Should we use CRISPR gene editing in human embryos?

Scientific enquiry is typically accompanied by a kaleidoscope of differing opinions, with champions of competing hypotheses seeking the final data that will prove the validity of their preferred ideas and the folly of those holding conflicting notions. Debate is not only healthy, but essential for the vitality of science and for the rigor with which new concepts are evaluated. In no field is this truer than in reproductive science. Consider the many differences of opinion with respect to scientific theory and clinical practice that sometimes divide us from our dear colleagues. So numerous are they in our discipline, that the journal in which we are writing today has little difficulty dedicating a section to a different “Fertile Battle” every other month. In this context, perhaps it seems strange that the scientific community has found an unusual degree of unanimity with respect to the question of germline genome editing. This was evidenced by the near universal dismay, disapproval, and disbelief that met the 2018 announcement of twin girls born in China after the treatment of in vitro fertilization (IVF) embryos using the genome editing technology known as CRISPR. If the scientist who led the work initially failed to appreciate the distinction between fame and infamy, he probably fully comprehends it today. The news of the “CRISPR births” stunned the scientific community and sent out societal shockwaves that continue to reverberate today. Presenting a heretical viewpoint can be an entertaining philosophical exercise for scientists, but with a subject as important and controversial as germline genome editing it is important to be clear that we, the investigators of this article, do not dissent from majority view. We fully agree that the clinical application of genome editing technologies to IVF embryos is premature given that significant questions concerning safety and efficacy remain to be answered and ethical considerations have not been fully resolved. The statement above should bring this debate to a rather early conclusion, leaving us an embarrassing 1000 words short of fulfilling our obligation to the journal. Luckily, however, there is more to be said. Adding IVF to genetics unites 2 topics that have long been favored themes in certain sections of the media. Think of “three-parent-embryos,” “designer babies,” “savior siblings,” and “slippery slopes.” It would be flippant to dismiss the genuine potential for technologies to be abused, and the legitimate ethical concerns that some may have. Nonetheless, it is not difficult to present a positive view of the possibilities that genome editing methods may offer us in the future. We will attempt to outline just a few of these below. It is sometimes argued that genome editing is “a step too far,” a radical intervention, which is unnecessary given that there are well-established alternatives for avoiding inherited disease transmission. It is true that prenatal diagnostics, using techniques such as amniocentesis and chorionic villus sampling to collect fetal DNA, are available for most common inherited conditions. Indeed, prenatal diagnosis has been used for some monogenic conditions since the late 1980s. However, the major limitation of this approach is that prevention of disease is only achieved by terminating the pregnancies that prenatal testing indicates to be affected. This strategy carries its own ethical questions and is also problematic from the perspective of some religious teachings. It is therefore unsurprising that prenatal diagnosis is unacceptable to some patients. Over 30 years ago, preimplantation genetic testing (PGT) was introduced as an alternative to prenatal testing, with the advantage that the vast majority of pregnancy terminations are avoided. This is achieved by sampling one or more cells from cultured preimplantation embryos, produced using IVF technology. The genetic material of the cell(s) is tested to establish the disease status of the embryo and only those found to be unaffected are considered for transfer to the uterus. Thus, any pregnancy established should be free of the familial disorder. Although many patients consider diagnosis at the preimplantation stage to be preferable to prenatal diagnosis, the reality is that PGT cannot help all patients. If the embryo is viewed as having an equal status to the fetus, which could be considered the case if following strict Catholic doctrine, then the destruction of several viable (but affected) embryos in a PGT cycle might not be deemed an improvement on the termination of a single affected pregnancy after prenatal testing. For this reason, some patients are unable to accept PGT. Another important limitation of PGT is that as many as 10% of cycles do not have any embryos that are both unaffected and suitable for transfer (1Gutiérrez-Mateo C. Sánchez-García J.F. Fischer J. Tormasi S. Cohen J. Munné S. et al.Preimplantation genetic diagnosis of single-gene disorders: experience with more than 200 cycles conducted by a reference laboratory in the United States.Fertil Steril. 2009; 92: 1544-1556Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). Of course, it is not uncommon for an IVF cycle to yield only 1 or 2 competent embryos, and if PGT shows those embryos to be affected, then the cycle cannot produce an unaffected child. Thus, a patient may endure the substantial costs (financial, physical, and emotional) of an IVF cycle with PGT, yet still be denied the outcome they desire. Genome editing has been successfully applied to human embryos in a research context, allowing the precise targeting and disruption of individual genes (2Fogarty N.M.E. McCarthy A. Snijders K.E. Powell B.E. Kubikova N. Blakeley P. et al.Genome editing reveals a role for OCT4 in human embryogenesis.Nature. 2017; 550: 67-73Crossref PubMed Scopus (252) Google Scholar). Unquestionably, this sort of approach provides us with powerful tools for increasing our understanding of early development. However, in a debate such as this, the inevitable focus is on the question of whether it is desirable to use genome editing technology clinically, removing mutations responsible for serious inherited disorders. It could be argued that the successful application of genome editing would “rescue” affected conceptions. Instead of being discarded at the preimplantation stage after PGT, or terminated at fetal stages after prenatal testing, their inherited mutation would be removed, the DNA restored to the same “wild-type” sequence carried by the vast majority of the global human population. From a PGT perspective, this would result in all embryos becoming eligible for transfer, and no human conception would have to be discarded. This would seem to circumvent some of the ethical quandaries associated with PGT and prenatal testing. Most of us can probably agree that it is preferable to avoid being affected by a genetic disorder, at least one that significantly impacts the length or quality of life, but why complicate things by adding IVF into the equation? Why not simply treat affected individuals after they are born? Unfortunately, many the >10,000 monogenic disorders caused by inherited mutations disrupt developmental processes and would already have had an irreversible impact within days or weeks of conception. In such cases, removal of mutations would need to be undertaken at a very early stage. An additional consideration is that heritable disorders often exert their effects on the phenotype by altering the function of millions of cells, contributing to multiple tissues and organs. This leaves scientists with the problem of how to deliver the components required for genome editing into huge numbers of cells located at numerous sites, deep inside the body. Somatic gene therapy is limited: often too late, with too many cells to correct or reach in vivo, and consequently with limited capacity to prevent the broad spectrum of inherited diseases that afflict children and adults. Theoretically, the application of genome editing to IVF embryos has several important advantages. Removal of a mutation during the preimplantation phase is likely to be sufficiently early in development that the disease phenotype is entirely avoided. Indeed, if treatment was undertaken before the 4-cell stage (when the major wave of embryonic gene expression begins), it is likely that the affected gene would not be transcribed until after the wild-type sequence had been restored. Genome editing at the preimplantation stage also has the advantage that the number of cells requiring targeting is very small. For example, at the zygote stage, delivery of genome editing reagents can be virtually guaranteed by microinjection into the single cell. For later preimplantation stages, microinjection may become impractical, but until around day 3, all blastomeres have part of their surface in contact with the external environment. This gives a high likelihood that genome editing components could be successfully introduced using various methods of transfection. Naturally, if genome editing is successfully accomplished in all cells of a preimplantation embryo, the phenotype associated with the inherited mutation should not occur. At first consideration, this seems the perfect outcome. However, if all cells are edited, this will include those destined to give rise to the germ cells. So, any alterations to the genome could be passed to subsequent generations, potentially remaining in the human gene pool forever. For some, the possibility of a heritable change to the human genome brings a sense of unease, but for others, the idea of permanently removing a mutation that may have brought misery to a family for generations seems less problematic, and maybe even desirable. Interestingly, a positive view of germline genome editing is often held by those who are perhaps best placed to consider its appropriateness, namely those who have experienced the devastation wrought by inherited disease within their own families. In recent years, distinguished organizations have carefully weighed the merits of heritable genome editing and the concerns. The National Academy of Sciences and National Academy of Medicine acknowledged that such an approach has potential to alleviate the suffering caused by inherited diseases, stating “There are circumstances in which genome editing in germline cells or embryos might be the only or most acceptable option for prospective parents who wish to have a genetically related child while minimizing the risk of transmitting a serious disease or disability” (3National Academy of Sciences, National Academy of MedicineHuman genome editing: science, ethics and governance. The National Academies Press, Washington, DC2017Google Scholar). Similarly, in the United Kingdom, the well-respected Nuffield Council on Bioethics agreed that “the use of heritable genome editing to influence the characteristics of future generations could be ethically acceptable in some circumstances, if it is consistent with, the welfare of a person who may be born as a consequence of interventions and it does not increase disadvantage, discrimination, or division in society” (4Nuffield Council on BioethicsGenome editing and human reproduction: social and ethical issues.http://nuffieldbioethics.org/project/genome-editing-human-reproductionDate accessed: August 7, 2023Google Scholar). It is true that some of these caveats are not trivial, but nonetheless, these learned opinions represent a consistent view that there are circumstances in which genome editing (even that which affects the germline) can be considered beneficial and appropriate. There is evidence that the public shares this opinion; a survey conducted by the Royal Society in the United Kingdom reported that 83% of participants were supportive of germline genome editing to treat incurable disease, although many drew the line at editing for “enhancement” (e.g. 60% were opposed to heritable gene editing to improve intelligence) (https://royalsociety.org/topics-policy/projects/genetic-technologies/). If genome editing is ever to be applied to human embryos, safety concerns will have to be addressed. There is mounting evidence that the cells of early preimplantation embryos struggle to process the double-strand DNA breaks induced by CRISPR/Cas9 technology, leading to unresolved damage with potentially serious consequences for the embryo (5Alanis-Lobato G. Zohren J. McCarthy A. Fogarty N.M.E. Kubikova N. Hardman E. et al.Frequent loss of heterozygosity in CRISPR-Cas9-edited early human embryos.Proc Natl Acad Sci USA. 2021; 118e2004832117Crossref PubMed Scopus (89) Google Scholar, 6Kubikova N. Esbert M. Titus S. Coudereau C. Savash M. Fagan J. et al.Deficiency of DNA double-strand break repair in human preimplantation embryos revealed by CRISPR-Cas9.Hum Reprod. 2023; 38: i46Crossref PubMed Google Scholar). Nonetheless, genome editing technologies continue to evolve and more nuanced methods, such as base editing and base editing, which can be considered “gentler” on the DNA, might assuage some fears over safety. Base editing, which avoids the creation of double-strand DNA breaks, has been applied to human embryos in vitro in a research context with promising results (7Zeng Y. Li J. Li G. Huang S. Yu W. Zhang Y. et al.Correction of the Marfan syndrome pathogenic FBN1 mutation by base editing in human cells and heterozygous embryos.Mol Ther. 2018; 26: 2631-2637Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). Even if genome editing is eventually adjudged to be safe, legitimate concerns may remain over equity of access, and the societal impact of genome editing. However, it is worth noting that from a health economics perspective, the permanent removal of a mutation from a family is likely to be far less costly than a lifetime of medical care (and potentially other forms of support), not just for one individual, but also for all their future descendants. This is food for thought for countries with national healthcare systems. As with many scientific advances, there is a risk that genome editing technologies could be abused—employed to modify non-disease characteristics or for the introduction of “enhancements.” Although this possibility can never be entirely ruled out, is it a reason to deny families a method that could prevent the suffering, emotional trauma, and death that accompanies many inherited diseases? Perhaps the introduction of mitochondrial replacement therapies (MRTs) in the United Kingdom could be taken as a model for how genome editing technologies could be introduced in the future. MRTs were developed to prevent the transmission of incurable mitochondrial diseases from mother to child. They involve the transfer of the nuclear DNA from an oocyte or zygote, produced by a woman carrying a mitochondrial DNA mutation, into the cytoplasm of an enucleated donor egg. The defective mitochondria are left behind and the nuclear DNA is placed in a cell populated by functional mitochondria, thus avoiding the cause of the disorder (8Hyslop L.A. Blakeley P. Craven L. Richardson J. Fogarty N.M. Fragouli E. et al.Towards clinical application of pronuclear transfer to prevent mitochondrial DNA disease.Nature. 2016; 534: 383-386Crossref PubMed Scopus (223) Google Scholar, 9Costa-Borges N. Nikitos E. Späth K. Miguel-Escalada I. Ma H. Rink K. et al.First pilot study of maternal spindle transfer for the treatment of repeated in vitro fertilization failures in couples with idiopathic infertility.Fertil Steril. 2023; 119: 964-973Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar). This has some parallels with genome editing, because the mitochondrial DNA of the resulting embryo has a different DNA sequence to that of the original oocyte/zygote and this change can be inherited by future generations through the maternal line. Like genome editing, implementation of MRT raised profound questions of ethics and safety. In the United Kingdom, an extensive consultation process was launched, engaging experts, religious groups, patients, and the public to understand their opinions. New legislation was then debated in the UK Parliament, ultimately culminating in The Human Fertilisation and Embryology (Mitochondrial Donation) Regulations 2015. This created a framework for a rigorous licensing process, with independent oversight, ensuring that clinics meet strict scientific, ethical, and safety standards, that the process of obtaining patient consent is adequate, and guaranteeing a commitment to long-term follow-up studies. It is not inconceivable that a similar chain of events could lead to the clinical application of heritable genome editing. In time, perhaps sooner than many anticipate, genome editing methods will be developed that are safe and efficacious. Before that occurs, efforts must be undertaken to demystify the process. It is of vital importance that scientists do not abdicate the responsibility of educating the public, for if they do, the void may be filled by the writers of tabloid newspaper headlines, or others with agendas where the needs of the patient are not adequately considered. A comprehensive and transparent public consultation should be undertaken, and robust regulation introduced to ensure responsible governance and to earn the confidence of the public. Of course, it must be acknowledged that establishing universally acceptable guidelines may prove difficult as attitudes toward embryo manipulation vary substantially across cultures, nations, governments, and religions. However, this is a worthy challenge because the potential prize is great. If practical regulations can be created, the benefits of genome editing may at last be realized, transforming the fates of afflicted families for generations to come. With thousands of heritable mutations known to cause disease, coupled with steady advancements in assisted reproductive technology (ART) and genetic technologies, embryo selection after ART and preimplantation genetic testing for monogenic disease (PGT-M) has become the predominant option for couples at risk of transmitting genetic disorders to their children. Despite its success, embryo selection comes with several limitations, most notably the finite number of embryos found in an IVF cycle that are both developmentally competent and unaffected by the familial mutation, of which there are often none. Germline gene editing has been proposed as a strategy for targeted correction of mutations in the germline that could potentially replace embryo selection. The development of this novel technique has the potential to rescue the number of transferable embryos, thus possibly leading to better outcomes. So far, the great majority of such attempts have utilized CRISPR-Cas9 technology (shortened to CRISPR). Both technical and ethical considerations, however, render germline gene editing unacceptable for clinical application at the current time. CRISPR induces double-stranded DNA breaks (DSBs) that can be repaired by 1 of 2 mechanisms—an error-prone pathway, called non-homologous end joining (NHEJ), which is the most common response. Less frequently, homology directed repair (HDR) can be employed to correct the DSB, which uses the embryo’s own repair machinery. The earliest stage of development, particularly the fertilization stage, is the most optimal for genome editing if it facilitates delivery of edits into all cells of the future organism. However, embryos also appear to be most vulnerable to DNA damage at this stage, presumably because of deficiency in DNA repair present before the activation of the embryonic genome. This observation alone makes a compelling argument against the clinical use of CRISPR-Cas9. However, when coupled with other technical and biological factors impeding the success of the technique in its current format, as well as the complicated ethics, one must wonder whether it would be safer to avoid gene editing in human embryos altogether. In fact, as far as we can tell, most of the evidence suggests that gene editing interventions in the germline have the potential to negatively impact embryonic development, and we think it should, therefore, be avoided in clinical form. Thus far, the utilization of CRISPR in the human germline has taken a mostly pre-clinical focus, aimed at investigating whether disease-causing mutations could be disrupted or repaired. Several studies now have established a proof-of-principle as well as technical and methodological framework for the application of gene editing technique in human embryos donated or created for research (2Fogarty N.M.E. McCarthy A. Snijders K.E. Powell B.E. Kubikova N. Blakeley P. et al.Genome editing reveals a role for OCT4 in human embryogenesis.Nature. 2017; 550: 67-73Crossref PubMed Scopus (252) Google Scholar, 10Ma H. Marti-Gutierrez N. Park S.W. Wu J. Lee Y. Suzuki K. et al.Correction of a pathogenic gene mutation in human embryos.Nature. 2017; 548: 413-419Crossref PubMed Scopus (1) Google Scholar, 11Kang X. He W. Huang Y. Yu Q. Chen Y. Gaoet X. et al.Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing.J Assist Reprod Genet. 2016; 33: 581-588Crossref PubMed Scopus (211) Google Scholar, 12Liang D. Mikhalchenko A. Ma H. Marti Gutierrez N. Chen T. Lee Y. et al.Limitations of gene editing assessments in human preimplantation embryos.Nat Commun. 2023; 14: 1219Crossref PubMed Scopus (3) Google Scholar, 13Liang P. Xu Y. Zhang X. Ding C. Huang R. Zhang Z. et al.CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes.Protein Cell. 2015; 6: 363-372Crossref PubMed Scopus (807) Google Scholar, 14Tang L. Zeng Y. Du H. Gong M. Peng J. Zhang B. et al.CRISPR/Cas9-mediated gene editing in human zygotes using Cas9 protein.Mol Genet Genomics. 2017; 292: 525-533Crossref PubMed Scopus (154) Google Scholar). Initial reports from China showed low efficiency, as well as significant off-target consequences and mosaicism 456 with CRISPR. Later studies using refined methods showed improved editing frequency (>90%) in early human embryos (2Fogarty N.M.E. McCarthy A. Snijders K.E. Powell B.E. Kubikova N. Blakeley P. et al.Genome editing reveals a role for OCT4 in human embryogenesis.Nature. 2017; 550: 67-73Crossref PubMed Scopus (252) Google Scholar, 10Ma H. Marti-Gutierrez N. Park S.W. Wu J. Lee Y. Suzuki K. et al.Correction of a pathogenic gene mutation in human embryos.Nature. 2017; 548: 413-419Crossref PubMed Scopus (1) Google Scholar, 11Kang X. He W. Huang Y. Yu Q. Chen Y. Gaoet X. et al.Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing.J Assist Reprod Genet. 2016; 33: 581-588Crossref PubMed Scopus (211) Google Scholar). It appears that editing efficacy increases when microinjection of the CRISPR components is performed at fertilization or early in development. Optimization of the half-life of CRISPR reagents can reduce mosaicism (10Ma H. Marti-Gutierrez N. Park S.W. Wu J. Lee Y. Suzuki K. et al.Correction of a pathogenic gene mutation in human embryos.Nature. 2017; 548: 413-419Crossref PubMed Scopus (1) Google Scholar). After generation of DSBs, HDR uses either the uncut homologous wild-type DNA sequence or alternatively a synthetic homologous template harboring the correct sequence for restoration of the damaged copy. When changing the timing of microinjection and HDR template availability, it also appears that human embryos prefer to use the endogenous template over the synthetic one, by a mechanism that is largely not understood. One could imagine this to be a protective strategy, inherent to embryonic cells, to prevent genotoxic damage. The downside for the prospect of clinical application of CRISPR is the fact that most embryos still resolve generated DSBs by mutagenic NHEJ rather than the HDR, which is required for correction of most mutations associated with human disease. NHEJ introduces additional mutations (small insertions and deletions known as indels) rather than correcting the existing ones. This predominant form of repair is, therefore, suboptimal for gene editing, except when the intent is to disrupt gene function entirely. Because HDR occurs infrequently (<10%) in cells of early human embryos, the success of editing is low. This is especially problematic for humans, because IVF provides only a limited number of embryos. The second concern precluding the clinical application of CRISPR in the human germline is the possibility that the technique causes DNA damage that is not readily repaired in early human embryos, which are already predisposed to genomic instability. Several papers report large deletions and structural abnormalities affecting chromosome segments after CRISPR (5Alanis-Lobato G. Zohren J. McCarthy A. Fogarty N.M.E. Kubikova N. Hardman E. et al.Frequent loss of heterozygosity in CRISPR-Cas9-edited early human embryos.Proc Natl Acad Sci USA. 2021; 118e2004832117Crossref PubMed Scopus (89) Google Scholar, 12Liang D. Mikhalchenko A. Ma H. Marti Gutierrez N. Chen T. Lee Y. et al.Limitations of gene editing assessments in human preimplantation embryos.Nat Commun. 