Here is an essay on ‘Chemicals and Cancer’ which states – How various chemical carcinogens can trigger the development of cancer.
Essay # 1. Identifying Chemicals that Cause Cancer:
We seem to be surrounded by a sea of chemical carcinogens. They are found in the air we breathe, the food we eat, the water and beverages we drink, the medications we take, the places where we work, and the homes in which we live. However, this assessment—while technically correct—conveys the misimpression that we are faced with severe hazards everywhere we look and that these dangers cannot be avoided.
In fact, many of the carcinogens we normally encounter are only weakly carcinogenic, and most of the more potent ones can be easily avoided by the general public. So rather than lumping all chemical carcinogens together, we need to consider them as individual molecules and make informed judgments about the dangers posed by each one.
Discovery of Chemical Carcinogens:
The first indication that chemicals might cause cancer came from the observations of doctors who, in their struggle to understand the nature of the disease, asked cancer patients a variety of questions about their backgrounds, experiences, and habits. This allowed physicians to gain some impressions, if not firm evidence, about the possible causes of cancer.
Such an approach led a London doctor, John Hill, to point to chemicals as a probable cause of cancer more than two hundred years ago. In 1761, Hill reported that people who routinely use snuff— a powdered form of tobacco that is inhaled—suffered an abnormally high incidence of nasal cancer, suggesting the existence of one or more cancer-causing chemicals in tobacco.
Several years later Percival Pott, another British physician, reported an unusual prevalence of oozing sores on the scrotums of men coming to his medical practice in London. While a less astute observer might have thought it was just one of the venereal diseases that were widespread at the time, Pott’s close examination of the sores revealed that they were actually a form of skin cancer.
Careful questioning revealed that the men with this condition shared something in common. They had all served as chimney sweepers in their youth. It was common practice at the time to employ young boys to clean chimneys because they fit into narrow spaces more readily than adults.
Pott therefore speculated that chimney soot chemicals had become dissolved in the natural oils of the scrotum, irritating the skin and eventually triggering the development of cancer. These ideas led to the first successful public health campaign for preventing a particular type of cancer- Scrotal cancer was virtually eliminated by promoting the use of protective clothing and regular bathing practices among chimney sweeps.
In the years since these pioneering observations, it has become increasingly evident that certain kinds of chemicals can cause cancer. Unfortunately, the ability of a particular chemical to cause cancer has often become apparent only after large numbers of cancers arise in people exposed to that chemical on a regular basis.
For example, in the early 1900s elevated rates of skin cancer were noted among workers in the coal tar industry, and an increased incidence of bladder cancer was seen in factories that produced aniline dyes. The experience in the aniline dye industry was especially dramatic and led to the discovery of several basic principles of chemical carcinogenesis, as will now be described.
Workers who Developed the First Cancer:
The late 1800s witnessed the birth of a series of new chemical industries that for the first time exposed large numbers of workers to high concentrations of toxic substances. A prominent example involved the industrial production of dyes used to color clothing and other fabrics.
Prior to the mid-1800s, most dyes were natural substances extracted from vegetable or animal sources. An accidental discovery made in 1856 by William Perkin, however, led to the birth of the synthetic dye industry and the first mass exposure of workers to potent carcinogenic chemicals.
Perkin was attempting to synthesize quinine, a drug for treating malaria, by carrying out chemical reactions on substances present in coal tar (a thick, black liquid formed during the distillation of coal). In one experiment, he extracted aniline from coal tar and oxidized it with potassium dichromate.
The result was a dark brown precipitate. Most nineteenth-century chemists would have discarded any such dark masses of material because scientists were generally interested in clear, crystalline products. But Perkin was instinctively curious and decided to investigate further. To his pleasant surprise, dissolving the dark sludge in alcohol yielded an intense purple solution that exhibited strong dying properties. Perkin had discovered aniline purple, the first synthetic dye.
Within a few years, coal tar had yielded several other dyes and the aniline dye industry was born. Chemists quickly discovered that a compound related to aniline called 2-naphthylamine is an ideal starting material for the synthesis of many dyes, and large-scale production began in Germany around 1890. Unfortunately, factory employees working with 2-naphthylamine soon began developing bladder cancer in alarming numbers (Figure 1).
In one small factory, all 15 workers developed the disease. A vigorous and protracted debate ensued as to whether 2-naphthylamine was actually responsible because bladder cancer also occurs among the general public and there was little precedence for using epidemiological data to infer cause and effect. Eventually the cancer-causing ability of 2-naphthylamine was demonstrated in convincing fashion, using both epidemiological and animal data, but it took almost 50 years before the large-scale production of this highly potent carcinogen was stopped.
Bladder cancer triggered by occupational exposure to 2-naphthylamine was the first example of a human cancer known to be caused by a specific chemical compound. The introduction of 2-naphthylamine into the workplace was also a key event because it marked the beginning of a massive increase in industrial chemical production and illustrated some important principles that are widely relevant to chemical carcinogenesis.
One of these principles involves the long delay that is typically observed between exposure to a chemical carcinogen and the onset of cancer. Few cases of bladder cancer were seen in factory workers until 10 years after initial exposure to 2-naphthylamine, and most cases took between 15 and 30 years to appear. Such a long delay is typical of chemical carcinogenesis and reflects the multiple events that take place on the road to developing cancer.
Another principle illustrated by the experience with 2-naphthylamine is dose dependence- Workers who were exposed to the chemical over a longer period of time, and hence had a larger total exposure, exhibited higher bladder cancer rates than workers with shorter exposures.
Figure 1 reveals that almost every worker in the group with the longest exposure to 2-naphthylamine eventually developed bladder cancer. If virtually everyone develops cancer under such conditions, it means that hereditary differences did not play a significant role in determining risk.
Finally, the experience with 2-naphthylamine illustrated the organ specificity that is a common feature of chemical carcinogenesis. Instead of causing cancer in general, chemical carcinogens tend to preferentially cause a few particular types of cancer. In the case of 2-naphthylamine, the bladder is the prime target.
We have already encountered other examples of organ specificity and will encounter additional examples later. Organ specificity is generally caused by the selective ways in which chemicals make contact with, or accumulate in, certain body tissues.
For example, chemicals that become concentrated in the urine are likely to produce bladder cancer, and carcinogens that are inhaled tend to cause lung cancer. The ability of cigarette smoke to cause many kinds of cancer in addition to lung cancer may seem to violate this principle; tobacco smoke, however, contains more than 40 different carcinogens, and some of these accumulate in tissues other than the lung.
Asbestos as a Cause of Cancer Deaths:
The natural mineral asbestos is a particularly striking example of an organ-specific carcinogen. Commercial use of asbestos began in the late 1800s, when large deposits of asbestos rock were discovered in Canada and shipped to the United States and to the newly industrializing countries in Europe. Crushing the rock yields a mixture of fine fibers that can be woven into materials that exhibit excellent insulating and fire-retarding properties.
