Enabling precision medicine through open standards in pharmacogenetics: data model development and Delphi study (Preprint)

• DOI: 10.2196/preprints.65689
• Publication Date: 2024-12-18T17:46:47Z
• Authors: Videha Sharma; John McDermott; Jessica Keen; Heidi Koikkalainen; Ian McNicoll; William Newman

BACKGROUND Pharmacogenetics, a subset of precision medicine, leverages genetic testing to improve the safety and efficacy of prescribed medicines. Despite compelling evidence supporting its clinical and economic impact, widespread implementation remains limited. An established barrier to implementation is the lack of interoperable health information technology solutions that integrate pharmacogenetic data into frontline clinical workflows, particularly for non-specialist clinicians. Open data standards, such as openEHR and Fast Healthcare Interoperability Resources (FHIR) could address this by creating standardised, vendor-neutral formats for pharmacogenetic test results and thus enable precision medicine at scale.

OBJECTIVE This study aimed to develop and validate open data standards to represent pharmacogenetic data using openEHR and map these standards to HL7 FHIR to enable interoperability within healthcare systems.

METHODS We developed a baseline openEHR data model by synthesizing existing literature, genomic sequencing technology outputs, and international FHIR specifications. Using openEHR’s Archetype Designer, an iterative design process was followed, informed by workshops with the Global Alliance for Genomics and Health (GA4GH). The model subsequently underwent two rounds of Delphi peer review involving 24 international experts across 10 countries providing a total of 30 reviews across both rounds. Mapping to HL7 FHIR was performed using both manual and automated approaches, including the FHIR-Connect tool.

RESULTS A standardised pharmacogenetic data model was developed, delineating "test results" from therapeutic implications to streamline clinical decision support (CDS). The final model incorporated globally recognized terminologies (e.g., SNOMED CT, HGNC) and was endorsed by the international expert Delphi review group. Published on the openEHR Clinical Knowledge Manager platform, the model achieved consensus for real-world deployment. HL7 FHIR mapping demonstrated successful bidirectional compatibility, supporting potential integration with English National Health Service (NHS) systems. Both manual and automated mapping approaches proved feasible, with the latter enabling scalable and reusable data transformations.

CONCLUSIONS The study established a robust framework for storing and exchanging pharmacogenetic test results, addressing key barriers to precision medicine implementation. By separating test results from therapeutic implications and leveraging openEHR and FHIR, the proposed standards ensure semantic harmonisation and interoperability. These findings lay the groundwork for scalable implementation of pharmacogenetics, supporting CDS integration into routine care. Future efforts should focus on refining genomic data models, enhancing policy frameworks, and promoting cross-disciplinary collaboration to advance precision medicine.

CLINICALTRIAL Not applicable