Editing the Human Genome: Progress and Controversies

One of the most provocative and interesting possibilities on the current biomedical scene is the question of whether we can or should alter the genomes of human embryos. It is fairly clear that we can. What is not decided is whether we should. Our three-year-old ability utilizing CRISPR to edit DNA quickly, economically, and precisely trumps Darwinian evolution and opens virtually unlimited opportunities to modify and/or create novel life forms. Inevitably, as Jennifer Doudna has pointed out, CRISPR will be used to edit the human embryonic genome, and those modifications would be passed on to future generations.

Dr. Legato: Let's begin with Dr. George Church who is the Director of the Harvard Medical School Center of Excellence in Genomic Sciences. Dr. Church, were the genome to be edited, how would the biological sex of the embryo impact the editing?

Dr. Church: It seems unlikely that the sex of the embryo would matter for the editing itself, except maybe for X-linked diseases. In addition, the gametes can be edited, with potentially lower risk to embryos and to the mother by avoiding in vitro fertilization (IVF) protocols. The sex could be a target for editing, for example family gender “balancing.” In the United States, some estimates are for 80% choice of girls for birth.

Dr. Legato: Dr. Virginia Miller is the Director of Women's Health Center at the Mayo Clinic. Dr. Miller what are your thoughts on this topic?

Dr. Miller: I agree with Dr. Church that the technique could be applied to either sex and to the gametes. However, I do not think we know the answer to how manipulation of the gamete (egg or sperm) or even autosomes of male and female embryos would affect the expression of the manipulated gene-directed phenotype. It will depend on which gene is being modified and how that particular gene might be affected by other conditions in regulatory systems that are dictated by the X and Y chromosomes or hormones produced by the ovaries and testes, the development of which are regulated by sex chromosomes.

In published reports of studies where the CRISPR techniques were used specifically to treat cystic fibrosis in adult humans, there were very strict inclusion and exclusion criteria or instructions regarding treatment of a woman who is pregnant or of a man who is in a relationship with a female partner who might become pregnant.

Those studies that have been conducted so far have been really cautious—and appropriately so—in terms of what might happen to the offspring of individuals treated with genetic therapy such as CRISPR. However, regarding the use of CRISPR to alter the gamete itself, I do not think we have enough information regarding how expression of the manipulated gene would differ in males and females across the life-span, that is, in the presence or absence of steroid hormones.

In using the CRISPR technique on a gamete, the outcome would seem to depend on whether the gene being modified is on an X chromosome or a Y chromosome, and how the genes on the autosome are modulated by the sex chromosome. We know, for example, that a whole cluster of genes on the Y chromosome can affect autosomal gene expression.

To me, it seems that it would be easier to manipulate the gamete rather than the embryo; it might be easier to track the outcomes. But again, I think these are questions that have yet to be explored experimentally.

Dr. Legato: Professor Henry T. Greely is the Director of the Center for Law and the Biosciences at Stanford University. Dr. Greely, can you comment?

Prof. Greely: I agree that they have yet to be explored experimentally, but I think Dr. Church is clearly right here in theory. If you can come up with cell lines of gametes or gamete progenitors and do the editing in those, at least you will know whether the editing took properly, whether there were off-target implications, and so on. This approach would avoid the mosaicism problem, which occurs when not all the cells in the embryo pick up the same changes.

Dr. Legato: Professor Greely, in an interview on March 14, 2016, you urged a balanced regulatory approach to gene editing, but you also pointed out that there is no time to lose. Furthermore, you felt that the editing of nonhuman genomes was an immediate concern and that gene editing in humans was in the far-distant future. Is that still your feeling?

Prof. Greely: I am not sure that I would say far-distant future, but it is in the distant future, mainly because we care so much more about humans and regulate human uses much more carefully. Lawyers get involved in human uses. Mothers have to be willing to take the risks involved, as do IVF clinics, physicians, and others. In contrast, if someone wants to modify the genome of mosquitoes or of Escherichia coli, both the laws and the practical barriers to those modifications represent infinitely smaller obstacles than they would for doing any work with humans.

I am much more concerned about our unpreparedness for dealing with nonhuman modifications through CRISPR, particularly if coupled with a technology called gene drive, which will move changes through populations quickly, than I am about issues with humans. Issues with humans are interesting, but we have time to work them out.

Dr. Legato: Mr. Sherif Moussa is a legal consultant for the Foundation for Gender-Specific Medicine.

Mr. Moussa: Professor Greely, could you further explain what you mean by a balanced regulatory approach?

Prof. Greely: The description would differ depending on whether you are talking about human or nonhuman uses. With humans, safety issues have to be taken extraordinarily seriously. I would envision a regulatory approach similar to that of the U.S. Food and Drug Administration (FDA), which looks at the costs versus the benefits, the risks versus the potential benefits, safety, and efficacy as a combined concept. That approach makes a lot of sense.

With nonhuman uses, we have two different problems. One is that our current regulatory regime for gene modifications in nonhumans—the so-called Coordinated Framework for Regulation of Biotechnology—does not work very well for CRISPR technologies. The U.S. Department of Agriculture recently decided that it did not have jurisdiction to look at a CRISPR change in mushrooms under the Coordinated Framework. The second issue is that even if the current regulatory approach did work, it is not very good at catching the 22-year-old college graduates doing work in their garages who, with as little as $300, can use CRISPR to start changing genomes.

Dr. Legato: Mr. Justin Killian is a legal consultant for the Foundation for Gender-Specific Medicine.

Mr. Killian: In addition to what Professor Greely has correctly pointed out about having a much longer lead time and a more organized process when dealing with something as momentous as changing the human genome, there is another fail-safe. With the human genome, there will undoubtedly be very strongly drafted bilateral contracts and the specter of potential litigation by another human party sometime in the future. This concept of a private mechanism of some due diligence would be completely absent when you are talking about a nonhuman modification, in my opinion.

Dr. Legato: Dr. Tetsuya Ishii is Professor of Life Science Section at Hokkaido University in Japan. Dr. Ishii, you urged a voluntary moratorium on research concerning the editing of the genome, and you recently summarized the attempts at regulation by 17 countries.

Dr. Ishii: Yes, but I would like to stress that the necessity for a moratorium is on clinical application. I think that basic research may be allowed in some limited cases, but clinical application should not at present be conducted in any country.

The U.S. National Institutes of Health (NIH) does not fund human embryo research, so the main focus is on China and Japan, where there are no strict regulations. I worry about what will take place in these Asian countries. The people are concerned about the future of genome editing research.

Dr. Legato: Dr. Doudna has suggested that the application of CRISPR techniques to the human genome was inevitable. Do you agree with that?

Dr. Ishii: Yes, I agree.

Prof. Greely: I think Tetsuya and I probably are not in disagreement. It certainly should not be done clinically unless or until it has been rigorously shown to be safe and effective, but most importantly safe. I am confident that Jennifer Doudna would agree with that as well.

Dr. Miller: In my view, many safety issues have not been addressed at all. We do not know whether targeting one of these genes is going to affect mutations in something else that we have yet to discover. I would ask what the reason is for editing the genome. Is it for therapeutic purposes, to cure disease, or is it for some other more esoteric reason that smacks of eugenics?

If the purpose is therapeutic, consider that conditions caused by a single gene modification are rare. When we talk about genetic associations with other disease entities, those are complex disease traits, and you cannot always pull up one gene that needs to be modified—take cardiovascular disease, for example, or Alzheimer's disease. Where do you begin? These are multifactorial diseases compared with cystic fibrosis or Huntington's disease, which are single gene–associated disease traits.

Dr. Legato: Do you foresee an environment in which cosmetic choices, if you will, could be made for a prospective child?

Dr. Miller: I think that will be a possibility. Whether that should be allowed is another issue—an ethical issue that needs to be wrestled with—and it is a very slippery slope.

Dr. Legato: What about access to the advantages of such modifications? It certainly would not be universal. How would you choose the candidates and regulate what is appropriate?

