DATA 5.0 -Data Acquisition, Translation & Analysis -A prospective urooncological data warehouse for the 21 st century

Viktoria Schütz ^1* Christine Geisler ¹ Mathias Rath ¹ Sarah Heike Böning ² Thomas Treber ³ Albrecht Stenzinger ⁴ Alexander Brobeil ⁴ Oliver Reinhard ³ Anette Duensing ^1,5 Stefan Duensing ^1,2 Markus Hohenfellner ¹ Magdalena Görtz ¹

• ¹ Department of Urology, Heidelberg University Hospital, Heidelberg, Baden-Württemberg, Germany
• ² Molecular Urooncology, Department of Urology, Heidelberg University Hospital, Heidelberg, Baden-Württemberg, Germany
• ³ Center for Digitalization and Information Technology, Heidelberg University Hospital, Heidelberg, Germany
• ⁴ Institute of Pathology, Heidelberg University Hospital, Heidelberg, Baden-Württemberg, Germany
• ⁵ Precision Oncology of Urological Malignancies, Department of Urology, Heidelberg University Hospital, Heidelberg, Baden-Württemberg, Germany

Background: Prospective data registration is the basis of clinical oncological research. Commonly, case documentation is restricted to studies investigating a defined hypothesis. Only few institutions prospectively register all oncological patients with a reliable, sustainable and continuous follow-up infrastructure. The Department of Urology of the Heidelberg University Hospital started it’s prospective tumor data base in 1992. Since then, the clinical course of all oncological in-patients is continuously registered within a life-long follow-up (success rate: 93%). Associated tumor tissue is stored in the Heidelberg Biobank. In 2005, the transfer of this invaluable registry from the initial InterSystemsCache®/KRAZTUR system to a modern data warehouse was initiated. However, the transfer of existing data into a new environment proofed to be technically challenging.Objective: To migrate the existing data into a modern data warehouse (DATA 5.0) while maintaining data extraction functions. Additional requirements included FHIR connectivity, big data analyses and AI applications. Methods: Together with SAP SE, DATA 5.0 was developed. Based on SAP HANA (High Performance Analytic Appliance) it allows data registration and analysis with third party analytical tools. The project was supported by members of the SAP SE executive board and funded by the Dietmar Hopp Foundation.Results: Data Acquisition, Translation & Analysis 5.0 (DATA 5.0), a web-based tool for data registration, preservation and analysis of treatment and follow-up data, was developed to proof-of-concept stage. DATA 5.0 was then implemented into clinical practice replacing the previous system. As of today, 15,345 oncological patients and 6.7 Mio. data points are registered. Conclusion: Prospective long-term data was successfully migrated into DATA 5.0, allowing data preservation, flexibility and capabilities for future data sources. DATA 5.0, together with associated tumor tissue, is a lighthouse platform for oncological research, with capability for third party analytical tools, big data analysis and AI applications including training of digital twin models.