Bioethics Forum Essay

Why Human Germline Editing Might Never Be Legal in the U.S.

A Russian scientist recently announced his intention to use the gene-editing tool CRISPR to edit and implant human embryos—a revelation that met with international outcry similar to the condemnation of the Chinese scientist He Jiankui last year when he announced that he had created the first gene-edited babies. Jiankui’s actions — deemed unethical for several reasons – led to a call for a moratorium on editing human germline cells (sperm, eggs, or embryos) to produce genetically modified babies.

Even the signatories of the moratorium, however, suggest that CRISPR might one day be safe enough for ethical, clinical use on the germline. What would it take for the first case of germline editing to proceed under applicable U.S. law and ethical standards?

Germline editing would be regulated as a gene therapy by the Food and Drug Administration. To comply with the relevant regulations, germline editing must undergo clinical testing to demonstrate safety and efficacy and win FDA approval before coming to market. Currently, federal law prevents the FDA from reviewing or approving any application involving manipulated human embryos. However, if and when this ban is lifted, the first case of germline editing would take place in the context of a clinical trial and therefore would be subject to the laws and ethical standards applicable to research.

Specifically, Subpart D of the Common Rule, which pertains to research involving children and is incorporated into FDA’s regulations, would probably apply to germline editing research. Under this provision, research that entails more than a minimal risk must provide a prospect of direct benefit to the child.

The Common Rule states that minimal risk means that “the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests”—a conservative standard that germline editing, for several reasons, is unlikely to meet.                                

First, CRISPR research has revealed relatively high frequencies of off-target edits (meaning that an untargeted gene is edited) and mosaicism (not all cells of an embryo are successfully edited) with the potential for significant deleterious effects. Risk of such inaccuracies cannot be completely ruled out in any embryo intended for implantation. Because a cell must be removed from an embryo in order to be sequenced, only a subset of an embryo’s cells can be tested for off-target edits and mosaicism. It is therefore practically impossible to confirm that a child produced from germline editing will be free from the more than minimal risks associated with these inaccuracies.

Furthermore, given the nascence of CRISPR gene-editing combined with our relative lack of understanding of the complexities of the human genome, there may be other, yet-undiscovered, greater-than-minimal risks associated with germline editing. It’s unclear how animal and lab testing could successfully identify all such risks before human trials are initiated.

Even assuming that all the technical risks potentially associated with germline editing can be resolved, the technology must be performed in conjunction with IVF. IVF, however, carries some risks, including multiple births and other complications. Even though IVF is routinely carried out in a clinical setting, as a procedure, IVF was not subject to FDA’s clinical research requirements and related legal and ethical standards. However, in the context of germline editing research, the risks associated with IVF may alone be enough to surpass the minimal risk threshold of Subpart D.

Because germline editing arguably involves more than a minimal risk, any related research must provide the prospect of direct benefit to the child-to-be. If gene-editing were applied to an existing child with a severe genetic disease, such as Tay Sachs, then such research would provide the possibility of direct benefit to the child, such as relief from suffering associated with the disease.

With germline editing, however, no sick child exists. Instead, there are embryos of a certain genetic makeup that if implanted would cause the existence of a child with a genetic disease. No such child needs to exist, however, given the other options available to couples affected by genetic disease.

For example, the vast majority of couples affected by genetic disease can produce and selectively implant some amount of healthy embryos using IVF and preimplantation genetic diagnosis. The relatively few couples who cannot produce healthy embryos, however, still have other options. Specifically, they can 1) use gamete donation to create a healthy child that is their partial genetic relation; 2) adopt a child; 3) or choose not to become parents. The unique benefit of germline editing over these existing options is to provide such a couple with the ability to have a genetically related child they likely otherwise would not have had. This benefit accrues to the parents, not the child-to-be.

The more than minimal risk that germline editing presents to the child-to-be is not outweighed by a direct benefit as required by relevant law. Therefore, even putting aside the many other ethical concerns associated with germline editing, it’s unclear that it could proceed in the U.S. under current law—a fact conspicuously absent from the CRISPR debates.

Jennifer M. Gumer, JD, is an attorney and adjunct professor of bioethics at Columbia University and Loyola Marymount University. She was a visiting scholar at The Hastings Center in July.