Adeno-associated virus (AAV) vectors have emerged as a dominant platform for gene delivery to the central nervous system (CNS) in neurodegenerative disease therapy. AAV is a small, non-pathogenic parvovirus that naturally infects humans but has not been associated with any known disease, making it inherently safe for therapeutic applications [1](https://pubmed.ncbi.nlm.nih.gov/33424157/). The virus's favorable safety profile, combined with its ability to transduce both dividing and non-dividing cells and maintain long-term transgene expression, has positioned AAV as the preferred vector for CNS gene therapy. [@aav2021]
The fundamental challenge in treating neurodegenerative diseases lies in delivering therapeutic genes across the [blood-brain barrier](/entities/blood-brain-barrier) (BBB) to the affected neuronal populations. AAV vectors offer a solution through their ability to be administered via multiple routes and their capacity for targeted engineering of capsid proteins to enhance CNS tropism. This mechanism page explores the molecular biology of AAV vectors, the engineering strategies used to optimize CNS delivery, and the clinical translation of these approaches for [Alzheimer's disease](/diseases/alzheimers-disease) (AD), [Parkinson's disease](/diseases/parkinsons-disease) (PD), amyotrophic lateral sclerosis (ALS), frontotemporal dementia (FTD), and Huntington's disease (HD). [@aavphpeb2018]
Adeno-associated virus (AAV) vectors have emerged as a dominant platform for gene delivery to the central nervous system (CNS) in neurodegenerative disease therapy. AAV is a small, non-pathogenic parvovirus that naturally infects humans but has not been associated with any known disease, making it inherently safe for therapeutic applications [1](https://pubmed.ncbi.nlm.nih.gov/33424157/). The virus's favorable safety profile, combined with its ability to transduce both dividing and non-dividing cells and maintain long-term transgene expression, has positioned AAV as the preferred vector for CNS gene therapy. [@aav2021]
The fundamental challenge in treating neurodegenerative diseases lies in delivering therapeutic genes across the [blood-brain barrier](/entities/blood-brain-barrier) (BBB) to the affected neuronal populations. AAV vectors offer a solution through their ability to be administered via multiple routes and their capacity for targeted engineering of capsid proteins to enhance CNS tropism. This mechanism page explores the molecular biology of AAV vectors, the engineering strategies used to optimize CNS delivery, and the clinical translation of these approaches for [Alzheimer's disease](/diseases/alzheimers-disease) (AD), [Parkinson's disease](/diseases/parkinsons-disease) (PD), amyotrophic lateral sclerosis (ALS), frontotemporal dementia (FTD), and Huntington's disease (HD). [@aavphpeb2018]
AAV is a member of the Dependoparvovirus genus within the Parvoviridae family. The virus has a simple structure consisting of a protein capsid approximately 25 nm in diameter surrounding a single-stranded DNA genome of approximately 4.7 kb [1](https://pubmed.ncbi.nlm.nih.gov/33424157/). The capsid is composed of three viral proteins (VP1, VP2, and VP3) in a 1:1:10 ratio, forming an icosahedral symmetry with 60 subunits. [@aavhsc2019]
The viral genome contains two inverted terminal repeats (ITRs) that serve as packaging signals and origins of replication. The rep gene encodes four proteins (Rep78, Rep68, Rep52, and Rep40) involved in replication and site-specific integration, while the cap gene encodes the capsid proteins. For vector production, the rep and cap genes are provided in trans, allowing the ITR-flanked transgene cassette to be packaged into the capsid. [@aavcapb2020]
AAV entry into cells is mediated by specific interactions between capsid proteins and cell surface receptors. Different serotypes utilize distinct receptor complexes, which determines their tissue tropism. The primary receptors for common CNS-targeted serotypes include: [@aav2019]
The choice of AAV serotype critically impacts transduction efficiency in the CNS. Natural serotypes vary significantly in their ability to cross the BBB and transduce [neurons](/entities/neurons) versus glial cells. [@convectionenhanced2019]
