Evaluating the Utility of EQ5D-5L in Patients with Brain Metastases: Secondary Analysis of NRG-CC001

Abstract

<h3>Purpose/Objective(s)</h3> EuroQoL EQ5D has become the preferred preference-based tool used in oncology clinical trials to assess patient health states following different interventions. It allows for an efficient measure of patient quality-of-life (QoL) that can be used to calculate quality-adjusted life years required to perform cost-effectiveness analyses. NRG-CC001 established the neuroprotective effect of hippocampal avoidant-whole brain radiotherapy (HA-WBRT) plus memantine over WBRT plus memantine in patients with brain metastases. EQ5D results, collected as part of patient reported outcomes, demonstrated no differences between study arms. Although EQ5D was used in this trial, it is unknown if this tool is sensitive to QoL changes experienced by patients with brain metastases. The primary objective of this study was to assess the utility of EQ5D in reflecting neurocognitive function (NCF) changes in this patient population. <h3>Materials/Methods</h3> We performed a post hoc analysis of NRG-CC001 trial data, focusing on EuroQoL EQ5D measures. We used independent sample t-tests to assess mean differences of EQ5D index and visual analog scale (VAS) score changes from baseline (at 2, 4, and 6 months) between patients who experienced NCF failure and those who did not. We also compared mean changes in EQ5D for patients who experienced intracranial progression to those without intracranial progression. <h3>Results</h3> EQ5D index score and VAS score changes between those patients who had NCF failure and those that had intact NCF showed no statistically significant differences at 2, 4, and 6 months (Table). When stratifying by age (<=61, >61), education (<= high school, >high school), and KPS (70, >70), there was no significant EQ5D difference by treatment arm or by NCF change. There was a trend towards significance when assessing changes in EQ5D index and VAS score at 4 months (p=0.06 and 0.05, respectively), and a significant correlation at 6 months for VAS score change (p=0.04) based on progression of disease versus not. <h3>Conclusion</h3> EQ5D was not sensitive to changes in NCF following whole brain radiation therapy for patients with brain metastases, but was sensitive to changes in intracranial progression. Given that NCF is well-established as a patient-centric outcome measure following brain radiation, the results of this study challenge the routine use of EQ5D in brain metastases clinical trials that are focused on preventing NCF failure. There is an urgent need to identify preference-based health status measures that are sensitive to changes in NCF and can be utilized in clinical trials for patients with brain metastases.