Why screening DNA for ‘designer babies’ probably won’t work

Picking embryos based on genetics might
not give prospective parents the “designer baby” they’re after.

DNA predictions of height or IQ might
help would-be parents select an embryo that would grow into a child who is, at
most, only about
three centimeters taller or about three IQ points smarter
than an average
embryo from the couple, researchers report November 21 in Cell. But offspring predicted by their DNA to be the tallest among
siblings were actually the tallest in only seven of 28 real families, the study
found. And in five of those families, the child predicted to be tallest was
actually shorter than the average for the family.

Even if it were ethical to select
embryos based on genetic propensity for height or intelligence, “the impact of
doing so is likely to be modest — so modest that it’s not likely to be
practically worth it,” says Amit Khera, a physician and geneticist at the
Center for Genomics Medicine at Massachusetts General Hospital in Boston who
was not involved in the new study.

For years, couples have been able to use
genetic diagnosis to screen out embryos carrying a disease-causing DNA variant.
The procedure, called preimplantation
genetic diagnosis
, or PGD, involves creating embryos through in vitro
fertilization. Clinic staff remove a single cell from the embryo and test its
DNA for genetic variants that cause cystic fibrosis, Tay-Sachs or other
life-threatening diseases caused by defects in single genes.

Many diseases such as diabetes and heart
disease, as well as traits like height and intelligence,
are considered complex because they are caused in part by tiny effects of
variants in hundreds or even
thousands of genes
(SN: 5/31/13; SN: 9/29/10) But researchers can boil
down those tiny effects of multiple genes into one “polygenic”
score (SN: 4/18/19)
. Khera was
involved, for example, in compiling 6 million genetic variants into a risk
score for heart disease.

Advances in this technology in recent
years, however, have sparked debates about whether people should be allowed to use
these scores to select embryos with complex traits, such as height or
intelligence. Some people say that selecting for “enhancements,” gender
preferences or other nonmedical traits smacks of eugenics, the practice of breeding
humans toward some desired outcome. The term is now often associated with
racism, genocide and forced sterilization campaigns. The findings in the new
study bring a dose of reality to these discussions about designer babies and
selecting embryos for reasons other than medical conditions.

“We wanted to use these numbers or these
methods to make the debate more quantitative, more evidence-based,” says Shai
Carmi, a statistical geneticist at the Braun School of Public Health of the
Hebrew University of Jerusalem.

Carmi and colleagues used data from
studies of longevity and schizophrenia to look at what might happen in embryo
selection. Some people in the longevity study were real-world spouses, so the
researchers simulated genetic profiles for the couples’ hypothetical offspring.
The researchers also created virtual couples by pairing unrelated people and simulating
the genetic makeup of those faux couples’ embryos. Polygenic scores predicting
height or IQ were then calculated for the embryos.

Among 10 embryos, the one that the
genetic scores suggested would be tallest was predicted to be only three
centimeters taller than the average height predicted for all 10 embryos, the
team found. Similarly, the highest intelligence polygenic score predicted the
resulting child would be three IQ points smarter than the average for all 10 embryos.
With only five embryos to choose from (a more realistic number based on numbers
of embryos generated in fertility studies), the gains drop to about 2.5 centimeters
and 2.5 IQ points, the researchers found.

That’s only the predicted result. To see
whether the predictions would hold up in the real world, Carmi and colleagues looked
at 28 large families, each with up to 20 children. In those families, the child
predicted by genetic score to be the tallest was actually the tallest in only about
a quarter of the families. In fact, the child predicted to be the tallest was
actually about three centimeters shorter on average than the family’s tallest
child, the researchers discovered.

The study “does add more data,” says
Susanne Haga, a human geneticist at Duke University School of Medicine. But there’s
too much variability in these complex traits to accurately predict an outcome
just by looking at DNA. “There’s still a great deal of
variability that’s not accounted for by the genes they were analyzing or
simulating in their polygenic score. Therefore, you’re still going to see a
wide distribution of height or IQ points.”

Differences in diet, lifestyle, exposure
to pollution, culture, undiscovered genetic variants and other unknown factors can
also influence how complex traits develop, Haga says. “We still have a ways to
go to understand the genetic mechanisms behind these traits, and the fact that
the environment plays such a critical role cannot be discounted,” she says.

Other researchers expressed concern that
the study’s findings could be misunderstood or misrepresented to promote
picking embryos for nonmedical traits.

In the new study, “the science is
rigorous,” says Nicholas Katsanis, a human geneticist at the Ann & Robert
Lurie Children’s Hospital of Chicago. But he worries it could encourage
fertility clinics to offer genetic screening for height and IQ to entice
would-be parents hoping to give their future children an edge, even though the
study suggests that edge isn’t really predictable.

“The fundamental question should not be asked in the first place, because the idea that we’re going to do genetic screening for anything other than medically actionable items is the definition of eugenics,” Katsanis says. “That we’re even contemplating this is disturbing.”

Source link

Show More

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: