Biorepositories are valuable resources to the cancer research community and arguably invaluable if they include detailed longitudinal health information such as personal risk factors, disease progression, and treatment outcomes. However, manual review of patient records is comparatively inefficient (expensive, time-intensive, and less reliable) than data extraction from the electronic health record (EHR). Using Research Data Network funds granted from the Patient Centered Outcomes Research Institute (PCORI), PathLink is being developed as an institution-wide biorepository linked to Vanderbilt's EHR data warehouses, the Synthetic and Research Derivatives, and utilizing pathological specimens that exist or would otherwise be discarded, without duplicating existing biorepositories and other core resources.

We are cataloging samples from clinical and research collections and creating a sustainable and prospective link to structured health information. PathLink will de-identify samples on-demand, permitting translational investigators real-time sample access while also preserving tissue for future clinical assays. The EHR derivatives already include fundamental review features and standardized data extraction mechanisms that can be easily leveraged for tissue-based research projects. Projects proposing a de-identified research approach can also be crosslinked to our germline genetic biobank of >200,000 subjects and will have lower barriers to access through a streamlined regulatory mechanism.

Proof-of-concept studies underway include the banking of tissues with molecular genetic pathology testing (n = 3642), a large subset already annotated with SNaPshot tumor mutation profiling. Macrodissected tumor cells were collected from cases accessioned from 2010 to 2013 and stored as a tumor lysate reserve (B-tubes) pending successful completion of clinical assays. After extraction and quality testing of genomic material, PathLink was seeded with the transfer of these remnant samples to the biorepository. Pilot studies suggest that more than 75% of B-tubes will contain sufficient genomic material of high quality for targeted gene sequencing (>250ng, n = 120). We will digitally scan corresponding tissue sections and create tissue microarrays from cases with adequate tissue.

In conclusion, PathLink is a new concept in biobanking, minimizing time and cost, and enabling more efficient translation of tissue-based discoveries from bench to clinical care and back again. A unique feature is that much of the collection is a virtual catalog of partner biorepositories, and most of the sample processing is executed by existing core facilities, minimizing overhead while also increasing interdepartmental collaboration. By July 2015, we expect to have catalogued at least 300,000 pathological specimens and have 3,000 tumor DNA samples ready for distribution to approved translational projects.

Citation Format: Jennifer M. Giltnane, Jana Shirey-Rice, Jodell Linder, Erica Bowton, James Cowan, Xiaoming Wang, Jon Scherdin, Melissa Basford, Kimberly Dahlman, Joseph Roland, Kerry Wiles, Cynthia Vnencak-Jones, Kay Washington, Jill Pulley. PathLink: Leveraging clinical informatics to empower translational pathology by connecting biospecimens to outcomes. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 222. doi:10.1158/1538-7445.AM2015-222