The Chemotherapy Of Cancer







9.4.1 Excessive cell proliferation 335

9.4.2 Loss of tissue-specific characteristics 335

9.4.3 Invasiveness 335

9.4.4 Metastasis 336


9.5.1 Internal factors 336 Mutation 336 Addition or loss of genetic material 336 Changed gene expression 336

9.5.2 External factors 336 Viruses 336 Chemicals 337 Radiation 338

9.5.3 Hereditary factors 338


9.6.1 Surgery 338

9.6.2 Radiation therapy (Radiotherapy) 339

9.6.3 Photodynamic therapy 339

9.6.4 Immunotherapy and vaccines 339

9.6.5 Chemotherapy 340 Achievements of chemotherapy 340 Discovery of drugs and their preclinical evaluation 340 Accessibility of tumour cells to drugs 340 Achieving selective toxicity 341 Limiting the toxicity of chemotherapeutic agents 341 Overview of the mode of action of chemotherapeutic agents 342 Drug resistance 343 Combination chemotherapy 343 Adjuvant chemotherapy 344


9.7.1 DHFR Inhibitors (antifolates) 344 Methotrexate (Maxtrex®) 344

9.7.2 Purine antimetabolites 346

9.7.3 Pyrimidine antimetabolites 347

9.8 DNA INTERACTIVE COMPOUNDS 348 9.8.1 Cross-linking agents 348 Nitrogen mustards 348 Aziridines 352 Methanesulphonates 352 Triazeneimidazoles 353 Imidazotetrazinones 354 Nitrosoureas 354 Metal complexes 356 Carbinolamines 356 Cyclopropanes 357 Procarbazine 358

9.8.2 Intercalating agents 359 Anthracyclines 359 Anthracenes 360 Phenoxazines 361

9.8.3 Topoisomerase inhibitors 362 Ellipticine 362 Camptothecin 362 Etoposide 363

9.8.4 DNA Cleaving agents 363 The Bleomycins 363 The Enediynes 365


9.9.1 Vinca alkaloids 365

9.9.2 The Taxanes 366


9.10.1 Hydroxyurea 367

9.10.2 Mithramycin

9.10.3 Mitotane


9.11.1 Azathioprine

9.11.2 Cyclosporin (Neoral®)


9.12.1 Breast cancer Anti-oestrogens Aromatase inhibitors

9.12.2 Prostatic cancer Oestrogen therapy LHRH analogues Anti-androgens



9.14.1 Interferon alpha

9.14.2 Tumour necrosis factor

9.14.3 Interleukin

9.14.4 Growth factors


9.15.1 Bioreductive prodrugs

9.15.2 Estramustine

9.15.4 Antibody—drug conjugates

9.15.5 ADEPT

9.15.6 GDEPT


9.17 FUTURE POSSIBILITIES 9.17.1 Gene targeting Antisense oligonucleotides Ribozymes

9.17.2 Oncogene—product inhibitors

9.17.3 Gene therapy

9.17.5 Resistance inhibitors

9.17.6 DNA—Repair inhibitors

9.15.3 Photoactivated prodrugs (Photodynamic

Therapy, PDT) 378

381 381 Antigene (Macromolecules and small molecules) 381

382 382 382 382

9.17.4 Growth factor and signalling pathway modulators 383

9.17.7 Telomerase inhibitors

9.17.8 Antimetastic agents

9.17.9 Blood-flow modifying agents

9.17.10 Vaccines

9.17.11 Chemopreventative agents ("Neutriceuticals")




Cancer is a disease in which the control of growth is lost in one or more cells leading to a solid mass of cells known as a tumour. A growing (primary) tumour will often become life-threatening by obstructing vessels and/or organs. However, death is most often caused by spread of the primary tumour to numerous other sites in the body which makes surgical intervention impossible. Other types of cancers such as leukaemia involve a build-up of large numbers of cells in the blood.

In the first three decades of this century cancer accounted for less than 10% of all UK deaths, infectious diseases being the main cause of mortality. Whilst dramatic progress has been made in controlling infections, similar progress has not been made with the treatment of cancer. Improved diet, living conditions and health care have increased the average life-span to the point where cancer, which is a disease of advanced years (70% of new cases of cancer in the UK occur in those over 60), has become more prevalent. Consequently, about 300,000 new cases of cancer occur each year in the UK. The annual number of deaths from cancer of all types is approximately 160,000 which constitutes approximately 25% of all UK deaths. Statistics show that approximately one in three of the population will suffer from some form of cancer during their lives and one in four will die from the disease. Furthermore, 1 in 10,000 children will be diagnosed annually as suffering from cancer, which means that there are 1300 new cases each year.

It is thought that exposure to an ever increasing number of chemicals (carcinogens) in both the environment and the diet may be significant. Occupational factors are thought to account for 6% of cancers, while life-style and diet may account for up to 30%. Genetic predisposition is also a factor in some types of the disease.


A tumour or neoplasm is an abnormal tissue mass or growth which results from neoplasia, a state in which the control mechanisms governing cell growth are deficient, leading to cell proliferation. Cancers are generally named according to the type of tissue in which they arise; for example, sarcoma describes those neoplasms occurring in mesodermal tissue which includes connective tissue, bone and muscle. Osteosarcoma refers to bone cancer, and tumours of the epithelial tissues such as the mucous membranes and glands (including cancers of the breast, ovary and lung) are referred to as carcinomas.

