Cancer epidemiology provides the tools and methods to understand the cancer problem in any given population, from the local level up to the global level. Incidence, prevalence, and mortality are the most commonly used measures to assess the cancer burden. Examining temporal trends in these measures or comparison of these measures between states, regions, or countries can uncover important causes of cancer. The study of cancer from an epidemiological perspective has uncovered numerous causes of cancer and has hence paved the way towards prevention and early detection. Perhaps the most well-known accomplishment of cancer epidemiology was the identification of tobacco as a cause of lung cancer in 1964. This finding revolutionized our understanding of cancer, as it was the first time that a common, modifiable behavior—tobacco smoking—was shown to result in cancer, and subsequently led to the development and implementation of a wide range of preventive measures throughout the second half of the 20th century that have demonstrably reduced both the use of tobacco and deaths from lung cancer. , This chapter reviews and summarizes the latest data on the most basic measures of the global cancer burden and then briefly describes cancer prevention and early detection recommendations, as well as survivorship care.
Cancer incidence and mortality are defined as the number of new cancer cases and deaths, respectively, that occur in a given population over a specified time period. The selection of the time period is arbitrary, although often these measures are expressed as an absolute number of new cases or deaths per year. While this may be helpful for planning health services in a given population, this simplified expression does not provide risk information, and it does not allow for comparisons of incidence and mortality to be made among different populations. For these uses, incidence and mortality are generally expressed as a rate or proportion of the number of new cancer cases or deaths over the number of persons at risk of developing or dying from cancer during a specified time period per 1000, 10,000, or 100,000 individuals. Often, incidence and mortality will be reported as an “age-standardized rate” (ASR) to facilitate comparisons among populations that have different age distributions.
The number of new cancer cases and deaths are captured by population-based cancer registries oriented toward a geographic or geopolitical area, such as a country, although these registries typically capture just a small proportion of the global population. Coverage also varies by country. Population coverage is typically greater for mortality than for incidence. Sometimes incidence and mortality rates are estimated in cohort studies. Cancer incidence and reporting can be influenced by the screening practices, diagnostic intensity, and primary prevention programs in the population under study. Cancer mortality data are influenced by the adequacy of death certification, including autopsy rates, by changes in cancer treatment effectiveness, and by the availability of prevention programs in the population under study.
Cancer prevalence is defined as the number of cancer cases in a population at a specific point in time over the number of persons in the population at that time point. Unlike the incidence rate, prevalence is not a measure of cancer risk. Nevertheless, it can be useful for planning health services. Ascertaining the prevalence of cancer can be done through population-based cancer registration or estimated from cross-sectional studies. Determinants of cancer prevalence include the incidence and prognosis of the cancer in question, as well as mortality from other competing causes.
Cancer survival is defined as the proportion of cancer patients surviving for a specified time after diagnosis. Despite numerous limitations, it is considered the best available measure for evaluating the effectiveness of cancer treatments. Survival is influenced by the natural history of the disease, the stage at diagnosis, and therapeutic efficacy. Survival data require long-term follow up of large number of patients, are sensitive to both the misclassification of the cause of death and to lead-time bias, and provide no insight into the quality of life lived. There are various measures of survival, each serving a different purpose and each with its own limitations. Observed survival is the probability of surviving for a specific time period, generally starting at the date of cancer diagnosis, and considers all causes of death. Corrected or cause-specific cancer survival excludes deaths due to causes other than the cancer of interest and therefore is a more valid estimate of the excess death due to a cancer. Relative cancer survival compares the observed survival of a group of cancer patients to the expected survival of a group from the general population with the same age and sex distribution.
Cancer Burden in the World
Cancer is a leading cause of death around the world. In 2018, there were an estimated 18.1 million (including non-melanoma skin cancer) new cancer cases and 9.6 million (including non-melanoma skin cancer) cancer deaths. Fig. 1.1 illustrates the distribution of these cases and deaths among the 10 most common cancers in 2018. Lung cancer is the most commonly diagnosed cancer worldwide, accounting for nearly 12% of all cancer cases in men and women combined. Unsurprisingly, it is also the leading cause of global cancer mortality, accounting for just over 18% of all cancer deaths. Lung, breast, colorectal, and prostate are the top four most common cancers, accounting for a combined 40.5% of all new cases worldwide. Lung, colorectal, stomach, and liver are the top four causes of cancer mortality, accounting for a combined 44% of global cancer deaths. Of note is the significant burden of liver and stomach cancer mortality among men. Also notable, cervical cancer, which is almost entirely preventable through cervical screening and HPV vaccination, remains a leading cause of cancer and cancer death among women.
