Tim E. Byers
INTRODUCTION
Cancer incidence registries now cover nearly all of the US population. State-based vital records systems and aggregate national systems regularly report trends in both cancer incidence and mortality, and national surveys routinely monitor cancer-related risk factors in the population. These surveillance systems have documented substantial changes in both risk factors for cancer and in cancer incidence and mortality rates in the United States over the past 3 decades. In 1996, the American Cancer Society (ACS) set an ambitious challenge for the United States: to reduce cancer mortality rates from their apparent peak in 1990 by 50% in the 25-year period ending in 2015.1 In 1998, the ACS then challenged the United States to also reduce cancer incidence rates from their peak in 1992 by 25% by the year 2015.2 In this chapter, we will examine trends in cancer risk factors as well as trends in cancer incidence and mortality rates in the United States over the 25-year period between 1990 and 2015.
CANCER SURVEILLANCE SYSTEMS
Collecting cancer incidence rates is largely a state-based activity in the United States, because cancer is a reportable disease in all states. The Centers for Disease Control and Prevention (CDC) organizes all state-based cancer registries within the National Program of Cancer Registries, which now reports collective data on cancer incidence from over 40 different state-based registries, providing data that meets strict quality standards.3 The National Cancer Institute has supported high-quality cancer incidence and outcomes registration in selected states and cities since 1973 within the Surveillance, Epidemiology, and End Results (SEER) Program.4 The most precise measures of long-term trends in cancer incidence come from SEER-9, a set of nine SEER registries that together include about 10% of the US population. The populations included in the SEER-9 registries document the most detailed history of cancer trends beginning in the 1970s based on highly standardized cancer case ascertainment, staging, treatment, and outcomes. Deaths from cancer are well ascertained in all states via state-based vital records, which are aggregated into annual national mortality reports by the CDC’s National Center for Health Statistics. 5Each year, the ACS, the National Cancer Institute, and the CDC publish a Report to the Nation on trends in cancer incidence and mortality in the United States.6 Trends in the prevalence of behavioral factors that affect cancer risk are tracked by the Health Interview Survey, an ongoing, in-person interview of a nationally representative sample of adults, and in annual reports by the Behavioral Risk Factor Surveillance System, a continuously operating telephone-based survey operated by state departments of health and organized by the CDC.7
MAKING SENSE OF CANCER TRENDS
Understanding the reasons for cancer trends requires understanding trends in cancer-related risk factors. For factors like tobacco, relating trends in exposure to trends in rates is easy, because those effects are large and single. However, for many other cancer risk factors, because effects are much smaller and multifactorial, simple correlations over time are less apparent. In most situations, all that maybe possible are crude qualitative relationships between temporal trends in cancer risk factors and subsequent trends in cancer rates. Statistical methods such as linear regression joinpoint analysis can tell us when inflections in cancer trends occur, but accounting for the precise reasons for changing rates is often impaired by our incomplete knowledge about the interacting impacts of variations in cancer screening, diagnosis, and treatment, and by uncertainties about latencies between interventions and outcomes.8
TRENDS IN CANCER RISK FACTORS AND SCREENING
Trends in major cancer risk factors have been mixed (Table 12.1). Although the downward trends in tobacco smoking among adults that began in the 1960s slowed after 1990, there has been a continuing downward trend in the number of cigarettes smoked per day by continuing smokers.9 Obesity trends have been adverse among both men and women since the 1970s, with more than a doubling of the prevalence of obesity between 1990 and 2010. Long-term trends in the use of hormone replacement therapy (HRT) are not routinely monitored in the Behavioral Risk Factor Surveillance System (BRFSS), but HRT use increased substantially in the last 2 decades of the 20th century. Then, following the 2002 publication of the Women’s Health Initiative trial, which showed clear adverse effects of HRT, there was a rapid and substantial drop in HRT use.10,11 The use of endoscopic screening for colorectal cancer (sigmoidoscopy or colonoscopy) has increased substantially in recent years, approximately doubling since the mid 1990s, so that, as of 2010, about two-thirds of Americans age 50 and older reported ever having had an endoscopic examination. Mammography use increased progressively through the 1990s, but mammogram rates then leveled off after 2000.12 Widespread prostate-specific antigen (PSA) testing began in the mid to late 1980s, then increased substantially during the 1990s. By 2002, a majority of US men age 50 and older reported having been tested.
CANCER INCIDENCE AND MORTALITY
In this chapter, we describe and discuss cancer trends for the time period 1990 through 2010 using cancer incidence data from the SEER-9 registry (Table 12.2 and Fig. 12.1) and US cancer mortality data from the National Center for Health Statistics (Table 12.3).4,5 All rates were age-adjusted to the US 2000 standard population by the direct method, using 10-year age intervals.
Lung Cancer
The lung is the second leading site for cancer incidence and the leading site for cancer death among both men and women in the United States.6 There are now more deaths from lung cancer in the United States than from the sum of colorectal, breast, and prostate cancers. Trends in lung cancer incidence and mortality have been nearly identical because there are few effective treatments for lung cancer, and survival time remains short. Lung cancer trends follow historic declines in tobacco use, lagged by about 20 years.13 Between 1965 and 1985, tobacco use among US adults dropped substantially, and more in men than in women. Lung cancer mortality rates began to decline among men in 1990, but rates increased among women throughout the 1990s. The stabilization of lung cancer incidence trends among women from 2000 to 2005 and the beginning of a decline in the period 2005 to 2010 foretells a coming persistent decline in lung cancer mortality among women in the United States.
The effectiveness of annual examinations by use of chest radiographs in reducing lung cancer mortality was studied as part of the Prostate, Lung, Colorectal, Ovary (PLCO) trial, and the effectiveness of annual screening by low-dose computed tomography (LDCT) of the lung fields was studied in the National Lung Screening Trial (NLST).14,15 In brief, screening with standard chest radiography finds more cancers earlier but does not affect mortality, whereas screening with LDCT reduces the risk of death from lung cancer by at least 20%.14,15 Therefore, both the ACS and the US Preventive Services Task Force have issued recommendations that favor informed decision making for lung cancer screening using LDCT.16,17
The major factor that will determine lung cancer incidence in the coming decade is the past history of tobacco use, but future screening will also reduce future mortality rates. Considering all factors, it is likely that over the coming decade the downward trends in mortality from lung cancer will continue at about the same rate among men, and soon will become more apparent among women.
Colorectal Cancer
The colorectum is the third leading site for cancer incidence and the second leading site for cancer death in the United States.6 Colorectal cancer incidence rates increased until 1985, when they began to decline. The reasons for this decline are not clear, but could be related to downward trends in cigarette smoking and the increasing use of both nonsteroidal anti-inflammatory drugs (NSAIDs) and HRT.18 The rapid decline in HRT use following the publication of the Women’s Health Initiative trial results in 2002 may adversely affect colorectal trends among women in the coming years, because HRT reduces the risk for colorectal cancer among women.11 Recent trials have demonstrated the potential for NSAIDs to reduce colorectal neoplasia, but adverse effects from these agents will limit their widespread use for that explicit purpose. Nonetheless, even the common sporadic use of NSAIDs for other indications will contribute to continuing declines in colorectal cancer incidence in the coming years.
Screening with either sigmoidoscopy or colonoscopy leads to the identification and removal of adenomas, thus preventing the development of colorectal cancer.19,20 Medicare included coverage for all recommended colorectal screening methods in 2001, and national publicity has substantially increased public interest in screening.21 Colorectal screening rates have increased over time, now with about two-thirds of adults over age 50 reporting having ever been screened by lower gastrointestinal endoscopy (see Table 12.1).
Decreasing rates of colorectal cancer incidence are occurring in spite of the obesity epidemic, which is an adverse force on colorectal cancer risk, because obesity may account for as much as 20% of colorectal cancer in the United States.22 Recently, however, obesity trends have stabilized in the United States. 23 As a result of the increased use of lower gastrointestinal endoscopy for colorectal screening and this stabilization of obesity trends, the incidence of colorectal cancer may exceed the ACS goal for 2015 of a 25% reduction, and there is a high likelihood that the rate of decline in deaths from colorectal cancer will be steep enough to reach the 2015 ACS mortality reduction goal of 50%.
