Effectiveness and Cost-effectiveness of Artificial Intelligence–assisted Pathology for Prostate Cancer Diagnosis in Sweden: A Microsimulation Study

• DOI: 10.1016/j.euo.2024.05.004
• Publication Date: 2024-05-23T19:29:10Z
• Authors: Xiaoyang Du; Shuang Hao; Henrik Olsson; Kimmo Kartasalo; Nita Mulliqi; Balram Rai; Dominik Menges; Emelie Heintz; Lars Egevad; Martin Eklund; Mark Clements
• Keywords: Overdiagnosis

Background and objective Image-based artificial intelligence (AI) methods have shown high accuracy in prostate cancer (PCa) detection. Their impact on patient outcomes cost effectiveness comparison to human pathologists remains unknown. Our aim was evaluate the cost-effectiveness of AI-assisted pathology for PCa diagnosis Sweden. Methods We modeled quadrennial prostate-specific antigen (PSA) screening men between ages 50 74 yr over a lifetime horizon using health care perspective. Men with PSA ≥3 ng/ml were referred standard biopsy (SBx), which cores either examined via AI followed by pathologist AI-labeled positive cores, or alone. The performance characteristics estimated an internal STHLM3 validation data set. Outcome measures included number tests, incidence mortality, overdiagnosis, quality-adjusted life years (QALYs), potential reduction pathologist-evaluated if used. Cost-effectiveness assessed incremental ratio. Key findings limitations In alone, workflow increased SBx procedures, deaths ≤0.03%, slightly reduced overdiagnosis. would reduce proportion evaluated 80%. At €10 per case, less result <0.001% lower QALYs results sensitive cost. Conclusions clinical implications According our model, significantly decrease workload pathologists, not affect quality life, yield savings Sweden when compared Patient summary patients relevant costs two assessing biopsies Sweden: (1) technology review pathologist; (2) alone all biopsies. found that addition save money, suggest adding costs.