Worldwide, men experience higher cancer incidence and mortality than women. Incidence for all cancers combined is 20% higher in men (218.6 vs. 182.6 per 100,000), and all-site mortality is 50% higher in men (122.7 vs. 83.1 per 100,000). There is substantial variation in these rates by world region. This variation reflects differences in risk factor exposures and in the availability of preventive, early detection, and treatment resources that exist across populations. In turn, risk factor exposures and access to treatment and resources are driven by a country’s level of socioeconomic development, or its human development index (HDI), a composite measure of income, education, and life expectancy. Both incidence and mortality are higher in high-income countries than in lower-income countries. Furthermore, in countries with a low HDI, infection-related cancers, such as cervical, stomach, and liver predominate. As a country transitions to higher levels of HDI, infection-related cancers typically decrease and lifestyle-related cancers, such as breast and colorectal, emerge.
Figs. 1.2 and 1.3 highlight the variation in cancer incidence and mortality around the world. A few patterns can be discerned from these maps. First, cancer incidence varies more for men than for women. Breast cancer was the top cancer in women in at least 154 countries, while cervical cancer was the most common cancer in 28 countries, mainly those in sub-Saharan Africa. Among men, there are 10 different cancers that claim the top spot across 185 countries. Second, cancer mortality varies more than cancer incidence. Among women, cervical cancer is the leading cause of cancer mortality in sub-Saharan Africa, while lung cancer leads cancer mortality in more developed countries, such as the United States and Canada, northern and eastern Europe, Australia, and China. Stomach cancer mortality predominates in parts of South America and liver cancer mortality predominates in Mongolia, Cambodia, and Guatemala. Among men, stomach cancer is the leading cause of cancer mortality in the Middle East and parts of South America, while liver cancer is the leading cause of cancer mortality in Egypt, parts of western Africa and southeast Asia. Deaths due to lip and oral cancer are the most common type of cancer mortality in India and Pakistan, and Kaposi sarcoma and leukemia lead cancer mortality in eastern and southern Africa.
Data from both high- and low-income countries suggest that mortality from cancer has decreased globally by approximately 1% annually between 2000 and 2010. Data from various regional studies and from Globocan confirm this. While declines have been observed in most countries for the most common cancers (lung [men only], breast, prostate, stomach, colorectal, and uterine), mortality rates are rising for liver cancer, and lung cancer in women, in many countries.
Projections from Globocan indicate that cancer incidence and mortality will rise rapidly worldwide over the next few decades due to the aging and growth of the population. By 2040, it is projected that there will be nearly 30 million new cases of cancer, with nearly 17 million deaths, occurring annually. Yet, many of these cases and deaths are preventable with knowledge we already have. However, additional investment in both research and cancer control, particularly in low- and middle-income countries, is urgently needed to address the current and even greater future burden of cancer.
Cancer Prevention and Early Detection
An estimated one-third to one-half of cancer occurring today in western (or “westernized”) populations is preventable through adoption of healthy lifestyles, including avoidance of known risk factors and adherence to screening recommendations.
Established modifiable lifestyle risk factors for cancer around the world are tobacco (including exposure to secondhand smoke); excess body weight; alcohol intake; consumption of red and processed meats; low consumption of fruits, vegetables, dietary fiber, and dietary calcium; physical inactivity; ultraviolet radiation (including tanning beds); and six cancer-associated infections (HIV, HPV, HBV, HCV, H. pylori , and HHV8). Recent data analyzed and published by the American Cancer Society suggest that as much as 42% of cancer cases and 45% of cancer deaths in the United States are attributable to the above modifiable risk factors. While the proportions of cancer cases and deaths attributable to any one or a combination of risk factors vary around the world as the prevalence of risk factors varies geographically, a significant proportion of cancer can be prevented through personal lifestyle choices.
Despite the well-established harms of tobacco, it continues to be the predominant cancer risk factor throughout much of the world. In 2017, tobacco accounted for nearly one-quarter of all cancer deaths globally. Infectious agents also remain an important cause of cancer in many parts of the world. HPV in sub-Saharan Africa is a particular problem, where its prevalence is estimated at 21% among women. Obesity, already an epidemic in high-income countries, is rapidly growing in prevalence around the world as low- and middle-income countries transition to higher levels of economic development and increasingly adopt “westernized” lifestyles.