The most commonly used form of asbestos has the formula (Mg,Fe)3Si2O5(OH)4, but many chemical variations exist. Numerous fireproof products, ranging from oven mittens and fireproof clothing to various kinds of construction materials, have been manufactured with asbestos.
Unfortunately, the widespread use of asbestos has had severe health consequences. Asbestos readily breaks down into a fine dust containing numerous sharp, needle-like fibers that are so tiny that they can only be seen with an electron microscope. These “needles of death” are easily inhaled and become lodged in the lung, where they cause scarring that kills people through suffocation.
Shortly after this disease, called asbestosis, was first recognized among asbestos workers in the 1920s, the same workers began to develop lung cancer. Because cigarette smoking was not yet popular, lung cancer was still rare, and the connection between the cancer outbreak and exposure to asbestos was therefore easy to detect.
Scientists eventually found that asbestos and cigarette smoke interact synergistically in causing lung cancer. As a result, smokers who have been heavily exposed to asbestos exhibit lung cancer rates that are 50 times higher than is observed in people who do not smoke or have significant exposure to asbestos.
An unusual property of asbestos is its ability to cause mesothelioma, a rare form of cancer derived from the mesothelial cells that cover the interior surfaces of the chest and abdominal cavities. This type of cancer was uncommon prior to the 1950s, when the first mesothelioma epidemic was reported in and around a group of asbestos mines in South Africa.
Mesotheliomas were subsequently detected in many locations around the world; virtually everywhere that asbestos is used. An increased risk for mesothelioma is exhibited mainly by asbestos workers and by individuals who experience significant exposure to asbestos either by living in neighborhoods surrounding asbestos factories or by working or living in asbestos- insulated buildings. At present, asbestos is the only clearly established cause of mesothelioma.
Microscopic examination of lung tissue obtained from asbestos workers has revealed that mesothelioma is caused by a rather unusual mechanism. Tiny, microscopic fibers of inhaled asbestos become embedded in the lung and gradually penetrate completely through the lung tissue, emerging into the chest cavity.
Here the asbestos fibers trigger a chronic irritation and inflammation that promotes the development of cancer in the mesothelial cells that cover the lungs and line the interior chest wall. In a similar fashion, asbestos fibers that have been inadvertently ingested can penetrate through the walls of the stomach and intestines, emerging into the abdominal cavity and triggering the development of abdominal mesotheliomas.
When the fatalities caused by asbestos- induced mesotheliomas and lung cancers are combined, asbestos ranks as the second most lethal commercial product (after tobacco) in terms of the number of cancer deaths caused.
Governmental actions to regulate the production and use of asbestos began in earnest in the 1960s and have become progressively more restrictive in many countries, so the incidence of asbestos-induced cancers should eventually begin to decline.
Nonetheless, mesothelioma deaths are still rising (Figure 2) and will probably continue to do so for several decades because a lag period of 30 or more years can intervene between asbestos exposure and developing cancer. Moreover, countries that formerly used asbestos still contain vast reservoirs of the carcinogen in existing buildings.
Workplace Exposure to Chemical Carcinogens as a Cause of Cancer:
As in the case of 2-naphthylamine and asbestos, carcinogenic chemicals are often identified only after a particular type of cancer starts to appear in people exposed to a specific substance in high doses. Once such observations point to a particular chemical as a potential carcinogen, follow-up animal testing is carried out to determine whether the substance really causes cancer.
Beginning around 1900 with 2-naphthylamine in the aniline dye industry, the list of known chemical carcinogens grew progressively as the Industrial Revolution proceeded throughout the twentieth century and unusual cancer patterns began to emerge among workers in various industries. Table 1 lists some of the main occupational carcinogens that were eventually discovered, including examples from the rubber, chemical, plastic, mining, fuel, and dye industries.
Workplace exposure to occupational carcinogens was substantial in the first half of the twentieth century before the cancer risks from such agents came to be fully appreciated. In a few extreme cases, all the workers exposed to the chemicals present in certain factories eventually developed cancer.
However, most of the currently known occupational carcinogens were identified by the 1970s and relatively few new ones have been identified since then. In 1970, an act of the United States Congress created the Occupational Safety and Health Administration (OSHA) to formulate regulations designed to protect the safety and health of workers.
OSHA has worked to eliminate the most dangerous chemicals from the workplace and to limit worker exposure to other chemicals. As a result, many occupationally induced cancers that were once prevalent in the United States have declined in frequency, and workplace exposure to carcinogens now accounts for less than 5% of all cancer deaths.
While a similar pattern is evident in many other industrialized nations, progress has been far from uniform. To illustrate some of the disparities, the use of asbestos in Nordic countries has decreased dramatically in recent decades, falling to a negligible value of 4 grams per person in 1996; in that same year, asbestos use in the former Soviet Union was 600 times higher at a value of 2400 grams per person. In general, exposure to industrial carcinogens is a greater problem in developing countries, where less progress has been made in regulating the workplace use of toxic chemicals.
Environmental Pollution not a Major Source of Cancer Risk:
Cancers arising in the workplace are usually triggered by sustained high-dose exposures to specific chemical carcinogens. Small amounts of the same chemicals are also released inadvertently into the environment, where they contaminate the air we breathe, the water we drink, and the food we eat. It has therefore become fashionable to blame industrial pollution for creating a growing cancer threat of epidemic proportions.
However, there are reasons to question such a far-reaching conclusion. First, cancer risk is related to carcinogen dose, and the doses of industrial carcinogens to which the public is exposed are generally orders of magnitude lower than is encountered in the workplace. For example, consider the pesticide ethylene dibromide (EDB), which is designated by the U.S. government as a probable human carcinogen based on its ability to cause cancer in animals.
Workers who have experienced high-dose exposures to this suspected carcinogen encountered about 150 milligrams (mg) per day, whereas EDB residues in food (before EDB was banned in 1984) exposed the average person to a daily dose of 0.00042 mg, which is 300,000 times less than the workers’ daily intake.
A similar situation exists for most of the other chemical contaminants in our food, air, or water, which do not represent a major cancer threat because they are present in concentrations thousands of times lower than typical industrial exposures.
Another reason to question the assumption that pollution represents a major cancer hazard is based on historical trends. If industrial pollution were a major cancer threat to the general public, one would have expected a significant increase in overall cancer rates during the twentieth century in response to the explosive growth in the use of industrial chemicals.
For example, the yearly production rates of plastics, pesticides, and synthetic rubber in the United States increased more than 100-fold between the 1940s and the 1970s. Any impact the chemicals used in these industries might have had in causing a cancer epidemic through environmental pollution should have been apparent by now.
In fact, when the data are adjusted for the increasing average age of the population, it is clear that a significant growth in age-adjusted cancer rates has not occurred (Figure 3). Most of the cancers that are common today were also common one hundred years ago, and the main exception, lung cancer, is triggered by cigarette smoke and has little to do with industrial pollution.
Risks from Low-Dose Exposures to Chemical Carcinogens as a Cause of Cancer:
The preceding arguments suggest that for most people other than those who receive high-dose exposures by working with or living near a concentrated source of toxic chemicals, the cancer risks posed by environmental chemicals are relatively small. This does not mean, however, that the risk for the average citizen is zero.
If the public were being routinely exposed to low carcinogen doses that cause a tiny fraction of the population to develop cancer, such small effects would be difficult to detect using traditional epidemiological methods. In fact, a weak carcinogen present in the environment could theoretically cause hundreds or even thousands of cancer cases each year in a country of several hundred million people without being noticed.
Assessing the actual risk, if any, from low-dose exposures to known or suspected carcinogens is a difficult task. To illustrate, let us briefly consider the dioxins, a family of chlorinated chemicals produced as a by-product during the burning of municipal wastes, the bleaching of paper, and the production of herbicides. Several epidemiological studies have demonstrated increased rates of cancer in workers exposed to large concentrations of dioxin, and high-dose animal studies have also shown increased cancer rates.
Although the quantities released into the environment are quite small, dioxins are stable molecules that tend to persist for long periods of time, accumulating in the food chain. For humans the main source of exposure is through the food we eat especially fatty meats.
The ingested dioxin molecules are fat soluble and are metabolized slowly, so they tend to accumulate in our body fat. The net result is that even tiny exposures to dioxin can lead to significant concentrations inside the body.
The large unanswered question is whether this accumulated dioxin represents a significant cancer risk. The most conservative approach has been to assume that even one molecule of a carcinogen can cause cancer—in other words, that there is no safe dose of dioxin. However, testing in rats has revealed that while high doses of dioxin cause cancer, low doses can sometimes decrease cancer rates compared to those observed in control animals.
Such data indicate that it is possible for low-dose carcinogen exposures to be safe, and perhaps even beneficial. Unfortunately, differences between humans and rodents in mode of exposure, metabolism, and genetics make it difficult to extrapolate such results to humans exposed to tiny amounts of dioxin. In other words, we don’t really know whether the tiny amount of dioxin that we typically encounter is a small cancer risk, poses no cancer risk, or perhaps even decreases our risk of developing cancer.
Another family of environmental contaminants that might represent a cancer hazard are the organochlorine pesticides, a group that includes the now-banned insecticide dichlorodiphenyltrichloroethane (DDT). Compounds of this type can mimic the action of estrogen, which is known to promote the development of breast cancer.
Animals exposed to organochlorines exhibit increased cancer rates, and several widely quoted epidemiological studies have detected a correlation between exposure to organochlorine pesticides and breast cancer rates in women. However, most epidemiological studies have failed to detect such a relationship. Some especially interesting data emerged from a study of several thousand women in Long Island, an area in which organochlorines were extensively used and in which breast cancer rates are higher than the national average.
To precisely quantify exposure to organochlorine pesticides, blood samples were obtained from breast cancer patients and from a group of comparable women without the disease. Measurements of the concentration of organochlorines in the blood failed to reveal any relationship between exposure to organochlorine compounds and the development of breast cancer.
Pollution of Outdoor and Indoor Air Creates Small Cancer Risks:
The general topic of air pollution provides yet another example of the difficulties encountered when trying to assess environmental cancer hazards involving low-dose exposures. In many cities, both large and small, the air is contaminated with fine particles of airborne soot emitted by cars and trucks, power plants, and factories.
A recent epidemiological study of 500,000 adults living in dozens of cities across the United States revealed that people located in cities with the largest amounts of this fine-particle soot have lung cancer death rates roughly 10% higher than in cities with minimal pollution.
Of course, a 10% increase is not very much; for comparison purposes, cigarette smokers increase their risk of developing lung cancer by 2500% or more, a value that is 250 times higher than the small increase in lung cancer risk that might be associated with air pollution.
Although discussions of air pollution usually focus on outdoor air, our main exposure to polluted air may be indoors. In studies involving more than a dozen different cities, researchers equipped people with air-quality monitoring devices that were small enough to carry around as people performed their normal daily activities.
For the average citizen, the greatest exposure to toxic airborne chemicals turned out to occur inside their homes (Figure 4). The sources of this indoor air pollution included ordinary consumer products such as cleaning compounds, paints, carpeting, gasoline, air fresheners, dry cleaning, and disinfectants.
Even in cities where the outdoor air was polluted by emissions from chemical processing plants, the concentration of many airborne carcinogens was higher inside homes than outdoors. However, these indoor concentrations were still much lower than those typically encountered in industrial workplaces, and it is difficult to know whether such low-dose exposures pose any cancer risks.
Thresholds can Cause Animal Studies to Overestimate Human Cancer Risks:
The difficulty in assessing the hazards of low-dose chemical exposures arises from limitations that are inherent to epidemiological and animal testing. The main problem with the epidemiological approach is that it is not sensitive enough to reliably detect small increases (less than a doubling) in cancer incidence, which is the type of effect that might be expected from low-dose carcinogen exposures. As a consequence, scientists often turn to animal testing.
Animal testing also has shortcomings that limit the ability to assess risks from low-dose carcinogen exposures. One problem is the need to obtain a sufficient number of cancer cases to generate statistically reliable results.
For this reason, animals are often exposed for their entire lifetime to the maximum tolerated dose (MTD) of a suspected carcinogen, which is defined as the highest dose that can be administered without causing serious weight loss or signs of immediate life-threatening toxicity.
At these high doses, many chemicals cause tissue destruction and cell death. The remaining cells proliferate to replace those cells that have been destroyed, and this enhanced cell proliferation creates conditions that are favorable for the development of cancer.
If the ability of a given chemical to cause cancer stems from this capacity to cause cell death at high doses, lower doses that do not kill cells may not cause cancer. The dose- response curve for such a carcinogen would exhibit a threshold—that is, a dose that must be exceeded before cancer rates begin to rise. Doses below the threshold would be safe in terms of cancer risk. Figure 5 shows how the existence of a threshold can cause the cancer risk of low-dose exposures to be overestimated.
The data in Figure 5 are for benzo[a]pyrene, a carcinogen present in gasoline exhaust fumes and in smoke generated by burning organic matter, including tobacco smoke. The graph on the left shows cancer rates for animals exposed to high doses of benzo[a]pyrene. If this data were the only information available, a straight line could be drawn through the data points and extrapolated to lower doses to estimate cancer risk for low-dose exposures to benzo[a]pyrene.
The graph on the right, however, shows what happens when actual experiments are performed using lower doses of benzo[a]pyrene. The shape of the overall curve exhibits a threshold, and the actual cancer risk associated with low-dose benzo[a]pyrene exposure is much lower than predicted.
The preceding example highlights that cancer biologists have traditionally had two ways of viewing the relationship between high-dose and low-dose cancer risks: the linear model and the threshold model. As shown in Figure 6, the linear model assumes a linear dose-response relationship with no threshold, whereas the threshold model assumes no cancer risk at lower doses followed by a linear dose-response relationship at higher doses.