Cancers of the blood or haemopoietic tissue are generally known as blastomas. These can involve lymphoid, erythroid or myeloid cells which generally fall into the sarcoma category. Leukaemias describe those cancers which originate in leucocytes, and may be myeloid, lymphatic or monocytic; in addition, these particular cancer types may be chronic or acute. Bone marrow cell tumours are referred to as myelomas, and in multiple myeloma (the most common bone marrow cancer) a clone of plasma cells is involved. Neoplasia of erythroid stem cells is known as primary polycythaemia.

The reticuloendothelial system is also susceptible to tumourogenesis. Lymphosarcoma is cancer of the lymphoid cells, whereas Hodgkin's disease is an example of a lymph adenoma which, although it mainly affects reticulum cells, can extend to eosinophils, fibroblasts and lymphocytes.


Metastasis is the ability of solid tumours to spread to new sites and establish secondary cancerous growths. Tumour cells may easily penetrate the walls of lymphatic vessels and distribute to draining lymph nodes. Cancer cells may also invade blood vessels directly, since capillaries have weak thin walls which offer little resistance. A tumour may also spread across body cavities from one organ to another; e.g. stomach to ovary. Most patients who die of cancer do so as a consequence of metastasis to vital organs.

At the point of clinical recognition of cancer, curative surgical or radiological treatment is only possible if metastasis of the primary tumour has not occurred. Therefore, early diagnosis is essential. Since about 50% of malignant tumours have metastasised prior to diagnosis, the condition is often beyond the reach of curative surgery or radiotherapy alone. It is in these cases that systemic chemotherapy can often help to reduce the total tumour mass.


Cancer is a disease of cells characterised by a reduction or loss of effectiveness in the normal cellular control and maturation mechanisms that regulate multiplication. A schematic diagram of the cell cycle is shown below:

( -1'i.'l I i.i Ijr Cinïipûntrtli an: iyrtlhfaiHbd and [he DNA generale

S - This phase involves ffeneiic iiiiviiv to product punîtes ira J pyrimidines fi, Cell prolifenti enzyme and DNA

synthesis occurs over a varying time

There are four main criteria common to all cancers.

9.4.1 Excessive cell proliferation

This usually results in the formation of a tumour. Normal adult tissues do not grow but maintain a steady number of cells. In some tissues, e.g. liver, this is achieved without proliferation because there is little cell loss. In the bone marrow, however, a steady number of cells are maintained by a fast rate of cell division balanced by a fast rate of cell loss. Often it requires only a slow increase in the rate of proliferation to gradually outgrow normal controlled cellular populations.

In the early stages of tumour growth cancer cells often resemble the original cells from which they are derived. Later, tumour cells lose the appearance and function of these tissues.

This is the ability to grow into adjacent tissue. The tumour not only expands in size but also infiltrates surrounding tissue. When nerve-endings are affected pain is experienced.

One of the major obstacles to successful cancer treatment is the ability of cancer cells to move around the body (metastasise) and develop into new tumours elsewhere.

9.4.2 Loss of tissue-specific characteristics

9.4.3 Invasiveness

9.4.4 Metastasis


It is now accepted that cancer is a "genetic" disease resulting from changes to the sequence information (and its expression) in specific genes. A number of ways in which these changes can be brought about by either internal, external or hereditary factors are described below.

9.5.1 Internal factors

Tumour formation may result from changes to the genetic information brought about by malfunction of the normal DNA processing systems within the cell. Mutation

Genetic mutations can take several forms. In a "point" mutation, only one base is altered, and the new codon that results will cause insertion of a different amino acid into that particular position of the protein. Should the protein be a growth factor, then tumourogenesis could result. In a "translocation" mutation, an entire DNA sequence is moved from one part of a gene or chromosome to another. Again, the loss of the proteins corresponding to the two original DNA sequences or the presence of the new protein may lead to tumourogenesis. The original genes are known as "proto-oncogenes"; that is, genes which will not cause cancer unless suitably activated (i.e. by translocation to form an "oncogene"). The proto-oncogene/oncogene theory has been proved in the case of Burkitt's Lymphoma and Chronic Myelogenic Leukaemia (CML) in which the precise sequences involved in the translocations have been identified. Addition or-loss of genetic material

During normal DNA handling processes such as repair, DNA bases may be inadvertently added or deleted. This can have a similar effect to point mutations in altering codons and ultimately modifying the structure of growth factors and control proteins. Changed gene expression

A problem with gene expression may occur, such as uncontrolled expression or amplification. Should growth factors or proteins that are responsible for receptor formation be involved, then tumourogenesis may result.

9.5.2 External factors Viruses

A link between viruses and cancer has been recognised since 1911, when Peyton Rous demonstrated that avian spindle cell carcinoma could be transmitted from one bird to another by a cell-free filtrate containing the virus (which now bears Rous' name—the Rous Sarcoma Virus). Since then, other viruses have been identified which are linked to human cancers.

