An Explanation for the Increasing Incidence of Testis Cancer: Decreasing Age at First Full-Term Pregnancy

For unknown reasons, the incidence of cancer of the testis has increased substantially among white male populations of several European countries, the United States, Australia, and New Zealand (1). In Japan, the rates of testicular cancer have increased recently as well (2). An increase in the rate in white populations was first noted among men born after 1920, but this increase was not consistently sustained for men born between 1930 and 1945 (1). However, for men born in the 1950s onward, the increase in incidence has been uninterrupted. Explanations for this secular increase have been elusive. To some investigators, reports of a concomitant decrease in sperm counts in young men have suggested that an environmental exposure (e.g., pesticides and plant phytoestrogens) might be responsible (3). A substantial body of experimental and epidemiologic evidence indicates that prenatal events or exposures are important risk factors for testis cancer (4). Excess maternal nausea and vomiting in the prenatal period, prenatal exogenous exposure to diethylstilbesterol (DES), and maternal obesity have been associated with increased risk of testis cancer and with the risk of cryptorchidism, which is by far the strongest known risk factor for testis cancer (5–7). These shared risk factors have suggested that in utero estrogen exposure might be a common cause of both cryptorchidism and cancer of the testis. Animal experiments have shown that estrogen treatment of pregnant mice can lead to undescended and hypogenetic testis (5). Similar abnormalities have been reported in male offspring of women exposed to DES and oral contraceptives during pregnancy (6). The risk of testis cancer associated with excess nausea and vomiting is greatest for nausea requiring medical treatment (5). The strongest risk factors for such hyperemesis gravidarum are earlier age at pregnancy, nulliparity, and high body weight (8). Increased levels of bioavailable estradiol are found in the first trimester of pregnancy in women with hyperemesis gravidarum compared with control subjects and in the first trimester of a woman's first pregnancy compared with her second pregnancy (8,9). The coincidence between the increasing incidence of testis cancer, particularly among men born since the 1950s, and decreasing age at first full-term pregnancy (FFTP) experienced by North American and presumably Western European women during the first half of this century (10) is striking. In the United States, the mean age at FFTP was 25.5 years for women born in 1910, but it fell more or less continuously to a nadir of less than 22 years for women born in 1940. It has since risen once again (10). We propose that the three decades of continuous decline in maternal age at FFTP provides at least a partial explanation for the increase in testis cancer incidence among the corresponding male offspring birth cohorts of these women. Table 1 provides age-specific and age-adjusted incidence rates (AAIRs) of testis cancer for white males in Los Angeles County from 1972 through 1994. The AAIRs for testis cancer increase steadily, with a peak between 1987 and 1989 that is more than double the rate of the early 1970s, but subsequently decline by about 15% in the first half of the 1990s. Women who were born between 1910 and 1940 and experiencing a declining age at FFTP would have male offspring born approximately between 1930 and 1960 (10). These sons would be entering their peak ages (25-35 years) for testis cancer risk in the 1950s. By the early 1990s, men in the peak incidence ages for testis cancer would increasingly be those whose mothers sustained their FFTP after 1960, when age at FFTP was again on the rise.