Cost effectiveness of axicabtagene ciloleucel versus tisagenlecleucel for adult patients with relapsed or refractory large B-cell lymphoma after two or more lines of systemic therapy in the United States

To assess from a US payer perspective the cost-effectiveness of the chimeric antigen receptor T (CAR T)-cell therapies axicabtagene ciloleucel (axi-cel) and tisagenlecleucel (tisa-cel) to treat relapsed or refractory (r/r) large B-cell lymphoma (LBCL) following ≥2 systemic therapy lines.

A three-state (i.e. pre-progression, post-progression, and death) partitioned survival model was used to estimate the quality-adjusted life-years (QALYs) and costs for patients on each treatment over a lifetime horizon. Progression-free survival (PFS) and overall survival (OS) were based on a matching-adjusted indirect treatment comparison (MAIC) that accounted for differences in trial population baseline characteristics. Mixture cure models (MCMs) were used to account for long-term survivors. Costs included drug acquisition and administration for the CAR T-cell therapies and conditioning chemotherapy, apheresis, CAR T-specific monitoring, transplant, hospitalization, adverse events, routine care, and terminal care. Health state utilities were derived from trial and published data. Sensitivity analyses included probabilistic sensitivity analyses (PSAs) and an analysis of extremes that assessed the results across a vast array of combinations of parametric OS and PFS curves across the two therapies.

Compared to tisa-cel, axi-cel resulted in 2.31 QALYs gained and a cost reduction of $1,407 in the base case. In the PSA, the cost per QALY gained was ≤$31,500 in 95% of the 1,000 simulations. In the analysis of extremes, the cost per QALY gained was ≤$7,500 in 99% of the 1,296 combinations of MCMs and ≤$40,000 in 95% of the 1,296 combinations of standard models.

In absence of head-to-head comparative data, we relied on a MAIC, which cannot account for all possible confounders. Moreover, some outcomes (i.e. transplantations, hospitalizations, adverse events (AEs)) were not adjusted in the MAIC.

In this simulation, axi-cel was a superior treatment option as it is predicted to achieve better outcomes at lower or minimal incremental costs versus tisa-cel.