2023; 14: 1219Crossref PubMed Scopus (3) Google Scholar, 15Kosicki M. Tomberg K. Bradley A. Repair of double-strand breaks induced by CRISPR–Cas9 leads to large deletions and complex rearrangements.Nat Biotechnol. 2018; 36: 765-771Crossref PubMed Scopus (51) Google Scholar, 16Adikusuma F. Piltz S. Corbett M.A. Turvey M. McColl S.R. Helbig K.J. et al.Large deletions induced by Cas9 cleavage.Nature. 2018; 560: E8-E9Crossref PubMed Scopus (190) Google Scholar, 17Cullot G. Boutin J. Toutain J. Prat F. Pennamen P. Rooryck C. et al.CRISPR-Cas9 genome editing induces megabase-scale chromosomal truncations.Nat Commun. 2019; 10: 1136Crossref PubMed Scopus (210) Google Scholar). Segmental abnormalities are known to be detrimental to embryonic viability and can produce congenital abnormalities in offspring. In 2020, Alanis-Lobato et al. (5Alanis-Lobato G. Zohren J. McCarthy A. Fogarty N.M.E. Kubikova N. Hardman E. et al.Frequent loss of heterozygosity in CRISPR-Cas9-edited early human embryos.Proc Natl Acad Sci USA. 2021; 118e2004832117Crossref PubMed Scopus (89) Google Scholar) observed loss-of-heterozygosity in the edited cells that spanned regions beyond the on-target locus (4 kb to at least 20 kb outside the POU5F1 gene), as well as segmental gains and losses of chromosome 6 (the site of the POU5F1 gene) present in approximately 22% of the samples, significantly higher than observed in the control group. Recently, Liang et al. (12Liang D. Mikhalchenko A. Ma H. Marti Gutierrez N. Chen T. Lee Y. et al.Limitations of gene editing assessments in human preimplantation embryos.Nat Commun. 2023; 14: 1219Crossref PubMed Scopus (3) Google Scholar) reported that embryonic cells are often subject to loss of heterozygosity extending out of the target site, which would indicate the absence of repair in the broken strand. The study showed that, besides technical challenges of gene editing, assessments resulting from the minute amount of DNA processed, allelic dropouts, and amplification biases, DSBs can produce large deletions at the target site extending outside of the targeted MYBPC3 locus. Furthermore, some embryonic stem (ES) cells derived from the targeted embryos show copy-neutral loss of heterozygosity at the cleavage site, likely caused by interallelic gene conversion (12Liang D. Mikhalchenko A. Ma H. Marti Gutierrez N. Chen T. Lee Y. et al.Limitations of gene editing assessments in human preimplantation embryos.Nat Commun. 2023; 14: 1219Crossref PubMed Scopus (3) Google Scholar). This study, along with work of others, indicate that human embryos have a DNA damage repair deficiency, presumably because of the lack of gene expression before the activation of the embryonic genome at around the cleavage stage. Continued mitotic division, despite the presence of DSBs, suggests cellular mechanisms that usually preserve genetic integrity lack stringency, ultimately failing to ensure repair. The results provide a strong warning against the therapeutic use of CRISPR-Cas9 in human embryos, and underline the importance of basic research to evaluate the safety of genome editing techniques in the human germline. If gene editing was to replace embryo selection, it is essential that the intervention specifically target the locus where the mutation is found. However, when evaluating data from CRISPR screens and validations, it is apparent that most CRISPR constructs have a degree of off-target activity, especially at DNA sequences that share close homology to the targeted locus. Algorithms predicting off-target activity are becoming more accurate with time, but they mostly lack data from actual human embryonic cells, because of scarcity of the material, with ES cells commonly used as a proxy. However, it appears that ES cells exhibit significantly lower targeting efficiencies and they lack the diversity of mutational profiles resulting from NHEJ, which make them unsuitable for predicting off-targets effects (2Fogarty N.M.E. McCarthy A. Snijders K.E. Powell B.E. Kubikova N. Blakeley P. et al.Genome editing reveals a role for OCT4 in human embryogenesis.Nature. 2017; 550: 67-73Crossref PubMed Scopus (252) Google Scholar, 12Liang D. Mikhalchenko A. Ma H. Marti Gutierrez N. Chen T. Lee Y. et al.Limitations of gene editing assessments in human preimplantation embryos.Nat Commun. 2023; 14: 1219Crossref PubMed Scopus (3) Google Scholar). Also worth noting is the fact that these tools depend on in silico predictions. Confirmatory tests to exclude the possibility of transferring an embryo harboring additional mutations would be required for clinical application of CRISPR. Such a test would necessitate whole-genome sequencing of the embryo to examine all putative off-target consequences, a