Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More
Genomic medicine has the power to transform and personalize the healthcare experience. However, despite this tremendous promise, genomic data is typically only used outside routine healthcare workflows to treat rare diseases or prioritize cancer diagnostics. Healthcare organizations face challenges ensuring privacy for genomic data across various workflows. Additionally, most doctors have not been trained to interpret the meaning of genomic results on treatment options and communicate these to patients.
The Estonian Genome Centre is hoping to change that. The research group is leading an ambitious pilot program to sequence the DNA of 200,000 Estonian citizens, securely manage the data at scale and weave insights into regular medical checkups and treatments. It is an effort designed to respect patient security at each step. Lessons from this pilot could eventually aid healthcare organizations around the world.
A focus on common ailments
At a recent press event, Lili Milani, head of the Estonian Genome Centre, explained how it hopes to transform everyday healthcare for all Estonians. She said countries like the United Kingdom, France and Sweden have all developed advanced personalized medicine programs, but these are primarily focused on cancer and rare diseases.
In contrast, Estonia is exploring how to apply personalized medicine to front-line doctors.
Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.
“We are using genomics to prevent common chronic common diseases,” Milani said.
Estonia’s death rate for heart disease is far higher than other Western countries like Finland, yet the adherence to medical treatment is about half.
“We believe that we could motivate people to improve health behavior with the right approach,” she said.
Protecting confidentiality at scale
The first step is turning raw blood samples into useful genomics data. Milani said it previously costed about $1,000 to analyze a whole human genome a few years ago. Since then, the cost has dropped to about $100 thanks to a recent breakthrough from Ultima Genomics.
However, most of the research has focused on analyzing a much smaller subset of DNA for single nucleotide polymorphisms (SNPs), which looks for about 700,000 significant variants across people. This type of analysis is much cheaper to do at scale. When they started in 2001, the cost was around $300 per sample and is now under $50.
The data is only allowed for health research or treatment with a patient’s consent. The data cannot be used to raise insurance rates and cannot be analyzed without a patient’s consent. Use for criminal investigations is also prohibited. Although this may help solve serious crimes, Milani said there is a larger concern that allowing its use for that purpose could discourage participation.
The Genome Centre developed a novel data encryption and key management infrastructure to protect the confidentiality and privacy of each participant at scale. The actual genomic data is stored separately from any personally identifying information (PII). When teams need to run a study, they pull in genomic data for each individual using a unique key to connect genomic results to other patient data. This is done in a special room without Internet access or phones.
The center also keeps the actual blood samples for future analysis, such as the discovery of new clinically relevant SNPs. Each sample is labeled using a unique barcode that can only be connected to an individual with their encryption key.
Communicating the uncomfortable
Milani’s team has conducted several studies to determine the best way to bring genomics information into the doctor’s office. The Genome Center analyzed subsets of their databases to identify individuals with an increased risk of high cholesterol or breast cancer. In theory, these are common conditions that can be prioritized based on family history. In practice, many family doctors have not made the connection.
“A genetic first approach could help prevent many of these diseases,” Milani said.
One concern was that patients might be stressed out to discover they were at higher risk for a particular disease. To address this, the group developed a comprehensive counseling program to accompany any discoveries. As a result, almost all patients reported a much higher emotion several months after the diagnosis.
“They were grateful they knew about the risk,” Milani said. “We discovered that if this information is accompanied by proper genetic counseling, then people can be relieved of the risk.”
Improving drug prescriptions
The team also explored how to incorporate pharmacogenetics data to improve patient prescriptions. Today, patients are generally given a prescription based on their sex and age rather than their specific metabolism. Genetic differences can reduce some drugs’ effectiveness or even promote side effects. Based on their metabolism, some patients can experience up to a massive difference in the concentration of a drug after ten hours.
The Clinical Pharmacogenetics Implementation Consortium has identified hundreds of potential gene-drug interactions. The Genome Centre has developed genetic reports for some of the most important ones and explored different ways to communicate this information to patients and their doctors.
Top hospitals are already doing this sort of testing for some drugs, particularly expensive ones, but it is uncommon and the process is slow. A doctor must take a sample, order the test, and wait a few weeks for the result before prescribing the medicine.
“Our vision is that the doctor can just query the genetic database to get the results in seconds rather than weeks,” Milani said.
Personalizing the future
Today, several companies sell tests for detecting genetic variants associated with higher risks for diseases, but these are not generally accompanied by counseling programs.
Although, sometimes the doctors do not know what to do with the data when patients bring it in.
“There are no recommendations or guidelines, so a lot of the doctors will get angry at the patient,” Milani said.
It does not help that about half the doctors in Estonia are over sixty years old.
Her team is working with Estonia’s National Personalized Medicine Initiative to develop training for family doctors and nurses in genetic epidemiology and personalized medicine. They are also training more medical geneticists and genetic counselors to help explain genetic risks and suggest lifestyle interventions.
“A lot of this also focuses on how giving this kind of information can help improve their health behavior,” Milani said. “Will they take the stairs the next time they come up?”
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.