Gabriela Ríos Ríos, SLS LLM and CLB student fellow, 2025

This is the final part of a three part blog post. 

The first part of this blog post explored the science behind the study published by Hashizume et al.[1] earlier this year, the second part dove deeper into the risks and limitations, now this third blog post will be dedicated to exploring the profound ethical questions it raises if the technique advances to clinical application and further exploring the legal landscape that currently governs such interventions across jurisdictions.

1. The ethical challenges: From Somatic to Germline Applications

A first ethical challenge is heritable impact. Somatic interventions affect only the treated individual, while germline interventions alter descendants who cannot consent and who may face risks we cannot yet characterize. Because uncertainty and the moral stakes grow with heritability, international reports converge on a cautious posture: allow tightly governed somatic applications; treat heritable uses as impermissible, or at minimum, not ready for clinical use. Authoritative frameworks (NASEM 2017[2]; WHO 2021[3]; ISSCR 2021[4]) and regional instruments (e.g. the Council of Europe’s Oviedo Convention, Article 13) reflect this asymmetry.

In practice, there are three routes to a germline change. First, embryo editing before implantation. Second, in-utero (fetal) delivery that reaches the fetal ovaries or testes, as some large animal studies show prenatal AAV vectors can cross to fetal germ cells and expose the pregnant patient, turning a somatic attempt into a potential germline edit[5]. Third, post-birth somatic delivery can, depending on the vector and route biodistributed to gonads. Regulators, therefore, require biodistribution studies and, when relevant, germline transmission safeguards in gene therapy trials, and keep heritable uses off-limits, insisting that any near-term exploration remain somatic and ideally, ex vivo[6].

A second challenge concerns risk, uncertainty and intergenerational responsibility. Even when somatic risks can be bounded, germline interventions raise questions about how much unknown long-term risk is acceptable and who bears it (future persons, families, and health systems). Leading reports propose high evidentiary thresholds (robust preclinical data, credible monitoring plans) and, for now, warn that heritable applications should not proceed to clinical trials. The ethical analysis conducted by the Nuffield Council on Bioethics allows that heritable editing could become permissible only under stringent conditions and with attention to social justice, not as a default extension of somatic success. A valid argument supporting germline editing in this specific case could come from how early neurodevelopmental differences in Down syndrome occur, as the only window to influence brain outcome might be at the blastocyst stage.

Third, consent and authority look different across the spectrum. For children, the accepted standard is parental permission with the child’s assent when developmentally possible[7], bounded by the harm principle as a threshold for overriding parental choices that would expose a child to serious risk or deny substantial benefit[8]. For adults with intellectual disabilities, international human rights (CRPD, Article 12) emphasize supported decision-making and respect for will and preferences, rather than defaulting to substitute decisions. These frameworks together suggest a narrow, medically grounded scope for pediatric somatic editing and a structured approach to adult participation[9].

Fourth, disability justice concerns complicate both somatic and germline contexts but become especially salient as one nears prenatal or embryonic uses. The expressivist critique holds that practices aimed at preventing the birth of people with certain genetic traits, through testing, selection or editing, send a social message that people who live with those traits are less valued and . For Down syndrome, this plays out in two concrete ways. In the prenatal setting, programs that emphasize screening and offer chromosome level prevention tools can be experiences by many in the Down syndrome community as an implicit statement that “people like me should not be born”, regardless of any individual parent’s motives. After birth, treating a comorbidity does not carry the same message because it targets a complication, not the person’s identity; by contrast, projects framed as “eliminating Down syndrome” in an existing child, or “normalizing cognition” risk re inscribing the expressivist harm by casting core identity features as defects to be corrected. Even if one does not accept the strongest versions of the argument, it rightly demands a practical commitment: inclusive governance; , careful language; co-designing studies with Down syndrome organizations; pairing any biomedical work with visible investments in education, support and inclusion; and promoting authentic engagement with disability communities to avoid stigmatizing effects and to align research aims with lived priorities. Seminal disability-rights analyses by Parens & Asch[10] and subsequent scholarship remain touchstones here[11]^/[12].

Fifth, justice and access. Somatic programs with clear clinical endpoints can, in principle, be assessed for fair selection, benefit sharing and long-term follow-up. Germline pathways raise additional concerns about distributional fairness, cross-border “forum shopping” and the social meaning of who gets invited into (or excluded from) genetic “prevention”.

