Cost-effectiveness of unselected multigene germline and somatic genetic testing for epithelial ovarian cancer

Abstract

Background: Parallel panel germline and somatic genetic testing of all patients with ovarian cancer (OC) can identify more pathogenic variants (PVs) that would bene fit from PARP inhibitor (PARPi) therapy, and allow for precision prevention in unaffected relatives with PVs. In this study, we estimate the cost-effectiveness and population impact of parallel panel germline and somatic BRCA testing of all patients with OC incorporating PARPi therapy in the United Kingdom and the United States compared with clinical criteria/family history (FH) -based germline BRCA testing. We also evaluate the cost-effectiveness of multigene panel germline testing alone. Methods: Microsimulation cost-effectiveness modeling using data from 2,391 (UK: n=1,483; US: n=908) unselected, population-based patients with OC was used to compare lifetime costs and effects of panel germline and somatic BRCA testing of all OC cases (with PARPi therapy) (strategy A) versus clinical criteria/ FH-based germline BRCA testing (strategy B). Unaffected relatives with germline BRCA1/BRCA2/RAD51C/RAD51D/BRIP1 PVs identi fied through cascade testing underwent appropriate OC and breast cancer (BC) risk-reduction interventions. We also compared the costeffectiveness of multigene panel germline testing alone (without PARPi therapy) versus strategy B. Unaffected relatives with PVs could undergo risk-reducing interventions. Lifetime horizon with payer/societal perspectives, along with probabilistic/one-way sensitivity analyses, are presented. Incremental cost-effectiveness ratio (ICER) and incremental cost per quality-adjusted life year (QALY) gained were compared with 30,000 pound/QALY (UK) and $100,000/QALY (US) thresholds. OC incidence, BC incidence, and prevented deaths were estimated. Results: Compared with clinical criteria/FH-based BRCA testing, BRCA1/BRCA2/RAD51C/RAD51D/BRIP1 germline testing and BRCA1/BRCA2 somatic testing of all patients with OC incorporating PARPi therapy had a UK ICER of 51,175 pound/QALY (payer perspective) and 50,202 pound/QALY (societal perspective) and a US ICER of $175,232/QALY (payer perspective) and $174,667/QALY (societal perspective), above UK/NICE and US cost-effectiveness thresholds in the base case. However, strategy A becomes cost-effective if PARPi costs decrease by 45% to 46% or if overall survival with PARPi reaches a hazard ratio of 0.28. Unselected panel germline testing alone (without PARPi therapy) is cost-effective, with payer-perspective ICERs of 11,291 pound/QALY or $68,808/QALY and societal-perspective ICERs of 6,923 pound/QALY or $65,786/QALY. One year 's testing could prevent 209 UK BC/OC cases and 192 deaths, and 560 US BC/OC cases and 460 deaths. Conclusions: Unselected panel germline and somatic BRCA testing can become cost-effective, with a 45% to 46% reduction in PARPi costs. Regarding germline testing, unselected panel germline testing is highly cost-effective and should replace BRCA testing alone.