Evidence-based comprehensive tobacco control strategies have been shown to reduce the prevalence of tobacco use and lead to reductions in the incidence and mortality of lung cancer. Such strategies include antitobacco mass media campaigns, tobacco taxation, marketing bans and plain packaging, smoke-free policies to protect nonsmokers, and promoting and ensuring access to cessation resources (e.g., through national quit lines). Many middle- and low-income countries struggle to fund and implement comprehensive tobacco control programs. The World Health Organization’s Framework Convention on Tobacco Control (FCTC) and MPOWER measures provide a foundation for these countries to implement effective tobacco control interventions. Approximately 63% of the world’s population is now covered by at least one MPOWER tobacco control measure. Newer tobacco products, such as electronic cigarettes (i.e., e-cigs), will need to be evaluated for their short- and long-term health effects but may represent a threat to decades of progress on controlling the tobacco epidemic. Prophylaxis against cancer-associated infectious agents, when available, can lead to significantly improved cancer outcomes. Administration of the HPV vaccine through national programs has resulted in significantly reduced risks of precancerous cervical lesions. Unfortunately, uptake of the vaccine is extremely low in most countries where the infection is highly prevalent. And while obesity is rising around the globe, there are few evidence-based strategies to mitigate its negative health effects, including cancer. Eating a plant-based diet, limiting sedentary activities, and moving more are important ways to maintain a healthy weight throughout life.
Both the American Cancer Society (ACS) and the World Cancer Research Fund (WCRF) have published cancer prevention recommendations for nutrition and physical activity that are based on regularly updated reviews of the evidence ( Table 1.1 ). These recommendations are in addition to avoiding tobacco, preventing overexposure to UV radiation, and adhering to age-appropriate cancer screening and vaccination schedules. Large observational studies , and a systematic review indicate that following the ACS or WCRF cancer prevention recommendations in a comprehensive manner significantly reduces the risks of developing and dying from cancer. Because lifestyle choices are influenced by one’s background and surrounding physical and social environment, implementation of evidence-based cancer control strategies that seek to make the healthy choice the default choice through policy, education, and delivery of community-based clinical services is greatly needed to prevent more cancers.
|Achieve and maintain a healthy weight throughout life.||Be a healthy weight.|
|Be physically active (Adults: 150-300 minutes of moderate-intensity or 75-150 minutes of vigorous-intensity activity each week. Reaching or exceeding the upper limit of 300 minutes is ideal).||Be physically active.|
|Follow a healthy eating pattern at all ages. (Include foods that are high in nutrients, a variety of vegetables and fruits, and whole grains. Limit red and processed meats, sugar-sweetened beverages, highly processed foods and refined grain products).||Eat a diet rich in wholegrains, vegetables, fruit and beans. Limit consumption of red and processed meats, fast foods, other processed foods high in fat, starches or sugars. Limit consumption of sugar sweetened drinks.|
|It is best not to drink alcohol. (People who choose to drink alcohol should have no more than 1 drink per day for women or 2 drinks per day for men).||Limit alcohol consumption.|
|Do not use supplements for cancer prevention.|
|For mothers: breastfeed your baby, if you can.|
|After a cancer diagnosis, follow each of these recommendations regarding healthy lifestyles, if you can.|
The goal of cancer screening is to detect cancer early, before symptoms occur, when it is more easily treated. Wilson and Junger first published several criteria for a screening test to be useful. The three primary criteria are (1) the test must detect the disease earlier than routine methods, (2) earlier treatment must lead to improved outcomes, and (3) the benefits of screening must be greater than the risks of any subsequent diagnostic and therapeutic treatments. Because screening programs are resource-intensive and require extensive health services infrastructure, they should only be undertaken when their efficacy has been clearly demonstrated, optimally through well-designed randomized controlled trials (RCTs). When RCTs are not available or feasible, as in the case of colonoscopies for colorectal cancer screening, observational data and/or meta-analyses are often relied upon as evidence of screening effectiveness. The ideal measure of the efficacy of a screening test is a reduction in cancer mortality among those being screened in comparison with those who do not undergo screening. Evaluations of screening tests may suffer from self-selection bias (i.e., the tendency for healthier individuals to attend screening exams compared with those who are less healthy), lead time bias (i.e., the perceived increase in survival time attributed to a screening test which accelerates the time of diagnosis compared with the onset of a symptomatic presentation, but without actually lengthening the time to cancer mortality), and length time bias (i.e., slower-growing cancers are more likely to be detected by screening than faster-growing, more aggressive cancers since they have a longer detectable preclinical phase), which can lead to over-diagnosis and over-treatment. Over-diagnosis is finding cancers that would never become clinically relevant in an individual’s lifetime, and over-treatment is the treatment of those cancers with the potential for causing important side effects. This has been an issue mainly in breast and prostate cancer screening.