Viruses may be either RNA retroviruses such as Human-T-Cell Leukaemia Virus (HTLV-1) or DNA viruses. It is believed that RNA viruses contain DNA-polymerases which facilitate the production of double-stranded viral DNA. On being incorporated into the host DNA, the viral DNA may cause transformation to a cancerous state via a number of different mechanisms including production of an oncogene from an existing gene, damage to a tumour suppresser gene or insertion of a completely new gene. For example, HTLV-1 introduces a gene known as tax which results in the overexpression of interleukin-2. This can lead to adult T-cell lymphomas and leukaemias with an increase in the number of activated lymphocytes, although these may take years to develop in susceptible individuals. hTlV-1 is endemic in South East Asia and the Caribbean; in the Far East it is also associated with nasopharyngeal cancers.

Epstein-Barr Virus (EBV) infects 90% of the world's population and is considered to be an "initiator" of cancer as opposed to a specific cause. For example, 90% of Burkitt's lymphoma cells test positive for EBV which allows infected lymphocytes to become immortal, leading to a potentially cancerous state. Burkitt's lymphoma is endemic in those parts of Africa with chronic malaria suggesting that the latter may be a co-factor in lymphoma development.

Hepatocellular carcinoma has been linked with Hepatitis B Virus (HBV), and is endemic in Southeast Asia and tropical Africa. The risk of tumour formation is greatest in those who are infected from an early age, and males are four times more likely to develop the cancer than females. It is believed that the X-gene in HBV codes for proteins which promote transcription.

There are fifty different types of Human Papilloma Virus (HPV), and HPVs 16 and 18 have been linked to cervical cancer. The virus produces several proteins, some of which enhance mitosis while others interfere with P53 (a tumour suppresser gene) or modify the interaction between cellular proteins and transcription factors. Chemicals

There is now convincing evidence that certain chemicals in the environment, encountered through the diet, lifestyle or occupation may be responsible for some cancers. The link between cigarette smoke and lung cancer is now well-established, and it is also known that carcinogenic polycyclic aromatic hydocarbons (PAHs) are formed in overcooked fried or barbecued meat. Carcinogenic amines are formed in the stomach as a result of the bacterial degradation of nitrites used as preservatives in meat and fish, and the potent carcinogenic aflatoxins are found in peanut butter as a result of the fungal infection of peanuts during growth. Occupation-associated cancer is not a feature of the nineteenth century alone as, in 1775, Sir Percival Pott noted the high frequency of scrotal skin cancer amongst young chimney sweeps. Infrequent washing meant that the tarry deposits produced by burning coal were in contact with the skin for long periods of time. More recently, vinyl chloride used by workers in the plastics industry has been associated with angiosarcoma of the liver. Furthermore, employees in the furniture industry have been prone to develop naso-pharyngeal malignancies brought about by the inhalation of particulate matter carrying organic compounds during leather and wood polishing processes. Most of these organic carcinogens are thought to work by covalently modifying DNA (either before or after metabolism). In addition to these organic carcinogens, certain dusts and minerals are known to cause cancer; for example, the link between asbestos and pleural and peritoneal tumours is well established. Radiation

Malignancies have been linked with exposure to a and p particles or X-rays which are known to damage DNA by fragmentation through the formation of free radicals. A link between nuclear fall-out from atomic bombs and cancer was firmly established after the Hiroshima bombing in World War II. It has also been postulated that children living close to nuclear power stations are at a higher risk of contracting leukaemias and brain tumours. The danger of the escape of radioactive materials from nuclear reactors was highlighted by the Chernobyl incident in Russia which caused widespread contamination of the food chain. It is also known that a build-up of radon gas produced by certain types of granite can endanger the occupants of houses built from this material. Radon is a naturally occurring radioactive gas that, once inhaled, enters the bloodstream and delivers radiation to all tissues; bone marrow is particularly sensitive and so leukaemias predominate. In the UK, local councils have been obliged to offer grants to affected householders so that buildings can be structurally modified to improve ventilation. Very recently it has been proposed that electromagnetic radiation from overhead power lines may be associated with childhood leukaemias, and that microwaves produced by mobile telephones held to the ear may cause brain tumours. In the former case, it has been suggested that rather than the electromagnetic radiation itself causing cancer, the high electric fields generated by power lines may concentrate radioactive radon gas into local pockets. Recently, the UK Government has been sufficiently concerned about these suggestions to fund a detailed investigation.

9.5.3 Hereditary factors

A number of genes have now been identified that, if inherited, can predispose individuals to certain types of cancer. For example, two genes, BRCA1 and BRCA2, have recently been identified and sequenced by UK and US researchers. These genes are inherited and are associated with breast cancer. Other genes associated with colon and bowel tumours are known to be inherited. This has lead to the introduction of diagnostic screening with subsequent genetic counselling for affected individuals.


Cancer treatment often encompasses more than one approach, and the strategy adopted is largely dependent on the nature of the cancer and how far advanced it is.

9.6.1 Surgery

The surgical removal of tumours is confined to those considered to be solid (for example breast, lung or colon tumours) as opposed to the leukaemias. If a tumour is small and reasonably well defined it can usually be removed by surgery. However, there is often additional treatment with chemotherapy or radiotherapy to try and eliminate any cells that may have either remained behind or metastasised. Increasingly, radiotherapy or chemotherapy may be administered prior to surgery, in order to shrink the tumour and facilitate its removal.

