page 673

55 Anticancer drugs

Overview

In this chapter, we deal with cancer and anticancer therapy, emphasising first the pathogenesis of cancer before proceeding to describe the drugs that can be used therapeutically. Finally, we consider the extent to which our new knowledge of cancer biology is leading to new treatments. The use of radioactive isotopes in cancer treatment is beyond the scope of this book.

Introduction

Cancer is a disease characterised by uncontrolled multiplication and spread of abnormal forms of the body’s own cells. It is the second most common cause of death in the developed nations (cardiovascular disease has the dubious distinction of heading that table) and one in three people will be diagnosed with cancer during their lifetime. In the UK, over 365 000 new cases were reported and mortality in 2006 was in excess of 154 000 (Cancer Research UK). Cancer is responsible for approximately one-quarter of all deaths in the UK, with lung and bowel cancer comprising the largest category, closely followed by breast and prostate cancer. Statistics from most other countries in the developed world tell much the same story. At first sight, incidence figures for the past 100 years or so give the impression that the disease is increasing in developed countries, but cancer is largely a disease of later life, and with advances in public health and medical science, many more people now live to an age where they are more liable to contract cancer.

The terms cancer, malignant neoplasm (neoplasm simply means ‘new growth’) and malignant tumour are synonymous. Both benign and malignant tumours manifest uncontrolled proliferation, but the latter are distinguished by their capacity for dedifferentiation, their invasiveness and their ability to metastasise (spread to other parts of the body). In this chapter, we shall be concerned only with the therapy of malignant neoplasia or cancer. The appearance of these abnormal characteristics reflects altered patterns of gene expression in the cancer cells, resulting from inherited or acquired genetic mutations.

There are three main approaches to treating established cancer—surgical excision, irradiation and drug therapy (often called chemotherapy)—and the relative value of each of these approaches depends on the type of tumour and the stage of its development. Chemotherapy may be used on its own or as an adjunct to other forms of therapy.

Compared with that of bacterial diseases, cancer chemotherapy presents a difficult problem. In biochemical terms, microorganisms are both quantitatively and qualitatively different from human cells (see Ch. 49), but cancer cells and normal cells are so similar in most respects that it is more difficult to find general, exploitable, biochemical differences between them. In recent years, the focus of cancer chemotherapy has broadened to include, as well as conventional cytotoxic drugs (which act on all cells, and rely on a small margin of selectivity to be useful as anticancer agents), several drugs that affect either the hormonal regulation of tumour growth, or the defective cell cycle controls that underlie malignancy (see below and Ch. 5). Overall, this has been one of the most fruitful fields of drug development in recent years, in which genomics and biopharmaceuticals have played a major role. The flow of innovation seems set to continue.

The Pathogenesis of Cancer

To understand the action and drawbacks of current anticancer agents and to appreciate the therapeutic hurdles that must be surmounted by putative new drugs, it is important to consider in more detail the pathobiology of this disease.

Cancer cells manifest, to varying degrees, four characteristics that distinguish them from normal cells. These are:

• uncontrolled proliferation
• dedifferentiation and loss of function
• invasiveness
• metastasis.

The Genesis of a Cancer Cell

A normal cell turns into a cancer cell because of one or more mutations in its DNA, which can be inherited or acquired, usually through exposure to viruses or carcinogens (e.g. tobacco products, asbestos). A good example is breast cancer; women who inherit a single defective copy of either of the tumour suppressor genes BRCA1 and BRCA2 (see below) have a significantly increased risk of developing breast cancer. However, carcinogenesis is a complex multistage process, usually involving more than one genetic change as well as other, epigenetic factors (hormonal, co-carcinogen and tumour promoter effects, etc.) that do not themselves produce cancer but which increase the likelihood that the genetic mutation(s) will eventually result in cancer.

There are two main categories of genetic change that are important:

1 The activation of proto-oncogenes to oncogenes. Proto-oncogenes are genes that normally control cell division, apoptosis and differentiation (see Ch. 5), but which can be converted to oncogenes that induce malignant change by viral or carcinogen action.
2 The inactivation of tumour suppressor genes. Normal cells contain genes that have the ability to suppress malignant change—termed tumour suppressor genes (antioncogenes)—and mutations of these genes are involved in many different cancers. The loss of function of tumour suppressor genes can be the critical event in carcinogenesis.
page 674

About 30 tumour suppressor genes and 100 dominant oncogenes have been identified. The changes that lead to malignancy are a result of point mutations, gene amplification or chromosomal translocation, often caused by viruses or chemical carcinogens.

The Special Characteristics of Cancer Cells

Uncontrolled Proliferation

Many healthy cells, in the bone marrow and the epithelium of the gastrointestinal tract for example, have the property of continuous rapid division, and it is not generally true that cancer cells proliferate faster than normal cells. Some cancer cells multiply slowly (e.g. those in plasma cell tumours) and some much more rapidly (e.g. the cells of Burkitt’s lymphoma). The significant issue is that cancer cells have escaped from the mechanisms that normally regulate cell division and tissue growth. It is this, rather than their rate of proliferation, that distinguishes them from normal cells.

What are the changes that lead to the uncontrolled proliferation of tumour cells? Inactivation of tumour suppressor genes or transformation of proto-oncogenes into oncogenes can confer autonomy of growth on a cell and thus result in uncontrolled proliferation by producing changes in several cellular systems (see Fig. 55.1), including:

• growth factors, their receptors and signalling pathways
• the cell cycle transducers, for example cyclins, cyclin-dependent kinases (cdks) or the cdk inhibitors
• the apoptotic machinery that normally disposes of abnormal cells
• telomerase expression
• local blood vessels, resulting from tumour-directed angiogenesis.
image

Fig. 55.1 Signal transduction pathways initiated by growth factors and their relationship to cancer development.

A few examples of proto-oncogenes and the products they code for are given in the table, with examples of the cancers that are associated with their conversion to oncogenes. Many growth factor receptors are receptor tyrosine kinases, the cytosolic transducers including adapter proteins that bind to phosphorylated tyrosine residues in the receptors. Ras proteins are guanine nucleotide-binding proteins and have GTPase action; decreased GTPase action means that Ras remains activated. EGF, epidermal growth factor; IGF, insulin-like growth factor; PDGF, platelet-derived growth factor; WTS, watch this space. *Her2 is also termed her2/neu.

Potentially all the genes coding for the above components could be regarded as oncogenes or tumour suppressor genes (see Fig. 55.2), although not all are equally prone to malignant transformation. It should be understood that malignant transformation of several components is needed for the development of cancer.

image

Fig. 55.2 Simplified outline of the genesis of cancer.

The diagram summarises the information given in the text. The genesis of cancer is usually multifactorial, involving more than one genetic change. ‘Other factors’, as specified above, may involve the actions of promoters, co-carcinogens, hormones, etc. which, while not themselves carcinogenic, increase the likelihood that genetic mutation(s) will result in cancer.

page 675

Resistance to apoptosis

Apoptosis is programmed cell death (Ch.5), and genetic mutations in the antiapoptotic genes are usually a prerequisite for cancer; indeed, resistance to apoptosis is a hallmark of the disease. It can be brought about by inactivation of proapoptotic factors or by activation of antiapoptotic factors.

Telomerase expression

Telomeres are specialised structures that cap the ends of chromosomes—like the small metal tubes on the end of shoelaces—protecting them from degradation, rearrangement and fusion with other chromosomes. Furthermore, DNA polymerase cannot easily duplicate the last few nucleotides at the ends of DNA, and telomeres prevent loss of the ‘end’ genes. With each round of cell division, a portion of the telomere is eroded, so that eventually it becomes non-functional. At this point, DNA replication ceases and the cell becomes senescent.

Rapidly dividing cells, such as stem cells and those of the bone marrow, the germline and the epithelium of the gastrointestinal tract, express telomerase, an enzyme that maintains and stabilises telomeres. While it is absent from most fully differentiated somatic cells, about 95% of late-stage malignant tumours do express the enzyme, and it is this that may confer ‘immortality’ on cancer cells.

The control of tumour-related blood vessels

The factors described above lead to the uncontrolled proliferation of individual cancer cells, but other factors, particularly blood supply, determine the actual growth of a solid tumour. Tumours 1–2 mm in diameter can obtain nutrients by diffusion, but any further expansion requires angiogenesis, the development of new blood vessels. Angiogenesis occurs in response to growth factors produced by the growing tumour (see Griffioen & Molema, 2000).

