Epidemiology of Cancer and Prevention Strategies

April 1, 1995

Cancer epidemiology is the study of the distribution, determinants, and frequency of malignant disease in specific populations [1]. The objective is to define causative factors to formulate preventive strategies for control of the disease.

Descriptive EpidemiologyAnalytic EpidemiologyClinical EpidemiologyReferences

Cancer epidemiology is the study of the distribution, determinants, and frequency of malignant disease in specific populations [1]. The objective is to define causative factors to formulate preventive strategies for control of the disease. Epidemiologic assessment provides the clinician with a quantification of cancer risk, outlines the basis for screening modalities for high-risk populations, and determines the efficacy of any preventive intervention.

Three types of epidemiologic research apply to the field of cancer. Descriptive epidemiology focuses on the trends and frequency of disease in a given population. Analytic epidemiology deals with identifying causes and the predisposing risk associated with the development of disease. Clinical epidemiology outlines screening programs and evaluates the impact of prevention strategies on overall outcome.

Descriptive Epidemiology

The American Cancer Society estimates that during 1995, there will be 1,252,000 new cancer cases and 547,000 deaths from cancer in the United States. In addition, about 120,000 new cases of carcinoma in situ (uterine, cervix, breast, and melanoma) plus more than 800,000 basal- and squamous-cell skin cancers will be diagnosed [2].

Cancer incidence and mortality rates are higher among males than females [2]. In addition, Americans over 65 years old have a tenfold greater risk of developing cancer than younger individuals. Despite an increase in the overall cancer mortality rate between 1950 and 1990, the mortality rates for all cancers combined have declined substantially for individuals under 45 years old but increased for individuals over 55 years old. Most of the increase is attributable to deaths from lung cancer. African-Americans have a higher cancer mortality rate than whites [3].

Currently, the leading cancers-those of the lung, breast, prostate, colon and rectum, and ovary-account for nearly 61% of the cancer burden in the United States (Table 1). If lung cancer is excluded, overall cancer deaths have declined over 14% since 1950 [3].

SiteEstimated new casesEstimated deaths
Colon and rectum138,20055,300
All sites1,252,000547,000

Analytic Epidemiology

The goal of analytic epidemiology is to identify the factors that predispose individuals to the development of disease and to quantitate risk. Cancer risk factors include environmental exposures, genetic susceptibility, and immunosuppressive state but may be secondary to prior history of malignancy, viral infection, or therapy. These risk factors can act at different steps during carcinogenesis. Some risk factors linked to the development of cancer are listed in Tables 2 and 3. Established risk factors for site-specific malignancies will be discussed below.



Occupational exposures




Genetic susceptibility


DrugCancer site
Alklylating agentsLeukemia
CyclophosphamideUrinary bladder
Androgen-anabolic steroidsLiver
Azathioprine, cyclosporineNon-Hodgkin's lymphoma
Phenacetin-containing analgesicsRenal pelvis
Synthetic (diethylstilbestrol)Vagina, cervix (adenocarcinoma)
Steroid contraceptivesLiver

Cancer of the Head and Neck

Between 1973 and 1989, the incidence rates of oral and pharyngeal cancer declined in white males of all ages while they increased in African-American males. African-American males younger than 65 have almost twice the incidence of oral and pharyngeal cancer seen in white males, but the trend changes for white men older than 65, who have higher rates than African-American men older than 65. The incidence rates of laryngeal cancer are significantly higher for African-American males of all ages than for white males [4] and are continuing to rise for females [3].

The most common head and neck malignancies are squamous-cell carcinomas of the upper aerodigestive tract. Tobacco exposure is a major etiologic factor. A 5- to 25-fold increased cancer risk has been documented for heavy smokers, compared with nonsmokers [4]. A clear dose-response relationship exists, and when smoking is combined with alcohol consumption, their effects on cancer risk appear to be synergistic. On the other hand, smoking cessation is associated with a declining risk [5].

Cigar and pipe smokers, as well as users of smokeless tobacco, are at increased risk for developing head and neck malignancies. Cigar and pipe smoking is associated with oral, pharyngeal, and laryngeal cancers [4]. Pipe smokers are predisposed to developing cancer of the lip [6]. Smokeless tobacco, primarily chewing tobacco and snuff, is known to cause cancer of the oral cavity, perhaps because of the high concentrations of tobacco–specific N-nitrosamines in smokeless tobaccos [6].

Marijuana smoking may be a carcinogen to the upper aerodigestive tract, possibly because the smoking increases exposure of the respiratory tract to tar and because of the rapid, deep inhalation used in marijuana smoking, which leads to particulate deposition [7].

Alcohol use has been associated with cancers of the oral cavity, pharynx, and larynx, as well. The effect of alcohol use is multiplicative to that of tobacco smoking, particularly for laryngeal cancers [4].

Several epidemiologic studies have suggested an inverse relationship between micronutrient intake and cancer incidence. A protective effect has been reported for vitamin C against oral and pharyngeal cancer [8], vitamin A against oral cancer [9], and beta-carotene against laryngeal cancer [10].

Additional risk factors for head and neck malignancies include occupational exposure to wood dust, organic compounds, and coal products [4]. An increased risk of developing laryngeal cancer has been related to asbestos exposure, exposure to nickel in smelting operations, occupational exposure to sulfuric acid, and the manufacture of mustard gas [4].

The occurrence of a head and neck malignancy itself places the affected individual at higher risk for a secondary primary cancer.

Cancer of the Lung

Cancer of the lung is the second most common malignancy affecting both sexes. In the early 1950s, it became the leading cause of cancer deaths in men. In the mid-1980s, women's lung cancer mortality rates surpassed those from breast cancer. For 1995, it is estimated that 169,900 new cases of lung cancer will be diagnosed and that 157,400 patients with lung cancer will die of the disease [2]. Lung cancer is considered the most rapidly increasing cause of death from cancer [11].

Cigarette smoking is a well-established pulmonary carcinogen. It is responsible for 90% of male and 78% of female lung cancer deaths [12]. A dose–response effect has been demonstrated between number of cigarettes per day smoked, duration of smoking, and subsequent cancer risk. This risk is 20-fold higher for one-pack-a-day smokers with over 30 years of tobacco use than it is for nonsmokers [6]. Younger age at initiation of smoking increases risk because it increases the overall duration of smoking, not because young people have an increased susceptibility [13]. Lifetime tar exposure is another good index of cancer risk [12]. There is an increased risk of developing airway obstruction [11,13] related to smoking unfiltered cigarettes [13]. Smoking cessation reduces the risk of lung cancer in former smokers, although the risk remains higher than the risk for those who have never smoked [13].

Environmental tobacco smoke or passive exposure to smoking has been linked to lung cancer. The risk of developing lung cancer in a nonsmoker married to a smoker has been estimated to be increased by about 30% [11,13].

Other environmental carcinogens include polycyclic hydrocarbons generated from the combustion of fossil fuels, radionuclides, and diesel exhaust [13]. Exposure to radon and its byproducts has been established as a cause of lung cancer. Radon interacts synergistically with cigarette smoke [11,13]. Annually, approximately 14,000 lung cancer deaths in the United States are attributable to radon exposure among smokers, making radon the second most important etiologic agent for lung cancer [13].

Among the occupational carcinogens, past exposure to asbestos predisposes to the development of mesothelioma in the absence of tobacco use [11]. Among cigarette smokers, the risk for lung cancer is 50-fold higher in workers exposed to asbestos without protection, than in the nonsmoking, unexposed population [11]. Workers exposed to asbestos include those in the shipbuilding, fireproofing, acoustic-control, and pipe-insulation industries [12]. Other workplace exposures are listed in Table 4.

As discussed for head and neck malignancies, an inverse association between beta-carotene intake and lung cancer risk has been established [11].

Cancer of the Esophagus

During the past two decades, the incidence rates of squamous-cell carcinoma of the esophagus have remained stable, whereas a steady increase for esophageal adenocarcinomas has been documented [14-16]. In the United States, esophageal cancer occurs most often in black men [15]. Mortality rates among blacks are three times higher than among whites [3].

The risk of squamous-cell carcinoma of the esophagus is increased by agents that cause chronic irritation of the epithelial lining. Alcohol and tobacco use are the main etiologic factors; their interaction is multiplicative [14]. Thermal injuries from hot drinks, fungal toxins in pickled vegetables, achalasia, radiation, and strictures from swallowing lye may predispose to the development of cancer of the esophagus. This cancer has also been associated with poor nutritional status (particularly deficiencies of vitamins A and C and riboflavin), tylosis, and Plummer-Vinson syndrome [14,15]. In the Asian esophageal cancer belt, smoked opiates have a strong etiologic role [14].

However, adenocarcinoma of the esophagus is more common in white individuals. The risk for its development is 30 to 125 times higher in individuals with Barrett's esophagus [15].

Cancer of the Stomach

The incidence of gastric carcinoma in the United States has been declining for the past 50 years [17]. Mortality rates have also been declining and may be leveling off [3]. Helicobacter pylori infection and conditions that result in achlorhydria with intragastric bacterial overgrowth and formation of N-nitroso compounds have been implicated in the development of gastric carcinoma [18].

Cancer of the Liver

Infection with hepatitis B virus has been causally related to the development of hepatocellular carcinoma. The relationship is limited to chronically active forms of the hepatitis B virus, that is, those that have hepatitis B surface antigen [19]. In populations with a high incidence of hepatocellular carcinoma, the risk for its development is more than 200 times greater in hepatitis B surface antigen carriers than among noncarriers [20].

Another etiologic factor is infection with hepatitis C virus, which progresses to chronic hepatitis and cirrhosis, ultimately leading to the development of hepatocellular carcinoma [21].

Environmental exposures linked to hepatic malignancies include exposure to aflatoxins in contaminated food and to vinyl chloride, consumption of alcoholic beverages, medications such as androgen-anabolic steroids and steroid contraceptives, and parasitic infection with Clonorchis sinensis [22].

Cancer of the Colon and Rectum

In the United States, the colon and rectum are among the leading sites of cancer. An individual's lifetime risk for developing colorectal cancer is 1 in 20. The risk increases with age, with only 3% of cases occurring before age 40 [23]. Between 1973 and 1990, the mortality rates decreased significantly among whites and increased significantly among African-Americans [3].

The sigmoid colon is the most common site of colon cancer in both sexes in industrialized countries. The disease moves proximally to the right side of the colon with increasing age [23].

Diet is a well-established risk factor. Diets high in fat, particularly from animal sources, are associated with increased risk, whereas high fiber intake is associated with reduced risk. Diets rich in calcium, vitamin D, fruits, and vegetables also have a protective effect [23]. Alcohol consumption is related to rectal cancer and, less consistently, to colon cancer [24].

Adenomatous polyps are thought to be premalignant lesions for colorectal cancer. Risk of developing colorectal cancer is associated with the number and size of adenomas, their degree of dysplasia, and their histology, with villous polyps indicating a higher risk [23]. Patients with chronic ulcerative colitis are at higher risk of developing colorectal cancer. The degree of predisposition is proportional to the extent of disease involvement and the duration of active disease. First-degree relatives of colorectal cancer patients are more susceptible to developing the disease themselves [23].

About 1% of these cases have specific genetic disorders. Familial adenomatous polyposis is an autosomal dominant condition characterized by the tendency to develop hundreds of polyps. If left untreated, the polyps may become invasive cancer. The genetic defect has been traced to chromosome 5q21–q22, where a tumor-suppressor gene is presumed to reside. The same defect is found in patients with Gardner's syndrome (familial adenomatous polyposis with extracolonic manifestations). Patients with hereditary nonpolyposis colorectal cancer have family histories of colorectal cancer and are affected at an early age, often with right-sided mucinous tumors. These patients are predisposed to developing second malignancies of the breast, ovary, and endometrium [23,25].

A history of colorectal cancer increases the risk for a second colorectal primary tumor by about 5% [23].

Cancer of the Kidney and Renal Pelvis

The incidence and mortality rates for cancer of the kidney and renal pelvis are two times higher for males than for females. Most kidney malignancies arise in the renal parenchyma [3]. Individuals with hypertension or obesity, those who have been exposed to tobacco smoke or asbestos or have used phenacetin, and perhaps those with previous renal injury are at higher risk of developing renal-cell carcinoma. The risk for developing carcinoma of the renal pelvis increases by two- to threefold with exposure to tobacco smoke and by tenfold with abuse of phenacetin-containing analgesics. Upper urinary tract infections and stones may predispose to cancer of the renal pelvis, as well [26].

Patients with von Hippel-Lindau syndrome also are predisposed to developing renal-cell carcinoma [22].

Cancer of the Bladder

Bladder cancer predominantly affects elderly people and tends to have a different natural history in younger patients [27]. Males are affected four times more often than females [3].

Cigarette smoking is the most important cause of bladder cancer, accounting for 25% to 60% of cases in industrialized countries. This activity has a clear dose-response effect on the development of transitional-cell, squamous-cell, and adenocarcinoma histologic subtypes [27].

Occupational exposures in the dye, leather, textile, rubber, paint, petroleum, and chemical industries increase the risk of developing bladder cancer [27].

Infection with Schistosoma haematobium with secondary chronic inflammatory response is causally related to the development of bladder cancer, particularly squamous-cell carcinomas [27].

Exposure to the antineoplastic drug cyclophosphamide (Cytoxan, Neosar) predisposes an individual to bladder cancer. The association between bladder cancer and the artificial sweeteners cyclamate and saccharin, however, is derived from animal studies and has not been supported by human data [27].

Cancer of the Prostate

In the United States, prostate cancer is the most common malignancy and the second leading cause of cancer death affecting the male population [3]. It is a disease of elderly men; less than 1% of cases are men younger than age 50. The incidence increases 50-fold in whites and 30-fold in African-Americans between the ages of 50 and 85 [28].

African-Americans have the highest incidence of prostate cancer in the world [28,29], whereas Japan has one of the lowest rates [28]. Japanese migrants to Hawaii develop a risk for prostate cancer higher than that of native Japanese but only half that of American whites [28].

African-American men tend to have metastatic disease at diagnosis, indeed, 40% more often than whites do. The overall survival rate for African-American men is 10% lower than that for white men, even when they are diagnosed at the same stage of disease [29].

Neither smoking nor alcohol consumption affect the risk of developing prostate cancer [29,30]. Cadmium exposure, at work or in the diet, has been etiologically related to prostate cancer [29]. A familial tendency for prostate cancer exists, and members of such families are affected at an earlier age [31].

Cancer of the Ovary

One in 70 American women will develop ovarian cancer in her lifetime. Epithelial ovarian cancer is infrequent before the age of 35, after which incidence rates progressively increase up to the age of 75 [32]. Ovarian cancer is the most often fatal gynecologic malignancy [2]. Factors associated with increased risk include infertility, nulliparity, and use of fertility drugs. Tubal ligation and hysterectomy with ovarian conservation and oral contraceptive use have been shown to have a protective effect. A risk reduction of about 50% has been documented after 5 years of oral contraceptive use [32]. The effect increases with duration of use and persists for 10 to 15 years after discontinuation [33].

Familial clustering of ovarian cancer poses an increased risk on the basis of genetic susceptibility. Three different entities-site-specific ovarian cancer, breast-ovarian cancer, and Lynch syndrome II (hereditary nonpolyposis colon cancer with proximal colonic predominance, endometrial cancer, and ovarian cancer)-are jointly referred to as hereditary ovarian cancer syndrome [25]. An autosomal dominant mode of inheritance with variable penetrance has been suggested for these diseases. Therefore, the probability of a woman in an affected family developing ovarian cancer is about 50%. Two to four generations of a family are usually affected. In such families, the disease develops at an earlier age than sporadic cases do, but there is no difference in prognosis given similar stage at diagnosis [34]. The tumors are usually serous cystadenocarcinomas [33].

Hereditary ovarian cancer cases account for less than 1% of all cases [32]. Recent genetic studies located a gene on chromosome 17q21 (BRCA1) that predisposes to familial breast-ovarian cancer [35]. However, the lifetime risk of a 35-year-old woman developing ovarian cancer ranges from 1.6% if she has no affected relative members to 5% if she has one affected first-degree relative to 7% if she has two affected relatives [33,36].

Genetic susceptibility for developing ovarian neoplasms has been documented. Women with Peutz-Jeghers syndrome have a 5% risk of developing ovarian tumors, patients with gonadal dysgenesis (46XY) may develop gonadoblastomas, and benign ovarian fibromas develop in patients who have inherited basal-cell nevus syndrome [22].

