Exploring Ancestries

Incorporating diverse populations into genetic research is crucial to understanding the biology of diseases. (2016)


By Charlotte Huff

Carol Horowitz was initially reluctant to pursue the connection when she learned about genetic variants that significantly boost the risk of developing severe kidney disease and that almost exclusively occur in men and women with African ancestry.

“My original reaction was, ‘I am not touching this.’ Because race is a social construct, and this could be really damaging,” says Horowitz, a physician and associate professor of population health science and policy at the Icahn School of Medicine at Mount Sinai in New York City.

Horowitz spoke with a genetics ethicist who also feared that linking health to a particular population’s DNA would set back efforts to reduce health disparities among racial and ethnic groups. But before nixing the idea, Horowitz sought feedback from a few community leaders, and a comment from a black pastor in Harlem particularly resonated, she says.

“His first response to me was, ‘Now, maybe when white doctors look at black people on dialysis, they won’t think that it’s because we didn’t try hard enough. They’ll realize that there’s more to sickness than bad behavior.’ ”

Horowitz and other researchers are part of a growing effort to sort out the relative significance of genetic variants, not just within population groups but also between groups of people with different geographical ancestral backgrounds.

In recent years, such research has revealed key drug insights, such as the inability of some people with an East Asian or a Pacific Islander background to effectively metabolize the frequently prescribed blood thinner Plavix (clopidogrel). Other initiatives, such as the study that Horowitz is involved with, which is called Genetic Testing to Understand and Address Renal Disease Disparities (GUARDD), are looking at whether some variants can boost vulnerability to a particular disease or condition in specific populations.

Genetic differences between populations — often described by researchers as ancestry groups — are simply a piece of the larger health puzzle, along with access to medical care, poor living conditions, and a myriad of other factors that also must be addressed, stresses Horowitz. “It’s not like we should give everybody a genetic test and not worry about the quality of schools,” she says. Genetics is “a piece of the puzzle. It’s an important piece, but it is a piece.”

Still, bypassing such research avenues would be a disservice to those who could potentially benefit, in the same way that cardiovascular researchers studied only men for many years, says Esteban Burchard, a pulmonologist who is researching asthma in Hispanics and African Americans as part of another genetics-linked initiative. Scientific advances sometimes constitute a double-edged sword, says Burchard, pointing to nuclear power as one example.

“Yes, there is a tremendous benefit to nuclear energy,” he says. “We use it in medicine every day. And yes, there are potential misuses of it. But we can’t stop scientific progress for fear of what someone will do with it.”

Building Diversity

Burchard, who is based at the University of California, San Francisco, is among several researchers who, in a 2011 Nature piece, spotlighted the low percentage of non-European participants in genome-wide association studies. Of more than 1,000 such studies, just 4 percent involved individuals of non-European descent. The genomic revolution, Burchard says, “is like a giant tidal wave of information. But whole swathes of the global population missed the wave because they weren’t included.”

Teri Manolio, the director of the division of genomic medicine at the National Human Genome Research Institute (NHGRI), disputes that figure, describing it as out of date. She couldn’t cite a more recent source, but Manolio estimates that non-European participation is approaching 20 percent, based on ongoing efforts in the U.S., as well as in other countries like China and Japan, to increase the diversity of study participants.

As some examples, Manolio points to the Human Heredity and Health in Africa (H3Africa) project and the genetics analysis arm of the Hispanic Community Health Study/Study of Latinos. The Precision Medicine Initiative announced by President Obama last year also includes a research cohort of at least one million people that will reflect America’s racial, ethnic, and socioeconomic diversity.

But Manolio and her NHGRI colleague and Genome board member Vence Bonham Jr. acknowledge ongoing challenges in boosting non-European participation, ones that have led them to target recruitment of racial and ethnic populations, underserved populations, or populations who experience poorer medical outcomes. (See “Recruiting Minority Participants,” page 34.) Among the challenges, they say, is a mistrust of the medical system rooted in prior injustices involving research. There also can be practical hurdles, says Bonham, a senior advisor to the NHGRI director on genomics and health disparities.

Some communities, he points out, are more immediately concerned about better access to basic medical care than the longer-term promise of genomics research, he says.

But education about genomics is making some headway. As researchers continue to identify variants that can affect patient care, such as drug dosing or which drugs to prescribe in certain populations, Bonham says that genomic research “may make the difference in getting the right drug to the right individual.”

Exploring Ancestries

Genetically speaking, we humans are far more similar than we are different. Each of our genomes contains 3.2 billion “letters,” which are 99.9 percent identical from one person to the next, says Timothy Thornton, one of the co-investigators in the genetics analysis arm of the Hispanic health study and an associate professor of biostatistics at the University of Washington School of Public Health. (Thornton also notes, however, that 0.1 percent of 3 billion is still a lot of variants.)

Demographic, health, and other information have already been compiled from the more than 16,000 Hispanics involved in the study. They reside in four urban areas of the U.S. and self-identify with various backgrounds: Cuban, Dominican, Puerto Rican, Mexican, Central American, or South American. Nearly 13,000 consented to genetic testing, resulting in more than 25 million genetic variants, according to Thornton.

As a consequence of the slave trade, as well as colonization of the Americas, both of which occurred in the last several hundred years, Thornton says, “you see that their genomes are almost a mosaic of these different ancestry groups.”

In a study published early this year in the American Journal of Human Genetics, Thornton and his colleagues used a color-coded graphic to illustrate that ancestral mosaic effect. Participants who self-identified as Cuban were overwhelmingly linked genetically to European ancestry, with a much smaller portion of African-American and American Indian background. On the other hand, Mexican participants had a genetic lineage that was roughly evenly split between European and American Indian, with a smaller genetic influence from Africa.

Burchard, who is working on two large studies involving African-American and Hispanic children with asthma, describes how he was struck years ago by the widely differing rates. Asthma diagnoses are most common among individuals of Puerto Rican origin and least common among those of Mexican background, with whites and blacks falling in between, he says. “Here you have this very common disease, but when you look at the extremes of the distribution of the prevalence, you had two Latino groups on the edges.”

In the years since, Burchard and other researchers have shown how non-genetic influences can play a role. In one 2013 study, which involved African-American and Hispanic children, Burchard and his colleagues found that exposure to air pollution before the age of 1 boosted the risk of later being diagnosed with asthma. Another analysis, which looked at asthma patterns in the Hispanic health study, found that age of immigration to the U.S. made a difference, although not in all Hispanic populations.

Puerto Ricans were most likely to suffer from the respiratory condition, regardless of whether they had been born in the U.S. or when they had moved to the country.

But among other Hispanic groups, moving to the U.S. before age 6 markedly boosted the risk, according to the findings published in 2016 in the American Journal of Respiratory and Critical Care Medicine. While the study didn’t look at causes, researchers questioned whether exposure to urban asthma triggers at a young age, such as pollution, mold, or cockroaches, could have boosted vulnerability.

But do genetic variants also come into play, predisposing people of Puerto Rican origin to asthma? Burchard and his colleagues tapped data from one of his ongoing research studies, Genetics of Asthma in Latino Americans (GALA), to look for some genetic clues. They found that genetic differences in Puerto Ricans but not in Mexicans appeared to play a role in the severity of the asthma, as well as in a lower response to albuterol, the most common asthma treatment.

With the help of large, racially mixed data-bases, such as that in the Hispanic health study, researchers can identify variants across ancestry groups that might otherwise be missed, according to Thornton. Earlier this year, Thornton and his colleagues reported on a variant associated with a lower platelet count (a factor related to blood clotting) specific to American Indians in the Hispanic health study, one that hadn’t been found in other studies because they lacked a sufficiently large sampling of individuals with some degree of American Indian ancestry to capture it, Thornton says.

The next step, Thornton believes, is to determine if that variant is clinically relevant. “It may not be that big a deal,” he says. Or it could be that people who carry that variant, along with other risk factors, could be more prone to clotting difficulties.

How Actionable?

As researchers explore potential genetic variants in population groups, it’s critical that they home in on those that are clinically relevant, Horowitz says. In some cases, the variant ultimately might not be meaningful, she says. “Or it might contribute so little to the risk of disease or the risk of medication response that we’re really just … scaring people for nothing.”

While both blacks and whites can struggle with high blood pressure, those with African ancestry have been found to have a fivefold risk of end-stage kidney failure compared with whites, a serious condition that can require dialysis or a kidney transplant. Two variants in the APOL1 gene have been identified that boost the risk of kidney disease and appear to occur almost exclusively in those with African ancestry, likely because they protect against an illness — dubbed sleeping sickness — that’s transmitted by tsetse flies in some parts of Africa. One in seven adults with African ancestry is believed to carry the two variants.

In 2014, Horowitz helped to launch the GUARDD study, which has recruited about 1,500 patients so far to educate and screen for the variants. Carolyn Caddle-Steele, who has high blood pressure and says the cardiovascular risk factor runs in her family, took further steps to improve her health after her APOL1 test came back positive.

The 51-year-old New York City grants manager has lost weight, has committed to exercising a few times each week, and has incorporated nutritious eating into her lifestyle. “There are certain things I would love to eat every day,” she says with a laugh. “I could, but I don’t.”

In a commentary published last year in PLOS Medicine, Burchard listed the APOL1 variants among examples of insights gained to date from genetically diverse data. Another variant, which has been traced back to Native American ancestry, helps to reduce the risk of breast cancer in one out of every five Hispanic women. And, as noted previously, genetic differences more common in East Asians and Pacific Islanders render Plavix less effective in these groups, due to difficulties with metabolization. In 2014, the Hawaii attorney general filed a lawsuit against the drug’s manufacturers, saying they had failed to alert patients about their potential genetic vulnerability.

Language Matters

As genetics research moves forward, its continued reliance on racial categories can be misleading and potentially stigmatizing, says Michael Yudell, who’s based at the Drexel University Dornsife School of Public Health in Philadelphia. Yudell co-authored an opinion piece, published last February in the journal Science, which called on the National Academies of Sciences, Engineering and Medicine to convene a panel to develop alternative terminology.

On a practical level, specific racial categories are not necessarily well defined by individual studies. Nor do the studies explain why racial categories are relevant, Yudell says. Another problem arises when people are asked to report their own race or ethnicity — the way they describe themselves might not necessarily align with their genetic mix, he says. “If you’re studying a particular [genetic] trait and its relationship to race,” he says, “how can you draw the conclusion that race may be a factor if there’s no consistency in the use of how race is identified?”

In a May 26, 2016, commentary in the New England Journal of Medicine, Bonham and his colleagues wrote, “Although self-identified race may correlate with geographical ancestry, it does not predict a patient’s genotype or drug response.”

Moreover, artificially distilling human diversity into racial categories, rather than looking at ancestries or populations, can influence how patients are perceived by clinicians, Yudell says. Cystic fibrosis can be missed because it’s considered a white disease; sickle cell disease might be perceived to develop only in black patients.

A racial framework also can risk harming patients, says Yudell. In a recent study, he says researchers first determined that roughly half of white medical students and residents held false beliefs about blacks, such as the misconception that blacks have thicker skin than whites. The students and residents who did so were more likely to rate a black patient’s pain as less severe.

Caddle-Steele, who sits on a subcommittee for the GUARDD study, has heard the arguments that looking for race-associated genetic differences could backfire. But she’s glad that researchers are pursuing such potential links, not only for herself and her daughter, who might want to get tested one day, but for the larger African-American community. Too many African-Americans don’t know they have elevated blood pressure or don’t understand the long-term risks, she says.

Caddle-Steele feels that “the more information we have, especially if that information is well researched and well documented,” the better. She adds that “if there is any other way that we can get that information out to the African-American community, I don’t have a problem with it.”