Abstract
Objectives
To determine the cost-effectiveness of combined durvalumab and tremelimumab in patients with metastatic colorectal cancer in the intention-to-treat (ITT) and biomarker-enriched populations using direct CCTG CO.26 phase-2 trial data.
Methods
A 4-state microsimulation model was used to evaluate the expected health outcomes in quality-adjusted life years (QALYs) and costs (2023 Canadian Dollars) over a lifetime horizon (5 years) from the Canadian public-payer perspective. Direct phase 2 CCTG CO.26 trial data informed model inputs, including overall survival Kaplan-Meier curves, progression-free survival Kaplan-Meier curves, and adverse event rates. Health-state utilities and costs of therapy, hospitalization, end-of-life care, sequencing panels, and physician care were obtained from published literature and Canadian costing databases. The incremental cost-utility ratios (ICURs) for the ITT and biomarker-enriched populations were determined.
Results
In the ITT population, expected QALYs for the treatment and best supportive care arms were 0.47 and 0.33 (incremental (Δ)0.14), respectively, and expected costs were $56 743 and $17 177 (Δ$39 566) for an ICUR of $277 661/QALY. In the plasma tumor mutation burden > 28 subgroup, expected QALYs were 0.43 and 0.21 (Δ0.21) and expected costs were $58 498 and $16 941 (Δ$41 557) for an ICUR of $193 945/QALY.
Conclusions
Combined durvalumab and tremelimumab is not cost-effective in refractory metastatic colorectal cancer per conventional cost-effectiveness thresholds. Cost-effectiveness is more favorable in the high-plasma tumor mutation burden subgroup, but costs of screening and cutoffs used must be considered.
Authors
Monish Ahluwalia Ambica Parmar Jonathan M. Loree Christopher J. O’Callaghan Dongsheng Tu Kelvin K.W. Chan