Dedifferentiation and Loss of Function

The multiplication of normal cells in a tissue begins with division of the undifferentiated stem cells giving rise to daughter cells. These daughter cells eventually differentiate to become the mature cells of the relevant tissue, ready to perform their programmed functions. For example, when fibroblasts mature, they secrete and organise extracellular matrix; mature muscle cells are capable of contraction. One of the main characteristics of cancer cells is that they dedifferentiate to varying degrees. In general, poorly differentiated cancers multiply faster and carry a worse prognosis than well-differentiated cancers.

Invasiveness

Normal cells are not generally found outside their ‘designated’ tissue of origin. This is because, during differentiation and tissue or organ growth, normal cells develop certain spatial relationships with respect to each other. These relationships are maintained by various tissue-specific survival factors that prevent apoptosis (see Ch. 5). In this way, any cells that escape accidentally lose these survival signals and die.

Consequently, although the cells of the normal mucosal epithelium of the rectum proliferate continuously as the lining is shed, they remain as a lining epithelium. A cancer of the rectal mucosa, by comparison, invades other tissues forming the rectum and often the tissues of other pelvic organs. Cancer cells have not only lost, through mutation, the restraints that act on normal cells, but they also secrete enzymes (e.g. metalloproteinases; see Ch. 5) that break down the extracellular matrix, enabling them to move around.

Metastasis

Metastases are secondary tumours (‘secondaries’) formed by cells that have been released from the initial or primary tumour and which have reached other sites through blood vessels or lymphatics, by transportation on other cells or as a result of being shed into body cavities. Metastases are the principal cause of mortality and morbidity in most cancers and constitute a major problem for cancer therapy.

As discussed above, dislodgment or aberrant migration of normal cells would lead to programmed cell death as a result of withdrawal of the necessary antiapoptotic factors. Cancer cells that metastasise have undergone a series of genetic changes that alter their responses to the regulatory factors that control the cellular architecture of normal tissues, enabling them to establish themselves ‘extraterritorially’. Tumour-induced growth of new blood vessels locally (see above) favours metastasis.

page 676

Secondary tumours occur more frequently in some tissues than in others. For example, metastases of mammary cancers are often found in lung, bone and brain. The reason for this is that breast cancer cells express chemokine receptors such as CXR4 (see Ch. 17) on their surfaces, and chemokines that recognise these receptors are expressed at high level in these tissues but not in others (e.g. kidney), facilitating the selective accumulation of cells at these sites.

General Principles of Cytotoxic Anticancer Drugs

In experiments with rapidly growing transplantable leukaemias in mice, it has been found that a given therapeutic dose of a cytotoxic drug1 destroys a constant fraction of the malignant cells. Thus a dose that kills 99.99% of cells, if used to treat a tumour with 1011 cells, will still leave 10 million (107) viable malignant cells. As the same principle holds for fast-growing tumours in humans, schedules for chemotherapy are aimed at producing as near a total cell kill as possible because, in contrast to the situation that occurs in microorganisms, little reliance can be placed on the host’s immunological defence mechanisms against the remaining cancer cells.

One of the major difficulties in treating cancer is that tumour growth is usually far advanced before cancer is diagnosed. Let us suppose that a tumour arises from a single cell and that the growth is exponential, as it may well be during the initial stages. ‘Doubling’ times vary, being, for example, approximately 24 h with Burkitt’s lymphoma, 2 weeks in the case of some leukaemias, and 3 months with mammary cancers. Approximately 30 doublings would be required to produce a cell mass with a diameter of 2 cm, containing 109 cells. Such a tumour is within the limits of diagnostic procedures, although it could easily go unnoticed. A further 10 doublings would produce 1012 cells, a tumour mass that is likely to be lethal, and which would measure about 20 cm in diameter if it were one solid mass.

However, continuous exponential growth of this sort does not usually occur. In the case of most solid tumours (for example of lung, stomach, uterus and so on), as opposed to leukaemias (tumours of white blood cells), the growth rate falls as the neoplasm grows. This is partly because the tumour outgrows its blood supply, and partly because not all the cells proliferate continuously. The cells of a solid tumour can be considered as belonging to three compartments:

1 Compartment A consists of dividing cells, possibly being continuously in cell cycle.
2 Compartment B consists of resting cells (G0 phase) which, although not dividing, are potentially able to do so.
3 Compartment C consists of cells that are no longer able to divide but which contribute to the tumour volume.

Essentially, only cells in compartment A, which may form as little as 5% of some solid tumours, are susceptible to the main current cytotoxic drugs, as is explained below. The cells in compartment C do not constitute a problem, but it is the existence of compartment B that makes cancer chemotherapy difficult, because these cells are not very sensitive to cytotoxic drugs and are liable to re-enter compartment A following chemotherapy.

Most current anticancer drugs, particularly cytotoxic agents, affect only one characteristic aspect of cancer cell biology—cell division—but have no specific inhibitory effect on invasiveness, the loss of differentiation or the tendency to metastasise. In many cases, the antiproliferative action results from an action during S phase of the cell cycle, and the resultant damage to DNA initiates apoptosis (see above). Furthermore, because their main target is cell division, they will affect all rapidly dividing normal tissues, and thus they are likely to produce, to a greater or lesser extent, the following general toxic effects:

• bone marrow toxicity (myelosuppression) with decreased leukocyte production and thus decreased resistance to infection
• impaired wound healing
• loss of hair (alopecia)
• damage to gastrointestinal epithelium (including oral mucous membranes)
• depression of growth in children
• sterility
• teratogenicity.
page 677

They can also, in certain circumstances, be themselves carcinogenic. Rapid cell destruction also entails extensive purine catabolism, and urates may precipitate in the renal tubules and cause kidney damage. Finally, in addition to specific toxic effects associated with individual drugs, virtually all cytotoxic drugs produce severe nausea and vomiting, which has been called the ‘inbuilt deterrent’ to patient compliance in completing a course of treatment with these agents.

Cancer pathogenesis and cancer chemotherapy: general principles image

• Cancer arises as a result of a series of genetic and epigenetic changes, the main genetic lesions being:
– inactivation of tumour suppressor genes
– the activation of oncogenes (mutation of the normal genes controlling cell division and other processes).
• Cancer cells have four characteristics that distinguish them from normal cells:
– uncontrolled proliferation
– loss of function because of lack of capacity to differentiate
– invasiveness
– the ability to metastasise.
• Cancer cells have uncontrolled proliferation often because of changes in:
– growth factors and/or their receptors
– intracellular signalling pathways, particularly those controlling the cell cycle and apoptosis
– telomerase expression.
• This may be supported by tumour-related angiogenesis.
• Most anticancer drugs are antiproliferative—most damage DNA and thereby initiate apoptosis. They also affect rapidly dividing normal cells and are thus likely to depress bone marrow, impair healing and depress growth. Most cause nausea, vomiting, sterility, hair loss and teratogenicity.

Anticancer Drugs

The main anticancer drugs can be divided into the following general categories:

• Cytotoxic drugs. The mechanism of action of these drugs is discussed more fully below and summarised in Table 55.1; they include:
– alkylating agents and related compounds, which act by forming covalent bonds with DNA and thus impeding replication
– antimetabolites, which block or subvert one or more of the metabolic pathways involved in DNA synthesis
– cytotoxic antibiotics, i.e. substances of microbial origin that prevent mammalian cell division
– plant derivatives (vinca alkaloids, taxanes, campothecins): most of these specifically affect microtubule function and hence the formation of the mitotic spindle.
• Hormones, of which the most important are steroids (e.g. glucocorticoids, oestrogens and androgens) as well as drugs that suppress hormone secretion or antagonise hormone action.
• Monoclonal antibodies: these are generally only of use in particular types of cancer.
• Protein kinase inhibitors: these drugs inhibit protein (usually tyrosine) kinases that transduce growth signals in rapidly dividing cells. They have a rather restricted use.
• Miscellaneous agents that do not easily fit into the above categories.

Table 55.1 An overview of anticancer drugs

image
page 678

The clinical use of anticancer drugs is the province of the specialist, who selects treatment regimens appropriate to the patient with the objective of curing, prolonging life or providing palliative therapy.2 There are over 80 drugs available in the UK, which are often used in combination. Here we discuss mechanisms of action and the main unwanted effects of commonly used anticancer agents. A recent textbook (Airley, 2009) provides detailed information.

Alkylating Agents and Related Compounds

Alkylating agents and related compounds contain chemical groups that can form covalent bonds with particular nucleophilic substances in the cell. With alkylating agents themselves, the main step is the formation of a carbonium ion—a carbon atom with only six electrons in its outer shell. Such ions are highly reactive and react instantaneously with an electron donor such as an amine, hydroxyl or sulfhydryl group. Most of the cytotoxic anticancer alkylating agents are bifunctional, i.e. they have two alkylating groups (Fig. 55.3).

image

Fig. 55.3 The effects of bifunctional alkylating agents on DNA.

Note the cross-linking of two guanines. A, adenine; C, cytosine; G, guanine; T, thymine.

The nitrogen at position 7 (N7) of guanine, being strongly nucleophilic, is probably the main molecular target for alkylation in DNA (Fig. 55.3), although N1 and N3 of adenine and N3 of cytosine may also be affected. A bifunctional agent, by reacting with two groups, can cause intra- or interchain cross-linking (Fig. 55.3). This interferes not only with transcription, but also with replication, which is probably the critical effect of anticancer alkylating agents. Other effects of alkylation at guanine N7 are excision of the guanine base with main chain scission, or pairing of the alkylated guanine with thymine instead of cytosine, and eventual substitution of the GC pair by an AT pair. Their main impact is seen during replication (S phase), when some zones of the DNA are unpaired and more susceptible to alkylation. This results in a block at G2 (see Fig. 55.3) and subsequent apoptotic cell death.

All alkylating agents depress bone marrow function and cause gastrointestinal disturbances. With prolonged use, two further unwanted effects occur: depression of gametogenesis (particularly in men), leading to sterility, and an increased risk of acute non-lymphocytic leukaemia and other malignancies.

Alkylating agents are among the most commonly employed of all anticancer drugs. A large number are available for use in cancer chemotherapy (some dozen are approved in the UK at the time of writing). Only a few commonly used ones will be dealt with here.

Nitrogen mustards

Nitrogen mustards are related to the ‘mustard gas’ used during the First World War; their basic formula (R-N-bis-(2-chloroethyl)) is shown in Figure 55.4. In the body, each 2-chloroethyl side-chain undergoes an intramolecular cyclisation with the release of a Cl−. The highly reactive ethylene immonium derivative so formed can interact with DNA (see Figs 55.3 and 55.4) and other molecules.

image

Fig. 55.4 An example of alkylation and cross-linking of DNA by a nitrogen mustard.

A bis(chloroethyl)amine (1) undergoes intramolecular cyclisation, forming an unstable ethylene immonium cation (2) and releasing Cl−, the tertiary amine being transformed to a quaternary ammonium compound. The strained ring of the ethylene immonium intermediate opens to form a reactive carbonium ion (in yellow box) (3), which reacts immediately with N7 of guanine (in green circle) to give 7-alkylguanine (bond shown in blue), the N7 being converted to a quaternary ammonium nitrogen. These reactions can then be repeated with the other –CH2CH2Cl to give a cross-link.

Cyclophosphamide is probably the most commonly used alkylating agent. It is inactive until metabolised in the liver by the P450 mixed function oxidases (see Ch. 9). It has a pronounced effect on lymphocytes and can also be used as an immunosuppressant (see Ch. 26). It is usually given orally or by intravenous injection but may also be given intramuscularly. Important toxic effects are nausea and vomiting, bone marrow depression and haemorrhagic cystitis. This last effect (which also occurs with the related drug ifosfamide) is caused by the metabolite acrolein and can be ameliorated by increasing fluid intake and administering compounds that are sulfhydryl donors, such as N-acetylcysteine or mesna (sodium-2-mercaptoethane sulfonate). These agents interact specifically with acrolein, forming a non-toxic compound. See also Chapters 9 and 57. Other nitrogen mustards used include melphalan and chlorambucil.

Estramustine is a combination of chlormethine (mustine) with an oestrogen. It has both cytotoxic and hormonal action, and is generally used for the treatment of prostate cancer.

Nitrosoureas

Examples include lomustine and carmustine. As they are lipid soluble and cross the blood–brain barrier, they may be used against tumours of the brain and meninges. However, most nitrosoureas have a severe cumulative depressive effect on the bone marrow that starts 3–6 weeks after initiation of treatment.

Other alkylating agents

Busulfan has a selective effect on the bone marrow, depressing the formation of granulocytes and platelets in low dosage and of red cells in higher dosage. It has little or no effect on lymphoid tissue or the gastrointestinal tract. It is used in chronic granulocytic leukaemia.

page 679

Dacarbazine, a prodrug, is activated in the liver, and the resulting compound is subsequently cleaved in the target cell to release an alkylating derivative. Unwanted effects include myelotoxicity and severe nausea and vomiting. Temozolomide is a related compound with a restricted usage (malignant glioma).

Procarbazine inhibits DNA and RNA synthesis and interferes with mitosis at interphase. Its effects may be mediated by the production of active metabolites. It is given orally, and its main use is in Hodgkin’s disease. It causes disulfiram-like actions with alcohol (see Ch. 56), exacerbates the effects of central nervous system depressants and, because it is a weak monoamine oxidase inhibitor, can produce hypertension if given with certain sympathomimetic agents (see Ch. 46). It causes the usual unwanted effects, and can be leukaemogenic, carcinogenic and teratogenic. Allergic skin reactions may necessitate cessation of treatment.

Other alkylating agents in clinical use include thiotepa and treosulfan.

Platinum compounds

Cisplatin is a water-soluble planar coordination complex containing a central platinum atom surrounded by two chlorine atoms and two ammonia groups. Its action is analogous to that of the alkylating agents. When it enters the cell, Cl− dissociates, leaving a reactive complex that reacts with water and then interacts with DNA. It causes intrastrand cross-linking, probably between N7 and O6 of adjacent guanine molecules, which results in local denaturation of DNA.

Cisplatin has revolutionised the treatment of solid tumours of the testes and ovary. Therapeutically, it is given by slow intravenous injection or infusion. It is seriously nephrotoxic, and strict regimens of hydration and diuresis must be instituted. It has low myelotoxicity but causes very severe nausea and vomiting. The 5-HT3 receptor antagonists (e.g. ondansetron; see Chs 15, 29 and 38) are very effective in preventing this and have transformed cisplatin-based chemotherapy. Tinnitus and hearing loss in the high-frequency range may occur, as may peripheral neuropathies, hyperuricaemia and anaphylactic reactions.

Carboplatin is a derivative of cisplatin. Because it causes less nephrotoxicity, neurotoxicity, ototoxicity, nausea and vomiting than cisplatin (although it is more myelotoxic), it is sometimes given on an outpatient basis. Oxaliplatin is another platinum-containing compound with a restricted application.

Anticancer drugs Alkylating agents and related compounds image

• Alkylating agents have groups that form covalent bonds with cell substituents; a carbonium ion is the reactive intermediate. Most have two alkylating groups and can cross-link two nucleophilic sites such as the N7 of guanine in DNA. Cross-linking can cause defective replication through pairing of alkylguanine and thymine, leading to substitution of AT for GC, or it can cause excision of guanine and chain breakage.
• Their principal effect occurs during DNA synthesis and the resulting damage triggers apoptosis.
• Unwanted effects include myelosuppression, sterility and risk of non-lymphocytic leukaemia.
• The main alkylating agents are:
– nitrogen mustards, for example cyclophosphamide, which is activated to give aldophosphamide, then converted to phosphoramide mustard (the cytotoxic molecule) and acrolein (which causes bladder damage that can be ameliorated by mesna). Cyclophosphamide myelosuppression affects particularly the lymphocytes
– nitrosoureas, for example lomustine, may act on non-dividing cells, can cross the blood–brain barrier and cause delayed, cumulative myelotoxicity.
• Platinum compounds (e.g. cisplatin) cause intrastrand linking in DNA. Cisplatin has low myelotoxicity but causes severe nausea and vomiting, and can be nephrotoxic. It has revolutionised the treatment of germ cell tumours.
page 680

Antimetabolites

Folate antagonists

The main folate antagonist is methotrexate, one of the most widely used antimetabolites in cancer chemotherapy. Folates are essential for the synthesis of purine nucleotides and thymidylate, which in turn are essential for DNA synthesis and cell division. (This topic is also dealt with in Chs 25, 49 and 53.) The main action of the folate antagonists is to interfere with thymidylate synthesis.

In structure, folates consist of three elements: a pteridine ring, p-aminobenzoic acid and glutamic acid (Fig. 55.5). Folates are actively taken up into cells, where they are converted to polyglutamates. In order to act as coenzymes, folates must be reduced to tetrahydrofolate (FH4). This two-step reaction is catalysed by dihydrofolate reductase, which converts the substrate first to dihydrofolate (FH2), then to FH4 (Fig. 55.6). FH4 functions as an essential co-factor carrying the methyl groups necessary for the transformation of 2′-deoxyuridylate (DUMP) to the 2′-deoxythymidylate (DTMP) required for the synthesis of DNA and purines. During the formation of DTMP from DUMP, FH4 is converted back to FH2, enabling the cycle to repeat. Methotrexate has a higher affinity than FH2 for dihydrofolate reductase and thus inhibits the enzyme (Fig. 55.6), depleting intracellular FH4. The binding of methotrexate to dihydrofolate reductase involves an additional bond not present when FH2 binds. The reaction most sensitive to FH4 depletion is DTMP formation.

image

Fig. 55.5 Structure of folic acid and methotrexate.

