History tells us that disease has been a part of the living process on this planet for eons. In fact, it is clear that life as we know it, by its very nature, requires that disease processes exist. It is natural, then, that a thinking, reasoning human organism should concern itself with disease and its effects on individuals as well as populations. Diseases that are self-limiting and readily con- trolled by natural life processes present no major problem for humans, animals, or plants. Our concern is with diseases that are potentially life-threatening or morbidly debilitating.
Since the dawn of civilization there have always been a few disease entities of great con- cern to humans. As evidenced by biblical writings, the disease most feared and abhorred by the population of the western civilized world at that time was leprosy. Later, in the Middle Ages and the Renaissance in Europe, the dreaded disease was the bubonic plague, or “black death.” Dur- ing the last century, a major killer associated with considerable human suffering was the “white death,” or tuberculosis. With the effective antimicrobial therapy developed in the twentieth cen- tury, infectious diseases now play a lesser role in “developed cultures” than in the past, although in relatively underdeveloped countries of the Third World, infectious diseases such as malaria and hookworm are still of paramount importance and concern. In modern times, however, espe- cially during the last half of the twentieth century, the most feared disease is cancer. One of the more succinct descriptions emphasizing the impact of this fear was presented at a symposium on cancer in 1936 by Glenn Frank, President of the University of Wisconsin.
But not all these tragic consequences together are the worst evil wrought by cancer. For every- body that is killed by the fact of cancer, multiplied thousands of minds are unnerved by the fear of cancer. What cancer, as an unsolved mystery, does to the morale of millions who may never know its ravages is incalculable. This is an incidence of cancer that cannot be reached by the physician’s medicaments, the surgeon’s knife, or any organized advice against panic. Nothing but the actual conquest of cancer itself will remove this sword that today hangs over every head.*
Although the United States was not the first country to proclaim the conquest of cancer as a national effort, the government’s financial backing of cancer research during the 1970s pro- vided the greatest single impetus in the history of this country to the scientific search for knowl- edge and understanding to control and eliminate cancer. In 1970, a special panel of consultants called together by the U.S. Senate submitted a “Report of the National Board of Consultants on the Conquest of Cancer” (1971); at that time, this was perhaps the best summary of the status of cancer as a disease and of cancer research in this country. This report showed that cancer is the primary health concern of the people of the United States. In several polls, approximately two- thirds of those questioned admitted fearing cancer more than any other disease. Of 200 million Americans living in 1970, some 50 million were destined to develop cancer, and approximately 34 million would die of the disease. According to the American Cancer Society (1993), about 85 million Americans living in 1993 will eventually develop cancer. About one-half of all deaths due to cancer occur prior to the age of 65, and cancer causes more deaths among children aged 1 to 14 than any other disease. About 20% of all deaths in this country are caused by cancer; it is second only to cardiovascular disease as the greatest killer of our population.
The committee of consultants that was convened in 1970 pointed out that in 1969 the bud- get of this country, on a per capita basis, provided $410 for national defense; $125 for the war in Vietnam; $19 for the space program; $19 for foreign aid; but only 89 cents for cancer research. During the same year, deaths from cancer were eight times the number of lives lost in all 6 years of the Vietnam War up to that time, 5¹⁄₂ times the number of people killed in automobile acci-dents in that year, and greater than the number of American servicemen killed in battle in all 4 years of World War II. Hodgson and Rice (1995) have indicated that the yearly cost to this na- tion’s economy because of cancer is nearly $73 billion, with the cost of medical care of cancer patients being more than $18 billion per year. These figures do not take into account the costs in suffering, mental anguish, and psychosocial trauma that haunt both cancer patients and their families.
We do not yet understand the basic nature of cancer; however, we know a great deal more about the disease today than we did 50 years ago. In 1930, the medical cure rate for those af- flicted with cancer was about one in five. Today, approximately two in five are cured, and the panel’s findings and subsequent studies have demonstrated that this could be improved to almost one in two simply by better application of the knowledge that exists today. In fact, in 1982, the National Cancer Institute’s SEER Program (see below) presented data to indicate that nearly 50% of white patients with cancer, excluding nonmelanoma skin cancer and carcinoma in situ (see Chapter 9), will survive to die of other diseases. Certain specific types of cancers that were 100% fatal prior to 1960 can now be cured in as many as 70% of the cases (see Chapter 16).
