While the era of isolated large medical record databases, unsystematic access to and performance of genetic testing and rising healthcare costs are at hand, we have an opportunity to develop solutions and tools that could unify and harmonize these efforts. The US Department of Veterans Affairs currently has one of the world’s largest genomic medicine programs fueled by volunteer efforts recruiting Veterans into the Million Veteran Program combining genomic, survey and medical record data for research purposes. This program, over 500,000 strong, has unearthed multitudes of challenges, as have many developed and new genomic efforts of similar kind. Through intellectual partnerships on committees and panels, we share similar discoveries, and seek common solutions. This proposal is targeted as sharing some of the challenges, goals for unification, and “wish lists” that should be taken into consideration globally if we are to develop systems that are comparable, translatable, actionable and most importantly, take into account the safety and individuality of the partnering patient.
Topics that would be considered for publication span from recruitment of patient and acquisition of samples, through databases, discoveries, validation of findings, translation into clinic – all aspects related to moving a discovery into practical clinical use. Subject areas that will be presented include:
- What are the standards required with genomic data, including nomenclature, metadata necessary given the evolution of analytics, structure of data integration within databases and presentation within medical records, tools required for seamless comparison analyses of a patient’s versus reference sequences?
- What are the standards required to develop a biobank to accrue, store and retrieve samples for study or diagnosis? What are the lessons learned from different biobanks worldwide. Can we learn from industry?
- What should one consider when performing phenotypic analyses of large patient data medical records. Why CPT coding isn’t good enough; how do semantics of medical records change over time and how knowledge changes meaning? When does a phenotype allow “just enough” analytical information but still gives enough flexibility for potential variances or overlaps with available diagnostic abilities?
- What are the ethical requirements that a patient partner should be aware of when joining a genomic study? How does one protect their data, or share their data with the safeguards that they expect or desire? Is the genomic/phenomic data truly “safe?” How do we educate them through sharing of their results?
- How does one address genomic/phenomic data findings to yield clinical validity and clinical utility? What are the foundational requirements that should be taken into account when developing such translational efforts? What is the boundary for acceptable risk? How does one seek and address the knowledge gaps?
- How does one develop and invest in clinical trials that bridge between genomic/phenomic research findings and clinical effectiveness? Is the use of retrospective mega-database trials and prospective trials adequate? Are pragmatic trials required? How does one address studies with a n=1 population?
- What are potential pathways to harmonize existing and developing genomic/phenomic or medical record databases? Are federated or unified models the pathway? What are the barriers to harmonization (data security, intellectual property, grant funding, employment stability)?
While the era of isolated large medical record databases, unsystematic access to and performance of genetic testing and rising healthcare costs are at hand, we have an opportunity to develop solutions and tools that could unify and harmonize these efforts. The US Department of Veterans Affairs currently has one of the world’s largest genomic medicine programs fueled by volunteer efforts recruiting Veterans into the Million Veteran Program combining genomic, survey and medical record data for research purposes. This program, over 500,000 strong, has unearthed multitudes of challenges, as have many developed and new genomic efforts of similar kind. Through intellectual partnerships on committees and panels, we share similar discoveries, and seek common solutions. This proposal is targeted as sharing some of the challenges, goals for unification, and “wish lists” that should be taken into consideration globally if we are to develop systems that are comparable, translatable, actionable and most importantly, take into account the safety and individuality of the partnering patient.
Topics that would be considered for publication span from recruitment of patient and acquisition of samples, through databases, discoveries, validation of findings, translation into clinic – all aspects related to moving a discovery into practical clinical use. Subject areas that will be presented include:
- What are the standards required with genomic data, including nomenclature, metadata necessary given the evolution of analytics, structure of data integration within databases and presentation within medical records, tools required for seamless comparison analyses of a patient’s versus reference sequences?
- What are the standards required to develop a biobank to accrue, store and retrieve samples for study or diagnosis? What are the lessons learned from different biobanks worldwide. Can we learn from industry?
- What should one consider when performing phenotypic analyses of large patient data medical records. Why CPT coding isn’t good enough; how do semantics of medical records change over time and how knowledge changes meaning? When does a phenotype allow “just enough” analytical information but still gives enough flexibility for potential variances or overlaps with available diagnostic abilities?
- What are the ethical requirements that a patient partner should be aware of when joining a genomic study? How does one protect their data, or share their data with the safeguards that they expect or desire? Is the genomic/phenomic data truly “safe?” How do we educate them through sharing of their results?
- How does one address genomic/phenomic data findings to yield clinical validity and clinical utility? What are the foundational requirements that should be taken into account when developing such translational efforts? What is the boundary for acceptable risk? How does one seek and address the knowledge gaps?
- How does one develop and invest in clinical trials that bridge between genomic/phenomic research findings and clinical effectiveness? Is the use of retrospective mega-database trials and prospective trials adequate? Are pragmatic trials required? How does one address studies with a n=1 population?
- What are potential pathways to harmonize existing and developing genomic/phenomic or medical record databases? Are federated or unified models the pathway? What are the barriers to harmonization (data security, intellectual property, grant funding, employment stability)?