A third possibility, called the hormetic model, has also begun to receive some attention. The hormetic model proposes that dose-response curves can also be U-shaped. The U-shape, known as hormesis, reflects a situation in which cancer rates actually decline at very low doses of carcinogen and then begin to go up as the dose is further increased.
While it is not clear how widely this model might apply to cancer risk, several carcinogenic agents have been reported to reduce cancer rates when administered to animals at low doses.
The hormetic and threshold models both include the concept of a “threshold”—that is, a dose that must be exceeded before cancer rates begin to rise. One possible explanation for the existence of thresholds is the ability of high-dose exposures to cause tissue destruction and cell death, which creates a unique set of conditions that do not exist at lower doses.
Another possible reason for thresholds is that carcinogens often act by causing DNA damage. The presence of damaged DNA triggers a group of repair mechanisms that attempt to correct the problem. Such repair mechanisms might be able to fix small amounts of DNA damage caused by low doses of carcinogens and may even help prevent cancer from arising in response to subsequent exposures to carcinogenic agents.
According to this view, carcinogen-induced mutations only begin to accumulate and initiate the development of cancer after a threshold dose has been exceeded and the capacity of these DNA repair pathways is overcome by massive DNA damage.
Humans and Animals Differ in their Susceptibilities to some Carcinogens:
Another problem that can complicate the extrapolation of animal data to humans is that animals often differ from one another, as well as from humans, in their susceptibility to different carcinogens. Consider the behavior of 2-acetylaminofluorene (AAF), which is a potent carcinogen in rats but does not cause cancer in guinea pigs. Based on this information alone, it would be difficult to predict whether or not AAF is likely to be carcinogenic in humans.
The reason for the differing behavior of AAF in rats and guinea pigs became apparent when it was discovered that AAF is actually a “precarcinogen” that needs to be metabolically activated before it can cause cancer. Rats, but not guinea pigs, contain the enzyme that catalyzes this metabolic activation. Biochemical analysis of human tissues has revealed that we also contain the activating enzyme, so it is likely that AAF is carcinogenic in humans just as it is in rats. Of course, if AAF had only been tested in guinea pigs, it never would have been suspected of being a carcinogen in the first place.
The artificial sweetener saccharin provides another illustration of the difficulties that can arise when extrapolating data from animal studies to humans. At the peak of its use in the 1970s, Americans consumed more than five million pounds of saccharin per year in artificially sweetened foods and beverages.
In 1981, the U.S. government labeled saccharin as a suspected human carcinogen and attempted to ban its use as a food additive because saccharin causes bladder cancer when fed to rats. Subsequent investigations, however, have revealed that saccharin causes bladder cancer in rats for reasons that do not apply to humans.
When rats ingest large amounts of any sodium salt, including sodium saccharin, a crystalline precipitate forms in the bladder that irritates the bladder lining, triggering cell proliferation and increasing the risk of developing cancer. But the precipitate only forms when there are large amounts of protein in the urine, and the urine protein concentration in rats is 100 to 1000 times greater than in humans. Subsequent studies have shown that other laboratory animals, such as hamsters, guinea pigs, and mice, do not develop bladder cancer when fed saccharin. As a consequence, saccharin was recently taken off the government’s list of suspected human carcinogens.
Because of the uncertainties involved in applying animal data to humans, caution is needed in labeling substances as human carcinogens when the information has been derived largely from animal studies.
The U.S. government therefore publishes a list that subdivides carcinogens into two distinct categories:
(1) Known to be human carcinogens and
(2) Reasonably anticipated to be human carcinogens.
The list of “known” human carcinogens contains several dozen chemicals for which the data from animal studies have been supplemented with enough human data to clearly establish a cancer risk for humans. The list of “anticipated” human carcinogens includes more than 100 additional chemicals for which the potential cancer hazard has been extrapolated largely from animal studies.
While many of these substances will almost certainly turn out to be human carcinogens, mistakes are possible because of the heavy reliance on animal data. For example, saccharin was listed as an “anticipated” human carcinogen for about 20 years before eventually being removed from the list.
Medications and Hormones can Cause Cancer:
We have now seen how difficult it can be to measure cancer risks associated with chemicals to which our exposures are small. Of course, this means that the hazards of such low-dose exposures must be rather small (or nonexistent) because larger risks would be readily detectable through epidemiological and animal testing.
The greatest cancer hazards are posed by chemical carcinogens that we encounter in high concentrations. Included in this category are several situations, including occupational exposure to industrial chemicals and inhalation of the carcinogenic chemicals present in tobacco smoke.
Another type of high-dose exposure to specific chemicals comes from the use of prescription drugs for treating certain illnesses. Prescription drugs are often taken for prolonged periods, so it is important to know whether the resulting high-dose chemical exposures can cause cancer.
One tragic example involves diethylstilbestrol (DES), a synthetic estrogen that was prescribed to pregnant women starting in the 1940s as a way of preventing miscarriages. Several decades later, women whose mothers had taken DES during pregnancy began developing vaginal cancer at alarmingly high rates. By that time, roughly five million women in the United States had already taken DES.
This episode illustrates the difficulty in establishing the risks associated with ingesting any new chemical compound, even when it appears to be safe and is prescribed for a specific medical purpose. In the case of DES, the drug’s ability to cause cancer did not become apparent until several decades after women had used DES, and the cancer did not affect the person who took the drug, but rather her children.
Although DES has now been banned as a prescription drug, a number of other medications can also cause cancer (Table 2). Most are prescribed for serious medical problems where the potential benefits of the drug in question are thought to outweigh the risk that cancer might arise.
For example, some drugs used for slowing or stopping tumor growth in cancer patients can themselves trigger development of a new cancer as a side effect. In a person who already has cancer, the small risk of causing another cancer (often many years in the future) is far outweighed by the possible benefits to be gained from a drug that might cure an existing cancer.
A similar situation exists with immunosuppressive drugs, which inhibit immune function and are given to organ transplant patients to prevent rejection of a transplanted organ, such as a heart or kidney. Two of the most commonly used immunosuppressive drugs, azathioprine and cyclosporin, increase the risk of developing cancer, but organ transplant patients depend on the transplanted organ for survival and the benefits of these drugs are thought to outweigh the risk of developing cancer.
Nonetheless, cancer has turned out to be a major cause of death in organ transplant patients and a need therefore exists for better immunosuppressive drugs that do not increase cancer risk. One drug under current investigation is rapamycin (also called sirolimus), an antibiotic with immunosuppressive activity. Animal studies have shown that besides suppressing immune function, rapamycin inhibits tumor growth under conditions in which another immunosuppressive drug, cyclosporin, stimulates tumor growth.
A possible explanation for the antitumor effect of rapamycin has come from the discovery that it inhibits angiogenesis, both by depressing the production of VEGF and by inhibiting the ability of endothelial cells to respond to VEGF. Suppression of angiogenesis by rapamycin may therefore limit the ability of newly forming tumors to obtain the blood supply they require for growth beyond a tiny size.
Essay # 2. Mechanisms of Chemical Carcinogenesis:
As the list of substances known to cause cancer has grown over the years, it has become increasingly apparent that carcinogens exhibit wide variations in structure and potency. At first this variability complicated our thinking about the origins of cancer because it was difficult to envision how such a diverse array of chemical substances could cause the same disease. Through an extensive series of studies, however, a common set of mechanisms and principles has begun to emerge that helps explain how the various kinds of carcinogens work.
Chemical Carcinogens can be Grouped into Several Distinct Categories:
Despite their structural diversity, chemical carcinogens can be grouped into a relatively small number of categories (Figure 7). Most are natural or synthetic organic chemicals—that is, carbon-containing compounds. They range from small organic molecules containing only a few carbon atoms to large, complex molecules constructed from multiple carbon-containing rings.
The vast majority fall into one of the following five categories:
1. Carcinogenic polycyclic aromatic hydrocarbons (or simply polycyclic hydrocarbons) are a diverse group of compounds constructed from multiple, fused benzene rings. Polycyclic hydrocarbons are natural components of coal tars, soots, and oils, and are also produced during the incomplete combustion of coal, oil, tobacco, meat, and just about any other organic material that can be burned.
The carcinogenic potency of polycyclic hydrocarbons varies widely, from weak or noncarcinogenic molecules to very potent carcinogens. The polycyclic hydrocarbons benzo[a]pyrene and dibenz[a,h]anthracene, isolated from coal tar in the 1930s, were the first purified chemical carcinogens of any kind to be identified.
2. Carcinogenic aromatic amines are organic molecules that possess an amino group (—NH2) attached to a carbon backbone containing one or more benzene rings. Some aromatic amines are aminoazo compounds, which means that they contain an azo group (N=N) as well as an amino group. Among the carcinogens in these categories are the aromatic amines benzidine, 2-naphthylamine, 2-acetylaminofluorene, and 4-aminobiphenyl, and the aminoazo dyes 4- dimethylaminoazobenzene and o-aminoazotoluene.
Many of these compounds were once employed in the manufacturing of dyes, although most are no longer used in significant quantities because of their toxicity. Some aromatic amines, such as 2-naphthylamine and 4-aminobiphenyl, are components of tobacco smoke. As in the case of polycyclic hydrocarbons, the carcinogenic potency of aromatic amines and aminoazo dyes varies from substances that are strongly carcinogenic to substances that are not carcinogenic at all.
3. Carcinogenic N-nitroso compounds are organic chemicals that contain a nitroso group (N=O) joined to a nitrogen atom. Members of this group include the nitrosamines and nitrosoureas, which are potent carcinogens when tested in animals. Most of these compounds are industrial or research chemicals encountered mainly in the workplace, although a few are present in cigarette smoke.
Nitrates and nitrites used in the curing of meats, which are not carcinogenic in themselves, can be converted in the stomach into nitrosamines, but no consistent relationship between these compounds and human cancer has been established.
4. Carcinogenic alkylating agents are molecules that readily undergo reactions in which they attach a carbon-containing chemical group to some other molecule. Unlike the three preceding groups of carcinogens, which are defined by their chemical structures (i.e., the presence of multiple benzene rings, amino groups, or nitroso groups), alkylating agents are defined not by their structural features but by their chemical reactivity—that is, their ability to join a chemical group to another molecule. The N-nitroso compounds, discussed in the preceding paragraph, are examples of carcinogens that function as alkylating agents.
Other examples include vinyl chloride (used in the production of plastics) and ethylene oxide (used in the production of antifreeze and other chemicals). Vinyl chloride and ethylene oxide are among the highest-volume chemicals produced in the United States. Other carcinogenic alkylating agents include sulfur mustard (a chemical warfare agent) and several drugs used in cancer chemotherapy.
5. Carcinogenic natural products are a structurally diverse group of cancer-causing molecules produced by biological organisms, mainly microorganisms and plants. Included in this category is aflatoxin, a carcinogenic chemical made by the mold Aspergillus. One of the most potent carcinogens known, aflatoxin sometimes contaminates grains and nuts that have been stored under humid conditions. Other carcinogenic natural products include plant-derived molecules such as safrole, a major component of sassafras root bark, and pyrrolizidine alkaloids, produced by a variety of different plants.
In addition to the preceding five classes of organic molecules, a small number of inorganic substances (compounds without carbon and hydrogen) are carcinogenic. Included in this group are compounds containing the metals cadmium, chromium, and nickel.
Some inorganic substances appear to be carcinogenic in the absence of chemical reactivity. For example, asbestos is a mineral composed of silicon, oxygen, magnesium, and iron, but its ability to cause cancer is related to the crystal structure and size of the microscopic fibers it forms rather than their precise chemical makeup.
Some Carcinogens need to be Activated by Metabolic Reactions Occurring in the Liver:
The chemicals illustrated in Figure 7 are considered to be “carcinogens” because humans or animals develop cancer when exposed to them. This designation does not mean, however, that every carcinogen triggers cancer directly. For example, consider the behavior of 2-naphthylamine, whose ability to cause bladder cancer in industrial workers.
As might be expected, feeding 2-naphthylamine to laboratory animals induces a high incidence of bladder cancer, but cancer rarely arises when 2-naphthylamine is directly inserted into an animal’s bladder.
The reason for this discrepancy is that when 2-naphthylamine is ingested (by animals) or inhaled (by humans), it first passes through the liver and is metabolically converted into chemical compounds that are the actual causes of cancer. Inserting 2-naphthylamine directly into the bladder bypasses the liver and the molecule is never activated, so cancer does not arise.
Many carcinogens share a similar need for metabolic activation before they can cause cancer. Carcinogens exhibiting such behavior are more accurately called precarcinogens, a term referring to any substance that is capable of causing cancer only after it has been metabolically activated. The activation of precarcinogens is generally carried out by liver proteins that are members of the cytochrome P450 enzyme family.
One function of these liver enzymes is to catalyze the oxidation of ingested foreign chemicals, such as drugs and pollutants, with the aim of making molecules less toxic and easier to excrete from the body.
The hydroxylation reaction illustrated in Figure 8 is one of several ways in which cytochrome P450 oxidizes foreign chemicals to make them more water soluble, thereby facilitating their excretion in the urine. Occasionally, however, oxidation reactions catalyzed by cytochrome P450 accidentally convert substances into carcinogens, a phenomenon known as carcinogen activation.
Evidence that cytochrome P450 is involved in carcinogen activation has come from numerous animal studies. One set of experiments involved a mutant strain of mice that produce abnormally large amounts of cytochrome P450 1A1, a form of cytochrome P450 that oxidizes polycyclic hydrocarbons.
