The Search for Cancer Cures Is Difficult but Not Hopeless

The difficulty of curing a cancer is similar to the difficulty of getting rid of weeds. Cancer cells can be removed surgically or destroyed with toxic chemicals or radiation; but it is hard to eradicate every single one of them. Surgery can rarely ferret out every metastasis, and treatments that kill cancer cells are generally toxic to normal cells as well. If even a few cancerous cells remain, they can proliferate to produce a resurgence of the disease; and, unlike the normal cells, they often evolve resistance to the poisons used against them. In spite of the difficulties, effective cures using anticancer drugs (alone or in combination with other treatments) have already been found for some formerly highly lethal cancers—notably Hodgkin's lymphoma, testicular cancer, choriocarcinoma, and some leukemias and other cancers of childhood. Even for types of cancer where a cure at present seems beyond our reach, there are treatments that will prolong life or at least relieve distress. But what prospect is there of doing better, and finding cures for the most common forms of cancer?

Current Therapies Exploit the Loss of Cell-Cycle Control and the Genetic Instability of Cancer Cells

Anticancer therapies need to take advantage of some property of cancer cells that distinguishes them from normal cells. One such property is the genetic instability that results from loss of chromosome maintenance or DNA repair mechanisms. Remarkably, it seems that most existing cancer therapies work because, unknown to the people who developed them, they exploit these molecular defects. Traditional anticancer therapies mostly rely on agents—drugs and ionizing radiation—that damage DNA and the machinery that maintains chromosomal integrity. Such treatments preferentially kill certain kinds of cancer cells because these mutants have a diminished ability to survive the damage. Normal cells, when treated with radiation, for example, will suffer damage to their DNA, but will then arrest their cell cycle until they have repaired it (Figure 23-43). Tumor cells that have defects in various cell-cycle checkpoints, on the other hand, lose the ability to arrest the cell cycle in these circumstances, and so continue to multiply immediately after irradiation. Almost all of these cells will therefore die after a few days as a result of the catastrophic DNA damage they sustain when they attempt to divide with defective chromosomes.

Figure 23-43

Effects of ionizing radiation on normal cells (A) and cancer cells (B). Cancer cells tend to be more susceptible than normal cells to the damaging effects of ionizing radiation because they lack an ability to arrest the cell cycle and make the necessary (more...)

Cancers Can Evolve Resistance to Therapies

Unfortunately, while some of the molecular defects present in cancer cells may enhance their sensitivity to such cytotoxic agents, others may increase their resistance. For example, much of the cell death induced by DNA damage occurs through apoptosis, so cancer cells that harbor defects in their cell death programs can sometimes escape the effects of cytotoxic therapy. Cancer cells vary widely in their response to radiation and to the different kinds of cytotoxic drugs, and it seems likely that this difference reflects the particular kinds of defects they have in DNA repair, cell-cycle checkpoints, and apoptosis pathways.

Genetic instability itself can be both good and bad for anticancer therapy. Although it seems to provide an Achilles' heel that many conventional therapies exploit, genetic instability can also make eradicating cancer more difficult. Because of the abnormally high mutability of many cancer cells, most malignant tumor cell populations are heterogeneous in many respects, which may make them difficult to target with a single type of treatment. Moreover, this mutability allows many cancers to evolve resistance to therapeutic drugs at an alarming rate.

To make matters worse, cells that are exposed to one anticancer drug often develop a resistance not only to that drug, but also to other drugs to which they have never been exposed. This phenomenon of multidrug resistance is frequently correlated with amplification of a part of the genome that contains a gene called Mdr1. This gene codes for a plasma-membrane-bound transport ATPase (belonging to the ABC transporter superfamily discussed in Chapter 11). The overproduction of this protein or some other members of the same family can prevent the intracellular accumulation of certain lipophilic drugs by pumping them out of the cell. The amplification of other types of genes can likewise give the cancer cell a selective advantage: thus the gene for the enzyme dihydrofolate reductase (DHFR) often becomes amplified in response to cancer chemotherapy with the folic-acid antagonist methotrexate.

New Therapies May Emerge From Our Knowledge of Cancer Biology

Our growing understanding of cancer cell biology and tumor progression is gradually leading to better methods for treating the disease, and not only by targeting defects in cell cycle arrest and DNA repair processes. As an example, estrogen antagonists (such as tamoxifen) and drugs that block estrogen synthesis are now widely used in patients to prevent or delay recurrence of breast cancer (and they are even being tested as agents to prevent new cancers from arising). Such antiestrogen compounds do not directly kill off the tumor cells. Nevertheless, they improve the patient's prospect of survival, presumably because estrogens are necessary for the growth of normal mammary epithelium and a proportion of breast cancers retain this hormone dependence.