9.6.2 Radiation therapy (Radiotherapy)

Radiation therapy utilises X-rays or radiopharmaceuticals (radionuclides) which act as sources of y-rays. X-Rays are delivered locally in a highly focused beam to avoid damage to healthy tissue, and there is still intensive research into the most effective treatment regimes in terms of the duration and frequency of exposure. The main radionuclides in use include cobalt-60, gold-198 and iodine-131. Gold-198 concentrates in the liver, and iodine-131 is used to treat thyroid cancers as iodine accumulates in this gland. A significant proportion of tumour cells are hypoxic (lack oxygen) and are thus less sensitive to damage by irradiation. Therefore, prior to and during radiation therapy, oxygen is often given to try and sensitise the cells. Radiosensitising drugs such as metronidazole have also been administered prior to treatment to try and improve the therapeutic outcome of radiation therapy. A process known as high linear energy transfer (HLET) has also been applied to the irradiation of hypoxic cells. In this procedure, the tumour is irradiated with neutrons (heavier than X or y-rays) which decay to a-particles, the latter causing cell damage in an oxygen-independent manner.

9.6.3 Photodynamic therapy

Photodynamic therapy (PDT) is a relatively new form of cancer treatment that is currently being evaluated in clinical trials. PDT involves the sytemic administration of a photosensitiser such as the haematoporphyrin derivative Photophrin® which is found to be selectively retained by malignant cells. A few days after administration of this agent, the tumour is irradiated with an intense light source of an appropriate wavelength which excites the photosensitiser. Upon decay to its ground state, available oxygen is transformed into singlet oxygen which is highly cytotoxic and damages the tumour. There is also evidence that, by damaging endothelial cells, PDT restricts tumour blood flow. Laser light sources are now well-developed, and the use of flexible optical fibres means that tumours in inaccessible parts of the body including the GI tract can be easily reached through key-hole surgery techniques and endoscopy; ovarian and stomach tumours have been treated experimentally in this way. Other less expensive non-laser light sources have been recently marketed and new types of photosensitisers are being developed. The use of PDT is therefore likely to escalate in the future.

9.6.4 Immunotherapy and vaccines

The aim of immunotherapy is to stimulate the body's natural response to fighting the cancer. Several neoplasms, including some types of breast cancer, have been found to possess specific tumour antigens, and this has led to the development of monoclonal antibodies specific for some tumour types. Although little success has so far been achieved by treating patients with antibodies alone, research is still ongoing into the development of vaccines that may either prevent tumour formation or modify the growth of established tumours. In the latter case, there has been recent publicity over the use of a vaccine to prolong the life of melanoma patients. Tumour-specific antibodies have also been used for drug targeting by attaching them to either drugs or enzymes (e.g. ADEPT), and these strategies are discussed later.

9.6.5 Chemotherapy Achievements of chemotherapy

Despite the limitations discussed below, cancer chemotherapy has made remarkable progress. Nitrogen mustards were the first agents to be clinically used, although their predecessors were initially used in warfare. Cisplatin, which was also discovered serendipitously, has provided a major advance in the treatment of testicular and ovarian carcinomas. In the latter case, the cure rate is very high for patients with early diagnosis. Taxol is a more-recently discovered natural product that is showing great promise in the chemotherapy of lung and refractory ovarian cancers. Similarly, the synthetic agent temozolomide is proving to be highly effective for melanomas and brain tumours (particularly those occurring in children). Discovery of drugs and their preclinical evaluation

Most clinically used anticancer drugs were discovered either through chance (e.g. cisplatin and the nitrogen mustards) or through screening programmes (e.g. vinblastine and taxol). Only recently, since a more detailed knowledge has been acquired of the fundamental biochemical differences between normal and tumour cells, has rational drug design become possible (e.g. Marimastat®). A combination of the power of screening techniques and rational drug design has recently been realised with the widespread trend towards the use of combinatorial chemistry to rapidly generate large numbers of molecules of diverse structure for screening.

New agents are nearly always evaluated initially in in vitro tumour cell lines. Unfortunately, this only measures the cytotoxicity of an agent and provides no indication of whether it is likely to have useful in vivo antitumour activity. It can, however, indicate whether the agent has selective cytotoxicity towards a particular tumour type which may suggest which in vivo experiments to carry out. Human tumour xenografts represent the most successful animal model in which human tumour fragments are transplanted into rodents. However, even these models do not necessarily correspond to equivalent tumours in humans, as numerous differences exist including the integrity of the blood supply to the transplanted tumour and general biochemical species differences. For example, a toxicity not detected in rodents may be serious enough in man to prevent clinical use of an agent. Such toxicities are often a consequence of interspecies variations in metabolism. Despite these problems, most drugs in clinical use today (with the exception of hormonal agents) were introduced as a result of activity demonstrated in animal models. It is worth noting that, sometimes, drugs that are active in humans show no effect in animal models. For example, hexamethylmelamine [2,4,6-tris(dimethylamino)-1,3,5-triazine] is the lead compound for the "melamine" class of antitumour agents and is metabolised to a carbinolamine species in man with activity against carcinomas of the bronchus, ovary and breast. However, it exhibits only minimal activity when tested in rodent models, as this species fails to carry out the crucial metabolic step. Accessibility of tumour cells to drugs

The accessibility of anticancer drugs to tumour cells varies. Whilst leukaemia cells are fully exposed to drugs in the blood stream, solid tumours have a less-reliable blood supply. Small tumours are usually reasonably well-supplied and are more susceptible to drug action than large tumours which frequently have poor capillary access, particularly in their centres which can be hypoxic. The degree of accessibility of the chemotherapeutic agent therefore explains the greater sensitivity of small primary tumours and early metastases to chemotherapy and highlights the importance of early diagnosis and treatment. It is noteworthy that brain tumours are particularly resistant to chemotherapy as few drugs are capable of crossing the blood brain barrier. Achieving selective toxicity

