.siil phi np^rcttonc

(paracetamol) (7.28), as well as to an inactive metabolite, the glucuronide of 2-hydroxyl phenacetin (7.29), in small amounts.

Paracetamol has replaced phenacetin in therapy, since it is usually free from toxic effects associated with phenacetin, e.g. methaemoglobin formation. However, extensive hepatic necrosis may occur when overdoses are ingested since the normal biotransformation pathway (conjugation with glutathione) is then saturated and a highly reactive metabolite is formed which binds irreversibly to hepatic tissue. More recent work has shown that phenacetin itself possesses antipyretic activity and that this activity is not dependent on metabolism to paracetamol.


Most chemically designed pro-drugs are composed of two parts in which the active drug is linked to a pharmacologically inert molecule. The chemical bond between the two parts of the pro-drug must be sufficiently stable to withstand the pharmaceutical formulation of the pro-drug whilst permitting chemical or enzymatic cleavage at the appropiate time or site. After administration or absorption of the pro-drug, the active drug is usually released either by catalysed hydrolysis by the liver or intestinal enzymes or simply by hydrolysis although reductive processes have also been utilized. Pro-drugs are most commonly used to overcome the biological and pharmaceutical barriers which separate the site of administration of the drug from the site of action (Figure 7.1).


The pharmaceutical problems that have been addressed using prodrug design include unpalatibility, gastric irritation, pain on injection, insolubility and drug instability.

7.3.1 Patient acceptability

Unpleasant tastes and odours may often affect patient compliance. For example very young children generally require liquid medication since they are usually not amenable to swallowing capsules or coated tablets. Despite the life-threatening toxicity the antibiotic


diagrammatic representation of the pro-drug concept where a pharmaceutically active drug is converted to an inactive compound to overcome pharmaceutical and biological barriers between the site of administration and the site of action.


diagrammatic representation of the pro-drug concept where a pharmaceutically active drug is converted to an inactive compound to overcome pharmaceutical and biological barriers between the site of administration and the site of action.

chloramphenicol (7.30) it is still administered orally for the treatment of typhoid fever and salmonella infections. However the drug has an extremely bitter taste and is entirely unsuitable for administration as a suspension to such patients. To overcome this problem orally administered chloramphenicol is usually formulated as the inactive tasteless palmitate (7.31) or cinnamate (7.32) esters. The active parent drug is released from these compounds by esterases present in the small intestine.

(7J0)l R=H; chloramphenicol (731); R = CH3<CHJ1(CO — (7,32); E - CjH,Œ =CHOO —

The bitter taste of the antibiotics clindamycin and erythromycin have been similarly masked using the palmitate ester and hemisuccinate ester pro-drugs, respectively. The antimicrobial metronidazole (7.33) is another example of a drug with an unacceptably bitter taste. To overcome this problem it is administered as a suspension of benzoylmetronidazole (Flagel S®) (7.34). Likewise, ethyl dithiolisophthalate (Ditophal®) has replaced the foul smelling liquid ethyl mercaptan for the treatment of leprosy. The odourless inactive diisophthalyl thioester is metabolised to the active parent drug by thioesterases.

The formulation of insoluble compounds for parenteral delivery represents a major problem as the insoluble drug will have a tendency to precipitate on injection in an organic solvent. The solubility of such compounds may be improved by the use of phosphate or hemi-succinate pro-drugs. For example the insoluble glucocorticoids such as betamethasone, prednisolone, methylprednisolone, hydrocortisone and dexamethasone are available for injection as the water-soluble pro-drug in the form of the disodium phosphate (RO.PO32- 2Na+) or sodium hemi-succinate (RO.CO.CH2CH2COO- Na+) salts. The phosphate esters are rapidly hydrolysed to the active steroid by phosphatases, whereas the hemi-succinate salts are less efficiently hydrolysed by esterases, possibly due to the presence of an anionic centre (COO-) near the hydrolysable ester bond. The poorly water-soluble anti-inflammatory steroidal alcohol dexamethasone has been shown to rapidly (t1/2=0 min) liberate the active steroid in vivo when injected as the water-soluble phosphate (7.35).


(733); R-H; metronidazole

7.3.2 Drug solubility




The water-soluble phosphate ester of the anti-inflammatory agent oxyphenbutazone (7.36) is rapidly hydrolysed in vivo and gives higher blood levels of oxyphenbutazone on oral or intramuscular administration than attained on administration of the same doses of the parent drug.