Finally, there’s a problem of line drawing: therapy versus “normalization”, disease prevention versus trait selection. Therapy means treating discrete medical problems, like repairing an AV canal, addressing hematologic disease, thyroid replacement, sleep apnea, or, if it becomes feasible and safe, reducing the high risk of Alzheimer type dementia in adulthood. These target complications, health and function, not personhood and identity, and can be pursued as non-heritable, somatic care. By contrast, “normalization” projects seek to shift global cognition or other core features toward a neurotypical baseline. Even when motivated by care, they risk casting identity-defining differences as defects, raising the expressivist concern and cutting against neurodiversity. The neurodiversity perspective reminds us that many people with Down syndrome and their families value characteristic cognitive styles, social strengths, and culture; their goal is support and removal of barriers, not a neurotypical makeover.

A practical ethics approach therefore, prioritizes disease specific, non-heritable interventions co-designed with people with Down syndrome and their families, and treats success as improved functions and well-being. Somatic editing anchored to well-characterized diseased with measurable outcomes is easier to justify; germline contexts magnify slippery slope worries because choices are made before anyone can express interests or dissent, and because the non-identity problem complicates “benefit” claims for future persons. The upshot from major reports is not moral paralysis but procedural humility, by keeping the clinical scope narrow, the evidentiary bar high, and the governance participatory.

Our decision to pursue or refrain from this technology will ultimately reflect deeper cultural values about human difference and diversity. The disability community has long argued that their genetic variations are not merely medical conditions to be corrected, but integral components of human cultural diversity that enrich our collective experience [13]. From this perspective, Down syndrome is not simply a chromosomal abnormality. The elimination of specific genetic variations, whether through selective pregnancy termination or chromosome editing, raises profound questions about which lives we deem worthy of coming into existence and how we value certain differences.

As so the ethical and regulatory questions become less abstract, and it becomes a matter of setting guardrails where limited plausibility intersects with non-trivial risk.

2. Guardrails: where plausibility meets risk

If the ethical analysis points anywhere, it points to a very narrow lane: somatic, non-heritable interventions aimed at specific medical problems, not at reshaping identity laden traits. Keeping effects confined to the treated person avoids passing uncertainty to future generations, tying aims to concrete clinical endpoints respects disability perspectives by targeting complications and not people.

Because the science we’ve reviewed shows both promise and real hazards (allelic mistargeting at intended loci, structural variant scars, and uncertainty in vivo delivery) the evidence bar has to be high before first in human work. In practice that means rigorous preclinical characterization of in target fidelity, checks for off-target and structural variants, conservative dose findings, and long-term follow-up plans. None of this is red tape for its own sake, it’s a direct answer to the genomic and procedural risks catalogued earlier and is broadly consistent with current regulatory expectations for genome editing products.

The legal landscape largely creates restraints. In the United States, the FDA oversees somatic gene editing, but federal funding and even federal approval of clinical trials for germline editing is prohibited. In much of Europe, The Council of Europe, under the Oviedo Convention, prohibits introducing heritable modifications. Latin America, meanwhile, presents a mosaic of regulations, often with limited enforcement. Read together, these realities effectively close the germline door and leave only a somatic gene therapy path (which is exactly where the ethical case is stronger anyway, but the benefits might be smaller).

Finally, responsible work here has to be transparent and participatory, by creating public registries, independent oversight, and early, genuine partnership with disability communities are essential to align the research aimed with lived priorities. If this moves at all, it should move in the open, with reasons the public can see and evaluate.

Current systems were not designed with chromosome-scale editing in mind, and they struggle to keep pace with rapid scientific advances. There is growing recognition that adaptive regulatory models—such as staged approvals based on accumulating safety data—will be needed to responsibly manage the risks and benefits of this technology as it moves toward clinical reality.

Conclusion: Are the risks worth the benefits?

Taken together, the Hashizume study is best read as a proof of principle in cells, not a clinical blueprint. It shows that chromosome level editing can, under controlled conditions, restore more typical gene expressions. It also shows why translation will be hard, allele mistargeting at intended loci, repair scars from double strand breaks, and a stubborn delivery problem that only grows as we move from dishes to organs.

