Balancing Risk: The Faustian Dilemma of Cancer Chemotherapy

by Ian Magrath

Photomicrograph of the "Elegant" nematode worm, Caenorhab-ditis elegans, which provided a model for the understanding of apoptotic processes in embryonic development (see panel on page 3), and led to the discovery of the first "death proteins" involved in apoptotic pathways. Picture published with the kind permission of James McCarter and Tim Schedl, Washington University.

Es ist so schwer, den falschen Weg zu meiden, Es liegt in ihr so viel verborgnes Gift. —Goethe

Cancer chemotherapy, as patients and those who care for them are only too well aware, can be a double-edged sword. This is perhaps not surprising for a treatment modality whose existence owes much to a substance developed for belligerent rather than benevolent purposes—mustard gas. Initially used in World War I, it was considered to be an advance over chlorine as a weapon of mass destruction because of its devastating effects on exposed body parts, and its ability to penetrate the protective clothing and gas masks available in 1917. Although it was soon observed that non-fatal victims rapidly developed low blood cell counts, it was an event in World War II that gave major impetus to the notion that a substance developed as a weapon might be used to combat human disease.

In 1943, American munitions ships and tankers berthed in Bari Harbor in southeastern Italy were bombed. Sixteen were sunk. Hundreds of oil-soaked men were rescued from the water and were soon observed to be suffering from unexpected symptoms. Eighty-three died. The cause of their demise was traced to contamination of the oily water with nitrogen mustard—one of the ships, the Liberty, had been carrying 100 tons of mustard munitions. Low blood counts were again observed among the exposed military personnel and autopsies in 53 of the victims revealed marked involution of lymphoid tissue. Studies at Yale University showed that this effect could be reproduced in mice and in 1946 Goodman and colleagues reported beneficial effects on lymphoid neoplasms in people. Nitrogen mustard is still used in the treatment of Hodgkin's lymphoma, although medical staff who handle or administer the drug must not only ensure that it is injected cleanly into a vein, with no leakage into subcutaneous tissues, but must also take steps to avoid inhalation of its vapors and to prevent it from coming into contact with their own skin. Although many chemotherapeutic agents have been developed since World War II, the vast majority of these, while generally not injurious to tissues on direct contact, have the potential to damage or destroy normal cells as well as cancer cells. They are, in effect, poisons whose dosage and administration must be carefully controlled if susceptible cancers are to be eradicated without causing irreparable harm to patients. Such potentially harmful agents must only be given under expert supervision.

Shedding Cancer Cells

Vinca rosea, the plant from which a frequently used chemotherapy agent, vincristine, is derived. Many drugs used in cancer chemo-therapy today are natural products.

Cancer cells are malignant by virtue of their capacity for unlimited expansion and their potential to spread to parts of the body beyond their sites of origin. Most conventional anti-cancer drugs are particularly toxic to dividing cells, and in all cancers, a fraction of the cell population must be capable of undergoing replication, since otherwise the cancer could not grow. In general, the higher the replicating fraction, the greater the likelihood that the cancer will be curable by chemotherapy (at least, by those drugs that primarily affect dividing cells). But many normal cell populations in the body are in a dynamic state of equilibrium between the production of new cells and the death of old or spent cells, such that, like cancer cells, some are also continuously dividing. Cell populations in which the turnover rate is high, such as blood-forming cells and lymphocytes (as observed in the victims of exposure to poison gas), cells of the hair follicles, and the lining cells of the mouth and gastrointestinal tract are particularly susceptible to anti-cancer drugs. Perhaps the most common side effect observed with the majority of drugs in use today is the increased risk of potentially life-threatening infections as a consequence of diminished numbers of white cells, which are responsible for protecting against infection and which, given their high turnover rates (sometimes higher than those of tumor cells), are particularly susceptible to chemotherapy. The reason that conventional anti-cancer drugs predominantly affect dividing cells is that the majority of them interfere either with DNA replication (duplicating the genetic material is an essential first step in cell division), or with the physical process of the separation of daughter cells. Cells die when something goes wrong with this process because of the existence of a "quality control" mechanism, whereby irreparable damage to the genetic material, or other disturbances which might lead to imperfect replicas of the parent cells, switch on a molecular pathway that induces suicide through the activation of enzymes that literally digest the cells' component parts. This process, which is active (i.e., consumes energy) and, unlike other kinds of cell death does not excite inflammation, is known as programmed cell death or apoptosis. The latter is a Greek word which refers to any process of shedding, such as the falling of leaves from trees in autumn (which itself involves apoptosis). The word apoptosis was suggested by a Greek scholar, James Cormack, of Aberdeen University, to the discoverers of programmed cell death, Kerr, Wyllie and Currie, as a fitting term for the "cellular dropout" they had described in 1972.

