Abstract-In the United States, new surgical heart valves can be approved on the basis of objective performance criteria (OPC). In contrast, the US Food and Drug Administration traditionally requires stricter criteria for transcatheter heart valve (THV) approval, including randomized, clinical trials. Recent US Food and Drug Administration approval of newgeneration THVs based on single-arm studies has generated interest in alternative study approaches for THV device approval. This review evaluates whether THV device approval could follow a pathway analogous to that of surgical heart valves by incorporating OPC and provides several considerations and recommendations. Factors to be taken into account in the construction of OPC include the maturity of THV technology, variability in transcatheter aortic valve replacement practice, end points included as OPC, follow-up terms for specific OPC, patient populations to which these OPC apply, and (statistical) methods for OPC development. 4 This review provides an overview of OPC, considers the potential role of OPC for THV device approval, and discusses the challenges associated with such an approach. Several recommendations for the future implementation of OPC for THV devices are provided.
Objective Performance CriteriaTraditionally, 2 methods of establishing a comparator for single-arm studies have been used: OPC and performance goals (PGs; Table 1). OPC are linearized event rates of safety end points that are derived from historical controls included in meta-analyses of available data. A PG, on the other hand, represents a single point estimate considered sufficient as a comparator. In general, the process of developing a PG is less formal than the process of developing OPC. Therefore, compared with PGs, OPC are thought to be superior because of the increased maturity of device technology and the more robust data from which the comparator is developed.5 For the FDA to allow a device to be considered for approval with OPC, it should fulfill a number of criteria: (1) The device technology should be sufficiently mature; (2) the meta-analysis from which event rates are derived should include an accumulation of studies and all relevant literature from different devices; (3) the OPC should not be developed by companies using only their own data; and (4) the OPC must be contemporary and updated regularly because technology may become obsolete over time, thereby attenuating the rigor and relevance of the OPC.
OPC DevelopmentThe OPC for surgical valves were conceived, subsequently developed, and then implemented after 2 major RCTs evaluating mechanical and biological prostheses did not identify significant safety concerns.1 It was thus recognized that traditional randomized evaluations were not feasible because of the large sample sizes required to identify infrequent events. Such large RCTs would also greatly impede the timely introduction of technological advancements to the market. 7 In 1993, the FDA convened an expert workshop to consider alternative study designs for regu...