The cost effectiveness of pembrolizumab versus chemotherapy or atezolizumab as second-line therapy for advanced urothelial carcinoma in the United States

Aims: Pembrolizumab demonstrated significantly prolonged overall survival (OS) vs. chemotherapy in the Phase III KEYNOTE-045 trial, and is approved in the US for the treatment of patients with locally advanced or metastatic urothelial carcinoma (mUC) who progressed after platinum-based chemotherapy. Using longer follow-up and individual patient-data from KEYNOTE-045, this study evaluates the cost-effectiveness of pembrolizumab vs. chemotherapy or atezolizumab from a US payer perspective. Materials and methods: A partitioned-survival model was developed over a 20-year time horizon. Progression-free survival (PFS) and OS for pembrolizumab and chemotherapy were extrapolated using a piecewise modelling approach, where patient-level data from KEYNOTE-045 were used for the initial period followed by parametric distributions. OS of atezolizumab was estimated by indirect treatment comparisons based on KEYNOTE-045 and IMvigor211. Different scenarios were explored in the absence of indirect comparisons on PFS and time-on-treatment (ToT) between pembrolizumab and atezolizumab. Drug acquisition/administration, disease management, adverse events, and terminal care costs were considered. Results: Compared with chemotherapy, pembrolizumab resulted in a mean gain of 1. 33 life-years and 1. 14 quality-adjusted life-years (QALYs) and an incremental cost of 106, 299, yielding an incremental cost-effectiveness ratio of 93, 481/QALY gained. Pembrolizumab dominated atezolizumab in extending patients’ life by 0. 89 years and 0. 76 QALYs, while reducing costs by 26, 458. Key drivers of cost-effectiveness included survival extrapolation, OS hazard ratio of pembrolizumab vs. atezolizumab, and time horizon. Pembrolizumab had a 66% and 100% probability of being cost-effective vs. chemotherapy and atezolizumab, respectively, at a 100, 000 willingness-to-pay threshold. Limitations and conclusions: Uncertainties remain with extrapolated PFS and OS for pembrolizumab, OS indirect comparison, and ToT for atezolizumab. Despite these limitations, the model used robust methods to estimate key clinical endpoints with patient-level data from longer follow-up of KEYNOTE-045. Pembrolizumab dominates atezolizumab and is very likely cost-effective vs. chemotherapy in 2 L mUC at a 100, 000 willingness-to-pay threshold.