Cancer of the Endometrium

Endometrial carcinoma is the most common gynecologic malignancy [2]. Its incidence is highest among white women, whereas its mortality rates are higher among African-American women. For the past two decades, the incidence rates have been declining, except among African-American women over 50 years old. Mortality rates have decreased for all ages [3]. Endometrial cancer, like ovarian cancer, is uncommon in before age 40; subsequently, incidence increases until age 70.

Established risk factors include use of unopposed estrogen replacement therapy, use of sequential oral contraceptives, obesity, nulliparity, and late menopause [35]. Breast cancer patients are at increased risk of developing endometrial carcinoma, among other cancers, probably because of a shared hormonal effect [37]. Furthermore, there is very clear evidence that tamoxifen given as adjuvant treatment for breast cancer increases the risk of endometrial cancer by three- to sevenfold [38]. Polycystic ovarian disease and estrogen-secreting ovarian tumors are also associated with an increased risk [35].

Use of combination oral contraceptives, which increase exposure to progesterone, and cigarette smoking, which may lower circulating estrogen levels, probably play a protective role [35].

Cancer of the Cervix

In the United States, the incidence and mortality rates for cancer of the cervix have decreased by more than 70% in the last 40 years [39]. The highest rates of invasive cervical cancer are found in Latin America, where women have a sixfold greater risk than do American white women. In the United States, the incidence rates for African-Americans and Hispanics are twice those of whites and Asians [40].

Cervical cancer tends to occur in women of lower socioeconomic classes. The risk is higher for women with multiple sexual partners; it has been reported to be three times higher for women who have had 10 or more partners than for women who have had one or no partners. Early age at first sexual intercourse increases the risk, perhaps because of increased susceptibility of the cervical epithelium to carcinogen exposure. Several studies have demonstrated that women who begin having sexual intercourse before age 16 have about twice the risk as those who begin after age 20. Multiparity has been related to cervical cancer risk, possibly because of cervical trauma during delivery and hormonal and nutritional changes during pregnancy. The risk is four times greater for Latin-American women who have borne 12 or more children than for those who have had only one child or no children [40].

Human papillomavirus types 16 and 18 have been causally related to cervical intraepithelial neoplasia. A history of genital warts (condyloma acuminatum), which are linked to human papillomavirus types 6 and 11, may explain the increased risk associated with multiple sexual partners [40]. Warts may also be a marker of infection with other types of human papillomavirus that are carcinogenic [41]. Other sexually transmitted viruses, like herpes simplex virus 2, may interact as etiologic factors [40].

Cancer of the Breast

Breast cancer is the most commonly diagnosed malignancy among women in the United States. For 1995, it is estimated that 182,000 women will be diagnosed with breast cancer and that 46,000 will die of the disease. Women's mortality rates for cancer of the breast are second only to those of cancer of the lung. Breast cancer incidence is low for women under age 40; only 6.5% of all such patients are less than 40 years old. However, the incidence rates triple by age 49 and double once again by age 69. Almost half of all cases of breast cancer are diagnosed in women age 65 and older [42]. The lifetime probability of a woman's developing breast cancer is one in eight, as determined by the revised methodology designed by the National Cancer Institute in collaboration with the American Cancer Society [42].

Between 1940 and 1982, breast cancer incidence rates in the United States increased by approximately 1% per year, largely in women over 40 years old. From 1982 through 1987, the rate of increase accelerated to around 4% per year and then leveled off. The rising rate is mainly attributable to early detection, due to the increase in breast cancer screening. The increase in breast cancer cases (with no change in incidence rates) among women 20 to 39 years old during 1970 to 1990 was due to a shift in the age distribution of the population. However, breast cancer mortality rates have remained fairly stable, with almost no change from 1950 to 1990 [42], increasing only about 0.2% per year [3].

Established hormone-related risk factors for breast cancer include early age at menarche, late age at menopause, nulliparity in women over age 40, and advanced age at first full-term pregnancy. The number of full-term pregnancies may increase the risk of breast cancer at younger ages but may be protective for women after the age of 50. Breast feeding and oophorectomy before menopause appear to be protective. Obesity in postmenopausal women elevates breast cancer risk, possibly due to peripheral conversion of androstenedione to estrogen in adipose tissue [35].

In premenopausal women, biopsy-proven proliferative benign breast disease with atypical hyperplasia is a marker for increased risk and has the potential for evolving into breast cancer [43]. A history of breast cancer predisposes to the development of a contralateral second primary [35], particularly if the initial tumor was lobular [37,44].

Breast cancer patients are at increased risk of developing malignant melanoma and cancers of the ovary, endometrium, colon, thyroid, and salivary glands because of similar hormonal and genetic factors. Elevated risks of leukemia, non-Hodgkin's lymphoma (NHL), and cancers of the lung and kidneys are believed to be a result of the treatment modalities used in breast cancer patients. Whereas hormonal treatment for an initial breast tumor reduces the risk in the contralateral breast by 50% [37,45], ionizing radiation at moderate to high doses increases the risk for breast cancer [35]. The risk depends on the woman's age at the time of exposure, and there is no increase in risk among women who were exposed after age 40 [22].

A family history of breast cancer predisposes members of affected families. Individuals with a history of premenopausal bilateral disease in first-degree relatives have the highest risk. The susceptibility is inherited in an autosomal dominant fashion with high penetrance and is manifested in both females and males [31]. A gene mapped to chromosome 17q21 (BRCA1) has been associated with early-onset familial breast cancer with a penetrance of 85% through age 70 [36,46].

Cancer of the Thyroid

Thyroid carcinoma is a rare malignancy with a high cure rate. It accounts for about 1% of all new cancers [3]. Women are affected two to three times more frequently than men [47]. Deaths from thyroid cancer account for only 0.2% of all cancer deaths per year [3].

