A Review of Cost-Effectiveness Studies of Pembrolizumab Regimens for the Treatment of Advanced Non-small Cell Lung Cancer

Abstract

Pembrolizumab monotherapy or combination therapy is an approved treatment for various advanced non-small cell lung cancer (NSCLC) indications. We review published cost-effectiveness analyses (CEAs) of pembrolizumab as treatment for NSCLC and provide in-depth assessment of their methodologies. Fourteen studies were selected through searches of the PubMed database. Modeling approaches, survival and cost estimation, and utility analyses were compared and evaluated. These publications covered regulatory-approved pembrolizumab NSCLC indications based on the following randomized clinical trials: KEYNOTE-010 (one publication), KEYNOTE-024 (six), KEYNOTE-042 (four), KEYNOTE-189 (two), and KEYNOTE-407 (one). Differences were observed in health states (progression free, progressed disease, and death vs stable disease, progressed disease, death, and treatment discontinuation), modeling approaches (partitioned survival vs Markov), survival extrapolation/transition probability estimation, inclusion of additional costs to drug, disease management and adverse event costs (e.g., programmed death-ligand 1 [PD-L1] testing, subsequent treatment, terminal care), treatment duration approaches (trial-based time on treatment vs treat to progression), utility sources (trial data vs literature), and utility analyses (time to death vs progression status). Certain aspects of variability across models were problematic, including deviation from observed treatment utilization within trials and predicted long-term mortality risks for pembrolizumab higher than historical real-world NSCLC mortality data prior to the availability of pembrolizumab. Consequently, results differed even among studies examining the same population and comparator within similar time intervals. Differences in methodology across CEAs may lead to distinct results and conclusions. Payers and policy makers should carefully examine study designs and assumptions and choose CEAs with greater validity and accuracy for evidence-based decision-making.