Bakground
The widespread adoption of next-generation sequencing data mandates the development of integration and standardization methods to evaluate best the clinical utility of genomic data in precision oncology.
Molecular Tumor Boards
Precision oncology is exemplified in Molecular Tumor Boards (MTBs) that collect, review, and interpret genomic and clinical data to match patient cases with targeted therapies. Often, the information collected and the knowledge generated by the MTBs is stored in repositories that do not comply with any data standards, making their actual use for any type of analysis and the interoperability across institutions challenging.
MTB at Johns Hopkins University
The Sidney Kimmel Comprehensive Cancer Center (SKCCC) at the Johns Hopkins University (JHU) has established a state-of-the-art multidisciplinary MTB that delivers top-quality precision oncology and personalized medicine services. The above data needs motivated us to develop a core data model for adequately supporting the MTB work through standardization and programmatic retrieval of all information utilized in the MTB review.
Model Development in Brief
We extended the minimal Common Oncology Data Elements model by adding profiles and data elements that would support the JHU MTB efforts. We further mapped elements to standardized terminologies and code sets as well as the Fast Healthcare Interoperability Resources (FHIR). The use of standards like FHIR is essential, especially given the nationwide efforts to maintain FHIR-based Electronic Health Record systems.
An interdisciplinary team of experts (clinicians, bioinformaticians, medical informaticians, and data managers) participated in the construction of the Precision-DM in 3 steps. First, they recognized all profiles to represent the information reviewed and generated by the JHU MTB. Second, they specified the complete list of elements under each profile with mappings to selected terminologies and FHIR (if applicable). Third, they shared the Precision-DM with an independent group of experts that provided feedback incorporated in the model.