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

This is Part II of a three part blog post. 

As explained in detail in the first part of this blog post, the study published by Hashizume et al.[1] demonstrated a technique that could selectively eliminate the extra chromosome in trisomy 21, by targeting the demolition and subsequent removal of an entire chromosome while leaving the necessary pair intact. The proposed approach corrects the underlying genetic imbalance, reversing cellular abnormalities and restoring normal gene expression patterns in laboratory-grown cells. However, while the promise of chromosome level editing to correct trisomy 21 is remarkable, a complicated set of issues need to be addressed and discussed while the technique advances, if it ever does, to in vivo application. The paper provides valuable data on several categories of risk that deserve scrutiny.

1. Risks and limitations

Genomic Risks:

Off-target effects:

The most immediate concern with any CRISPR-based therapy is off-target genetic damage. Despite the allele-specific approach’s precision compared to non-specific targeting, whole-genome sequencing revealed 5-6 structural variants (“SVs”) in post-edit trisomy clones and approximately 1 SV in post-edit disomy clones, evidence that the remaining chromosomes bear scars from the process.

Importantly, when the team aimed to cut only the extra chromosome 21, but 5 of the 13 cut sites the guides occasionally snipped one of the two normal copies, because those DNA sequences differ by only a single letter. This means that the CRISPR system sometimes mistakenly targeted the “good” copies of chromosome 21 rather than just the extra copy. Even when the researchers refined their approach to six gRNAs with demonstrated specificity, some residual off-target activity persisted.

Mosaicism:

Mosaicism cuts both ways. A small share of people with Down syndrome are naturally mosaics, meaning only some of their cells carry the extra chromosome, and, as a group, they often have milder clinical features than those with full trisomy 21. Severity varies with the proportion and tissue distribution of trisomic cells. From that lens, nudging tissues toward a higher fraction of disomy cells could plausibly soften some symptoms. But therapy induced mosaicism at today’s per cell rescue rates would likely be patchy across organs and may include cells with repair related rearrangements, such cellular heterogeneity might lead to unpredictable developmental outcomes or long-term instability in treated tissues.

Procedural Risks:

From lab to therapy:  

Even if genomic risks could be mitigated, substantial procedural challenges remain. The paper reports extremely low delivery efficiency—only 1.07% of induced pluripotent stem cells and 13.9% of fibroblasts successfully received the CRISPR-Cas9 system via electroporation (a technique that uses millisecond electrical pulses to momentarily open pores in cell membranes, allowing the CRISPR plasmid to enter). This inefficiency would severely limit therapeutic applications, particularly for tissues with limited regenerative capacity like neurons.

Perhaps most critically, the study demonstrates efficiency only in isolated cell cultures. The leap to treating intact tissues or whole organisms introduces enormous additional complexities. The researchers showed some promise in non-dividing cells, achieving a modest 3.2% chromosome elimination rate, but this falls far short of what would be needed for meaningful therapy.

These substantial risks must be weighed carefully against the potential benefits of CRISPR-based chromosome elimination therapy. As the next section will explore, the demonstrated cellular improvements following successful chromosome elimination suggest significant therapeutic potential, but is it enough to justify these risks?

2. Human use in Practice

As with any powerful technology, the ethical landscape of chromosome editing is complex and evolving. Right now, research is confined to somatic cells in laboratory settings, which keeps risks relatively contained and allows for careful studying of safety and efficacy. However, the technology’s trajectory points toward more challenging scenarios. In the near future we might see attempts to treat adults with Down syndrome-related conditions, such as leukemia, using edited blood cells.

It’s important to note that beyond trisomy 21, whole chromosome “rescue” would plausibly apply only to the small set of extra chromosome conditions (aneuploidies) compatible with live births, primarily, trisomy 13 and 18 and the sex chromosome trisomies. Other full autosomal trisomies are almost uniformly embryonic or fetal lethal, the rare liveborn cases are typically mosaic or involve partial/segmental trisomies.[2]^/[3].

However, before we argue the ethics in the abstract, it helps to ask a practical question: when, if ever, could this be used in humans? There are four key moments in which the technique presented by Hashizume et al. could be attempted, each having a mix of what’s plausible and what could go wrong.

Editing before implantation is the most conceptually attractive point to act: removing the extra copy of chromosome 21 at the eight-cell or blastocyst stage could, in theory, prevent downstream effects. In practice, however, early embryos often respond unpredictably to breaks, with mosaic outcomes and large, unintended alterations. Because any change at this stage would be heritable, the scientific uncertainties and regulatory prohibitions align: this route is technically imaginable but currently inadvisable[4]^/[5].

By the end of the first trimester, some features associated with Down Syndrome, especially structural heart differences, are already set[6], while other systems, including the lungs and brain, continue to develop. That timing makes early fetal intervention appealing in theory but difficult in execution. Delivering a complex editing payload safely to a high share of cells in specific organs, remains a major constraint as any dose given during pregnancy reaches to patients, the pregnant person and the fetus. The delivery vehicles that carry CRISPR (often AAV viruses or nanoparticles) can spread to the mother’s organs and can also cross the placenta. Because exposure of maternal tissue or the fetal germline would turn a somatic treatment into a potential germline one, routes and doses must be very conservative. Additionally, many adults already have antibodies against AAV, which can neutralize the vector before it reaches the fetus [7]^/[8]. Even if delivery were feasible, the organs most people hope to influence, particularly the brain, are precisely those that are the hardest to reach at scale.

