Ovarian cancer is the fifth leading cause of cancer death for women in the U.S. and the 7th most fatal worldwide. Reproductive and hormonal exposures are known to influence risk of ovarian cancer. We therefore investigated whether these exposures also influence ovarian cancer survival.

We studied survival in relation to hormonal and reproductive history in invasive epithelial ovarian cancer cases and controls from the New England based Case-Control Studies (NECC, 1,642 cases and 2,100 controls). Cases and controls were identified between 1992 and 2008 in New Hampshire and eastern Massachusetts. At the time of enrollment, participants were interviewed in person on a wide range of known and suspected risk factors, including hormone use, childbearing, lifestyle habits and anthropometric characteristics. Polytomous logistic regression was applied to evaluate whether a risk factor was associated differently with fatal (those that died within 3 years of diagnosis) versus non-fatal cases while controlling for other covariates and P-values for heterogeneity (comparing the Odds Ratios (ORs) for fatal ovarian cancer to the ORs for non-fatal ovarian cancer) were calculated based on the maximum likelihood estimates.

A total of 360 of the 1642 women with ovarian cancer (22%) died within 3 years. We found that fatal ovarian cancer was associated with increasing age (multivariate OR 1.05 per 1 year increase in age, 95% CI 1.04 − 1.07, P-value < 0.0001), post-menopausal status (multivariate OR 1.31 vs pre-/dubious menopause, 95% CI 0.97 − 1.77, P-value < 0.0001) and oral contraceptive pill use was consistently more protective across all duration of use categories (P-value = 0.01) as compared with the non-fatal cases. Associations with tubal ligation, total number of ovulatory years, body mass index and family history of breast/ ovarian cancer did not differ significantly between fatal and non-fatal ovarian cancer cases.

These results suggest that the most aggressive subset of invasive epithelial ovarian cancer exhibits important differences in its reproductive and hormonal risk factor profile as compared with non-fatal cases, and that selected protective factors such as oral contraceptive pill use may be especially important for the prevention of the most aggressive subgroup of disease.

Citation Information: Cancer Prev Res 2011;4(10 Suppl):A108.

Copyright © American Association for Cancer Research
2011