Difficulties in the formulation of the anticonvulsant drug phenytoin (7.37) as a soluble injectable dosage form has led to the development of water-soluble pro-drugs which have been shown to have a superior in vivo performance in rats. The pro-drug is prepared by the reaction of phenytoin with an excess of formaldehyde to give the 3-hydroxymethyl intermediate (7.38), which is unstable in the absence of excess reagent. Conversion of the intermediate (7.38) to the disodium phosphate ester pro-drug (7.39) gives a water-soluble derivative. This is metabolised in vivo by phosphatases to (7.38), which rapidly breaks down (t1/2=2s) at 37°C, (pH 7.4), to give the active drug, phenytoin.

Many drugs are unstable and may either breakdown on prolonged storage or are degraded rapidly on administration. This is a particular problem on oral administration as drugs are often unstable in gastric acid. Although enteric coatings may be used, it is also possible to utilise pro-drug design to overcome this problem.

For example, the antibiotic erythromycin is destroyed by gastric acid and, as an alternative to enteric-coated tablets, it is administered orally as a more stable ester.

7.3.3 Drug stability

The inactive erythromycin estolate (laurylsulphate salt of the propionyl ester), when administered as a suspension, is rapidly absorbed and the propionyl ester converted by body esterases to the active erythromycin. The propionyl ester gives higher blood levels after oral administration on an equi-dose basis than the acetate or butyrate esters. The ethyl succinate ester has also been used.

5-Aminosalicylic acid (mesalazine) is useful in the treatment of ulcerative colitis and to a lesser degree in the management of Crohn's disease. It cannot be administered orally since firstly, it is unstable in gastric acid and secondly, it would not reach its site of action in the ileum/colon since it would be absorbed in the small intestine. Sulphasalazine (7.40), where mesalazine is covalently linked with sulphapyridine, is broken down in the colon by bacteria to the two components and in this way 5-aminosalicylic acid is delivered to the required site of action.



However, sulphapyridine is responsible for the majority of side-effects attributable to this combination and is thought to have little therapeutic activity. An alternative pro-drug, osalazine (7.41), consisting of two molecules of 5-aminosalicylic acid has been developed to overcome this problem. Reduction of the azo bond by the colonic microflora therefore liberates two molecules of 5-aminosalicylic acid. Mesalazine has also been administered orally as tablets coated with a pH-dependent acrylic-based resin which disintegrates in the terminal ileum/colon as the environment pH rises above pH 7.



(7.41); nsalazine

(7.41); nsalazine

Microbial metabolism of pro-drugs has also been utilised in the delivery of corticosteroids to the colon. Such compounds are generally readily absorbed from the upper gastrointestinal tract and therefore delivered ineffectively to the colon. Administration of corticosteroids, such as dexamethasone (7.42), as glycoside pro-drugs overcomes these problems by reducing systemic uptake in the small intestine. Pro-drugs, such as dexamethasone-P-D-glucoside (7.43), are hydrolysed by the specific glycosidases produced by the colonic bacteria and the parent corticosteroid absorbed from the lumen of the large intestine resulting in much higher concentrations in the colonic tissues.

More recently macromolecular pro-drugs have been investigated as means of overcoming instability and undesirable systemic uptake. For example, 5-aminosalicyclic acid has been linked to poly(sulphonamidoethylene) to give another mesalazine pro-drug known as polyasa (7.44) which has been shown to have less side effects than sulphasalazine and is therefore better tolerated by patients found to be allergic to or intolerant of sulphasalazine. (7.40)


Macromolecular prodrugs have also been investigated as a means of reducing degradation of drugs by gastrointestinal enzymes. For example, the coupling of the B chain of insulin to water soluble copolymers such as N-(2-hydroxypropyl)methacrylamide or poly(N-vinylpyrrolidone-co-maleic acid) appears to reduce the susceptibility of the insulin B chain to degradation by brush border peptidases in vitro.


There are a number of pharmacological problems which may be addressed by pro-drug design. These problems may be either related to pharmacokinetic, pharmacodynamic or toxic properties of the drug. Inappropriate pharmacokinetics may result in an undesirable rate of onset or duration of action of a drug. Poor pharmacodynamics may be a consequence of inefficient or unpredictable drug absorption from the gastrointestinal tract, inappropriate distribution and variable bioavailability as a consequence of presystemic metabolism or the inability to reach the site of action from the systemic circulation, e.g. penetration of the blood brain barrier. Toxic side effects may be due to non-specific drug delivery to the site of action.

Many drugs are either poorly or unpredictably absorbed from the gastrointestinal tract resulting in variation in efficacy between patients. Pro-drug design has been utilised in a number of cases to optimise the absorption of such drugs thereby improving their bioavailability.