Exposure to ionizing radiation (up to 2,000 cGy) for the treatment of head and neck ailments predisposes to the development of cancer of the thyroid. There is a linear dose-response relationship, and risk is inversely related to age at exposure. A number of excess cases are seen 5 to 9 years after the insult, and increased risk persists even 35 years after exposure [47].

Benign thyroid disease, specifically adenomas and goiters, are also risk factors for cancer of the thyroid. Follicular carcinomas are more common in iodine-deficient areas, while papillary histology predominates in areas with high iodine intake. Medullary thyroid carcinoma is familial and occurs as part of multiple endocrine neoplasia II. The gene for this disease has been mapped to chromosome 10 [47].

Cancer of the Skin

Nonmelanoma skin cancer, the most common malignancy in the United States [48], refers collectively to basal-cell carcinoma and squamous-cell carcinoma of the skin. The incidence rates for nonmelanoma skin cancer, particularly squamous-cell carcinoma, increase with age. Men are affected twice as often as women [49].

Nonmelanoma lesions tend to develop at sites of prior inflammation or scars. Environmental exposure to arsenic or radiation, prior therapy with psoralen plus ultraviolet A light, infection with human papillomavirus, and immune suppression may predispose to these malignancies [48]. Cigarette smoking increases the risk for squamous-cell carcinoma [49].

Ultraviolet radiation is the most important risk factor. It accounts for most cases and interacts with other factors as an etiologic agent [48]. Squamous-cell carcinoma is thought to be related to cumulative sun exposure, whereas basal-cell carcinoma is related to intermittent exposure, particularly before age 40 [49]. The risk of nonmelanoma and melanoma skin cancers is higher for whites and for people with poor tanning ability or a tendency to sunburn, fair skin, red or blond hair, and blue eyes [49,50].

Individuals with nonmelanoma skin cancer are at increased risk for developing new primary lesions [48]. The risk remains stable over time and increases with the number of skin cancers diagnosed. New lesions tend to be of the same histology as the original tumor. These patients may be at increased risk for cutaneous melanoma [49].

The incidence of cutaneous malignant melanoma has been increasing. Australia has the highest incidence rate worldwide. The incidence is highest among persons in high socioeconomic classes. Having pale skin or a large number of melanocytic nevi and freckles, sunburn from intermittent intense solar ultraviolet irradiation (particularly early in life), and living near the equator are risk factors for cutaneous melanoma [50,51]. Ultraviolet irradiation from sunlamps or sunbeds increases the risk for melanoma as well [51].

Patients treated with cytotoxic therapy and those with transplant-associated immunosuppression or xeroderma pigmentosum are at increased risk for melanoma [50,51]. Patients with albinism and those previously exposed to psoralen and ultraviolet A radiation are at high risk for nonmelanoma skin cancer, but the risk for melanoma is not significantly increased [51].

There is a familial tendency toward dysplastic nevi. It is inherited in an autosomal dominant fashion with high penetrance, located in chromosome 1p36 [50].


Leukemias are a diverse group of hematologic malignancies. In general, the incidence rates of leukemias decreased slightly between 1973 and 1986. Leukemias tend to afflict more men than women and more whites than nonwhites. Acute myelogenous leukemia (AML) usually occurs after age 40, whereas acute lymphocytic leukemia is common during childhood, and its rates increase again after about age 60. After 1970, mortality rates for leukemias leveled off [52].

Exposure to benzene has been etiologically linked to leukemias; such exposure is estimated to increase leukemia risk 2 to 4.5 times. Workers in the rubber and shoe leather industries are exposed to benzene, which was a greater hazard before 1970, when occupational safety standards were implemented [52].

Radiation exposure at moderate to high doses increases the risk for leukemia. The effect of ionizing radiation was determined by studies of atomic bomb survivors and patients irradiated for medical purposes [52]. The increased risk for leukemia begins 2 to 4 years after exposure, peaks at 6 to 8 years, and declines to normal within 25 years [22]. Development of chronic lymphocytic leukemia (CLL) is not influenced by exposure to ionizing radiation [52].

A small increase in leukemia risk has been noted for residents living near power plants. Childhood leukemia, but not adult leukemia, has been linked to exposure to electromagnetic fields. However, it seems that the actual wiring configuration is a better indirect measure of exposure than is residency near power plants. Cigarette smoking increases the risk for AML by 1.5- to twofold [52].

Cytotoxic therapy increases the risk for developing a secondary leukemia, usually AML or a dysplastic syndrome. Older patients may be at increased risk, and risk decreases 10 years after treatment. Prolonged therapy with epipodophyllotoxins increases the risk, which is dependent on dose and schedule of administration. The latency period for acute leukemias related to epipodophyllotoxin therapy is shorter than for those associated with alkylating therapy. While abnormalities at chromosome 11q23 have been noted in patients treated with epipodophyllotoxins, exposure to alkylating agents has been related to abnormalities on chromosomes 5 and 7. Overall, 70% to 90% of patients with secondary leukemias related to prior cytotoxic chemotherapy display clonal aberrations and chromosomal deletions, including del 5, del 7, del 5q, and del 7q. Another cytogenetic abnormality related to prior therapy is acquired monosomy 7 [53].

Infection with human T-lymphotropic virus type I accounts for the high incidence of adult T-cell leukemia and lymphoma in areas of Japan and the Caribbean [22].

Patients with aplastic anemia are at increased risk for leukemia. A family history of NHL, Hodgkin's disease, or CLL increases the risk as much as five times [53]. Genetic susceptibility for leukemia is seen in patients with Down's syndrome, autosomal recessive syndromes with chromosomal instability such as Bloom's and Fanconi's anemia, and ataxia telangiectasia [22].


