Genomics: Insight

Research Question: 

How does CRISPR-Cas9 editing affect the long-term genomic stability of hematopoietic stem cells, and what evidence shows it can cause large-scale chromosomal damage linked to increased cancer risk?

Introduction:

The advent of the CRISPR-Cas9 system has revolutionized biotechnology by enabling precise genomic alterations. By utilizing a guide RNA (gRNA) to direct the Cas9 endonuclease to a specific DNA sequence, researchers can induce double-strand breaks (DSBs) to knock out genes or insert therapeutic sequences.¹ This technology provides hope for  “ex vivo” gene therapy, a technique that removes cells from a patient's body, genetically modifies them in a lab, and then reintroduces them back to the body, specifically in hematopoietic stem cells (HSCs), which can be returned to a patient to treat blood disorders like sickle cell disease or beta-thalassemia. 

However, with growing evidence that CRISPR can cause mutations, chromosomal breaks, and large-scale genomic instability, hope for its implementation has decreased. While initial concerns focused on “off-target” effects, where the Cas9 cuts at unintended sites with similar sequences, recent research highlights the precariousness of unintended consequences at the on-target site.¹

One of the most concerning potential issues related to the use of CRISPR-Cas9 techniques in HSCs is the possibility that large-scale genomic alterations induced by gene editing could be perpetuated in those stem cells and, over time, lead to the development of cancer in the patient. While such a study has not yet occurred in HSCs because a significant period of time is required and CRISPR is relatively new, related research in the mouse model system has revealed that gene editing through CRISPR-Cas9 in a mouse liver led to a type of cancer called hepatocellular carcinoma developing in 71% of those animals, a rate that far exceeds that of cancer development in control groups.⁹ Although the model was not performed in HSCs, these findings indicate that cancer development resulted from gene editing within the cells and suggest that CRISPR may cause large-scale genomic damage.⁹ Therefore, research investigating the potential for chromosomal damage from CRISPR-Cas9 editing in HSCs is important.

Evidence Linking CRISPR-Cas9 Activity to Large Deletions at Cas9 Target Sites

Initial studies, such as Kosicki et al. (2018)³, are beginning to elucidate the scope of CRISPR-mediated structural changes in mammalian genomes. The study conducted an intensive analysis of the CRISPR-Cas9 editing within both human and mouse cells. In the study, they discovered that large deletions frequently occurred at the Cas9-targeted loci.¹ When these deletions occurred, they extended far beyond the intended site, often by kilobases, far exceeding the anticipated footprint of typical non-homologous end joining repair.¹ Most of the changes were undetectable by base-level PCR-based genotyping,⁴ revealing a large blind spot in early quality control methods. Kosicki’s study suggests that even precise gRNA targeting does not always ensure genomic stability, as DSB repair can create complex local rearrangements that may dispose of cells early and lead to malignant transformation.⁸

Comparing Four Key Studies on Unintended CRISPR-Induced Genome Alterations 

(Table 1)

To deepen understanding of the evidence for structural changes induced by CRISPR, the following table summarizes the four main studies in this review and their findings.

These studies consistently show that CRISPR can cause unexpected, large changes in DNA, not just the tiny and precise edits scientists aim for. Even in alternative CRISPR systems and across multiple cell types, including already diseased cancer cells, researchers observed that large deletions, rearrangements, and chromosomal instability can all occur. In cells that are already compromised, this instability is concerning because it may promote tumor progression. This risk is even more significant in HSC therapy because a single affected stem cell could expand in the patient's blood system and be detrimental to the patient's health. The presence of these unintended edits suggests a link between CRISPR-induced gene disruption and heightened cancer risk, especially when editing takes place in a cell that already has a heightened level of genetic vulnerability. 

Structural Variants and Expanding Evidence of CRISPR-Induced Genomic Instability

In vitro findings from multiple studies prompted researchers to examine the structural damage that occurs in living organisms. A 2022 study examining CRISPR-edited animal models confirmed that structural variants, such as deletions, inversions, and duplications, also arise in vivo, a process performed within the living organism.² This study showed how genome instability is not just limited to artificial laboratory conditions,² but can also occur in complex, actively dividing tissues. Findings such as this raise concerns about long-term safety, particularly in therapeutic contexts. 

As a way to add specific evidence from immune cells to the experiment, Wu et al. (2022) demonstrated that CRISPR-edited T cells undergoing the “ex vivo” process, a process performed outside the living body, could accumulate and propagate different structural variants, such as the examined large-scale deletions and duplications. Even these unusual events that occur in a single edited cell can be amplified as that cell proliferates. When looking at HSCs, it's important to consider that they are naturally long-lived and capable of extensive self-renewal. If a structural variant confers a growth advantage or disrupts a tumor suppressor gene, it could further lead to a clonal expansion and malignancy.⁸

Adding to these experiments, large-scale analyses in 2023 further enforce the prevalence of the unintended genomic alterations. Studies of CRISPR-edited human cells reported relatively unusual high frequencies of deletions, inversions, and chromosomal translocations at the target sites.⁶ Some changes were longer and beyond the intended locus, affecting the neighboring genes of distant chromosomal regions. Long-read sequencing revealed standard short-read methods often as inaccurate as they underestimate the frequency and size of these structural changes.⁴ This overall suggests that the routine screening approaches are not stable and may miss clinically relevant genomic damage. 

Additionally, safety-focused investigations that use genome-wide sequencing document more complex genomic disruptions, such as chromothripsis⁸ (in which chromosomes undergo extensive shattering and rearrangement). This large-scale instability is strongly associated with and relates to oncogenesis, the process of normal cells transforming into cancer cells through alterations, cancer in multiple cell types. Genome-wide, the scope and purpose of detected abnormally large structural variants at off-target sites, which again further challenge the thought that the precise guide RNA design alone guarantees safety.

Although newer technologies, such as new base editors and prime editors, may reduce the reliance on the double-strand breaks and generate fewer large-scale deletions, emerging research still cautions the blind spots that remain. Base editors are engineered CRISPR systems that chemically change one DNA base into another without cutting both strands of the DNA helix. Usually, the process is done with a modified CRISPR Cas9 protein and deaminase enzyme, which changes specific nucleotide sequences, allowing for a more precise editing process. Prime editors are an even newer development and extend upon base editing by adding a modified Cas9 and a reverse transcriptase enzyme, which is then guided by a specialized prime-editing guide RNA. This system directs the cell to copy a new genetic sequence directly into the genome and is more precise and less disruptive than traditional CRISPR methods. 

Even the improved systems may produce large-scale genomic changes that evade normal detection assays.⁷ Therefore, while next-generation editors continue to develop and show further progress, they do not ultimately eliminate the need for comprehensive, genome-wide safety assessment. 

When viewed critically, there are several examples where CRISPR-Cas9 editing can introduce significant structural instability and a lack of reliability beyond the small insertions or deletions it makes. For HSCs, the implications are particularly serious because they are long-lived, self-renewing, and responsible for generating all blood lineages, including single edited cells that harbor an undetected structural variant, which can also expand over time. Issues like these can theoretically increase the risk of leukemia or other hematologic malignancies after years of therapy.⁸ As a result, evaluating editing efficiency at the intended locus helps ensure safety. Long-term genomic stability may become a central metric in assessing CRISPR-based therapies intended for clinical use.³

Because these cells are long-lived and proliferative, any structural damage that bypasses initial screening could, in theory, lead to clonal expansion and, ultimately, to leukemic cells. 

Conclusion

In summary, while the CRISPR-Cas9 system offers hope for regenerative medicine and biotechnology, the “molecular scissors” metaphor, though apt for its precision in targeting, often oversimplifies the complexity of the cellular response to double-strand breaks and therefore requires a better understanding of the potential genotoxicity CRISPR may cause.

Research consistently demonstrates that CRISPR-mediated edits are not always the precision they are intended to be. The emergence of large-scale deletions, chromosomal translocations, and chromothripsis, an event in which one or more chromosomes break into pieces and are then stitched back together, suggests that “on-target” success can be accompanied by profound structural instability. A major concern persists regarding the “blind spots” in standard quality control.⁴

Applying this technology to hematopoietic stem cells (HSCs) has uniquely high stakes. Because these cells are long-lived and proliferative, any structural damage that bypasses initial screening could, in theory, lead to clonal expansion and, ultimately, to leukemic cells. Therefore, “successful” editing at the target locus is an insufficient metric for clinical safety.

Ultimately, the goal is not to stifle innovation but to ensure that the “cures” of tomorrow do not inadvertently become the “pathologies” of the future. Future research must move beyond the simple “cut-and-paste” efficiency and focus on the long-term stability of the entire genome.

References: 

1. Hoban MD, Cost GJ, Mendel MC, et al. Comprehensive analysis and accurate quantification of unintended large gene modifications induced by CRISPR-Cas9 gene editing. PLoS One. 2023;— Article PMC9586483. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC9586483/
2. Owens DDG, et al. Microhomologies are prevalent at Cas9-induced larger deletions. Nucleic Acids Res. 2019;47(14):7402–7417. doi:10.1093/nar/gkz459. Available from: https://academic.oup.com/nar/article/47/14/7402/5498629.
3. Kosicki 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(8):765–771. doi:10.1038/nbt.4192. Available from: https://pubmed.ncbi.nlm.nih.gov/30010673/
4. Genome editing with HDR-enhancing DNA-PKcs inhibitor AZD7648 causes large-scale genomic alterations. Nat Biotechnol. 2024;43:1778–1782. doi:10.1038/s41587-024-02488-6. Available from: https://doi.org/10.1038/s41587-024-02488-6
5. Chen T, Barzi M, Furey N, Kim HR, Pankowicz FP, Legras X, et al. CRISPR/Cas9 gene therapy increases the risk of tumorigenesis in the mouse model of hereditary tyrosinemia type I. JHEP Rep. 2025;7(4):101327. doi:10.1016/j.jhepr.2025.101327. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC11985117/
6. Höijer I, Emmanouilidou A, Östlund R, Van Schendel R, Bozorgpana S, Tijsterman M, et al. CRISPR-Cas9 induces large structural variants at on-target and off-target sites in vivo that segregate across generations. Nat Commun. 2022;13:627. doi:10.1038/s41467-022-28244-5. Available from: https://www.nature.com/articles/s41467-022-28244-5
7. Large DNA deletions occur during DNA repair at 20-fold lower frequency for base editors and prime editors than for Cas9 nucleases. Nat Biomed Eng. 2024; (s41551-024-01277-5). doi:10.1038/s41551-024-01277-5. Available from: https://www.nature.com/articles/s41551-024-01277-5 — (from your list).
8. Whole genomic analysis reveals atypical non-homologous off-target large structural variants induced by CRISPR-Cas9-mediated genome editing. Nat Commun. 2023;14:40901. doi:10.1038/s41467-023-40901-x. Available from: https://www.nature.com/articles/s41467-023-40901-x — (Nature Communications link from your list).
9. Aussel C, Cathomen T, Fuster-García C. The hidden risks of CRISPR/Cas: structural variations and genome integrity. Nat Commun. 2025;16:7208. doi:10.1038/s41467-025-62606-z. Available from: https://www.nature.com/articles/s41467-025-62606-z