The development of more-effective chemotherapeutic agents is critically dependent upon the discovery of exploitable biochemical differences between normal and tumour cells. Such differences should allow a more rational approach to drug design rather than relying on the empirical manner in which many of the present day drugs have evolved. Examples of such a fundamental difference are presently limited, but perhaps the best known is the discovery that some lymphoid malignancies are dependent on an exogenous supply of asparagine whereas healthy cells can synthesise their own; this led to the clinically-useful agent asparaginase (see Section 9.13). A more recent example is the development of agents such as Marimastat® that target metalloproteinase enzymes crucial for the process of metastasis. There is hope that complete selectivity may one day be achieved through gene targeting in which genetic differences between tumour cells and healthy cells are exploited (see Section 9.17.1). Indeed, there is speculation that some of the DNA-binding agents such as the mustards that bond covalently to GC sequences of DNA may exploit the fact that some oncogenes are particularly GC-rich. However, a more common view at present is that most clinically-useful anticancer drugs (including the mustards) are generally cytotoxic but are more damaging to faster growing cells. Although it is true that certain types of cancer cells grow faster than other tissues in the body, in the majority of tumours the rate of cell division is still slower than that of normal bone-marrow, skin epithelium or the mucosa of the mouth. This explains the consistent pattern of side effects accompanying chemotherapy which are dose-limiting in practice. Limiting the toxicity of chemotherapeutic agents

It has proved possible to limit bone marrow toxicity by exploiting cell "kinetic" differences between normal and tumour stem cells. Stem cells constitute the smallest, yet most important, compartment in a proliferating system. They are capable of an indefinite number of divisions and are responsible for maintaining the integrity and survival of a cell population. Thus, the increase in size of a lymphoma cell population compared with a normal stem cell population can be explained by the fact that only 20% of the bone marrow stem cells are usually in active cycle at any one time, the remaining 80% being in the resting phase (G0). The dividing marrow stem cell population may be significantly reduced in size by chemotherapy, however within 3-4 days the remaining stem cells can move from G0 into active cycle. This means that very high doses of drugs may be given for 24-36 hour periods interspersed with adequate recovery intervals.

Studies of the cell kinetic patterns of tumour growth have suggested a classification for cytotoxic agents based on their ability to reduce the stem cell population of normal bone-marrow and lymphoma cells in mice. Two classes of antitumour agents are distinguished:

Class Cell-cycle-specific agents which kill cells in only one phase of the

1: cycle; e.g. S-Phase (the period of DNA synthesis): 6-

mercaptopurine, cytosine arabinoside and methotrexate; or M phase (mitosis): vinblastine and vincristine. An increased dose of these drugs will not kill more bone marrow stem cells than killed by the initial dose.

Class Non-cell-cycle-specific agents that kill cells at all phases of the cell

2: cycle; e.g. cyclophosphamide, melphalan, chlorambucil, cisplatin, BCNU, CCNU, 5-fluorouracil, actinomycin D and daunorubicin. An increased dose of these drugs will increase the number of bone marrow stem cells killed. Although cell killing can occur in all phases, it is possible that some agents are more active in a given phase of the cycle.

With a few exceptions, such as the cumulative toxicities associated with adriamycin (cardiac), bleomycin (pulmonary) and cisplatin (renal), common toxicities of anticancer drugs are usually reversible within 2-3 weeks. Mucositis (associated with actinomycin D, adriamycin, bleomycin, methotrexate, 5-fluorouracil and daunorubicin) is reversible over a period of 5-10 days, reflecting the rapid recovery of normal tissues. Nausea and vomiting, which accompany the administration of certain drugs, may be partially overcome with modern anti-emetic agents such as the 5HT3 antagonists (see section 9.18). Overview of the mode of action of chemotherapeutic agents

Different mechanisms of action include interference with metabolism (e.g. methotrexate), interruption of cell division (e.g. the effect of vinblastine and taxol on the spindle), steroid receptor antagonism (e.g. effect of tamoxifen on the oestrogen receptor) or the inhibition of steroid biosynthesis (e.g. aromatase inhibitors). The recently developed Marimastat® works by targeting metalloproteinease enzymes involved with metastasis. A large number of clinically-useful antitumour agents have DNA as their target. Some agents block the synthesis of DNA (e.g. 6-mercaptopurine), whilst others act by becoming incorporated into DNA and then interfering with its function (e.g. 6-thioguanine). However, the majority of clinically-useful anticancer agents interact directly with DNA. As the primary genetic material, DNA has many desirable characteristics as a drug target. It is an active participant in a wide range of biological processes and disruption is likely to have a detrimental effect on cell growth. Furthermore, drugs targeted to nucleic acids act at the earliest possible stage of gene expression and should be highly efficient on a molar basis.

