For consumers, the idea of getting a genetic test to determine risks for hereditary diseases is becoming an increasingly common proposition, but new research suggests that sometimes the accuracy of those results may depend on what ethnicity you are.
Take, for instance, hypertrophic cardiomyopathy, one of the most common hereditary heart diseases. It is also a silent disorder that has caused countless young athletes to suffer sudden collapse or cardiac arrest during team practices or sporting events.
African Americans have traditionally been considered at higher risk for the disorder. But a study out Wednesday in the New England Journal of Medicine concluded that common ways to determine that level of risk may be skewed because studies have traditionally had low numbers of black participants. It turns out that genetic characteristics based on ethnic differences caused some people to be diagnosed with a predilection for the disease - even though those markers were actually benign.
The researchers suggest these findings indicate a need for diversity in genomic research in interpreting these differences.
“Historically we've had less African American representation in these studies,” said Arjun Manrai, lead author of the study and research fellow at the Harvard Medical School. “Our paper highlights ethnicity as a key way to get a handle on classification of genetic variants.” Genetic variants are differences in DNA structure that determine human features and are unique to every person.
Manrai first looked at the data from the National Heart, Lung, and Blood Institute Exome Sequencing Project, which includes genomic data from 4,300 white Americans and 2,204 African Americans. He expected to find that 1 out of 500 individuals would have the genetic variants that cause the disease - that's the rate at which the disorder occurs in the general population. But instead, he found that 1 out of 4 individuals had those mutations and those individuals were disproportionately African Americans.
“This was the initial shocking revelation,” he said.
His team decided to figure out if those mutations were misclassified as harmful. They first looked at the initial studies that identified these genetic variants as disease-causing, focusing on five specific abnormalities that appear most frequently in the NHLBI population. They found that these studies had small sample sizes and none had representative samples of African Americans in their control groups.
They then compared genetic sequences of African Americans and whites through the 1000 Genomes Project, which has genome data from 14 populations worldwide, confirming the five variants they identified occurred most commonly among African Americans.
At the same time, they examined data from the Laboratory for Molecular Medicine operated by Partners HealthCare Personalized Medicine, a clinical lab that diagnoses and performs genetic testing for patients. By using the lab's classification system that includes data regarding the frequency of genetic variants in control populations, they determined that these five variants were actually benign. Four had been classified by the Human Gene Mutation Database in the most pathogenic category.
In the Partners' clinic records for the past decade, the authors found seven patients of African or unspecified ancestry between 2005 and 2007 who were told they had the disease-causing genes based on these misclassified variants. According to their calculations, inclusion of even a small number of African Americans in study control groups could have prevented the misclassifications.
In regard to hypertrophic cardiomyopathy, the misdiagnoses of risk may have resulted in unnecessary hardships for the patients and the families. But the lack of diversity in scientific studies and control groups can have other significant implications. Esteban Burchard, a professor at the University of California San Francisco, has studied how genetic differences can lead to higher rates of asthma among African Americans, but the drugs designed to treat diseases, he wrote, often work better in people of European origins.
Burchard published a study last year that showed less than 5 percent of lung disease studies funded by the National Institutes of Health in the last two decades have statistically meaningful number of participants from ethnic minorities.
“It's like basing your whole world … on one opinion or one biologic resource,” Burchard said. “And that's a problem because we miss the variation in genetics that is present worldwide.”
Genomic data from diverse populations is needed to find mutations specific to different ethnicities that indicate disease or in some cases, demonstrate responsiveness to treatments, he said.
Manrai points toward relatively new projects, such as the NHLBI Exome Sequencing Project and the 1,000 Genomes that now have genomic sequences from diverse populations, although there is still a need for data from Native Americans and Asian Americans.
“There is now an opportunity to use those resources to study hypertrophic cardiomyopathy and other diseases as well as reassess a lot of studies that support the genetic studies that might be decades old,” Manrai said. “Since studies in the past do not have perfect mixes and they shape the current literature, it's important to evaluate those studies with current data.”