Estimation of Health State Utility Values for Patients Undergoing First-Line Treatment of Advanced-Stage Classical Hodgkin Lymphoma in the HD21 Trial

Introduction: HD21 (NCT02661503) is a phase 3, randomized, open-label study evaluating 4 or 6 cycles of BrECADD vs eBEACOPP as a first-line treatment (tx) of advanced classical Hodgkin lymphoma.Results showed that BrECADD had a more favorable toxicity profile and improved progression-free survival compared with eBEACOPP. Health-related quality of life (HRQoL) was assessed using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire - Core 30 (EORTC QLQ-C30); however, preference-based measures required for economic evaluations, such as the EQ-5D-3L, were not administered. This study aimed to identify and apply an appropriate algorithm to map the collected EORTC QLQ-C30 data to EQ-5D-3L utility values and estimate utility values for relevant health states. Methods: A targeted literature review (TLR) was conducted to identify relevant studies mapping the EORTC QLQ-C30 to the EQ-5D using the Embase, Medline, and Oxford Health Economic Research Centre databases; studies of adults with Hodgkin lymphoma and other blood cancers where EORTC QLQ-C30 was directly or indirectly mapped to EQ-5D were included. Once an appropriate mapping model was identified, it was applied to the HD21 EORTC QLQ-C30 data to derive EQ-5D-3L health utility scores using the UK value set. All analyses were conducted on the EQ-5D-evaluable population, which consisted of those with non-missing calculated EQ-5D-3L health utility values at baseline and at least one post-baseline assessment. A longitudinal linear mixed-effect model was used to estimate health utility values for the following health states: progression-free (PrF) on BrECADD tx, PrF on eBEACOPP tx, PrF off BrECADD tx, PrF off eBEACOPP tx, and progressed. On-tx was defined as visits while patients were undergoing chemotherapy, while off-tx was defined as visits after completion of chemotherapy. The model included EQ-5D-3L health utility as the dependent variable and the following covariates: treatment arm (BrECADD/eBEACOPP), on/off-tx, BrECADD on treatment (interaction term; yes/no), progressive disease (yes/no), age (continuous), and baseline utility (continuous). Least-square (LS) mean (95% confidence interval [CI]) utility values were estimated for the health states above. Results: The TLR (search date: 25 Jan 2024) identified 150 records. Of the 5 full-text articles that were eligible for inclusion, the study by Young et al. 2015 (PMID: 25997920) was chosen as the most appropriate due to its large sample size (N=771, 74% with a blood cancer), wide usage in the literature, and use of an indirect approach, allowing for flexibility in the country value sets to be used. In HD21, 751 patients were randomized to BrECADD and 749 to eBEACOPP (intent-to-treat [ITT] population), of which 342 and 347 patients, respectively, formed the EQ-5D-evaluable population. Demographic and disease characteristics were generally comparable between treatment groups and between the ITT and EQ-5D-evaluable populations, except that the latter consisted of mostly (99.9%) patients from Germany, where the QLQ-C30 was mainly administered. At baseline, mean (95% CI) utility values were similar for the BrECADD and eBEACOPP groups (0.62 [0.59-0.65] vs. 0.65 [0.62-0.68], respectively). In both treatment groups, after patients completed chemotherapy, mean health utility values increased and remained stable across visits while patients remained PrF. On/off-tx, progressive disease, age, and baseline utility score were significant predictors of utility (P<0.05). The mean utility values predicted by the linear mixed-effect model aligned closely with the mapped values for both treatment arms. PrF utility estimates (LS mean [95% CI]) were similar for BrECADD vs eBEACOPP, but higher off-tx (after completing chemotherapy; BrECADD 0.76 [0.74-0.78], eBEACOPP 0.73 [0.71-0.76]) vs on-tx (BrECADD 0.59 [0.57-0.61], eBEACOPP 0.57 [0.55-0.59]). The utility estimate for the progressed state (0.66 [0.57-0.75]) fell between the PrF on-tx and off-tx estimates. Conclusions: Completion of chemotherapy resulted in improved utility in the PrF state; thus, improvement of progression-free survival by BrECADD could lead to an increase in health utilities.