Background: Approximately half of epithelial ovarian cancers are fatal within 3 years; however, roughly 35% of women survive for at least 10 years. In the Nurses' Health Study, New England Case-Control Study, Australian Ovarian Cancer Study, and NIH-AARP Diet and Health Study, we investigated potential differences in the associations with ovarian cancer risk factors by tumor aggressiveness, defined on the basis of time from diagnosis until death.
Methods: We calculated relative risks (RR) and 95% confidence intervals (CI) for associations of known or suspected ovarian cancer risk factors with rapidly fatal (death within 3 years of diagnosis) and less aggressive tumors (all others) using Cox proportional hazards competing risks analysis (NHS and AARP) or polytomous logistic regression (NECC, AOCS). Results were combined using random effects meta-analysis.
Results: Increasing age was associated with greater risk of rapidly fatal versus less aggressive disease (RR, 5-year increase: 1.39; 95% CI, 1.29-1.49 vs. RR, 1.09; 95% CI, 1.03-1.16, respectively; Pdiff < 0.0001). Oral contraceptive use was associated with a greater decreased risk of rapidly fatal (RR, 5-year increase: 0.69; 95% CI, 0.58-0.82) versus less aggressive disease (RR, 0.81; 95% CI, 0.74-0.89; Pdiff, 0.002). Conversely, increasing parity was associated only with less aggressive disease (RR per child, 0.87; 95% CI, 0.81-0.93).
Conclusion: In this analysis of 4,342 cases, there were clear differences in risk factors for rapidly fatal versus less aggressive ovarian tumors.
Impact: Differences in risk factor associations by tumor aggressiveness suggests the developmental pathways through which the tumors develop and may be important for developing primary preventive strategies for the most aggressive cancers.