By Deborah Borfitz
May 18, 2021 | Genomic technologies are expected to be a mainstay of medicine within the next decade, but it is not entirely clear how to bring the resulting information into clinical care. The Institute for Genomic Health was established at Mount Sinai Health System two years ago to bridge that gap, says Founding Director Eimear Kenny, Ph.D., professor of medicine, and genetics and genomic sciences in the Icahn School of Medicine at Mount Sinai.
The idea is to launch pilot programs in the clinic based on discoveries from genomic research to evaluate their real-world utility, says Kenny. Where genomics data proves useful, the necessary infrastructure then gets built in terms of educating medical professionals and creating navigational paths for patients to bring that information to bear in clinical decision-making.
In addition to a large discovery research program using artificial intelligence and data science, the Institute for Genomic Health also runs clinical trials, she says. Whole genome sequencing is being tested as a frontline modality for pediatric diagnostics, for example, as is the use of digital apps in genetic counseling for empowering parents to take next steps.
New clinical programs have already emerged to deliver genomic screening information to patients at heightened risk of hereditary breast and ovarian cancer and cardiac conditions for which treatments are available. This summer, a study gets underway to learn how the sharing of polygenic risk scores for common diseases plays out with patients—as well as how health outcomes correlate with biological markers of ancestry and self-reported ethnicity and race.
In a study newly published in Cell (DOI: 10.1016/j.cell.2021.03.034), Kenny and her health system collaborators showed for the first time how genomic data linked to electronic health records (EHRs) capturing demography could be used to predict which populations are more susceptible to certain disorders, including cancers, asthma, diabetes, and cardiovascular disease. Researchers drew from Mount Sinai's BioMe BioBank program, a growing repository now containing genomic information on about 60,0000 patients across the five boroughs of New York City.
While limited to one urban health system, the EHR-linked biobank is highly diverse, says Kenny. One-quarter of participants have been linked to a genetic founder population, suggesting between 10% and 15% of New Yorkers are predisposed to certain genetic diseases.
Founder populations are commonly thought of as isolated, but the city is home to a half dozen such groups, she reports. The largest ones are among individuals of Ashkenazi Jewish descent, 1 out of every 80 of whom are uniquely at risk for three mutations associated with breast cancer, and those of Puerto Rican ancestry, possibly as many as 100 walking around the city with a hereditary collagen disease causing extremely short stature and a host of potential orthopedic complications.
Use of genomics data has increasingly come into the clinical care arena not just for diagnosing genetic disorders and family planning but also for pharmacogenomics and preventive health purposes, says Kenny, who has worked in population genetics for two decades. The classic example is testing for the pathogenic variants BRCA 1 and 2 in patients at high familial risk of breast and ovarian cancer.
By better understanding who is at risk for what diseases, researchers hope to broaden and personalize the types of screenings that get done as part of routine clinical care in the future, she says. That should result in earlier detection of diseases, if not thwarting their occurrence entirely.
Ethnic and racial labels are useful but not sufficient in either research or clinical care aimed at improving health outcomes, says Kenny. People are increasingly of mixed ethnic and racial ancestry, do not necessarily self-identify the same way around the world based on those parameters, and the information recorded in medical records may not even be coming directly from patients.
Depending on the health outcome, biological markers of ancestry might play a greater or lesser role than social and behavioral determinants of health based on race and ethnicity, Kenny says. At Mount Sinai, a decision 12 years ago to invest in the building of the BioMe BioBank was crucial to exploring those relationships.
BioMe BioBank was created so researchers could explore the utility of genomic information and its impact on health outcomes, says Kenny. “Most health systems don’t have that data in any way that’s accessible… it’s locked up, like a PDF.”
As happened with imaging data, Kenny predicts, genomic information will soon become digitized and widely available for improving the quality of care rendered by health systems. Only last year, the U.K. launched a large-scale initiative that involves genotyping five million Britons who will be followed over their life course to understand how the data will impact their health management, she notes.
Enterprise-level genomics has arrived, says Kenny, driving the creation of global communities and federated sharing of information, methods, standards, and best practices for data analysis. The common understanding among collaborators is that what is learned about one biologically distinct population benefits people everywhere.
Large databases of genomic data are essential to reaching statistically sound conclusions about disease risk based on the “totality of genes and variants,” she says. Most common diseases reflect an aggregate effect across many genetic variants and their relative presence or absence determines an individual’s polygenic risk score.