Many penicillins are not absorbed efficiently when administered orally and their lipophilic esters have been used to improve absorption. However, simple aliphatic esters of penicillins are not active in vivo and therefore activated esters are necessary for release of the active penicillin from the inactive pro-drug. Ampicillin (7.45), a wide spectrum antibiotic, is readily absorbed orally as the inactive pro-drugs, pivampicillin (7.46), bacampicillin (7.47) and talampicillin (7.48) which are then converted by enzymic hydrolysis to ampicillin.

7.4.1 Drug absorption


The preferred pro-drug is pivampicillin since minimal hydrolysis occurs in the intestine before absorption into the systemic circulation. Pivampicillin, the pivaloyloxymethyl ester, contains an acyloxymethyl function which is rapidly hydrolysed by enzymes to the hydroxymethyl ester. This hemi-ester of formaldehyde, spontaneously cleaves with release of ampicillin and formaldehyde. In a similar manner, bacampicillin and talamipicillin are cleaved and decompose to give ampicillin together with acetaldehyde and 2-carboxybenzaldehyde, respectively.

Acyclovir (7.49) has been widely used for the treatment of herpes simplex and herpes zoster infections. This pro-drug is activated through phosphorylation by the viral thymidine kinase to acyclovir monophosphate which is then converted to the triphosphate, which inhibits DNA polymerase, by host cellular enzymes. However the use of this drug has been limited to some extent by low oral absorption; only 20% of a 200 mg dose being absorbed and little improvement being seen with doses above 800 mg. This has led to the development of a range of acyclovir prodrugs including '6-deoxyacyclovir' (BW A515U; (7.50)) which has been used for prophylaxis of herpes-virus infections in patients with haematological malignancies. It is well absorbed orally and produces plasma concentrations of the drug which are much higher than those obtained by oral administration of acyclovir. The drug (7.50) is converted to acyclovir in viro by xanthine oxidise.

An alternative orally active pro-drug is valaciclovir (7.51), the L-valyl ester of acyclovir, which is rapidly hydrolysed by first pass intestinal and hepatic metabolism. The mechanism of this biotransformation has yet to be fully elucidated but is thought to be enzymatic in nature.

Xanthine Oxidase


(7.50): Ci-dcoxyacydovir

(7,49}; acyclovir



(7.51); valaciclovir

More recently famciclovir (7.52) has been licensed in the United Kingdom for the treatment of herpes zoster infections. Famciclovir is an orally absorbed 6-deoxy, diacetyl ester pro-drug of penciclovir (7.53). This pro-drug is rapidly deacetylated and oxidised in the intestinal wall and liver to give a systemic availability of pencyclovir of 77% on oral administration. In vitro studies suggest that aldehyde oxidase, rather than xanthine oxidase, is involved in the conversion of famciclovir to penciclovir in the human liver.

Penciclovir is selectively phosphorylated by viral thymidine kinase in the same way as acyclovir. Although penciclovir triphosphate, generated by the phosphorylation of the monophosphate by cellular enzymes, is 100 times less efficient at inhibiting the DNA polymerase from herpes virus it has similar activity to acyclovir. This may in part be explained by the 10- to 20- times greater intracellular stability of penciclovir triphosphate compared to acyclovir triphosphate.

Several 2',3'-dideoxynucleoside analogues such as zidovudine (azidothymidine, AZT) (7.54) and 2',3'-didehydro-3'-deoxythymidine (D4T) (7.55) have potent antiviral activity against human immunodeficiency virus (HIV). These compounds are phosphorylated intracellularly to the 5'-triphosphate derivatives which inhibits the viral reverse transcriptase. To achieve effective metabolic antagonism against reverse transcriptase the plasma concentration of these compounds must be maintained. However, this has proved difficult because of the rapid elimination and metabolism of these compounds. Furthermore, the undesirable side effects associated with such compounds has been attributed to elevated plasma concentrations of these drugs. In an attempt to overcome these problems and to improve oral bioavailability a number of workers have recently investigated the potential of ester pro-drugs of these compounds. These studies have demonstrated that such prodrugs increase the circulating half-life whilst limiting the elevation of the plasma concentration of the parent nucleoside. Some of the ester pro-drugs were also shown to have higher absolute oral bioavailabilities than the parent nucleoside drug.

