Programming Technology for Inserting Large-Scale Genes Entirely Overcomes Clinical Entry Hurdles
Background
Genome editing technologies that induce single nucleotide substitutions (base editing) or insertions/deletions of a few dozen base pairs have already entered Phase 1/2 clinical trials, proving their potential as therapeutics. However, for severe single-gene diseases with multiple damaged exons or complex genetic diseases, fundamental treatment is difficult through methods that only change a few bases. There is an urgent need for technology that can precisely insert an entire complete gene coding sequence of several kilobases (kb) or more into a target location to perform normal functions fully.
Currently commercialized CRISPR-Cas9-based homology-directed repair (HDR) methods show a sharp decline in efficiency in non-dividing cells. When delivering exogenous DNA templates, risks of cytotoxicity and genomic toxicity due to random insertion also follow. To stably insert large gene cargoes into desired sites within the genome, a next-generation integration system that goes beyond existing cleavage-repair mechanisms is required.
Key Findings
Recently, the academic community has been advancing large-scale gene integration strategies focusing on recombinases, CRISPR-associated transposases (CAST), genome writing systems, and retrotransposon-based platforms. Three genome engineering experts analyzed the advantages, disadvantages, and technical bottlenecks hindering clinical application for each technology.
Recombinase-based systems demonstrate precision in joining large fragments of tens of kb or more via attachment sites (attP/attB) without double-strand breaks (DSB). However, a preliminary step is required to first construct landing pads that the enzyme can recognize within the human genome. Recently, active attempts have been made to bypass this problem by combining prime editors, which are based on single-strand breaks, with large-scale serine recombinases.
CAST and retrotransposon systems use guide RNA (gRNA) to directly find specific sequences and have demonstrated the potential to insert DNA cargoes of 5–10kb or more without double-strand breaks. However, they still fall below clinical standards in terms of off-target insertion rates and overall integration efficiency. A common challenge is that the total size of the enzyme complex and the cargo DNA is too massive to be loaded into viral vectors or lipid nanoparticles (LNPs). The innate and adaptive immune responses triggered by foreign protein complexes upon in vivo injection are also cited as variables that must be resolved.
To overcome these, experts suggested the primary development paths as: reducing enzyme size through protein engineering, designing variants with increased binding specificity through structural determination based on cryo-electron microscopy (cryo-EM), and optimizing in vitro transcription (IVT) mRNA and non-viral delivery vehicles.
Significance and Outlook
Once large-scale gene insertion technology is perfected, the need to design customized guides for each individual patient's mutation type will disappear. By replacing the entire disease gene with a normal sequence or inserting expression cassettes into safe genomic harbors, a single therapeutic can respond to genetic diseases caused by hundreds of different mutations.
However, for in vivo application, chemical and biological innovations to overcome the capacity limits of delivery vehicles are essential. The establishment of evaluation protocols using ultra-precise whole-genome sequencing (WGS) to thoroughly verify the risk of oncogenicity that may occur if large foreign sequences are inserted into off-target sites must also be carried out in parallel. If such safety standards are established, the next-generation large-scale gene insertion platform will reveal the potential to fundamentally reshape the entire production process of cell and gene therapeutics, not only for rare genetic diseases but also for CAR-T cell manufacturing.
Nature Biotechnology, Published online: 17 September 2026; doi:10.1038/s41587-026-03319-6While technologies for smaller gene edits, such as base editing, have already entered clinical trials, larger gene cargoes face different challenges, including low editing efficiencies, delivery vehicle size constraints, and possible immunogenicity. Programmable genome engineering systems using recombinases; transposases, including CRISPR-associated transposases (CASTs); genome writing systems; and retrotransposons promise different avenues for therapy, but each system has unique limitations. What are the most promising strategies to further develop these technologies, and which parameters need to be further improved to make them efficient and safe for therapy? We asked three experts in the field to share their thoughts on the challenges and opportunities.
- It opens the door for developing single therapeutics for diseases with large defective gene sizes and diverse mutation locations, such as Duchenne muscular dystrophy (DMD), Hemophilia A, and Cystic Fibrosis. In the case of the dystrophin gene, which reaches 2.2Mb, existing base editing was limited to the level of exon skipping, but if large-scale insertion technology is refined, a functional mini-gene can be permanently anchored at once.
- From an industry perspective, the process for next-generation cell therapeutics will be dramatically simplified. Previously, lentiviral vectors had to be used to insert Chimeric Antigen Receptors (CAR), or complex multi-editing processes had to be undergone. Applying a high-efficiency large-scale insertion system would allow desired immune receptors, safety switches, and regulatory factors to be precisely anchored at specific genomic sites in T-cells in a single step, reducing production costs and significantly improving batch-to-batch quality consistency.