Dr. Church: It would probably be similar to orphan drugs initially. I would be surprised if this did not become quite inexpensive and widespread eventually.

I have personally experienced exponential drops in price of three-million-fold for DNA sequencing and a roughly similar amount for various aspects of synthetic biology. I would not be completely surprised if this, like cell phones and other revolutions, could also drop in price. That is one way it could be delivered equitably. I think it also requires some political will, but it could be that it may depend mainly on technological reductions in price.

Prof. Greely: Access would probably depend on the specific use. If you are talking about successful, safe, and effective methods to cure or prevent disease, I do not see why, at least in developed countries, that would not be universally available.

Almost every developed country has a universal healthcare system. I reside in one of the few exceptions, but I am hopeful we will become civilized eventually. If it turns out that a CRISPR modification is the best way to prevent or cure a disease, then I think it will probably have fairly broad usage. If, on the other hand, it is an enhancement issue, then the access will likely be quite different and will be based on ability to pay in most if not all places.

I am not terribly worried about enhancement uses yet, frankly. There is almost nothing we know how to enhance through gene editing at this point. I am not sure that will change very quickly, in part because, as pointed out earlier, many of the things we would like to change are multigenic and involve not only many different genes with different alleles, but also the effects of many different environmental influences.

Dr. Legato: That brings me to the question of how to predict the impact of the environment on edited DNA. I think it will require very long-term follow-up.

Dr. Miller: I agree, because we know that the environment modifies the human genome at all stages of development. There is no reason to think that those same environmental pressures are not going to affect an altered genome. Environmental stimuli affect demethylation processes and other epigenetic modifications, and these change with age and hormonal status. These processes will continue regardless of what the genome is, and we do not know how gene editing might affect them.

Prof. Greely: I agree, but I also think it depends on the type of genetic editing that is performed. Consider for example Huntington's disease, in which if you have more than about 37 CAG repeats in a row in the Huntington's gene, you will get the disease and die from it.

Let us say that the goal of gene editing is to take someone who has 47 CAG repeats in the Huntington's gene and reduce that number to 22 CAG repeats. I do not know whether there is any reason to think that the changed allele, which is much, much more common in the general population than is the disease-causing allele, will have any special epigenetic properties.

Dr. Church: If the mutation is back to a common allele, then the environmental effects should be similar to the environmental effects on the common allele present in most people.

Prof. Greely: If instead you were creating a completely new, never-before-seen-in-humans allele, then I think the issue would be quite serious. The only exception would be if there were something about the process of “CRISPR-ing” that left its mark on the chromosome in a way that made it open to different epigenetics. That, however, is the sort of thing we should be able to test empirically, and to test in nonhumans as well as in human cell lines.

Dr. Legato: The legal implications of parents' rights to manipulate the genome of their embryo have not been explored. It is likely that a debate will evolve similar to the one surrounding the rights of parents to determine the sex of children with developmental sex disorders such as ambiguous genitalia.

Dr. Church: Yes, that does sound like a precedent. Another might be mitochondrial germline therapy, which has been done in many cases. Yet another might be cures for infertility by manipulating gamete precursor cells (not embryos).

Dr. Legato: Mr. Killian and Mr. Moussa, what are your thoughts on the legal implications of a parent's right to manipulate the genome of the embryo, and how do you think that will be addressed if the opportunity becomes available?

Mr. Killian: As Professor Greely has correctly pointed out, this is really a hypothetical discussion at this point, given that human uses are likely so far off in the future. That said, I believe we should come to a point at which it is possible to provide parents with the opportunity to screen for certain characteristics and some disease-causing characteristics and eliminate them. That would probably trigger a very vigorous dialogue from a lot of different parties.

What will be interesting to me from the perspective of someone who has been engaged in commercial litigation is at what point will lawmakers seek to regulate this, and, frankly, will they have the scientific expertise to do so effectively? I do not know, but it would be an unprecedented type of regulation if there were to be an attempt to regulate it comprehensively before it is a concrete possibility.

