Comparative Health Economic Evaluation of Alzheimer’s Diagnostic Treatment Strategies in the Era of Amyloid-Targeting Therapies: A Conceptual Model Framework
Author(s)
Arturo Cabra, BS, MSc1, Montserrat Chivite, MSc2, Gill Farrar, PhD3, Emilija Veljanoska, MSc4, Anna Tokarz, MSc5, Sofia Gomes, PharmD6, Agota Szende, PhD7.
1GE HealthCare, Miami, FL, USA, 2GE Healthcare, Madrid, Spain, 3GE Healthcare, Little Chalfont, United Kingdom, 4Fortrea, Munich, Germany, 5Fortrea, Warsaw, Poland, 6Fortrea, Coimbra, Portugal, 7Fortrea, Leeds, United Kingdom.
1GE HealthCare, Miami, FL, USA, 2GE Healthcare, Madrid, Spain, 3GE Healthcare, Little Chalfont, United Kingdom, 4Fortrea, Munich, Germany, 5Fortrea, Warsaw, Poland, 6Fortrea, Coimbra, Portugal, 7Fortrea, Leeds, United Kingdom.
OBJECTIVES: Amyloid-targeting therapies (ATTs) for Alzheimer’s disease (AD), lecanemab and donanemab, require confirmation of β-amyloid pathology prior to initiation. Diagnostic modalities include amyloid PET, cerebrospinal fluid (CSF), and plasma-biomarkers. This study describes the conceptualization of a health economic model assessing diagnostic accuracy, clinical outcomes, and cost-effectiveness of ¹⁸F-flutemetamol PET (with/without quantification), CSF, and plasma-biomarkers from a US payer perspective.
METHODS: A Markov model was developed to simulate diagnostic and treatment pathways in a cohort of patients aged 40-79 years with mild cognitive impairment (MCI) or suspected AD. The model incorporates diagnostic accuracy (sensitivity and specificity) determining patient distribution and transitions across disease states based on treatment strategy. The model follows the natural patient disease trajectory, informed by published literature, and includes six health states: MCI+/-AD, mild AD, moderate AD, severe AD, other dementia, and death. In health states beyond the scope of ATTs, transition probabilities from the symptomatic care arm were applied. Three diagnostic approaches (PET, CSF, plasma-biomarkers) and three treatment strategies (symptomatic care, lecanemab, donanemab) were evaluated. Quantitative PET analysis and treatment monitoring (via PET or MRI, the latter being mainly used to diagnose and monitor amyloid-related imaging abnormalities) were also considered. Model inputs were informed by published literature.
RESULTS: The model enables cost-effectiveness analysis by estimating incremental cost-effectiveness ratios for diagnostic-treatment combinations. Preliminary findings suggest that ¹⁸F-flutemetamol PET with quantification achieves the highest diagnostic accuracy (95%), outperforming CSF (85.4%) and plasma-biomarkers (86.4%). Quantitative PET also supports treatment monitoring, resulting in the application of a stopping rule and optimal resource allocation.
CONCLUSIONS: This model provides a robust framework for evaluating diagnostic-treatment strategies in AD. Incorporating β-amyloid quantification with ¹⁸F-flutemetamol PET enhances diagnostic precision, supports treatment monitoring, and may improve clinical outcomes and resource allocation. These findings can inform payer decision-making, optimizing diagnostic pathways in the context of ATTs.
METHODS: A Markov model was developed to simulate diagnostic and treatment pathways in a cohort of patients aged 40-79 years with mild cognitive impairment (MCI) or suspected AD. The model incorporates diagnostic accuracy (sensitivity and specificity) determining patient distribution and transitions across disease states based on treatment strategy. The model follows the natural patient disease trajectory, informed by published literature, and includes six health states: MCI+/-AD, mild AD, moderate AD, severe AD, other dementia, and death. In health states beyond the scope of ATTs, transition probabilities from the symptomatic care arm were applied. Three diagnostic approaches (PET, CSF, plasma-biomarkers) and three treatment strategies (symptomatic care, lecanemab, donanemab) were evaluated. Quantitative PET analysis and treatment monitoring (via PET or MRI, the latter being mainly used to diagnose and monitor amyloid-related imaging abnormalities) were also considered. Model inputs were informed by published literature.
RESULTS: The model enables cost-effectiveness analysis by estimating incremental cost-effectiveness ratios for diagnostic-treatment combinations. Preliminary findings suggest that ¹⁸F-flutemetamol PET with quantification achieves the highest diagnostic accuracy (95%), outperforming CSF (85.4%) and plasma-biomarkers (86.4%). Quantitative PET also supports treatment monitoring, resulting in the application of a stopping rule and optimal resource allocation.
CONCLUSIONS: This model provides a robust framework for evaluating diagnostic-treatment strategies in AD. Incorporating β-amyloid quantification with ¹⁸F-flutemetamol PET enhances diagnostic precision, supports treatment monitoring, and may improve clinical outcomes and resource allocation. These findings can inform payer decision-making, optimizing diagnostic pathways in the context of ATTs.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EE140
Topic
Economic Evaluation
Disease
Neurological Disorders