OBJECTIVE:Pembrolizumab + chemotherapy substantially extends life expectancy for metastatic non-small cell lung cancer (NSCLC) patients. Its cost-effectiveness (CE) was previously evaluated based on interim trial analyses (follow-up ∼1 year). The present analysis describes CE incorporating additional follow-up based on protocol-specified final trial analyses (1-1.5 years additional follow-up), from a US healthcare payer perspective.
METHODS:A partitioned survival model is used to compare pembrolizumab + chemotherapy vs chemotherapy using data from the KN189 (non-squamous patients) and KN407 (squamous patients) clinical trials. An indirect treatment comparison vs pembrolizumab monotherapy is made for patient subgroups with PD-L1 TPS ≥50% and 1-49% based on data from the KN024 and KN042 trials. Efficacy, treatment utilization, health utility, and safety data are derived from trials and projected over 20 years. Costs for drugs, non-drug disease management, and adverse events are also incorporated.
RESULTS:Overall, versus chemotherapy alone, pembrolizumab + chemotherapy is projected to increase life expectancy by 1.12 years (3.35 vs 2.23) and 0.67 years (3.17 vs 2.50) in non-squamous and squamous patients, respectively. Resultant ICERs ($158,030/QALY and $178,387/QALY) are below a US 3-times GDP per capita threshold ($195,000/QALY). ICERs vs chemotherapy also generally fall below the threshold within PD-L1 sub-groups (except in squamous PD-L1 < 1%, which may have differed due to small sample size) while ICERs vs pembrolizumab monotherapy in PD-L1 ≥ 50% and 1-49% sub-groups generally exceed it (except in squamous PD-L1 1-49%); largely a result of the higher drug acquisition cost of pembrolizumab + chemotherapy relative to differences in life expectancy.
CONCLUSIONS:Taken together, with longer-term trial follow-up and in the context of prior literature, in the US, one of the two options for pembrolizumab use (either pembrolizumab + chemotherapy or pembrolizumab monotherapy), represents a cost-effective treatment in virtually all non-squamous and squamous metastatic NSCLC patient populations and PD-L1 sub-groups evaluated.