So What’s Wrong with Whole Genome Sequencing?
Whole genome sequencing is the process of determining the entirety, or nearly the entirety, of the DNA sequence of an organism's genome. It has been offered to determine an individual’s risk of future disease based on a subject’s DNA mismatch from “normal.” However, other than relatively rare monogenic diseases (caused by one gene), most illnesses and traits result from multiple gene interactions coupled with environmental and lifestyle.
Accuracy – or Tea Leaf Reading?
Whole genome sequencing might prove useful to someone already sick to concoct an individualized precision-medicine cocktail. But for predicting future disease, based on the available data we now have, its validity is akin to reading tea leaves.
Even if the interplay of environmental, lifestyle and multiple genetic variables could be addressed, all this data can do is project a population -- not individual -- risk of disease. That’s the first problem with extrapolating the new knowledge, it’s not precise. But there is more.
The extent of ethnic influences on genetic expression is not yet known. Whole genome sequencing data has focused on white Europeans, constraining extrapolation to other ethnicities. Recently, 100,000 whole genome sequences, half from African-Americans, Hispanic, or Asian, have been sequenced, supposedly making the database more relevant to the non-white population. But much of the data is only available to researchers associated with certain American universities. Exactly how much of the 50% of the data collected belongs to which of the subgroups is not readily known, rendering reliance by non-whites risky.
The Death of Privacy
Some enthusiasts ignore the accuracy concerns and call for the creation of a national DNA database where we would assemble the whole genome package of all our citizens. – the next step in Pandora’s progress. The British have been planning (plotting?) this since 2013. This whole genome sequence data supposedly would be anonymized (but the anonymization would be reversible and simple to untangle). Its use is proposed to be available to private companies, including Google, without people’s knowledge and consent. Suffice it to say, the abuses, including privacy violations, are rather horrific to contemplate.
“Genomed” at Birth
Compiling whole-scale genomic information for a willing and consenting adult might fall under the realm of autonomy, although the cost for such testing would create its own disparities and social justice concerns. Engaging in the same practice for newborns, as proposed by the British Newborn Genomes Programing project, poses its own problems. While the information could disclose risks of hundreds of conditions, some are without cure or preventive measures. That knowledge might well be considered a burden – not a blessing.
In the US, some 29 diseases are routinely screened for at birth. But these are diseases where early diagnosis and treatment are available, increasing the chances of healthy development and reducing death. Many such screens are mandatory, although some opt-outs do exist.
Using whole genome sequencing to achieve this end would disclose diseases, both with treatment or preventatives and those without. Of what use would the latter of knowledge be?
Consider the discussion of Huntington’s Chorea by my colleague, Dr. Dinerstein, disclosing that most adults wouldn’t even avail themselves of knowing they would have a disease without cure or treatment or preventive measures even if it were available.
Can We Handle the Knowledge?
Another objection is, do we have the person-power to address findings that might be treatable? At recent conferences discussing the issue, objections were voiced by the healthcare industry, which anticipates an overflow of conditions without sufficient personnel to treat them, along with questions regarding the training of healthcare workers [and genetic counselors].
And What About the Rights of the Child?
Amongst issues that were not raised were the rights of the child. It is one thing to identify childhood diseases that a parent might be able to prevent. But the Whole Genome Project doesn’t limit itself to illness by age – knowledge of the panoply of diseases for which a baby is at risk in their future life would be available to the parent and, by extension, the siblings. Perhaps a child, on reaching adulthood, wouldn’t want their parent to know they are at risk of depression, Alzheimer’s, or diabetes? Should that knowledge be made available?
Pre-Implantation Genetic Testing
Pre-implantation Genetic Testing (PGT) is now used to identify monogenetic diseases in newborns – testing to determine illness caused by a single gene. It determines which embryos are suitable for implantation. In the UK, its use limited - legal to prevent serious illness.
“Genomic Prediction is the creator of LifeView, an advanced genetic testing platform for embryos for … IVF. We help couples have the healthiest child possible, protecting your baby from genetic risks that run in your family."