As would be expected if cytochrome P450 1A1 were involved in carcinogen activation, cancer rates are elevated in the mutant mice that produce large amounts of this enzyme. Cancer rates can be reduced in these same animals by using inhibitors that block the action of cytochrome P450 1A1.
Elevated amounts of cytochrome P450 1A1 are found in the livers of people who smoke cigarettes, apparently because tobacco smoke stimulates production of the enzyme by the liver. This means that in addition to containing dozens of chemicals that cause cancer, cigarette smoke also induces the production of liver enzymes that make the situation worse by activating carcinogenic activity in chemicals that might not otherwise cause cancer.
About one person in ten inherits a form of cytochrome P450 1A1 that is produced in especially large amounts in response to tobacco smoke. If such a person smokes cigarettes, he or she has an even higher risk of developing lung cancer than other smokers.
The role played by liver enzymes in carcinogen activation explains why chemicals being assayed for mutagenic activity in the Ames test are first incubated with a liver homogenate to mimic any reactions that might take place in the body. The requirement for metabolic activation also helps explain why some chemicals only cause cancer in certain organisms.
For example, 2-acetylaminofluorene (AAF) is carcinogenic in rats but not in guinea pigs because guinea pigs lack the enzyme needed to convert AAF into an active carcinogen. Because of such differences in liver enzymes between organisms, it is important that suspected carcinogens be tested in more than one animal species.
Many Carcinogens are Electrophilic Molecules that React Directly with DNA:
Despite the variations in molecular structure exhibited by the carcinogens illustrated in Figure 7, many share the same property: When metabolized in the liver, they are converted into highly unstable compounds with electron-deficient atoms. Such molecules are said to be electrophilic (“electron-loving”) because they readily react with substances possessing atoms that are rich in electrons.
DNA, RNA, and proteins all have electron-rich atoms, making each a potential target for electrophilic carcinogens. Of the three, DNA is the prime candidate because the Ames test has shown that most carcinogens cause DNA mutations. An experiment designed to determine whether DNA is in fact the direct target of chemical carcinogens is summarized in Figure 9.
In this study, animals were injected with various polycyclic hydrocarbons that differed in carcinogenic potency. Cells were then isolated from the treated animals and measurements were made to determine which intracellular molecules (if any) had become bound to the polycyclic hydrocarbons.
The data revealed a direct relationship between the carcinogenic potency of different polycyclic hydrocarbons and their ability to become covalently linked to DNA; in other words, those polycyclic hydrocarbons that became extensively bound to DNA were the most effective at causing cancer.
Before a polycyclic hydrocarbon can interact with DNA, it must be activated. For example, consider the behavior of benzo[a]pyrene, which is normally a nonreactive, non-mutagenic compound. After entering the body, metabolic reactions catalyzed by cytochrome P450 in the liver convert benzo[a]pyrene into activated derivatives containing an epoxide group (Figure 10).
An epoxide is a three-membered ring containing an oxygen atom covalently bonded to two carbon atoms; these two carbons are electron deficient and therefore tend to react with atoms that are electron rich, such as the amino nitrogen found in the DNA base guanine. Reaction of the epoxide group with guanine causes the benzo[a]pyrene to become covalently bonded to DNA, thereby forming a DNA-carcinogen complex called a DNA adduct. The presence of the bound carcinogen distorts the DNA double helix and thereby causes errors in base sequence (mutations) to arise during DNA replication.
Epoxide formation is also involved in the activation of other classes of chemical carcinogens. For example, aflatoxin and vinyl chloride, which differ significantly from polycyclic hydrocarbons in chemical structure, are both oxidized by cytochrome P450 into epoxides that, like benzo[a]pyrene, react with DNA bases to form DNA adducts (Table 3).
However, the various epoxides do not all react with the same DNA bases. In fact, depending on the carcinogen involved, almost every electron-rich site in the various DNA bases can serve as a target for carcinogen attachment (Figure 11). And epoxides are not the only electrophilic groups that react with DNA.
Some carcinogens are activated by reactions that create other types of electrophilic groups, such as positively charged nitrogen atoms (nitrenium ions) or carbon atoms (carbonium ions), or compounds containing an unpaired electron (free radicals). Like epoxides, these electrophilic groups also attack electron-rich atoms in DNA.
The preceding mechanisms illustrate that, despite their chemical diversity, many carcinogens share the property of being converted into electrophilic molecules that in turn become linked to DNA. This ability to form DNA adducts is one of the best predictors of a molecule’s capacity to cause cancer.
In addition, carcinogens can inflict DNA damage in several other ways; for example, they may generate crosslinks between the two strands of the double helix, create chemical linkages between adjacent bases, hydroxylate or remove individual DNA bases, or cause breaks in one or both DNA strands (Figure 12).
Chemical Carcinogenesis is a Multistep Process:
An attack on DNA by an activated carcinogen is just the first of several steps involved in creating a cancer cell. The idea that cancer arises through a multistep process was first proposed in the early 1940s by Peyton Rous to explain a phenomenon he encountered when studying the ability of coal tar to cause cancer in rabbits.
Rous had observed that repeated application of coal tar to rabbit skin caused tumors to develop, but the tumors disappeared when application of the coal tar was stopped. Subsequent application of an irritant such as turpentine, which does not induce cancer by itself, caused the tumors to reappear.
This pattern suggested to Rous that coal tar and turpentine play two different roles, which he called initiation and promotion. According to his theory, initiation converts normal cells to a precancerous state and promotion then stimulates the precancerous cells to divide and form tumors.
Because coal tar is a mixture of various chemicals, clarification of the initiation/promotion hypothesis required the isolation and study of individual coal tar components. One such chemical is the polycyclic hydrocarbon dimethylbenz[a]anthracene (DMBA). DMBA is a potent carcinogen, but feeding mice a single dose rarely causes tumors to arise. However, if the skin of a mouse fed a single dose of DMBA is later painted with a substance that causes skin irritation, cancer develops in the treated area (Figure 13).
Besides turpentine, the irritant most commonly used for triggering tumor formation in such experiments is croton oil, a substance derived from seeds of the tropical plant Croton tiglium. Croton oil does not cause cancer in the absence of prior exposure to a carcinogen such as DMBA, nor will cancer arise if DMBA is administered after the croton oil.
These observations support the concept that chemical carcinogenesis is a multistep process in which an initiator (in this case, DMBA) first creates an altered, precancerous state and then a promoting agent (in this case, croton oil) stimulates the development of tumors.
The Initiation Stage of Carcinogenesis is Based on DNA Mutation:
A year or more can transpire after feeding animals a single dose of DMBA and yet tumors will still arise if an animal’s skin is then irritated with croton oil. Thus a single DMBA treatment creates a permanently altered, initiated state in cells located throughout the body, and a promoting agent (croton oil) can then act on these altered cells to promote tumor development.
