Over the last few decades, the international community has issued several bioethical guidelines and legally binding documents, ranging from UN Declarations to regional charters to national legislation, about editing the human germline--the DNA that is passed down to future generations. There was a broad consensus that modifications should be prohibited. But now that CRISPR-cas9 and related methods of gene editing are taking the world by storm, that stance is softening--and so far, no thorough public discussion has emerged.
There is broad agreement in the scientific and ethics community that germline gene editing must not be clinically applied unless safety concerns are resolved. Predicting that safety issues will indeed be minimized, the National Academy of Sciences issued a report this past February that sets up several procedural norms. These may serve as guidelines for future implementation of human embryo editing, among them that there are no "reasonable alternatives," a condition that is left deliberately vague.
I regard the conditional embrace of germline gene editing as a grave mistake: It is a dramatic break with the previous idea of a ban, departing also from the moratorium that the UNESCO International Bioethics Committee had recommended in 2015. But in a startling move, the Academy already set the next post, recommending "that genome editing for purposes other than treatment or prevention of disease and disability should not proceed at this time" (my emphasis). It recommended public discussions, but without spelling out its own role in facilitating them.
To proceed ethically, I argue that the international community, through the United Nations and in line with the ban on human reproductive cloning, should explicitly ban embryo gene editing as a method of human reproduction. Together with guidelines adjusted for non-reproductive and non-human applications, a prohibition would ensure two important results: First, that non-reproductive human embryo research could be pursued in a responsible way in those countries that allow for it, and second, that individual scientists, public research institutes, and private companies would know the moral limit of possible research.
Basic human embryo research is required, scientists argue, to better understand genetic diseases and early human development. I do not question this, and I am convinced that existing guidelines can be adjusted to meet the moral requirements in this area. Millions of people may benefit from different non-reproductive pathways of gene editing. Germline gene editing, in contrast, does not offer any resolutions to global or local health problems – and that alone raises many concerns about the current state of scientific research.
I support a ban because germline gene editing for reproductive purposes concerns more than safety. The genetic modification of a human being is irreversible and unpredictable in its epigenetic, personal, and social effects. It concerns the rights of children; it exposes persons with disabilities to social stigmatization; it contradicts the global justice agenda with respect to healthcare; and it infringes upon the rights to freedom and well-being of future persons.
Apart from questions of justice, reproductive germline gene editing may well increase the stigmatization of persons with disabilities. I want to emphasize here, however, that it directly violates the rights of individual future persons, namely a future child's right to genetic integrity, to freedom, and potentially to well-being, all guaranteed in different UN Declarations of Human Rights. For all these reasons, it is an unacceptable path forward.
The way the discussion has been framed so far is very different from my perspective that situates germline gene editing in the broader framework of human rights and responsibilities. In short, many others never questioned the goal but instead focused on the unintentional side-effects of an otherwise beneficial technique for human reproduction. Some scientists see germline gene editing as an alternative to embryo selection via Preimplantation Genetic Diagnosis (PGD), a procedure in which multiple embryos are tested to find out which ones carry disease-causing mutations. Others see it as the first step to human enhancement.
Some physicians argue that in the field of assisted reproduction, not every couple is comfortable with embryo selection via PGD, because potentially, unchosen embryos are discarded. Germline gene editing offers them an alternative. It is rarely mentioned, however, that germline gene editing would most likely still require PGD as a control of the procedure (though without the purpose of selection), and that prenatal genetic diagnosis would also be highly recommended. In other words, germline gene editing would not replace existing protocols but rather change their purpose, and it would also not necessarily reduce the number of embryos needed for assisted reproduction.
In some (rare) cases, PGD is not an option, because in the couples' condition, all embryos will be affected. One current option to avoid transmitting genetic traits is to use a donor sperm or egg, though the resulting child would not be genetically related to one parent. If these parents had an obligation, as some proponents argue, to secure the health of their offspring (an argument that I do not follow), then procreation with sperm or egg donation would even be morally required, as this is the safest procedure to erase a given genetic trait.
There are no therapeutic scenarios that exclusively require reproductive gene editing even if one accepts the right to reproductive autonomy. The fact is that couples who rightly wish to secure and protect the health of their future children can be offered medical alternatives in all cases. However, this requires considering sperm or egg donation as the safest and most reasonable option – the condition the NAS Report has set.
Scientists in favor of germline gene editing argue against this: the desire for genetic kinship, they say, is a legitimate expression of a couple's reproductive freedom, and germline gene editing offers them an alternative to have a healthy child. In the future, proponents say, these (very few) couples who wish for genetically related offspring will be faced with the dilemma of either accepting the transmission of a genetic health risk to their children or weighing the benefits and risks of gene editing.
But here is a blind spot in the whole discussion.
Many scientists and some bioethicists think that reproductive freedom includes the right to a genetically related child. But even if we were to presuppose such a right, it is not absolute in the context of assisted reproduction. Although sperm or egg donation may be undesirable for some couples, the moral question of responsibility does not disappear with their reproductive rights. At a minimum, the future child's rights must be considered, and these rights go further than their health rights.
It is puzzling that in claiming their own reproductive freedom, couples would need to ignore their children's and possibly grandchildren's future freedom – including the constraints resulting from being monitored over the course of their lives and the indirect constraints of the children's own right to reproductive freedom. From a medical standpoint, it would be highly recommended for them, too, to have children through assisted reproduction. This distinguishes germline gene editing from any other procedure of assisted reproduction: we need the data from the second and third generations to see whether the method is safe and efficacious. Whose reproductive freedom should count, the parents' or the future children's?
But for now, the question of parental rights may well divert the discussion from the question of responsible gene editing research; its conditions and structures require urgent evaluation and adjustment to guide international research groups. I am concerned that we are in the process of developing a new technology that has tremendous potential and ramifications – but without having considered the ethical framework for a responsible path forward.
Editor's Note: Check out the viewpoints expressing enthusiastic support and mild curiosity.
Scientists have been using MRI images (like the ones pictured here) to train artificial intelligence to detect cancers earlier and with more accuracy. Here, MIT's AI system, MIRAI, looks for patterns in a patient's mammograms to detect breast cancer earlier than ever before. news.mit.edu