Cancer Surveillance Research Implementation Plan

1. Introduction

The ultimate goal of research sponsored by the National Cancer Institute (NCI) is to reduce the burden exacted by cancer by preventing or curing it. Basic laboratory and epidemiologic research are elucidating the causes, prevention, and cures of cancer. Intervention research is demonstrating how the cancer burden can be affected by manipulating risk factors and health-related behaviors known to cause, prevent, or cure cancer. Surveillance research is describing how cancer rates respond to interventions known to be effective. Control of cancer requires such discoveries as well as translation of these discoveries into effective interventions and delivery of these interventions to the population. It also requires synthesis of knowledge in order that informed decisions can be made regarding the cancer research agenda and public health policy. Genes involved in the progression of the major cancers have been identified and their translation into effective interventions will be important in controlling cancer. During the last two decades, screening has been introduced for three of the most common cancers: breast, colon, and prostate. In fact, the observed increases in incidence rates in the 1980s for breast and prostate cancer may be attributable in part to increases in screening. In addition, new therapies for preventing and treating certain cancers have been introduced, such as taxol, tamoxifen, and herceptin. At least one of these—tamoxifen—has been shown to have the potential to prevent breast cancer. Perhaps most importantly, education of the public regarding the relationship between cancer and lifestyle has had a positive impact on cancer-related behaviors, such as a reduction in smoking rates. These major breakthroughs in our understanding of cancer, and the resulting growth in our arsenal of weapons against cancer, are beginning to be reflected in declines in cancer incidence. Already, we are observing declines in the mortality rate for many cancers, including cancers of the breast, prostate, male lung, and colorectum. The annual age-adjusted cancer death rate of the U.S. population fell between 1992 and 1996, the first sustained decline since national record keeping was instituted in the 1930s.

Figure 1.
Percent Change in U.S. Cancer Mortality and SEER Cancer
Incidence Rates for Each 5-Year Period, 1977-1996

Figure 2.
SEER Cancer Incidence1, Percent Change 1992-1996 Trends for
Top 10 Sites, All Ages, All Races

1.1
NCI’s Cancer Surveillance Research Program

The role of NCI’s Cancer Surveillance Research Program (CSRP) is to study cancer rates and trends, track the impact of cancer on the general population, and provide information that will generate hypotheses and address questions about changes in trends over time. Appropriate decision making in science and in public health depends on reliable information about the effects of our efforts to control cancer. An effective surveillance program drives the cancer control research agenda by identifying opportunities for investment with high payoff in terms of reduced morbidity and mortality.

The CSRP is the most authoritative source of information on cancer incidence, mortality, and stage-specific survival in the United States. Its rigorous quality standards have made it the model for cancer surveillance activities throughout the world. Since 1973, the CSRP’s Surveillance, Epidemiology, and End Results (SEER) Program has tracked the impact of cancer on the general population, amassing data on more than 2.5 million cancer cases. These data increasingly are being used to answer questions about cancer causation, prevention, treatment, and control. For more than 25 years, SEER and its predecessor programs have enabled the NCI to elucidate environmental carcinogens, to track the cancer-related effects of tobacco on men and women, to locate geographic areas with higher than average rates of cancer, to study patterns and outcomes of cancer care, to estimate the cost of cancer, and to identify risk groups for research and public health intervention programs. All of this has been accomplished while maintaining the highest level of confidentiality and privacy. The CSRP has done an outstanding job of characterizing the cancer burden and identifying trends in incidence, mortality, and survival rates for specific cancer sites. Because of NCI’s cancer surveillance program, we know that overall cancer incidence and mortality rates in the United States have begun to decline for the first time in this century, even though rates for certain cancers (e.g., female lung cancer and non-Hodgkin’s lymphoma) are rising.

Opportunities now exist to expand the surveillance program to ensure that it will provide information needed to better characterize the cancer burden, interpret observed changes in trends over time, and enhance researchers’ ability to generate hypotheses. Recent advances in information technology provide opportunities to link cancer surveillance data with different types of health-related information on populations, which can provide a powerful tool for analyzing factors that influence cancer rates (risk factors, screening, treatment, and health practices) and for planning and evaluating population-based prevention and control interventions.

1.2
The Surveillance Implementation Group

To identify what cancer surveillance research is most needed and how best to advance our knowledge of cancer based on the opportunities available, the Director of the NCI established the Surveillance Implementation Group (SIG), which included 42 leading scientists and experts (see Appendix A for a list of the SIG members) from within the NCI, other federal agencies, and the extramural community as well as representatives of major NCI review and advisory committees. The SIG was charged with providing advice and recommendations for expanding and enhancing NCI’s Cancer Surveillance Research Program. The SIG was asked to identify research directions and priorities and to produce an implementation plan that was national in scope and that presented a comprehensive, focused, coherent vision for NCI-funded surveillance research. The framework for the SIG’s deliberations was based on the recent reviews of the cancer control and prevention programs at the NCI, but the group was directed to go beyond these reports to provide scientific guidance about cancer surveillance programs of the future. The SIG took as its primary directive the recommendations of the Cancer Control Program Review Group (CCPRG), convened in December 1996 by the Director of the NCI and the NCI Board of Scientific Advisors (BSA) to evaluate the full scope of current and past activities of the Institute’s cancer control research program. Regarding NCI’s surveillance program, the CCPRG recognized the high quality of the data collection, research, and reporting activities and noted that SEER data have been used nationally for many reports on cancer trends and patterns and to facilitate data collection for epidemiologic, cancer control, and genetic studies. To improve the current surveillance program, the CCPRG recommended that the NCI should:

Expand the SEER Program to include additional populations, more data from patients’ medical records and patients themselves, and population data from the SEER regions to monitor individual and societal mediators of cancer.
Use the SEER expanded data and expertise to produce a timely report card on the cancer burden.

