CLINICAL UTILITY OF ENHANCED LIVER FIBROSIS (ELF)-GUIDED REFERRAL STRATEGIES: A DECISION CURVE ANALYSIS IN US VETERANS
Author(s)
Artem T. Boltyenkov, MBA, PhD1, Toana Kawashima, MS2, Wanyi Chen, PhD3, Donald B Chalfin, MD4, Stephanie Chang, MD5, Ramsey Cheung, MD5.
1Head, Global HEOR, Siemens Healthcare Diagnostics Inc., Lexington, SC, USA, 2Siemens Healthineers, Issaquah, WA, USA, 3Siemens Healthineers, Walpole, MA, USA, 4Siemens Healthineers, Newark, DE, USA, 5VA, Palo Alto, CA, USA.
1Head, Global HEOR, Siemens Healthcare Diagnostics Inc., Lexington, SC, USA, 2Siemens Healthineers, Issaquah, WA, USA, 3Siemens Healthineers, Walpole, MA, USA, 4Siemens Healthineers, Newark, DE, USA, 5VA, Palo Alto, CA, USA.
OBJECTIVES: To evaluate the clinical utility of non-invasive testing (NIT) strategies incorporating the Enhanced Liver Fibrosis (ELF) test for optimizing specialist referral in patients with metabolic dysfunction-associated steatotic liver disease (MASLD), with emphasis on threshold-dependent decision-making.
METHODS: We conducted a prospective observational cohort study of 400 adults with MASLD within a US Veterans Affairs health system (2023-2024). Advanced fibrosis (F3) was defined using vibration-controlled transient elastography (VCTE ≥9.5 kPa), with magnetic resonance elastography (MRE ≥3.6 kPa) used for imputation when VCTE was unavailable. Decision Curve Analysis (DCA) quantified clinical utility across threshold probabilities (1%-20%) using Net Benefit (NB). Six referral strategies were evaluated: clinical baseline (BMI), FIB-4 (1.3 cutoff), ELF (9.0 and 9.8 cutoffs), and sequential and additive FIB-4+ELF pathways. Models were adjusted for demographic and clinical covariates.
RESULTS: Among 365 evaluable patients, advanced fibrosis prevalence was 19.2%. Optimal strategies varied by decision threshold. At low (1%) and high thresholds (17.5%-20%), ELF using a 9.0 cutoff yielded the highest net benefit. Across intermediate thresholds (2.5%-15%), two-test strategies combining FIB-4 and ELF (sequential or additive) consistently demonstrated maximal clinical utility. In contrast, strategies based on clinical factors alone, universal referral, or FIB-4 alone did not achieve maximal net benefit at any evaluated threshold. Overall, ELF-based strategies also improved efficiency by reducing unnecessary referrals while maintaining detection of advanced fibrosis.
CONCLUSIONS: Across all clinically relevant decision thresholds, ELF-based approaches, whether used as a standalone test or in combination with FIB-4, consistently provide superior clinical utility for referral decision-making. These findings support incorporation of ELF into MASLD diagnostic pathways to improve targeting of high-risk patients and optimize the use of specialist care resources.
METHODS: We conducted a prospective observational cohort study of 400 adults with MASLD within a US Veterans Affairs health system (2023-2024). Advanced fibrosis (F3) was defined using vibration-controlled transient elastography (VCTE ≥9.5 kPa), with magnetic resonance elastography (MRE ≥3.6 kPa) used for imputation when VCTE was unavailable. Decision Curve Analysis (DCA) quantified clinical utility across threshold probabilities (1%-20%) using Net Benefit (NB). Six referral strategies were evaluated: clinical baseline (BMI), FIB-4 (1.3 cutoff), ELF (9.0 and 9.8 cutoffs), and sequential and additive FIB-4+ELF pathways. Models were adjusted for demographic and clinical covariates.
RESULTS: Among 365 evaluable patients, advanced fibrosis prevalence was 19.2%. Optimal strategies varied by decision threshold. At low (1%) and high thresholds (17.5%-20%), ELF using a 9.0 cutoff yielded the highest net benefit. Across intermediate thresholds (2.5%-15%), two-test strategies combining FIB-4 and ELF (sequential or additive) consistently demonstrated maximal clinical utility. In contrast, strategies based on clinical factors alone, universal referral, or FIB-4 alone did not achieve maximal net benefit at any evaluated threshold. Overall, ELF-based strategies also improved efficiency by reducing unnecessary referrals while maintaining detection of advanced fibrosis.
CONCLUSIONS: Across all clinically relevant decision thresholds, ELF-based approaches, whether used as a standalone test or in combination with FIB-4, consistently provide superior clinical utility for referral decision-making. These findings support incorporation of ELF into MASLD diagnostic pathways to improve targeting of high-risk patients and optimize the use of specialist care resources.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
MSR5
Topic
Health Service Delivery & Process of Care, Medical Technologies, Methodological & Statistical Research
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity)