Pembrolizumab with or without bevacizumab for recurrent or metastatic cervical cancer: A cost-effectiveness analysis.

ABSTRACT

OBJECTIVE: To determine the cost-effectiveness of the addition of pembrolizumab in various combinations in patients with recurrent/metastatic cervical cancer. METHODS: A decision-analysis model evaluated the cost-effectiveness of chemotherapy plus pembrolizumab and bevacizumab (CPB) relative to chemotherapy plus pembrolizumab (CP) and chemotherapy plus bevacizumab (CB) in cervical cancer patients. Data from KEYNOTE-826 was used to estimate quality-adjusted life-years (QALYs). Drug cost estimates were obtained using average wholesale prices. Incremental cost-effectiveness ratios (ICERs) were calculated to determine cost/QALY. The willingness-to-pay threshold (WTP) was set a $100,000/QALY. Sensitivity analyses were performed on cost and effectiveness for pembrolizumab-containing regimens. RESULTS: Cost of treatment with CB, CP, and CPB were $416 million (M), $713 M, and $1.51 billion, respectively. Relative to CB, the ICER for CP was $92,678. CPB was dominated. Sensitivity analyses were performed varying the cost and efficacy of CP and CPB. If overall survival (OS) with CP decreased from 24.4 months to 23.4 months, the ICER would exceed the WTP. If the OS from CP is assumed to be 20.4 months, the ICER increases to $187,746. The ICER for CP improves to $63,670 when the model is restricted to PD-L1 positive cancers. With CP eliminated, CPB becomes cost-effective relative to CB if the cost of pembrolizumab per cycle decreases from $12,080 to $2913 for the baseline model and to $4644 for the PD-L1 model. CONCLUSIONS: CP is cost-effective relative to CB for recurrent or metastatic cervical cancer. The efficacy of CPB would need to far exceed both CB and CP to be cost-effective. Restricting the model to patients with PD-L1 positive tumors dramatically improves the ICER for CP relative to CB by $30,000/QALY.