Abstract 2463: Prototype open-access biomarker knowledgebase for genetic tests for breast cancer

Cancer biomarkers have become integral components of many clinical protocols and research studies. Nonetheless there still is an urgent need for more accurate, less invasive, and cost-effective biomarker tests to advance precision oncology. Despite considerable success in the preclinical setting, most candidate biomarker tests have yet to receive regulatory clearance or approval and gain market acceptance. To better understand the challenges facing modern biomarker research and facilitate access to biomarker-related information, this study aims to delineate the paths to approval of genetic tests for breast cancer. Public databases were searched to identify approved tests and respective biomarkers. A prototype format for unified biomarker information was developed iteratively due to the highly variable presentation of this data. A knowledgebase of FDA-approved genetic biomarkers for breast cancer was built as a pilot implementation of this format. This model is extensible to a larger database of biomarkers to be hosted in OncoMX (https://oncomx.org), which integrates cancer mutation, differential expression, literature mining, and pathway information. Fields in the prototype database were populated based on findings from the FDA website, PubMed, and other open-source databases. The resulting dataset is structured with the corresponding readme following the BioCompute Object (BCO) model. This BCO-compliant approach enables provenance capture and transparency of data processing, both of which are critical to large-scale data integration efforts. The adherence to BCO standards is expected to enhance the usability of the resulting data and to streamline the subsequent integration of multiple biomarker datasets. In the rapidly evolving field of oncologic biomarker research, this open-source biomarker knowledgebase offers clinicians and researchers a streamlined access to genetic tests information including corresponding genes, guides to evidence, clinical trial data, links to archival databases, and context for the behavior of the implicated gene in other disease and normal processes. Citation Format: Dara Baker, Natalie Fedorova Abrams, Amanda Bell, Maureen Colbert, Hayley Dingerdissen, Evan Holmes, Samir Gupta, Robel Kahsay, Heather Kincaid, David Liu, A. S. M. Ashique Mahmood, Frédéric B. Bastian, Marc Robinson-Rechavi, Elena Schwartz, K. Vijay-Shanker, Daniel Crichton, Raja Mazumder. Prototype open-access biomarker knowledgebase for genetic tests for breast