Both compounds are shown as polyglutamates. In tetrahydrofolate, one-carbon groups (R, in orange box) are transported on N5 or N10 or both (shown dotted). The points at which methotrexate differs from endogenous folic acid are shown in the blue boxes.

image

Fig. 55.6 Simplified diagram of action of methotrexate and fluorouracil on thymidylate synthesis.

Tetrahydrofolate polyglutamate FH4(glu)n functions as a carrier of a one-carbon unit, providing the methyl group necessary for the conversion of 2′-deoxyuridylate (DUMP) to 2′-deoxythymidylate (DTMP) by thymidylate synthetase. This one-carbon transfer results in the oxidation of FH4(glu)n to FH2(glu)n. Fluorouracil is converted to FDUMP, which inhibits thymidylate synthetase. DHFR, dihydrofolate reductase.

Methotrexate is usually given orally but can also be given intramuscularly, intravenously or intrathecally. The drug has low lipid solubility and thus does not readily cross the blood–brain barrier. It is, however, actively taken up into cells by the folate transport system and is metabolised to polyglutamate derivatives, which are retained in the cell for weeks (or even months in some cases) in the absence of extracellular drug. Resistance to methotrexate may develop in tumour cells by a variety of mechanisms (see below). Methotrexate is also used as an immunosuppressant drug to treat rheumatoid arthritis and other autoimmune conditions (see Ch. 26).

Unwanted effects include depression of the bone marrow and damage to the epithelium of the gastrointestinal tract. Pneumonitis can occur. In addition, high-dose regimens—doses 10 times greater than the standard doses, sometimes used in patients with methotrexate resistance—can lead to nephrotoxicity, caused by precipitation of the drug or a metabolite in the renal tubules. High-dose regimens must be followed by ‘rescue’ with folinic acid (a form of FH4).

page 681

Chemically related to folate, raltitrexed also inhibits thymidylate synthetase and pemetrexed, thymidylate transferase.

Pyrimidine analogues

Fluorouracil, an analogue of uracil, also interferes with DTMP synthesis (Fig. 55.6). It is converted into a ‘fraudulent’ nucleotide, fluorodeoxyuridine monophosphate (FDUMP). This interacts with thymidylate synthetase but cannot be converted into DTMP. The result is inhibition of DNA but not RNA or protein synthesis.

Fluorouracil is usually given parenterally. The main unwanted effects are gastrointestinal epithelial damage and myelotoxicity. Cerebellar disturbances can also occur. Another drug, capecitabine, is metabolised to fluorouracil as is tegafur.

Cytarabine (cytosine arabinoside) is an analogue of the naturally occurring nucleoside 2′-deoxycytidine. The drug enters the target cell and undergoes the same phosphorylation reactions as the endogenous nucleoside to give cytosine arabinoside trisphosphate, which inhibits DNA polymerase (see Fig. 55.7). The main unwanted effects are on the bone marrow and the gastrointestinal tract. It also causes nausea and vomiting.

image

Fig. 55.7 The mechanism of action of cytarabine (cytosine arabinoside).

For details of DNA polymerase action, see Figure 49.8. Cytarabine is an analogue of cytosine.

Gemcitabine, an analogue of cytarabine, has fewer unwanted actions, mainly an influenza-like syndrome and mild myelotoxicity. It is often given in combination with other drugs such as cisplatin.

Purine analogues

The main anticancer purine analogues include fludarabine, pentostatin, cladribine, clofarabrine, nelarabrine, mercaptopurine and tioguanine.

Fludarabine is metabolised to the trisphosphate and inhibits DNA synthesis by actions similar to those of cytarabine. It is myelosuppressive. Pentostatin has a different mechanism of action. It inhibits adenosine deaminase, the enzyme that transforms adenosine to inosine. This action interferes with critical pathways in purine metabolism and can have significant effects on cell proliferation. Cladribine, mercaptopurine and tioguanine are used mainly in the treatment of leukaemia.

Anticancer drugs Antimetabolites image

• Antimetabolites block or subvert pathways of DNA synthesis.
• Folate antagonists. Methotrexate inhibits dihydrofolate reductase, preventing generation of tetrahydrofolate interfering with thymidylate synthesis. Methotrexate is taken up into cells by the folate carrier and, like folate, is converted to the polyglutamate form. Normal cells affected by high doses can be ‘rescued’ by folinic acid. Unwanted effects are myelosuppression and possible nephrotoxicity.
• Pyrimidine analogues. Fluorouracil is converted to a ‘fraudulent’ nucleotide and inhibits thymidylate synthesis. Cytarabine in its trisphosphate form inhibits DNA polymerase. They are potent myelosuppressives.
• Purine analogues. Mercaptopurine is converted into fraudulent nucleotide. Fludarabine in its trisphosphate form inhibits DNA polymerase and is myelosuppressive. Pentostatin inhibits adenosine deaminase—a critical pathway in purine metabolism.

Cytotoxic Antibiotics

This is a widely used group of drugs that mainly produce their effects through direct action on DNA. As a rule, they should not be given together with radiotherapy, as the cumulative burden of toxicity is very high.

Doxorubicin and the anthracyclines

The main anticancer anthracycline antibiotic is doxorubicin. Other related compounds include idarubicin, daunorubicin, epirubicin and mitoxantrone (mitozantrone). Amascrine has a similar action to this group.

Doxorubicin has several cytotoxic actions. It binds to DNA and inhibits both DNA and RNA synthesis, but its main cytotoxic action appears to be mediated through an effect on topoisomerase II (a DNA gyrase; see Ch. 49), the activity of which is markedly increased in proliferating cells. The significance of the enzyme lies in the fact that, during replication of the DNA helix, reversible swivelling needs to take place around the replication fork in order to prevent the daughter DNA molecule becoming inextricably entangled during mitotic segregation. The ‘swivel’ is produced by topoisomerase II, which nicks both DNA strands and subsequently reseals the breaks. Doxorubicin intercalates in the DNA, and its effect is, in essence, to stabilise the DNA–topoisomerase II complex after the strands have been nicked, thus halting the process at this point.

page 682

Doxorubicin is given by intravenous infusion. Extravasation at the injection site can cause local necrosis. In addition to the general unwanted effects, the drug can cause cumulative, dose-related cardiac damage, leading to dysrhythmias and heart failure. This action may be the result of generation of free radicals. Marked hair loss frequently occurs.

Dactinomycin

Dactinomycin intercalates in the minor groove of DNA between adjacent guanosine–cytosine pairs, interfering with the movement of RNA polymerase along the gene and thus preventing transcription. There is also evidence that it has a similar action to that of the anthracyclines on topoisomerase II. It produces most of the toxic effects outlined above, except cardiotoxicity. It is mainly used for treating paediatric cancers.

Bleomycins

The bleomycins are a group of metal-chelating glycopeptide antibiotics that degrade preformed DNA, causing chain fragmentation and release of free bases. This action is thought to involve chelation of ferrous iron and interaction with oxygen, resulting in the oxidation of the iron and generation of superoxide and/or hydroxyl radicals. Bleomycin is most effective in the G2 phase of the cell cycle and mitosis, but it is also active against non-dividing cells (i.e. cells in the G0 phase; Fig. 5.4). It is often used to treat germline cancer. In contrast to most anticancer drugs, bleomycin causes little myelosuppression: its most serious toxic effect is pulmonary fibrosis, which occurs in 10% of patients treated and is reported to be fatal in 1%. Allergic reactions can also occur. About half the patients manifest mucocutaneous reactions (the palms are frequently affected), and many develop hyperpyrexia.

Mitomycin

Following enzymic activation, mitomycin functions as a bifunctional alkylating agent, binding preferentially at O6 of the guanine nucleus. It cross-links DNA and may also degrade DNA through the generation of free radicals. It causes marked delayed myelosuppression and can also cause kidney damage and fibrosis of lung tissue.

Anticancer drugs Cytotoxic antibiotics image

• Doxorubicin inhibits DNA and RNA synthesis; the DNA effect is mainly through interference with topoisomerase II action. Unwanted effects include nausea, vomiting, myelosuppression and hair loss. It is cardiotoxic in high doses.
• Bleomycin causes fragmentation of DNA chains. It acts on non-dividing cells. Unwanted effects include fever, allergies, mucocutaneous reactions and pulmonary fibrosis. There is virtually no myelosuppression.
• Dactinomycin intercalates in DNA, interfering with RNA polymerase and inhibiting transcription. It also interferes with the action of topoisomerase II. Unwanted effects include nausea, vomiting and myelosuppression.
• Mitomycin is activated to give an alkylating metabolite.

Plant Derivatives

Several naturally occurring plant products exert potent cytotoxic effects and have earned a place in the arsenal of anticancer drugs on that basis.

Vinca alkaloids

The vinca alkaloids are derived from the Madagascar periwinkle (Catharanthus roseus). The principal members of the group are vincristine, vinblastine and vindesine. Vinorelbine is a semisynthetic vinca alkaloid with similar properties that is mainly used in breast cancer. The drugs bind to tubulin and inhibit its polymerisation into microtubules, preventing spindle formation in dividing cells and causing arrest at metaphase. Their effects become manifest only during mitosis. They also inhibit other cellular activities that involve the microtubules, such as leukocyte phagocytosis and chemotaxis, as well as axonal transport in neurons.

The vinca alkaloids are relatively non-toxic. Vincristine has very mild myelosuppressive activity but causes paraesthesias (sensory changes), abdominal pain and muscle weakness fairly frequently. Vinblastine is less neurotoxic but causes leukopenia, while vindesine has both moderate myelotoxicity and neurotoxicity. All members of the group can cause reversible alopecia.

Paclitaxel and docetaxel

These taxanes are derived from a naturally occurring compound found in the bark of the yew tree (Taxus spp.). They act on microtubules, stabilising them (in effect ‘freezing’ them) in the polymerised state, achieving a similar effect to that of the vinca alkaloids. Paclitaxel is given by intravenous infusion and docetaxel by mouth. Both have a place in the treatment of breast cancer, and paclitaxel, given with carboplatin, is the treatment of choice for ovarian cancer.

Unwanted effects, which can be serious, include bone marrow suppression and cumulative neurotoxicity. Resistant fluid retention (particularly oedema of the legs) can occur with docetaxel. Hypersensitivity to both compounds is liable to occur and requires pretreatment with corticosteroids and antihistamines.

Camptothecins

The camptothecins irinotecan and topotecan, isolated from the stem of the tree Camptotheca acuminata, bind to and inhibit topoisomerase I, high levels of which occur throughout the cell cycle. Diarrhoea and reversible bone marrow depression occur but, in general, these alkaloids have fewer unwanted effects than most other anticancer agents.

Etoposide

Etoposide is derived from mandrake root (Podophyllum peltatum). Its mode of action is not clearly known, but it may act by inhibiting mitochondrial function and nucleoside transport, as well as having an effect on topoisomerase II similar to doxorubicin (see above). Unwanted effects include nausea and vomiting, myelosuppression and hair loss.

Anticancer drugs Plant derivatives image

• Vincristine inhibits mitosis at metaphase by binding to tubulin. It is relatively non-toxic but can cause unwanted neuromuscular effects.
• Etoposide inhibits DNA synthesis by an action on topoisomerase II and also inhibits mitochondrial function. Common unwanted effects include vomiting, myelosuppression and alopecia.
• Paclitaxel stabilises microtubules, inhibiting mitosis; it is relatively toxic and hypersensitivity reactions occur.
• Irinotecan inhibits topoisomerase I; it has relatively few toxic effects.
page 683

Hormones

Tumours arising in hormone-sensitive tissues (e.g. breast, uterus, prostate gland) may be hormone dependent, an effect related to the presence of hormone receptors in the malignant cells. Their growth can be inhibited by hormones with opposing actions, by hormone antagonists or by agents that inhibit the endogenous hormone synthesis. Hormones or their analogues that have inhibitory actions on target tissues can be used in treatment of tumours of those tissues. Such procedures alone rarely effect a cure but do retard tumour growth and mitigate the symptoms of the cancer, and thus play an important part in the clinical management of sex hormone-dependent tumours.

Glucocorticoids

Glucocorticoids such as prednisolone and dexamethasone have marked inhibitory effects on lymphocyte proliferation (see Ch. 26) and are used in the treatment of leukaemias and lymphomas. Their ability to lower raised intracranial pressure, and to mitigate some of the side effects of anticancer drugs, such as nausea and vomiting, makes them useful as supportive therapy when treating other cancers, as well as in palliative care.

Oestrogens

Diethylstilbestrol and ethinyloestradiol are two oestrogens used clinically as physiological antagonists in the palliative treatment of androgen-dependent prostatic tumours. The latter compound has fewer side effects. These tumours are also treated with gonadotrophin-releasing hormone analogues (see below).

Oestrogens can also be used to recruit resting mammary cancer cells (i.e. cells in compartment B; see above) into the proliferating pool of cells (i.e. into compartment A), thus facilitating killing by other, cytotoxic drugs.

Progestogens

Progestogens such as megestrol, norehisterone and medroxyprogesterone have been useful in endometrial neoplasms and in renal tumours.

Gonadotrophin-releasing hormone analogues

As explained in Chapter 34, analogues of the gonadotrophin-releasing hormones, such as goserelin, buserelin, leuprorelin and triptorelin, can, under certain circumstances, inhibit gonadotrophin release. These agents are therefore used to treat advanced breast cancer in premenopausal women and prostate cancer. The effect of the transient surge of testosterone secretion that can occur in patients treated in this way for prostate cancer must be prevented by an antiandrogen such as cyproterone.

Somatostatin analogues

Analogues of somatostatin such as octreotide and lanreotide (see Ch. 32) are used to relieve the symptoms of neuroendocrine tumours, including hormone-secreting tumours of the gastrointestinal tract such as VIPomas, glucagonomas, carcinoid tumours and gastrinomas. These tumours express somatostatin receptors, activation of which inhibits cell proliferation as well as hormone secretion.

Hormone Antagonists

In addition to the hormones themselves, hormone antagonists can also be effective in the treatment of several types of hormone-sensitive tumours.

Antioestrogens

An antioestrogen, tamoxifen, is remarkably effective in some cases of hormone-dependent breast cancer and may have a role in preventing these cancers. In breast tissue, tamoxifen competes with endogenous oestrogens for the oestrogen receptors and therefore inhibits the transcription of oestrogen-responsive genes. Tamoxifen is also reported to have cardioprotective effects, partly by virtue of its ability to protect low-density-lipoproteins against oxidative damage.

Unwanted effects are similar to those experienced by women following the menopause. Potentially more serious are hyperplastic events in the endometrium, which may progress to malignant changes, and the risk of thromboembolism.

Other oestrogen receptor antagonists include toremifene and fulvestrant. Aromatase inhibitors such as anastrozole, letrozole and exemestane, which suppress the synthesis of oestrogen from androgens, are also effective in the treatment of breast cancer. Aminoglutethimide, which blocks the generation of all steroids, has been largely replaced by the aromatase inhibitors.

Antiandrogens

The androgen antagonists, flutamide, cyproterone and bicalutamide, may be used either alone or in combination with other agents to treat tumours of the prostate. They are also used to control the testosterone surge (‘flare’) that is seen when treating patients with gonadorelin analogues (see above).

Adrenal hormone synthesis inhibitors

Several agents that inhibit synthesis of adrenal hormones have effects in postmenopausal breast cancer. The drugs used are trilostane and (rarely today) aminoglutethimide, which inhibit the early stages of sex hormone synthesis. Replacement of corticosteroids is necessary with these agents.

Anticancer agents Hormones image

Hormones or their antagonists are used in hormone-sensitive tumours:

• Glucocorticoids for leukaemias and lymphomas.
• Tamoxifen for breast tumours.
• Gonadotrophin-releasing hormone analogues for prostate and breast tumours.
• Antiandrogens for prostate cancer.
• Inhibitors of sex hormone synthesis for postmenopausal breast cancer.
page 684

Monoclonal Antibodies

Monoclonal antibodies are immunoglobulins, of one molecular type,3 produced by hybridoma cells in culture, that react with defined target proteins expressed on cancer cells. Some are humanised, meaning that they are hybrids or chimeras of human antibodies with a murine or primate backbone4 (and hence are less likely to be immunogenic in their own right; see Ch. 59 for more details). In some cases, binding of the antibody to its target activates the host’s immune mechanisms and the cancer cell is killed by complement-mediated lysis or by killer T cells (see Ch. 6). Other monoclonal antibodies attach to and inactivate growth factor receptors on cancer cells, thus inhibiting the survival pathway and promoting apoptosis (Fig. 5.5).