Soon after birth, the picture shifts. Narrow somatic uses, especially ex vivo approaches where cells are edited outside the body and then returned, seem more credible than attempting a whole-body approach. Blood cells are the most realistic targets[9][10], and some Down syndrome associated comorbidities, such as leukemias, and in some cases lymphomas live in those compartments. By contrast, trying to reach enough cells in the brain, heart, or lungs to change overall development runs into a practical coverage problem[11], meaning the fraction of target cells that actually receive and execute the edit. Organ level change usually requires very high coverage in specific cell types, which is hard because (i) vectors like AVV or nanoparticles distribute unevenly, and the brain ads a blood-brain barrier; (ii) dose limits and immune responses cap how much you can give or re-dose; and (iii) this approach requires multiple components to reach the same cell.

In childhood and adulthood, consent considerations improve for some patients, yet the biology does not become easier. The central nervous system remains difficult to access broadly and safely, and most of today’s delivery vehicles (especially AAV vectors) can’t be given repeatedly, one dose prompts neutralizing antibodies that often persist for years and block later doses[12]; ex vivo strategies for blood or immune conditions continue to make the most sense.

Across all stages, the central constraint is less the cleverness of the method than the ability to deliver it safely to enough of the right cells. Where that hurdle is tractable (narrow, somatic, ex vivo contexts with concrete clinical endpoints), a cautious path may emerge. Where it is not (embryo editing or broad fetal/neonatal/adult organ-wide interventions), the risk-benefit profile remains unfavorable. The recent trisomy 21 “rescue” study in cells is an impressive proof-of-principle but it underscores how far we are from organ-scale use on people.

Read against that practical map, the ethical questions sharpen, as we move from somatic to germline contexts, who bears the risks, how is consent obtained, and what counts as benefits all change, and the justifications that may hold for narrow somatic uses no longer carry unmodified into prenatal or embryonic settings.

Opposition is likely to be the weakest where an intervention clearly treats disease or prevents serious harm after birth, like repairing life threatening cardiac or pulmonary problems, treating hematologic disease, or, if credible evidence emerges, reducing the markedly elevated lifetime risk of Alzheimer’s disease in adults with Down syndrome. In pediatric ethics and practice, parental discretion is broad but not unlimited, when refusing a low burden, high benefit intervention exposes a child to a significant risk of serious harm, clinicians and, if necessary, the state, may override refusal. Applying chromosome level editing to modify global cognition is a different category, it currently lacks proven safety and efficacy, targets identity laden traits, and should, if ever contemplated, remain optional and grounded in robust supported decision making rather than compelled care.

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

[2] Warburton, Dorothy, Louis Dallaire, Maya Thangavelu, Lori Ross, Bruce Levin, and Jennie Kline. “Trisomy Recurrence: A Reconsideration Based on North American Data.” American Journal of Human Genetics 75, no. 3 (September 2004): 376–85. https://doi.org/10.1086/423331

[3] Nagaoka, So I., Terry J. Hassold, and Patricia A. Hunt. “Human Aneuploidy: Mechanisms and New Insights into an Age-Old Problem.” Nature Reviews Genetics 13 (June 18, 2012): 493–504. https://doi.org/10.1038/nrg3245

[4] Michael Kosicki et al., “Repair of Double-Strand Breaks Induced by CRISPR–Cas9 Leads to Large Deletions and Complex Rearrangements,” Nature Biotechnology 36, no. 8 (2018): 765–71, https://doi.org/10.1038/nbt.4192.

[5] Gregorio Alanis-Lobato et al., “Frequent Loss of Heterozygosity in CRISPR-Cas9-Edited Early Human Embryos,” Proceedings of the National Academy of Sciences of the United States of America 118, no. 22 (2021): e2004832117, https://doi.org/10.1073/pnas.2004832117.

[6] Jill P. J. M. Hikspoors et al., “A Pictorial Account of the Human Embryonic Heart between 3.5 and 8 Weeks of Development,” Communications Biology 5, no. 1 (2022): 226, https://doi.org/10.1038/s42003-022-03153-x.

[7] John S. Riley et al., “Preexisting Maternal Immunity to AAV but Not Cas9 Impairs in Utero Gene Editing in Mice,” The Journal of Clinical Investigation 134, no. 12 (n.d.): e179848, https://doi.org/10.1172/JCI179848.

[8] Rrita Daci and Terence R. Flotte, “Delivery of Adeno-Associated Virus Vectors to the Central Nervous System for Correction of Single Gene Disorders,” International Journal of Molecular Sciences 25, no. 2 (2024): 1050, https://doi.org/10.3390/ijms25021050.

[9] Haydar Frangoul et al., “CRISPR-Cas9 Gene Editing for Sickle Cell Disease and β-Thalassemia,” New England Journal of Medicine 384, no. 3 (2021): 252–60, https://doi.org/10.1056/NEJMoa2031054.

[10] Haydar Frangoul et al., “Exagamglogene Autotemcel for Severe Sickle Cell Disease,” New England Journal of Medicine 390, no. 18 (2024): 1649–62, https://doi.org/10.1056/NEJMoa2309676.

[11] Petr O Ilyinskii et al., “Readministration of High-Dose Adeno-Associated Virus Gene Therapy Vectors Enabled by ImmTOR Nanoparticles Combined with B Cell-Targeted Agents,” PNAS Nexus 2, no. 11 (2023): pgad394, https://doi.org/10.1093/pnasnexus/pgad394.

[12] Daci and Flotte, “Delivery of Adeno-Associated Virus Vectors to the Central Nervous System for Correction of Single Gene Disorders.”