In most cases, the genetic contribution to disease comes from hundreds if not thousands of variants across the genome, explains Kenny. While each of the variants on their own has a small impact, collectively they can explain upwards of 10%—and in some cases 25% or more—of the heritability of a disease.
Patients represented in the BioMe BioBank are served by Mount Sinai’s seven health campuses and hundreds of laboratory centers across New York City. “If we go back two generations, 65% of participants have grandparents born outside of mainland U.S. with ties to 160 countries around the world,” says Kenny, which “gives us insights into how some of these questions play out globally.”
Which findings specific to New Yorkers are transferable to people in, say, Uganda or Hong Kong, has yet to be investigated, Kenny adds. For particularly common diseases—e.g., type 2 diabetes, heart disease, and kidney disease—social and environmental factors are probably having a bigger health impact than genetics.
The Value Add
The new study used machine learning to identify 17 distinct ethnic communities from among 30,000 initial participants in BioMe BioBank, says Kenny. After linking to their EHR data, the researchers found the 25% who were genetically predisposed to certain genetic diseases.
Any medically actionable findings were shared back with participants who opted to receive that information, she points out. In 2018, the institutional review board at Mount Sinai approved the inclusion of the share-back option in informed consent documents for genomic studies. Unsurprisingly, upwards of 90% of research participants opt to be so notified.
One of the more interesting study findings relates to founder effects in Puerto Rican populations, Kenny says, which she believes is “widely under-appreciated.” She was one of the researchers who co-discovered the founder variant for a disorder called Steel syndrome that was thought to be rare (DOI: 10.7554/eLife.25060).
The recessive disorder, a mutation of the COL27A1 gene, impacts height and has been clinically described for decades although clinicians in both Puerto Rico and New York City seemed to be wholly unfamiliar with it, says Kenny. In addition to short stature, clinical features include bilateral hip and radial head dislocations, fusion of the carpal bones, scoliosis, high arches, and dysmorphic features.
Steel syndrome impacts 1 in 2,000 Puerto Ricans, she says. That means at Mount Sinai, which treats about 190,000 individuals of Puerto Rican descent annually, up to 100 patients are likely afflicted. “We didn’t even have a molecular test until a couple of years ago.”
Another noteworthy finding of the latest Cell study is that risk variants of the APOL1 gene, which increases the risk of developing renal and cardiovascular diseases, are most frequently seen in populations sharing African genetic ancestry, Kenny says. The gene was discovered only 10 years ago in African American populations and it has an “evolutionary history for protection against a form of sleeping sickness caused by the Trypanosoma parasite.”
The APOL1 haplotype, a combination of three variants, is quite common among people of African American descent—13% of the population carries the risk, says Kenny. But it has been mainly studied in the U.S. and Africa, and “there are many other populations around the world who may or may not identify as having African ancestry but nevertheless have African genetic ancestry due to slavery and other forms of migration who can also harbor the risk variants.”
As Kenny and her colleagues reported a few years back in correspondence published in The New England Journal of Medicine (DOI: 10.1056/NEJMc1800748), elevated frequencies of the APOL1 haplotype have been found in 15 other populations around the world who are not typically screened. This suggests kidney disease and related coexisting conditions are being underdiagnosed and undertreated and highlights the knowledge gap that genetic ancestry data can help fill.
Mount Sinai researchers have also recently published (DOI: 10.3390/jpm11010049) on the value of genomic screening in identifying individuals at high risk for hereditary transthyretin amyloidosis. Among BioMe BioBank participants, the associated TTR V1421 variant was detected in 32 African Americans and Hispanic/Latinx individuals, none of whom had been previously diagnosed with the multisystem, rapidly progressing disease. The majority had “well-described red flags” of amyloidosis and more than half followed up with a cardiologist after receiving their TTR V1421 results.
More broadly, their recent research reveals “how patients with origins from different countries in the Americas can have different rates of disease,” Kenny says. “For example, people of Puerto Rican and Mexican descent are broadly classified as Hispanic or Latinx, yet the former population has one of the highest rates of asthma in the world, while the latter population has one of the lowest."
Next steps for Mount Sinai are to turn the findings into pilot studies using biological markers of ancestry to target people for preventive screenings, Kenny says. The need is growing for “more sophisticated, fluid understanding of genetic ancestry in concert with self-reported identity” that factors in people’s lived experiences that can themselves impact their health.