Rates and Risk Factors
Rates and Risk Factors
In this chapter, we describe each major group of childhood cancer and report the observed incidence and mortality rates for New Jersey and the United States. We review demographic patterns and known genetic factors associated with major cancers and summarize the environmental, behavioral, and lifestyle factors that have been linked to various childhood cancers.
Classification of Childhood Cancers
The types and distribution of childhood cancer differ notably from those occurring in adult populations. The International Association of Cancer Registries (IACR) has established a unique system for categorizing childhood cancers, and this report follows that system. Whereas adult cancers are usually organized by site of the primary tumor, childhood cancers are classified according to their histology (microscopic identification of cells and tissue). The histological classification for childhood cancers is useful for the purposes of understanding the origins of the cancer and for identifying the best treatment.
Multiple Primary Tumors
Children who have been treated for one primary cancer are at increased risk of a second primary cancer later in life, i.e. a second primary cancer which is unrelated to the first cancer and not a result of the first cancer spreading or returning. Radiation and chemotherapies may increase the risk of subsequent primary cancers. Close monitoring of children for additional cancers is recommended after initial treatment for cancer. Each primary malignancy is counted as a separate cancer when calculating incidence rates, thus the total number of cancers can be greater than the total number of affected children. The increased risk of a second primary cancer persists into adulthood.
Table 1a presents age-adjusted incidence rates in New Jersey for specific childhood cancers by sex for 1979-1995. We also present US rates for 1979-1994, the available comparison years. Note that while the national data are intended to represent the US rates, they cover a sample of only ten percent of the US population. These data are made available by the National Cancer Institute's (NCI) Surveillance, Epidemiology and End Results (SEER) program. Table 1b presents age-adjusted incidence rates for New Jersey and the United States by sex for white and black children. In general, incidence rates were higher for white children than for black children and for boys than for girls. These findings hold true for the national data as well. For white children, incidence rates were slightly higher in New Jersey than in the US, while among black children incidence rates in NJ were slightly lower than the corresponding US rates, with the exception of lymphomas among males.
The incidence rates in Tables 1a and 1b are cited throughout the following sections. Appendix A discusses the statistical issues related to the data presented in this report. Appendix B presents the population counts used as denominators to generate NJ incidence rates.
Distribution of Specific Cancers
Figure 1 presents the distribution of specific cancers among New Jersey children from birth to age 14 for the period 1979-1995. Thirty one percent of childhood cancers between 1979 and 1995 were leukemias, similar to national and international proportions. The second most frequently diagnosed cancer in New Jersey and nationally was cancer of the central nervous system, comprising about 20 percent of all childhood cancers in New Jersey. The third most frequently diagnosed cancer in New Jersey children was lymphoma, comprising about 12 percent of all childhood cancers. Figure 2 indicates the incidence rates of the four most common pediatric cancer groups, leukemias, central nervous system cancers, lymphomas and soft-tissue sarcomas. The figure illustrates their relative magnitude (Please note that in order to increase clarity, other figures in this report have varying scales).
Presentation of Incidence Trends, 1979-1995
Figures 3-12, which are discussed further below, show the trends in New Jersey and United States childhood cancer three-year incidence rates centered on the years 1980-1994. They also show New Jersey three-year mortality rates and United States single-year mortality rates. Three-year rates are shown in order to reduce the amount of random fluctuation typically seen when the numbers of cases are small. Pooling data for three years increases the number of cases used for each rate thereby reducing the impact of year to year changes. Appendix C presents the rates used to create the graphs.