(7.52 k famciclovir

(7.53); penciclovir

(7.52 k famciclovir

(7.53); penciclovir



The use of these nucleoside analogues as antiviral and anti-neoplastic agents is also limited by their absolute requirement for kinase mediated intracellular phosphorylation. Nucleotide phosphates are unable to readily penetrate membranes and therefore have little therapeutic utility. This has led to the development of masked-phosphate prodrugs of anti-HIV nucleoside analogues, such as (7.56), which facilitate intracellular delivery of the bio-active free phosphate. These compounds have been shown to be 25 times more potent and 100 times more selective than the parent nucleosides. Unlike the parent drugs they also retain good activity against kinase-deficient cells. Such strategies also have important implications for the development of much wider ranges of compounds to combat the emergence of resistance to certain nucleoside analogues.

In another example, the antihypertensive effects on oral administration of the angiotensin-converting enzyme inhibitor enalaprilat (7.57) have been improved by conversion to the more efficiently absorbed ethyl ester, enalapril (7.58). In the active form, less than 12% is absorbed whereas the inactive derivative has an improved absorption of between 50% and 75%. The pro-drug enalapril is converted in vivo to the active enalaprilat by hydrolysis in the liver following absorption from the gastrointstinal tract.


Animal studies have shown that the oral absorption of certain basic drugs may be increased by the preparation of 'soft' quaternary salts. The 'soft' quaternary salt is formed by reaction between an a-chloromethyl ester (7.59) and the amino group of the drug. The quaternary salt formed is termed a 'soft' quaternary salt since, unlike normal quaternary salts it can release the active basic drug on hydrolysis.

'Soft' quaternary salts have useful physical properties compared with the basic drug or its salts. Water solubility may be increased compared with other salts, such as the hydrochloride, but more important there may be an increased absorption of the drug from the intestine. Increased absorption is probably due to the fact that the 'soft' quaternary salts have surfactant properties and are capable of forming micelles and unionized ion pairs with bile acids etc., which are able to penetrate the intestinal epithelium more effectively. The pro-drug, after absorption, is rapidly hydrolysed with release of the active parent drug as illustrated below.

Such an approach has also been utilised to achieve improved bioavailability of pilocarpine on ocular administration. Pilocarpine is rapidly drained from the eye resulting in a short duration of action. The 'soft' quaternary salt (7.60) has a lipophilic side-chain which has been shown to improve absorption in rabbits and gives a more prolonged effect at one tenth of the concentration of pilocarpine. The action of this compound has been shown to be due to the release of pilocarpine on hydrolytic cleavage of the ester followed by release of formaldehyde.

Topical administration is also used in the treatment of glaucoma with adrenaline (7.61). which lowers the intraocular pressure. Enhanced therapeutic efficacy may be achieved using a more lipophilic prodrug dipivefrin (7.62) which is 100 time more active than adrenaline as a consequence of more efficient corneal transport, followed by deesterification by the corneal tissue and release of adrenaline in the aqueous humor. Consequently lower doses of dipivefrin than adrenaline can be administered to achieve the same therapeutic effect. This offers advantages in reducing the side-effects associated with the use of adrenaline including cardiac effects due to systemic absorption and the accumulation of melanin deposits in the eye.

The modification of a drug to a pro-drug may lead to enhanced efficacy for the drug by differential distribution of the pro-drug in body tissues before the release of the active form.

For example, more extensive distribution of ampicillin occurs in the body tissues when the methoxymethyl ester of hetacillin (a 6-side-chain derivative of ampicillin) is administered, than is obtained with ampicillin itself. Conversely, decreased tissue distribution of a drug may occur, as was observed when adriamycin as its DNA-complex was administered as a pro-drug. Decreased tissue distribution restricts the action of a drug to a specific target site in the body and may therefore decrease its toxic side-effects, resulting from its reaction at other sites. Anticancer drugs can suppress growth in normal as well as neoplastic tissue. Improved selective localization has been achieved using non-toxic pro-drugs which release the active drug within the cancer cell as a result of either the enhanced enzyme activity in the cell or enhancement of reductase activity in the absence of molecular oxygen in hypoxic cells.

The pro-drug cyclophosphamide (7.63) is used for the treatment of certain forms of cancer and as an immunosuppressant after organ transplant. It does not possess alkylating properties and consequently is not a tissue vesicant since the electron-withdrawing properties of the adjacent phosphono-function decrease the nucleophilic properties of the P-chloroethylamino-nitrogen atom and prevent formation of the reactive alkylating ethyleniminium ion. The pro-drug requires hepatic mixed-function oxidase-mediated metabolic activation to generate 4-hydroxycyclophosphamide

(7.61): R = H; adrenaline (7.C);RsCOC(CH,K; dipi^frin

7.4.2 Drug distribution

(7.64). The 4-hydroxycyclophosphamide exists in equilibrium with its open ring tautomer aldophosphamide (7.65) which undergoes P-elimination to produce the alkylating cytotoxic phosphoramide mustard (7.66) in the target cells.


