Reproductive Factors and Epithelial Ovarian Cancer Risk by Histologic Type:A Multiethnic Case-Control Study

• Published in: American journal of epidemiology 2003-10, Vol.158 (7), p.629-638
• Main Authors: Tung, Ko-Hui, Goodman, Marc T., Wu, Anna H., McDuffie, Katharine, Wilkens, Lynne R., Kolonel, Laurence N., Nomura, Abraham M. Y., Terada, Keith Y., Carney, Michael E., Sobin, Leslie H.
• Format: Article
• Language: English
• Subjects: Adenocarcinoma, Clear Cell - classification; Adenocarcinoma, Clear Cell - epidemiology; Adult; Age of Onset; Aged; Asian Continental Ancestry Group; Biological and medical sciences; breast feeding; Case-Control Studies; contraceptives; Contraceptives, Oral - therapeutic use; Educational Status; Endometrial Neoplasms - classification; Endometrial Neoplasms - epidemiology; Ethnic Groups - statistics & numerical data; Female; Female genital diseases; Gravidity; Gynecology. Andrology. Obstetrics; Hawaii - epidemiology; histology; Humans; Los Angeles - epidemiology; Medical sciences; Menstruation; Middle Aged; Neoplasms, Cystic, Mucinous, and Serous - classification; Neoplasms, Cystic, Mucinous, and Serous - epidemiology; Neoplasms, Glandular and Epithelial - classification; Neoplasms, Glandular and Epithelial - epidemiology; oral; Ovarian cancer; ovarian neoplasms; Ovarian Neoplasms - epidemiology; Ovarian Neoplasms - pathology; ovulation; parity; Postmenopause; Pregnancy; Reproduction; Risk Assessment; Risk Factors; Tumors
• ISSN: 0002-9262; 1476-6256; 0002-9262
• DOI: 10.1093/aje/kwg177

Associations of reproductive factors with ovarian cancer may differ by histologic type. Data from a multiethnic, population-based, case-control study, conducted in Hawaii and Los Angeles, California, between 1993 and 1999, were used to assess this hypothesis. A structured questionnaire was administered to 558 histologically confirmed epithelial ovarian cancer cases and 607 population controls. Factors suppressing ovulation, including pregnancy and oral contraceptive use, were inversely associated with the risk of all histologic types. Nonmucinous but not mucinous tumors were significantly associated with menstruation years (odds ratio = 1.5 for the highest vs. the lowest quartile) and lifetime ovulatory cycles (odds ratio = 2.8 for the highest vs. the lowest quartile). Duration of breastfeeding (odds ratio = 0.4 for the highest vs. the lowest quartile) was significantly and inversely related to nonmucinous tumors but not to mucinous tumors. Among all tumor types, endometrioid tumors were the most strongly related to pregnancy and tubal ligation, while clear cell tumors were the only type that was associated with noncontraceptive hormone use. The risk factors were similar for borderline and invasive tumors, except for age at diagnosis. Mucinous tumors, both borderline and invasive, were more common in Asian women than in Caucasian and other women. Our data suggest that histologic types of epithelial ovarian cancer are etiologically distinct.