ECONOMIC MODEL ARCHITECTURE IN NICE EVALUATIONS FOR ULTRA-RARE NON-ONCOLOGY CONDITIONS: A COMPARATIVE ANALYSIS OF STRUCTURAL CHOICES AND CLINICAL TRIAL ENDPOINT USE ACROSS THE STA AND HST PATHWAYS
Author(s)
Sandra Milev, MSc1, Benjamin White, MSc1, Anshul Shah, B.pharm, MSc2, Monique Martin, MSc, MBA2.
1Red Nucleus, Yardley, PA, USA, 2Red Nucleus, London, United Kingdom.
1Red Nucleus, Yardley, PA, USA, 2Red Nucleus, London, United Kingdom.
OBJECTIVES: To characterize economic model structures and assess primary endpoint incorporation in NICE single technology appraisal (STA) evaluations for ultra-rare non-oncology conditions, as a companion to a previously published HST portfolio analysis enabling cross-pathway comparison.
METHODS: NICE STA appraisals (2017-2025) for non-oncology conditions with disease prevalence <1/50,000 were systematically reviewed. For each appraisal, economic model type, number of alive health states, primary endpoint classification and structural role, and ERG/committee commentary were extracted. Findings were compared with the entire HST portfolio (n=31).
RESULTS: Among the 36 included appraisals, Markov cohort predominated (67%; n=24), followed by decision-tree plus Markov (11%; n=4), cost-comparison analyses (8%; n=3), partitioned survival (6%; n=2), individual patient simulation (6%; n=2), and semi-Markov (3%; n=1). The median number of alive health states was 4 (range 1-9) versus 6 (range 2-12) in HST. Structural concerns were raised in 25% of appraisals (9/36), double the HST rate (13%; 4/31), concentrated in neuromuscular indications (6/8; 75%), driven by state architectures that failed to reflect clinically meaningful functional trajectories or disease progression. The primary endpoint served as a direct structural driver in 76% of evaluable appraisals (25/33) versus 45% in HST (14/31). Surrogate endpoints were accepted in 22% (8/36), consistently characterized as sources of high uncertainty due to limited prospective validation.
CONCLUSIONS: Compared with HST, NICE STA evaluations for ultra-rare conditions favor Markov cohort models of lower structural complexity, with more frequent direct use of primary endpoints as structural drivers and a higher rate of structural concern. The elevated structural concern rate, concentrated in neuromuscular disorders, underscores persistent tensions between trial endpoint design and model validity, reinforcing the need for HTA-relevant outcome selection at trial design and prospective surrogate validation to optimize market access.
METHODS: NICE STA appraisals (2017-2025) for non-oncology conditions with disease prevalence <1/50,000 were systematically reviewed. For each appraisal, economic model type, number of alive health states, primary endpoint classification and structural role, and ERG/committee commentary were extracted. Findings were compared with the entire HST portfolio (n=31).
RESULTS: Among the 36 included appraisals, Markov cohort predominated (67%; n=24), followed by decision-tree plus Markov (11%; n=4), cost-comparison analyses (8%; n=3), partitioned survival (6%; n=2), individual patient simulation (6%; n=2), and semi-Markov (3%; n=1). The median number of alive health states was 4 (range 1-9) versus 6 (range 2-12) in HST. Structural concerns were raised in 25% of appraisals (9/36), double the HST rate (13%; 4/31), concentrated in neuromuscular indications (6/8; 75%), driven by state architectures that failed to reflect clinically meaningful functional trajectories or disease progression. The primary endpoint served as a direct structural driver in 76% of evaluable appraisals (25/33) versus 45% in HST (14/31). Surrogate endpoints were accepted in 22% (8/36), consistently characterized as sources of high uncertainty due to limited prospective validation.
CONCLUSIONS: Compared with HST, NICE STA evaluations for ultra-rare conditions favor Markov cohort models of lower structural complexity, with more frequent direct use of primary endpoints as structural drivers and a higher rate of structural concern. The elevated structural concern rate, concentrated in neuromuscular disorders, underscores persistent tensions between trial endpoint design and model validity, reinforcing the need for HTA-relevant outcome selection at trial design and prospective surrogate validation to optimize market access.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EE414
Topic
Economic Evaluation, Health Technology Assessment
Disease
Rare & Orphan Diseases