Cyclophosphamide is also metabolised by aldehyde dehydrogenase to the inactive carboxyphosphamide (7.67). Since this reaction provides a detoxification pathway, the effectiveness of cyclophosphamide is found to inversely correlate with the dehydrogenase activity of the target cells. The action of this alkylating species would be expected to be restricted to the target tissue but unfortunately in practice the action of the drug is more widespread and it shows toxicity to normal tissue, one of the apparent effects being alopecia.

Recently the organic thio-phosphate pro-drug amifostine has been introduced as a cytoprotective agent to reduce the toxic effects of cyclophosphamide on bone marrow. Amifostine uptake into normal cells occurs by facilitated diffusion and is therefore more rapid than the uptake into tumour cells by passive diffusion. As tumour cells are often hypoxic, poorly vascularised and have a low pH environment they also have reduced alkaline phosphatase activity. Amifostine (7.68) exploits these differences in uptake and enzyme activity to ensure that the pro-drug is only dephosphorylated to the active drug in healthy tissues. The active drug therefore selectively deactivates the reactive cytotoxic species produced by cyclophosphamide in non-tumour tissue without compromising the efficacy of the chemotherapy.

ami fasti rc active drug ami fasti rc active drug

In addition, the acrolein produced from (7.65) was initially found to cause bladder trouble. This problem has been overcome by either administration of cyclophosphamide together with an alkyl sulphide (sodium 2-mercaptoethanesulphonate, mesna, Uromitexan®) to remove acrolein as it is formed by addition to the P-carbon atom by a Michael reaction, or use of a modified cyclophosphamide (7.69) which does not form acrolein after ring opening.

Cl am

The anticancer effect of the pro-drug procarbazine (7.70) has also been attributed to to the formation of a cytotoxic species in the target cells. In this case, procarbazine is metabolised by the mixed function oxidase to azoprocarbazine (7.71) which undergoes further cytochrome P450 mediated oxidation to azoxy procarbazine isomers (7.72, 7.73) which liberate the diazomethane alkylating agent (7.74) in the target cells.

Tjcg*t Cells


A series of other non-toxic nitrogen mustard pro-drugs have also been designed to regenerate the parent alkylating agent in neoplastic tissues by taking advantage of the difference in the level of enzymatic amidase between normal and neoplastic cells. AfA-diallyl-3-(1-aziridino)propionamide (DAAP) is active against certain forms of leukaemia but does not cause leucopenia, a common toxic side-effect observed with other bifunctional alkylating agents. This observation suggests that DAAP is selective in its action against dividing (neoplastic) cells where a high amidase level occurs.

7.4.3 Site-specific drug delivery

Pro-drugs have more recently been used to achieve site-specific drug delivery to various tissues. Such pro-drugs are designed to ensure that the release of the active drug only occurs at its site of action thereby reducing toxic side-effects due to high plasma concentrations of the drug or non-specific uptake by other body tissues. This has led to the development of systems for site-specific delivery to the brain and to cancer cells.

The blood-brain barrier is inpenetrable to lipid insoluble and highly polar drugs. Although lipophilic pro-drugs may be used to overcome this physiological barrier, the increased lipid solubility may enhance uptake in other tissues with a resultant increase in toxicity. Furthermore, therapeutic levels of such lipophilic pro-drugs can only be maintained if there is a constant plasma concentration. These problems may be overcome by utilising a dihydropyridine—pyridinium salt type redox system. This approach was first used to enhance the penetration of the nerve gas antagonist pralidoxine into the CNS using (7.75) a non-polar pro-drug which crosses the barrier, where it is rapidly oxidized to the active form and trapped in the CNS.

More recently this approach has been developed as a general rationale for the site-specific and sustained delivery of drugs which either do not cross the blood-brain barrier readily or are rapidly metabolized. Phenylethylamine and dopamine have been used to illustrate the principles involved and in vivo work has been described in animal experiments.

The delivery system is prepared by condensing phenylethylamine with nicotinic acid to give (7.76) which is then quaternized to give (7.77). The quaternary ammonium salt (7.77) is then reduced to the 1,4-dihydro-derivative (7.78). The prodrug (7.78) is delivered directly to the brain, where it is oxidized and trapped as the pro-drug (7.77). The quaternary ammonium salt (7.77) is slowly cleaved by enzymic action with sustained release of the biologically active phenylethylamine and the facile elimination of the carrier molecule. Elimination of the drug from the general circulation is by comparison accelerated, either as (7.77) or (7.78) or as cleavage products. This rationale removes excess drug and metabolic products during or after onset of the required action. This is in contrast to normal penetration of the brain by a drug from plasma, where plasma levels must be maintained to produce the required effect and which can cause systemic side-effects.

In animal experiments the anti-inflammatory effect of topically applied hydrocortisone has been increased, and its systemic effects after absorption decreased, by use of the prodrug spirothiazolidine derivative (7.79). These beneficial effects are due to restriction of the action of hydrocortisone within the skin. After absorption, (7.79) is hydrolysed in a stepwise manner with eventual release of hydrocortisone within the skin from the accumulated pro-drug, resulting in a more intense anti-inflammatory effect and a decrease in its rate of leaching into the blood stream to produce systemic effects. The sustained release of hydrocortisone is due to retardation of the intermediate hydrolytic product (7.80) by disulphide formation (7.81) between its thiol group and a thiol group of the skin, followed by a slow breakdown of (7.81) to release hydrocortisone.


Success in cancer chemotherapy probably lies in utilizing differences in rates of growth between the rapidly-dividing tumour cells and the slower non-cycling normal tissue cells, as evidenced by responsiveness to chemotherapy of leukaemia and the high growth solid tumours. However, a different approach is needed for low growth solid tumours.

The blood supply to large solid tumours is disorganized and internal regions may be non-vasculated and the cells, termed hypoxic, deprived of oxygen. Hypoxic tissues are known to have greater powers of reduction than oxygenated areas and the reduced species are expected to be stable in the absence of molecular oxygen, which could theoretically reverse the reduction process. This knowledge has been used in the development of a rationale for targeting drugs to the internal hypoxic regions of solid tumours, these regions being relatively inaccessible to drugs that are rapidly metabolized or strongly bound to tissue components. This approach could provide a selective chemical drug-delivery system when used in combination with treatments likely to be limited by the presence of hypoxic cells.

Certain aromatic or heterocyclic nitro-containing compounds can be reduced in an hypoxic environment to produce intermediates which then fragment into alkylating species. The 2-nitro-imidazole compound misonidazole (7.82) is selectively cytotoxic to cultured hypoxic cells. Reduction of the nitro group to the hydroxylamine (R-NH2OH) probably occurs, with further fragmentation occurring to give the DNA-alkylating species including glyoxal ((CHO)2).

Nitracine (7.83) is another selective alkylating agent for hypoxic mammalian cells in culture after reduction, although the identity of the active species is unknown. Although nitracine is 105 times more potent than misonidazole in this system, it lacks activity in murine or human xenografted tumours.

Research has also been direct towards the bioactivation of aromatic nitrogen mustards, where the mechanism of action is predictable and the activation step occurs by reduction of a substituent group in the aryl ring. The alkylating ability of the P-chloroethylamine side-chain is dependent on the electron density on the nitrogen. The p-nitro substituent in (7.84), by exerting an electron withdrawing effect, reduces the electron density on the nitrogen thereby inhibiting the formation of the alkylating carbonium ion. Reduction of the nitro group in (7.84) in an hypoxic environment removes its electron withdrawing effect and restores the ability of the compound to form the alkylating species via an SN1 reaction pathway. Whether reduction gives the hydroxylamine (7.85), or the amine (7.86) is uncertain, but both species have been calculated to have greater activity than the nitro compound.

The aziridine (7.87) may be activated in a similar manner and has been shown to be selectively toxic to hypoxic cells. It should be noted that the presence of additional groups in the aryl ring may affect the actual electron density on the nitrogen atom, and hence the reactivity of the alkylating species generated, despite the activation process occurring on reduction.

Soluble macromolecular pro-drug delivery systems have also been developed to improve the pharmacokinetic profile of pharmaceutical agents by the controlled release of the active agent. It has been suggested that such soluble polymeric carriers have the potential to improve the activity of conventional antitumor agents. Recently the potential of A^-(2-hydroxypropyl)-methacrylamide (HPMA) copolymers as carriers for the anti-tumour agent doxorubicin (DOX) has been investigated. Doxorubicin was linked to the polymeric carrier by peptidyl spacers designed to be cleaved by lysosomal thiol dependent proteases, which are known to have increased activity in metastatic tumours. Such conjugates have been shown to have a broad range of antitumour activities against leukamic, solid tumour and metastatic models. Fluorescein labelled HPMA copolymers have beeen shown to accumulate in vascularised stromal regions, particularly in new growth sites in the tumour periphery. Treatment of C57 mice bearing subcutaneous B16F10 melanomas with DOX-HPMA copolymer conjugate improved the treated to control lifespan by three fold with respect to that obtained on aggressive treatment with free doxorubicin. It has been suggested that these macromolecular pro-drugs reduce toxicity by controlled drug release following passive accumulation and retention within solid tumours.