Mr. Moussa: I agree with what Mr. Killian said. I envision this as being somewhat of a political issue, and I very much agree that it will yield a very vigorous debate.

I wonder if that might mirror in some ways at least what we see in the abortion debate, where the usual suspects will take different sides: a left-wing feminist might say it is a woman's choice, it is a personal decision (my body, my choice), whereas the right-wing, religious contingent would certainly have a different view, and I can see the Catholic Church opining on this from the perspective of the right.

I think this is a huge, loaded issue that could unfold along somewhat predictable lines.

Prof. Greely: I think we have been overlooking an interesting precedent. For more than 25 years, people have been doing pre-implantation genetic diagnosis (PGD). When you do IVF, you can take one or a few cells from a three-day or five-day embryo, test the genetic variations of those cells, and use that information to guide a decision about what embryos to transfer. That is embryo selection rather than embryo editing. You are not going to have an embryo with some combination of genes that the parents involved did not have to give. For 25 years, embryo selection allowed parents to make decisions about which embryos to pick based on disease traits, on the child's sex, and on immune system characteristics, in terms of the child being a potential cord blood donor for a relative, for example.

About 3,000 kids last year were born in the United States after PGD. It is uncommon, but it is not rare, and I think it raises many of the same issues as gene editing.

The interesting thing about it is that there is no legislation in the United States that regulates what traits can and cannot be selected using PGD. Ultimately, if either gene editing or widespread PGD were to become common—which I talk about in a new book entitled The End of Sex—then I think there would be legislative activity. It would probably be largely a political issue, though there are constitutional arguments that might be made in one direction or the other. Predicting what the Supreme Court will do next year, let alone in 20 years, is a real crapshoot.

Mr. Killian: Professor Greely, I agree with you that trying to predict what direction the Court might take is certainly difficult at this juncture. I do think, however, that although PGD does exist, the public reaction to the idea of editing will be significantly different and more vociferous from some groups than it is to the process of embryo selection, simply because people will view it as being a different type of process. And though this is related, I believe it will trigger a debate of a different scope and intensity.

Prof. Greely: I agree that there will be some increase of the discussion, but I think that when the general population becomes more aware that PGD exists—when it becomes common enough so that it impinges on the public consciousness—we will begin to have that debate.

I give talks on these topics quite often, and I typically find that only 5% of my audience has ever heard of PGD, and these are people who are interested enough to come to a talk about reproductive technologies.

Dr. Ishii: Professor Greely, do you see any difference between PGD for infertility treatment and PGD for disease prevention?

Prof. Greely: Yes, I do.

Dr. Ishii: PGD can be used as pre-implantation genetic screening (PGS) for infertility treatment, and it can also be used for disease prevention in offspring. Are there different legal implications for the parental rights to the embryo?

Prof. Greely: I am not sure if there are legal differences at this point related to the use of pre-implantation testing just to screen for which embryos are most likely to be successful or to prevent the selection and birth of an embryo with a serious genetic disease, or to pick one with non-disease-related traits that you prefer—or, for that matter, if you want to select for hair color or sex. I do not think there are currently legal differences. Those will, though, undoubtedly raise different concerns in the population and in the legislature, and are likely to be regulated differently, at least in some states and countries.

Avoiding serious health issues is the one selection criterion that is most likely to be allowed. Cosmetic selection is least likely to be allowed, and sex selection is somewhere in between, I suspect.

Dr. Church: One scenario in which people might prefer editing to IVF or PGD is in cases where they feel that the embryo has a right to life. With abortion, you lose about 25% of the embryos when you have two recessive carrier parents. With PGD, you lose more like 80%, but with sperm editing you do not put any embryos at risk. On that count, I think there might actually be less friction for some people.

There also could be lower risks for all people independent of their philosophy, in the sense that with a sperm cell precursor stem cell, you can get a clonal analysis. You can sequence the DNA of the cells in advance, before you implant them into the father, and you can ensure that they are not mosaic but instead are homogenous and are “correct,” which might mean having some common allele to replace an allele associated with a disease. Essentially, you are trying to fix the germline of one of the parents by eliminating a recessive, deleterious DNA variant.

