When the American Society of Clinical Oncology surveyed its members to ask what they got when they order a cancer gene panel from a lab, approximately 50 percent said the lab faxes over a PDF or sends a PDF file as an attachment. Only 22 percent said the lab can send the results as discrete data and that their EHR can store it.
The average genomic report is 30 pages long and dense, noted Jeremy Warner, M.D., M.S., assistant professor of medicine and biomedical informatics at Vanderbilt University School of Medicine. Integrating genomics into clinical workflow is a step into unknown territory, he added. Health systems have to make sure the report is not interruptive of the patient-doctor relationship. “I can’t imagine reading a 30-page PDF in front of a patient in the office,” said Warner, who was speaking at an Oct. 26 HL7 meeting on the future of cancer genomics, interoperability and precision patient care.
In order to help providers get some context to go along with test results, Warner and colleagues have developed a SMART on FHIR Precision Cancer Medicine app that compares a patient’s diagnosis-specific somatic gene mutation to a population-level set of comparable data. Context-specific links within the app connect to Gene Wiki, My Cancer Genome, and HemOnc.org. For example, if a lung cancer patient has the EGFR exon 19 deletion, the app can show the patient and clinician how common EGFR mutations are compared to other gene mutations and how often the exon 19 deletion is found compared to other alterations. But Warner was just one of several speakers at the HL7 conference who talked about how difficult it is for providers to make sense of genomics results as well as the challenges they face getting them in structured format and bringing them together with other clinical data.
Phillip LaJoie, director of operations at 2bPrecise, an Allscripts subsidiary with a cloud-based genomics and precision medicine solution, agreed with Warner that physicians don’t have the time to read that 30-page PDF. “It gets lost in the documents section of the EHR, never to be seen again,” he said. “We need to take the information on mutations and translate it into language that is clinically understandable,” he said. The National Institutes of Health (NIH), is piloting the use of 2bPrecise. The results of genomic sequencing tests will be made available and actionable at the point of care through Allscripts Sunrise EHR. The solution uses clinical-genomic ontology and data harmonization to return merged, semantically harmonized knowledge in a machine-readable and structured format, he explained.
Jonathan Hirsch, founder and president of Syapse, a precision medicine software company, spoke about how most EHRs import genomic data today. Many testing labs say they integrate with clinical systems, he noted. What they mean is that instead of faxing, they transmit a PDF. “That is not an improvement. They are actually turning structured data into unusable garbage,” Hirsch said. “How are you going to overlay decision support and clinical trials information if you have garbage data coming in?” He said Syapse is trying to solve some of these issues with its customers, including Intermountain Healthcare, by focusing on four components: data integration, decision support, clinical work flow, and a learning health system framework.
Barbara McAneny, M.D., a New Mexico-based oncologist, is board chair of the National Cancer Care Alliance and a member of the board of trustees of the American Medical Association. She gave a provider-level view of how overwhelming the genomic data is at the point of care and the challenges of operationalizing personalized medicine. “The amount of data that comes at me scares the daylights out of me,” she said. One problem is that there is no decision support accompanying the results. If there are possible drug therapies or clinical trials available, she said, there are often time-consuming challenges going through prior authorization processes with payers or in getting patients in rural areas into clinical trials.
McAneny described her work on a grant-funded project called Community Oncology Medical Home (COME HOME) that included seven community oncology practices working to develop and implement a community oncology medical home model. It developed a standardized, computer-based decision support tool with 12 clinical treatment pathways. The appropriate laboratory, radiologic, and molecular testing required at each stage of diagnosis for the 12 cancers being followed are incorporated into the clinical treatment pathways.
“Oncologists need real-time decision support,” she said, “and we have to put it in the right place in the clinical pathway. The trick is embedding these in EHRs — and EHRs don’t like you to embed anything in them.”
Watch this space for more coverage of the HL7 genomics conference, including a pilot project at Intermountain to use HL7’s FHIR standard to link family health history pedigrees to an external web-based provider of risk calculations for multiple types of cancer.