Because the carcinogenic potency of most chemicals correlates with their ability to bind to DNA and cause mutations (see Figure 9), the permanent alteration is thought to be a mutation. Carcinogens that act in this way are said to be genotoxic because they cause gene damage. The ability to cause gene mutations explains how a single exposure to an initiating carcinogen can create a permanent, inheritable change in a cell’s properties.
Referring to carcinogen-induced mutations as “permanent,” however, implies that DNA damage cannot be repaired, which seems to contradict what we know about the existence of DNA repair mechanisms. Such mechanisms are in fact capable of repairing mutations created by initiating carcinogens as long as the damage is repaired in a timely fashion. Once a damaged DNA molecule has been replicated, as occurs each time a cell divides, mutations become very difficult, if not impossible, to repair and the initiated state therefore becomes permanent.
Figure 14 illustrates why this is the case, using the carcinogen methylnitrosourea as an example. Methylnitro- sourea attacks the base guanine (G) in DNA, creating a methylated guanine derivative. If the cell’s DNA is replicated before repair mechanisms correct the defect, the methylated guanine tends to form an incorrect base pair with thymine (T) during DNA replication rather than pairing with its correct partner, cytosine (C).
During the next round of DNA replication, the incorrectly inserted T will form a base pair with its normal complementary base, adenine (A), creating an AT base pair. The DNA molecule now contains an AT base pair where a GC base pair had originally existed. Since DNA repair mechanisms would not recognize anything abnormal about an AT base pair, the error will persist.
The preceding scenario demonstrates an important principle that applies to many mutations- If DNA replication occurs before mutations are repaired, base-pair alterations tend to arise during replication that cannot be subsequently detected as mutations by cellular repair mechanisms. For this reason it is crucial that mutations be repaired swiftly, before subsequent rounds of DNA replication create a permanent mutation.
Tumor Promotion involves a Prolonged Period of Cell Proliferation:
In contrast to initiation, which requires only a single exposure to an initiating carcinogen, promotion is a gradual process that depends on prolonged or repeated exposure to a promoting agent. If the promoting agent is removed during the early stages of tumor formation, tumors stop growing and may even disappear.
How do we explain the ability of promoting agents to trigger an event that is potentially reversible, at least in its early stages? Studies of a wide variety of promoting agents have revealed that their main shared property is the ability to stimulate cell proliferation. When an initiated cell is exposed to a promoting agent, the cell starts dividing and the number of initiated cells increases.
In the early stages of this process, cell proliferation depends on the presence of the promoting agent, and the cells will stop dividing if the agent is removed. As cell division continues, however, natural selection favors those newly forming cells whose proliferation is faster and autonomous, eventually leading to the formation of a malignant tumor whose growth no longer depends on external promoting agents. The time required for promotion contributes to the long delay that often transpires between exposure to an initiating carcinogen and the development of cancer.
The way in which specific promoting agents stimulate cell proliferation was first established for phorbol esters, the class of tumor promoters found in croton oil. In terms of its tumor promoting activity, the most potent phorbol ester is tetradecanoyl phorbol acetate (TPA).
TPA binds to and activates an enzyme called protein kinase C, which plays a key role in one of the cell’s normal pathways for controlling cell proliferation (Figure 15). In the normal operation of this pathway, external signaling molecules bind to cell surface receptors whose activation leads to the production of an intracellular signaling molecule called diacylglycerol (DAG).
DAG in turn activates protein kinase C, which triggers events leading to cell division. Phorbol esters mimic the action of DAG, binding to and activating protein kinase C. Unlike DAG, however, which is converted to inactive forms, phorbol esters continually activate the protein kinase C molecule. Activation of protein kinase C by TPA is a highly selective interaction; small changes in the chemical structure of TPA yield derivatives that exhibit diminished ability to bind to protein kinase C and, as a result, decreased ability to function as tumor promoters (Figure 16).
In addition to phorbol esters, a variety of other foreign substances stimulate cell proliferation and thereby acts as promoting agents. Some of these molecules resemble phorbol esters in being able to interact with protein kinase C. The fungal toxin teleocidin and the algal toxin aplysiatoxin are two such agents that function by activating protein kinase C, even though their chemical structures differ significantly from those of phorbol esters.
Other promoting agents stimulate cell proliferation indirectly, causing tissue damage and cell destruction that make it necessary for the remaining cells of the affected tissue to proliferate to replace the cells that have been damaged and destroyed. Asbestos and alcohol are two previously discussed substances that function in this way.
Not all tumor promoters are foreign substances. Because cell proliferation occurs in normal cells as well as in tumor cells, molecules produced by the body for the purpose of stimulating normal cell division may also function inadvertently as tumor promoters. For example, estrogen is a naturally produced steroid hormone that can contribute to the development of breast and ovarian cancer by stimulating the proliferation of cells in these tissues.
The hormone testosterone stimulates the proliferation of cells in the prostate gland and plays a comparable role in promoting the development of prostate cancer. Of course, the intended function of estrogen and testosterone is to stimulate the growth and division of normal cells, not cancer cells.
But if a breast or prostate cell has acquired an initiating mutation caused by a carcinogen or by an error in DNA replication, any normal hormone or growth factor that stimulates the proliferation of the mutant cell will act inadvertently as a tumor promoter.
In addition to foreign chemicals and natural hormones, certain components of the diet may also act as promoting agents—that is, agents that increase cancer risk by stimulating cell proliferation rather than by creating mutations. Dietary fat and alcohol are two examples that fit this category. In general, any chemical associated with an increased cancer risk that is found not to be genotoxic can be suspected of acting as a promoting agent.
Tumor Progression involves Repeated Cycles of Selection for Rapid Growth and Other Advantageous Properties:
When Rous first proposed that chemical carcinogenesis is a multistep process, he identified only two stages- initiation and promotion. It has gradually become apparent that a third stage, known as tumor progression, follows initiation and promotion (Figure 17). The concept of tumor progression refers to the gradual changes in the properties of proliferating tumor cell populations that occur over time as cells acquire more aberrant traits and become increasingly aggressive.
The underlying explanation for tumor progression is that cells exhibiting traits that confer a selective advantage—for example, increased growth rate, increased invasiveness, ability to survive in the bloodstream, resistance to immune attack, ability to grow in other organs, resistance to drugs, and evasion of death-triggering mechanisms (apoptosis)—will be more successful than cells lacking these traits and will gradually come to predominate.
While it is easy to see why cells exhibiting such traits tend to prevail through natural selection, that does not explain how the aberrant traits originate in the first place. One way of creating new traits is through additional mutations. If a particular mutation causes a cell to divide more rapidly, cells produced by the proliferation of this mutant cell will outgrow their companions and become the predominant cell population in the tumor.
Such a process is called clonal selection because the cells that predominate represent a clone—that is, a population of cells derived from a single initial cell by successive rounds of cell division.
One member of a clonal population may acquire another mutation that makes it grow even faster and the process repeats again, generating an even faster growing clone of cells. Multiple cycles of mutation and selection can occur in succession, each creating a population with enhanced growth rate or some other advantageous property.
