Cost-effectiveness of population screening for Lynch syndrome and hereditary breast and ovarian cancer syndrome in Singapore.

Abstract

164 Background: Hereditary breast and ovarian cancer syndrome (HBOC) and Lynch syndrome (LS) are the most prevalent hereditary cancer syndromes. Population-based germline genetic testing (GT) has been proposed to improve carrier detection beyond family history (FH)-guided approaches, but economic evidence from Asian settings is lacking. We evaluated the cost-effectiveness of population-based GT strategies for HBOC and LS diagnosis in Singapore. Methods: We conducted cost-utility analyses from Singapore's healthcare system perspective with a lifetime horizon using decision trees and Markov models. For HBOC diagnosis, we compared population-based GT for BRCA1/2 , FH-guided GT, and no testing in cancer-unaffected women aged 30 years. For LS diagnosis, we compared population-based GT for MLH1 , MSH2 , MSH6 and PMS2 ; FH-guided GT, and no testing in cancer-unaffected individuals aged 30 years. Both models included cascade testing of first-degree relatives. Confirmed carriers received syndrome-specific enhanced surveillance and risk-reducing interventions. The willingness-to-pay threshold was S$75,000 per quality-adjusted life-year (QALY). Deterministic, probabilistic, and scenario analyses assessed uncertainty. Budget impact analyses estimated the five-year costs of the optimal strategy. Results: Population-based BRCA1/2 testing was the optimal strategy with an ICER of S$50,541/QALY relative to FH-guided BRCA1/2 testing with 99.93% probability of cost-effectiveness and averted one additional cancer per 1,000 women tested. Implementation of population-based BRCA1/2 testing requires an additional annual investment of S$1.69 million and identifies 182 additional BRCA1/2 carriers per annual cohort. For LS diagnosis, population-based GT was not cost-effective (ICER: S$179,564/QALY relative to FH-guided GT). FH-guided GT for LS had an ICER of S$75,210/QALY versus no testing but exhibited substantial decision uncertainty (45.44% vs 54.50% probability of cost-effectiveness), which resolved when cascade testing was extended to second-degree relatives (ICER: S$63,809/QALY). Cost-effectiveness outcomes were highly sensitive to surveillance adherence, cascade testing uptake, and LS prevalence. Conclusions: Population-based GT is cost-effective for BRCA1/2 but not for LS. FH-guided GT for LS approaches cost-effectiveness, particularly when cascade testing is extended to second-degree relatives. Future analyses should evaluate the integration of LS-associated genes into multigene hereditary cancer panels alongside BRCA1/2 to leverage shared infrastructure.