Monoclonal antibodies are relatively recent additions to the anticancer armamentarium. Unlike most of the cytotoxic drugs described above, they offer the prospect of highly targeted therapy without many of the side effects of conventional chemotherapy. This advantage is offset in most instances as they are often given in combination with more traditional drugs. Several monoclonals are in current clinical use. Their high cost is a significant problem.

Rituximab

Rituximab is a monoclonal antibody that is licensed (in combination with other chemotherapeutic agents) for treatment of certain types of lymphoma. It lyses B lymphocytes by binding to the calcium channel-forming CD20 protein and activating complement. It also sensitises resistant cells (see below) to other chemotherapeutic drugs. It is effective in 40–50% of cases when combined with standard chemotherapy.

The drug is given by infusion, and its plasma half-life is approximately 3 days when first given, increasing with each administration to about 8 days by the fourth administration.

Unwanted effects include hypotension, chills and fever during the initial infusions and subsequent hypersensitivity reactions. A cytokine release reaction can occur and has been fatal. The drug may exacerbate cardiovascular disorders.

Alemtuzumab is another monoclonal antibody that lyses B lymphocytes, and is used in the treatment of resistant chronic lymphocytic leukaemia. It may also cause a similar cytokine release reaction to that with rituximab.

Trastuzumab

Trastuzumab (Herceptin) is a humanised murine monoclonal antibody that binds to an oncogenic protein termed HER2 (the human epidermal growth factor receptor 2), a member of the wider family of receptors with integral tyrosine kinase activity (Fig. 55.1). There is some evidence that, in addition to inducing host immune responses, trastuzumab induces cell cycle inhibitors p21 and p27 (Fig. 5.2). Tumour cells, in about 25% of breast cancer patients, overexpress this receptor and the cancer proliferates rapidly. Early results show that trastuzumab given with standard chemotherapy has resulted in a 79% 1-year survival rate in treatment-naive patients with this aggressive form of breast cancer. The drug is often given with a taxane such as docetaxel.

Two mechanistically related compounds are panitumumab and cetuximab, which bind to epidermal growth factor (EGF) receptors (also overexpressed in a high proportion of tumours). They are used for the treatment of colorectal cancer usually in combination with other agents.

Unwanted effects are similar to those with rituximab.

Bevacizumab

Bevacizumab is a humanised monoclonal antibody that is also used for the treatment of colorectal cancer but would be expected to be useful for treating other cancers too. It neutralises VEGF (vascular endothelial growth factor), thereby preventing the angiogenesis that is crucial to tumour survival. It is administered by intravenous infusion and is generally combined with other agents. It is also given by direct injection into the eye to retard the progress of acute macular degeneration (AMD), a common cause of blindness associated with increased retinal vascularisation.

Protein Kinase Inhibitors

Imatinib

Hailed as a conceptual breakthrough in targeted chemotherapy, imatinib is a small-molecule inhibitor of signalling pathway kinases. It inhibits an oncogenic cytoplasmic kinase (Bcr/Abl, see Fig. 55.1 and Fig. 55.8) considered to be a unique factor in the pathogenesis of chronic myeloid leukaemia (CML), and also inhibits platelet-derived growth factor (a receptor tyrosine kinase; Fig. 55.1). It has greatly improved the hitherto poor prognosis of patients with CML, and is also used for the treatment of some gastrointestinal tumours not susceptible to surgery.

image

Fig. 55.8 The mechanism of action of anticancer monoclonal antibodies and protein kinase inhibitors.

Many tumours overexpress growth factor receptors such as EGFR, the proto-oncogene HER2 or VEGFR. Therapeutic monoclonals can prevent this by interacting directly with the receptor itself (e.g. trastuzumab, cetuximab) or with the ligand (e.g. bevacizumab). An alternate way of reducing this drive on cell proliferation is by inhibiting the downstream signalling cascade. The receptor tyrosine kinases are good targets as are some oncongenic kinases such as bcr/abl.

The drug is given orally. The half-life is long, about 18 h, and the main site of metabolism is in the liver, where approximately 75% of the drug is converted to a metabolite that is also biologically active. The bulk (81%) of the metabolised drug is excreted in the faeces.

Unwanted effects include gastrointestinal symptoms (pain, diarrhoea, nausea), fatigue, headaches and sometimes rashes. Resistance to imatinib, resulting from mutation of the kinase gene, is a growing problem. It results in little or no cross-resistance to other kinase inhibitors (see below).

Other mechanistically similar drugs which inhibit the bcr-abl kinase include dasatinib and nilotinib while erlotinib targets the EGFR kinase and sunitinib another tyrosine kinase. Sorafenib inhibits all these kinases. Several kinase inhibitors are currently in development, and are expected to make a significant contribution to cancer therapy in the foreseeable future.

Anticancer drugs Monoclonal antibodies and protein kinase inhibitors image

• Many tumours overexpress growth factor receptors that therefore stimulate cell proliferation and tumour growth. This can be inhibited by:
– monoclonal antibodies which bind to the extracellular domain of EGF (e.g. panitumumab) the oncogenic receptor HER2 (e.g. trastuzumab) or which neutralise the growth factors themselves (e.g. VEGF; bevacizumab)
– protein kinase inhibitors which prevent downstream signalling triggered by growth factors by inhibiting specific oncogenic kinases (e.g. imatinib; bcr/abl) or by inhibiting specific receptor tyrosine kinases (e.g. EGF receptor; erlotinib) or several receptor-associated kinases (e.g. sorefenib).
• Some monoclonals act directly on lymphocyte cell surface proteins to cause lysis (e.g. rituximab), thereby preventing proliferation.
page 685

Miscellaneous Agents

Crisantaspase

image Crisantaspase is a preparation of the enzyme asparaginase, given intramuscularly or intravenously. It converts asparagine to aspartic acid and ammonia, and is active against tumour cells, such as those of acute lymphoblastic leukaemia, that have lost the capacity to synthesise asparagine and therefore require an exogenous source. As most normal body cells are able to synthesise asparagine, the drug has a fairly selective action and has very little suppressive effect on the bone marrow, the mucosa of the gastrointestinal tract or hair follicles. It may cause nausea and vomiting, central nervous system depression, anaphylactic reactions and liver damage.

Hydroxycarbamide

image Hydroxycarbamide (hydroxyurea) is a urea analogue that inhibits ribonucleotide reductase, thus interfering with the conversion of ribonucleotides to deoxyribonucleotides. It is mainly used to treat polycythaemia rubra vera (a myeloproliferative disorder of the red cell lineage) and (in the past) chronic myelogenous leukaemia. Its use (in somewhat lower dose) in sickle cell anaemia is described in Chapter 25. It has the familiar spectrum of unwanted effects, bone marrow depression being significant.

Bortezomib

image Bortezomib is a boron-containing tripeptide that inhibits cellular proteasome function. For some reason, rapidly dividing cells are more sensitive than normal cells to this drug, making it a useful anticancer agent. It is generally used for the treatment of myeloma (a malignant bone marrow tumour).

Thalidomide

image Investigations of the notorious teratogenic effect of thalidomide showed that it has multiple effects on gene transcription, angiogenesis and proteasome function, leading to trials of its efficacy as an anticancer drug. In the event, it proved efficacious in myeloma, for which it is now widely used. The main adverse effect of thalidomide, apart from teratogenesis (irrelevant in myeloma treatment), is peripheral neuropathy, leading to irreversible weakness and sensory loss. It also increases the incidence of thrombosis and stroke.

A thalidomide derivative lenalidomide is thought to have fewer adverse effects, but unlike thalidomide, can cause bone marrow depression and neutropenia.

page 686

Biological response modifiers

image Agents that enhance the host’s response are referred to as biological response modifiers. Some, for example interferon-α (and its pegylated derivative), are used in treating some solid tumours and lymphomas, and aldesleukin (recombinant interleukin-2) is used in some cases of renal tumours. Tretinoin (a form of vitamin A) is a powerful inducer of differentiation in leukaemic cells and is used as an adjunct to chemotherapy to induce remission.

Resistance to Anticancer Drugs

The resistance that neoplastic cells manifest to cytotoxic drugs is said to be primary (present when the drug is first given) or acquired (developing during treatment with the drug). Acquired resistance may result from either adaptation of the tumour cells or mutation, with the emergence of cells that are less susceptible or resistant to the drug and consequently have a selective advantage over the sensitive cells. The following are examples of various mechanisms of resistance. See Mimeault et al. (2008) for an up-to-date appraisal of this issue.