1.2.1
Vision for the Cancer Surveillance Research Program

Vision Statement for NCI’s Cancer Surveillance Research Program

The National Cancer Institute’s Cancer Surveillance Research Program (CSRP) is the internationally recognized standard of excellence for the comprehensive surveillance of cancer. It characterizes the cancer burden borne in the United States over time by integrating traditional statistics on persons with cancer with the widest possible collection of cancer-related data in the general population. Based on a sound foundation of applied and methodologic research, the CSRP measures progress in reducing the Nation’s cancer burden. It simultaneously provides a stimulus for etiologic research and identifies opportunities for intervention research and public health applications.

The SIG members interpreted the recommendations of the CCPRG in light of their vision for the CSRP. The members agreed that the implementation plan should extend beyond the SEER Program expansion recommended by the program review group. They also concurred that addressing the many questions relating to the national cancer burden will require substantial changes within the current NCI surveillance system. Clearly, the SEER Program—a continuing model of excellence in cancer surveillance throughout the world—will remain the core of the expanded system. As new tools are developed, the SEER system will be connected to data collection mechanisms that probe deeply into the causes of cancer rates and trends with a consistent focus on defined populations as the point of reference.

These defined population studies in the SEER areas (or other areas of high-quality registration) will collect data on prevention, risk factors, screening, and treatment interventions. This will involve data collection on cohorts of patients over time, and will include information on health status, patterns of care, and quality of life. These data will provide information for understanding specific questions concerning cancer rates, such as: Why are colorectal cancer mortality rates decreasing? What impact does early detection have on colorectal cancer mortality rates? These data collection models, similar to NCI’s Breast Cancer Surveillance Consortium and Prostate Cancer Outcomes Study, will require increasing connections of the surveillance program with the epidemiology and treatment programs at the NCI. As other registries around the country match the SEER standard, data can be pooled. Any expansion of SEER activities will be taken in concert with a newly developed long-term surveillance research plan that includes collaborations with other organizations involved in cancer surveillance. Finally, the strong research structure serving as the underpinning of SEER will continue, with major methodologic efforts in modeling rates and trends along with new efforts in geographic information systems, development of improved approaches to generating national estimates of cancer burden, and new research on familial and genetic components of cancer surveillance.

The SIG emphasized that in pursuing the vision for the surveillance program and the goal of complete cancer registration, the NCI and its CSRP must collaborate with partners in national cancer surveillance such as the Centers for Disease Control and Prevention (CDC), including the National Center for Health Statistics (NCHS) and the National Center for Chronic Disease Prevention and Health Promotion (NCCDPHP); the Health Care Financing Administration (HCFA); the American Cancer Society (ACS); the Commission on Cancer of the American College of Surgeons (ACoS); the North American Association of Central Cancer Registries (NAACCR); the National Coordinating Council for Cancer Surveillance (NCCCS); extramural grantees; and others. The involvement of these partners is essential because each organization has its own purpose and niche in the overall national cancer surveillance scheme. The SIG also recommended that the NCI provide feedback to the Institute scientists, extramural researchers, and health policy administrators to improve their understanding of how interventions and public health applications of research results are affecting the national cancer burden.

1.2.2
Process of the Surveillance Implementation Group

To accomplish its charge, the full SIG met five times during 1998—April, August, September, November, and December—as well as by conference calls for smaller working groups. After establishing consensus on the vision for the CSRP, the SIG analyzed NCI’s portfolio of cancer surveillance research, the balance of current research across topic areas, and the distribution of funding across pro-jects and funding mechanism. The NCI spent $39.9 million on cancer surveillance research in Fiscal Year 1998, which represents approximately 1 percent of NCI’s total budget. About half of the CSRP budget was allocated to the SEER Program.

The SIG then determined which of the CCPRG’s recommendations already had been partially or completely implemented, identified emerging issues from research in progress, and generated a list of recommendations. Through a consensus-building process, the SIG members modified and refined this list until it contained 12 cancer surveillance research opportunities within five priority areas. The research opportunities range from expansion of data collection to development of analysis tools to establishment of linkages among cancer and other health-related databases. These recommendations provided the foundation for development of this Cancer Surveillance Research Implementation Plan.

The next chapter of this report provides an overview of NCI’s current surveillance research program. Chapter 3 describes the SIG’s proposed strategy for implementing the 12 research opportunities for enhancing the surveillance program. A brief rationale for each of the research opportunities is provided in Chapter 3, along with an estimate of the approximate level of investment required and a time frame for initiating the research.

Figure 3.
Fiscal Year 1998 Budget Allocation for NCI’s Cancer Surveillance Research Program