Overview

On March 28, 2025, the team led by Guangping Gao from Horae Gene Therapy Center, University of Massachusetts Chan Medical School, published a review article titled “The curious case of AAV immunology” in Molecular Therapy. This review primarily discusses the complexity of adeno-associated virus (AAV) immunology, including the interactions between AAV and the host immune system, the clinical challenges posed by immune responses, strategies for optimization, and future research directions.

Recombinant adeno-associated virus (rAAV) has many unique properties that make it amenable to delivering genetic material to cells in vivo. So, it is the leading gene delivery vector for the treatment of numerous genetic diseases. However, the immune system perceives these AAV vectors as foreign and potential pathogens. The resulting immune responses cause significant safety issues, negatively impact the durability of the therapeutic gene transduction, and can prevent redosing when needed. Therefore, it is imperative to elucidate the molecular mechanisms underlying rAAV-induced immune activation and to develop immunomodulatory strategies that balance therapeutic efficacy with safety. This represents a core direction for future research in the field.

About Genevoyager

Genevoyager (Wuhan) Co., Ltd. (Genevoyager) is a leading provider of one-stop CRO/CDMO services for viral vectors, proteins, and vaccines. With a proprietary technology platform for large-scale AAV manufacturing and cGMP-grade facilities for viral vector, protein, and vaccine production, Genevoyager is dedicated to advancing healthcare by supporting its partners to develop safe, effective, affordable, and accessible gene therapy products, protein-based drugs, and therapeutic vaccines that address unmet medical needs.

The company boasts a highly skilled team of 300 professionals, covering expertise in viral vector research and development, preclinical efficacy evaluation, CMC development, QA/QC, and GMP production. Genevoyager's cGMP facility in Jiangxia District, Wuhan spans 68,000ft2 and encompasses 5 viral vector production lines and 1 fill & finish line. The facility commenced operations in October 2023 and has facilitated the completion of multiple project deliveries. 

Genevoyager possesses a proprietary insect baculovirus system known as One-Bac 4.0. This system employs a novel approach for regulating AAV expression cassettes, coupled with critical process optimizations. It attains higher yield, better infectivity, and higher full-capsid ratio for large-scale manufacturing. The single-batch production capacity of the One-Bac 4.0 system can reach up to 1E+18 vg.

Interactions between rAAV, host cells, and the immune system

How does the immune system recognize AAV?

AAV vectors are associated with low immunogenicity and lack of pathogenicity; however, the innate immune response to AAV vectors can still influence their efficacy and safety. More adverse immune responses to AAV have been observed as clinical trials have used increasingly higher dosages of vectors, which may be due to triggering innate immune pathways in conjunction with provoking the adaptive immune system to induce strong responses.

 

Many studies have indicated that AAVs are recognized by various PRRs, including Toll-like receptor (TLR)2, TLR9, cGas, and RIG-I/MDA5 (Figure 1). TLR9, cGAS, and RIG-I/MDA5 activate host interferon (IFN) responses. TLRs signal through adaptor proteins such as MyD88 and TRIF, leading to the activation of IRF3, IRF7, and nuclear factor (NF)-κB and the production of type 1 IFN and proinflammatory cytokines. The IFN response can activate natural killer (NK) cells and other innate immune effectors, potentially reducing transduction efficiency. Complement is a major component of the innate immune response to AAV and can be activated through three major pathways: the classical, alternative, and lectin pathways. The classical and alternative pathways are most relevant to gene therapy because they have been implicated in AAV-induced complement activation in both preclinical and clinical studies. However, it is possible that the lectin pathway could also contribute to host responses against AAV, possibly being activated by the abundance of mannosylated N-glycans on AAV capsids or following endothelial injury.

 

Despite low levels of innate immune responses, strong adaptive immune responses against both wtAAV and rAAV have been observed. Adaptive immune responses create immunological memory to specific antigens and can be induced by either the AAV capsid or the transgene for rAAV. The HLA genotypes of AAV recipients are likely critical in determining the outcome of their adaptive responses, evidenced by the observation that children who developed AAV2-related non A-E acute hepatitis during the COVID-19 pandemic were highly enriched in the HLA-DRB1∗04:01 allele.

 

Figure 1: Immunogenic motifs in the AAV virion(Source: Keeler AM et al, Mol Ther., 2025)

Antibody responses to capsids

Seropositivity levels for naturally occurring viruses are found in different populations, and many neutralizing antibodies (Nabs) that have reactivity against AAVs are prevalent in these populations. However, it is unclear how these antibodies may affect gene therapy vectors. One study evaluated Nabs and binding antibodies (Babs) and found a positive correlation between them, but the seroprevalence of capsid-specific Nabs was higher than Babs. The role of Nabs in blocking the expression of transgene has been well modeled in animal studies and observed in patients, but the role of Babs has been unclear. The presence of these antibodies is durable, as high-titer Nabs can still be detected years after AAV gene therapy. One study found that high Nabs against the delivered AAV2 capsid, as well as some cross-reactive Nabs to AAV5 and AAV8, persisted for 15 years after treatment, which limits redosing strategies. These findings are crucial to optimizing AAV therapy strategies.