Here, the ethics must not be an afterthought but a compass. We have a duty to develop therapies that can alleviate suffering, but also protect individuals and communities from harm. As we contemplate following this path, we will be forced to ask ourselves big societal questions: how do we balance the hope of alleviating suffering against the risks of unintended harm? Who decides which conditions require these interventions, and how do we ensure respect and the rights of people with disabilities? We need to concentrate on technical innovation while dedicating similar efforts on developing robust ethical, legal and societal frameworks that can keep pace with the accelerating capabilities of genome engineering. Embryo stage and broad fetal uses concentrate uncertainty and heritable impact and are therefore both technically and normatively ill-suited for clinical pursuit. Post-birth somatic uses, especially ex vivo approaches in blood and immune compartments, align better with that we can responsibly justify, as the effects are confined to the treated person, endpoints are concrete and risks can be monitored. The clinical universe is also small, as beyond trisomy 21, plausible candidates are largely trisomy 13 or 18 and sex chromosome aneuploidies, so ambition should be trimmed to fit reality rather than rhetoric. If work moves forward, it should do so narrowly and in the open, somatic, non heritable aims, rigorous preclinical evidence of fidelity and safety, long term follow up and disability inclusive governance. In paralell, the right scientific bets are clear, safer, non-DSB modalities, better in vivo delivery, and honest replication in relevan human cells and tissues.

I land then on trisomic rescue belonging in the laboratory for now, and if ever in patients only in tightly bounded, somatic, ex vivo studies with clear medical endpoints. Everything upstream, embryo editing, identity shaping goals, organ wide fetal or neonatal interventions, should remain off the table until the science, the law and the social mandate say otherwise. Our choices here signal not just what we can do, but what kind of diversity we intend to welcome.

[1] Hashizume et al., “Trisomic Rescue via Allele-Specific Multiple Chromosome Cleavage Using CRISPR-Cas9 in Trisomy 21 Cells.”

[2] Human Genome Editing: Science, Ethics, and Governance, with Committee on Human Gene Editing: Scientific, Medical, and Ethical Considerations et al. (National Academies Press, 2017), https://doi.org/10.17226/24623.

[3] WHO Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing. Human Genome Editing: Recommendations, 1st ed (World Health Organization, 2021).

[4] International Society for Stem Cell Research, ISSCR Guidelines for Stem Cell Research and Clinical Translation, Version 1.1, May 2021, 2021.

[5] Beltran Borges et al., “Prenatal AAV9-GFP Administration in Fetal Lambs Results in Transduction of Female Germ Cells and Maternal Exposure to Virus,” Molecular Therapy Methods & Clinical Development 32, no. 2 (2024), https://doi.org/10.1016/j.omtm.2024.101263.

[6] U.S. Department of Health and Human Services et al., S12: Nonclinical Biodistribution Considerations for Gene Therapy Products, 2023.

[7] Aviva L. Katz et al., “Informed Consent in Decision-Making in Pediatric Practice | Pediatrics | American Academy of Pediatrics,” accessed September 16, 2025, https://publications.aap.org/pediatrics/article/138/2/e20161485/52519/Informed-Consent-in-Decision-Making-in-Pediatric?utm_source=chatgpt.com?autologincheck=redirected.

[8] Douglas S. Diekema, “Parental Refusals of Medical Treatment: The Harm Principle as Threshold for State Intervention,” Theoretical Medicine and Bioethics25, no. 4 (2004): 243–64, https://doi.org/10.1007/s11017-004-3146-6.

[9] Aviva L. Katz et al., “Informed Consent in Decision-Making in Pediatric Practice | Pediatrics | American Academy of Pediatrics.”

[10] Erik Parens and Adrienne Asch, “Disability Rights Critique of Prenatal Genetic Testing: Reflections and Recommendations,” Mental Retardation and Developmental Disabilities Research Reviews 9, no. 1 (2003): 40–47, https://doi.org/10.1002/mrdd.10056.

[11] A Asch, “Prenatal Diagnosis and Selective Abortion: A Challenge to Practice and Policy.,” American Journal of Public Health 89, no. 11 (1999): 1649–57, https://doi.org/10.2105/ajph.89.11.1649.

[12] Non-Invasive Prenatal Testing: Ethical Issues (Nuffield Council on Bioethics, 2017).

[13] Felicity Boardman, “Human Genome Editing and the Identity Politics of Genetic Disability,” Journal of Community Genetics 11, no. 2 (April 2020): 125–27, https://doi.org/10.1007/s12687-019-00437-4.