*Quoted from the welcome by President Glenn Frank to participants in “A Symposium on Cancer,” Uni- versity of Wisconsin School of Medicine, Madison, Wisconsin, September 7–9, 1936. University of Wis- consin Press, Madison, 1938.
In all likelihood, all multicellular organisms are afflicted or have the potential to be afflicted with the disease we call cancer. Paleopathologists have shown that cancerous lesions occurred in dinosaur bones long before the advent of Homo sapiens (Bett, 1957). In view of the numerous reports of spontaneous and induced cancers in plants and animals, vertebrates as well as inverte- brates, it is probable that cancer has been with us for much of the evolutionary period of life on earth. Ancient Egyptians knew of the existence of cancer in humans, and in one papyrus, the Edwin Smith papyrus, a glyph clearly refers to a clinical cancer of the breast (Fig. 1.1). In addi- tion, autopsies of mummies have shown the existence of bone tumors and the probability of other cancerous processes.
By the era of Hippocrates in the fourth century B.C., many types of cancers were clinically recognized and described, such as cancer of the stomach or uterus. Hippocrates felt that in many instances little could be done for the cancer patient and, more importantly, that it was to this disease that one of his cardinal rules, Primum non nocere (first do no harm), applied. Hippo- crates coined the term carcinoma, which referred to tumors that spread and destroyed the pa- tient. This was in contrast to the group he termed carcinos, which included benign tumors, hemorrhoids, and other chronic ulcerations. He proposed that cancer was a disease of an excess
Figure 1.1 The hieroglyphic symbol for the word tumor, referring to the surgical treatment of cancer of the breast as described in the Edwin Smith papyrus, dated earlier than 1600 B.C. The reader is referred to Breasted’s translation (1930) of the document for further information.
of black bile, which was manufactured by both the spleen and stomach but not the liver. This concept of the causation of cancer remained the predominant theory for almost 2000 years. Hippocrates as well as other physicians during the next two millennia tended not to treat ulcer- ated or deep-seated cancers, because “if treated, the patients die quickly; but if not treated, they hold out for a long time.”
Almost 600 years later, Galen distinguished “tumors according to nature,” such as en- largement of the breast with normal female maturation; “tumors exceeding nature,” which in- cluded the bony proliferation occurring during the reuniting of a fracture; and “tumors contrary to nature,” which today we may define as benign or malignant tumors (Chapter 2). This distinc- tion, proposed some 1800 years ago, is still reasonably correct. Galen also suggested the similar- ity in gross outline between a crab and the disease we know today as cancer.
The concepts of Hippocrates and Galen dominated medical practice during the Middle Ages. With the advent of the Renaissance and during the seventeenth and eighteenth centuries, the “black bile” theory of the causation of cancer was disputed by a number of physicians (in- cluding Ramazzini), and the surgery of neoplasms became somewhat more extensive. Several treatises on mastectomies for breast cancer, including dissection of regional lymph nodes, were written. Ramazzini attributed the high occurrence of breast cancer among nuns to the celibate life of these women. This was the first example of occupation-associated cancer, an observation that has withstood the test of time. In addition, in 1761, John Hill of London suggested that tobacco in the form of snuff was a cause of nasal tumors or polyps.
It was not until the nineteenth century, however, that physicians and scientists began to study cancer systematically and intensively. The anatomist Bichat extended the principles of Ga- len, which had reigned supreme for more than 1600 years. Bichat (1821) described the anatomy of many neoplasms in the human and suggested that cancer was an “accidental formation” of tissue built up in the same manner as any other portion of the organism. Seventeen years later, Johannes Müller (1838) extended the findings of Bichat through the use of the microscope. Al- though the cellular theory was just being formulated during this period, Müller independently demonstrated that cancer tissue was made up of cells. At the time little was known about cell division, and Pasteur and others had not yet demonstrated the doctrine Omnis cellula e cellula, that is, “Every cell from a cell.”
A student of Müller, Rudolf Virchow (1863), dramatically extended our descriptive knowledge of cancer; although he proposed a number of theories that were later disproved, he was the first to point out a relation between chronic irritation and some cancers.
Early in this rapid advance of our knowledge of cancer, two possible pathogenetic bases for the origin of cancer were proposed—that normal cells are converted to cancer cells, or that cancer cells exist from embryonic life but do not express themselves until later in the organism’s existence. Müller (1838) supported the latter concept, as did Julius Cohnheim, who in 1877 ad- vanced the “embryonal rest theory” of cancer. On the other hand, many pathologists, such as Laënnec, argued that a number of cancers resemble the normal tissues of the body and that “there are as many varieties of these as there are kinds of normal tissues.” Laënnec did, however, recognize that a number of tumors bore no direct resemblance to any normal tissue found in the adult organism. Laënnec’s studies supported the cellular theory (see above) and actually added to it the words ejusdem naturae, which, combined with the original statement, may be translated as “Every cell arises from a cell of the same kind” (cf. Shimkin, 1977).
In 1829, Recamier published Recherches du Cancer, in which he specifically introduced the term metastases and described clearly how cancer spreads by this method (Chapter 2). An- other major advance was the demonstration by Waldeyer (1872) that metastases were the result of cell emboli. In addition, he was able to show that cells from primary cancers infiltrated blood and lymphatic vessels.
After major advances had been made in the knowledge of the biology of human cancer, experimental oncology emerged as a separate area of study. The first example of the successful transplantation of an experimental tumor was reported by Novinsky (1877), who succeeded in transplanting a nasal cancer from an adult dog to several puppies and then maintained the cancer in vivo for at least one or two generations. By 1900, some animal neoplasms had been carried through many generations of grafts with few alterations in the microscopic appearance of the cancers.
Students interested in a more detailed and readable discussion of some aspects of the his- tory of the science of oncology are referred to Shimkin’s Contrary to Nature (1977), which shows by extensive illustration and relatively complete documentation the development of on- cology from ancient times to many of the major discoveries through 1975.
During the nineteenth century, many hypotheses of the origin and development of cancer were presented. In general, these hypotheses may be categorized as follows:
1. The irritation hypothesis
2. The embryonal hypothesis
3. The parasitic hypothesis
The first hypothesis encompassed what little was known at the time of the effects of chem- ical agents, mostly crude, and of radiation in the genesis of cancer. The relation of some ulcer- ations, both internal and external, to cancer appeared to support and strengthen this hypothesis. Cancers arising in old scars and those occurring after both acute and chronic injury were also cited in support of the irritation hypothesis.
Perhaps the most common example in support of the embryonal hypothesis is the nevus, or common mole of the skin. In most instances nevi are present from birth, and a very small per- centage of such structures become cancerous. Many cancers that appear to resemble embryonic tissue, such as the teratoma (Chapter 2) occurring in the adult, also support this hypothesis.
Prior to the nineteenth century, Hippocrates’ “black bile” theory of cancer causation served to inhibit any concepts of an infectious etiology of cancer. However, in view of the rapid advances in our understanding of infectious disease during the last century by Pasteur and nu- merous others, physicians and scientists have searched for an infectious origin of cancer during the last 100 years. Several reports appeared at the end of the nineteenth century, including that of Doven, who described a bacterium, Micrococcus neoformans, which he isolated from several neoplasms and believed to be the cause of all types of cancer (cf. Bett, 1957; Oberling, 1952). As it turned out, this organism was merely a common staphylococcus. It was not until the twen- tieth century that the infectious hypothesis became scientifically sound. Even with the dawn of this century, more than 50 years were to pass before proper scientific recognition was given to the parasitic hypothesis (see Chapter 4).
Today the demographics and statistics of cancer in the human race have become topics of great popular concern and study. An interesting prelude to the extensive statistical and epidemiologic investigations (Chapters 11 and 12) of human cancer was the book The Mortality from Cancer Throughout the World, by F. L. Hoffman (1915), which in part comprised a report to the Pruden- tial Insurance Company of America on the “statistics” of cancer and its application to the life insurance industry. A number of the points raised by Hoffman in relation to the increase in can- cer incidence seen in the world at that time, the mortality from cancer in different occupations, and the geographical consideration of cancer statistics have all proved to be major factors in our understanding of cancer as a disease in the human race today.
The incidence of cancer in the human population may be defined as the rate of diagnosis of the disease in the human population. This can be expressed in a variety of ways, as demonstrated by several tables and figures in this chapter. For the student it is important to gain a clear under- standing of the distinction between cancer incidence rates and cancer mortality rates, the latter being discussed later in this chapter. Cancer survival rates are related to cancer mortality rates by the success or failure of the therapy of the disease.