Seropositive status can exclude patients from being enrolled in clinical trials and/or make redosing ineffective for patients in trials. A major hurdle in the field is the lack of standardization of antibody assays. This makes it hard to compare responses, but it is also a hurdle for seeking approval and optimizing strategies. Research shows that certain domain areas or “antigenic hot spots” have been identified on AAV capsids, including the icosahedral 3-fold axes, the 5-fold axes, and the 2-/5-fold wall. These regions are prone to antibody binding and may be closely associated with the antigenicity of AAV. Therefore, reducing the antigenic hot spots on AAV9 capsids is a potentially impactful area for the development of immune evasion strategies.

T cell responses to capsids

Responses of T cells against the AAV capsid are activated through cross-presentation. These capsid-specific CD8+ T cell responses are dose-dependent but can be controlled by treatment with steroids. Several muscle-directed clinical trials have observed induction of specialized T cells, called T regulatory cells (Tregs), which induce tolerance and allow for long-term stable transgene expression. The immune response against the transgene is associated with the patient’s genetic background and the design of the vector. The CRIM-negative (like Pompe disease) or the use of shortened gene (like the microdystrophin of DMD) is prone to cause anti-drug antibodies (ADAs) and T cell responses. Self-complementary AAV (scAAV) enhances expression but also increases immunogenicity. Specific HLA haplotypes and pro-inflammatory microenvironments (such as in DMD muscle tissue) can further exacerbate this risk. In addition, disease type and immune tolerance status collectively influence therapeutic outcomes. Therefore, individualized assessment of immune risk is of great importance for improving treatment efficacy.

Clinical challenges and optimizing strategies

The immune responses limit the ability to re-dose patients in gene therapies. They can become more complicated due to the dichotomy of T cell responses, which means that T cell responses can be responsible for either clearance or persistence of transgene expression. Factors that affect how the immune system responds include route of injection, AAV capsid selection/design, disease phenotype, HLA status, immunosuppression regimen, and individual differences. Currently, the strategies to reduce AAV’s immunogenicity focus on engineering either the AAV capsid or its gene cargo.

Gene optimization: Hyperactive variants of AAV-delivered transgenes and codon optimization have been investigated as a strategy to reduce vector dose, but it often increases CpG motifs, which increases TLR9 activation. Removing CpGs or adding a TLR9-inhibitory sequence into the AAV genome can decrease the immunogenicity. Blocking transcription from the 3′-ITR can reduce the activation of RNA sensors, and blocking transgene expression in APCs using microRNA de-targeting has been successfully demonstrated to limit adaptive immune responses.

Capsid engineering: Engineering antigenic hotspots can reduce neutralizing antibody recognition, while tissue-specific capsids can enhance targeting efficiency. Tyrosine mutations can boost transgene expression and reduce antigen presentation. Chemical modifications (such as N-acetylgalactosamine ligands and PEGylation) and exosome enveloping can help evade antibody neutralization and induce tolerance. Moreover, exosome-AAV delivery vehicles may even allow for in vivo engineering of lymphocytes, expanding therapeutic potential. These strategies aim to balance transduction efficiency with immune evasion, providing multidimensional solutions for clinical optimization.

The route of injection role in the immune response

Different routes of administration of AAV vectors will lead to distinct immune responses for a variety of reasons (Figure 2). The interplay between tissue-specific immune characteristics and routes of administration significantly impacts the efficacy of gene therapy, necessitating tailored optimization strategies to balance therapeutic benefits and immune risks.

Eye: As an immune-privileged site, subretinal injection can circumvent systemic antibody interference but may trigger local inflammation, such as gene therapy-associated uveitis (GTAU). Intravitreal injection is thought to be more immunogenic due to potential vector dissemination into the bloodstream and access to lymphatic tissues.

Central nervous system: AAV9 could cross the blood-brain barrier (BBB) and target the CNS. However, AAV administration to the CNS results in increased antibody titers in both the serum and CNS, as well as cellular infiltration into the CNS.

Intramuscular: Nabs are minimally able to impact gene therapy vectors. However, intravenous delivery for muscle indications is greatly limited by pre-existing Nabs. T-cell responses exhibit dichotomy: regulatory T cells promote long-term transgene expression, whereas in Duchenne muscular dystrophy, strong T-cell responses against the transgene are observed.

Systemic/liver: The liver microenvironment supports immune tolerance; however, in clinical practice, corticosteroids are often required to suppress transient transgene loss and elevated liver enzymes. To date, three liver-targeted gene therapies have been approved.

Ear: inner ear blood-labyrinth barrier limits systemic immune exposure; however, local macrophages and delivery route–dependent immune responses still exist (with low antibody levels observed after round window injection, whereas posterior semicircular canal administration can trigger cellular infiltration).

 

 

Figure 2: Immune responses in AAV-vectored gene therapy vary by routes of administration
(Source: Keeler AM et al, Mol Ther., 2025)

Immunotoxicities posed by immune responses against rAAV

Immunotoxicities have been observed within days after injection as well as 1 year post-delivery, with early responses within days to weeks after gene therapy being more associated with the innate immune system and events occurring in the weeks to months after treatment being more associated with the adaptive immune system (Figure 3).

Immunotoxicities due to innate immune responses: High-dose intravenous administration, particularly of AAV9, can lead to complement activation, resulting in thrombotic microangiopathy (TMA) and atypical hemolytic uremic syndrome (aHUS). The mechanism involves endothelial damage caused by the membrane attack complex of complement C5b-9. Reported cases are often associated with recent infections or vaccinations. Complement inhibition therapies and immunomodulation (e.g., rituximab) can prevent unwanted immune responses, but patient genetic predispositions (such as complement factor I variants) may exacerbate the risk.

Immunotoxicities due to adaptive immune responses: T-cell responses against the capsid or the transgene can lead to hepatotoxicity, immune-mediated myositis, and myocarditis, with hepatotoxicity often managed with corticosteroids.

Myocarditis and ganglion toxicity: In clinical trials for the treatment of DMD, myocarditis is related to transgene immune responses. Patients generally present with increased troponins and more inflammatory infiltrates. Dorsal root ganglion (DRG) toxicity was observed in one patient in an amyotrophic lateral sclerosis (ALS) clinical trial, but the mechanism remains controversial. Other studies have suggested that non-protein-encoding RNAi constructs can result in DRG toxicity.

Multiple factors: The occurrence of toxicity is closely associated with vector dose, route of administration (intravenous/intrathecal), individual differences, and disease phenotype. Although most toxicities can be mitigated by immunosuppression (e.g., corticosteroids, sirolimus), certain cases (such as DRG damage) remain mechanistically unclear, requiring further distinction between immune and non-immune factors.

 

Figure 3: Timeline of immune responses and immunotoxicity following treatment with AAV

(Source: Keeler AM et al, Mol Ther., 2025)

Management options to mitigate adverse effects associated with immunotoxicities

Several strategies have been undertaken to modulate immune responses to AAV gene therapies both clinically and pre-clinically. There is no current standard immunosuppression regimen for AAV gene therapy, but several factors are considered when designing an appropriate immunosuppression strategy.

Current strategies: Corticosteroids such as prednisolone remain the cornerstone, combined with rapamycin, tacrolimus, or rituximab (anti-CD20) to suppress B/T cell responses, thereby alleviating hepatotoxicity and complement-mediated TMA/aHUS (e.g., triple immune suppression regimen reducing antibody generation). Complement inhibitors (such as eculizumab, ravulizumab, APL-9) act by targeting C5 or C3 to block the formation of the membrane attack complex, thereby reducing the risk of endothelial damage.

Experimental strategies: Focused on overcoming pre-existing immunity and redosing. 1)Antibody clearance: Plasma exchange combined with immunoadsorption; bacterial enzymes (Imlifidase, IdeS; IceMG) cleaving IgG/IgM; FcRn inhibitors shortening antibody half-life. 2)B-cell targeting: Rituximab combined with BAFF inhibitors or ibrutinib (BTK inhibitor) to deplete B cells, synergized with MyD88 signaling inhibition to reduce antibody production. 3)Innate immune modulation: TLR9 inhibitory oligonucleotide sequence or ImmTOR (nanoparticles containing rapamycin) to induce tolerance, achieving multiple redosing in animal models.

Induction of tolerance: Engineering Tregs (e.g., capsid Tregitopes, CAR/TRuC-Tregs) or infusion of polyclonal Tregs to suppress anti-capsid/transgene immune responses. In hemophilia A mouse models, TRuC Tregs were more effective than CAR Tregs in suppressing anti-drug antibody (ADA) formation; B cell antibody receptors (BARs) have also been evaluated in the context of Tregs by transducing Tregs with BAR constructs, which have shown success at tolerizing against hemophilia A ADA development.

 

Conclusions

AAV is sometimes referred to as “Almost A Virus” because it does not possess many common properties of viruses. The innate immune response to high viral load administered in clinical applications is surprisingly mild, yet enough to robustly activate the adaptive immune response. As vector dosages have increased in clinical trials, so have the immunotoxicities related to AAV gene therapy. Beyond the safety dose threshold, the risk of immunotoxicity increases. Moving forward, collaborative efforts among multiple disciplines will be required to optimize strategies, elucidate molecular mechanisms mediating virus–host interactions, develop immune modulation technologies, and enhance cell-specific targeting efficiency to achieve dose reduction while balancing efficacy and safety. These directions aim to minimize immune rejection, enable redosing, and advance the development of safer and more effective gene therapies.