Genome sequencing is enabling new insights into the genetic aspects of health and disease that have touched just about every aspect in biomedicine.
It is also, like the “skin”-colored crayons of yore, disproportionately focused on the Caucasian segment of the population.
And that is a loss for everyone.
How lack of representation in research harms those that are not represented is obvious.
Unless a concerted effort is made to ensure broad inclusion of different racial groups, existing health care disparities will carry forward seamlessly into the genomics area.
For example, although African American and Puerto Rican children have higher rates of asthma than other populations, the overwhelming majority of clinical research on lung diseases has been done in Caucasians.
The unsurprising result is that asthma drugs do not work as well for African American and Puerto Rican children, although those groups have the highest prevalence of pediatric asthma in the U.S. Compared to Caucasian or Mexican American children, African American and Puerto Rican children are four to five times more likely to die from asthma.
The same dynamic is playing out in genomics-based medicines.
A recent report in Nature Genetics, for example, reported that in a study of schizophrenia risk variants in roughly 23,000 East Asians and 35,000 controls, though many risk variants overlapped, polygenic risk scores, which try to predict individuals’ overall risk of developing schizophrenia, did not transfer well to their dataset. Risk score methodology has been developed using European-ancestry genomics data.
In October, the FDA approved Trikafta (elexacaftor/ivacaftor/tezacaftor, Vertex Pharmaceuticals Inc.) for patients ages 12 and older who have at least one copy of the F508del mutation in the CFTR gene.
The approval made Science magazine’s Breakthrough of the Year list. “For those who have the mutation – about 90% of all CF patients – it could convert CF from a progressive disease into a more manageable chronic illness,” the magazine asserted in its reporting.
But in reality, it’s 90% of Caucasian CF patients.
Around the time of Trikafta’s approval, Meghan McGarry, assistant professor of pediatrics at the University of California, San Francisco; Esteban Burchard, a UCSF pulmonologist and geneticist studying health disparities in asthma; and their colleagues reported that Dominican and Puerto Rican individuals who suffer from CF were less likely than Caucasians to have the genetic variant that Trikafta is approved for.
More than two-thirds of Dominican patients, and 10% of Puerto Rican patients had none of the known functional mutations in the CF gene that are seen in 95% of CF cases.
Puerto Rican and Dominican populations also differed substantially from each other in their specific variants, which was a surprise to McGarry: “I thought that the Dominican patients and the Puerto Rican patients would be more similar in their mutations,” she told BioWorld.
The variant that Trikafta’s label specifies, p.Phe508del, is present in nearly 90% of CF patients in the mainland U.S., but occurred in only a third of Puerto Rican patients and in 10% of Dominican patients.
“Ideally, it probably works for other mutations,” McGarry said. But “how are you going to figure that out?”
Not from the studies that led to Trikafta’s approval, which “don’t even list race and ethnicity of their study populations,” she said.
“Ideally, you would give every patient a month or three months of that medication,” she said. But at a list price of $300,000 annually “it’s too expensive to do that.”
As things stand, Trikafta is both a legitimate breakthrough and an illustration that in the absence of active mitigation, the more things change, the more they stay the same.
McGarry and her colleagues are among those working for such change. So is Karestan Koenen, professor of psychiatric epidemiology at the Harvard T.H. Chan School of Public Health (HSPH) and principal investigator for NeuroGAP-Psychosis, and her colleagues.
Mend the GAP
NeuroGAP, or Neuropsychiatric Genetics in African Populations, was formed to address “the lack of specifically African representation in specifically psychiatric studies.” Koenen told BioWorld.
That lack hampers progress in psychiatric genomics as a whole. In Ethiopia, for example, “major work has been done on psychosis, but those people aren’t represented in psychiatric genomics” – a situation that NeuroGAP wants to help rectify.
Koenen said that in addition to the project’s size, it is notable for its commitment to enhancing African capacity for genomics research. She is co-principal investigator with seven African academics, and “the places we do work will be empowered to do more complex research and be able to analyze the complex data they are generating from their own people,” she said.
The approach is in sharp contrast to what is sometimes termed “safari research,” where, from the point of view of the African collaborators, researchers “have come to do research… and they just come here, take the data and leave, and not only are we not any better off for it, but we never hear from them again.”
Koenen noted that self-interest is as good a reason as social justice for researchers to pay attention to minority populations.
Ten percent of the African genome is missing from the current reference genome, but “Africans have the greatest diversity,” she told BioWorld. The major metabolic disease target PCSK9, for example, was first identified in case studies that included several African families.
Another recent study in Caribbean Hispanics showed that African-derived ApoE, the strongest genetic predictor of Alzheimer’s disease (AD) risk in Caucasians, has only half as much influence on AD risk as European-derived ApoE. The E2 allele was less protective, and the E4 allele less damaging. Understanding why might help figure out treatment or prevention options for those with African and European ancestry alike.
“If we don’t study African populations, we’re missing a lot of variation… a lot of data [that] can inform drug development that is for all populations,” Koenen said.
Equitable representation is an important goal, but also a challenge. As McGarry’s research into Dominican and Puerto Rican populations illustrates, in part, the more diversity researchers include, the clearer it becomes that there are populations within populations.
In a paper published in October in Cell, a Singaporean-led multinational team took advantage of the nation-state’s combination as historical immigration hub and top-notch modern research enterprise to sequence nearly 5,000 Singapore residents of Chinese, Indian and Malay descent. The work, which uncovered more than 50 million novel single nucleotide polymorphisms (SNPs) and small insertions/deletions (indels) is part of the SG10K project, a national effort aiming to whole-genome-sequence 10,000 Singaporeans. Fourteen of those loci, the authors wrote, “harbored robust associations with complex traits and diseases… [highlighting] the value of our data as a resource to empower human genetics discovery across broad geographic regions.”