Table 1.1 lists the ten most common cancers in the world (excepting nonmelanotic skin cancer) in 1985 on the basis of estimates by the International Agency for Research on Cancer (Parkin et al., 1993; Pisani et al., 1993). The basis of these studies was an investigation of cancer incidence and mortality patterns in 23 geographical areas of the world, as depicted in Figure 1.2. In this worldwide survey, the crude cancer incidence rates (see below) for all (or most) of the countries in a given area were estimated and then the weighted average was calculated, where the weights are the populations of the individual countries as determined in 1985. Wherever possible, incidence rates were derived from population-based cancer registries or other reliable sources (Parkin et al., 1993). In countries in which no incidence data were available but mortality rates from cancer could be obtained, estimates of incidence were determined with a set of conversion factors (Parkin et al., 1993). Similarly, for countries or regions in which valid cause-specific mor- tality information is not available, mortality rates have been estimated from incidence data. For a given cancer site, mortality is empirically related to incidence by the following relationship:
M = I [k – Si]
where Si is the relative survival at year i of follow-up, while k is a constant depending on i. M
and I are the mortality and incidence rates respectively (Pisani et al., 1993). Muir (1990) has also
indicated that the expected increase in cancer in the world each year is approximately 6.4 mil- lion new cases. This number is almost equally divided between Third World countries and those that are more developed.
In the United States, the most accurate cancer incidence rates are those of the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program, now covering 12 geographic areas with population-based registries. These include the metropolitan areas of Atlanta, Detroit, New Orleans, Seattle–Puget Sound, and Oakland–San Francisco as well as the states of Connecticut, Iowa, New Jersey, New Mexico, Utah, and Hawaii and the Common- wealth of Puerto Rico. Several more population areas are due to be added to the SEER program in the near future. This program involves a sample of approximately 10% of the U.S. population and thus has a base analogous to that of the world incidence study seen in Table 1.1. While this is a very large sample of the U.S. population and has been shown to reflect trends in cancer statistics for the entire country, some race-, sex-, and site-specific differences in the magnitude of trends and levels of mortality occur in the SEER data as compared with those from the entire U.S. population (Frey et al., 1992).
On the basis of previous data from the SEER program, the American Cancer Society esti- mated that, in the United States in 1993, approximately 1.17 million new cases of cancer, ex- cluding nonmelanoma skin cancer, would occur, essentially equally divided between males and females. For a child born in 1985, the probability at birth of developing cancer (excluding non- melanotic skin cancer) at some time during its life span is about 33%. The probability for that individual of eventually dying of cancer is about 20% (Centers for Disease Control, 1986). In general, males have a higher age-specific incidence of cancer than females when all sites com- bined are considered. Even in the best-controlled epidemiological studies as exemplified by the SEER program, these data may be incomplete, since some cases of cancer are never diagnosed. The failure to diagnose cancer is related not only to the lack of contact of an individual with a physician but also to the frequent lack of interaction of a patient with the best methods of cancer diagnosis, found only in modern hospitals. Earlier studies (cf. Bauer et al., 1973) demonstrated that the likelihood of discovering an undiagnosed or incorrectly diagnosed case of cancer in- creases dramatically as the number of hospital admissions increases. Thus, as medical care for the U.S. population improves in efficiency and availability, it is hoped that the patient who seeks medical advice and has undiagnosed cancer will become a relative rarity in our society.
Except for cancer of the skin—the most common and, in most cases, the most curable of human cancers—75% of all cancers in humans in the United States occur in only ten anatomic sites: colon and rectum, breast, lung and bronchus, prostate, uterus, lymph organs, bladder, stomach, blood, and pancreas. In the U.S. male, one of the most common sites is the lung, ac- counting for 17% of such cancers in 1993 (American Cancer Society, 1993). A similarly com- mon site of cancer incidence in the U.S. male is the prostate, which accounts for 27% of the 10 most common cancers. In the U.S. female, cancer of the breast accounts for 32% of these can- cers; in both males and females in the United States, the incidence of cancer of the colon and rectum is approximately 13% of all cancers.
