A model for estimating ovarian cancer risk: application for preventive oophorectomy

It is important to identify women in the population who have a high risk of ovarian cancer and who might benefit from prophylactic bilateral salpingo-oophorectomy. The probability that a woman will develop ovarian cancer depends on her current age, her reproductive history and her genetic status.

We simulated the distribution of ovarian cancer risk for the 2011 Ontario female population. We generated (at random) individual risks of ovarian cancer to age 80 for 6,301,340 women, based on the published risk factors, mutation frequencies and population age-specific incidence rates (SEER database). Risk factors included parity, breastfeeding, oral contraceptives, tubal ligation and family history. Genetic factors included 11 single nucleotide polymorphisms (SNPs) and BRCA1/2 mutations.

Of the 6,301,340 women simulated as the general population of Ontario, the (complete) model predicts that 65,805 women (1.0%) will develop ovarian cancer by age 80. There were 46,069 women (0.7%) with a risk of ovarian cancer above 5%. BRCA1/2 mutation carriers accounted for 67.4% of the women at greater than 5% risk (31,028 women). Among ovarian cancer patients at greater than 5% risk, a BRCA1/2 mutation was present in 89.2%. In contrast, SNPs contribute to a very small proportion of the ovarian cancer patients who were at greater than 5% risk.

Approximately 12.9% of all ovarian cancers in Ontario occur in the 0.7% of women in the general population who have a lifetime ovarian cancer risk in excess of 5%, the majority of whom carry a mutation in BRCA1 or BRCA2.