Although mutations play a central role in tumor progression, they are not the whole story. Cancer cell properties are also influenced by alterations in the expression of normal genes. The term epigenetic change is used to refer to any such alteration in gene expression that does not involve mutating the structure of a gene itself.
Cells possess a variety of mechanisms for altering gene expression. Among them, activating or inhibiting the transcription of individual genes into messenger RNA is especially important in cancer cells. For example, many of the traits required for cancer cell invasion and metastasis are produced by epigenetically turning on or turning off the transcription of normal genes rather than by gene mutation.
Because the DNA base sequence is not being altered, epigenetic changes are easier to reverse than mutations. The question therefore arises as to whether a cancer cell can be epigenetically reprogrammed to reverse some of the changes responsible for malignant behavior. One way of addressing this question experimentally is to transfer the nucleus of a cancer cell into a different cytoplasmic environment to see if its gene expression patterns can be converted to a more normal state.
When nuclei are taken from mouse cancer cells and transplanted into mouse eggs whose own nuclei have been removed, the eggs divide and proceed through the early stages of embryonic development, even though the cells possess cancer cell nuclei. Especially striking results have been reported when mouse melanoma cells (a cancer of pigment cells) are used as a source of nuclei for transplantation.
Eggs receiving melanoma nuclei divide and produce embryonic cells that give rise to normal-appearing cells and tissues of adult mice (Figure 18). Nonetheless, mice containing such cells are not completely normal; the mice still exhibit an increased susceptibility to developing cancer.
Such results indicate that the DNA of a cancer cell nucleus can be reprogrammed to a more normal state, but a propensity for cancer to arise still remains. In other words, epigenetic and genetic changes both play important roles in tumor development.
To sum up, tumor progression is a phase of carcinogenesis that involves the gradual acquisition of DNA mutations and epigenetic changes in gene expression, accompanied by natural selection of cells that have acquired advantageous properties generated by these mechanisms.
The net result is a population of cells whose properties, including growth rate and the ability to invade and metastasize, slowly change over time. The time required for tumor progression contributes to the lengthy delay commonly observed between exposure to carcinogenic chemicals and the development of cancer. These principles, derived largely from studies of chemical carcinogenesis, apply to cancers triggered by other cancer-causing agents as well.
Carcinogenesis is a Probabilistic Event that Depends on Carcinogen Dose and Potency:
The realization that chemical carcinogenesis is a multistep process involving several distinct stages and mechanisms can cause some confusion about the meaning of the term carcinogen. In common usage, any agent that increases the risk of developing cancer in animals or humans is considered to be a carcinogen. In this sense, either an initiating or a promoting agent would qualify as a carcinogen.
For clarification, the term incomplete carcinogen is sometimes employed when referring to a chemical that exerts only one of these two actions. Some chemicals possess both initiating and promoting activities, and can therefore cause cancer by themselves; such chemicals are called complete carcinogens.
The ability to function as a complete carcinogen may be dose dependent. For example, certain polycyclic hydrocarbons act as initiating agents at lower doses but are complete carcinogens at higher doses. In normal human experience, people are exposed to chemical mixtures, such as tobacco smoke or coal tar that contain both initiating and promoting carcinogens. In such cases, the mixture acts as a complete carcinogen.
The multistep nature of chemical carcinogenesis also complicates the question of what scientists mean when they say that something “causes” cancer. For example, exposure to an incomplete carcinogen (i.e., an initiating or promoting agent) will not, by itself, cause cancer. Even a complete carcinogen rarely causes cancer in every exposed person or animal.
When it is stated that a particular carcinogen causes cancer, what is really meant is that the agent in question increases the probability that cancer will arise. The magnitude of the increased risk depends on several factors, including the dose and potency of the agent involved and the issue of whether it is acting as an incomplete or complete carcinogen (complete carcinogens obviously carry a greater risk).
The reason for carcinogen dose dependence should now be more apparent. As the dose of an initiating carcinogen is increased, more DNA adducts and other types of DNA damage accumulate. To initiate the development of cancer, this damage must affect certain critical genes.
The probability that one of these cancer-related genes will happen to undergo mutation is quite small because carcinogens trigger random DNA damage and the critical genes constitute only a tiny fraction of the total DNA. The higher the dose of carcinogen, however, the greater the overall DNA damage and hence the greater the chance that a critical gene will be affected by a random mutation.
The likelihood that a particular carcinogen will cause cancer also depends on its potency. Carcinogen potency is generally assessed in animals by determining how large a dose is needed to cause cancer in 50% of the animals tested. Such testing has revealed that a ten-million-fold difference in strength separates the strongest carcinogens from the weakest (Figure 19).
Two properties are especially important in explaining this enormous variation in potency. The first involves the activation reactions catalyzed by cytochrome P450, which are more effective in converting certain types of chemicals into active carcinogens than they are for other chemicals. The second factor is related to the electrophilic strength of different carcinogens.
Some substances are strongly electrophilic and are highly reactive with DNA, whereas others are weaker electrophiles and are less reactive with DNA. The probability that a carcinogen will happen to mutate a critical gene randomly is much greater for carcinogens that are stronger electrophiles because they trigger more mutations.
The random nature of mutation helps explain why everyone who is exposed to carcinogens does not develop cancer. For many years, tobacco companies tried to cast doubt on the relationship between cigarette smoking and lung cancer by pointing out that some cigarette smokers live long lives without ever developing cancer. The ability of carcinogens to trigger random DNA mutations provides a simple explanation for such observations: It is largely a matter of chance.
Tobacco smoke contains numerous carcinogens that cause random DNA damage, but for cancer to develop, a mutation must arise in a critical cancer-related gene. An apt metaphor is the game of Russian roulette, in which a single bullet is placed in a gun containing six chambers and the cylinder is then spun. When the trigger is pulled, the probability of firing a bullet that can kill is 1 in 6.
A similar principle applies to smoking cigarettes, a potentially lethal practice governed by the laws of probability. Each cigarette has a small but finite probability of randomly creating a mutation that can cause cancer. Like Russian roulette, the more the game is played, the greater the chance that lethal damage will occur. So when it is stated that smoking cigarettes (or exposure to any other carcinogen) “causes” cancer, it simply means that a person’s risk of developing cancer is increased. For this reason, agents exhibiting the potential to cause cancer are sometimes referred to as cancer risk factors.
In concluding, brief mention should be made of the fact that immunosuppressive drugs increase cancer risk in a fundamentally different way from the chemical carcinogens we have been discussing. Immunosuppressive drugs are given to organ transplant patients to inhibit the immune system and thereby minimize the possibility that a transplanted organ will be rejected.
Because they inhibit immune function, some immunosuppressive drugs increase cancer risk by diminishing the likelihood that immune surveillance will destroy newly forming cancer cells. These immunosuppressive drugs differ from typical carcinogens in that they increase cancer risk indirectly, targeting the immune system rather than acting directly on the cells destined to become cancerous.