Quality of life assessment of cabozantinib in patients with advanced hepatocellular carcinoma in the CELESTIAL trial

• Published in: European journal of cancer (1990) 2022-06, Vol.168, p.91-98
• Main Authors: Freemantle, Nick, Mollon, Patrick, Meyer, Tim, Cheng, Ann-Lii, El-Khoueiry, Anthony B., Kelley, Robin K., Baron, Ari D., Benzaghou, Fawzi, Mangeshkar, Milan, Abou-Alfa, Ghassan K.
• Format: Article
• Language: English
• Subjects: Anilides - therapeutic use; Cabozantinib; Carcinoma; Carcinoma, Hepatocellular - drug therapy; Carcinoma, Hepatocellular - pathology; Confidence intervals; Generalized linear models; Hepatocellular; Hepatocellular carcinoma; Humans; Inhibitor drugs; Life assessment; Liver cancer; Liver neoplasms; Liver Neoplasms - drug therapy; Liver Neoplasms - pathology; Patient health questionnaire; Patient Reported Outcome Measures; Patients; Placebos; Pyridines - therapeutic use; Quality assessment; Quality of Life; Quality-adjusted life years; Risk assessment; Statistical analysis; Statistical models; Targeted cancer therapy; Visual analog scale
• ISSN: 0959-8049; 1879-0852
• DOI: 10.1016/j.ejca.2022.03.021

The CELESTIAL trial (NCT01908426) demonstrated overall survival benefit for cabozantinib versus placebo in patients with advanced hepatocellular carcinoma (aHCC) who had received prior sorafenib treatment. This analysis of CELESTIAL compared the impact of cabozantinib versus placebo on health-related quality of life (HRQoL). Health status was assessed using the EuroQol five-dimension five-level (EQ-5D-5L) questionnaire over the 800-day follow-up period. EQ-5D-5L health states were mapped to health utility scores using reference values for the UK population. Quality-adjusted life years (QALYs) were calculated for each treatment group as the area under the curve for the plot of health utility score over time. The between-treatment group difference in restricted mean QALYs was calculated by generalized linear models and adjusted for baseline differences. A difference of 0.08 in health utility score (or in QALY) was deemed a minimally important difference and to be clinically significant. At week 5, the difference in mean health utility score between cabozantinib and placebo was −0.097 (95% confidence interval [95% CI]: −0.126, −0.067; p ≤ 0.001). Between-group differences in health utility scores diminished over time and were generally non-significant. The cabozantinib group accrued more QALYs than the placebo group over follow-up. Differences in mean QALYs (cabozantinib minus placebo) were statistically and clinically significant, ranging from +0.092 (95% CI: 0.016, 0.169) to +0.185 (95% CI: 0.126, 0.243) in favour of cabozantinib, depending on the reference value set used. These HRQoL findings support a positive benefit–risk profile for cabozantinib in previously treated patients with aHCC. •This analysis of CELESTIAL compared the impact of cabozantinib vs placebo on HRQoL•Randomisation to cabozantinib led to a small short-term decrement in HRQoL•Cabozantinib accrued greater HRQoL vs placebo over follow-up to 3 years•This result adds to a positive benefit–risk profile for cabozantinib