Cost-effectiveness of an artificial intelligence predictive model for guiding androgen deprivation therapy in intermediate-risk prostate cancer
Menée aux Etats-Unis à l'aide d'un modèle de simulation, cette étude estime le rapport coût-efficacité d'un modèle prédictif utilisant l'intelligence artificielle pour identifier, parmi les patients atteints d'un cancer de la prostate à risque intermédiaire de récidive, ceux pouvant bénéficier d'un traitement anti-androgénique après une radiothérapie
The ArteraAI Prostate Test (ArteraAI Inc.) is the first predictive biomarker for benefit of adding short-term androgen deprivation therapy (ADT) to radiotherapy for intermediate-risk prostate cancer. We evaluated the cost-effectiveness of ArteraAI to guide short-term ADT with a Markov model simulating 15-year outcomes for 71-year-old patients with intermediate-risk prostate cancer receiving radiotherapy using NRG/RTOG 9408 data on which ArteraAI was validated. Three strategies were compared: 1) all patients receive ADT (ADT-for-all), 2) only patients with unfavorable intermediate-risk prostate cancer receive ADT (National Comprehensive Cancer Network [NCCN]), and 3) only ArteraAI-positive patients receive ADT (ArteraAI). Costs and utilities obtained from Medicare claims and published literature were used to calculate incremental cost-effectiveness ratios (ICERs). A willingness-to-pay threshold of $100,000/QALY was chosen. The ADT-for-all strategy was dominated by the NCCN strategy. Compared with the NCCN strategy, the ArteraAI strategy lowered costs by $12,296 and improved effectiveness by 0.01 QALYs, and thus was dominant.
JNCI Cancer Spectrum , article en libre accès, 2026