At the frontlines of precision medicine: Data and IT challenges
Numerous challenges, such as capturing genetic data in EHRs, need to be addressed to achieve the promised value of precision medicine.
David Jones, Ph.D., offers a vision of how precision medicine will evolve in the next 20 years to transform healthcare.
By 2042, many babies will have their genome sequenced soon after birth, revealing their risk for hundreds of diseases and conditions, both rare and common, says Jones, chief scientist at Intermountain Health in Salt Lake City. Healthcare providers will have this genetic data at their fingertips to help them develop customized prevention and treatment plans for each patient, he predicts.
“We’re going to move from a reactionary style of patient care, where we wait until people get sick before doing something, to a more proactive approach,” the scientist says. We’ll know the diseases people may get, so we can get those people screened regularly and prevent them from getting those diseases.”
The first steps
Initial steps toward achieving that long-term goal are being made today. For example, genetic tests are available to identify a modest number of genes and gene mutations linked to certain diseases and conditions, including cancers and high cholesterol. In addition, physicians can order pharmacogenomic tests to determine if patients, based on their genomic profile, likely will respond to a drug or have side effects. About 100 drugs have been identified by the U.S. Food and Drug Administration as having gene-drug interactions.
Most of the early progress in precision medicine has been in cancer care. For instance, oncologists are increasingly able to prescribe targeted medicines based on the molecular (or genetic) profile of a patient’s tumor.
Other medical specialties are working to catch up. Scientists at Cedars-Sinai Health System in Los Angeles, for example, developed a targeted drug, which is in phase 2 trials, for those with Crohn’s disease and ulcerative colitis with a certain molecular profile.
“Our great hope is that we can say to patients with Crohn’s disease, ‘You have this molecular profile; therefore, you’re much more likely to respond to this drug than that drug,” says Dermot McGovern, M.D., Ph.D., director of the precision health initiative, Cedars-Sinai.
But there’s still a long way to go to achieve all the benefits of precision medicine. “We’re in the early innings,” says Tom Andriola, chief digital officer and co-director at The Institute for Precision Health at UCI Health in Irvine, Calif. “There’s a tremendous amount of value that still needs to be unlocked.”
In addition to UCI Health, Cedars-Sinai and Intermountain Health, a growing number of organizations, including NorthShore University Health System in suburban Chicago and The Christ Hospital Health Network in Cincinnati, are making significant steps in implementing precision medicine.
Artificial intelligence is one tool these organizations are using to drive precision medicine implementation. But a variety of data and IT challenges, from capturing genetic data in the EHR to data equity and privacy issues, need to be addressed to achieve the promised value of precision medicine.
Artificial intelligence has an important role to play in the data-driven science of precision medicine.
“The future of personalized care is about how we can take advantage of all this data being generated inside and outside the healthcare enterprise to help providers and patients make decisions about how to stay healthy,” says Tom Andriola, chief digital officer and co-director at The Institute for Precision Health at UCI Health in Irvine, Calif...READ MORE
Genetic screening program
In one pioneering effort, some primary care patients at NorthShore University HealthSystem are given the option of getting a genetic test to learn if they are at risk for certain hereditary conditions, including cancer, high cholesterol and thrombophilia (blood clots). A pharmacogenomics panel can also determine how the patient responds to some medicines.
The health system is partnering with the tech company Sema4 on this initiative.
Since NorthShore launched the population genetic screening program launched in 2019, over 20,000 patients have obtained the tests and had their hereditary results recorded as discrete data in the electronic health record.
“Our patients can get this information through their healthcare provider rather than have to turn to direct consumer testing where that information may live outside the electronic health record,” says Peter Hulick, M.D, medical director at NorthShore’s Mark R. Neaman Center for Personalized Medicine.
The health system’s personalized medicine team oversees the return of genetic test results, which helps ensure patients receive the appropriate referrals, such as for genetic counseling, screenings or medical specialist care. “The goal is not just to get a bunch of people tested; we need to put that information to good use and have patients act on it,” Hulick says.
In one case, a 27-year-old woman’s genetic test came back positive for change in the BRCA1gene, indicating an increased risk of breast and ovarian cancers, Hulick says. A mammogram and breast MRI revealed she had breast cancer. Fortunately, the cancer was caught in the early stages when it is most treatable.
Because NorthShore records genetic test results in the EHR, the information can help guide the care that patients receive. Clinical decision support tools provide real-time alerts to providers across the health system so they can leverage genomic insights when seeing a patient.
For instance, cardiologists, oncologists, psychologists and other specialists can be alerted to pharmacogenomic results, which can help them identify the best medicine to prescribe or medicines to avoid due to potential side effects.
“Over half of the encounters where genetic information is being used are in areas outside of primary care (where the testing was performed), such as in specialty care and the emergency department,” Hulick says.
Expanding genomic knowledge
The human genome contains about 20,000 to 25,000 genes. But so far, only about 20 genes or gene mutations have been identified to cause specific diseases, Intermountain Health’s Jones says.
Research projects are underway around the globe to collect and study genomic data in an effort to identify more disease-causing genes and gene mutations.
“One of the things that limits pharmaceutical companies is that they don’t know who to study.”
One of these research projects is Intermountain’s HerediGene, which aims to study the genomes of 500,000 adults and children who volunteer to provide blood samples.
Since the project was launched in 2019, more than 100,000 people have enrolled in HerediGene. Intermountain and its vendor partner, deCODE Genetics, have announced HerediGene’s first discovery: six genetic variants associated with vertigo, a condition that causes dizziness and balance problems.
A next-generation sequencing technology, which allows for sophisticated analysis of DNA on a massive scale, and Intermountain Health’s EHR, which has been in place for 35 years, enable scientists to pinpoint genetic commonalities in groups of people with the same disease, Jones says.
Discoveries made through HerediGene could lead to new treatments for rare or complex diseases.
“One of the things that limits pharmaceutical companies is that they don’t know who to study,” Jones says. “Let’s say they have a drug that may prevent Parkinson’s disease. But they don’t know who to test the drug on because they don’t know who is going to get Parkinson’s.”
Once the genes or genetic mutations that cause Parkinson’s are identified, then pharmaceutical companies could test the drug on individuals with this molecular profile who have not yet developed Parkinson’s, Jones says.
Genetic discoveries are hammering home the need to tailor care to each patient. Experts now know that many diseases present in different ways in different people.
For instance, a 2015 study found that type 2 diabetes patients could be divided into three distinct subgroups. Each subgroup has unique genetic variants, and the disease progresses differently in each subgroup.
“Up to today, individuals have been treated based on what was best for groups of patients,” UCI Health’s Andriola says. “Now, with our ability to understand through data the phenotype, genotype and behavior of the individual, we will be able to deliver truly personalized care to each individual.”
The role of the EHR
Developing truly personalized medicine will require the embedding of genetic data into EHRs, stresses Burns C. Blaxall, Ph.D., executive director of precision medicine at The Christ Hospital Health Network in Cincinnati. “This allows your data to live in your medical record forever. That way, for example, if you have a hereditary cancer risk, your doctor knows you need to screen more often.”
But incorporating genetic data into the EHR can prove to be challenging.
“Health systems are transforming their care models to increasingly leverage genomic and other patient data to drive improvements in care, but the vast majority of physicians today were not trained in an environment where large-scale data was available.”
Healthcare providers often enter contracts with labs and precision medicine companies for genetic testing.
Patients’ results are typically shared as PDFs via a portal or as faxed documents. And many EHR systems are not designed to incorporate genetic data submitted in this format.
“A lot of times, what people do is scan the results as a PDF into a non-searchable area of the medical record, called the media tab,” Blaxall says. “It makes trying to find the data difficult.”
To help address the problem, staff at Christ Hospital worked with partners to create order and result interfaces that allow test results to be electronically sent as discrete data to Christ Hospital’s EHR.
At NorthShore, bioinformatics and programming staff tackled this challenge by developing software called Flype, which works through an application programming interface with the systems of labs and precision medicine companies. Only clinically actionable information from patients’ genetic results is placed into the EHR, which helps prevent physicians from becoming overwhelmed with too much information, according to a paper on the software.
Storing genetic results as discrete data in the EHR helps to ensure timely follow-up care, Hulick says. “It allows us to quickly identify patients who have not yet gotten the care they need related to their genetic risks,” Hulick says.
This approach is also essential to adopting machine learning algorithms to assist in implementing and advancing personalized medicine.
Change management is the hardest part of scaling precision medicine, says James Lu, M.D., Ph.D., CEO and cofounder of population genomics company Helix.
“What health systems are doing is transforming their care models to increasingly leverage genomic and other patient data to drive improvements in care,” he says. “But the vast majority of physicians today were not trained in an environment where large-scale data was available.”
Lu stresses the need for physician education on how to leverage genetic data. And he calls for establishing workflows that provide physicians with the data in a way that is helpful, easily grasped and not distracting.
The precision medicine team at The Christ Hospital has spent two years embedding clinical decision support tools into provider workflows that alert them to patient-specific genomic information at the point of care. For example, one tool alerts oncologists when a patient’s tumor should be tested for genetic mutations, which can guide treatment decisions. Another tool advises physicians when pharmocogenetic testing is recommended for a patient.
The health network also has embedded precision medicine staff in various departments to provide support to providers.
“For example, we hired seven pharmacogenomics-trained pharmacists and embedded them in 30 primary care offices,” Blaxall says. “These pharmacists didn’t even use the word ‘pharmacogenomics’ for almost a year. They developed relationships with the physicians and then introduced the concept of pharmacogenomics.”
Going beyond genetics
Many precision medicine experts believe a patient’s genetic data needs to be considered alongside other relevant data, including medical history, lifestyle habits (e.g., physical activity collected from wearable devices), and environmental factors (e.g., whether the patient lives in a food desert).
Hulick uses the analogy of a jigsaw puzzle. “We’re trying to put together as many pieces as we can to get the clearest picture of someone’s health risks, and certainly one’s environment and lifestyle play a prominent role,” he says.
The industry is in the nascent stages of determining how to best combine and present all this data in a way that is useful to providers caring for patients. Machine learning algorithms will be part of the answer, according to a 2021 commentary in the journal Cell by National Institutes of Health.
UCI’s Andriola believes the availability of all this data will improve doctor-patient relationships and empower patients to make informed decisions about their health. “I envision a future where the interactions between health consumers and their primary care physicians will look more like my meetings with my financial adviser,” he says.
For instance, based on a wide variety of data, doctors and patients could pinpoint and discuss a handful of health risks, such as a change in blood pressure or genetic risk for high cholesterol, that need to be watched or acted on. “Those are the kinds of conversations [based on specific data] that I have now with my financial advisor around money,” Andriola says. “In the future, I hope to have similar conversations with my primary care doctor around my health.”
This content is is part of a series of articles and related material highlighting the people, processes, and technologies currently advancing Genomics Data and Precision Medicine. See related content