DNA-interactive drugs exert their effect by a number of different mechanisms. DNA cross-linking may occur on the same strand (intrastrand cross-linking) or between DNA strands (interstrand cross-linking). Interstrand cross-linking is observed with alkylating agents such as the nitrogen mustards; cisplatin is an example of an agent that causes intrastrand cross-links. An important feature of intrastrand cross-linking is that it bends the DNA at the adduct site. Intercalation is a type of non-covalent interaction in which planar molecules insert between the stacked base-pairs of DNA (e.g. mitoxantrone). The intercalating moiety is inserted perpendicular to the DNA helix, causing the bases to separate vertically. This lengthens the DNA and distorts the sugar-phosphate backbone. Unwinding of the helix at the intercalation site disrupts the action of enzymes such as RNA polymerases and the topoisomerases which control DNA supercoiling.

A further mechanism involves groove-binding in which the agent interacts either covalently or non-covalently in the minor or major grooves. The two grooves of DNA differ in electrostatic potential, hydration and hydrogen bonding characteristics, as well as in width and depth. Most DNA-binding control proteins are thought to interact in the major groove as it is more information-rich than the minor groove. Experimental agents such as the lexitropsins and other netropsin/distamycin analogues bind non-covalently in the minor groove, whereas analogues of CC-1065 and the pyrrolobenzodiazepines bind covalently. The minor-groove binding distamycin-mustard conjugate, tallimustine, is presently in clinical trials. Most of the mustards and the recently marketed temozolomide (see Section act in the major groove. Finally, some agents can interact with DNA (by one of the above mechanisms) and then cause strand scission (strand breakage) at the binding site usually through the production of free radicals (e.g. bleomycin and doxorubicin). As in the case of DNA cleavage by ionising radiation, the reaction is usually oxygen dependent and the sugars or bases of the sugar-phosphate backbone are attacked. If the breaks occur in close proximity but on opposite strands then double-strand breakage occurs. Alkylation of DNA bases or of the phosphate backbone can also result in strand scission, but this is oxygen independent.

The development of drug resistance may seriously interfere with treatment. Initial selective cytotoxicity towards a tumour can sometimes be followed by a rapid recovery in tumour growth, and this resistance may increase after each administration. Drug resistance has been observed in a number of drug-sensitive tumour types including breast, choriocarcinoma and lymphoblastic leukaemias. It has been demonstrated that in some resistant tumour cells there is an altered expression of particular proteins; for example, glutathione transferase production may increase. This enzyme is believed to be responsible for resistance to alkylating agents by catalysing their covalent interaction with glutathione rather than DNA. Another mechanism of resistance involves the active transport of anticancer drugs out of cells. The discovery of the multidrug resistance gene (MDR) has led to an understanding of how tumours can become resistant to a wide variety of agents, and future therapies may even include strategies to down-regulate the MDR gene as a means to enhance the effectiveness of existing anticancer drugs.

This problem of resistance has been overcome in several tumour types by using a combination of different cytotoxic agents as described below.

Most attempts at treating tumours with single agents have been disappointing. A single drug kills the population of cells that is most sensitive to it and leaves a resistant fraction unharmed and still dividing. This led, in 1960, to the first use of a combination of drugs for treating testicular tumours. The "cocktail" principle was then rapidly extended to other tumour types. Each drug included in a particular combination should be active as a single agent but have different toxic (dose-limiting) side-effects from the others. Multiple drug therapy also enables the simultaneous attack of different biological targets thus enhancing the effectiveness of the treatment. The successful application of this technique to the treatment of acute lymphoblastic leukaemia is illustrated below with the comparative response rates of single agents and various combinations: Drug resistance Combination chemotherapy


% Complete Remission

Methotrexate (M, see Section Mercaptopurine (MP, see Section 9.7.2) Prednisone (P)

Vincristine (V, see Section 9.9.1) Daunorubicin (D, see Section P, V

22 27 63 57 38 94 94

A variety of combination schedules using different drugs and dosage regimes are now available and, in many cases, are accepted as superior to single drug therapy. Furthermore, many cancers may already be disseminated at the time of clinical detection, and so combination chemotherapy may be commenced concurrently with local treatment (e.g. radiotherapy, surgery) to maximise benefit for the patient. The micrometastases associated with the primary tumour are often very sensitive to chemotherapy since they have a good blood supply that facilitates drug access. They are also less likely to develop drug resistance than an older tumour. Adjuvant chemotherapy

In treating cancer, it is sometimes necessary to co-administer other agents that may enhance the activity of the anticancer drug or counteract any side-effects produced. Common side-effects include nausea, vomiting and other gastrointestinal disturbances, hair loss and myelosuppression. To counteract the gastrointestinal problems, anti-emetics will usually be administered (see Section 9.18). Myelosuppression is more problematic in that it can lead to an increased risk of infection, and so antibiotic and/or antifungal therapy may be required.


Antimetabolites function by blocking crucial metabolic pathways essential to cell growth. Selectivity is based on the concept that some cancer cells can be faster growing than many normal cell populations with the exception of cells such as those in the bone marrow or parts of the GI tract. While this may be true in the case of leukaemias, older solid tumours often have a very small fraction of cells in active growth. All antimetabolite agents in current clinical use, including the antifolates and the purine and pyrimidine antimetabolites, interfere with DNA synthesis.