| Serotype | Species Origin | BBB Crossing | Neuronal Tropism | Glial Tropism | Key Applications | [@longterm2018]
|----------|---------------|--------------|------------------|---------------|------------------| [@multiyear2020]
| AAV1 | Primate | Low | High | Moderate | Motor [cortex](/brain-regions/cortex), research | [@preexisting2019]
| AAV2 | Human | Low | Moderate | Low | Historical trials, striatum | [@aavaadc2019]
| AAV5 | Bovine | Low | High | Moderate | Substantia nigra, research | [@aavsod2021]
| AAV8 | Simian | Moderate | Very High | Low | Broad CNS coverage | [@aavmihtt2019]
| AAV9 | Human/Primate | High | High | Moderate | IV delivery, preferred | [@aav2020]
| AAVrh10 | Rhesus | Moderate | Very High | Moderate | Excellent for PD | [@clinical2022]
| AAV-PHP.eB | Engineered | Very High (mouse) | High | Moderate | Research only | [@bbb2020]
| AAV-PHP.S | Engineered | High (mouse) | High | High | Research only | [@gene2021]
Serotype selection depends on the disease target, delivery route, and desired cell population. AAV9 has become the preferred serotype for intravenous delivery due to its ability to cross the BBB in humans, while AAVrh10 shows excellent CNS tropism for direct brain injection [5](https://pubmed.ncbi.nlm.nih.gov/30092717/). [@crispr2022]
Beyond natural serotypes, significant effort has focused on engineering AAV capsids with enhanced properties for CNS delivery. Three main strategies have emerged: [@promoter2021]
AAV-PHP.eB and AAV-PHP.S: Engineered through directed evolution in mice, these variants show dramatically enhanced BBB crossing (up to 100-fold improvement over AAV9 in mice) [2](https://pubmed.ncbi.nlm.nih.gov/30083074/). However, their enhanced CNS transduction is species-dependent, and they do not show the same efficiency in non-human primates or humans. [@safety2022]
AAVHSC: Derived from hematopoietic stem cell-derived capsids, AAVHSC variants show improved CNS tropism and reduced liver transduction compared to wild-type serotypes [3](https://pubmed.ncbi.nlm.nih.gov/29249656/).
AAV.CAP-B10: Engineered for improved CNS delivery in non-human primates, showing 10-100 fold higher transduction than AAV9 in the brain following intravenous administration [4](https://doi.org/10.1016/j.ymthe.2020.11.006).
AAV-Pan: A panel of capsids selected for broad CNS distribution, including the ability to transduce the spinal cord following systemic delivery.
The route of administration fundamentally determines AAV distribution in the CNS and the cell types that can be transduced.
IV delivery offers the least invasive approach but requires AAV vectors that can cross the BBB. AAV9 is the primary serotype used for this route, as it has demonstrated BBB crossing in both preclinical models and human trials [5](https://pubmed.ncbi.nlm.nih.gov/30092717/). Following IV injection, AAV9 transduces various brain regions, with particular tropism for motor neurons when administered at high doses.
Advantages: Non-invasive, whole-brain coverage potential Disadvantages: High doses required, liver sequestration, variable BBB penetration
IT delivery involves injection into the cerebrospinal fluid (CSF) space surrounding the spinal cord. This approach bypasses the BBB at the level of the spinal cord and can efficiently transduce motor neurons and supporting cells. IT delivery has been used in clinical trials for ALS and spinal muscular atrophy [6](https://pubmed.ncbi.nlm.nih.gov/30647051/).
Advantages: Lower doses than IV, direct access to spinal cord Disadvantages: Limited rostral distribution, requires lumbar puncture
ICV injection delivers vector directly into the lateral ventricles, enabling distribution throughout the ventricular system and parenchyma via CSF flow. This route has been explored for pediatric neurological disorders.
CED involves bulk flow infusion under pressure to distribute vectors through brain tissue. This approach allows for targeted, high-concentration delivery to specific brain regions while minimizing systemic exposure. CED has been used in clinical trials for PD and brain tumors [7](https://pubmed.ncbi.nlm.nih.gov/23797173/).
