of formula (lb) in which R¹ represents a group of formula (II), R² represents a hydroxy group, R³ and R⁴ both represent hydrogen atoms, and R^5a represents a methyl group; those compounds of formula (Ic) in which R¹ represents a hydroxy group, R² represents a group of formula (II), R³ and R⁴ both represent hydrogen atoms, and R^^a represents a methyl group; and those compounds of formula (Id) in which R* represents a group of formula (II), R³ and ⁴ both represent hydrogen atoms, and R^5a represents a methyl group.

We prefer that the steric configuration of the compounds of the present invention should be the same as that of the natural steroids to which they are related, those configurations being well known in the art.

The compounds of the present invention may be prepared by a variety of methods, each well known in themselves. For example, they may be prepared as illustrated by the following reaction scheme A for those compounds where ² represents a group derived from spermidine (compounds derived from other polyamines may be prepared in an analogous manner): Reaction Scheme A:

(VLT) In the above formulae, R^, R³, R⁴ and R⁵ and the dotted line and circle are as defined above, and Boc represents a protecting group, e.g. a t-butoxycarbonyl group.

Ste Al:

In this step, a steroid derivative of formula (IS) is reacted with a protected spermidine compound of formula (TV) to give an imine of formula (V).

The reaction is normally and preferably effected in the presence of a solvent. There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: ethers, such as diethyl ether, tetrahydrofuran, diisopropyl ether, or dioxane; aromatic hydrocarbons such as benzene, toluene or xylene; halogenated hydrocarbons such as dichloromethane, chloroform or carbon tetrachloride; and mixtures of any two or more of these solvents.

The reaction can take place over a wide range of temperatures, and the precise reaction temperature is not critical to the invention. In general, we find it convenient to carry out the reaction at a temperature of from 5 to 60°C, preferably at about room temperature. The time required for the reaction may also vary widely, depending on many factors, notably the reaction temperature and the nature of the reagents and solvent employed. However, provided that the reaction is effected under the preferred conditions outlined above, a period of from 10 hours to several days will usually suffice.

The reaction is preferably effected in an oxygen-free environment, preferably in the presence of an inert gas, such as nitrogen.

Following the reaction, the desired compound may be separated from the reaction mixture, and, if desired, purified, by conventional means. For example, the solvent may be removed, e.g. by evaporation, and the residue washed, and extracted, e.g. with chloroform, after which the extract may be washed again, dried and then the solvent removed, to leave the desired compound. Step A2:

In this step, the inline of formula (V), prepared as described in Step Al, is reduced to give the polyamino steroid compound of formula (VI).

The reaction may be carried out using a reducing agent. The nature of the reducing agent employed in this reaction is not critical, and any such agent commonly used in reactions of this type may equally be used here. Examples of such reducing agents include: metal hydrides, such as lithium borohydride, sodium borohydride, sodium cyanoborohydride, lithium aluminium hydride and diisopropylaluminium hydride. The reaction is normally and preferably effected in the presence of a solvent.

There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: hydrocarbons, such as benzene, toluene, xylene, hexane or heptane; ethers, such as diethyl ether, tetrahydrofuran or dioxane; amides, such as dimethylformarnide, dimethylacetamide or hexamethylphosphoric triamide; alcohols, such as methanol, ethanol or isopropanol; and mixtures of any two or more of these solvents.

The reaction can take place over a wide range of temperatures, and the precise reaction temperature is not critical to the invention. In general, we find it convenient to carry out the reaction at a temperature of from that of ice-cooling to heating, e.g. to the reflux temperature of the reaction medium, preferably at about room temperature. The time required for the reaction may also vary widely, depending on many factors, notably the reaction temperature and the nature of the reagents and solvent employed. However, provided that the reaction is effected under the preferred conditions outlined above, a period of from 0.5 hour to several days will usually suffice.

The reaction is preferably effected in an oxygen-free environment, preferably in the presence of an inert gas, such as nitrogen.

The desired compound may then be separated from the reaction mixture and, if desired, purified by conventional means. For example, one suitable procedure comprises: terminating the reaction by the addition of water; removing the solvent, e.g. by evaporation; extracting the desired compound, e.g. with chloroform; washing and drying the extract; and removing the solvent, e.g. by evaporation. The compound may then be purified, e.g. by column chromatography.

Step A3:

In this step, the compound of formula (VI), prepared in Step A2, is deprotected to give the final product of formula (VLT).

The nature of the reaction employed will depend on the nature of the protecting group, but these conditions are well known to those skilled in the art. These conditions are described, for example, in T. W. Green & P. G. M. Wuts: Protective Groups in Organic Synthesis, Second Edition, 1991 (John Wiley & Sons Ed.).

Reaction Scheme B:

Those compounds of the present invention where R^ represents a group derived from spermidine may, for example, be prepared as illustrated by the following reaction scheme B (compounds derived from other polyamines may be prepared in an analogous manner):

Reaction Scheme B:

(VϋJ) (IX)

(XIV)

In the above formulae, R³, R⁴, R^, Boc and the dotted line and circle are as defined above, R^2a represents any of the groups that may be represented by R , provided that any reactive group is protected, and Tf represents a protecting group, for example a trifluoroacetate group.

Ste Bl:

In this step, the hydroxy group in the compound of formula (VLTI) is protected, for example using a trifluoroacetate group. The reaction involved will depend on the nature of the protecting group, as is well known to those skilled in the art. Details may be found in T. W. Green & P. G. M. Wuts: Protective Groups in Organic Synthesis, Second Edition, 1991 (John Wiley & Sons Ed.).

Step B2: In this step, the protected compound of formula (IX), prepared as described in

Step Bl, is converted to an imino compound of formula (X).

The reaction is carried out in the presence of a catalyst, and any catalyst commonly used in this type of reaction may equally be employed here. Examples of suitable catalysts include tetrakis(triphenylphosphine)palladium(0), palladium(II) chloride, bis(triphenylphosphine)palladium(Ii) chloride, palladium(II) acetate and tris(dibenzylideneacetone)dipalladium(0). Of these, we prefer palladium(II) acetate or tetrakis(triphenylphosphine)palladium(0), most preferably palladium(H) acetate.

We prefer to employ a ligand for the catalyst, for example BINAP [2,2'- bis(diphenylphosphino)- 1 , 1 '-binaphthyl] . The reaction is preferably effected in an oxygen-free environment, preferably in the presence of an inert gas, such as nitrogen.

The reaction is normally and preferably effected in the presence of a solvent. There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: ethers, such as diethyl ether, tetrahydrofuran, diisopropyl ether, or dioxane; aromatic hydrocarbons such as benzene, toluene or xylene; halogenated hydrocarbons such as dichloromethane, chloroform or carbon tetrachloride; and mixtures of any two or more of these solvents.

The reaction can take place over a wide range of temperatures, and the precise reaction temperature is not critical to the invention. In general, we find it convenient to carry out the reaction at a temperature of from about room temperature to heating, e.g. to the reflux temperature of the reaction medium, preferably at the reflux temperature of the reaction medium. The time required for the reaction may also vary widely, depending on many factors, notably the reaction temperature and the nature of the reagents and solvent employed. However, provided that the reaction is effected under the preferred conditions outlined above, a period of from 2 hours to several days will usually suffice.

The product of this step is normally not separated. Instead, it is converted, without intermediate isolation, into the aminoesterone (XI), in Step B3.

Step B3: In this step, the imino compound of formula (X) prepared as described in Step

B2 is converted to an aminoesterone of formula (XI).

This reaction may be effected simply by acidification of the reaction mixture prepared in Step B2. There is no particular restriction on the acid used for the reaction. Examples of suitable acids include: inorganic acids, such as hydrochloric, sulphuric and nitric acids; and organic acids, such as trifluoromethylsulphonic acid, acetic acid and p- toluenesulphonic acid.

Although the reaction may be carried out over a wide range of temperatures, we prefer to carry it out at about room temperature.

Following the reaction, the desired compound may be separated from the reaction mixture and, if desired, further purified by conventional means. For example, one suitable recovery procedure comprises: neutralising the reaction mixture by adding a base, such as sodium or potassium carbonate; extracting the compound from the reaction mixture; removing the solvent from the extract, e.g. by evaporation; and then purifying the resulting compound by column chromatography. Step B4:

In this step, the aminoesterone (XI) prepared as described in Step B3 is reacted with a malonic acid amide of formula (XH), to give a ketone compound of formula (XTJDL). The reaction is normally and preferably effected in the presence of a solvent.

There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: hydrocarbons, such as benzene, toluene, xylene, hexane or heptane; ethers, such as diethyl ether, tetrahydrofuran or dioxane; amides, such as dimethylformamide, dimethylacetamide or hexamethylphosphoric triamide; alcohols, such as methanol, ethanol or isopropanol; and mixtures of any two or more of these solvents.

The reaction may be terminated by the addition of water, after which the desired compound may be recovered from the reaction mixture and, if necessary, purified by conventional means. For example, one suitable recovery technique comprises: extracting the desired compound from the reaction mixture, for example with chloroform; washing and drying the extract; and then removing the solvent, for example by evaporation.

Step B5: In this step, where R^2a in the compound of formula (XILT) prepared as described in Step B4 represents an oxo group, this oxo group is converted to a hydroxyl group, to give a compound of formula (XIV).

The reaction is effected by means of a reducing agent, and any such agent capable of reducing a ketone to an alcohol in a conventional reaction may equally be used in this reaction. Examples of such reducing agents include: metal hydrides, such as lithium borohydride, sodium borohydride, sodium cyanoborohydride, lithium aluminium hydride and diisopropylaluminium hydride. The reaction is normally and preferably effected in the presence of a solvent. There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: hydrocarbons, such as benzene, toluene, xylene, hexane or heptane; ethers, such as diethyl ether, tetrahydrofuran or dioxane; amides, such as dimethylformamide, dimethylacetamide or hexamethylphosphoric triamide; alcohols, such as methanol, ethanol or isopropanol; and mixtures of any two or more of these solvents.

The reaction can take place over a wide range of temperatures, and the precise reaction temperature is not critical to the invention. In general, we find it convenient to carry out the reaction at a temperature of from that of ice-cooling to heating, e.g. to the reflux temperature of the reaction medium, preferably at about room temperature. The time required for the reaction may also vary widely, depending on many factors, notably the reaction temperature and the nature of the reagents and solvent employed. However, provided that the reaction is effected under the preferred conditions outlined above, a period of from 0.5 hour to 24 hours will usually suffice.

At the end of this time, water may be added to terminate the reaction, after which the desired product may be recovered and, if desired, purified from the reaction mixture by conventional means. For example, one suitable recovery procedure comprises: extracting the desired compound, for example with chloroform; drying the extract; and purifying the residue, for example by column chromatography.

Step B6:

In this step, the protecting group, represented by Boc, is removed from the compound of formula (XIV) to give the amine compound of formula (XV). The nature of the reaction employed will depend on the nature of the protecting group, but these conditions are well known to those skilled in the art. These conditions are described, for example, in T. W. Green & P. G. M. Wuts: Protective Groups in Organic Synthesis, Second Edition, 1991 (John Wiley & Sons Ed.). Step B7:

In this step, the compound of formula (XV) is reduced to give the final product, the compound of formula (XVI).

The reaction is effected by means of a reducing agent, and any such agent capable of reducing a ketone to an alcohol in a conventional reaction may equally be used in this reaction. Examples of such reducing agents include: borane; metal hydrides, such as lithium borohydride, sodium borohydride, sodium cyanoborohydri.de, lithium aluminium hydride and diisopropylaluminium hydride.

The reaction is normally and preferably effected in the presence of a solvent. There is no particular restriction on the nature of the solvent to be employed, provided that it has no adverse effect on the reaction or on the reagents involved and that it can dissolve the reagents, at least to some extent. Examples of suitable solvents include: hydrocarbons, such as benzene, toluene, xylene, hexane or heptane; ethers, such as diethyl ether, tetrahydrofuran or dioxane; amides, such as dimethylformamide, dimethylacetamide or hexamethylphosphoric triamide; alcohols, such as methanol, ethanol or isopropanol; and mixtures of any two or more of these solvents.

The reaction can take place over a wide range of temperatures, and the precise reaction temperature is not critical to the invention. In general, we find it convenient to carry out the reaction at a temperature of from that of ice-cooling to heating, e.g. to the reflux temperature of the reaction medium, preferably at the reflux temperature. The time required for the reaction may also vary widely, depending on many factors, notably the reaction temperature and the nature of the reagents and solvent employed. However, provided that the reaction is effected under the preferred conditions outlined above, a period of from 0.5 hour to 24 hours will usually suffice. At the end of this time, methanol may be added to terminate the reaction, after which the desired product may be recovered and, if desired, purified from the reaction mixture by conventional means. For example, one suitable recovery procedure comprises: extracting the desired compound, for example with chloroform; drying the extract; and purifying the residue, for example by column chromatography. Reaction Scheme C:

The starting material of formula (XII) used in Step B4 may be prepared as illustrated in the following Reaction Scheme C:

H Step CI

H₂N^' H₂N^{' '}Boc (XVII) (XVIII)

In the above formulae, Et represents an ethyl group and Boc is as defined above.

Step Cl:

In this step, one of the amino groups in the 1,4-diaminobutane of formula (XVLT) is protected, e.g. by a t-butoxycarbonyl group. The nature of the reaction employed will depend on the nature of the protecting group, but the conditions employed are well known to those skilled in the art. These conditions are described, for example, in T. W. Green & P. G. M. Wuts: Protective Groups in Organic Synthesis, Second Edition, 1991 (John Wiley & Sons Ed.). Step C2:

In this step, the protected amino compound of formula (XVIJJ) prepared as in Step CI is coupled with monoethyl malonate, to give the malonate amide of formula (XIX). The reaction is preferably carried out in the presence of a coupling aid, for example l-(3-dime ylaminopropyl)-3-ethylcarbodiimide hydrochloride and under an inert atmosphere. It is normally and preferably carried out in a solvent, the nature of which is not critical, for example dimethylformamide and preferably at about ambient temperature.

Step C3:

In this step, the ester group of the malonate amide of formula (XIX), prepared as described in Step C2, is hydrolysed to give the acid of formula (XLT), which is used in Step B4 of Reaction Scheme B.

This is a conventional ester hydrolysis and may be carried out using well known techniques, for example by alkaline hydrolysis, for example in the presence of potassium carbonate in aqueous methanol at about room temperature.

Reaction Scheme D:

The starting material of formula (TV), used in Reaction Scheme A, may be prepared as shown in the following Reaction Scheme D:

H Step D2 ^Boc- CN Step D3 H (XX)

Bo

In the above formulae, Boc is as defined above.

Step Dl:

This step is identical to Step CI of Reaction Scheme C, and maybe carried out in the same way and using the same reagents and reaction conditions.

Step D2:

In this step, the protected dia ino compound of formula (XVLH), produced in Step Dl above is reacted with acrylonitrile, in a pseudo-Michael reaction, to give the nitrile of formula (XX). The reaction is preferably effected in a solvent, the nature of which is not particularly critical. An alcohol, such as methanol or ethanol, is preferred. The reaction temperature is likewise not particularly critical, but the reaction is preferably effected with ice-cooling. Step D3:

In this step, the compound of formula (XX) prepared as described in Step D2 above is first protected at its unprotected amino group and then the cyano group is reduced to an amino group. The protection step is preferably effected using a butoxycarbonyl protecting group and this reaction is well known in the art, as described, for example, in T. W. Green & P. G. M. Wuts: Protective Groups in Organic Synthesis, Second Edition, 1991 (John Wiley & Sons Ed.).

The cyano group is then reduced to an aminomethyl group. This may be achieved by hydrogenation in the presence of a catalyst. Conventional hydrogenation catalysts may be used, for example, Raney nickel, palladium-on-charcoal, palladium- black, ruthenium or platinum oxide, of which Raney nickel is preferred.

The reaction is preferably effected in a suitable solvent, for example an ethanolic solution of an alkali metal hydroxide. It is preferably effected under superatmospheric pressure and at slightly elevated temperatures, e.g. from 25 to 40°C.

The compounds of the present invention may be used for the prevention or treatment of cancers, especially those associated with the sex hormones, such as breast cancer or prostate cancer. They may also be used for the treatment or prevention of diabetic retinopathy. The compounds of the present invention can be administered in various forms, depending on the disorder to be treated and the age, condition and body weight of the patient, as is well known in the art. For example, where the compounds are to be administered orally, they may be formulated as tablets, capsules, granules, powders or syrups; or for parenteral administration, they may be formulated as injections (intravenous, intramuscular or subcutaneous), drop infusion preparations or suppositories. For application by the ophthalmic mucous membrane route, they may be formulated as eye drops or eye ointments. These formulations can be prepared by conventional means, and, if desired, the active ingredient may be mixed with any conventional additive, such as an excipient, a binder, a disintegrating agent, a lubricant, a corrigent, a solubilising agent, a suspension aid, an emulsifying agent or a coating agent.

Examples of vehicles which may be employed include: organic vehicles including; sugar derivatives, such as lactose, sucrose, glucose, mannitol and sorbitol; starch derivatives, such as corn starch, potato starch, a-starch, dextrin and carboxymethylstarch; cellulose derivatives, such as crystalline cellulose, low-substituted hydroxypropylcellulose, hydroxypropylmethylcellulose, carboxymethylcellulose, calcium carboxymethylcellulose and internally bridged sodium carboxymethylcellulose; gum arabic; dextran; Pullulane; and inorganic vehicles including silicate derivatives, such as light silicic anhydride, synthetic aluminium silicate and magnesium aluminate metasilicate; phosphates, such as calcium phosphate; carbonates, such as calcium carbonate; and sulphates, such as calcium sulphate.

Examples of lubricants which may be employed include: stearic acid; metal stearates, such as calcium stearate and magnesium stearate; talc; colloidal silica; waxes, such as bee gum and spermaceti wax; boric acid; adipic acid; sulphates, such as sodium sulphate; glycol; fumaric acid; sodium benzoate; DL-leucine; fatty acid sodium salts; lauryl sulphates, such as sodium lauryl sulphate and magnesium lauryl sulphate; silicates, such as silicic anhydride and silicic acid hydrate; and the aforementioned starch derivatives. Examples of binders which may be employed include: polyvinylpyrrolidone; macrogol; and the same compounds as are mentioned above for the vehicles.

Examples of disintegrating agents which may be employed include: the same compounds as are mentioned above for the vehicles; and chemically modified starches and celluloses, such as sodium crosscarmellose, sodium carboxymethylstarch and bridged polyvinylpyrrolidone.

Examples of stabilizers which may be employed include: paraoxybenzoates, such as methylparabene and propylparabene; alcohols, such as chlorobutanol, benzylalcohol and phenylethyl alcohol; benzalkonium chloride; phenols, such as phenol and cresol; thimerosal; dehydroacetic acid; and sorbic acid. Examples of corrigents which may be employed include: sweetening agents, acidifiers and spices.

Although the dosage will vary depending on the symptoms, age and body weight of the patient, the nature and severity of the disorder to be treated or prevented, the route of administration and the form of the drug, in general, for oral administration, a daily dosage ranging from a minimum of 1 mg (preferably 10 mg) to a maximum of 600 mg (preferably 400 mg and more preferably 200 mg) of the compound is recommended for an adult human patient, and this may be administered in a single dose or in divided doses. The invention is further illustrated by the following non-limiting Examples, of which Examples 1 and 2 illustrate the preparation of certain compounds of the present invention, whilst Examples 3 and 4 illustrate the activity of these compounds.

EXAMPLE 1

17-Dehydroxy-17-,3-(4-ammobntylamino)propylamino1oestradiol

(a) Protection of 1,4-diaminobntane

2-(t-Butoxycarbonyloxyimino)-2-phenylacetonitrile (25 g) in tetrahydrofuran (75ml, anhydrous) was added dropwise, over 2 hours, to an ice chilled solution of 1,4- diaminobutane 1 (25g) in tetrahydrofuran (75ml, anhydrous). This was allowed to stir for 18 hours. The solvent was removed by evaporation; the crude mixture taken up in water (300ml) and acidified to pH~6 using HC1 solution (2M). This was then washed with dichloromethane. The aqueous layer was then basified to ~pH 12 with NaOH (2M) and extracted with chloroform (2 x 200ml). The combined organic layers were washed with brine (2 x 150ml), dried (anhydrous sodium sulphate), filtered and evaporated down giving a clear light yellow/green oil 2 (15.1 g, -79% yield) which was sufficiently pure to be carried through to the next stage. Analysis: 1H-NMR (300MHz, CDC1₃) 1.16 (2H, br s, -NH₂), 1.44 (9H, s, -Bu⁴), 1.44- 1.54 (4H, m, CH₂), 2.71 (2H, m, CH₂), 3.13 (2H, m, CH₂), 4.70 (1H, br s, NH); ¹³C- NMR (75MHz, CDC1₃) 27.48, 28.40, 30.29, 40.43, 41.83, 78.99, 155.98; MS (CI) 189 (100%, M+l).

(b) Psendo-Michael Addition of 4-bntoxycarbonylaminobntylamine to Acrylonitrile.

Acrylonitrile (2.27ml) was added, over 2 minutes, to an ice chilled solution of 4- butoxycarbonylaminobutylamine (6.5g), prepared as described in step (a) above, in methanol (4ml), with vigorous stirring and under an inert atmosphere. After 1.5 hours, the solvent was removed giving a clear oil (8.4g, 99% yield).

Analysis: 1H-NMR (300MHz, CDC1₃) 1.21 (1H, br s, R₂NH), 1.44 ( 9H, s, -Bu¹), 1.52 (4H, m, CH₂), 2.53 (2H, t, CH₂), 2.66 (2H, m, CH₂), 2.92 CH₂), 3.13 (2H, , CH₂), 4.78 (1H, br s, ROCONHR); ¹³C-NMR (75MHz, CDC1₃) 18.69, 27.22, 27.74, 28.41, 40.31, 44.99, 48.71, 79.04, 118.72, 156.01. MS (CI) 242 (60%, M+l)

(c) Butoxycarbonyl Protection of Nitrile

Di-t-butyl dicarbonate (7.91g) was added to an ice chilled, stirring solution of 2- (4-butoxycarbonylaminobutylamino)ethyl cyanide (8.33g), prepared as described in step (b) above, in dichloromethane (60ml) under an inert atmosphere. This was allowed to stir for 18 hours, before water (400ml) was added and the mixture was extracted into ethyl acetate (400ml & 2 x 200ml). The combined organic extracts were washed with a saturated aqueous solution of sodium hydrogencarbonate (500ml). The organic layer was then washed with brine (~400ml), dried (anhydrous sodium sulphate), filtered and evaporated down giving a clear oil (12.0g, 98%yld) Analysis: 1H-NMR (300MHz, CDC1₃) 1.44 (9H, s, -Bu^l), 1.50 (9H, s, -Bu^l), 1.56 (4H, m, CH₂), 2.62 (2H, m), 3.13 (2H, m, CH₂), 3.28 (2H, t, CH₂), 3.46 (2H, m, CH₂), 4.56 (1H, br s, NH); MS (El) 705 (100%, 2M+23).

(d Hydrogenation of 2-rN-butoxycarbonyl-N-(4-butoxycarbonyIaminobutyI)- aminol ethyl cyanide

Raney Nickel (~2.5g) was added to a solution of the product of step (c) (8g) in a 1M ethanolic solution of sodium hydroxide (507ml) in a hydrogenation vessel. The vessel was sealed, purged with nitrogen (3 times) and then with hydrogen (3 times). The reactor was charged with hydrogen (3bar) and stirred at 3bar hydrogen at 30°C for 18 hours. The reaction mixture was combined with a previous batch (3g) and the mixture evaporated down under reduced pressure to ~200ml. Water (400ml) was added and the aqueous layer was extracted with chloroform (3 x 250ml). The combined organic layers were dried (anhydrous sodium sulphate), filtered and evaporated down, under reduced pressure to give 2-[N-butoxycarbonyl-N-(4-butoxycarbonylaminobutyl)- aminojpropylamine, as a clear oil (11.05g for combined reactions, quantitative yield).

Analysis: 1H-NMR (300MHz, CDC1₃) 1.31 (1H, br s, NH), 1.45 (18H, m, 2-But), 1.53 (4H, m, CH₂), 1.64 (2H, m, CH₂), 2.69 (2H, m, CH₂), 3.14 (6H, m, 3 CH₂); ¹³C-NMR (75MHz, CDC1₃) 27.40, 28.38, 28.43, 39.34, 40.16, 46.47, 78.99, 155.96.

(e) Preparation of imine by coupling a side chain to oestrone

Tetraisopropyl titanate (1.05g) was added to a solution of oestrone (lg) and 2- [^-butoxycarbonyl-N-(4-butoxycarbonylammobutyl)amino]propylamine (l .69), prepared as described in step (d) above in tetrahydrofuran (33ml, anhydrous) under an inert atmosphere at room temperature. This was allowed to stir for 18 hours before a small sample was taken for NMR analysis. ¹H-NMR studies showed that the reaction had gone in favour of the product by 65%. The reaction was left to stir for a further 24 hours. Analysis showed trace amounts of starting material. The reaction was left to stir for another 24 hours. Further tetraisopropyl titanate (280mg) was added and the reaction left to stir for 70 hours. Another small sample was removed and analysed by NMR, showing completion of the reaction. The whole sample was quenched by the addition of water (10ml). The solvent was removed by rotary evaporation, the crude mixture added into water (150ml) and extracted with chloroform (2 x 150ml). The combined organic extract was washed with brine (150ml), dried (anhydrous sodium sulphate), filtered and evaporated down giving a clear yellow/brown oil 6 (1.88g). This was of sufficient purity to use in the next stage.

Analysis: 1H-NMR (300MHz, CDC1₃) 0.86 (3H, s, Me), 1.35 - 1.50 (20H, br m, -Bu* & CH₂), 1.95 (2H, m, alkyl), 2.07 (1H, m, alkyl), 2.23 (2H, m, alkyl), 2.37 (2H, m, alkyl), 2.84 (2H, m, alkyl) 3.08-3.30 (4H, m, alkyl), 6.58 (1H, s, Ar), 6.63 (1H, d, Ar), 7.13 (lH, d, Ar).

(f) Reduction of imine

Sodium borohydride (200mg) was added in 4 portions to a stirred solution of the product of step (e) above (1.88g) in ethanol (25ml). This was allowed to stir for 24 hours under nitrogen. Water (5ml) was added and all solvent was removed by rotary evaporation. The crude mixture was taken up in water (75ml) and extracted with chloroform (2 x 100ml). The combined organic extracts were washed with brine (75ml), dried (anhydrous sodium sulphate) and evaporated down to give a clear light brown oil (1.76g). The crude mixture was purified by flash column chromatography (l%triethylamine/5% methanol/dichloromethane). An early fraction was isolated, which, when analysed by 1H-NMR, showed deprotected product. This indicated decomposition on the column. Later fractions afforded the desired product of high purity and in good percentage yield (61%). Both these compounds were kept for the next stage.

Analysis: 1H-NMR (300MHz, CDC1₃) 0.74 (3H, s, Me), 1.04 - 1.55 (20H, m, alkyl), 1.71 (2H, m, alkyl), 1.83 (IH, m, alkyl), 2.01 (2H, m, alkyl), 2.20 (2H, m, alkyl) 2.57 - 2.65 (2H, m, alkyl), 2.78 (2H, m, alkyl), 3.11 (2H, m, alkyl), 6.54 (IH, s, Ar), 6.59 (IH, d, Ar), 7.10 (2H, d, Ar).

(g) Deprotection to give 17-dehydroxy-17-f3-(4-aminobutylamino)propylaminol- oestradiol

Hydrogen chloride (3ml, 4M in dioxane) was added over 2 minutes to a stirred solution of carbamate (400mg) in dichloromethane (5ml, anhydrous) under an inert atmosphere. This was refluxed at 90°C for 3 hours and was allowed to cool. Diethyl ether (30ml) was added and the solution filtered. The filtrate was washed with diethyl ether (30ml), sucked dry for 30 minutes and placed in a vacuum oven for 18 hours with a nitrogen bleed. The sample submitted for analysis showed traces of dioxane. The sample was again placed in a vacuum oven for 24 hours, and dried to constant weight, yielding the title compound as a white powder in a 90% yield. Analysis: 1H-NMR (300MHz, d₆-DMSO) 0.66 (3H, s, Me), 1.24 - 1.40 (6H, m, alkyl), 1.53 - 1.85 (7H, m, alkyl), 2.10 - 2.32 (6H, m, alkyl), 2.71 - 3.06 (11H, m, alkyl), 6.45 (IH, s, Ar), 6.53 (IH, d, Ar), 7.06 (IH, d, Ar), 8.10 (2H, br s, NH2), 8.93 (IH, br s), 9.08 (IH, br s), 9.35 (IH, br s); ¹³C-NMR (75MHz, d₆-DMSO) 11.86, 21.96, 22.26, 22.72, 23.95, 24.58, 25.53, 26.83, 28.91, 36.04, 38.76, 42.24, 42.96, 43.75, 44.02, 45.77, 50.95, 66.75, 112.67, 114.82, 125.92, 129.78, 136.87, 154.93; MS (CI) 400 (100% M+l).

EXAMPLE 2

3-Dehvdroxy-3-F3-(4-aminobutylamino propylamino1oestradiol

(a) Coupling of Malonate Ester with 4-(t-butoxycarbonylamino)butylamine to form malonate amide

mono-Ethyl malonate (5g), l-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (7.24g), N-hydroxybenzotriazole (5.12g) and triethylamine (3ml) were added to a stirred solution of 4-(t-butoxycarbonylamino)butylamine, prepared as described in Example 1(a), (7.12g) in dimethylformamide (120ml, anhydrous), under an inert atmosphere. This was stirred at room temperature for 3 days. A small sample was washed with a saturated aqueous solution of sodium hydrogencarbonate and extracted with ethyl acetate. The organic layer was dried and concentrated down to give a sample for ΝMR & MS analysis. This confirmed the correct product had formed and that the reaction had gone to completion. The whole sample was taken up in ethyl acetate (300ml) and washed with a saturated aqueous solution of sodium hydrogencarbonate (600ml). The aqueous layer was washed with ethyl acetate and the combined organic layers were washed with brine (400ml), dried (anhydrous sodium sulphate) and evaporated down, giving the desired product, as a light brown oil (13.2g). NMR analysis showed traces of dimethylformamide remaining.

Analysis: 1H-NMR (300MHz, CDC1₃) 1.29 (3H, t, alkyl), 1.44 (9H, s, alkyl), 1.54 (4H, , alkyl), 3.14 (2H, m, alkyl), 3.30 (4H, s, alkyl), 4.19 (2H, q, alkyl), 4.73 (IH, br s, NH), 7.23 (IH, br s, NH); ¹³C-NMR (75MHz, CDC1₃) 14.00, 26.57, 27.47, 28.37, 39.16, 40.08, 41.23, 61.48, 79.09, 156.04, 165.08, 169.53; MS (ES) 325 (100%, M+23).

(b) Hydrolysis of Malonate amide to acid

An aqueous potassium carbonate solution (66ml, 1M) was added to a solution of the malonate amide prepared as described in step (a) above (13.2g) in methanol (330ml) at room temperature. This was vigorously stirred for 2 hours and was monitored by thin layer chromatography (chloroform, phosphomolybdate stain). Upon completion, the methanol was removed under reduced pressure and at a water bath temperature of -25°C. Hydrochloric acid (~100ml, concentrated) was added until fizzing had stopped. The mixture was extracted with ethyl acetate (2 x 200ml). The aqueous layer was saturated with salt and re-extracted with ethyl acetate (150ml). The combined organic layers were dried (anhydrous sodium sulphate), filtered and evaporated down to give a clear pale green oil (9.6g). NMR and MS analysis showed dimethylformamide and acetic acid remaining. The oil was left to crystallise on standing. This was slurried in hexane (100ml) and filtered. The solid was placed in a vacuum oven overnight with a nitrogen bleed to give the desired compound as a white crystalline solid (8.0g). NMR analysis showed almost complete removal of residual solvents.

Analysis: 1H-NMR (300MHz, CDC1₃) 1.44 (9H, s, alkyl), 1.52-1.60 (4H, m, alkyl), 3.14 (2H, , alkyl), 3.32 (4H, m, alkyl), 4.78 (IH, br s, NH), 7.51 (IH, br s, NH); ¹³C-NMR (75MHz, CDCI₃) ; MS (ES) 547 (100%, 2M-1). (c Triflate derivatisation of Oestrone

Triflic anhydride (2.97ml) was added over 2 minutes to a stirring solution of oestrone (4.5g) in 2,6-lutidine (9ml) and dichloromethane (90ml, anhydrous), at 0°C, under an inert atmosphere. This was left to stir for 1 hour 20 minutes. Water (100ml) was added and the layers were separated. The organic layer was washed with an aqueous solution of cupric sulphate (3 x 100ml) and brine (100ml). This was dried and concentrated under reduced vacuum to give the desired triflate as a brown oil, which crystallised on standing (6.56g).

Analysis: 1H-NMR (300MHz, CDC1₃) 0.92 (3H, s, Me), 1.41-1.70 (6H, m, alkyl), 1.96- 2.58 (7H, m, alkyl), 2.94 (2H, dd, alkyl), 6.99 (IH, s, aryl), 7.04 (IH, d, aryl), 7.34 (IH, d, aryl); ¹³C-NMR (75MHz, CDC1₃) 13.79, 21.56, 25.68, 26.08, 29.38, 31.47, 35.80, 37.74, 440.9, 47.85, 50.36, 118.31, 121.24, 217.20, 139.29, 140.27, 147.58.

(d) Conversion of Oestrone Triflate to Aminoestrone

Caesium carbonate (6.94g) was added to a solution of oestrone triflate prepared as described in step (c) above (6.1g), benzophenone imine (3.30g), palladium (LI) acetate (103mg) and (±)-2,2'-bis(diphenylphosphino)-l,r-binaphthyl (427mg) in tetiahydrofuran (60ml, anhydrous), under an inert atmosphere. This was stirred/refluxed for 18 hours. A small sample was taken up in chloroform and washed with water, the organic layer was evaporated down and analysed by NMR. This showed little of the triflate starting material remaining. The whole mixture was allowed to cool and was filtered through a Celite (trade mark) filter aid. The mother liquor was made up to 160ml with tetrahydrofuran (-100ml, anhydrous) and hydrochloric acid (42ml, 2M) was added. This was allowed to stir for 3 hours before being evaporated down to a volume of ~25ml. The solution was basified using an aqueous solution of potassium carbonate (2M) and extracted with chloroform (2 x 150ml). The organic layer was evaporated down and purified by flash column chromatography (10 - 50% ethyl acetate/hexane) with benzophenone eluting first followed by traces of the triflate starting material, followed by the desired aminoestrone product as a cream solid (2.5g). Analysis: 1H-NMR (300MHz, CDC1₃) 0.91 (3H, s, Me), 1.47-1.67 (6H, m, alkyl), 1.92- 2.55 (7H, m, alkyl), 2.84 (2H, m, alkyl), 6.45 (IH, s, aryl), 6.52 (IH, d, aryl), 7.08 (IH, d, aryl); ¹³C-NMR (75MHz, CDC1₃) 13.88, 21.59, 26.60, 29.49, 31.58, 35.89, 38.50, 43.97, 48.06, 50.40, 113.09, 115.37, 126.23, 130.07, 137.41, 144.20.

(e) Coupling of Aminoestrone with Malonic Acid Derivative

l-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (1.18g) was added to a solution of the aminoestrone prepared as described in step (d) above (690mg), N-hydroxybenzotriazole (590mg) and the acid prepared as described in step (b) above (863mg) in dimethylformamide (38ml, anhydrous). This was allowed to stir for 15 minutes to dissolve. Triethylamine (lOOdrops) was then added and the solution left to stir for 18 hours at room temperature. The reaction was monitored by Ihin layer chromatography (30% ethyl acetate/hexane, ninhydrin stain). Water (150ml) was added and the solution was extracted into chloroform (2 x 150ml). The organic extract was washed with water (100ml), dried (anhydrous sodium sulphate), and evaporated down. Residual dimethylformamide was removed by vacuum distillation (-40°C) leaving the desired compound, as an orange oil (2.0g). ΝMR and MS analysis showed a little dimethylformamide remaining.

Analysis: 1H-ΝMR (300MHz, CDC1₃) 0.91 (3H, s, Me), 1.47-1.67 (19H, m, alkyl), 1.90-2.55 (7H, m, alkyl), 2.89 (2H, m, alkyl), 3.14 (2H, m, alkyl), 3.31 (2H, m, alkyl), 4.65 (IH, br s, NH), 7.09 (IH, br s, NH), 7.25 (IH, d, aryl), 7.29 (IH, d, aryl), 7.32 (IH, s, Aryl); MS (CI) 425 (100%, M-100). (f) Reduction of 17-oxo group to hydroxyl group

Sodium borohydride (1.5g) was added portion-wise to a solution of the ketone prepared as described in step (e) above (3g) in ethanol (75ml), stirring at room temperature. This was allowed to stir for 3 hours before acetic acid (3ml) in water (5ml) was slowly added to quench the reaction. More water (150ml) was added and the solution was extracted with chloroform (3 x 150ml). The organic extracts were dried (anhydrous sodium sulphate) and evaporated under reduced pressure, giving a brown oil. The oil was purified by flash column chromatography (50-100% ethyl acetate/hexane) giving the desired product, as a clear oil (1.42g). NMR & MS analysis showed the oil to be of high purity.

Analysis: 1H-NMR (300MHz, CDC1₃) 0.78 (3H, s, Me), 1.15-2.35 (25H, m, alkyl), 2.89 (2H, m, alkyl), 3.13 (2H, m, alkyl), 3.32 (2H, m, alkyl), 3.72 (IH, m, alkyl), 4.61 (IH, br s, NH), 6.81 (IH, br s, NH), 7.26 (3H, m, aryl), 9.15 (IH, br s, OH) MS (CI) 529 (100%, M+2).

(2) Deprotection

Trifluoroacetic acid (2.2ml) was added via syringe to a stirring solution of the alcohol produced as described in step (f) above (580mg) in chloroform (30ml, anhydrous). The reaction was allowed to stir for 2 hours and was monitored by thin layer chromatography (ethyl acetate, ninhydrin). The sample was evaporated to dryness and slurried in diethyl ether (50ml) for 72 hours, forming a white semi-solid. NMR analysis of this showed absence of any BOC signals, indicating completion of reaction. The solid was taken up in water (15ml) and basified to pH -9-10 with a saturated aqueous solution of potassium carbonate. This was extracted with ethyl acetate (3 x 50ml), the organic layer was dried (anhydrous sodium sulphate), filtered and evaporated down under reduced pressure, giving a white solid (270mg). The aqueous layer was saturated with salt and re-extracted with ethyl acetate (2 x 50ml). The organic layer was again dried, filtered and evaporated down under reduced pressure, giving cream solid (lOOmg). NMR analysis of these samples confirmed the clean desired amide.

Analysis: 1H-NMR (300MHz, d₄-MeOD) 0.63 (3H, s, Me), 1.00-2.22 (16H, m, alkyl), 2.62 (2H, m, alkyl), 3.15 (3H, m, alkyl), 3.50 (IH, , alkyl), 7.09 (IH, d, Aryl), 7.17 (2H, m, aryl).

(h) Reduction of Amide to 3-dehvdroxy-3-[3-(4-aminobutylamino)propylamino1- oestradiol

Borane-tetrahydrofuran (5ml) was added to a stirring solution of the amide prepared as described in step (g) above (370mg) in tetrahydrofuran (10ml, anhydrous), under an inert atmosphere. The solution was stirred until fizzing had ceased. The solution was refluxed for 20minutes before the second addition of borane- tettahydrofuran (5ml). This was refluxed under inert atmosphere for 5 hours before a small sample was added to methanol (1ml) and evaporated down for LR analysis. This confirmed the reaction had gone to completion. The mixture was allowed to cool before methanol (10ml) was slowly added to quench the reaction. The solution was azeoxroped with methanol (3 x 10ml) after stirring in methanol for 30 minutes. A small sample was evaporated down for NMR analysis, showing clean product. Hydrochloric acid (3.1ml, 1M) and water (6ml) were added and the solution was refluxed for 20 minutes. Further hydrochloric acid (3ml, 0.5M) was added and the solution was refluxed for 3 hours. The solution was decanted, basified with a concentrated aqueous solution of potassium carbonate to pH -10-11 and extracted with chloroform (3 x 10ml). The combined organic extracts were dried, filtered and evaporated down giving a foamy solid. NMR analysis showed impurity present. The sample was dissolved in ethanol (2ml) and hydrogen chloride (3.5ml, 2M in ether) and stirred for 30 minutes. The solution was filtered, the filtrate was washed with diethyl ether and analysed by NMR, showing impurities and solvent remaining. The sample was then slurried in diethyl ether for 1 hour and dried overnight in a vacuum oven, giving the desired 3-dehydroxy-3-[3-(4- aminobutylamino)propylamino]oestradiol as a white solid (~150mg). NMR and MS analysis still showed impurity present at about ~6%.

Analysis: 1H-NMR (300MHz, d₆-DMSO) 0.45 (3H, s, Me), 0.86-1.19 (m, alkyl), 1.18- 1.35 ( , alkyl), 1.89 (m, alkyl), 2.05 (m, alkyl), 2.32 (m, alkyl), 2.49 (m, alkyl), 3.13 (m, alkyl), 3.29 (t, alkyl), 3.84 (IH, m, alkyl),6.98 (IH, s, Aryl), 7.02 (IH, d, aryl), 7.16 (IH, d, aryl), 7.89 (br s), 9.00 ( br s); ¹³C-NMR (75MHz, d₆-DMSO) 10.15, 22.30, 22.71, 25.71, 26.41, 28.95, 29.81, 36.44, 37.95, 40.08, 42.69, 47.38, 49.47, 79.91, 119.68, 122.44, 134.08, 138.11, 140.39; MS (CI) 400 (100%, M+l).

EXAMPLE 3

Anti-proliferation assay

MTT (3,[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) dye uptake assay for determination of proliferative rates

Cells (10,000) were plated (180/xl) into 96 well multiwell plates in DMEM (Dulbecco's Modified Eagle Medium) supplemented with 10% serum. Wells reserved as blanks contained medium (200μl) in the absence of any cells. Following a 24 hour attachment period, 20/xl of test compound was added at concentrations ranging from lOOnM to 50μM. Starting cell density was determined at this time (day 0) and after 96 hours drug exposure using the MTT dye uptake assay.

MTT (50μ,l) was added to the culture medium and the plate was incubated at 37°C for 4 hours. After incubation, the MTT and medium were removed from the wells and dimethyl sulphoxide (lOOμl) was added to each well. The plate was placed on an orbital mixer platform (Stuart Scientific) and agitated for 10 minutes. The absorbance at 570nm was read using a Molecular Devices plate reader and associated software (Softmax Pro).

Migration and Invasion assay

In vitro invasion or migration assays were performed using Costar Transwell cell culture chamber inserts (Coming Costar Corp.). These consist of an insert with a porous polycarbonate membrane (8 μM) resting in a tissue culture treated 12 well plate. The membranes were coated with Matrigel (0.5 ml, 100 μg/cm²) and allowed to dry overnight. Cells (2.9 x 10⁵, 0.5 ml), in serum free medium, were added to the upper chamber. Investigational drugs were included with the cells at various concentrations, determined from MTT cell proliferation assays not to inhibit cell proliferation over the time period of the invasion or migration assay. Fibroblast conditioned medium (1.5 ml) (for tumour cells) or vascular endothelial growth factor [VEGF] (for endothelial cells) was added to the lower chamber to act as a chemo-attractant. (Fibroblast conditioned medium was produced from sub-confluent human fibroblast cells that had been incubated for 24 hours). Plates were incubated at 37°C for 20 hours and any cells that had not invaded through the matrigel were removed by wiping the upper surface of the membrane with cotton wool. The membrane was then cut out of the insert and the cells fixed in methanol and stained with crystal violet. Following removal of any excess stain by washing in running water, the membranes were air dried and de-stained with 0.1 M sodium citrate in 50 % ethanol. Optical density (dependent on the number of cells that had invaded or migrated) was then read at 570 nm on a plate reader. Clonogenic Cell Survival Assay

Cells were plated (lxlO³ cells/well) in 24 well plates in RPMI1640 medium supplemented with 10% foetal calf serum and 1% penicillin/streptomycin. Following a 24 hour attachment period at 37°C, the medium was removed from each well and replaced with 1ml medium containing the appropriate drug at a range of concentrations appropriate to the drug being tested. Control cells contained medium and solvent alone. All concentrations of drug and control cells were tested in triplicate. The cells were left for 96 hours at 37°C, after which the drug-containing medium was removed and replaced with fresh media. Cells were left for 6 days at 37°C, and then the medium was removed from each well. Cells were fixed with 1ml methanol for 5 minutes at room temperature. After incubation, the methanol was removed from the cells and 1ml 0.5% crystal violet solution (lg crystal violet, 50ml methanol, 150ml distilled H₂0) was added per well. The plate was placed on an orbital mixer platform and left for 5 minutes at room temperature. After the incubation, the crystal violet solution was removed from each well and all the wells were washed twice in distilled water to remove excess stain. The plates were allowed to air dry at room temperature and were visually examined for differences between drug and drug concentrations. If necessary, the stain was reabsorbed to obtain a numerical reading. Briefly, 1ml of 0.1M sodium citrate solution was added per well and the plate was placed on the orbital mixer platform for 20 minutes to allow gentle agitation. 200μl of each sample was taken into a well in a 96 well plate and the absorbance was read at 570nm using a Molecular Devices plate reader and associated software (Softmax Pro). Samples were diluted in 0.1M sodium citrate, if necessary.

Cell Counts Cells were plated (lxlO³ cells/well) in 24 well plates in RPMI1640 medium supplemented with 10% foetal calf serum and 1% penicillin/streptomycin. Following a 24 hour attachment period at 37°C, the medium was removed from each well and replaced with 1ml medium containing the appropriate drug at a range of concentrations appropriate to the drug being tested. Control cells contained medium and solvent alone. Cell counts were performed daily. Briefly, medium was removed from each well, cells were trypsinised and then counted using a coulter counter set at the appropriate limits for the cells under investigation. All concentrations of drug and control cells were tested in triplicate. Cell growth curves were plotted using the counts obtained for a 7- day period. The cell count data supported the data obtained independently by the MTT assay.

From the results of the MTT assay, the IC5_Q values (indicated in μM) were calculated, and the results for Compounds A and B are shown in the following Table 1, whilst those for Compounds C and D are shown in Table 2. Compound A is 17- dehydroxy-17-[3-(4-aminobutylamino)propylamino]oestradiol (prepared as described in Example 1); Compound B is 3-dehydroxy-3-[3-(4-aminobutylamino)propylamino]- oestradiol (prepared as described in Example 2); Compound C is Oestrogen + spermine and has the formula:

And Compound D is Dihydrotestosterone + spermidine, which has the formula:

Compounds C and D were prepared by standard procedures known in the art Table 1

EXAMPLE 4

In this Example, Compound B was tested for its anti-angiogenesis effect.

Protocol for Cell Migration assay

The assay is well known in the art but is briefly outlined below. This assay was performed using Costar Transwell (12mm diameter 12.0μM pore size) polycarbonate membrane polystyrene 12-well plates.

Human umbilical vein epithelia cells (HUVEC, CRL 1761) were centrifuged (5minutes, 4°C @ lOOOrpm) and resuspended in serum free medium. Trypsin inhibitor was added to cells and centrifuged to wash off trypsin and inhibitor. The cells were resuspended in serum free medium. The cells were counted and diluted to appropriate concentration in serum free medium. 0.5ml cells (e.g. 5 x 10⁵ cells per well i.e. 1 x 10⁶ cells/ml) were added to each insert. Compound B was added to the cell suspension. The cells were incubated for 20 hours at 37°C. 1.5ml serum free medium containing VEGF (vascular endothelial growth factor) (Clonetics Endothelial media systems EGM- 2 Bullet Kit (CC-3202) was added to the wells to act as attractant.

The chambers were removed and the upper side of the membrane was wiped with cotton wool to remove cells. Using fine forceps and a scalpel blade, the membrane was carefully cut to remove it from the plastic insert. The membranes were placed in 12 well plates and fixed in methanol for 10 minutes. The methanol was removed and the membranes stained in crystal violet solution (0.5g crystal violet, 75ml distilled water, 25ml methanol) for 10 minutes. Each membrane was individually washed in running tap water and then allowed to air dry for approximately 1 hour.

200μl of destaining solution (2.94g sodium citrate, 50ml distilled water, 50ml ethanol) was added to each well and left for 20 minutes. This removes the stain from each membrane and the amount of stain is proportional to the stain of the cells and therefore proportional to the number of cells that have invaded/migrated through the membrane. 200μl of staining solution from each well was transferred to a 96 well plate and read at 570nm (650nm reference). The results are shown in Table 3.

Table 3

Endothelial cell migration is a critical component of the process of angiogenesis. In the described migration assay, in an apparent dose-dependent fashion, Compound B inhibited the migration of HUVEC cells at 1 and 5μM. Considering that the anti-migration effects are lower than the anti-proliferative effects described in Table 1 of Example 3 (IC50 30 M), it supports the fact that these anti-migration effects are not simply due to inhibition of cell proliferation but rather that they are the result of anti-angiogenesis effects.

Claims

CLAIMS:

1. Compounds of formula (I):

in which:

one of R and R² represents a group of formula (II):

-NH-(CH₂)_n-NH-[(CH₂)_m-NH]_χ-[(CH₂)_p-NH]_y-H (IT) where n, m and p are the same as or different from each other and each represents a number from 2 to 6; and x and y are the same as or different from each other and each represents zero or a number from 1 to 2;

and the other of R* and R² represents a hydrogen atom, an oxo group, a hydroxyl group, a C_j - Cg alkanoyl group or said group of formula (II);

R³ and R⁴ are the same as or different from each other and each represents an oxo group, a hydroxy group, a mercapto group, a hydrogen atom, a halogen atom, an alkoxy group or an aryloxy group;

R⁵ represents a hydrogen atom or a methyl group or is not present; the dotted line indicates that there may be a single or double bond between the respective carbon atoms; and

the dotted circle indicates that there may be none, one, two or three double bonds between pairs of adjacent carbon atoms;

and esters thereof.

2. Compounds according to Claim 1, in which one of R and R² represents said compound of formula (LT) and the other represents a hydrogen atom, an oxo group, a hydroxyl group, or a C_j - Cg alkanoyl group.

3. Compounds according to Claim 1 or 2, in which said group of formula (LT) is H₂N(CH₂)₄NH-, H₂N(CH₂)₃NH(CH₂)₄NH-, -HN(CH₂)₃NH(CH₂)₄NH₂ or

H₂N(CH₂)₃NH(CH₂)₄NH(CH₂)₃NH-.

4. Compounds according to Claim 1, having the formula (la):

in which R , R R³ and R⁴ are as defined in Claim 1.

5. Compounds according to Claim 4, in which one of R and R² represents said compound of formula (II) and the other represents a hydrogen atom, an oxo group, a hydroxyl group, or a Ci - Cg alkanoyl group.

6. Compounds according to Claim 4 or 5, in which said group of formula (II) is H₂N(CH₂)₄NH-, H₂N(CH₂)3NH(CH₂)4NH-, -HN(CH₂)₃NH(CH₂)₄NH₂ or

H N(CH₂)₃NH(CH₂)4NH(CH₂)₃NH-.

7. Compounds according to Claim 4, in which represents a hydroxyl group, R² represents a group of formula (LT), and R³ and R⁴ both represent hydrogen atoms.

8. Compounds according to Claim 1, having the formula (lb):

in whichR¹, R², R³ andR⁴ are as defined in Claim 1, and R^5a represents ahydrogen atom or a methyl group.

9. Compounds according to Claim 8, in which one of R and R^2, represents said compound of formula (LT) and the other represents ahydrogen atom, an oxo group, a hydroxyl group, or a Ci - C^ alkanoyl group.

10. Compounds according to Claim 8 or 9, in which said group of formula (LT) is

H₂N(CH₂)₄NH-, H₂N(CH₂)₃NH(CH₂)4NH-, -HN(CH₂)₃NH(CH₂)₄NH₂ OR H₂N(CH₂)₃NΓH(CH₂)₄NH(CH₂)₃NH-.

11. Compounds according to any one of Claims 8 to 10, in which R^5a represents a methyl group.

12. Compounds according to Claim 8, in which R represents a group of formula (LT),

R² represents a hydroxy group, R³ and R⁴ both represent hydrogen atoms, and R^5a represents a methyl group.

13. Compounds according to Claim 1, having the formula (Ic):

(in which R¹, R , R³ and R⁴ are as defined above, and R^^a represents a hydrogen atom or a methyl group.

14. Compounds according to Claim 13, in which one of R¹ and R² represents said compound of formula (LT) and the other represents a hydrogen atom, an oxo group, a hydroxyl group, or a Ci - Cg alkanoyl group.

15. Compounds according to Claim 13 or 14, in which said group of formula (II) is H₂N(CH₂)₄NH-, H₂N(CH₂)₃NH(CH₂)₄NH-, -HN(CH₂)₃NH(CH₂)₄NH2 or

H₂N(CH₂)₃NH(CH₂)₄NH(CH₂)₃NH-.

16. Compounds according to any one of Claims 13 to 15, in which R^^a represents a methyl group.

17. Compounds according to Claim 13, in which R represents a hydroxy group, R² represents a group of formula (II), R³ and R⁴ both represent hydrogen atoms, and R-^>a represents a methyl group.

18. Compounds according to Claim 1, having the formula (Id):

(in which R , R³ and R⁴ are as defined above, and R^^a represents a hydrogen atom or a methyl group.

19. Compounds according to Claim 18, in which said group of formula (LT) is H₂N(CH₂)₄NH-, H₂N(CH₂)₃NH(CH₂)₄NH-, -HN(CH2)3NH(CH₂)₄NH2 or

H2N(CH2)3NH(CH2)4NH(CH₂)₃NH-.

20. Compounds according to Claim 18, in which R represents a group of formula (LT), R³ and R⁴ both represent hydrogen atoms, and R^^a represents a methyl group.

21. The use of compounds according to any one of Claims 1 to 20 for the manufacture of a medicament for the treatment or prevention of tumours and or diabetic retinopathy.

22. The use of compounds according to any one of Claims 1 to 20 for the manufacture of a medicament for the inhibition of angiogenesis.