Dr. Miller: Dr. Church, I am having trouble wrapping my head around manipulation of the gametes, especially if it would involve changing a gene that is on the X chromosome. Could you comment on that in light of X-inactivation in females?

Dr. Church: If you engineer the male sperm progenitor cells—let us say that they have a gene for a highly deleterious X-linked disease—then they will produce sperm that contain either a Y chromosome or an X chromosome as usual, but the X chromosome will no longer have the deleterious allele. It is as if they are no longer a carrier of the disease or, in the case of males, affected. Treated males can have healthy children of both male and female type.

Dr. Miller: But how do you deal with that if it is the female that is the carrier for a male-related X-linked disease?

Dr. Church: In that scenario, there is an equivalent procedure for manipulating oocyte precursor stem cells, and you would engineer the X chromosome genetically so that both chromosomes are normal. It is the same as if it were a non-sex—an autosomal—chromosome.

Dr. Miller: Would this not require IVF and not be able to be done in vivo?

Dr. Church: No, not at all. If you change the precursor cells in either the mother or the father, then those precursors would produce normal gametes, and sexual reproduction and everything else would be completely normal from that point on. There would be no need for IVF. This is a forward-looking statement, by the way, based on things that have been done in animals and assuming that they can be translated to humans.

Prof. Greely: However, you have just shifted the site of the mosaicism problem from the embryo to the testes or the ovary, have you not?

Dr. Church: With stem cells, we have published an example in which you can get a clonal stem cell. ¹ You can take a few cells from that clone, sequence them, and then pick the clones that are perfectly edited, with no off-target effects. Those are fairly common in even a small set of stem-cell clones, and then you implant one that has checked out during such quality control.

Prof. Greely: You would wipe out all of the naturally occurring sperm progenitor cells in the male's testes?

Dr. Church: You would displace them. Correct, you would have to.

Prof. Greely: That might not be a very popular move.

Dr. Church: Why not?

Prof. Greely: I personally am past the age for it, but if I were not, then I would not be excited about a treatment that involved taking a few of my sperm progenitor cells and then wiping out all the ones that are in my testes and putting back the manipulated ones to repopulate it. I think that is going to pose a marketing problem.

Dr. Church: First of all, it probably would be marketed to men who are infertile, so that would be a distinct advantage for them. For men who have a 25% chance of having a baby with Tay-Sachs or some other inherited disease, they would have a choice between having that procedure, which would have already been vetted in infertile men, or abortion, or PGD. I think some men will chose one option and some will choose another. I agree, though, it will not be for everyone.

Prof. Greely: My guess is that there will be a lot more PGD than there will be testicular eradication.

Dr. Church: I do not know, but even if it is done in a small subset of people, it would still be a meaningful medical advance. Consider that roughly half the U.S. population believes in right-to-life or some similar sentiment.

Prof. Greely: Yes, but the overlap representing people who are right-to-life, and so would not do PGD but are likely to have no religious objections to genome editing, I think is going to be a small set. It is an empirical question. I have not seen any survey data on it.

Dr. Church: I agree.

Dr. Miller: I am still having difficulty with the idea of modifying eggs and whether the genetic modification would affect the ability of the egg to mature and be released to be fertilized.

Dr. Church: You are modifying stem cells. It has been done in other animals. It is just a matter of determining whether it is suitable for clinical use.

Dr. Miller: Are the viability of the embryos and fetuses comparable? Is the ability of the eggs derived from manipulated stem cells to be fertilized comparable to that of non-manipulated eggs?

Dr. Church: It is too early to say.

Prof. Greely: The best work I know of is from Dr. Mitinori Saitou at Kyoto University, Japan, who is working with mice. He derived both oocytes and sperm cells from both mouse embryonic stem cells and mouse-induced pluripotent stem cells, and he used each to lead to the birth of baby mice.

