Trends in AAV Targeting of Non-Liver Tissues

Introduction

Recombinant adeno-associated virus (AAV) vectors have become the dominant platform for in vivo gene therapy, with more than 700 active clinical and preclinical programs, with 30% in active clinical development. AAV gene therapy’s biggest clinical wins to date have come disproportionately from tissues the virus naturally reaches well. This primarily includes the liver, where natural serotypes (AAV1-AAV9) transduce efficiently after a simple intravenous infusion, and secondly the eye, which is accessible by direct local injection. However, many of the highest-burden genetic diseases reside elsewhere: in the brain and spinal cord, in skeletal and cardiac muscle, and in other tissues that natural AAV serotypes reach poorly,  if at all, when dosed systemically. 

Over the past three years, the field has been moving away from natural serotypes and toward deliberately engineered, receptor-informed capsids

Achieving therapeutic transduction in these difficult-to-reach tissues using natural serotypes has historically required very high systemic doses, and high doses mean more vector accumulating in the liver as an unwanted off-target site, raising both toxicity risk and the sheer manufacturing cost of producing enough vector per patient. Over the past three years, the field has been moving away from natural serotypes and toward deliberately engineered, receptor-informed capsids paired with tissue-specific uptake elements, a combination designed to concentrate therapeutic effect in the target tissue while detargeting the liver.

Capsid Engineering: From Empirical Screening to Receptor-Informed Design

The most visible engine of this shift has been directed evolution platforms that screen enormous, diverse capsid libraries directly in vivo, typically in non-human primates, to find rare variants with genuinely enhanced tissue tropism. Voyager Therapeutics has built one such advanced approach with its TRACER platform. It is an RNA-based functional screening system that has produced capsids delivering more than one-thousand-fold higher brain transgene expression than conventional AAV9 in non-human primates. Second-generation TRACER variants show even greater liver detargeting and transduction of fifty to seventy-five percent of cells across diverse brain regions.

Capsida Biotherapeutics has pursued a similar strategy and reached an important regulatory milestone, holding what the company describes as the first FDA IND clearance for an engineered, intravenously delivered gene therapy that crosses the blood-brain barrier while simultaneously detargeting the liver. 2025 non-human primate data for its Friedreich’s ataxia program CAP-004 showed transduction of more than eighty percent of cerebellar Purkinje cells alongside meaningful cardiac and retinal expression.

In a parallel trend, rather than screening empirically and hoping a winning capsid translates from mice to humans, researchers have identified specific receptors, including alkaline phosphatase and transferrin receptor 1 that mediate transcytosis across the blood-brain  barrier.  This   allows   capsids   to  be  rationally designed around a known human-relevant mechanism rather than in vivo primate testing. 

Receptor mediated transcytosis is achieved by tagging the viral capsids with targeting peptide motifs, single domain camelid nanobodies or bispecific antibodies for dual tissue targeting. For example, 4D Molecular Therapeutics generates capsids with enhanced tropism for specific CNS cell types and for the retina, advancing multiple programs into clinical development, while JSR Therapeutics utilizes miniaturized antibodies against targeted receptors on their Just AAV capsids.

Myotropic and Cardiotropic Capsids

Skeletal and cardiac muscle present their own distinct targeting challenge. Since natural serotypes transduce these tissues inefficiently relative to the liver, very high systemic doses for neuromuscular indications such as Duchenne muscular dystrophy are required. 

Tabebordbar, Lagerborg, Stanton and colleagues at the Broad Institute and Harvard used an in vivo RNA-based selection strategy where RGD motif-containing capsid variants achieved ten- to twenty-nine-fold greater muscle transduction than AAV9 in mice. Kate Therapeutics has since extended this lineage under the name of MyoAAV-LD variants.

A two-step engineering strategy reported in 2026 stripped a capsid of essentially all-natural tissue tropism by domain-swapping between AAV9 and AAV2 and then deliberately reintroduced muscle specificity by inserting a myogenic peptide into this de-targeted backbone, yielding a capsid designated as AAV.eM whose muscle-specific, liver-detargeted profile held up consistently across mouse strains and non-human primates. These MyoAAV family and AAV.eM rely on peptide motifs binding to overexpressed cell surface integrin on muscle tissues that facilitate endocytosis of the viral vectors. 

Since natural serotypes transduce muscle tissues inefficiently relative to the liver, very high systemic doses are required for neuromuscular indications.

The heart has proven to be a particularly stubborn target historically. Notable early cardiac gene therapy trials include the CUPID program in 2007. It demonstrated biological activity but no real clinical benefit. That long drought finally broke in October 2025 when Henry, Chung, Alvisi and colleagues at Ask Bio reported the first positive phase 1 human data for a cardiotropic AAV vector, AB-1002, showing good tolerability and preliminary signs of efficacy in eleven heart failure patients followed for twelve months, a result widely described as a genuine inflection point for cardiac gene therapy.

Other research has yielded cardiac-focused variants such as AAV9-CR (cardiac-restricted), which was selected from random mutagenesis libraries for enhanced binding to cardiomyocyte surface receptors. More recently, the BRC family of vectors, derived from the Multiple Capsid Mutation Strategies (MCMS) library, demonstrated not only a significant improvement in brain transduction but also a substantial reduction in liver targeting, with preserved cardiac transduction. Another breakthrough comes from modified AAVrh.10 variant with mutations that reduce hepatic glycan binding while maintaining or even enhancing affinity for cardiac proteoglycans. Concurrently, the JUST-AAV platform has produced dual-targeting capsids that carry antibodies or DARPins (designed ankyrin repeat proteins) for both cardiac troponin I and muscle-specific kinase, achieving 52-fold higher heart transduction models compared to AAV9 in preclinical research. Indirect mechanisms of cardiac targeting are also explored.

This includes incorporating the AAV genome with miR-142-3p target sequences that are abundant in dendritic cells and macrophages for immune evasion. Also, incorporating miR-122 target sequences into the AAV transgene 3’ UTR for reducing off-target liver expression by promoting the degradation of transgene mRNA in hepatocytes which in turn, results in preserving cardiac expression by lowering immune evasion. 

Ocular Targeting

The eye was the field’s first major non-liver success, anchored by the approval of Luxturna for inherited retinal disease. Recent engineering efforts have focused on improving intravitreal delivery, a less invasive route than subretinal injection but one that historically struggles to penetrate the eye’s inner membrane. 

This has been addressed by tuning how engineered capsids interact with heparan sulfate proteoglycan (HSPG) on the membrane surface. Suprachoroidal injection techniques are separately expanding how much of the posterior segment a single injection can reach.

Conclusion

These engineered AAV variants are not merely incremental improvements but represent a paradigm shift, indicating that synthetic capsids could outperform their natural predecessors in both safety and efficacy. The post-liver era of AAV gene therapy will likely prioritize the CNS first due to high disease burden and recent proof-of-concept, followed by cardiac and skeletal muscle driven by muscular dystrophy populations, and then sensory organs like retina and cochlea for further delivery optimizations, and, finally, visceral organs of pancreas, kidney, lung which remain under researched. As machine learning accelerates capsid design and cross-species validation matures, the field is converging on the ambitious goal of injectable, one-time gene therapies that reach only the intended diseased tissue, leaving the liver and dose-limiting toxicity challenges behind.

As such, CMC teams should note that while the manufacturers and industry consortia are pushing towards serotype-agnostic platform steps for AAV manufacturing, newly engineered synthetic capsids may pose a different set of challenges and opportunities in both upstream and downstream processing. These capsids also introduce new analytical aspects that challenge many platform analytical methods. Viral vector developers and manufacturers may require additional know-how to serve these unique processes alongside broader AAV offerings. CROs and CDMOs should consider how to onboard in-depth viral capsid analytical and process development expertise to serve these needs of future complex processes where synthetic viral vectors are conjugated to peptides, nanobodies and antibodies.

Resources:

^[1] 1.Hoffman JA, Denton N, Sims JJ, Meggersee R, Zhang Z, Olagbegi K, Wilson JM. Modulation of AAV9 Galactose Binding Yields Novel Gene Therapy Vectors and Predicts Cross-Species Differences in Glycan Avidity. Hum Gene Ther. 2024 Sep;35(17-18):734-753. doi: 10.1089/hum.2024.050. Epub 2024 Aug 13. PMID: 39001819.

^[2] Asokan A, Shen S. Redirecting AAV vectors to extrahepatic tissues. Mol Ther. 2023 Dec 6;31(12):3371-3375. doi: 10.1016/j.ymthe.2023.10.005. Epub 2023 Oct 6. PMID: 37805712; PMCID: PMC10727976.

^[3] Xu X. Strategies of AAV capsid engineering for targeted delivery to brain, muscle, and retina. Front Mol Biosci. 2026 Jan 6;12:1750807. doi: 10.3389/fmolb.2025.1750807. PMID: 41567865; PMCID: PMC12816303.

^[4] Tabebordbar, M., Lagerborg, K. A., Stanton, A., King, E. M., Ye, S., Tellez, L., Krunnfusz, A., Tavakoli, S., Widrick, J. J., Messemer, K. A., Troiano, E. C., Moghadaszadeh, B., Peacker, B. L., Leacock, K. A., Horwitz, N., Beggs, A. H., Wagers, A. J., & Sabeti, P. C. (2021). Directed evolution of a family of AAV capsid variants enabling potent muscle-directed gene delivery across species. Cell, 184(19), 4919-4938.e22. https://doi.org/10.1016/j.cell.2021.08.028

^[5] JCR Pharmaceuticals. (2025). JUST-AAV platform achieves 52-fold improvement in muscle gene delivery while reducing liver exposure. MedPath Trial Report.

^[6] Song Q, Wu J, Zhang Q, Wu S, Luo X, Xu M, Xu Y, Wang G, Yu H, Chen H, Lu Z, Ma L. Directed evolution of novel AAV variants using the MCMS library for enhanced CNS tropism and reduced liver targeting in mice. Mol Ther Methods Clin Dev. 2025 Jun 25;33(3):101522. doi: 10.1016/j.omtm.2025.101522. PMID: 40687372; PMCID: PMC12272419.

^[7] Wang L, Bell P, Somanathan S, Wang Q, He Z, Yu H, McMenamin D, Goode T, Calcedo R, Wilson JM. Comparative Study of Liver Gene Transfer With AAV Vectors Based on Natural and Engineered AAV Capsids. Mol Ther. 2015 Dec;23(12):1877-87. doi: 10.1038/mt.2015.179. Epub 2015 Sep 28. PMID: 26412589; PMCID: PMC4700115.

^[8] Belfort BDW, Jia JD, Garza AR, Insalaco AM, McGinnis JP, Pekarek BT, Ortiz-Guzman J, Tepe B, Chen H; aSCENT-PD Investigators; Liu Z, Arenkiel BR. Comparative analysis of AAV serotypes for transduction of olfactory sensory neurons. Front Neurosci. 2025 Feb 13;19:1531122. doi: 10.3389/fnins.2025.1531122. PMID: 40018362; PMCID: PMC11865030.

^[9] Issa SS, Shaimardanova AA, Solovyeva VV, Rizvanov AA. Various AAV Serotypes and Their Applications in Gene Therapy: An Overview. Cells. 2023 Mar 1;12(5):785. doi: 10.3390/cells12050785. PMID: 36899921; PMCID: PMC10000783.

^[10] Deverman BE, Ravina BM, Bankiewicz KS, Paul SM, Sah DWY. Gene therapy for neurological disorders: progress and prospects. Nat Rev Drug Discov. 2018 Sep;17(9):641-659. doi: 10.1038/nrd.2018.110. Epub 2018 Aug 10. Erratum in: Nat Rev Drug Discov. 2018 Oct;17(10):767. doi: 10.1038/nrd.2018.158. PMID: 30093643.

