Two delivery platforms now dominate in vivo gene therapy and gene editing: adeno-associated virus and lipid nanoparticles. Both can deliver nucleic acid payloads to target cells. Both have produced approved therapeutics. And both are being actively developed across an overlapping set of disease indications. This convergence has made the comparison between them a practical question for development teams, investors, and commercial strategists — not just a scientific exercise.
The honest answer is that neither platform is superior in general. They are suited to different biological problems. Understanding where each wins, where each struggles, and where the frontier between them is actively contested requires looking at the underlying biology of delivery, expression, and immune interaction.
What Lipid Nanoparticles Are
Lipid nanoparticles are synthetic delivery vehicles — spherical assemblies of ionizable lipids, phospholipids, cholesterol, and polyethylene glycol-lipid conjugates — that encapsulate nucleic acid cargo and facilitate its delivery into cells. LNPs were first developed for small interfering RNA delivery and reached their first regulatory approval with Onpattro (patisiran) in 2018 for transthyretin amyloidosis. Their profile as a delivery platform expanded dramatically during the COVID-19 pandemic, when LNP-mRNA vaccines from Moderna and Pfizer-BioNTech demonstrated large-scale safety and efficacy in billions of people — the largest clinical dataset ever generated for any LNP product.
LNPs are now being explored for delivery of mRNA, siRNA, antisense oligonucleotides, base editors, prime editors, and CRISPR-Cas9 ribonucleoprotein complexes in therapeutic applications. The cargo flexibility of LNPs — their ability to encapsulate different nucleic acid types without major reformulation — is one of their defining advantages over viral vectors.
Payload Flexibility: LNPs Win
The most consequential advantage of LNPs over AAV is payload capacity. AAV is constrained to approximately 4.7 kilobases of single-stranded DNA — a limit set by the physical dimensions of the capsid. This constraint is manageable for many monogenic diseases where the therapeutic gene fits within the packaging limit, but it categorically excludes others.
LNPs have no analogous size constraint for mRNA payloads. mRNA molecules ranging from a few hundred to tens of thousands of nucleotides have been encapsulated and delivered by LNPs. For gene editing applications using CRISPR-Cas9, LNPs can simultaneously deliver both the guide RNA and the Cas9 mRNA — a combined payload that would exceed AAV’s packaging limit if delivered as DNA. This payload flexibility has made LNPs the leading delivery platform for base editing and prime editing programs targeting the liver, where LNP delivery is highly efficient.
Re-dosability: LNPs Win
AAV gene therapy is, for most practical purposes, a one-time treatment. After the first administration, the patient’s immune system generates capsid-directed neutralizing antibodies that would inactivate a second dose of the same serotype. Re-dosing is theoretically possible with a different serotype, but this strategy is constrained by cross-reactive immunity and the requirement that the patient lack pre-existing antibodies to the alternative serotype.
LNPs do not display protein antigens on their surface — they present a PEGylated lipid exterior that is less immunogenic than viral capsids. Repeated dosing of LNP-siRNA and LNP-mRNA products has been demonstrated in clinical practice: Onpattro is administered by intravenous infusion every three weeks in treated patients, and the COVID-19 mRNA vaccines demonstrated tolerability of multiple doses at short intervals in billions of recipients. For diseases requiring ongoing or repeated treatment — particularly those where a single gene delivery does not achieve permanent correction — LNP re-dosability is a significant clinical advantage.
Duration of Expression: AAV Wins
LNPs delivering mRNA produce transient protein expression. mRNA is inherently unstable — it is degraded by cellular ribonucleases with a half-life that varies by sequence and modification but is typically on the order of hours to days in vivo. For applications where a brief pulse of protein expression is sufficient — vaccine immunization, acute gene editing with subsequent disappearance of the editing machinery — this transience is not a limitation and may be a safety advantage. For applications requiring sustained protein replacement over months or years, mRNA delivery requires repeated dosing to maintain therapeutic levels.
AAV, delivering a DNA episome that persists in non-dividing cells, achieves long-term transgene expression from a single administration. Decade-long expression has been documented in hemophilia B patients treated with AAV8-factor IX. This durability profile is unmatched by any LNP-mRNA approach currently in clinical development, and it remains the primary reason AAV dominates in diseases requiring sustained protein replacement in post-mitotic tissues.
Liver Targeting: Comparable, with Nuances
Both AAV and LNPs are effective at targeting the liver after intravenous administration, and both platforms have produced approved liver-directed therapeutics. The mechanisms differ: AAV relies on capsid-receptor interactions on hepatocyte surfaces, with serotypes like AAV8 and AAV5 achieving high hepatocyte transduction efficiency. LNPs accumulate in the liver through a mechanism involving adsorption of apolipoprotein E from plasma onto the LNP surface, followed by LDL receptor-mediated uptake in hepatocytes — a pathway that is efficient and reproducible across LNP formulations using ionizable lipids optimized for hepatic delivery.
For liver-directed gene silencing using siRNA or gene editing using base editors, LNPs are currently the leading clinical platform. For liver-directed gene addition requiring durable protein expression — as in hemophilia — AAV remains the standard, though LNP-mRNA approaches for hemophilia requiring repeat dosing are in clinical development.
Extrahepatic Delivery: AAV Wins, for Now
LNP delivery outside the liver remains a significant technical challenge. Standard intravenous LNP formulations accumulate predominantly in the liver, with secondary distribution to the spleen and, at lower levels, other tissues. Achieving efficient extrahepatic delivery — to the CNS, muscle, lung, eye, or other tissues — requires formulation modifications, alternative routes of administration, or targeting ligands that redirect LNP biodistribution away from hepatocytes.
AAV, through serotype diversity and route-of-administration flexibility, has demonstrated efficient delivery to the CNS (AAV9 via IV or intrathecal), retina (AAV2 via subretinal or intravitreal injection), muscle (AAV1, AAV6 via intramuscular), and other tissues. This extrahepatic versatility is currently one of AAV’s most durable competitive advantages over LNPs, and it underlies the continued dominance of AAV in CNS, neuromuscular, and ocular gene therapy programs.
Research into targeted LNP formulations using tissue-specific lipid compositions, ionizable lipid chemistry optimized for non-hepatic cells, and active targeting ligands is active and advancing. LNP delivery to the lung via inhalation is the most clinically advanced extrahepatic LNP application, with programs targeting cystic fibrosis and other pulmonary diseases. CNS LNP delivery remains earlier stage, with intrathecal and intracerebral administration routes under investigation.
Manufacturing and Scale: Different Profiles
LNP manufacturing is continuous-flow and scalable using microfluidic mixing technology. The lipid components are synthesized or procured as raw materials, combined in a defined ratio with the nucleic acid cargo during mixing, and formulated into the final drug product. The process is relatively rapid and does not require cell-based production, living organisms, or biological containment beyond what is standard for pharmaceutical manufacturing. Scale-up of LNP processes has been demonstrated at extraordinary scale by the COVID-19 vaccine programs.
AAV manufacturing is cell-based, requiring transfection of producer cells, multi-day incubation for virus production, cell harvest, lysis, and multi-step downstream purification. Batch sizes are smaller, timelines are longer, and the analytical complexity of characterizing a biological particle — including full/empty capsid ratio, genome integrity, and potency — is greater than for a synthetic nanoparticle. Manufacturing cost per dose for AAV gene therapy remains high relative to most LNP products, and manufacturing scale-up challenges have contributed to the high prices of approved AAV therapeutics.
Where the Platforms Will Compete
The frontier between AAV and LNP will continue to shift as both technologies evolve. LNPs will expand their extrahepatic reach through formulation innovation, gaining ground in tissues that AAV currently dominates. AAV will benefit from engineered capsids that reduce pre-existing immunity barriers and improve tissue targeting. Base editing and prime editing — which currently favor LNP delivery for liver indications — may shift toward AAV-based delivery if payload optimization overcomes the packaging constraint, or remain with LNPs if re-dosability proves essential for editing efficacy.
For development teams choosing between platforms today, the biology of the target disease should drive the decision: persistent expression in post-mitotic tissue points toward AAV; repeated or transient treatment in the liver points toward LNPs; extrahepatic targets beyond the liver currently favor AAV pending further LNP innovation. Neither platform is in decline. Both are expanding. The developers who understand where each genuinely excels will make better platform decisions than those treating the choice as a technology preference.
MKA Insights