As discussed in an exchange earlier this year on this blog, one of the concerns about editing the human germline is that the risks to the next generation and future generations are not predictable, and the experiments to address those safety concerns cannot be done ethically. Go here, and to the embedded links, to review. Recently, Paige Cunningham and Michael Sleasman of the Center for Bioethics and Human Dignity (CBHD) weighed in on the blog at desiringgod.org.
At the session on CRISPR at the CBHD summer conference this past June 18-20, the various arguments against human germline editing were rehearsed and debated. They were expressed similarly to the discussions on this blog and the desiringgod.org blog. But the “safety argument” was challenged by no less an authority than plenary speaker Maureen Condic, Ph.D., Associate Professor of Neurobiology and Anatomy at the University of Utah. Her rejoinder: in a carefully circumscribed case, the safety argument against human germline editing likely fails. That is, an exception that could be plausibly argued to have an acceptable risk-benefit ratio for human treatment could be found.
The criteria for such an exception: point gene mutation, gene function clearly understood, expression in only one tissue, clear ability to assess whether the target gene, and only the target gene, had been edited. And an example? Sickle cell anemia. Point mutation in the hemoglobin gene. Expression only in red blood cell precursors. Known gene function, easily monitored. The “low hanging fruit” of human germline editing.
I still resist this argument. If one envisions editing the hemoglobin gene in an intact human embryo, then I don’t see how the sufficient control of the process could be demonstrated without subjecting at least one—and likely more than one—human embryo to risks that include severe genetic harm and, frankly, death or oblivion, at least in the form of never being implanted. Even if the process were repeated many, many times in non-human species, it would still arguably be a huge step of faith that results would be the same in man. If editing germ cells (sperm or egg) prior to fertilization is in view, the efficiency and selectivity of the editing process could be confirmed in the cells, but it is not at all clear to me that the stability and selectivity of the edit would be maintained through the process of fertilization, much less through subsequent generations. One step to confirm potential safety would be to create human embryos—a lot of them—to get good data. If a proposal like this went through an IRB I sat on, I would vote “no.”
(I assume that FDA would assert regulatory jurisdiction and require that initial efforts be classified as human subject research. But perhaps not, because human subject research protections currently do not include embryos. Dr. Condic suggested that convincing FDA to change that would be a good project for an enterprising bioethicist.)
Counterarguments to my resistance include: individual parents could choose to accept the risk on behalf of their offspring; preimplantation genetic diagnosis (PGD) could be used to ensure that genetically damaged embryos are not brought to term. That, of course, assumes that one accepts PGD as ethical.
So I worry that the “safety argument,” which is also the most readily accepted basis for public policy arguments, will be heard but ultimately rejected. Eventually, someone will try this. But we should not accept human germline editing, for a lot of reasons reviewed on this blog and by Cunningham and Sleasman in their post.