The field of precision oncology uses genomic testing to identify the genetic differences between tumor cells and normal cells, which provides the ability to make individualized diagnoses and can guide therapy selection. It sounds straightforward, but the genomic complexity of cancers and large numbers of genomic variants that contribute to the initiation and progression of malignancies makes it a very difficult process.
As a result, many organizations, including The Jackson Laboratory (JAX), have combed the research literature to collect and curate data about the variants and their properties. The resulting knowledgebases—JAX’s is called the (CKB)—provide clinicians with extensive information, including references, to help them interpret cancer patient genomic data. The separate efforts have introduced differences in data formatting and annotation and don’t allow for a comprehensive overview of all documented variants. Now, a team led by researchers from the Oregon Health and Science University, Washington University of St. Louis, and including , Ph.D., MBA, and Sara Patterson, Ph.D., who lead the JAX-CKB program, has produced a meta-knowledgebase to improve and harmonize variant interpretations with clinical significance. JAX CKB provided a subset of their CORE data, representing approximately five percent of the total genes found in CKB BOOST.
The project, part of the Global Alliance for Genomics and Health (GA4GH)’s umbrella program to share and standardize genomic data, worked with data provided from a total of six leading knowledgebases, including JAX-CKB. In “A harmonized meta-knowledgebase of clinical interpretations of somatic genomic variants in cancer,” published in Nature Genetics, the team reports aggregating a total of 12,856 variant interpretations within a single framework. The work revealed highly disparate representations of variant interpretation, including in the primary literature from which the curated knowledgebase data is compiled.
Using the meta-knowledgebase, the team demonstrated improved matching between a patient cohort and variant interpretations of potential clinical significance, from an average of 33% per individual knowledgebase to 57% in aggregate. The authors argue that the findings highlight the need for a cooperative effort on a global scale to standardize clinical variant interpretations for more robust implementation of precision oncology and other genomic medicine programs. The meta-knowledgebase is available to the research and clinical communities through a web application at search.cancervariants.org.
A leader in the effort to aggregate and standardize clinical cancer variant data, JAX CKB has been adopted by more than 40,000 oncology stakeholders worldwide. Working with a Microsoft-driven AI system, the CKB team is able to scan, identify and rank the thousands of biomedical journal papers published daily for review. Because of the issues involved with discrepant available data, the CKB platform is rigorously curated to ensure data integrity and adherence to existing clinical oncology guidelines. JAX CKB can be accessed at .