Implementation of screening programs for common cancers, specifically cervical, breast, prostate, lung, and colorectal, have been shown to reduce the mortality associated with these cancers. However, implementation of these tests has been inconsistent around the world, and screening programs vary greatly among countries in the manner in which they are conducted. Some countries such as Australia, Finland, and the UK that can systematically call and recall target populations offer organized screening programs on a national level. Other countries without this ability, like the United States, France, and Germany have unorganized or opportunistic screening programs in place, which reach fractions of the recommended populations.
As lung cancer is the number one cause of cancer and cancer death worldwide, there has been long-standing interest in developing a screening test for this highly prevalent and lethal cancer. In 2011, the National Lung Screening Trial demonstrated an approximate 20% reduction in lung-cancer mortality in heavy smokers through the annual use of low-dose computed tomography (LDCT). More recently, the NELSON trial results suggest an even greater mortality benefit of 26% in high-risk men and 61% in high-risk women. However, concerns over the potentially high false-positive rate of the test provide justification for the application of informative risk prediction models to identify those who would most benefit from LDCT screening for early detection while minimizing false positives. This is an active area of research in lung cancer screening. Screening in the United States is opportunistic and rates are extremely low at 3.9% (in 2015). No country to date has implemented an organized lung cancer screening program, although many countries in Europe and east Asia have started small trials or regional demonstration projects to determine the feasibility of lung screening in their respective countries.
Screening for cervical cancer through use of the pap smear is one of the most successful tests ever offered in medicine, as observational data show that it has resulted in decreases of cervical cancer incidence and mortality on the order of 75%–90% in high-income countries where large-scale, population-based cervical screening programs have been in effect since the middle of the 20th century. Much of the developing world has yet to see such cervical screening programs implemented due to low availability of the health services resources needed for screening, follow up, and subsequent treatment for screen positives. Hence, cervical cancer continues to be a leading cause of cancer and cancer death in areas such as sub-Saharan Africa. The HPV test is beginning to replace the pap smear for cervical screening. Several RCTs demonstrate that HPV testing is superior to pap testing in reducing incidence of cervical precancers and cancers. Although not yet adopted in the United States, primary HPV testing for cervical screening has already been implemented in some countries such as Australia and the Netherlands or is in the process of being implemented in others throughout Europe. Visual inspection with ascetic acid (VIA) was shown to result in a 31% mortality reduction among screened women in India in a large RCT and can be offered in low-resource settings when primary HPV testing is unavailable. HPV vaccination is an important component of cervical cancer prevention, but does not remove the need for routine cervical cancer screening according to age and clinical history.
Systematic reviews of RCTs of screening mammography have demonstrated an approximate 20% mortality reduction in women ages 40–74 years, , although most of the RCTs were initiated prior to or in the early 1980s and so are limited by the quality of mammography available at the time the trials were conducted, with many cancers having to be 1 cm or greater to be detected. Observational data are believed to better reflect contemporary screening practices in which the detection of sub-centimeter tumors is commonplace. These data suggest a 48% reduction in breast cancer mortality with modern screening mammography. This is supported by modeling studies that estimate a 29%–54% mortality reduction. Although the age to start screening (40 vs. 50 years) and the frequency with which to screen (annual vs. biennial) may differ by the organization making the recommendation, all agree that there is a mortality benefit to beginning screening at age 40 years and that annual screening results in fewer breast cancer deaths than does biennial screening. Large-scale breast cancer screening programs relying upon mammography are again limited to higher-income countries in North and South America, Europe, Australia, and New Zealand. Although evidence supporting the use of clinical breast examination alone for screening is mixed, it is encouraged in specific situations in basic- and limited-resource settings.
