BACKGROUND: Inconsistencies in reporting and defining clinical trial endpoints have been previously documented. In oncology, ongoing debate surrounds the utility of, meaningfulness of, and evidence behind common surrogate endpoints, particularly Progression Free Survival and Overall Response Rate, which hinders design of trials to improve outcomes. Questions emerging from this debate could be answered by comparing endpoint frequency, definition, and application within and across trials, but no data source exists for such analysis. This study presents a method for developing such a dataset.

OBJECTIVE: Design a replicable, systematic approach to build a dataset for analyzing cancer trial endpoints and evidence.

METHOD: The method employed a five-step process for dataset development. NSCLC disease was selected given the high volume of therapy development programs. At each step, data including summary characteristics and statistics were extracted and recorded in the database.

Step 1 identified therapy FDA approvals from 2016-2020 via systematic search of the FDA website. In Step 2, the Drugs@FDA database was searched to identify approval packages. Step 3 synthesizes the clinical evidence supporting approvals through systematic analysis of FDA Review Documents. Step 4 identified the published protocol for each study from Step 3 using ClinicalTrials.gov. Step 5 identified publications reporting results from Step 4 using the National Clinical Trial identifier numbers and keywords.

RESULTS: &

CONCLUSIONS: This method yielded a dataset that comprises 33 NSCLC therapeutic FDA approvals between 2016-2020, including 6 first approvals and 27 indication line extensions. By indexing documents as described above, we can establish a baseline dataset with which to evaluate endpoints and inform research questions and policy discussions on endpoint fidelity, utility, and meaningfulness.