Recently, companies have expanded single-gene tests to encompass a broader range of genes - to assess not just a specific disease but the likelihood of a particular trait. Companies like Genomic Prediction advertise polygenic embryo screening, or PGT-P, to assess genetic data involving multiple genes (although not the whole genome) to predict the “healthiest” baby. [2] These PGT-P tests evaluate a series of genes arguably related to “healthful” traits but could be gerry-rigged to select for intelligence, aesthetics, or athleticism in the future.
Genomic Prediction and its competitors use algorithms in their analysis, relying in some cases on adult populations that may not even be relevant. Even if a PGT-P trait was relatively accurate in assessing relative (population) risk of disease, the value in predicting absolute risk (actual risk in individual) is very small. A 50% increased population risk translates into an absolute risk of that trait in an individual of 0.5% - a hundred-fold reduction.
“At present there is no evidence that PGT-P can reliably predict the relative risk of an embryo developing a certain trait of disease.”
– Dr. Francesca Forzano, Consultant in clinical genetics Guy's & St Thomas' NHS Foundation Trust, London
The best the tests can currently do is approximate which embryo is the hardiest and has the least risk for future biologic disease at the time of selection. The effects of environmental stressors, including pregnancy, aren’t included in the analysis – giving the would-be parents a false sense of security.
It's not surprising that the European Society of Human Genetics, the European Society of Human and Reproduction Embryology, and the American College of Medical Genetics have released statements advising against using polygenic risk scores for embryo selection. In the UK, its use is illegal.
Solomon’s Choice
Some bioethicists, like Hank Greeley, oppose the technique, leaving open the possibility of its future use, should they prove more accurate, claiming well-informed parents should be able to make their own decisions.
But the process also determines which embryos are rejected. Even if the testing proves more accurate, its use might be a societal tragedy. Selecting the ostensibly “healthiest” embryos means the less “healthy” ones are rejected. Superficially, this might seem beneficial, individually as well as societally. But this process also selects against geniuses such as Stephen Hawking, who suffered AMLS, or schizophrenic Nobel Prize winner John Nash.
Determining the “healthiest” or “best” baby to be born might evidence hubris rather than wisdom on the part of he or she doing the selecting.
Future Risk
The problems raised by PGT-P, which confines itself to groups of genes, are magnified by attempts to perform whole-genome sequencing at the embryo stage. These additional tests promise to identify the entire genomic structure of the would-be child before it is even born. Now, instead of selecting the healthiest baby, parents might specify one that is also beautiful and brawny – weighting whichever traits they deem most important to them.
Relevant molecular and statistical techniques published last month in Nature Medicine touted a high degree of validity for this technique in predicting genetic risks for adult-onset diseases, including diabetes and breast cancer, when used in an embryo.
Ethical questions that should have been addressed regarding using whole-genome sequencing first in adults and then babies, now raise themselves for embryos.
This applies to privacy rights belonging to the embryo once it is born. Even assuming predictions can be accurately made, fore-knowledge of diseases that may surface in adulthood, like atrial fibrillation, cancer, or dementia, is not something an individual might want their parents, employers, or colleagues to know. And once the knowledge-genie is out of the bottle, it’s impossible to stuff it back in.
So, we keep trucking along with the technology before fully considering the ethical implications that would be unleashed. And initial thoughts may well change. [3]
Present Risk Ignored: And Virtues of the “Old-Fashioned Way”
The excitement of the new always seems to eclipse the banalities of the old. Certainly, infertile parents may require IVF, but as the hype around PGT-P increases, fertile would-be parents might be tempted to avail themselves of IVF to enable themselves of the service. That would be unwise.
IVF is associated with
- increased vascular complications and irregular heartbeat during pregnancy
- higher likelihood of kidney injury
- placental abruption
- ovarian hyperstimulation syndrome.
IVF, then, should not be the first choice for conception.
Instead of searching for more knowledge, perhaps we should concentrate on safely using the knowledge we have.
[1] Genesis chapter 2
[2] at least four companies are advertising the service.
[3 Even “bioethicist” Julian Savelescu, who, as recently as June 2021, advocated a parental obligation of “procreative beneficence,” requiring selecting the child with the chance of the “best life” using PRS, changed his tune by October.