The age-specific incidence of cancer at the four most frequent sites for males and females as reported by the most recent SEER publication (Miller et al., 1992) is seen in Figure 1.3. In the male, the incidence of these cancers increases dramatically after age 40 and continues through- out life except for cancer of the lung and bronchus. The reason for this decline in men over age 80 has been speculated to be the result of either (1) the relatively low incidence of smoking in this group as young individuals, since smoking became most popular after the 1930s, or (2) the lethality of the disease and its greater incidence in the 60- to 80-year-old age range. The former explanation does not coincide with the study by Harris (1983), who reported that, in a represen- tative sample of the U.S. population, maximum exposure to cigarette smoking probably oc-
curred among men now in their seventh and eighth decades. In that study it was also postulated that the peak exposure to smoking probably occurred in women presently in their fifth and sixth decades; this would also conform with the decreasing incidence of lung cancer in females over age 70. Today, deaths from lung cancer exceed those from breast cancer in women (see below). The reason for the slight decrease in breast cancer and in cancer of the uterus at older ages is not readily apparent.
On a worldwide basis, the incidence of various types of potentially fatal cancers is some- what different from that in the United States, as noted in Table 1.1, when compared with Figure
1.3. As in the United States, the total cancer burden in the world includes cancer of the lung and cancer of the breast as exhibiting the highest incidences in males and females, respectively. When all areas were considered together for both sexes, the most frequent cancer worldwide in 1980 was cancer of the lung (Table 1.1). Stomach cancer remains the most common cancer in some parts of the world, including Japan, China, other east Asian countries, and the former So- viet Union, and incidence rates still remain relatively high in both Europe and Latin America. Nevertheless, stomach cancer is declining in frequency almost everywhere in the world, with the estimated number of cases decreasing by 1.9% since the previous estimates in 1975. This is es- pecially noteworthy since a population increase of 9.4% occurred during the same period. Un- like stomach cancer, lung cancer has been increasing in incidence between 1975 and 1980, by 11.8% in males and 16% in females worldwide. Thus, lung cancer is the most common fatal cancer in the human race as we approach the twenty-first century. Another striking finding from these data is that cancers of the breast and cervix, both limited to females, have a higher inci- dence than the two most common cancers in males—lung and stomach. Although it is possible that the increasing incidence of breast cancer worldwide may be the result of changes in the way the estimates are calculated (Parkin et al., 1993), cancer of the uterine cervix is also increasing, especially in underdeveloped nations, in which methods for early diagnosis are not as well de- veloped as in the “developed” areas of the world.
In the United States, the most dramatic changes in the incidence of cancer also reflect those seen in the world. This is exemplified in Figure 1.4, showing the change in incidence of specific cancer types during the period 1950–1989. As seen from the figure, the most striking differences are again seen in cancer of the lung and bronchus, with the percentage change being greatest in females during this period. Among other cancers that increased dramatically during this period, non-Hodgkin’s lymphomas, melanomas, and cancers of the kidney, testes, and pros- tate increased by more than 100%. The incidence of cancer of the breast, while increasing mark- edly during this period, has not seen a dramatic change like those indicated above. Not listed in the figure is the recent finding of the dramatic increases seen in the incidence of brain cancers in the period 1973–1985. The most dramatic increases during this period were noted for persons aged 75 to 79, 80 to 84, and 85 years and older, where the relative increases were 187%, 394%, and 501% respectively (Greig et al., 1990).
There are even more striking differences in the incidences of specific cancers in certain areas and countries of the world. The rates for a variety of cancers in the highest- and lowest- incidence areas are shown in Table 1.2. The incidence rates of some cancers may vary as much as 300-fold in different areas of the world. Such variability led some epidemiologists more than two decades ago, when this information first became available, to suggest that most cancers in humans are the result of environmental factors, since inherent genetic and related factors could not explain such large differences in incidence of specific cancers in various parts of the world (cf. Wynder and Gori, 1977).
Changes in the incidence of cancer may be due either to an absolute change in the inci- dence—resulting, for example, from some alteration in the environment or to better diagnostic methods—which give rise to an apparent increase in the incidence of a specific cancer. One ex-
ample of the latter was described by Linos and associates (1981) for a bone marrow cancer, mul- tiple myeloma. Although a number of incidence studies indicated that multiple myeloma had increased significantly in recent years in the United States, no such increase occurred in the pop- ulation of a carefully monitored county in Minnesota over a 20-year period. Since no other study that reported increased incidence rates analyzed a well-defined population over a long period, it is likely that the increased incidence of multiple myeloma reported in several other parts of the country were the result of improved diagnosis. This factor is also important when one attempts to compare incidence, mortality, and survival rates over defined periods of time .