The greatest hopes lie, however, in finding more powerful and selective ways to directly exterminate cancer cells. Now that we can pinpoint their genetic lesions, can we use our knowledge of cell biology to kill them off? In recent years, a wide variety of adventurous new ways to attack tumor cells have been suggested, many of which have been shown to work in model systems—typically reducing or preventing tumor growth in mice. Many of these protocols will turn out to be of no medical use, because they do not work in humans, have bad side effects, or are simply too difficult to implement. But some seem likely to succeed. For example, some tumor cells are heavily dependent on a particular protein that they overproduce (although it may not be unique to them). Blocking the activity of this protein may be an effective means of treating cancer if it does not unduly damage normal tissues. For example, about 25% of breast tumors express unusually high levels of the Her2 protein, a receptor tyrosine kinase, related to the EGF receptor, that normally plays a part in the development of the mammary epithelium. Thus, shutting off Her2 function might be expected to slow or halt the growth of breast tumors in humans; in fact, this approach is currently being tested with some success in clinical trials, using as the blocking agent a monoclonal antibody that recognizes Her2.

Another approach to destroying tumors targets the delivery of a toxic compound directly to the cancer cells by exploiting proteins like Her2 that are abundant on their surface. Antibodies against such proteins can be armed with a toxin, or made to carry an enzyme that cleaves a harmless ‘prodrug’ into a toxic molecule. In the latter case, one molecule of enzyme can then generate a large number of toxic molecules. A virtue of this strategy is that the toxic drug generated enzymatically can then diffuse to neighboring tumor cells, increasing the odds that they too will be killed, even if the antibody did not bind to them directly.

Treatments Can Be Designed to Attack Cells That Lack p53

These treatments all target properties or molecules that tumor cells possess—but what about molecules that they lack? One ingenious way to target tumor cells exploits their loss of p53. Certain viruses, including the papillomaviruses and the adenoviruses, encode proteins that bind to and inactivate the host cell's p53 (see Figure 23-35). This enables these viruses to outwit the p53-mediated defenses of the host cell and replicate their own genomes freely inside it. As part of their lytic life style, adenoviruses replicate continuously inside an undefended cell, and then burst out when their numbers are sufficient, killing the cell and infecting its neighbors. An adenovirus has been constructed that lacks the gene that encodes the p53-blocking protein; this defective virus can therefore only replicate in cells in which p53 is already inactivated—including many types of cancer cells. If this modified adenovirus is injected into a tumor, the virus might be expected to replicate in and kill only the cancer cells that lack p53, leaving normal cells unharmed. This strategy is also undergoing clinical trials.

Tumor Growth Can Be Choked by Depriving the Cancer Cells of Their Blood Supply

Another promising approach to destroying tumors does not directly target cancer cells at all. Because tumors require the formation of new blood vessels to grow to more than a millimeter or so in size, treatments that block angiogenesis should block tumor growth in many different types of cancer. As discussed in Chapter 22, growth of new vessels requires local signals—angiogenic growth factors—and the action of these molecules can, in principle, be blocked. Clinical trials with angiogenesis inhibitors are now taking place. Endothelial cells that are in the process of forming new vessels also turn out to express distinctive cell-surface markers, providing a promising way in which they might be attacked without harming the existing blood vessels in non-cancerous tissues.

Small Molecules Can Be Designed to Target Specific Oncogenic Proteins

The new treatments outlined above are still at an experimental stage. For most of them, it remains to be seen how effective they are in curing human cancers or in slowing their progress; past experience has taught us to be cautious. There is at least one recent dramatic success, however, that raises high hopes.

As we saw earlier, chronic myelogenous leukemia is associated, in more than 95% of cases, with a particular chromosomal translocation, visible as a characteristic abnormality in the karyotype—the Philadelphia chromosome (see Figure 23-5). This is the consequence of chromosome breakage and rejoining at the sites of two specific genes, called Abl and Bcr. The fusion of these genes creates a hybrid gene that codes for a chimeric protein, consisting of the N-terminal fragment of Bcr fused to the C-terminal portion of Abl (Figure 23-44). Abl is a protein tyrosine kinase involved in cell signaling. The substitution of the Bcr fragment for its normal N terminus makes it hyperactive, so that it stimulates inappropriate proliferation of hemopoietic precursor cells that contain it and inhibits these cells from dying by apoptosis, which many of them would normally do. Excessive numbers of white blood cells are consequently made and released into the bloodstream, producing leukemia.

Figure 23-44

The conversion of the Abl proto-oncogene into an oncogene in patients with chronic myelogenous leukemia. The chromosome translocation responsible joins the Bcr gene on chromosome 22 to the Abl gene from chromosome 9, thereby generating a Philadelphia (more...)

The chimeric Bcr-Abl protein is an obvious target for therapeutic attack. Searches for synthetic drug molecules that can inhibit the activity of protein kinases discovered one, called STI-571, that blocks Bcr-Abl (Figure 23-45). When this molecule, now renamed Gleevec, was given to a series of 54 patients with chronic myeloid leukemia that had resisted other treatments, all but one of them showed an excellent response, with return of their white blood cell counts to normal and, in some cases, an apparent eradication of the cells carrying the Philadelphia chromosome. After a year of treatment, 51 out of the initial 54 patients were still well. Results were not so good for patients who had already progressed through further mutations to the acute phase of myeloid leukemia, where genetic instability has set in and the march of the disease is far more rapid. These patients showed a response at first and then relapsed: the cancer cells were able to evolve a resistance to the Gleevec. Nevertheless, the extraordinary success of Gleevec for the patients in the chronic (early) stage of the disease is enough to prove the principle: once we understand precisely what genetic lesions have occurred in a cancer, we can begin to design effective rational methods to treat it.

Figure 23-45

How Gleevec (STI-571) blocks the activity of Bcr-Abl protein and halts chronic myeloid leukemia. (A) The chemical structure of Gleevec. The drug can be given by mouth; it has side effects but they are usually quite tolerable. (B) The structure of the (more...)

Understanding of Cancer Biology Leads Toward Rational, Tailored Medical Treatments

All medical progress depends on accurate diagnosis. If one cannot identify a disease correctly, one cannot discover its causes, predict its outcome, select the appropriate treatment for a given patient, or conduct trials on a population of patients to judge whether a proposed treatment is effective. Cancers, as we have seen, are an extraordinarily heterogeneous collection of diseases. Nevertheless, new techniques provide tools to make diagnosis precise and specific. We have the means now to characterize each individual tumor at a molecular level in unprecedented detail. For example, by using DNA microarrays (as described in Chapter 8) to analyze the mRNA present in a tissue, the levels of expression of thousands of genes can be determined simultaneously in a single sample and compared with the levels in normal control tissue. Each case of a given form of cancer, such as breast cancer, has its own gene expression profile, but when the profiles of many patients are compared it is found that they can be grouped into a smaller number of distinct classes whose members share common features. The different classes of gene expression profiles, reflecting the consequences of different sets of oncogenic mutations, correlate with different prognoses and different responses to therapy. These correlations are just beginning to be discovered, interpreted, and acted upon.

With our greatly increased understanding of molecular genetic mechanisms, we can for example, aim to determine for each tumor the precise defects in DNA metabolism—the alterations in DNA replication, DNA recombination, DNA repair, chromosome maintenance, and/or checkpoint controls—that have presumably in most cases helped its cells to acquire the multiple mutations required for tumor growth and invasiveness. These defects should make the tumor unusually vulnerable to particular types of attack on its DNA and its DNA-handling machinery. The observation that the cells of certain tumors are killed unusually easily by irradiation or by exposure to drugs that damage DNA supports this view. By designing precisely targeted drugs that exploit a particular weakness more precisely than these traditional treatments, we should be able to attack the cancer cells more effectively. In such ways, the molecular analysis of cancer promises to transform cancer treatment by enabling us to tailor therapy much more accurately to the individual patient.

The discovery of a range of cancer-critical genes has marked the end of an era of groping in the dark for clues to the molecular basis of cancer. It has been encouraging to find that there are, after all, some general principles and that some key genetic abnormalities are shared by many forms of the disease. But we are still far from fully understanding the most common human cancers. We know the DNA sequences of many cancer-critical genes, and the physiological functions of an increasing number of them. It is beginning to be possible to devise precisely targeted, rational treatments. But we still need a better understanding of how the relevant molecules interact to govern the behavior of the individual cell, a better understanding of the sociology of cells in tissues, and a better understanding of the many processes that govern the genesis and spread of cancer cells through mutation and natural selection.

Looking back on the history of cell biology and contemplating the speed of recent progress, we can be hopeful. The desire to understand that drives basic research will surely reveal new ways to use our knowledge of the cell for humanitarian goals, not only in relation to cancer, but also with regard to infectious disease, mental illness, agriculture, and other areas that we can scarcely foresee.

Summary

Our growing understanding of cancer cell biology should lead to better ways of diagnosing and treating this disease. Anticancer therapies can be designed to destroy cancer cells preferentially by exploiting the properties that distinguish them from normal cells, including the defects they harbor in their DNA repair mechanisms, cell-cycle checkpoints, and apoptosis pathways. Tumors can also be attacked through their dependence on their blood supply. By understanding the normal control mechanisms and exactly how they are subverted in specific cancers, it becomes possible to devise drugs to target cancers more precisely. As we become better able to determine which genes are amplified, which are deleted, and which are mutated in the cells of any given tumor, we can begin to tailor treatments more accurately to each individual patient.