Optimizing MASLD Diagnostic Protocols: A Microsimulation Analysis of Test Sensitivity Requirements
Author(s)
Osvaldo Ulises Garay, MSc1, Maria Karagianni, PhD1, Roland Fritz, PhD1, Brigitta Monz, MA, MPH, MD2.
1Roche Diagnostics International, Rotkreuz, Switzerland, 2Roche Diagnostics GmbH, Mannheim, Germany.
1Roche Diagnostics International, Rotkreuz, Switzerland, 2Roche Diagnostics GmbH, Mannheim, Germany.
OBJECTIVES: Evaluating the diagnostic performance of an intervention that occurs over time requires a shift from analyzing a single test to assessing the performance of the entire diagnostic protocol with repeated testing. This is especially pertinent for patients with Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD), for whom diagnostic strategies involve re-testing over years. Given MASLD's high incidence and low progression, prioritizing test sensitivity could cause significant financial strain due to costs associated with false positives. Our aim was to determine the minimal individual diagnostic protocol sensitivity needed to diagnose 90% of MASLD patients with significant or advanced fibrosis (F2-F3) before progression to cirrhosis (F4) or F2-F3-related mortality, across various scenarios.
METHODS: We developed a microsimulation model tracking 10,000 MASLD individuals through fibrosis stages (F0-F4). Detection was defined as F2-F3 prior to reaching F4 or death at F2-F3. Model parameters, including age, stage distribution, progression/regression rates, and mortality, were derived from published literature. Our base case assumed full adherence and annual testing, while scenario analyses explored the impact of reducing adherence and testing frequencies. We assumed a fixed test specificity of 90%.
RESULTS: The simulated cohort had an average age of 50 and a life expectancy of 81.4 years. Of these individuals, 5,601 developed F2-F3 fibrosis at some point in their lives, and 719 progressed to cirrhosis. Our base case scenario achieved a 90% case detection rate with a sensitivity of 32%. In case the adherence was reduced or testing was less frequent than annually, a higher test sensitivity to achieve the same detection target was observed.
CONCLUSIONS: Balancing diagnostic sensitivity and specificity within longitudinal monitoring protocols is complex. This model indicates that diagnostic protocols using tests with relatively low sensitivity can effectively identify a high proportion of MASLD patients.
METHODS: We developed a microsimulation model tracking 10,000 MASLD individuals through fibrosis stages (F0-F4). Detection was defined as F2-F3 prior to reaching F4 or death at F2-F3. Model parameters, including age, stage distribution, progression/regression rates, and mortality, were derived from published literature. Our base case assumed full adherence and annual testing, while scenario analyses explored the impact of reducing adherence and testing frequencies. We assumed a fixed test specificity of 90%.
RESULTS: The simulated cohort had an average age of 50 and a life expectancy of 81.4 years. Of these individuals, 5,601 developed F2-F3 fibrosis at some point in their lives, and 719 progressed to cirrhosis. Our base case scenario achieved a 90% case detection rate with a sensitivity of 32%. In case the adherence was reduced or testing was less frequent than annually, a higher test sensitivity to achieve the same detection target was observed.
CONCLUSIONS: Balancing diagnostic sensitivity and specificity within longitudinal monitoring protocols is complex. This model indicates that diagnostic protocols using tests with relatively low sensitivity can effectively identify a high proportion of MASLD patients.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
MT33
Topic
Medical Technologies
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity)