Reduced Breast and Ovarian Cancer Through Targeted Genetic Testing: Estimates Using the NEEMO Microsimulation Model

Background: The effectiveness and cost-effectiveness of genetic testing for hereditary breast and ovarian cancer largely rely on the identification and clinical management of individuals with a pathogenic variant prior to developing cancer. Simulation modelling is commonly utilised to evaluate genetic testing strategies due to its ability to synthesise collections of data and extrapolate over long time periods and large populations. Existing genetic testing simulation models use simplifying assumptions for predictive genetic testing and risk management uptake, which could impact the reliability of their estimates. Our objective was to develop a microsimulation model that accurately reflects current genetic testing and subsequent care in Australia, directly incorporating the dynamic nature of predictive genetic testing within families and adherence to cancer risk management recommendations. Methods: The populatioN gEnEtic testing MOdel (NEEMO) is a population-level microsimulation that incorporates a detailed simulation of individuals linked within five-generation family units. The genetic component includes heritable high- and moderate-risk monogenic gene variants, as well as polygenic risk. Interventions include clinical genetic services, breast screening, and risk-reducing surgery. Model validation is described, and then to illustrate a practical application, NEEMO was used to compare clinical outcomes for four genetic testing scenarios in patients newly diagnosed with breast cancer (BC) and their relatives: (1) no genetic testing, (2) current practice, (3) optimised referral for genetic testing, and (4) genetic testing for all BC. Results: NEEMO accurately estimated genetic testing utilisation according to current practice and associated cancer incidence, pathology, and survival. Predictive testing uptake in first- and second-degree relatives was consistent with known prospective genetic testing data. Optimised genetic referral and expanded testing prevented up to 9.3% of BC and 4.1% of ovarian cancers in relatives of patients with BC. Expanding genetic testing eligibility to all BC patients did not lead to improvement in life-years saved in at-risk relatives compared to optimised referral of patients eligible for testing under current criteria. Conclusions: NEEMO is an adaptable and validated microsimulation model for evaluating genetic testing strategies. It captures the real-world uptake of clinical and predictive genetic testing and recommended cancer risk management, which are important considerations when considering real-world clinical and cost-effectiveness.