Budget Impact Model of Enhanced Lung Cancer Screening With AI/ML Tech-Based Software as a Medical Device (SaMD) on a US Cohort and Private Payer Perspective
Author(s)
Antoine Disset, PhD1, Charles Voyton, .2, Danny Quach, .3, Eric Lam, PhD4.
1VP market access & Gov Affairs, median technologies, Valbonne, France, 2Median Technologies, Valonne, France, 3University of Illinois at Chicago College of Pharmacy, Chicago, IL, USA, 4Avania, Boston, MA, USA.
1VP market access & Gov Affairs, median technologies, Valbonne, France, 2Median Technologies, Valonne, France, 3University of Illinois at Chicago College of Pharmacy, Chicago, IL, USA, 4Avania, Boston, MA, USA.
Presentation Documents
OBJECTIVES: Lung cancer remains a leading cause of cancer-related mortality worldwide, highlighting the need for cost-effective screening solutions. This study evaluates the budget impact and clinical outcomes of a US-based lung cancer screening cohort implementing eyonis® LCS, an AI/ML-based Software as a Medical Device (SaMD) designed to detect and characterize lung nodules, compared to LDCT-only screening.
METHODS: A comprehensive Health Economic Outcomes Model (HEOM) was constructed, incorporating a 5-stage Markov process to simulate lung cancer progression. The model compared two approaches: AI-unaided and AI-aided screening with eyonis® LCS. Patient management was based on Lung-RADS classifications, incorporating sensitivity, specificity, and downstream invasive procedures and treatments with associated costs (e.g., PET-CT, biopsy). A Per Member Per Month (PMPM) framework was used to evaluate outcomes and cost implications annually over five years.
RESULTS: AI-aided screening with eyonis® LCS outperformed LDCT alone, showing a 16% increase in true positive detections and a 67% reduction in false negatives. Over five years, the model projected $52 million in savings for a cohort of 1 million covered lives, primarily due to fewer unnecessary invasive diagnostic procedures (-89% PET scans, -88% biopsies in Year 1) and earlier-stage cancer detection. The average PMPM cost reduction in Year 1 was -$1.55, driven by decreases in late-stage cancer management expenses and diagnostic inefficiencies. Savings were most pronounced in Year 1, with sustained incremental benefits through Year 5 as fewer patients progressed to advanced cancer stages.
CONCLUSIONS: AI-aided screening with eyonis® LCS demonstrates significant clinical utility and cost savings from Year 1 through Year 5 compared to Standard of Care. By improving early detection and characterization, reducing unnecessary procedures, and lowering late-stage cancer care costs, AI-aided screening offers a valuable solution for national lung cancer screening programs. These findings support adopting eyonis® LCS to enhance patient outcomes and optimize healthcare resources.
METHODS: A comprehensive Health Economic Outcomes Model (HEOM) was constructed, incorporating a 5-stage Markov process to simulate lung cancer progression. The model compared two approaches: AI-unaided and AI-aided screening with eyonis® LCS. Patient management was based on Lung-RADS classifications, incorporating sensitivity, specificity, and downstream invasive procedures and treatments with associated costs (e.g., PET-CT, biopsy). A Per Member Per Month (PMPM) framework was used to evaluate outcomes and cost implications annually over five years.
RESULTS: AI-aided screening with eyonis® LCS outperformed LDCT alone, showing a 16% increase in true positive detections and a 67% reduction in false negatives. Over five years, the model projected $52 million in savings for a cohort of 1 million covered lives, primarily due to fewer unnecessary invasive diagnostic procedures (-89% PET scans, -88% biopsies in Year 1) and earlier-stage cancer detection. The average PMPM cost reduction in Year 1 was -$1.55, driven by decreases in late-stage cancer management expenses and diagnostic inefficiencies. Savings were most pronounced in Year 1, with sustained incremental benefits through Year 5 as fewer patients progressed to advanced cancer stages.
CONCLUSIONS: AI-aided screening with eyonis® LCS demonstrates significant clinical utility and cost savings from Year 1 through Year 5 compared to Standard of Care. By improving early detection and characterization, reducing unnecessary procedures, and lowering late-stage cancer care costs, AI-aided screening offers a valuable solution for national lung cancer screening programs. These findings support adopting eyonis® LCS to enhance patient outcomes and optimize healthcare resources.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
OP9
Topic
Organizational Practices
Disease
SDC: Oncology, SDC: Respiratory-Related Disorders (Allergy, Asthma, Smoking, Other Respiratory)