PRM11 Real-World Data to Calculate Cost-Effectiveness of Monoclonal Antibodies: Problems and Solutions

Abstract

Real-world data is considered to be the gold standard by decision makers to inform on cost-effectiveness of new drugs. Unfortunately real-world data are often lacking in important parameters needed to inform on cost-effectiveness, and RCT data can be used to address this problem. Illustrated by two cases this abstract will show that the manner in which RCT and real-world data are combined can have a profound influence on the resulting ICER. Two case studies in which real-world data on cetuximab for the indication of locally advanced head and neck cancer and panitumumab for the indication of chemo-refractory metastatic colorectal cancer was collected retrospectively served as examples. The problem: In the case of cetuximab, patient selection in daily practice resulted in too much disparity in baseline characteristics between the treated and control group. The solution: survival data for both treatment groups from the pivotal RCT was used and corrected according to the results seen in the real-world data to better represent survival in daily practice. Using unadjusted RCT data resulted in a difference of approximately 5,000 euro/QALY in the ICER. In the case of panitumumab no data were available on progression of disease in the control group. The solution: progression free survival and survival after progression of the control group were drawn from the pivotal RCT and adjusted according to the survival observed in the control group of the outcomes research. Unadjusted RCT data resulted in an ICER that was approximately 20,000 euro/QALY higher. RCT data are often necessary to supplement missing data that cannot be collected through outcomes research, however the manner in which RCT data is used can have a profound effect on the resulting cost-effectiveness.