Risk Factors for Ovarian Cancer HS310: Epidemiology and Biostatistics

Ovarian cancer is the fifth most common type of cancer among women, and ranks first among the different types of gynecological cancer mortality in Canada (Canadian
Cancer Statistics, 2004). Ovarian cancer mostly affects elderly and middle aged women, with the highest incidence rates reported in North America and Northern Europe, and the lowest in Japan and in developing countries (Greenlee et al., 2000). Over the past 40 years, the rate of ovarian cancer mortality has increased among women 65 years of age and older. The incidence of ovarian cancer increases with age, from 1.5/100,000 annually among women 20 -30 years of age to 49/100,000 in women 60- 69 year of age (Mant and
Vessey, 1995). In Canada, 2,500 new cases are reported annually, and 1,550 women die from this disease each year. The lack of screening tests for early diagnosis of ovarian cancer is an important determinant of the mortality rate from this disease. Because of the absence or subtlety of early symptoms, more than 70 % of patients are diagnosed with advanced stages of ovarian cancer (Goff et al., 2000) The increase in the incidence of ovarian cancer near the age of menopause can be related to the diminution of oocytes or germ cells, the reduction in circulating estrogen levels, or a significant increase in the pituitary gland’s production of the gonadotropic hormones (follicle-stimulating hormone [FSH] and luteinizing hormone [LH]). Ovulation, growth factors, cytokines, and environmental agents may play a role in the initiation and progression of ovarian cancer (Vanderhyden et al., 2003). Ovarian cancer begins with transformation of cells that comprise the ovaries, including surface epithelial cells, germ cells, and the sex cord or stromal cells. The surface epithelial cells are the primary locus of ovarian cancer, with almost 90% of all ovarian cancers derived from the surface epithelium (Weiss et al., 1977). The etiology of ovarian cancer is not completely understood. Known risk factors for ovarian cancer are related to hormonal and reproductive factors, including nulliparity, late age at menopause, and family history of ovarian and breast cancer. Oral contraceptive use has been shown to be protective against ovarian cancer, with risk decreasing with duration of use. Epidemiologic studies, animal experiments, and receptor binding studies suggest that malignant ovarian tumors may be endocrine-related and hormone-dependent (Lacey et al., 2002, Dowsett 2000, Davis et al., 1993, Gillesby and Zacharewski 1998). These studies have led to the hypothesis that exogenous estrogenic compounds may play a role in the etiology of this cancer.
In order to better understand the role of such agents, it is important to examine all potential risk factors for ovarian cancer. This overview will also serve to identify gaps in knowledge and research needs for ovarian cancer that will address primary prevention and early detection of this disease. Elucidation of the etiology of ovarian cancer may be accomplished both through animal experiments and population based epidemiological studies of women at risk of ovarian cancer. The etiology of ovarian cancer is not very clear. However, some hypotheses about the etiology of epithelial ovarian tumors have been advanced (Modugno et al., 2003). The incessant ovulation hypothesis suggests that proliferation and repair of the surface epithelial cells following ovulation may increase the chance of an abnormal repair process, leading to a malignant cell and subsequent development of ovarian cancer
(Fathalla, 1971). Under this hypothesis, any factor that inhibits or prevents ovulation would reduce the risk of ovarian cancer. The pituitary gonadotropin hormone hypothesis suggests that high levels of gonadotropins can directly, or indirectly through the increased production of estrogen, stimulate ovarian surface epithelial cells in inclusion cysts and leads to malignant transformation (Konishi et al., 1999). Any factor that increases the level of these hormones can increase the risk of ovarian cancer (Modugno et al., 2003). Other risk factors for ovarian cancer include: personal or family history of breast, ovarian, endometrial, prostate or colon cancer; hereditary nonpolyposis colorectal cancer or syndrome; increasing age; infertility; nulliparity; no history of birth control pill usage, use of high dose estrogens for long periods; lifestyle, and exposure to certain environmental contaminants. Each of these factors will be considered in more detail below. There is epidemiologic evidence that late menopause may be associated with an increased risk of ovarian cancer (Franceschi et al., 1991). Other studies have found no relationship between age at menarche and the risk of ovarian cancer (Purdie et al., 1995). The frequency of ovulation and the development of epithelial ovarian cancer appear to be related (Purdie et al., 2003). Incessant ovulation is an important determinant of ovarian cancer risk, with factors that suppress ovulation, such as pregnancy, use of oral contraceptive pills (OCP), and lactation, reducing the risk of ovarian cancer (Purdie et al.,
2003). Evidence continues to accumulate that nulliparity is associated with an inc reased risk of ovarian cancer (Bristow et al., 1996). Infertile women who received fertility drugs for extended periods of time are at higher risk of developing ovarian cancer (Rossing et al., 1994). Overall, multiparous women have a 30 to 70% lower risk as compared with nulliparous women (Mosgaard et al., 1997). Several investigators have hypothesized that exposure to environmental chemicals such nas pesticides and herbicides can be linked to ovarian cancer. A case-control study of women with previous exposure to triazine herbicides showed a 2-3 fold increase in the risk of epithelial ovarian cancer, in comparison with unexposed women (Donna et al., 1989). An association between atrazine and ovarian tumours has been observed in two
Italian studies (Donna et al., 1984, 1989), suggesting that atrazine may be carcinogenic to humans. Aside from phenoxy acid herbicides, atrazine is the most commonly used herbicide worldwide (Short & Colborn, 1999), with application in the cultivation of corn, fruits, vegetables and grapes for producing wine. Atrazine has been classified as a possible human carcinogen by the International Agency for Research on Cancer (IARC,
1999). A study by Wetzel et al. also demonstrated that feeding Sprague-Dawley rats subjected to high doses of atrazine lengthened their estrous cycle, increased the number of days in estrous, and, given in conjunction with estrogen, induced an earlier onset of mammary tumors. Atrazine is a genotoxic compound, which induces DNA damage in some cells, and increases mammary and ovarian tumor incidence in rats following oral exposure (Wetzel et al., 1994). Ovarian cancer is the most common cause of death from gynecologic malignancies in thewestern world. Considerable effort has been expended to date to link variations in incidence in ovarian cancer rates with environmental, endocrinological, and genetic factors. There is growing evidence to suggest that factors such as family history, hormone therapy, lifestyle, and environmental exposure to certain chemicals may increase the risk of ovarian cancer. If current knowledge of the epidemiology of ovarian cancer is to be used as a basis for disease prevention, more attention should be paid to women at risk because of their family history, and educational strategies should be implemented to promote awareness of the protective effect of oral contraceptives. Pesticides with endocrine disrupting activity remain in use, legally or illegally, in different countries, including organochlorine, organophosphorous, arsenic and mercury compounds, phenoxy acid herbicides, atrazine, and dithiocarbamates. Scientific research to date suggests a link between atrazine and risk of ovarian cancer. Overall, certain environmental and occupational exposures may be associated with ovarian cancer. This association may be modified by life-style or genetic susceptibility. Although the collective evidence of an association between ovarian cancer and environmental agents is limited, environmental exposures should be considered in future studies of ovarian cancer.