Presentation (CTI)

OBJECTIVES: Non-small cell lung cancer (NSCLC) has a variable impact on health-related quality of life (HRQoL), influenced by disease progression and other disease-related factors, resulting in heterogeneous utilities reported in the literature. This study assessed the variability in health state utility values (HSUVs) applied across NICE technology appraisals (TAs) for second-line plus (2L+) NSCLC therapies across progression-free (PF), progressed-disease (PD), and time-to-death (TTD) health states.
METHODS: A targeted review was conducted of TAs in the 2L+ indication, published on the NICE website from 2015 -2025, to identify HSUVs accepted by NICE and associated estimation methods.
RESULTS: Fifteen appraisals were identified, all in advanced NSCLC; 13 used progression-based utilities and two used TTD-approaches. PF HSUVs ranged 0.693-0.81, while PD HSUVs ranged 0.46-0.665. PF HSUVs were mainly derived by mapping (n=5), direct elicitation (n=4), or precedent in prior TAs (n=4); in the latter a PF utility of 0.713 was consistently cited from TA713/TA484. Most PD utilities (n=6) were midpoint estimates based on values from literature/prior TAs and in-trial (n=4) or external trial (n=2) data. Two ALK+ TAs stratified PD utilities by treatment status (on-treatment: 0.65-0.732; off-treatment: 0.46-0.582), while two TAs reported PD HSUVs as plausible ranges for ALK+ NSCLC (0.74-0.78) and EGFR T790M+ NSCLC (0.64-0.678). For those reporting TTD-based utilities (n=2), one TA estimated PF/PD utilities ≥30 days before death (0.763/0.675) and <30 days before death (0.284/0.320), while the other TA reported on- and off-treatment utilities to decline from 0.77 and 0.68 respectively, to 0.39 and 0.35, respectively, between 30 weeks and ≤5 weeks before death.
CONCLUSIONS: Over the past decade, NICE favoured progression-based utilities in 2L+ NSCLC appraisals, although the accepted HSUVs remain heterogeneous. Limited adoption of TTD-approaches suggests inconsistent capture of end-of-life HRQoL across appraisals. Standardised HSUV derivation could strengthen future NSCLC economic evaluations.