Poison in the Cure
Drugs that cause damage to replicating cells (and often have other side effects too) can cause severe toxicity, or, as demonstrated on the battle fields of Flanders and in Bari Harbor, even death, if given in too high a dose. The goal of cancer chemotherapy is to give enough of a drug, or a combination of drugs, to eradicate cancer, but not enough to induce severe or irreversible toxicity. Medical oncologists have spent a good part of the second half of the last century learning, through empirical clinical trials, where the dividing line between acceptable and unacceptable toxicity lies. The line is not fixed, and indeed, varies from one patient to another, so that one can only deal in probabilities in the context of populations (cohorts) of patients - i.e., the percentage of patients who will develop toxicity of a given degree with a given dose of drug. Greater risks are warranted when the stakes are high, i.e., when the likelihood of curing the patient is considered low and the disease is one which progresses rapidly.

The therapeutic ratio (the risk-benefit ratio) can be altered in favor of the patient by the combined use of several chemotherapeutic drugs active in the disease, because each will have a somewhat different range of toxic side effects, if often overlapping; several relatively minor side effects are more tolerable than a single major toxicity. Drug combinations are also more efficient therapeutically, since each drug damages the tumor cell in a different way, making it more difficult to survive the cumulative damage or to develop resistance to subsequent treatment (one treatment administration is rarely enough). Resistance arises because of the presence of mutations which, through any of a broad range of mechanisms, negate the effects of the drug. Combination therapy has been remarkably effective in some types of cancer, even when advanced, e.g., in leukemias and lymphomas, childhood cancers, testicular cancers and choriocarcinoma, although in many other cancers, chemotherapy has limited benefit or is able to control the disease temporarily, but not to eradicate it, i.e., there is inherent resistance to available chemotherapeutic drugs—perhaps due to a general resistance to apoptosis.

Balancing Risk
The success of chemotherapy depends upon three main factors. The particular drug combination used, the sensitivity of the cancer itself (which varies according to the extent of disease and the biochemical attributes of individual cells) and various patient characteristics which influence the amount of active drug that finally reaches the cancer cell. Most drugs are chemically modified (metabolized) in the body before the active element is formed and are also converted into inactive elements. The drug and its derivatives (metabolites) are eliminated from the body, each with its own time frame. Inherited genetic factors influence the efficiency of the various enzymes involved in drug activation and detoxification, as well as the ability to repair damage done by the drug to both normal and tumor cells. Both are relevant to the outcome of therapy and the degree of toxicity encountered. Genetic variability in both tumors and patients provides an explanation for the inability to predict outcome in individual patients. Interestingly, some of the inherited characteristics which influence the efficacy of treatment and the degree of toxicity are also relevant to the impact of environmental toxins (carcinogens) that cause or predispose to cancer.

Estimates of the likelihood of therapeutic benefit can be significantly refined by defining risk factors. These are characteristics of the cancer (e.g., its size, degree of spread, or molecular genetic abnormalities) or of the patient (e.g., age, sex, general state of health, or inherited ability to metabolize a drug or repair genetic damage) that are known to affect outcome with a given treatment regimen. Risk factors are of considerable value in determining the most appropriate therapy for the patient, for risk in medicine, as in all walks of life, must be consonant with potential benefit—the greater the likelihood of treatment failure (and, therefore, death) the greater the acceptable risk of significant toxicity. This is the premise on which the principle of risk adaptation of therapy is based—patients with a low risk of dying (i.e., who respond well to a given treatment regimen) can be successfully treated with less intensive, less toxic (and less expensive) therapy while patients at high risk (i.e., much less likely to respond to a given treatment regimen) may appropriately be given more intensive and therefore more toxic therapy.

It is important that therapy is not reduced to sub-optimal doses in low risk patients, or made sufficiently intensive that the likelihood of severe toxicity is unacceptable even in high risk patients. Drug doses or scheduling may have to be modified in individual patients who encounter excessive toxicity. Balancing therapeutic effect with toxic risk is not easy, but in the more readily treatable diseases, e.g., childhood malignant lymphoma, successive clinical trials with increasingly intensive regimens for higher risk patients have resulted in survival rates of approximately 90% in both high- and low-risk patients, although the therapy given to each group differs markedly in intensity. Of course, whether a patient succumbs to a life-threatening toxicity is, to a significant degree, dependent upon the quality of the medical care received, particularly the immediate administration of antibiotics to treat or prevent an infection arising in a patient with a low white blood count. Skilled supportive care is critical to the success of chemotherapy regimens associated with a significant risk of toxicity.

Geographical Variation in Risk
While the principles of chemotherapy outlined above apply to patients anywhere in the world, in developing countries the balance between therapeutic benefit and toxicity may differ greatly from that in countries with greater resources. The response of tumors to chemotherapy in different world regions may vary because of differences in the pattern of genetic abnormalities present in tumors; because of differences in factors that modify drug metabolism or tolerance, including genetic variability in different populations (sometimes enhanced by higher rates of consanguinity); and because of co-morbidities, i.e., the presence of unrelated health problems (e.g., malnutrition, malaria, hepatitis or tuberculosis). Excessive toxicity may dictate modifications in dose, schedule or even discontinuation of individual drugs or the entire treatment program. Treatment may also be compromised because of prohibitive cost, the quality of supportive care, or the ability or willingness to adhere to the planned therapy (on the part of doctor or patient).

In developing countries, cancers are generally much more extensive at the time of presentation because of delays in diagnosis or limited access to appropriate therapeutic facilities, such that a greater proportion of patients fall into a high risk category—requiring, if still potentially curable, more toxic and more expensive therapy. Radiotherapy or surgery alone is rarely curative in patients with disseminated tumors. Thus, the lack of resources creates a vicious cycle—limited resources result in poor access to care and consequently more advanced tumors, which in turn require more resources for their management. The INCTR's mission is, in essence, to help to demonstrate that this vicious cycle can be broken—by ensuring, through professional and public education, and, where appropriate, screening programs, that more patients reach appropriate treatment facilities early in the course of their disease, while at the same time enhancing the capacity to deliver appropriate therapy safely, such that potentially curable patients receive the care they need. Patients for whom cure is not an option should be given palliative care—another area where limited resources lead to suffering which could frequently be easily ameliorated. Prevention, including education about behavior that increases cancer risk and screening high-risk populations for accessible, simply treated pre-malignant lesions (e.g, cervical and oral cancers) will, if successful, help to slow the rise in the burden of cancer resulting from increasing population size, aging, higher tobacco consumption and dietary changes. Clearly, the need for augmentation of resources, and for simultaneous efforts on multiple fronts will require effective, coordinated collaboration with many institutions and organizations.

The Future - Targeting Drugs at the Lesions Responsible for Cancer
Advances in the understanding of the molecular genetic lesions that are the immediate cause of cancer provide promise that the next generation of chemotherapeutic drugs will be much less toxic. These genetic lesions provide an "Achilles heel" which is, in essence, specific to cancer cells, such that drugs targeted at them should be equally specific. Doubtless, combinations of such drugs will be required, since cancer cells contain multiple genetic lesions and are adept at developing drug resistance. Targeted drugs, which are likely to have the relatively low toxicity rates of the order of magnitude seen with present antibiotics, will be particularly valuable in developing countries, where toxicity is less readily managed. While likely to be expensive, savings on the management of side effects, and higher cure rates, will at least partially offset increased cost. This development in combating cancer is analogous to the tendency in human warfare to use targeted "smart weapons" in order to limit damage to civilian populations. Warfare, however, is a tragedy of the human condition, and it is small consolation that weapons developed to kill people have, on occasion, been effectively directed against diseases that kill people.

Developing Worms Elucidate Cancer Pathways The protein components of the molecular pathways leading to apoptosis were first identified in the nematode worm, Caenor-habditis elegans, by Horvitz, after Sulston and Brenner had shown that apoptosis is critical to organ development, and to determining the number of cells (959) in the adult worm. The work of all three was recognized this year by the award of a Nobel prize. The nematode proteins have their counterparts in a wide range of multicellular organisms, including humans, indicating that apoptosis is a fundamental biological process. Indeed, apoptosis is, in essence, a regulator of the numbers of cells in a variety of cell populations, and as such is a critical element not only in sculpting the form (and in the case of the nervous system, the neurological connections) of the developing embryo, but also in the control of numerous physiological processes. In mammals, apoptosis is involved in the shedding of cells from inner lining surfaces, such as those of the gastrointestinal tract and endometrium, as well as in the regulation of the expansion of immunologically competent cells participating in an immune response. Virus-infected cells can be killed through the activation of apoptotic pathways. The induction of programmed cell death in cells in which the genetic material is damaged, or in which normal replication is hindered, is particularly relevant to tumor cells, since genetic abnormalities are the immediate cause of neoplasia. In order for genetically modified cells to survive, one or more of the apoptotic pathways must be inactivated by the genetic abnormalities themselves. The tumor cells are then able to pass the check points of cell division and differentiation where abnormalities are detected and sufficiently damaged cells diverted into an apoptotic pathway. Interestingly, tumor cells sometimes mimic immunologically competent cells and develop the capability of activating apoptotic pathways in normal cells, for example, lymphocytes that would otherwise destroy the tumor cells. There are many paths that lead to apoptosis, and while some may be damaged in tumor cells, others are activated by chemotherapeutic agents to which the tumor is sensitive, or by radiation therapy.