BACKGROUND: The European Society for Medical Oncology (ESMO) and the National Comprehensive Cancer Network (NCCN) provided different recommendations for the second-line treatment of advanced endometrial carcinoma (EC). ESMO recommends the combination of lenvatinib and pembrolizumab (LP) for all patients, while the NCCN suggests LP for patients with proficient mismatch repair (pMMR) and pembrolizumab for those with deficient mismatch repair (dMMR).

OBJECTIVES: This study aims to evaluate the cost-effectiveness of the universal LP regimen recommended by ESMO as a second-line treatment for advanced EC, compared to the biomarker-based treatment regimen proposed by the NCCN, within the context of Taiwan's National Health Insurance (NHI).

METHODS: A 20-year partitioned survival model was employed using the efficacy results from the KEYNOTE-775 and KEYNOTE-158 trials. The prices of LP referred to the fee schedule set by Taiwan NHI. Direct medical costs were estimated based on NHI claim data. Health state utilities and disutilities associated with adverse events were derived from relevant literature sources. The willingness-to-pay threshold was set as three times the GDP per capita in 2022 (equivalent to NT$2,925,582). Quality-adjusted life-years (QALYs) and costs were discounted at a rate of 3%. Deterministic and probabilistic sensitivity analyses were conducted to assess the uncertainty.

RESULTS: The ESMO-recommended universal LP regimen, when compared to the NCCN-recommended treatment regimen, led to a cost reduction of NT$821,018 but a decrease of 0.6 QALYs. The analysis revealed a negative net monetary benefit (-NT$937,915) and an exceedingly low acceptance probability (2%). The time horizon, health state utilities, and medication cost of lenvatinib emerged as the primary sources of uncertainty influencing the results.

CONCLUSIONS: Within the framework of Taiwan's NHI, the NCCN's biomarker-based regimen emerges as a more cost-effective treatment option for patients with advanced EC compared to ESMO's recommendation.