The correlation between the costs and clinical benefits of PD-1/PD-L1 inhibitors in malignant tumors: An evaluation based on ASCO and ESMO frameworks

Background: Life expectancy for patients with malignant tumors has been significantly improved since the presence of the programmed cell death protein-1/programmed cell death protein ligand-1 (PD-1/PD-L1) inhibitors in 2014, but they impose heavy financial burdens for patients, the healthcare system and the nations. The objective of this study was to determine the survival benefits, toxicities, and monetary of programmed cell death protein-1/programmed cell death protein ligand-1 inhibitors and quantify their values. Methods: Randomized controlled trials (RCTs) of PD-1/PD-L1 inhibitors for malignant tumors were identified and clinical benefits were quantified by American Society of Clinical Oncology Value Framework (ASCO-VF) and European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). The drug price in Micromedex REDBOOK was used to estimate monthly incremental drug costs (IDCs) and the correlation between clinical benefits and incremental drug costs of experimental and control groups in each randomized controlled trial, and the agreement between two frameworks were calculated. Results: Up to December 2022, 52 randomized controlled trials were included in the quantitative synthesis. All the randomized controlled trials were evaluated by American society of clinical oncology value framework, and 26 (50%) met the American society of clinical oncology value framework “clinical meaningful value.” 49 of 52 randomized controlled trials were graded by European society for medical oncology magnitude of clinical benefit scale, and 30 (61.2%) randomized controlled trials achieved European Society for Medical Oncology criteria of meaningful value. p-values of Spearman correlation analyses between monthly incremental drug costs and American society of clinical oncology value framework/European society for medical oncology magnitude of clinical benefit scale scores were 0.9695 and 0.3013, respectively. In addition, agreement between two framework thresholds was fair (κ = 0.417, p = 0.00354). Conclusion: This study suggests that there might be no correlation between the cost and clinical benefit of programmed cell death protein-1/programmed cell death protein ligand-1 inhibitors in malignancy, and the same results were observed in subgroups stratified by drug or indication. The results should be a wake-up call for oncologists, pharmaceutical enterprises and policymakers, and meanwhile advocate the refining of American Society of Clinical Oncology and European Society for Medical Oncology frameworks.