Hodgkin's disease is the most common malignancy in young adulthood (ages 15 to 24 years). The incidence rates of lymphomas have declined for Americans age 65 and older [3]. The disease is more common in males than in females; reproductive and hormonal factors may have a protective effect against Hodgkin's disease among females [54]. High socioeconomic status, white race, and a family history of Hodgkin's disease increase the predisposing risk. The risk also is higher in families with few children and for individuals born earliest within the family. Epstein-Barr virus is found in approximately half of affected patients.

In the United States, the incidence of NHL has increased approximately 60% in the past two decades. Although acquired immunodeficiency syndrome (AIDS) has contributed to this increase, it is not solely responsible. The patients with AIDS who develop NHL are usually young men, whereas an overall increased incidence of NHL has been noted in elderly people of both genders. Among patients with NHL, the survival rates are better for whites and females [55].

Patients treated with cytotoxic drugs, particularly alkylating agents and ionizing radiation for a prior neoplasia, have a three- to ninefold increased risk for NHL. The latency period is about 5 to 6 years [55].

Environmental exposures that may cause NHL include exposure to pesticides, which particularly increase the risk for intermediate-grade NHL. Workers exposed to organic solvents may have a three times greater risk for NHL, and prolonged exposure increases that risk. Exposure to wood and cotton dust also predispose to the development of NHL. Use of hair dyes, especially long duration of use and young age at first use, increases risk as well. Smoking may also be a predisposing factor [55]. Agricultural laborers are exposed to oncogenic animal viruses that may be linked to NHL [56].

Infection with H pylori is associated with a sixfold increased risk for gastric NHL [55]. Epstein-Barr virus infection is strongly associated with Burkitt's lymphoma in areas where the latter is endemic. Concurrent malarial infection in patients with the African form of Burkitt's lymphoma induces an immunodeficiency state that potentiates the oncogenic effect of Epstein-Barr virus [22].

Immunodeficiency syndromes, autoimmune diseases with persistent antigenic stimulation, organ transplantation, and immunosuppressive therapy with azathioprine or cyclosporine are associated with increased risk for developing NHL [22,56].

Clinical Epidemiology

Epidemiologic research plays an important role in the development of cancer screening modalities and prevention strategies. Cancer prevention focuses on decreasing incidence by lowering risk through changes in lifestyle patterns and behavior. Primary prevention attempts to stop the development of cancer. Secondary prevention aims to improve cure rates by cancer screening and early diagnosis and treatment.

Cancer Screening

Cancer screening involves testing to detect early-stage cancer in asymptomatic individuals. Ideally, screening tests should be easily administered, noninvasive, and inexpensive. To be beneficial, early detection should alter prognosis and improve survival.

Cervical Cancer: Cytologic screening for cervical cancer by Pap smears has had a major impact on the mortality rates for this malignancy. Successful screening relies on the detection of preinvasive lesions. When cervical cancer is diagnosed in situ, the cure rate is about 99% [39]. The current recommendation is that screening for cervical cancer (annual Pap smears) should start at age 18, or earlier in sexually active women. Examinations can be performed less frequently after three consecutive exams are deemed normal by a physician and can be discontinued at age 65 if findings were previously normal [39]. Screening at intervals of 2 years offers the same protection as annual exams, but intervals of longer than 2 years between screenings are associated with an increased risk for invasive cervical cancer [57].

Breast Cancer: The combination of mammographic screening and clinical breast examination may reduce mortality rates from breast cancer by 30% to 40% in women over 50 years old. Nevertheless, guidelines for breast cancer screening vary. The American Cancer Society and the National Cancer Institute recommend annual clinical breast exams starting at age 40 and mammography every 1 to 2 years until age 50. Women older than 50 should have mammograms every year. The American College of Physicians, the American College of Surgeons, and the US Preventive Services Task Force recommend that mammography screening begin at age 50 and be performed at 1- to 2-year intervals until age 75 and more frequently if any abnormalities are diagnosed. Earlier breast cancer screening is advisable for those with increased risk for the disease [39].

Colorectal Cancer: The goal of screening tests for colorectal cancer is to detect adenomatous polyps that might become invasive so that these can be removed, thus improving overall survival through the detection of early-stage disease. Fecal occult blood testing has reduced mortality from colorectal cancer by 38% [58], and having a screening flexible sigmoidoscopic examination at least once every 10 years may decrease an individual's risk of death from rectal or distal colon cancer by 60% to 70% [59,60].

Prostate Cancer: Digital rectal examination has been the traditional method of detecting abnormal areas in the prostate gland, but when used as a screening modality for prostate cancer, its effectiveness in diagnosing tumors confined to the prostate is uncertain [61]. Digital rectal examination plus measurement of serum prostate-specific antigen concentrations has been shown to enhance the detection rate for prostate cancer, but an optimal screening strategy for this disease has not yet been developed [62]. Transrectal ultrasonography and guided biopsy may be used in cases with abnormal findings [63].

Other Malignancies: The efficacy of screening programs for other malignancies has not been determined. Screening strategies for other cancer sites are under investigation. Special attention must be paid to patients who have had a malignancy and who may, therefore, need screening for second malignancies at other sites [64].


A relatively new approach to cancer prevention is under investigation through chemoprevention trials. Cancer chemoprevention is defined as the reversal of carcinogenesis in the premalignant phase [65]. The observation that retinoids, acting as modulators of cell differentiation, are effective in suppressing oral carcinogenesis and, therefore, in preventing second primary tumors in squamous-cell carcinoma of the head and neck has led to the evaluation of these agents as chemopreventive therapy for tumors of the upper aerodigestive tract in high-risk populations [65,66]. Studies of adjuvant hormonal therapy with tamoxifen for breast cancer have shown a 50% reduction of contralateral disease [37]. A national tamoxifen chemoprevention trial is being conducted to evaluate risk reduction for primary breast cancer in women at high risk [67]. With the development of new molecular techniques, chemoprevention trials will be aided by the identification of markers for premalignant lesions.



1. Hennekens CH, Buring JE: Epidemiology in Medicine, pp 3–15. Boston, Little Brown & Co, 1987.

2. Wingo PA, Tong T, Bolden S: Cancer statistics, 1995. CA Cancer J Clin 45:8–30, 1995.

3. Miller BA, Ries LAG, Hankey BF, et al (eds): SEER Cancer Statistics Review: 1973–1990. Bethesda, Maryland, National Cancer Institute. NIH Publication No. 93–2789, 1993.

4. Spitz MR: Epidemiology and risk factors for head and neck cancer. Semin Oncol 21(3):281–288, 1994.

5. Blot WJ, McLaughlin JK, Winn DM, et al: Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res 48: 3282–3287, 1988.

6. Tobacco, in Tomatis L, Aitio A, Day NE, et al (eds): Cancer: Causes, Occurrence and Control, pp 169–180. Lyon, International Agency for Research on Cancer, IARC Publication No. 100, 1990.

7. Caplan GA, Brigham BA: Marijuana smoking and carcinoma of the tongue: Is there an association? Cancer 66:1005–1006, 1990.

8. Block G: Vitamin C status and cancer: Epidemiologic evidence of reduced risk. Ann NY Acad Sci 669:280–290, 1992.

9. Marshall J, Graham S, Mettlin C, et al: Diet in the epidemiology of oral cancer. Nutr Cancer 3:145–149, 1982.

10. Mackerras D, Buffler PA, Randall DE, et al: Carotene intake and the risk of laryngeal cancer in coastal Texas. Am J Epidemiol 128:980–988, 1988.

11. Beckett WS: Epidemiology and etiology of lung cancer. Clin Chest Med 14(1):1–15, 1993.

12. Szabo E, Mulshine J: Epidemiology, prognostic factors, and prevention of lung cancer. Curr Opin Oncol 5(2):302–309, 1993.

13. Samet JM: The epidemiology of lung cancer. Chest 103(suppl 1):20–29S, 1993.

14. Muñoz N: Epidemiological aspects of oesophageal cancer. Endoscopy 25(9):609–612, 1993.

15. Mayer RJ: Overview: the changing nature of esophageal cancer. Chest 103(suppl 4):404–405S, 1993.

16. Cameron AJ: Epidemiologic studies and the development of Barrett's esophagus. Endoscopy 25(9):635–636, 1993.

17. Parsonnet J, Friedman GD, Vandersteen DP, et al: Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 325:1127–1131, 1991.

18. Stockbrugger RW: Epidemiology and pathology of precancerous lesions of the stomach. Eur J Cancer Prev 2(suppl 2):59–63, 1993.

19. Viruses and other biological agents, in Tomatis L, Aitio A, Day NE, et al (eds): Cancer: Causes, Occurrence and Control, pp 184–200. Lyon, International Agency for Research on Cancer, IARC Publication No. 100, 1990.

20. Beasley RP, Hwang LY, Lin CC, et al: Hepatocellular carcinoma and hepatitis B virus. Lancet 2:1129–1133, 1981.

21. Kiyosawa K, Furuta S: Clinical aspects and epidemiology of hepatitis B and C viruses in hepatocellular carcinoma in Japan. Cancer Chemother Pharmacol 31(suppl 1):S150–156, 1992.

22. Fraumeni JF Jr, Devesa SS, Hoover RN, et al: Epidemiology of cancer, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, pp 150–181. Philadelphia, JB Lippincott, 1993.

23. DeCosse JJ, Tsioulias GJ, Jacobson JS: Colorectal cancer: Detection, treatment, and rehabilitation. CA Cancer J Clin 44: 27–42, 1994.

24. Willett WC: Micronutrients and cancer risk. Am J Clin Nutr 59 (suppl):1162–1165S, 1994.

25. Rustgi AK: Hereditary gastrointestinal polyposis and nonpolyposis syndromes. N Engl J Med 331:1694–1702, 1994.

26. McCredie M: Epidemiology of kidney cancer in Australia. Med J Aust 157(8):508–510, 1992.

27. Cohen SM, Johansson SL: Epidemiology and etiology of bladder cancer. Urol Clin North Am 19(3):421–428, 1992.

28. Karr JP: Prostate cancer in the United States and Japan. Adv Exp Med Biol 324:17–28, 1992.

29. Burks DA, Littleton RH: The epidemiology of prostate cancer in black men. Henry Ford Hosp Med J 40(1–2):89–92, 1992.

30. Talamini R, Franceschi S, LaVecchia C, et al: Smoking habits and prostate cancer: A case-control study in northern Italy. Prev Med 22(3):400–408, 1993.

31. Li FP: Molecular epidemiology studies of cancer in families. Br J Cancer 68:217–219, 1993.

32. Herbst AL: The epidemiology of ovarian carcinoma and the current status of tumor markers to detect disease. Am J Obstet Gynecol 170(4):1099–1105, 1994.

33. Kerlikowske K, Brown JS, Grady DG: Should women with familial ovarian cancer undergo prophylactic oophorectomy? Obstet Gynecol 80:700–707, 1992.

34. Bewtra C, Watson P, Conway T, et al: Hereditary ovarian cancer: A clinicopathological study. Int J Gynecol Pathol 11(3):180–187, 1992.

35. Kelsey JL, Whittemore AS: Epidemiology and primary prevention of cancers of the breast, endometrium, and ovary. A brief overview. Ann Epidemiol 4(2):89–95, 1994.

36. Miki Y, Swensen J, Shattuck-Eidens D, et al: A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266:66–71, 1994.

37. Horn-Ross PL: Multiple primary cancers involving the breast. Epidemiol Rev 15(1):169–176, 1993.

38. Fisher B, Costantino JP, Redmond CK, et al: Endometrial cancer in tamoxifen–treated breast cancer patients: Findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. J Natl Cancer Inst 86:527–537, 1994.

39. The National Strategic Plan for the early detection and control of breast and cervical cancers, pp 5–9. Washington, DC, US Department of Health and Human Services, 1993.

40. Brinton LA: Epidemiology of cervical cancer-overview, in Muñoz N, Bosch FX, Shah KV, et al (eds): The Epidemiology of Human Papillomavirus and Cervical Cancer, pp 3–23. Lyon, International Agency for Research on Cancer, IARC Scientific Publication No. 119, 1992.

41. Muñoz N, Bosch FX: HPV and cervical neoplasia: Review of case-control and cohort studies, in Muñoz N, Bosch FX, Shah KV, et al (eds): The Epidemiology of Human Papillomavirus and Cervical Cancer, pp 251–261. Lyon, International Agency for Research on Cancer, IARC Scientific Publication No. 119, 1992.

42. Miller BA, Feuer EJ, Hankey BF: The significance of the rising incidence of breast cancer in the United States, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Important Advances in Oncology 1994, pp 193–207. Philadelphia, JB Lippincott, 1994.

43. Bodian CA: Benign breast diseases, carcinoma in situ, and breast cancer risk. Epidemiol Rev 15(1):177–187, 1993.

44. Vandenbroucke A, Bourdon C: Epidemiological survey of preinvasive breast cancer. Eur J Cancer Prev 2(suppl 3):3–10, 1993.

45. Early Breast Cancer Trialists' Collaborative Group: Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. Lancet 339:1–15, 1992.

46. Easton DF, Bishop DT, Ford D, et al: Genetic linkage analysis in familial breast and ovarian cancer: Results from 214 families. Am J Hum Genet 52:678–701, 1993.

47. Franceschi S, Boyle P, Maisonneuve P, et al: The epidemiology of thyroid carcinoma. Crit Rev Oncog 4(1):25–52, 1993.

48. Weinstock MA: Epidemiology of nonmelanoma skin cancer: Clinical issues, definitions, and classification. J Invest Dermatol 102(6):4–5S, 1994.

49. Karagas MR: Occurrence of cutaneous basal cell and squamous cell malignancies among those with a prior history of skin cancer. J Invest Dermatol 102(6):10S–13S, 1994.

50. Franceschi S, Cristofolini M: Cutaneous malignant melanoma: Epidemiological considerations. Semin Surg Oncol 8(6):345–352, 1992.

51. Elwood JM: Recent developments in melanoma epidemiology, 1993. Melanoma Res 3(3):149–156, 1993.

52. Sandler DP: Epidemiology and etiology of acute leukemia: An update. Leukemia 6 (suppl 4):3–5, 1992.

53. Vogel VG, Fisher RE: Epidemiology and etiology of leukemia. Curr Opin Oncol 5(1):26– 34, 1993.

54. Glaser SL: Reproductive factors in Hodgkin's disease in women: A review. Am J Epidemiol 139(3):237–246, 1994.

55. Weisenburger DD: Epidemiology of non-Hodgkin's lymphoma: Recent findings regarding an emerging epidemic. Ann Oncol 5 (suppl 1):19–24, 1994.

56. Pearce N, Bethwaite P: Increasing incidence of non-Hodgkin's lymphoma: Occupational and environmental factors. Cancer Res 52 (suppl 19):5496–5500S, 1992.

57. Shy K, Chu J, Mandelson M, et al: Papanicolaou smear screening interval and risk of cervical cancer. Obstet Gynecol 74:838–843, 1989.

58. Mandel JS, Bond JH, Church TR, et al: Reducing mortality from colorectal cancer by screening for fecal occult blood. N Engl J Med 328:1365–1371, 1993.

59. Selby JV, Friedman GD, Quesenberry CP, et al: A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 326:653–657, 1992.

60. Levin B: Screening sigmoidoscopy for colorectal cancer. N Engl J Med 326:700–702, 1992.

61. Friedman GD, Hiatt RA, Quesenberry CP, et al: Case-control study of screening for prostatic cancer by digital rectal examinations. Lancet 337:1526–1529, 1991.

62. Krahn MD, Mahoney JE, Eckman MH, et al: Screening for prostate cancer. A decision analytic view. JAMA 272:773–780, 1994.

63. Catalona WJ, Smith DS, Ratliff TL, et al: Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 324:1156–1161, 1991.

64. Offit K, Brown K: Quantitating familial cancer risk: A resource for clinical oncologists. J Clin Oncol 12:1724–1736, 1994.

65. Lippman SM, Hong WK: Retinoid chemoprevention of upper aerodigestive tract carcinogenesis, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Important Advances in Oncology 1992, pp 93–109. Philadelphia, JB Lippincott, 1992.

66. Hong WK, Lippman SM, Itri LM, et al: Prevention of second primary tumors with isotretinoin in squamous–cell carcinoma of the head and neck. N Engl J Med 323:795–801, 1990.

67. Nayfield SG, Karp JE, Ford LG, et al: Potential role of tamoxifen in prevention of breast cancer. J Natl Cancer Inst 83: 1450–1459, 1991.