They were not always healthy mice—too large a percentage were unhealthy. This is clearly not something it would be responsible to try in humans at this point, but if you have a long enough lead time, I think the idea of deriving oocytes or oocyte progenitor cells from either human embryonic stem cells or human-induced pluripotent stem cells is quite plausible. However, safety tests would have to be rigorous, extensive, and years in the works.

Mr. Moussa: I think it was Professor Greely who said that the sort of conversation this would spark would mirror the abortion debate and that it would be naive to think otherwise. The Catholic Church, for example, strongly objects to IVF, right? It would likely view gene editing as even more of an intervention.

Prof. Greely: Right, and the Catholic Church's objection to IVF has had what effect on the use of IVF in the United States? I would say little to zero. And what effect on regulation of IVF in the United States? Absolutely zero.

Mr. Moussa: I think it informs a lot of people on that side of the abortion debate. I understand that, for example, religious bans on birth control are widely ignored by U.S. Catholics, right?

Prof. Greely: As are restrictions on IVF. Right-to-life people like healthy babies, and as long as the focus is on healthy babies, I think we will continue to see the almost complete absence of significant regulation of assisted reproduction in the United States.

Gene editing will tempt regulation more, and what the endpoint turns out to be is not clear, but we are in a strange country that has a very strong right-to-life movement and an almost completely unregulated assisted reproduction industry.

Dr. Legato: What is the history of lawsuits related to assisted reproduction? Does it ever go wrong, or do parents ever perceive that it has gone wrong?

Prof. Greely: Sure it goes wrong from time to time. I do not know of any suits that have gone to trial or have made it to the appellate level, which is what lawyers mainly see. Ninety percent of the law is invisible because it never gets reported if cases do not get appealed.

But I have heard discussions of cases in which PGD has turned out to be wrong, and parents have sued or threatened suit and settlements were reached. Whether the PGD was wrong because of a laboratory error or because of mosaicism is not entirely clear.

If you can show that the activity was done in a negligent way and that it has imposed significant extra costs, you are likely to be able in almost every state to win what would be called a wrongful birth suit. In a few states, including California and New Jersey, the child can also sue under what is called a wrongful life suit. In most states, however, a wrongful life suit by a child does not apply. But the parents, if faced with extra costs to care for an unhealthy child as a result of the negligence of the provider, can almost always get damages for what is called wrongful life. A few states have banned that, but the vast majority of states allow wrongful life suits.

Dr. Legato: Are they common?

Prof. Greely: I do not know the best answer. The law is not very good at keeping track of how many suits are filed in different areas. It is much better at keeping track of how many appellate cases are reported. There are appellate cases in almost every state, but there are not many of them. I suspect that they are not common, but also not rare.

Dr. Miller: If a person is the product of IVF, how is that recorded in one's medical record—your personal medical record?

Dr. Church: I do not know the answer to that.

Prof. Greely: I do not know either, but it is one of the frustrations with which we must currently deal. Louise Joy Brown's birthday was in July, and that marked 38 years of IVF. We do not have the detailed type of follow-up on IVF babies that one would really like, in part because they are not research projects. They are babies, and they did not consent to be research subjects.

Dr. Miller: That is fascinating.

Prof. Greely: But it would not surprise me if we do not know if it is not noted on their charts. But I do not know the answer to that question.

Dr. Miller: Would genetic engineering of the gametes, or whatever procedure might be done, become part of an individual's medical record? While these individuals, when they are born, would not be considered research subjects, they become part of an epidemiological record that could be used voluntarily for research studies on the incidence and prevalence of disease just by counting hospital visits, for example. These individuals would be identified because they are part of a target population.

I do not anticipate that these types of manipulations are going to be so widespread that they are going to affect population-based expression of the incidence or prevalence of disease, but I cannot say for sure. I suppose they would reduce disease occurrence, but how would information be generated to know what factors would contribute to those reductions, whether they were environmental or directly attributable to the gene manipulation?

Dr. Church: For diseases such as Tay-Sachs, we know that within populations that practice genetic counseling, the rate of homozygotes being born is reduced by at least a factor of 10.

Dr. Miller: But that is due to genetic counseling and perhaps choices not to have children.