9.7.1 DHFR inhibitors (Antifolates) Methotrexate (Maxtrex®)

Tetrahydrofolic acid is produced by the action of the enzyme dihydrofolate reductase (DHFR) on dihydrofolic acid (9.1), and is required for the synthesis of thymine which becomes incorporated into DNA. Slight modification of the structure of folic acid produced the lead antimetabolite aminopterin; methotrexate (9.2), which was shown to be more selective, followed in the 1950s. Methotrexate binds more strongly to the active site on DHFR than the natural substrate by a factor of 104 due to the presence of an amino rather than a hydroxyl moiety which increases the basic strength of the pyrimidine ring.

The most basic centre in the methotrexate molecule is at the N1 and adjacent C2-NH2 position as confirmed by 13C NMR measurements at C2. Examination of the drug-enzyme complex by X-ray diffraction has shown that the pyrimidine ring is situated in a lipophilic cavity with the cation of N1/C2-NH2 binding to an aspartate-26 anion of the enzyme. Other binding points revealed by X-ray include hydrogen bonding between C4-NH2 and the carbonyl groups of both Leu-4 and Ala-97, and ionic interactions between the a-COOH of the glutamate residue and the basic side chain of Arg-57. The p-aminobenzoyl

DiMttoJcMc Mmac

tetrahydrofolic and

(9.1): cti hydro folic acid



No reaction

(9.2); methotrexate residue lies in a pocket formed on one side by the lipophilic side chains of Leu-27 and Phe-30, and, on the other side, by Phe-49, Pro-50 and Leu-54. A neighbouring pocket lined by Leu-4, Ala-6, Leu-27, Phe-30 and Ala-97 accommodates the pteridine ring. The nicotinamide (NADPH) portion of the fully extended co-enzyme lies sufficiently close to the pteridine ring to facilitate transfer of a hydride anion from the pyridine nucleus to the C6-position.

These studies reveal that methotrexate occupies the reverse position at the active site of the enzyme compared to the substrate. Although the p-aminobenzoate and glutamate portions of both are identically bound, with dihydrofolate N-1 is unbound, C2-NH2 and C4-OH bind only to water molecules, N3 is hydrogen bonded to Asp-26, N5 is unbound and N8 interacts with Leu-4 via van der Waals forces. This results in the substrate being comparatively loosely bound, a surprising consequence of the differences in position and strength of the most basic centres in the substrate and inhibitor molecules.

In lymphoblastic leukaemia, methotrexate produces a remarkable remission of symptoms for about a year, however the cells eventually develop resistance by increasing the production of DHFR. Choriocarcinoma, a fast growing tumour of pregnancy with a previously high death rate, is rapidly and completely cured with methotrexate. Dramatic cures have also been obtained in Burkitt's lymphoma which is a highly malignant carcinoma of the lymph glands. Disadvantages of the use of methotrexate include its adverse effect on the production of red blood cells which leads to macrocytic anaemia. Gastrointestinal ulceration and potential damage to the kidneys and liver also require careful monitoring. In high dose intermittent schedules the adverse effect on bone marrow can be relieved by the periodic administration of the calcium salt of N5-formyltetrahydrofolic acid (9.3, folinic acid, leucovorin) which enables blockade of tetrahydrofolic acid production to be bypassed (folinic acid "rescue" therapy).

Many derivatives of methotrexate have been synthesised with a view to reducing its toxicity. The phenyl substituted 3',5'-dichloro derivative is significantly less toxic, perhaps due to its ease of metabolism to 3',5'-dichloro-7-hydroxymethotrexate which is equally effective at inhibiting DHFR. Analogues containing a fluorine atom have also been



(9.3); leucovorin synthesised so that their interaction with the DHFR enzyme may be studied by F-NMR both in vitro and, more recently, in vivo.

9.7.2 Purine antimetabolites

Purine analogues inhibit a later stage in DNA synthesis than DHFR inhibitors. Their major problem is a lack of selective toxicity, since purines are involved in many cellular processes apart from nucleic acid synthesis.

6-Mercaptopurine (9.4, 6-MP) is well-established in the treatment of childhood leukaemias, especially chronic myelocytic leukaemia where the remission rate is about 50%. The free base form is converted by sensitive tumour cells to the ribonucleotide 6-mercatopurin-9-yl (MPRP) which results from interaction of the base with 5-phosphoribosyl transferase. Resistance to 6-MP usually arises due to loss of this enzyme within the tumour.

Although MPRP inhibits several enzymatic pathways in the biosynthesis of purine nucleotides including the conversion of inosine-5'-phosphate to adenosine-5'-phosphate, the main inhibitory action appears to occur at an earlier stage when 5'-phosophoribosylpyrophosphate is converted into phosphoribosylamine by phosphoribosylpyrophosphate amido-transferase. Allopurinol may be used as an adjuvant therapy to inhibit xanthine oxidase mediated degradation of 6-MP to thiouric acid which may cause renal damage.

Another cytotoxic drug used for treating myeloblastic leukaemia, 6-thioguanine (9.5), is also metabolised to the 9-(1'-ribosyl-5'-phosphate) by tumour cells. However, in contrast to MPRP, this does not inhibit an enzyme but is further phosphorylated to the triphosphate and then incorporated into DNA as a "false" nucleic acid. The lack of selectivity towards tumour cells is due to rapid incorporation of 6-thioguanine into the DNA of bone marrow cells. Another analogue, 8-azaguanine exerts its anticancer action in a similar way.

(9.4); G-mercaptopuririe (Puri-nethoJ®) (9.5); 6-thioguanine (Lartvis®)

9.7.3 Pyrimidine antimetabolites

Cytarabine (9.6, ARA-C) and 5-fluorouracil (9.7, 5-FU) are two of the best known pyrimidine antimetabolites. Cytarabine is one of the most effective single agents available for treating myeloblastic leukaemia, achieving a remission rate of 25%. When combined with 6-thioguanine or the cytotoxic antibiotic daunorubicin, remission rates are increased to 50%. Other combinations that include vincristine have led to claims of 70% remission, and some long term survivals of myeloblastic leukaemia have been reported.

A disadvantage of cytarabine therapy arises from its rapid hepatic deamination by cytosine deaminase to give an inactive uracil derivative. This short half-life is counteracted by continuous infusion methods of administration. The rapid deamination has led to the quest for pyrimidine nucleoside deaminase inhibitors which might be co-administered, although this approach has not yet met with success. Deaminase-resistant O-acyl derivatives have also been prepared which can be hydrolysed by esterases. The agent fluoro-2',5'-anhydrocytosine arabinoside is a novel cyclonucleotide that is hydrolysed non-enzymatically and has significant antitumour activity against stomach and pancreatic adenocarcinomas.

Of the many halogeno-pyrimidines investigated, only fluoro derivatives have any appreciable antitumour activity. 5-Fluorouracil (5-FU) is highly effective as a 5% cream in treating skin cancer, and has also found clinical use in the palliative treatment of certain solid tumours such as those of the GI tract, breast, and pancreas. 5-FU is initially metabolised to the 2'-deoxyribonucleotide, 5-fluoro-2'-deoxyuridylic acid (FUdRP), which is a potent inhibitor of thymidylate synthetase. The latter causes the transfer of a methyl group from the co-enzyme methylenetetrahydrofolic acid to deoxyuridylic acid which is converted to thymidylic acid and incorporated into DNA. 5-FU has been shown to have an affinity for thymidylate synthetase several thousand times greater than that of the natural substrate. This remarkable effect is associated with the unique properties of the fluorine atom whose van der Waals radius compares favourably with that of hydrogen although the bond strength is considerably greater. Additionally, the high electronegativity of fluorine affects the electron distribution, conferring a lower pKa on the molecule compared to uracil. These two features combine to enable FUdRP to fit the active site of the enzyme extremely well although the fluorine cannot be removed thus effectively inhibiting the enzyme. Further studies have suggested that a nucleophilic sulphydryl group at the active site forms a covalent bond to FUdRP leading to a 'dead end' adduct of the enzyme, co-enzyme and 5-FU. Structure-activity studies



(9.6); cytarabine (Alexan®) {9.7); 5-fluorouraciJ (Efudix®)

have shown that the increased size but lower electronegativity of other halogen atoms reduce activity. It has been postulated that the high selectivity of 5-FU, especially in skin treatments, may reflect the fact that certain types of cancer cells lack the relevant enzymes to degrade it.


A large number of clinically-useful anticancer drugs exert their effect by interacting with DNA. Some agents intercalate between the base pairs of DNA, whereas others alkylate in either the minor or major grooves. Other types cross-link the DNA strands together in either an "intra" or "interstrand" fashion. As with other classes of anticancer drugs, the selective toxicity towards cancer cells may arise solely from the difference in growth rate of populations of cancer cells compared to normal cells, which also explains their toxicity toward bone marrow and cells of the GI tract. However, there is speculation that some DNA-interactive agents may selectively target certain DNA sequences (for example GC-rich regions) which may inhibit the growth of cancer cells. The agents below are categorised according to their mechanism of action.

9.8.1 Cross-linking agents Nitrogen mustards

The mustards were developed further as chemical warfare agents during World War II as an advance over sulphur mustard gas [S(CH2CH2Cl)2] that made its debut in World War I. Clinicians observed that the leukocyte count dropped in victims surviving sulphur mustard exposure. However, whilst some leukaemias proved responsive, any benefit was greatly outweighed by the toxic effect of the compound. The bioisosteric nitrogen mustards, which had an improved therapeutic index, were introduced into the clinic by Goodman in 1946. The first aliphatic example was mechlorethamine hydrochloride (9.8, mustine) which was found to effectively depress the white blood cell count and was used for treating certain leukaemias.

The high chemical reactivity of the nitrogen mustards and their vulnerability to attack by a wide range of nucleophilic centres accounts for the observed toxicities. Under physiological conditions, the aliphatic nitrogen mustards undergo initial internal cyclisation through elimination of chloride to form cyclic aziridinium (ethyleneiminium) ions. This cyclisation involves intramolecular catalysis through neighbouring group participation and results in a positive charge on the nitrogen which is delocalised over the two adjacent carbon atoms. This cation, although relatively stable in aqueous biological fluids, is highly strained and reacts readily with any nucleophile. The clinically-useful antitumour activity is thought to result from attack of the N7-atom of a guanine residue in the major groove of DNA. The process is then repeated with the second p(2)-chloroethyl group and a second guanine N7-atom located on the opposite strand. This gives rise to an interstrand cross-link that effectively locks the two strands of DNA together.



Get The Body Of Your Dreams

Get The Body Of Your Dreams

Everybody wants to lose weight. This is one fact that is supported by the countless weight loss programs on the market along with the numerous weight loss products, ranging from snack bars, powdered juices, shakes and even slimming soaps and lotions.

Get My Free Ebook

Post a comment