Recent research has been directed towards alternative approaches to obtain site-specific activation of pro-drugs for cancer chemotherapy using antibody-directed enzyme prodrug therapy (ADEPT) (Figure 7.2). The ADEPT approach employs an enzyme, not normally present in the extracellular fluid or on cell membranes, conjugated to an antitumour antibody which localises in the tumour via an antibody-antigen interaction on administration. Once any unbound antibody conjugate has been cleared from the systemic circulation, a pro-drug, which is specifically activated by the enzyme conjugate, is administered. The bound enzyme-antibody conjugate ensures that the pro-drug is only converted to the cytotoxic parent compound at the tumour site thereby reducing systemic toxicity. It has been shown that in systems utilising cytosine deaminase to generate 5-fluorouracil from the 5-fluorocytosine pro-drug at tumour sites, 17 times more drug can be delivered within a tumor than on adminstration of 5-fluorouracil alone.

The ADEPT approach has been recently investigated as a means of overcoming the side effects of taxol which is an effective treatment for breast cancer but also attacks healthy tissues. The system utilises a P-lactamase enzyme antitumour antibody conjugate


diagrammatic representation of the ADEPT approach to cancer chemotherapy which employs an antitumour antibody conjugated to an enzyme. The conjugate is localised at the tumour site via an antibody-antigen interaction and converts a subsequently administered pro-drug into a cytotoxic agent which attacks the tumour.


diagrammatic representation of the ADEPT approach to cancer chemotherapy which employs an antitumour antibody conjugated to an enzyme. The conjugate is localised at the tumour site via an antibody-antigen interaction and converts a subsequently administered pro-drug into a cytotoxic agent which attacks the tumour.

and a pro-drug (PROTAX) which consists of taxol linked via a short chain to cepham sulphoxide. Taxol is selectively released at the tumour site by the localised P-lactamase enzyme which is not normally found in any other tissues. In studies on cultured human breast cells it has been shown the prodrug is almost as effective as taxol on cells which have been treated with the enzyme-bound antibody, however PROTAX alone is only a tenth as toxic to cancer cells as taxol and is therefore less likely to harm healthy cells.

More recently advances in molecular biology have led to the development of a virus-directed enzyme pro-drug therapy (VDEPT) using suicide genes. Suicide genes encode for nonmammalian enzymes which can convert a pro-drug into a cytotoxic agent. Cells which are genetically transduced to express such genes essentially commit metabolic suicide in the presence of the appropriate pro-drug. Typical suicide genes include herpes simplex thymidine kinase and Escherichia coli cytosine deaminase. Viral vectors are used to carry the gene into both tumour and normal cells. Tumour specific transcription of the suicide gene is achieved by linking the foreign gene downstream of a tumour-specific transcription unit such as the proximal ERBB2 promoter. The ERBB2 oncogene is overexpressed in approximately a third of all breast and pancreatic tumours by transcriptional upregulation of the ERBB2 gene with or without gene amplification. In recent studies a chimeric minigene consisting of the proximal ERBB2 promoter linked to a gene coding for cytosine deaminase has been constructed and incorporated into a double-copy recombinent retrovirus. In vitro studies using pancreatic and breast cell lines have been used to demonstrate significant cell death on treatment of cells which expressed ERBB2 with the viral vector and 5-fluorocytosine, whereas cells which did not express ERBB2 were not affected.

Pro-drug design has also been successfully used to modify the duration of action of the parent drug by either reducing the clearance of the drug or by providing a depot of the parent drug.

The pro-drug bitolterol (7.88), which is the di-p-toluate ester of A-t-butyl noradrenaline (7.89), has been shown in dogs to provide a longer duration of bronchodilator activity than the parent drug. Furthermore, the pro-drug is preferentially distributed in lung tissues rather than plasma or heart so that the bronchodilator effect, following subsequent biotransformation of the pro-drug, is not associated with undesirable cardiovascular effects and is slow and prolonged.

The phenothiazine group of drugs, acting as tranquillizers, have been converted to long acting pro-drugs which are administered by intramuscular injection. Not only is the frequency of administration reduced but the problem associated with patient compliance is also eliminated. Flupenthixol (7.90) when administered as the decanoate ester (7.91) in an oily vehicle for the treatment of schizophrenia is released intact from the depot and subsequently hydrolysed to the parent drug, possibly after penetration of the blood-brain barrier. Maximum blood levels are observed within 11-17 days after injection and the plateau serum levels averaged 2-3 weeks in duration.

7.4.4 Sustaining drug action

NHQCHik (7,88); R = p-toluovJ; biloltero! (7.89): K = il

Similarly, perphenazine (7.92) has been used as the enanthrate ester (7.93) and pipothiazine (7.94) as the undecanoate (7.95) and palmitate (7.96) esters.

Vasopressin has been used for the treatment of bleeding varicose veins in the lower end of the oesophagus (oesophageal varices), a condition which affects about 1000 individuals annually. The vasoconstrictor action of the drug stops the bleeding, but the action is of short duration and cannot be prolonged by the use of higher doses due to

the development of toxic side-effects. Glypressin, Gly-Gly-Gly-Lys-vasopressin, is an inactive pro-drug of vasopressin and after injection the glycyl residues are steadily cleaved off by enzymic action to release the active drug. A sustained low level of vasopressin is obtained in this manner, which is sufficient to produce the required vasoconstriction effect on portal blood pressure whilst minimizing the possibility of unwanted effects caused by high blood pressure.


The examples given in this chapter illustrate the importance of the pro-drug concept as a means of overcoming pharmaceutical and pharmacological problems encountered during drug development. In addition, recent advances in biotechnology have made it possible to utilise pro-drug design to develop chemical drug delivery systems which provide various means of targetting the delivery of parent drugs to specific sites within the body. Clearly, the increasing demands for more efficacious and less toxic drugs will ensure that prodrug approaches continue to be exploited in the development of future drug substances.


Bagshawe, K.D., Sharma, S.K., Springer, C.J. and Rogers, G.T. (1994) Antibody directed enzyme prodrug therapy (ADEPT). Annals of Oncology 5, 879-891.

Bodor, N. and Farag, H.H. (1983) Improved delivery through biological membranes. 11. A redox chemical drug-delivery system and its use in brain-specific delivery of phenylethylamine. Journal of Medicinal Chemistry 26, 313-18.

Denny, W.A. and Wilson, W.R. (1986) Considerations for the design of nitrophenyl mustards as agents with selective toxicity for hypoxic tumour cells. Journal of Medicinal Chemistry 29, 879-87.

Druzgala, P., Winwood, D., Drewniak-Deyrup, M, Smith, S., Bodor, N. and Kaminski, J.J. (1992) New water-soluble pilocarpine derivatives with enhanced and sustained muscarinic activity. Pharmaceutical Research 9, 372-377.

Duncan, R. (1992) Drug polymer conjugates-potential for improved chemotherapy. Anti-Cancer Drugs 3, 175-210.

Easterbrook, P. and Wood, M.J. (1994) Successors to acyclovir. Journal of Antimicrobial Chemotherapy 34, 307-311.

Harris, J.D., Gutierrez, A.A., Hurst, H.C., Sikora, K. and Lemoine, N.R. (1994) Gene therapy for cancer using tumour-specific prodrug activation. Gene Therapy 1, 170175.

Huber, B.E., Richards, C.A. and Austin, E.A. (1994) Virus-directed enzyme/prodrug therapy-selectively engineering drug sensitivity into tumors. Annals of New York Academy of Sciences 716, 104-114.

Huennekens, F.M. (1994) Tumor targeting: activation of prodrugs by enzyme-monoclonal antibody conjugates. Trends in Biotechnology l2, 234-239.

McGuigan, C., Sheeka, H.M., Mahmood, N. and Hay, A. (1993) Phosphate derivatives of d4T as inhibitors of HIV. Bioorganic and Medicinal Chemistry Letters 3, 12031206.

Riley, T.N. (1988) The prodrug concept and new drug design and development. Journal of Chemical Education 65, 947-953.

Seymour, L.W., Ulbrich, K., Steyger, P.S., Brereton, M., Subr, V., Strohalm, J. and Duncan, R. (1994) Tumor tropism and anticancer efficacy of polymer-based doxorubicin prodrugs in the treatment of subcutaneous murine B16F10 melanoma. British Journal of Cancer 70, 636-641.

Sinkula, A.A. and Yalkowsky, S.H. (1975) Rationale for design of biologically reversible drug derivatives: prodrugs. Journal of Pharmaceutical Sciences 64, 181210.

Stella, V.J., Charman, W.N.A. and Naringrekar, V.H. (1985) Prodrugs. Do they have advantages in clinical practice? Drugs 29, 455-73. See references to other reviews cited therein.

Waller, D.G. and George, C.F. (1989) Prodrugs. British Journal of Clinical Pharmacology 28, 497-507.

0 0

Post a comment