Gender of offspring and maternal ovarian cancer risk

Abstract

Objective.

A single live birth compared to nulliparity significantly reduces the risk for ovarian cancer, but exactly how pregnancy reduces ovarian cancer risk is unknown. We sought to determine whether offspring gender, which differentially alters maternal hormonal milieu, may be associated with maternal ovarian cancer risk.

Methods.

Parous women (n = 511) with incident ovarian cancer were compared to parous community controls (n = 1136) participating in a population-based case-control study of ovarian cancer (Delaware Valley, 1994–1998). In subgroup specific models for women with one, two, or three births, multivariate logistic regression was used to assess the relationship between ovarian cancer and offspring gender, adjusting for age, race, education, oral contraceptives, breast feeding, tubal ligation, and ovarian cancer family history.

Results.

Compared to having all girls, women with all boys tended to have a reduced risk of ovarian cancer (OR = 0.80 95% CI: 0.58, 1.10), while women with boys and girls conferred the greatest protection (OR = 0.58, 95% CI: 0.43, 0.79). Among women with two births, the association was observed for those with one boy and one girl (OR = 0.63, 95% CI: 0.40, 1.00), but not for those with two male offspring (OR = 1.12, 95% CI: 0.68, 1.85). This result was consistent among women with three births (OR = 0.42, 95% CI: 0.21, 0.84; OR = 0.47, 95% CI: 0.23, 0.95; OR = 0.49, 95% CI: 0.20, 1.21; for one, two, and three boys, respectively, compared to all girls).

Conclusion.

Compared to having all girls, bearing both male and female offspring may be associated with a decrease in maternal ovarian cancer risk, although the biologic relevance of this observation is unclear.

Introduction

Ovarian cancer is responsible for more deaths than any other cancer of the female reproductive system, with less than 50% of women in the United States surviving 5 years after diagnosis [1]. It is estimated that 22,220 American women will be diagnosed with ovarian cancer in the year 2005, and approximately 16,210 of these women will die [1], due, in part, to the lack of effective prevention and early detection methods. Increased understanding of the etiology of ovarian cancer may inform preventative strategies.

It is well-documented that a single live birth compared to nulliparity substantially reduces the risk for ovarian cancer, and the risk declines for each additional full-term pregnancy [2]. Exactly how pregnancy reduces ovarian cancer risk is unknown. Several hypotheses posed to explain the association include: reduction in number of ovulations [3], lower circulating gonadotropins [4], reduced inflammation [5], and altered circulating steroid hormones [6].

Fetal sex differentially alters the maternal hormonal milieu during normal pregnancy [7], [8], [9], [10], [11], [12], [13]. In particular, circulating concentrations of estriol [14], alpha-fetoprotein, and human chorionic gonadotropin [7], [8], [9], [10], [13] are lower in mothers of male fetuses; whether these differences relate to ovarian cancer risk is unknown. Since fetal sex is associated with some circulating maternal hormones, it is possible that an altered hormonal milieu might modify the protective effect of childbearing on hormonally linked cancers such as ovarian cancer. Epidemiologic studies of offspring gender and ovarian cancer risk are lacking. Bearing male offspring was associated with a reduced risk of breast cancer, another hormonally-related cancer, in one study [15], although not consistently [16], [17], [18], [19]. These observations led us to investigate the association between offspring gender and maternal ovarian cancer risk in a large population-based case-control study of ovarian cancer.