Breast Cancer
The breast is the leading site of cancer incidence and the second leading site for cancer death among women in the United States.6 Over the period 1990 to 2001, no substantial changes in incidence rates were observed, but after 2000, breast cancer incidence began to decline. The decline in breast cancer incidence observed after 2002 seems to have been the result of the sudden decline in the use of HRT following the 2002 publication of the Women’s Health Initiative results.10,11 It is likely that persisting lower rates of HRT use will cause a continued decline in breast cancer incidence in the coming years. Countering this favorable trend, however, are the adverse effects of the obesity epidemic. Obesity, a major risk factor for postmenopausal breast cancer, increased substantially between 1990 and 2005, now with over 25% of US women being obese. However, the slowing of the obesity epidemic since 2005 may have substantial beneficial effects on the future trends in breast cancer incidence.
After persistent increases in the use of mammography over a 20-year period, mammography rates declined modestly between 2000 and 2004, and then leveled off. The downgrading of the evidence recommendations by the US Preventive Services Task Force for mammography for women age 40 to 49 and recommendations for every other year mammographies for women age 50 and older have resulted in lower mammogram utilization, which is likely to continue into the coming decade.17 This trend will have an adverse effect on breast cancer mortality, but will tend to reduce breast cancer incidence somewhat because of a lack of detection of very early stage cancers.
The antiestrogens tamoxifen and raloxifene have both been shown to reduce the risk of incident breast cancer.24 The safety profile for tamoxifen discourages its widespread use, but there is a more favorable risk/benefit balance of raloxifene. Nonetheless, neither of these drugs is commonly used for breast cancer prevention among postmenopausal women in the United States.
The average decline in breast cancer death rates of 2% per year since 1990 is the combined result of earlier diagnosis and better treatment.25 Progress in breast cancer treatment is continuing, especially in the development and application of hormone-targeted therapies. Aromatase inhibitors have largely replaced tamoxifen therapy for breast cancer treatment for postmenopausal women. Because all antiestrogens substantially reduce the incidence of second primary cancers in the contralateral breast, they impact both therapy and prevention. In the coming decade, the longer term effects of decreased HRT use, increased antiestrogen use, reversal of the obesity trends, and continued improvements in therapies will likely lead to continued decreases in both the incidence and mortality rates from breast cancer.
Prostate Cancer
The prostate is the leading site for cancer incidence and the second leading site for cancer death among men in the United States.6 The incidence of prostate cancer has been extremely variable over the period 1990 to 2010. The incidence spike observed in the early 1990s actually began in the late 1980s, coincident with the advent of PSA testing. The reasons for the 2.8% annual downward trend in prostate cancer mortality since 1990 are uncertain, however, because the ongoing PSA screening trials have not yet demonstrated a mortality benefit from screening anywhere as large as the downward mortality decline observed since 1990.26,27 In fact, the US trial findings suggest that there was virtually no mortality benefit within the first decade following the initiation of screening.28 Therefore, it is not possible to know how much of this favorable trend was related to early diagnosis, how much was related to improvements in treatment, or how much might have been related to other factors, such as changes in the way cause of death has been listed on death certificates.
The Prostate Cancer Prevention Trial provided an important proof of principle that antiandrogen therapies can reduce prostate cancer risk.29 Although the net benefits of finasteride for prevention are not clearly demonstrated from this trial, other agents that interfere with androgen effects on prostate cancer growth could prove to be useful for prostate cancer chemoprevention in the future. Prostate cancer incidence trends will likely continue to be largely driven by rates of PSA screening in the coming decade. Longer term results of a clearer benefit to mortality from either the PLCO trial in the United States or the European PSA trial would help to better specify screening recommendations.