• Decreased accumulation of cytotoxic drugs in cells as a result of the increased expression of cell surface, energy-dependent drug transport proteins. These are responsible for multidrug resistance to many structurally dissimilar anticancer drugs (e.g. doxorubicin, vinblastine and dactinomycin; see Gottesman et al., 2002). An important member of this group is P-glycoprotein (P-gp/MDR1; see Ch. 8). The physiological role of P-glycoprotein is thought to be the protection of cells against environmental toxins. It functions as a hydrophobic ‘vacuum cleaner’, picking up foreign chemicals, such as drugs, as they enter the cell membrane and expelling them. Non-cytotoxic agents that reverse multidrug resistance are being investigated as potential adjuncts to treatment.
• A decrease in the amount of drug taken up by the cell (e.g. in the case of methotrexate).
• Insufficient activation of the drug. Some drugs require metabolic activation to manifest their antitumour activity. If this fails, they may no longer be effective. Examples include conversion of fluorouracil to FDUMP, phosphorylation of cytarabine and conversion of mercaptopurine to a fraudulent nucleotide.
• Increase in inactivation (e.g. cytarabine and mercaptopurine).
• Increased concentration of target enzyme (methotrexate).
• Decreased requirement for substrate (crisantaspase).
• Increased utilisation of alternative metabolic pathways (antimetabolites).
• Rapid repair of drug-induced lesions (alkylating agents).
• Altered activity of target, for example modified topoisomerase II (doxorubicin).
• Mutations in various genes, giving rise to resistant target molecules. For example, the p53 gene and overexpression of the Bcl-2 gene family (several cytotoxic drugs).

Treatment Schedules

Treatment with combinations of several anticancer agents increases the cytotoxicity against cancer cells without necessarily increasing the general toxicity. For example, methotrexate, with mainly myelosuppressive toxicity, may be used in a regimen with vincristine, which has mainly neurotoxicity. The few drugs we possess with low myelotoxicity, such as cisplatin and bleomycin, are good candidates for combination regimens. Treatment with combinations of drugs also decreases the possibility of the development of resistance to individual agents. Drugs are often given in large doses intermittently in several courses, with intervals of 2–3 weeks between courses, rather than in small doses continuously, because this permits the bone marrow to regenerate during the intervals. Furthermore, it has been shown that the same total dose of an agent is more effective when given in one or two large doses than in multiple small doses.

Drug action during the cell cycle

image Cells that are constantly replicating constitute the ‘growth fraction’ of the tumour. Some anticancer drugs act at particular phases on the cell cycle, as shown below, and in principle this information could be of value in selecting individual agents or combinations for clinical use. However, not all authorities agree that treatment schedules based on these principles are better than purely empirical schedules.

• Phase-specific agents. Many cytotoxic drugs act at different points in the cycle. For example, the vinca alkaloids act in mitosis, whereas cytarabine, hydroxycarbamide, fluorouracil, methotrexate and mercaptopurine act in S phase. Some of these compounds also have some action during G1 phase and thus may slow the entry of a cell into S phase, where it would be more susceptible to the drug.
• Cycle-specific agents. These act at all stages of the cell cycle but do not have much effect on cells out of cycle (e.g. alkylating agents, dactinomycin, doxorubicin and cisplatin).
• Cycle non-specific agents. These act on cells whether in cycle or not (e.g. bleomycin and nitrosoureas).

Control of Emesis and Myelosuppression

Emesis

The nausea and vomiting induced by many cancer chemotherapy agents constitute an inbuilt deterrent to patient compliance (see also Ch. 29). It is a particular problem with cisplatin but also complicates therapy with many other compounds, such as the alkylating agents. 5-hydroxytryptamine (HT)3 receptor antagonists such as ondansetron or granisetron (see Chs 15 and 29) are effective against cytotoxic drug-induced vomiting and have revolutionised cisplatin chemotherapy. Of the other antiemetic agents available, metoclopramide, given intravenously in high dose, has proved useful and is often combined with dexamethasone (Ch. 32) or lorazepam (Ch. 43), both of which further mitigate the unwanted effects of chemotherapy. As metoclopramide commonly causes extrapyramidal side effects in children and young adults, diphenhydramine (Ch. 26) can be used instead.

Myelosuppression

Myelosuppression limits the use of many anticancer agents. Regimens contrived to surmount the problem have included removal of some of the patient’s own bone marrow prior to treatment, purging it of cancer cells (using specific monoclonal antibodies; see below) and replacing it after cytotoxic therapy is finished. A protocol in which aliquots of stem cells, harvested from the blood following administration of the growth factor molgramostim, are expanded in vitro using further haemopoietic growth factors (Ch. 25) is now frequently used. The use of such growth factors after replacement of the marrow has been successful in some cases. A further possibility is the introduction, into the extracted bone marrow, of the mutated gene that confers multidrug resistance, so that when replaced, the marrow cells (but not the cancer cells) will be resistant to the cytotoxic action of the anticancer drugs.

page 687

Future Developments

As the reader will have judged by now, our current approach to cancer chemotherapy embraces an eclectic mixture of drugs and techniques, all designed to target selectively cancer cells. Real therapeutic progress has been achieved, although ‘cancer’ as a disease (actually many different diseases with a similar outcome) has not been defeated and remains a massive challenge for future generations of researchers. In this therapeutic area, probably more than in any other, the debate about the risk–benefit of treatment and the patient quality of life issues has taken centre stage and remains a major area of concern. These sensitive issues have been explored by Duric & Stockler (2001) and Klastersky & Paesmans (2001).

The quest for less toxic forms of therapy is, of course, central to anticancer initiatives, and many new drugs or novel combination regimens are in clinical trial or at earlier stages of development (see, for example, Kurtz et al., 2003). What follows is a selection of new and different approaches to the treatment of cancer that may bear fruit over the next decade.

Angiogenesis and metalloproteinase inhibitors

Tumour cells produce metalloproteinases and angiogenic factors that facilitate tumour growth, invasion of normal tissue and metastases. Targeting the mechanisms involved could provide us with drugs that block metastases. Several existing drugs already target this process (e.g. bevacizumab) and it is likely that this area will see further development (see Griffioen & Molema, 2000; Thijssen et al., 2007).

Cyclo-oxygenase inhibitors

There is strong epidemiological and experimental evidence suggesting that chronic use of cyclo-oxygenase (COX) inhibitors (see Ch. 26) protects against cancer of the gastrointestinal tract and possibly other sites as well. The COX-2 isoform is overexpressed in about 85% of cancers, and prostanoids originating from this source may activate signalling pathways that enable cells to escape from apoptotic death. The COX-2 inhibitor celecoxib reduces mammary and gastrointestinal cancer incidence in animal models and causes regression of existing tumours. It is in trial in humans as an inhibitor of a familial type of colon tumour. Overall, COX-2 is now considered to be a potentially important target for anticancer drug development although, ironically, some argue that the mechanism of action is unrelated to COX inhibition. The literature is daunting and often controversial; see Karamouzis & Papavassiliou (2004) for recent comment.

Antisense oligonucleotides

Genetic approaches are seen by many experts as the hope for the future. Antisense oligonucleotides (see Ch. 59) are synthetic sequences of single-stranded DNA complementary to specific coding regions of mRNA, which can inhibit gene expression. An antisense drug, augmerosen, downregulates the antiapoptotic factor Bcl-2. In an early clinical trial, it sensitised malignant melanoma to standard anticancer drugs.

Gene therapy

The introduction of engineered genes, antisense oligonucleotides or siRNA by gene therapy (see Ch. 59) offers, in principle, enormous advantages over conventional approaches in terms of selective toxicity to cancer cells. There are many technical problems yet to be solved with the delivery of the genes, (e.g. p53 or growth factor antisense DNA) into the target cells. There have been clinical trials, some of which showed modest success (see, for example, Wolf & Dwayne Jenkins, 2002, on ovarian cancer trials), but progress has been disappointingly slow.

Reversal of multidrug resistance

Several non-cytotoxic drugs (e.g. verapamil) that inhibit P-glycoprotein can reverse multidrug resistance. Other drugs with this action are being investigated. In addition, the use of antibodies, immunotoxins, antisense oligonucleotides (see above) or liposome-encapsulated agents may be useful in the elimination of cells with multidrug resistance (reviewed by Gottesman & Pastan, 1993).

Telomerase is known to be important in maintaining cancer cell viability. Several strategies for controlling its activity have been reviewed by Keith et al. (2004).

References and Further Reading

General textbook

Airley R. Anticancer drugs. Chichester: Wiley-Blackwell; 2009. (Recent textbook covering all aspects from basic pharmacology to clinical use)

Mechanisms of carcinogenesis

Buys C.H.C.M. Telomeres, telomerase and cancer. N. Engl. J. Med.. 2000;342:1282-1283. (Clear, concise coverage)

Chambers A.F., Groom A.C., MacDonald I.C. Dissemination and growth of cancer cells in metastatic sites. Nat. Rev. Cancer. 2, 2002. 563-557. (Review; stresses the importance of metastases in most cancer deaths, discusses the mechanisms involved in metastasis and raises the possibility of targeting these in anticancer drug development)

Griffioen A., Molema G. Angiogenesis: potentials for pharmacologic intervention in the treatment of cancer, cardiovascular diseases and chronic inflammation. Pharmacol. Rev.. 2000;52:237-268. (Comprehensive review covering virtually all aspects of angiogenesis and the potential methods of modifying it to produce an antineoplastic effect)

Haber D.A., Fearon E.R. The promise of cancer genetics. Lancet. 1998;351:1-8. (Excellent coverage; detailed tables of mutations in proto-oncogenes and tumour suppressor genes in human cancers)

Mimeault M., Hauke R., Batra S.K. Recent advances on the molecular mechanisms involved in the drug resistance of cancer cells and novel targeting therapies. Clin. Pharmacol. Ther.. 2008;83:673-691. (Comprehensive review covering all aspects of this field)

Talapatra S., Thompson C.B. Growth factor signalling in cell survival: implications for cancer treatment. J. Pharmacol. Exp. Ther.. 2001;298:873-878. (Succinct overview of death receptor-induced apoptosis, the role of growth factors in preventing it and potential drugs that could be used to promote cell death)

Weinberg R.A. How cancer arises. Sci. Am.. 1996;Sept:42-48. (Simple, clear overview, listing main oncogenes, tumour suppressor genes and the cell cycle; excellent diagrams)

Zörnig M., Hueber A.-O., Baum W., Evan G. Apoptosis regulators and their role in tumorigenesis. Biochim. Biophys. Acta. 2001;1551:F1-F37. (Extensive review describing the genes and mechanisms involved in apoptosis, and summarising the evidence that impaired apoptosis is a prerequisite for cancer development)

page 688

Anticancer therapy

Gottesman M.M., Fojo T., Bates S.E. Multidrug resistance in cancer: role of ATP-dependent transporters. Nat. Rev. Cancer. 2002;2:48-56. (Outlines cellular mechanisms of resistance; describes ATP-dependent transporters, emphasising those in human cancer; considers resistance reversal strategies)

Houghton A.N., Scheinberg D. Monoclonal antibody therapies—a ‘constant’ threat to cancer. Nat. Med.. 2000;6:373-374. (Lucid article; very useful diagram)

Krause D.S., Van Etten R. Tyrosine kinases as targets for cancer therapy. N. Engl. J. Med.. 2005;353:172-187. (Excellent review on tyrosine kinases as targets; good diagrams and tables as well as a highly readable style)

Kurtz J.-E., Emmanuel A., Natarajan-Ame S., et al. Oral chemotherapy in colorectal cancer treatment: review of the literature. Eur. J. Int. Med.. 2003;14:18-25. (Discusses potential new leads in colorectal cancer; good tables summarising recent advances and clinical trials)

Norman K.L., Farassati F., Lee P.W.K. Oncolytic viruses and cancer therapy. Cytokine Growth Factor Rev.. 2001;12:271-282. (Describes mechanisms of action and efficacy of three oncolytic viruses in clinical trial)

Overall C.M., López-Otin C. Strategies for MMO inhibition in cancer: innovations for the post-trial era. Nat. Rev. Cancer. 2002;2:6577-7672. (Review of matrix metalloproteinases and their role in tumour metastasis; also discusses various approaches that could be used to target metalloproteinases, thus producing new anticancer drugs)

Reed J.C. Apoptosis-based therapies. Nat. Rev. Drug Discov.. 2002;1:111-121. (Excellent coverage, useful tables, good diagrams)

Savage D.G., Antman K.H. Imatinib mesylate—a new oral targeted therapy. N. Engl. J. Med.. 2002;346:683-693. (Review with detailed coverage of this drug for chronic myelogenous leukaemia; very good diagrams)

White C.A., Weaver R.L., Grillo-López. Antibody-targeted immunotherapy for treatment of malignancy. Annu. Rev. Med.. 2001;52:125-145. (Clear, comprehensive review; includes tables of monoclonals and radiolabelled monoclonals in clinical trial)

New directions and miscellaneous

Adjei A.A. Blocking oncogenic Ras signaling for cancer therapy. J. Natl. Cancer Inst.. 2001;93:1062-1074. (Gives details of Ras processing, activation, mutations, cytoplamsic targets and physiological role, and outlines therapeutic implications)

Anderson W.F. Gene therapy scores against cancer. Nat. Med.. 2000;6:862-863. (Short crisp article)

Armstrong A.C., Eaton D., Ewing J.C. Cellular immunotherapy for cancer. Br. Med. J.. 2001;323:1289-1293. (Brief discussion of rationale and possible future exploitation of tumour cell and dendritic cell vaccines and T cell therapy)

Carter P. Improving the efficacy of antibody-based cancer therapies. Nat. Rev. Cancer. 2001;1:118-128. (Review considering the possible future use of monoclonal antibodies to treat cancer; lists antibodies in advanced clinical trials)

Duric V., Stockler M. Patients’ preferences for adjuvant chemotherapy in early breast cancer. Lancet Oncol.. 2001;2:691-697. (The title is self-explanatory; deals with patients’ assessment of quality of life issues)

English J.M., Cobb M.H. Pharmacological inhibitors of MAPK pathways. Trends Pharmacol. Sci.. 2002;23:40-45. (Lists mitogen-activated protein kinases and discusses small-molecule inhibitors under investigation)

Favoni R.E., de Cupis A. The role of polypeptide growth factors in human carcinomas: new targets for a novel pharmacological approach. Pharmacol. Rev.. 2000;52:179-206. (Thorough review that describes 14 growth factor families, their signalling pathways and their possible role in cancer; it also deals with drug action on signalling pathways)

Gottesman M.M., Pastan I. Biochemistry of multidrug resistance mediated by the multidrug transporter. Annu. Rev. Biochem.. 1993;62:385-427.

Karamouzis M.V., Papavassiliou A.G. COX-2 inhibition in cancer therapeutics: a field of controversy or a magic bullet? Expert Opin. Investig. Drugs. 2004;13:359-372. (Good review of the field of COX inhibitors in cancer therapy)

Keith W.N., Bilsland A., Hardie M., Evans T.R. Drug insight: cancer cell immortality—telomerase as a target for novel cancer gene therapies. Nat. Clin. Pract. Oncol.. 2004;1:88-96.

Klastersky J., Paesmans M. Response to chemotherapy, quality of life benefits and survival in advanced non-small lung cancer: review of literature results. Lung Cancer. 2001;34:S95-S101. (Another paper that addresses quality of life issues surrounding chemotherapy)

Smith I.E. New drugs for breast cancer. Lancet. 2002;360:790-792. (Succinct coverage)

Thijssen V.L., van Beijnum J.R., Mayo K.H., Griffioen A.W. Identification of novel drug targets for angiostatic cancer therapy; it takes two to tango. Curr. Pharm. Des.. 2007;13:3576-3583.

Wolf J.K., Dwayne Jenkins A. Gene therapy for ovarian cancer (review). Int. J. Oncol.. 2002;21:461-468. (Very readable review of ovarian cancer and basic concepts in gene therapy, coupled with a round-up of data on compounds in clinical trial)

Useful Web resources

http://www.cancer.org/ (The US equivalent of the Web site below. The best sections for you are those marked Health Information Seekers and Professionals)

http://www.cancerresearchuk.org (The Web site of Cancer Research UK, the largest cancer charity in the UK. Contains valuable data on the epidemiology and treatment of cancer, including links to clinical trials. An excellent resource)

1The term cytotoxic drug applies to any drug that can damage or kill cells. In practice, it is used more restrictively to refer to drugs that inhibit cell division and are therefore potentially useful in cancer chemotherapy.

2You will have gathered that many anticancer drugs are toxic. ‘To be an oncologist,’ one practitioner quipped, ‘one has to hate cancer more than one loves life.’

3As opposed to the ‘polyclonal’ antibodies produced by the body in response to a foreign antigen, which comprise a complex (and variable) molecular species.

4The nomenclature can be confusing: by convention the suffix ‘-mab’ denotes a ‘monoclonal antibody’; ‘-momab’ a mouse; ‘-ximab’ a chimeric; ‘-zumab’ a humanised; and ‘-umab’ a human antibody.