Stereotactic injection into specific brain regions allows precise targeting of affected areas. This approach has been used for PD (putamen), HD (striatum), and AD (hippocampus). While highly targeted, distribution is limited to the injection site.
Following cellular entry and endosomal escape, AAV genomes must undergo trafficking to the nucleus. The single-stranded genome is converted to double-stranded DNA in the nucleus, a rate-limiting step that influences expression kinetics. Self-complementary AAV variants can bypass this step by packaging inverted terminal repeat-flanked genomes that form double-stranded DNA upon entry.
AAV vectors typically remain episomal in the nucleus, forming circular concatemers that persist in the cell without integrating into the host genome. This maintains transgene expression in non-dividing cells (neurons, astrocytes) for years, as demonstrated in long-term follow-up of clinical trials [8](https://pubmed.ncbi.nlm.nih.gov/29490361/). Expression typically peaks 2-4 weeks post-delivery, depending on the promoter and target tissue [19](https://pubmed.ncbi.nlm.nih.gov/33575283/).
The choice of promoter determines which cell types express the transgene and the expression level:
One of AAV's key advantages is sustained expression in the CNS. Long-term studies in non-human primates have demonstrated stable expression for over 6 years [9](https://pubmed.ncbi.nlm.nih.gov/32415379/), supporting the durability of AAV-mediated gene therapy for chronic neurodegenerative diseases.
A significant challenge for AAV gene therapy is pre-existing humoral immunity in the human population. Serological studies indicate that 30-60% of adults have neutralizing antibodies against AAV capsids, which can reduce transduction efficiency or prevent readministration [10](https://pubmed.ncbi.nlm.nih.gov/28797034/). Assessment of neutralizing antibody titers is now standard in clinical trial eligibility screening.
T cell-mediated immune responses against transduced cells have been observed in some clinical trials, particularly when high vector doses trigger robust immune activation. In some cases, transduced cells expressing viral proteins (from residual rep gene expression in producer cells) can trigger cytotoxic T cell responses, leading to loss of transgene expression [20](https://pubmed.ncbi.nlm.nih.gov/35245891/).
While AAV shows favorable safety, concerns include:
PD gene therapy programs have targeted multiple pathways [15](https://pubmed.ncbi.nlm.nih.gov/35197623/):
| Program | Vector | Target | Mechanism | Stage |
|---------|--------|--------|-----------|-------|
| VY-AADC (Voyager) | AAV2-AADC | AADC enzyme | Restore L-DOPA conversion | Phase 1/2 [11](https://pubmed.ncbi.nlm.nih.gov/30682476/) |
| PR001 (Prevail) | AAV9-[GBA](/entities/gba) | GBA | Restore glucocerebrosidase | Phase 1/2 |
| AAV-GDNF (BrainLight) | AAV2-GDNF | GDNF | Neurotrophic support | Phase 1 |
| ProSavin (Oxford Biomedica) | Lentiviral | GAD, AADC, TH | Multi-enzyme restoration | Phase 1/2 |
The AAV2-AADC program (VY-AADC) showed promising results in Phase 1/2 trials, with improved motor function and reduced levodopa requirements in advanced PD patients [11](https://pubmed.ncbi.nlm.nih.gov/30682476/).
AD gene therapy approaches have focused on:
ALS gene therapy programs target multiple genetic forms [15](https://pubmed.ncbi.nlm.nih.gov/35197623/):
HD gene therapy programs include:
FTD programs are nascent but include AAV-mediated delivery of:
Lentiviral vectors remain an important platform for CNS gene delivery, particularly for applications requiring larger cargo capacity or integration [14](https://doi.org/10.1016/j.tibtech.2020.06.007).
| Property | AAV | Lentiviral |
|----------|-----|------------|
| Genome size | ~4.7 kb | ~8 kb |
| Integration | Rare (episomal) | Integrates (stable) |
| Expression duration | Long-term (years) | Very long-term (integration) |
| Pre-existing immunity | Common (30-60%) | Low |
| Manufacturing | Complex (triple transfection) | Complex (stable cell lines) |
| Dose requirements | Higher for CNS | Lower |
| Safety profile | Excellent | Good |
| Clinical experience | Extensive | Moderate |
When to use AAV: Long-term expression, non-dividing cells, favorable safety track record, established manufacturing processes
When to consider lentivirus: Large transgenes (>4.7 kb), applications requiring integration, situations where pre-existing AAV immunity is a concern
The majority of current CNS gene therapy clinical trials use AAV due to its superior safety profile and well-established manufacturing processes.
Continued engineering of AAV capsids aims to improve:
To overcome the ~4.7 kb packaging limit, dual-vector systems that reconstitute large transgenes from two AAV genomes are being developed, though efficiency remains a challenge.
Combining AAV delivery with CRISPR-Cas systems enables precise genome editing [18](https://doi.org/10.1038/s41587-022-01244-4). Approaches include:
Current AAV approaches are limited to single administration. Strategies to enable repeat dosing include:
The translation of AAV gene therapy for neurodegenerative diseases has accelerated significantly, with over 50 active clinical trials targeting various CNS disorders. The field has evolved from primarily academic investigations to a mature commercial ecosystem with multiple pharmaceutical companies advancing programs through clinical development.
Parkinson's Disease Clinical Trials: The most advanced AAV gene therapy programs are in PD, with several compounds having completed Phase 2 trials. The AAV2-AADC program (VY-AADC/VY-222) demonstrated meaningful improvements in motor function, reduced levodopa requirements, and improved quality of life metrics in patients with advanced PD. Long-term follow-up studies extending beyond 5 years show sustained therapeutic benefit without significant safety concerns.
ALS Gene Therapy: Multiple Phase 1/2 trials have evaluated AAV-delivered SOD1 silencing, with results demonstrating acceptable safety profiles and biomarker evidence of target engagement. The WAVE Life Sciences program showed dose-dependent reduction in CSF SOD1 levels, supporting the mechanism of action.
Huntington's Disease: The AAV5-miHTT program (Roche/Voyager) completed Phase 1b/2 trials demonstrating dose-dependent reduction in mutant huntingtin protein in CSF. This represents a critical proof-of-concept for allele-selective gene silencing in HD.
Neurotrophic Factor Delivery: AAV-mediated delivery of GDNF, BDNF, and NGF represents a strategy to support neuronal survival and function. While early trials showed biological activity, clinical efficacy has been limited by challenges in achieving adequate distribution throughout target brain regions. Next-generation approaches focus on improved delivery vectors and combination strategies.
Enzyme Replacement: For lysosomal storage disorders and metabolic deficiencies, AAV provides a platform for sustained enzyme delivery to the CNS. The GBA programs for PD aim to restore glucocerebrosidase activity, potentially modifying disease progression in patients with GBA mutations.
Gene Silencing: AAV-delivered shRNA and miRNA constructs enable sustained reduction of disease-causing proteins. Applications include mutant SOD1 in ALS, mutant HTT in HD, and alpha-synuclein in PD. Challenge remains in achieving adequate silencing while preserving normal protein function.
Gene Editing Integration: The convergence of AAV delivery with CRISPR-Cas systems enables precise genome modification. Current approaches focus on non-homologous end joining for gene disruption, with future directions including homology-directed repair for precise corrections and base editing for single-nucleotide modifications.
Biomarker development is critical for clinical trial success and patient selection:
The clinical translation of AAV gene therapy carries significant implications for patient care:
Disease Modification Potential: Unlike symptomatic treatments, gene therapy offers the potential for disease modification through sustained target engagement. Long-term expression profiles (6+ years in NHP studies) support durable therapeutic effects.
Treatment Timing: Early intervention before substantial neuronal loss may provide maximum benefit. This creates challenges for patient identification and trial design in slowly progressive diseases.
Combination Approaches: AAV gene therapy may synergize with small molecule drugs, biologics, and cell-based therapies. Combination strategies are being explored to address multiple disease pathways simultaneously.
Access and Equity: Current AAV gene therapies carry substantial costs ($1-3M per treatment). Healthcare system preparation for gene therapy reimbursement and delivery infrastructure is ongoing.
The regulatory pathway for AAV gene therapy has become increasingly defined: