AAV vs LNP Delivery
AAV and lipid nanoparticles represent two fundamentally different approaches to delivering genetic medicines into the human body. AAV vectors — the workhorse behind Zolgensma, Luxturna, and Hemgenix — use engineered viruses to deliver permanent gene copies but face payload limits, pre-existing immunity, and re-dosing barriers. LNPs — proven at massive scale through COVID-19 mRNA vaccines and now powering Intellia's in vivo CRISPR programs — deliver transient cargo with no size limit and the ability to re-dose. As the field moves from gene addition to gene editing, the choice of delivery vehicle is becoming as important as the editing tool itself.
Last updated: March 30, 2026
AAV (Adeno-Associated Virus)
Engineered, non-pathogenic viral vectors that deliver single-stranded DNA payloads into target cells. AAV integrates rarely, providing long-term episomal expression in non-dividing cells. Multiple serotypes (AAV1-9, AAVrh10, etc.) offer tissue tropism for different organs.
LNP (Lipid Nanoparticle)
Synthetic lipid-based particles (~80-100 nm) that encapsulate mRNA or ribonucleoprotein (RNP) cargo for delivery into cells. Proven at billion-dose scale through COVID-19 vaccines (Pfizer/BioNTech, Moderna) and now the leading delivery vehicle for in vivo CRISPR gene editing.
Key Specifications
| Feature | AAV (Adeno-Associated Virus) | LNP (Lipid Nanoparticle) |
|---|---|---|
| Vector type | Engineered non-pathogenic virus | Synthetic lipid-based nanoparticle (~80-100 nm) |
| Payload capacity | ~4.7 kb (single-stranded DNA) | No practical size limit (mRNA, RNP, guide RNA) |
| Expression duration | Long-term (years) in non-dividing cells; wanes in dividing cells | Transient (days to weeks — ideal for hit-and-run gene editing) |
| Re-dosing | Not possible due to anti-capsid immunity | Possible — no anti-vector immunity |
| Pre-existing immunity | 30-50% of population has neutralizing antibodies | Low (some anti-PEG antibodies in population) |
| Tissue targeting | Serotype-dependent: AAV9 (CNS/muscle), AAV8 (liver), AAV2 (retina) | Primarily liver (ApoE-mediated); lung, spleen with modified lipids |
| FDA-approved products | Zolgensma, Luxturna, Hemgenix, Elevidys, Roctavian | COVID vaccines (Comirnaty, Spikevax) — no gene therapy approvals yet |
| Manufacturing cost | Very high ($100K-500K+ per patient dose for high-dose systemic therapy) | Low-moderate ($50-500 per dose at scale) |
| Immunogenicity | High — capsid triggers strong humoral and cellular immune responses | Low — fully synthetic, no viral components |
| Key limitation | Payload size + no re-dosing + pre-existing immunity | Liver tropism + transient expression + endosomal escape efficiency |
AAV (Adeno-Associated Virus)
Advantages
- Proven clinical track record: 5+ FDA-approved gene therapies (Zolgensma, Luxturna, Hemgenix, Elevidys, Roctavian)
- Long-term expression in non-dividing cells — Luxturna efficacy sustained 7+ years
- Tissue-specific tropism via serotype selection: AAV9 (CNS/muscle), AAV8 (liver), AAV2 (retina)
- Efficient transduction of hard-to-reach tissues: CNS, retina, muscle, heart
- Well-established large-scale manufacturing processes (though expensive)
Limitations
- Strict 4.7 kb packaging limit — cannot deliver large genes (dystrophin, Factor VIII full-length) without truncation
- Cannot re-dose: anti-AAV neutralizing antibodies form after first exposure, blocking subsequent doses
- Pre-existing immunity: 30-50% of population has anti-AAV antibodies from natural exposure, disqualifying them
- Dose-dependent hepatotoxicity: high-dose AAV trials have seen liver inflammation, TMA, and deaths (Audentes AT132)
- Expression can wane in dividing cells — not truly permanent in pediatric liver or growing tissues
LNP (Lipid Nanoparticle)
Advantages
- No payload size limit — can deliver mRNA, guide RNA, Cas9 mRNA, base editors, prime editors of any size
- Re-dosing possible: no anti-vector immunity (though anti-PEG antibodies are a consideration)
- COVID vaccine-validated manufacturing at billion-dose scale — cost and supply chain proven
- Transient expression ideal for gene editing: deliver CRISPR, make permanent edit, cargo degrades — limits off-target window
- Intellia's LNP-delivered CRISPR (nex-z) achieved 93% TTR knockdown sustained 3+ years from single IV dose
Limitations
- Primarily targets liver via ApoE-mediated uptake — reaching non-liver tissues remains a major challenge
- Transient expression: not suitable for gene addition therapies requiring permanent expression
- Dose-dependent liver toxicity: ALT elevations observed in clinical trials at higher doses
- Endosomal escape efficiency is low (~1-2%) — most cargo degrades in endosomes, requiring high doses
- Anti-PEG antibodies (from prior PEG exposure) may reduce efficacy in some patients
The Verdict
AAV and LNP are complementary, not competing technologies — the right choice depends on the therapeutic goal. AAV excels when permanent gene expression is needed in non-liver tissues (CNS, retina, muscle), explaining its dominance in gene addition therapies like Zolgensma and Luxturna. But AAV's 4.7 kb limit, inability to re-dose, and 30-50% patient exclusion due to pre-existing immunity are fundamental constraints. LNPs are the clear winner for in vivo gene editing: transient CRISPR delivery makes the permanent DNA edit while the editing machinery degrades, minimizing off-target risk. Intellia's clinical validation (93% TTR knockdown, 3+ years durable) and billion-dose COVID manufacturing scale make LNPs the future of scalable genetic medicine. The field is moving decisively toward LNP-delivered gene editing for liver diseases, while next-generation LNP formulations (ionizable lipids, selective organ targeting) race to unlock non-liver tissues.
Sources & References
- [1]FDA approved gene therapy products
- [2]Intellia nex-z 3-year durability data
- [3]AAV immunogenicity and pre-existing antibodies (Nature Reviews)
- [4]LNP delivery for CRISPR gene editing (Nature Reviews Drug Discovery)
- [5]COVID mRNA vaccine LNP manufacturing scale
- [6]Zolgensma prescribing information (Novartis)
- [7]Selective organ targeting (SORT) LNPs