The Value of Performance-Based Risk Sharing Arrangements in the United States: A Case Study of First-Line Pembrolizumab Therapy in Advanced Non-Small Cell Lung Cancer

Abstract

OBJECTIVES: The objective of this study was to evaluate the health and financial implications of implementing a PBRSA along with lower patient cost sharing for two distinct PBRSAs in a case study of first-line pembrolizumab (PEMB) therapy in patients with advanced NSCLC. METHODS: We evaluated two distinct PBRSAs based on surrogate endpoints from the pivotal KEYNOTE-024 trial for a hypothetical treated population of 50 patients. The first PBRSA is an example of a conditional treatment continuation (CTC-PBRSA) arrangement in which a 25% rebate or manufacturer (MFTR) risk share of PEMB costs is applied to all in patients the first four months and subsequently dropped for patients who achieve or maintain an objective response. The second PBRSA is an outcome guarantee arrangement (OG-PBRSA), where the payer is reimbursed based on the duration of progression free survival realized by each patient relative to an outcome target of ten months. To evaluate the potential health and cost implications of these PBRSAs along with lower patient cost sharing we developed a partitioned survival model composed of three health states: PFS, disease progression, and death. Parametric survival curves were fit using published survival data from pivotal clinical trials and survival was extrapolated for a lifetime time horizon using one-week increments. Model parameters such as survival, drug utilization rates, adverse event rates, utilities, and costs were derived from clinical trials and other published sources. The analysis was performed from a third-party payer perspective. RESULTS: In our hypothetical treated population of 50 patients, three additional patients were treated with PEMB under our PBRSAs at an additional drug cost of approximately $600,000, excluding drug rebates, compared to the base case. The payer received drug rebate amounts of $460,000 under the CTC-PBRSA and $113,000 under the OG-PBRSA. Average lifetime QALYs were 1.64 for the base case and 1.67 for both of our PBRSAs, which produced an incremental cost per QALY of $85,000 and $321,000 for the CTC-PBRSA and OG-PBRSA compared to traditional reimbursement practices, respectively. From our sensitivity analysis, the initial cost of implementing a pilot PBRSA was found to be the most influential parameter for both of our PBRSAs. CONCLUSIONS: The results of our decision model showed the CTC-PBRSA was more cost-effective than the OG-PBRSA with similar MFTR risk sharing of drug costs relative to the base case. Our findings suggest the barriers to the use of PBRSAs in the US health care environment are not insurmountable, and their use along with lower patient cost sharing may lead to cost savings and improved health outcomes. However, the generalizability of our results is limited to the specific terms of our PBRSAs, which reaffirms the need to carefully structure and implement PBRSAs based on the needs and demographics of the patient population of interest.