Cost-Effectiveness of Risk-Reducing Surgery for Breast and Ovarian Cancer Prevention: A Systematic Review

Abstract

Policymakers require robust cost-effectiveness evidence of risk-reducing-surgery (RRS) for decision making on resource allocation for breast cancer (BC)/ovarian cancer (OC)/endometrial cancer (EC) prevention. We aimed to summarise published data on the cost-effectiveness of risk-reducing mastectomy (RRM)/risk-reducing salpingo-oophorectomy (RRSO)/risk-reducing early salpingectomy and delayed oophorectomy (RRESDO) for BC/OC prevention in intermediate/high-risk populations; hysterectomy and bilateral salpingo-oophorectomy (BSO) in Lynch syndrome women; and opportunistic bilateral salpingectomy (OBS) for OC prevention in baseline-risk populations. Major databases were searched until December 2021 following a prospective protocol (PROSPERO-CRD42022338008). Data were qualitatively synthesised following a PICO framework. Twenty two studies were included, with a reporting quality varying from 53.6% to 82.1% of the items scored in the CHEERS checklist. The incremental cost-effectiveness ratio/incremental cost-utility ratio and cost thresholds were inflated and converted to US$2020, using the original currency consumer price index (CPI) and purchasing power parities (PPP), for comparison. Eight studies concluded that RRM and/or RRSO were cost-effective compared to surveillance/no surgery for BRCA1/2, while RRESDO was cost-effective compared to RRSO in one study. Three studies found that hysterectomy with BSO was cost-effective compared to surveillance in Lynch syndrome women. Two studies showed that RRSO was also cost-effective at ≥4%/≥5% lifetime OC risk for pre-/post-menopausal women, respectively. Seven studies demonstrated the cost-effectiveness of OBS at hysterectomy (n = 4), laparoscopic sterilisation (n = 4) or caesarean section (n = 2). This systematic review confirms that RRS is cost-effective, while the results are context-specific, given the diversity in the target populations, health systems and model assumptions, and sensitive to the disutility, age and uptake rates associated with RRS. Additionally, RRESDO/OBS were sensitive to the uncertainty concerning the effect sizes in terms of the OC-risk reduction and long-term health impact. Our findings are relevant for policymakers/service providers and the design of future research studies.