br To account for uncertainty in
To account for uncertainty in the risk of carrying a BRCA mutation, and variability in the uptake of risk-reducing surgery and subsequent use of HRT, we conducted sensitivity analyses on these rates, as well as costs to approximate those applicable to the United States, and indi-rect opportunity costs, which could range from as little as a few weeks to a few months, depending on postoperative recovery [26–28]. We conducted a Monte Carlo simulation to estimate the number of women who would be diagnosed with ovarian and breast cancer ac-cording to each strategy, as well as deaths secondary to premenopausal BSO without HRT.
BRCA mutation testing for female first-degree relatives of women with HGSC yielded a higher average quality-adjusted life expectancy at acceptable cost compared to no testing, with an ICER of $7888 (CAD) per QALY gained. BRCA mutation testing of first-degree relatives was more effective and less costly than universal RRBSO in the absence of testing (19.20 QALYs vs. 18.52 QALYs, and $10,135 and $14,231, re-spectively), and therefore BRCA mutation testing is the dominant strat-egy. Table 2 summarizes the average discounted quality-adjusted life expectancy gains and costs associated with each strategy.
Our results were stable over a wide range of costs, to estimate those in the United States health care system, and variables such as BRCA mu-tation rates among women with HGSC, and the proportion having risk-reducing surgery in the context of a known BRCA mutation. Compliance with HRT must be very high in order to mitigate the downstream conse-quences known to be associated with premenopausal BSO. Fig. 2 illus-trates that the proportion using HRT must be higher than 79.3% for universal RRBSO to be a more effective strategy than BRCA mutation testing first for these women. The utility associated with premeno-pausal BSO in the absence of HRT must also be very high for this strategy to be more effective than BRCA mutation testing. Sensitivity analysis re-vealed that when this utility is greater than 0.956, universal RRBSO (without BRCA mutation testing) is preferable to BRCA mutation testing
Selected data for Calpain Inhibitor I case.
Variables Estimate for Range
Lifetime mortality risk, premenopausal RRBSO 12.5% 10–15%
Proportion having risk-reducing surgery with 33% 30–50%
Proportion with intraoperative or postoperative 6% 2–6%
complications from RRBSO 
Costs are expressed in Canadian dollars (CAD).
first. If we estimate life expectancy gain (unadjusted for quality of life), and assume 100% compliance with HRT for those undergoing premeno-pausal BSO, then universal RRBSO without BRCA mutation testing is the most effective strategy. Although universal RRBSO without BRCA muta-tion testing is more costly in this scenario, it has a favorable ICER of $10,057 relative to BRCA mutation testing first.
We conducted a Monte Carlo simulation to estimate the number of breast and ovarian cancer cases that would be diagnosed in a lifetime according to each of the three strategies, as well as the number of deaths secondary to premenopausal BSO in the absence of HRT. In Canada, there are approximately 2800 women diagnosed with ovarian cancer every year, and about 50% (n = 1400) of them will have HGSC. These women likely have at least 1 female first-degree relative (assuming 2 generations, such as a sister or daughter, based on total fertility rates ranging from 1.51–2.46 in the last 50 years ). By simulating a cohort of 1400 female first-degree relatives of women with HGSC (through 1000 trials) our model estimates breast cancer diagnoses in 179, 146, and 95 women, and ovarian cancer diagnoses in 40, 19, and 2 women, associated with no testing, universal BRCA mutation testing for all fe-male first-degree relatives, and universal RRBSO without BRCA testing, respectively, over a lifetime. However, the model also estimates that
Average discounted costs and life expectancy gains in base case.
Strategy Costs Effectiveness (QALYs) ICER
Fig. 1. Framework for Markov decision-analytic model. Dominated means that the strategy is more costly and less effective than an alternate (previous) strategy.
Fig. 2. Sensitivity analysis on proportion of women using hormone replacement therapy after premenopausal risk-reducing bilateral salpingo-oophorectomy.
80 women will die as a result of universal RRBSO when offered before menopause because of low use of HRT. These results are summarized in Table 3.
Although women with HGSC have a 1 in 5 chance of carrying a BRCA mutation, many of them do not undergo genetic testing, for a variety of reasons [30–32]. Based on a large cross-sectional study in the United States, it is estimated that only 10% of eligible ovarian cancer patients actually undergo genetic testing . HGSC is associated with a poor prognosis, and many women will die as a result of their disease before being tested for BRCA mutations. Furthermore, there is presumably a large legacy cohort of thousands of women diagnosed with HGSC who never had the opportunity to undergo genetic testing, because they were diagnosed before BRCA testing was routinely offered for HGSC his-tology, and they had no other risk factors to warrant genetic testing.