SAN FRANCISCO (GenomeWeb) – Researchers at the Mayo Clinic are trying to figure out how to ensure that the results of pharmacogenomic tests that they perform are used by patients once they leave the clinic – and with that goal in mind, they are collaborating with a group at the Medical University of Vienna that has developed technology to make PGx data portable.
Robert Freimuth, who heads the Informatics of Genomic Medicine lab at the Mayo, described the work in a presentation at the recent American Medical Informatics Association Joint Summits on Translational Science in San Francisco and in a follow-up interview.
The system is known as Medication Safety Code (MSC) and consists of a card that patients can keep in their wallets. On the card is both a brief list of the relevant gene and drug interactions as well as a barcode that can be scanned with a mobile device to get a detailed account of the patient's specific results and dosage recommendations.
Each year, more than 1.3 million patients from all 50 US states and 140 countries visit the Mayo Clinic. Many of them are treated at the Mayo for a short time and then return home. Sharing the electronic medical records created within the Mayo system with patients' providers at home has proven to be challenging, and the Mayo team does not have a way to ensure that test results are adequately communicated to those providers. "We don't want patients to just benefit when they are at Mayo," Freimuth said, so "we need the ability to share results with their providers."
So, Freimuth said he began working with Matthias Samwald, an assistant professor at the Medical University of Vienna, who began the MSC system. Samwald's MSC system is also a component of a larger European study called Ubiquitous Pharmacogenomics, which launched last year.
The MSC system enables clinicians to bypass the hurdles of electronic medical record systems and send results to the physician via the patient. Reading the results requires just a way to scan the code and an Internet connection. When providers scan the code, they are able to access only the clinical PGx results.
Freimuth said he thinks the MSC system could "empower patients to share results directly with their providers," taking a more active role in their care, and can enable healthcare providers who have not yet implemented PGx testing themselves to still prescribe medications based on a patient's specific PGx results.
The Mayo team has not yet implemented the MSC system clinically. Freimuth said that they are figuring out the regulatory requirements of the system, which would involve giving results to patients. In the meantime, Mayo is populating its knowledgebase of gene-drug interactions — creating both a public database as well as a private one that is behind a firewall where it will store patient data.
The Mayo has been an early adopter of PGx testing, including preemptive PGx testing — identifying PGx genotypes before a drug is prescribed. It is part of both the Pharmacogenomics Research Network (PGRN) as well as the Electronic Medical Records and Genomics (eMERGE) Network, two National Institutes of Health funded consortiums. In addition, as part of a collaboration project between the two consortiums, the Mayo has been participating in studies evaluating how to do PGx testing preemptively and deposit the results in patients' EMRs. It began the pilot of the so-called RIGHT protocol, which stands for Right Drug, Right Dose, Right Time – Using Genomic Data to Individualize Treatment, in 2014 for around 1,000 patients, who were deemed likely to be prescribed statin therapy within three years.
The Mayo has also been building a clinical decision support system for PGx testing that provides alerts to physicians when a new PGx result becomes available or when they prescribe a drug to a patient.
The Mayo Clinic uses an 84-gene next-generation sequencing panel for its PGx testing, although it does not yet report back on the results of all of those genes. The panel was developed as part of the PGRN consortium and is used at the other PGRN sites.
For the initial RIGHT protocol, results from five genes were reported in patients' EMRs. Since then, however, the Mayo team has developed clinical decision support for 19 drug-gene pairs, Freimuth said.
As a first step to test the MSC system, Freimuth said he and his team focused on generating codes for the initial 1,013-patient RIGHT cohort and the initial five genes: SLCO1B1, CYP2C19, CYP2C9, VKORC1, and CYP2D6.
In addition, he said, they focused on building up the web interface. Although very basic information will be contained on the back of the physical MSC card itself, "there's no way to possibly include everything on the back of the card that we want to convey," Freimuth said.
That is where the barcode comes in. When physicians scan the barcode, they are able to see the patients' specific results. Clicking on the result takes them to Mayo's public knowledgebase — a private knowledgebase used internally at Mayo is behind a firewall — where they can get more specific information on what a certain genotype means and recommended dosage for specific drugs based on that genotype.
Freimuth said that a goal of the MSC system is to "overcome the challenges that are presented by the lack of standards on how to transmit genomic information and communicate genomic information to other providers."
The Mayo has also expanded the RIGHT protocol to 10,000 patients, and Freimuth said his team is also in the process of generating MSC codes for those patients as well. For those patients, clinical results from 19 drug-gene interactions will be deposited in the patients' EMRs, while the remainder of the genomic information will be included in a research database. Similarly, Freimuth's team will expand its MSC system to include the same 19 drug-gene interactions. He said he anticipates this next phase will be completed by the end of June.
Also as part of the RIGHT study, the Mayo researchers will evaluate outcomes data in terms of how physicians use the PGx results as well as how well both patients and physicians understand the results. Those results would also help inform how information is communicated via the MSC system.
Freimuth said that before the MSC system can be fully rolled out to patients, though, there are still a number of steps Mayo must undertake. For one, he said, there are a number of pieces of infrastructure that still need to get built. The computational infrastructure needs to be refined and the clinical decision support tools need to be more fully integrated into the workflow. In addition, he said he would work with the education team as well as the content management system to make sure that the web-based educational content is all up to date and working properly.
The big remaining piece will be to figure out the regulatory requirements. Although patients who are currently being tested under the RIGHT protocol can access their PGx results in their medical records through a patient portal, Freimuth said there are key differences between that set up and the MSC system. For one, those results are the full signed out lab report, he said. The MSC is the "bare minimum of results that would be needed to start a dialog with the clinician," he said.
The intent of the MSC is not to take the place of the full lab report, but simply to alert physicians outside of the Mayo that their patient has had PGx testing and there are important findings to consider. When the code is scanned, aside from being given more details about the test results, the physician is also given a way to contact the Mayo to learn more. "That tie-in lets the clinicians contact us and request the patient's full record or a PGx consultation," Freimuth said.
Freimuth said that the team is currently working with its internal IRB and ethics boards to identify the risks of the system. "Before we put this in practice we'll engage with the FDA" to determine to what extent the agency needs to be involved or how such a process should be validated.
Freimuth added that he sees the MSC system as a short-term solution to the problem of communicating genomic results between healthcare providers.
"Over the long term, we'd like to see the development of more robust standards that enable the exchange of this information electronically without having to rely on the patient to actually perform the transmission," he said.