^[11] Chamberlain K, Riyad JM, Weber T. Cardiac gene therapy with adeno-associated virus-based vectors. Curr Opin Cardiol. 2017 May;32(3):275-282. doi: 10.1097/HCO.0000000000000386. PMID: 28169951; PMCID: PMC5544589.  

^[12] Wang, D., et al. (2026). Engineering novel AAV capsids by global de-targeting and subsequent muscle-specific re-targeting in mice and non-human primates. Molecular Therapy Advances. https://doi.org/10.1016/j.omtm.2026.101504 

^[13] Voyager Therapeutics. (2024, May 8). Voyager Therapeutics presents data for second-generation TRACER-generated capsids and CNS gene therapy programs. Press Release. https://ir.voyagertherapeutics.com 

^[14] Voyager Therapeutics. (2022, October 4). Voyager Therapeutics announces license option agreement with Novartis for next-generation TRACER AAV capsids. Press Release. https://ir.voyagertherapeutics.com

^[15] Capsida Biotherapeutics. (2025, May 14). Systemic AAV gene therapy with next-generation engineered capsids for CNS and cardiac symptoms in Friedreich’s ataxia. Presentation at the American Society of Gene and Cell Therapy 28th Annual Meeting. 

^[16] Kate Therapeutics. (2024). MyoAAV-LD variants for systemic neuromuscular gene therapy. Presentation at the American Society of Gene and Cell Therapy 27th Annual Meeting. 

^[17] Wang, D., et al. (2026). AAV.Zero and AAV.eM: Global detargeting and muscle-specific re-targeting strategy. Molecular Therapy Advances. https://www.cell.com/molecular-therapy-family/advances 

^[18] Henry TD, Chung ES, Alvisi M, Sethna F, Murray DR, Traverse JH, Roessig L, Roberts L, Reddy S, Chen Y, Ozkan TG, Webb S, Mittal M, Ervin L, Sadek H, Mikhail S, Haghighi K, Jiang C, Samulski RJ, Kranias EG, Tretiakova AP, Hajjar RJ. Cardiotropic AAV gene therapy for heart failure: a phase 1 trial. Nat Med. 2025 Nov;31(11):3845-3852. doi: 10.1038/s41591-025-04011-z. Epub 2025 Oct 21. PMID: 41120766; PMCID: PMC12618250. 

^[19] Inside Precision Medicine. (2025, October 22). Heart-targeting AAV advances cardiac gene therapy in first-in-human trial. https://www.insideprecisionmedicine.com 

^[20] Yang L, Liu Z, Chen G, Chen Z, Guo C, Ji X, Cui Q, Sun Y, Hu X, Zheng Y, Li Y, Gao F, Chen L, Zhou P, Pu WT, Guo Y. MicroRNA-122-Mediated Liver Detargeting Enhances the Tissue Specificity of Cardiac Genome Editing. Circulation. 2024 May 28;149(22):1778-1781. doi: 10.1161/CIRCULATIONAHA.123.065438. Epub 2024 May 28. PMID: 38805581. 

^[21] Majowicz A, Maczuga P, Kwikkers KL, van der Marel S, van Logtenstein R, Petry H, van Deventer SJ, Konstantinova P, Ferreira V. Mir-142-3p target sequences reduce transgene-directed immunogenicity following intramuscular adeno-associated virus 1 vector-mediated gene delivery. J Gene Med. 2013 Jun-Jul;15(6-7):219-32. doi: 10.1002/jgm.2712. PMID: 23658149. 

^[22] Kolesnik VV, Nurtdinov RF, Oloruntimehin ES, Karabelsky AV, Malogolovkin AS. Optimization strategies and advances in the research and development of AAV-based gene therapy to deliver large transgenes. Clin Transl Med. 2024 Mar;14(3):e1607. doi: 10.1002/ctm2.1607. PMID: 38488469; PMCID: PMC10941601.