Other Cancers
Even though mortality rates have been declining by about 2% per year from the four most common causes of cancer death (lung, colorectal, breast, and prostate), very little progress has been made in reducing death rates from the other half of all adult cancers in the United States. Continuing progress in tobacco control will have beneficial effects on many other types of cancer linked to tobacco, and stopping the obesity epidemic will have favorable effects on many obesity-related cancers that have been increasing in recent years, such as adenocarcinoma of the esophagus and renal cancer.30 Melanoma incidence rates have been increasing substantially in recent years, likely the result of the combined effects of previous sun exposure and increased awareness and surveillance for pigmented skin lesions, but recent advances in therapy for metastatic melanoma may foretell future declines in melanoma morality. Declining rates of stomach cancer incidence and mortality over several decades may be related to the combined effects of historic improvements in nutrition and the declining prevalence of chronic infection with Helicobacter pylori. Liver cancer incidence has been substantially increasing in recent years, likely resulting from historic trends in chronic infection with hepatitis B and C viruses. As a result, liver cancer will likely continue to rise in the United States over the coming decade.
The incidence of thyroid cancer has been increasing in the United States for the past several decades, but thyroid cancer mortality rates have been stable, a pattern most likely due to increased detection from improved diagnostic techniques. Invasive cervical cancer is uncommon in the United States because of widespread screening using Pap smears. Although the vaccination for human papillomavirus (HPV) has been shown to be highly effective in protecting against the serotypes that together account for 70% of cervical cancer cases, so far, HPV vaccine coverage has been low among young women in the United States.6 For many of the other cancers, such as cancers of the pancreas, brain, ovary, and the hematopoietic malignancies, risk factors are poorly understood, and there are no effective early detection methods. For these cancers, the current hope for improvement resides in the development of better methods for early cancer detection and treatment.
PREDICTING FUTURE CANCER TRENDS
In the United States, cancer is now the leading cause of death under age 85 years. Over the first half of the ACS 25-year challenge period, overall cancer incidence rates have declined by about 0.2% per year, and mortality rates have declined by about 1% per year. The trends in both incidence and mortality from the four leading cancer sites are summarized in Figure 12.1. Using simple linear extrapolation, it therefore seems that the ACS challenge goals of reducing cancer incidence by 25% and mortality by 50% over 25 years may be only half achieved.31,32 Clearly, though, estimating future trends only by linear extrapolation is a crude way to foretell future events. Projecting cancer trends into the more distant future using complex modeling is possible, however, as knowledge about changes in major cancer risk factors can lead to reasonable predictions about the direction and approximate slope of future trends. One method to incorporate knowledge about trends in risk factors into estimates of future cancer trends is to estimate the impact of changes in the attributable risk (also called the preventable fraction) in the population for each risk factor. By making assumptions about latency period, then tying changes in factors to changes in cancer incidence and mortality, cancer trends resulting from risk factor changes can be predicted. For example, if there were a factor that explained 30% of a particular cancer, then cutting that exposure in half would eventually lead to a projected 15% reduction in rates (50% of 30%). This method was used to project cancer mortality trends to 2015 and seems to have projected trends that are quite similar to those observed in recent years.33
Progress in cancer prevention, early detection, and treatment since 1990 has been persistent, and there are many reasons to be optimistic about the future. Just how much steeper the future downward slope in cancer death rates can be driven will depend on the extent to which we can discover new factors causing cancer, and effectively deploy ways to better act on our current knowledge about how to prevent and control cancer. Especially important will be progress in reversing the epidemics of tobacco use and obesity, and ensuring that the coming improvements to health care access will lead to access to state-of-the-art cancer screening and therapy for all.
REFERENCES
1. American Cancer Society Board of Directors. ACS Challenge goals for U.S. Cancer Mortality for the Year 2015. Proceedings of the Board of Directors. Atlanta, GA: American Cancer Society, 1996.
2. American Cancer Society Board of Directors. ACS Challenge goals for U.S. Cancer Incidence for the Year 2015. Proceedings of the Board of Directors. Atlanta, GA: American Cancer Society, 1998.
3. Centers for Disease Control and Prevention. National Program of Cancer Registries (NPCR) Web site. http://cdc.gov/cancer/npcr.
4. National Cancer Institute. Surveillance, Epidemiology, and End Results Program Web site. http://seer.cancer.gov.
5. Centers for Disease Control and Prevention. U.S. mortality data. http://wonder.cdc.gov/ucd-icd10.html
6. Jemal A, Simard E, Dorell C, et al. Annual report to the nation on the status of cancer, 1975–2009, featuring the burden and trends in human papillomavirus (HPV)–associated cancers and HPV vaccination coverage levels. J Natl Cancer Inst 2013;105:175–201.
7. Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System Web site. http://cdc.gov/brfss.
8. Ward E, Thun M, Hannan L, et al. Interpreting cancer trends. Ann N Y Acad Sci 2006;1076:29–53.
9. Centers for Disease Control and Prevention. Smoking & Tobacco Use Web site. http://cdc.gov/tobacco.
10. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–333.
11. Hersh A, Stefanick M, Stafford R. National use of postmenopausal hormone therapy: annual trends and response to recent evidence. JAMA 2004;291:47–53.
12. Ryerson AB, Miller J, Eheman CR, et al. Use of mammograms among women aged ≥40 years—United States, 2000–2005. MMWR 2007;56:49–51.
13. Giovino GA. Epidemiology of tobacco use in the United States. Oncogene 2002;21:7326–7340.
14. Oken M, Hocking W, Kvale P, et al. Screening by chest radiograph and lung cancer mortality. JAMA 2011;306:1865–1873.
15. The National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365:395–409.
16. Smith R, Brooks D, Cokkinides V, et al. Cancer screening in the United States, 2013: a review of current American Cancer Society guidelines, current issues in cancer screening, and new guidance on cervical cancer screening and lung cancer screening. CA Cancer J Clin2013;63:88–105.
17. U.S. Preventive Services Task Force. U.S. Preventive Services Task Force Web site. http://uspreventiveservicestaskforce.org.
18. Martinez ME. Primary prevention of colorectal cancer: lifestyle, nutrition, exercise. Recent Results Cancer Res 2005;166:177–211.
19. Atkin W, Edwards R, Kralj-Hans I, et al. Once-only flexible sigmoidoscopy screening in prevention of colorectal cancer: a multicentre randomized controlled trial. Lancet 2010;375:1624–1633.
20. Schoen RE, Pinsky PF, Weissfeld JL, et al. Colorectal-cancer incidence and mortality with screening flexible sigmoidoscopy. N Engl J Med 2012;366:2345–2357.
21. Cram P, Fendrick A, Inadomi J, et al. The impact of celebrity promotional campaign on the use of colon cancer screening: the Katie Couric effect. Arch Intern Med 2003;163(13):1601–1605.
22. World Cancer Research Fund/American Institute for Cancer Prevention. Policy and Action for Cancer Prevention. Food, Nutrition, and Physical Activity: A Global Perspective. Washington, DC: AICR; 2009.
23. Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–1555.
24. Vogel V, Constantino J, Wickerham D, et al. Effects of tamoxifen vs raloxifene on the risks of developing invasive breast cancer and other disease outcomes: the NSABP Study of Tamoxifen and Raloxifene (STAR) P-2 trial. JAMA 2006;295:2727–2741.
25. Berry D, Cronin K, Plevritis S, et al. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 2005;353:1784–1792.
26. Andriole GL, Crawford ED, Grubb RL 3rd, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009;360(13):1310–1319.
27. Schröder FH, Hugosson J, Roobol MJ, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009;360(13):1320–1328.
28. Andriole G, Crawford D, Grubb R, et al. Prostate cancer screening in the randomized Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial: mortality results after 13 years of follow-up. J Natl Cancer Inst 2012;104:125–132.
29. Thompson I, Goodman P, Tangen C, et al. Long-term survival of participants in the prostate cancer prevention trial. N Engl J Med 2013;369:603–610.
30. International Agency for Cancer Research. Weight Control and Physical Activity. Handbook 6. Lyon, France: IARC Press; 2002.
31. Sedjo R, Byers T, Barrera E, et al. A midpoint assessment of the American Cancer Society challenge goal to decrease cancer incidence by 25% between 1992 and 2015. CA Cancer J Clin 2007;57:326–340.
32. Byers T, Barrera E, Fontham E, et al. A midpoint assessment of the American Cancer Society challenge goal to halve the U.S. cancer mortality rates between the years 1990 and 2015. Cancer 2006;107:396–405.
33. Byers T, Mouchawar J, Marks J, et al. The American Cancer Society challenge goals. How far can cancer rates decline in the U.S. by the year 2015? Cancer 1999;86:715–727.