Dr. Church: That is correct, but the point is that it is tracked. Furthermore, the cause of the reduction is not environmental.

Dr. Miller: Yes, I see, but by environmental, I meant in terms of the geographic incidence.

Prof. Greely: I do think that if we ever get to genome editing of children before birth, whether it is through editing gametes or editing embryos, which I agree with Dr. Church seems more likely, it would be really, really useful to have a registry and to have lengthy follow-up.

This does raise some tricky research-related ethics issues because, after all, the child has not consented, and to provide a clinical service with the condition that the patient take part in subsequent follow-up research is questionable. It would certainly be great, though, to have long follow-up of everyone who is born following genome editing, just as it would be to have that follow-up for PGD or IVF.

Dr. Legato: Do the parents have a legal right to make that decision for the baby or the child?

Dr. Church: That is very common in medical research (and non-research clinical practice) when the children are below the age of consent. It is a commonly accepted Institutional Review Board (IRB) practice.

Prof. Greely: Although the age of consent is 18, typically at some stage when the kids are teenagers or so, you begin to give them a veto as well. The answer to your question, though, is yes, the parents could consent to that. The question is what happens if the parents do not consent to it and you would still like to get that information. Plus there may be some confounding variables that differ between parents who consent and parents who do not.

Dr. Legato: That is very interesting. What about making it a condition of the manipulation that the parents agree to have long-term follow-up for the sake of the careful regulation and assessment of the intervention?

Prof. Greely: As long as it is an experimental provision, then I think you could do that, although remember that, in general, human subjects have a right to change their mind and withdraw from any research. The parents could say it in order to get the procedure done and then after the baby was born, presumably, they could withdraw.

Even trickier would be if, in the future, genome editing were to become a general clinical procedure. I think it would be difficult with respect to the research ethics and common rule to make access to an FDA-approved clinical procedure depend on one's willingness to take part afterward in a research project, particularly when you are not volunteering yourself, but instead are volunteering an as-yet unborn child.

Mr. Killian: Yes, I agree, and I think that this would have to come as almost a concerted effort to develop professional best practice. It is something that should come from practitioners and be strongly promoted as beneficial to the individual and to the emerging field for patients as a whole.

I do not know that legally you would find much precedent or support for conditioning a clinical option on irrevocable—even if you could make it irrevocable—consent to have your child followed until he or she could attain sufficient wisdom to opt out on his or her own.

Prof. Greely: I suppose a state legislature or Congress could pass a law to that effect. A common rule is a regulation, not a statute or a constitutional provision. Such a statute might be open to some constitutional attack, although it is not obvious to me which way that suit would come out. The legislature might be able to do it, but I think that is probably unlikely.

Dr. Legato: Dr. Miller's question about whether the way these children are formed—IVF and so on—is ever recorded in the medical record is a very interesting one.

Mr. Killian: Speaking from my understanding of general practices, it would only be recorded insofar as it is part of a comprehensive history that is given by one or both parents at some point to the child's pediatrician. It may or may not be asked as a standard question.

Dr. Legato: Dr. Miller, if the information were recorded, could those children then be subjected to epidemiologic research?

Dr. Miller: If the data were linked to the medical record, then that kind of information would be gathered. If it were just a questionnaire, for example, without being linked to the medical record, then there would be no way to know.

I think that for at least some studies now, with the electronic medical record, once you consent to be a part of an epidemiological study, consent includes access to your medical record, and that kind of information could be gathered.

I am not sure how the Centers for Disease Control and Prevention (CDC) collects some of their data. Some of it is through self-report questionnaires, and there is always a question of how accurate those self-report measures are.

Dr. Legato: Dr. Ishii, would you like to comment on the attempts to regulate this whole area in the countries you surveyed?

Dr. Ishii: Yes, I can discuss that. There is a problem in assisted reproduction. Obstetricians are involved in the reproductive process, but even if a problem occurs, nobody follows up. I think the responsibility will be on the obstetricians as well as pediatricians. There is a big clinical gap between genome editing prior to reproduction and seeking healthy babies that are already here.

We are talking about genome editing of the human embryo. There are many purposes. Is it for infertility treatment or disease prevention? The situation depends on the parents' purpose because parents consent to germline genome editing.

Dr. Legato: Follow-up seems to be a tremendously important issue in all of this. Does everyone agree?

Dr. Miller: I think so. Looking long term in animals is one thing. Looking in humans is another, as it goes on for multiple years, with comorbidities that can also have an effect.

Dr. Legato: Environmental impact has to be considered as well.

Dr. Miller: Yes. Would there be any social pressure or discrimination toward individuals that have undergone genetic engineering if it became known?

Mr. Moussa: Possibly, and especially if the treatment ends up being beyond some people's financial means. I think it would be frustrating if people felt that others could buy themselves an advantage. That becomes a psychological issue. Is that what you mean?

Dr. Miller: That is an interesting part of it, but that was not exactly what I meant. If it became known that a person had some type of genetic manipulation for other than a medical purpose, would that person be under some discriminatory pressure in terms of how they are viewed by society as a whole?

Dr. Legato: Given our experience with organic food and genetically altered crop plants and other substances, I think there would be pressure. Do you not agree?

Dr. Church: I think there are exceptions for life and death. For example, despite problems with genetically modified foods, there is almost zero resistance to genetically modified organisms producing human insulin for diabetics. These patients might be treated with sympathy or they might be treated as being among an elite group. That is to say they, like many wealthy people, would be recognized as having the advantages of wealth.

I am not saying that is a good thing. I am just saying that they may not be discriminated against. They may be favored. In the movie Gattaca, the people who had been genetically chosen or influenced were considered elite. They were not discriminated against; instead, they were favored. That is fiction, of course, but that is an opinion.

Mr. Moussa: We are at an interesting moment in history right now. People are increasingly becoming aware of income disparity, and this is an important issue at the moment. Where that goes in 5 or 10 years, or beyond, who knows? Where we are at this moment, though, might be viewed as the culmination of unfair income equality: not only does this guy have a lot more money than I do and a lot more comforts and luxury, but in addition, he is healthier for no good reason. I see the potential for it to breed a lot of resentment, but I think we are in a peculiar kind of political moment right now. It could go several different ways.

Prof. Greely: This goes back to the earlier discussion of medical records. This could be one reason that some people would be very interested in either keeping those records confidential or not noting the gene editing on them at all. One way to avoid discrimination is to “pass.”

Dr. Legato: I think that is very relevant.

Mr. Killian: I would also add that as long as we are on a not quite universal healthcare model, consider the interest that insurers would take in those who had been genetically assured of not having certain conditions, insofar as it would be possible to do that. Is that another economic benefit that might be visited on some, whereas others would not have that opportunity?

Dr. Legato: Dr. Church, do you think there will be a category of people that are precluded from having certain vulnerabilities to certain diseases?

Dr. Church: We already have adult gene therapies aimed at eliminating diseases, such as making people resistant to HIV infections, so that possibility is already in motion. Whether that would be extended to infants, it is not clear whether there would be motivation for that. But there might be motivation for something else.

Dr. Legato: Do Professor Greely, Mr. Killian, or Mr. Moussa have any thoughts about insurance consequences?

Prof. Greely: The health insurance consequences should be limited, even if the United States does not move beyond its current level with respect to providing healthcare. Genetic risks are not allowed to be viewed as preexisting conditions. That actually goes back as far as 1996 in the U.S. Health Insurance Portability and Accountability Act legislation. Under the 2010 Affordable Care Act (“Obamacare”), which may or may not exist in a year, even individually underwritten policies are not allowed to medically underwrite. With existing law, therefore, it would be difficult for a health insurer to use this against or in favor of someone.

Remember, using it in favor of one group is effectively using it against another group. The most relevant federal law, the Genetic Information Nondiscrimination Act (GINA), does not cover life insurance, disability insurance, and long-term care insurance, although some state laws cover them.

Dr. Legato: This has been a very valuable and interesting conversation, and I would like to thank you all for participating.