Numerous tests exist for colorectal cancer screening. The gold standard test is colonoscopy because of its high sensitivity and specificity for the identification of adenomas and its ability to both diagnose and reduce risks associated with biopsied precancerous lesions. Although there is no evidence from RCTs supporting the use of colonoscopy, there is observational evidence from the Nurses’ Health Study and the Health Professionals Follow-up Study demonstrating a 68% reduction in colorectal cancer mortality Flexible sigmoidoscopy, while not typically offered in the United States, is offered in other countries and is supported by RCTs showing a 26%–31% reduction in colorectal cancer mortality. , Use of guaic-based stool tests is supported by several RCTs, with mortality reductions of up to 32%. Stool-based fecal immunohistochemical tests (FIT) are more accurate than the guaic-based stool tests although RCT evidence of any mortality benefit associated with their use is lacking. A FIT test that includes DNA markers is also available, but again, data supporting a mortality benefit are lacking. Two newer screening tests are virtual colonoscopy (CT [computed tomography] colonography) and a blood test that detects circulating methylated SEPT9 DNA, a marker for colorectal cancer. These tests are still evolving and the balance between benefits and harms for each of these tests remains unclear. As with cervical and breast cancer screening, large-scale colorectal cancer screening programs are limited to high-income countries.
Screening for prostate cancer with the prostate-specific antigen (PSA) test has been highly controversial and recommendations for the test have changed repeatedly. An initial RCT conducted in the United States did not suggest a mortality benefit but results were called into question when it was determined that there was a high rate of PSA testing in the control arm of the trial. , A second trial in Europe did find a significant 20% reduction in prostate-cancer mortality as a result of PSA screening. However, there was a high risk of over-diagnosis of prostate cancer in subjects lacking clinical symptoms. And while the 12- to 14-year follow up results from the Swedish and Netherlands sites of the ERSPC trial found enhanced benefit of PSA screening, the 12- to 15-year follow-up results from the Finland and Spanish sites failed to demonstrate significant effect on prostate-cancer mortality. , The latest evidence suggests that the benefits and harms of screening are more closely matched than previously thought. Consequently, most organizations now recommend individualized decision-making for men ages 55–69 years.
Going forward, genomic- and proteomic-based approaches have the potential to refine risk assessment and stratification, allowing for more tailored screening and early detection strategies. However, the use of such approaches is still in its early stages.
Cancer chemoprevention is the use of drugs, vaccines, and natural compounds to inhibit, reverse, or delay the onset of carcinogenesis. This approach may also be referred to as molecular prevention, or in the case of established precancers, cancer interception. Table 1.2 lists medications that have been FDA-approved for the treatment of precancerous lesions and/or cancer risk reduction. Almost half of the agents are for treatment of precancerous skin lesions (i.e., actinic or solar keratosis). This is due, at least in part, to the accessibility and visibility of the target organ. Tamoxifen, approved for reducing the risk of invasive breast cancer, was initially approved for use in advanced breast cancer. Its use in women at high risk of breast cancer nearly halves the risk of invasive disease. Despite this, uptake of the drug among eligible women has been poor due to concerns over potential toxicity. Celecoxib, a nonsteroidal antiinflammatory (NSAID), was approved to reduce the number of colorectal adenomas in adults at very high risk of colorectal cancer due to familial adenomatous polyposis (FAP). However, given concerns over reported cardiovascular toxicity associated with the use of celecoxib in RCTs at the time, the labeling indication was voluntarily withdrawn by Pfizer. Experiences with both tamoxifen and celecoxib highlight the importance of the balance between risks and benefits for preventive agents. Despite the early concerns with celecoxib, NSAIDs in general remain a promising class of agents for cancer chemoprevention and continue to be actively investigated for the prevention of colorectal and numerous other cancers. Aspirin is perhaps the most promising and well-studied NSAID. Aspirin has been shown to reduce the incidence and mortality of colorectal cancer in randomized trials of cardiovascular disease prevention. , While it has yet to be approved for general population use to prevent or reduce the risk of cancer, the United States Preventive Services Task Force (USPSTF) has recognized colorectal cancer prevention as a benefit of aspirin use in those aged 50–59 years who are at increased risk of cardiovascular disease. Enhancing our understanding of the type, timing, and sequence of molecular changes that underlie the development and progression of precancerous lesions, as is being facilitated by efforts to map various precancerous genomes, will help drive the identification of novel chemopreventive agents. Additionally, a reverse migration strategy, where drugs approved for treatment against established cancers are tested earlier in the carcinogenic process (as happened with tamoxifen), offers another potential pathway to making this strategy a reality. Of course, any potential chemopreventive or interceptive agent must undergo rigorous evaluation in Phase I–IV trials to define the most safe and effective dosing regimen. Potential agents for molecular cancer prevention currently under investigation include COX-2 inhibitors, retinoids, HER2 receptor kinase inhibitors, IGF inhibitors, metformin, statins, PARP inhibitors, and innovative vaccine and inflammation- and immunoprevention-based approaches.