ABSTRACT
When seriously ill patients have exhausted all treatment options available as part of usual care, the use of investigational agents may be warranted. Food and Drug Administration's (FDA) Expanded Access (EA) pathway provides a mechanism for these patient's physicians to pursue use of an investigational agent outside of a clinical trial when trial enrollment is not a feasible option. Though FDA has recently implemented processes to significantly streamline the regulatory portion of the process, the overall pathway has several time-consuming components including communication with the pharmaceutical company and the associated institutional requirements for EA use (contracting, Institutional Review Board [IRB], pharmacy, billing). Here, we present our experience building infrastructure at the Vanderbilt University Medical Center (VUMC) to support physicians and patients in pursuing EA, called the Access to Investigational Medicines (AIM) Platform, aligning the needs and responsibilities of institutional stakeholders and streamlining to ensure efficiency and regulatory compliance. Since its launch, the AIM team has experienced steady growth, supporting 40 EA cases for drugs/biologics, including both single patient cases and intermediate-size EA protocols in the emergent and non-emergent setting. As the EA pathway is a complex process that requires expert facilitation, we propose prioritizing EA support infrastructure at major academic medical centers as an essential regulatory knowledge function.
ABSTRACT
Background: Coronavirus Disease 2019 (COVID-19) instigated a flurry of clinical research activity. The unprecedented pace with which trials were launched left an early void in data standardization, limiting the potential for subsequent data pooling. To facilitate data standardization across emerging studies, the National Heart, Lung, and Blood Institute (NHLBI) charged two groups with harmonizing data collection, and these groups collaborated to create a concise set of COVID-19 Common Data Elements (CDEs) for clinical research. Methods: Our iterative approach followed three guiding principles: 1) draw from existing multi-center COVID-19 clinical trials as precedents, 2) incorporate existing data elements and data standards whenever possible, and 3) alignment to data standards that facilitate data sharing and regulatory submission. We also supported rapid implementation of the CDEs in NHLBI-funded studies and iteratively refined the CDEs based on feedback from those study teams. Results: The NHLBI COVID-19 CDEs are publicly available and being used for current COVID-19 clinical trials. CDEs are organized into domains, and each data element is classified within a three-tiered prioritization system. The CDE manual is hosted publicly at https://nhlbi-connects.org/common_data_elements with an accompanying data dictionary and implementation guidance. Conclusions: The NHLBI COVID-19 CDEs are designed to aid data harmonization across studies to achieve the benefits of pooled analyses. We found that organizing CDE development around our three guiding principles focused our efforts and allowed us to adapt as COVID-19 knowledge advanced. As these CDEs continue to evolve, they could be generalized for use in other acute respiratory illnesses.
