简短解答
People imagine the hard part of gene editing is the editing. It is not. The hard part is getting the machinery into the right cells inside a living body. We can reach the liver reliably, the eye and blood cells with more effort, and muscle, brain and lung barely at all. Almost every disease that looks 'nearly solved' but never arrives is stuck on this.
Delivery determines tissue tropism, dose, immunogenicity and duration of editor exposure. Lipid nanoparticles achieve efficient hepatocyte delivery and are transient, favouring safety. AAV offers broad tropism with engineered capsids but has a ~4.7 kb packaging limit, provokes pre-existing and induced immunity, and causes dose-dependent hepatotoxicity that has been fatal at high systemic doses. Ex vivo electroporation of ribonucleoprotein avoids systemic exposure entirely but requires conditioning and cell manufacturing.
What each route can and cannot reach
| Route | Reaches | Main limitation |
|---|---|---|
| Lipid nanoparticle | Liver, very efficiently | Little else, without targeting innovations |
| AAV | Eye, liver, some muscle and central nervous system | Size limit; immunity; toxicity at high doses |
| Ex vivo electroporation | Blood and immune cells | Requires conditioning chemotherapy and manufacturing |
| Inhaled delivery | Airway, in principle | Mucus and epithelial defences; largely preclinical |
| Direct injection | Eye, some local tissue | Only where a needle can go |
This table is effectively a map of which diseases have clinical programmes and which do not.
Sources
- National Human Genome Research Institute
Talking Glossary of Genomic Terms ↗