HOUT-27. THE CHALLENGE OF HEALTH UTILITY VALUES FOR GLIOBLASTOMA PATIENTS WITH LONG-TERM SURVIVAL

Abstract BACKGROUND The EF-14 trial was the first randomized controlled trial to report a GBM survival rate greater than 10% at 5-years. The quality-adjusted life year (QALY) has been developed to estimate the economic impact of new treatments. For the assessment of QALYs, health-utility values are needed. Utilities are preference weights that are determined either through direct elicitation methods or generic preference-bases measures, however in oncology they do not account for patients with long-term survival. This approach may understate QALYs of long-term survivors, as patients alive 5-years after diagnosis will be assigned the same utility value as patients recently diagnosed. METHODS We performed a comprehensive review of the published literature regarding health-utility values in GBM patients through a Pubmed search using glioblastoma AND health utility OR glioblastoma AND health utilities OR glioblastoma and health preference. RESULTS A total of 77 publications were screened with only 3 relevant publications in health-economic context. All three publications use utilities for GBM derived from the same source (Garside et al.). These utilities were obtained from the NHS Value of Health Panel (VoHP) of which 36 members rated a total of 9 descriptive health-state scenarios. This approach assumes the constant presence of adverse events during the respective state and a constant decline in quality of life of patients in progressive state, which may particularly be inadequate for long-term survivors with sustained stable disease. CONCLUSIONS Health economic evaluations of GBM treatments use utilities from Garside. There are no published utilities for GBM in a general population sample, adequately adjusting the utilities with time from diagnosis. Deriving and validating health preference-based measures from existing condition-specific questionnaires (e.g. the QLQ C-30 with its brain cancer module BN20) specifically for GBM would be beneficial and allow for adjusted utility values with time from diagnosis as already developed in multiple myeloma.