Impact of AAV Capsid-Specific T-Cell Responses on Design and Outcome of Clinical Gene Transfer Trials with Recombinant Adeno-Associated Viral Vectors: An Evolving Controversy

Abstract

Recombinant adenovirus-associated (rAAV) vectors due to their ease of construction, wide tissue tropism, and lack of pathogenicity remain at the forefront for long-term gene replacement therapy. In spite of very encouraging preclinical results, clinical trials were initially unsuccessful; expression of the rAAV vector-delivered therapeutic protein was transient. Loss of expression was linked to an expansion of AAV capsid-specific T-cell responses, leading to the hypothesis that rAAV vectors recall pre-existing memory T cells that had been induced by natural infections with AAV together with a helper virus. Although this was hotly debated at first, AAV capsid-specific T-cell responses were observed in several gene transfer trials that used high doses of rAAV vectors. Subsequent trials designed to circumvent these T-cell responses through the use of immunosuppressive drugs, rAAV vectors based on rare serotypes, or modified to allow for therapeutic levels of the transgene product at low, non-immunogenic vector doses are now successful in correcting debilitating diseases.

Introduction

Gene transfer remains the treatment of choice to correct permanently the phenotype of diseases caused by single gene defects, as well as other acquired diseases. Recombinant adenovirus-associated (rAAV) vectors due to their ease of construction, wide tissue tropism, and lack of pathogenicity remain at the forefront as delivery vehicles for long-term gene replacement therapy.^1,2 Transfer of rAAV vectors to experimental animals, including mice,^3,4 dogs,^5,6 and nonhuman primates^7,8 using a variety of AAV serotypes at different doses and applied by different routes, results in general in sustained gene transfer. Nevertheless, in initial clinical trials, gene transfer by rAAV vectors at doses that achieved therapeutic levels of the transgene product were not successful⁹ unless vectors were applied to immune-privileged sites.^10

–14 The initial trial that allowed for continued precise monitoring of the transgene product was designed for treatment of hemophilia B through rAAV2 vectors expressing coagulation factor (F)IX.⁹ In this trial, transfer of a high dose of an rAAV2-FIX vector to hepatocytes resulted in expression of sufficient levels of FIX to offer therapeutic benefits. Nevertheless, the initial rise in FIX levels was followed by a self-limited, asymptomatic increase in liver transaminases and a gradual FIX decline back to baseline.⁹ The next patient given a reduced dose of rAAV2-FIX vector again showed transaminitis, although he did not show an increase in FIX levels. In this patient, peripheral blood mononuclear cells (PBMCs) collected at different times after rAAV2-FIX vector transfer were probed for T-cell responses to FIX and the AAV capsid antigens. CD8⁺ T-cell responses to AAV capsid were shown to develop concomitantly with the increases in transaminases.¹⁵ T-cell responses to the transgene product were not observed.

These results were interpreted as evidence that in humans, rAAV vectors induce a recall response of AAV capsid-specific CD8⁺ memory T cells that had been primed during childhood due to natural infections with AAV and a helper virus. Although many hotly debated this hypothesis, it was substantiated in subsequent clinical trials. Acknowledging that in humans rAAV-vector-induced T-cell responses could terminate transgene expression influenced the design of subsequent trials, which by now in some patients are achieving correction of phenotypes of debilitating diseases such as hemophilia B.¹⁶

AAV Biology

AAVs are dependoviruses of the Parvovirideae family. They are single-stranded DNA viruses with an approximately 4.7 kb genome that is flanked by inverted terminal repeats (ITR). The genome is composed of two genes: rep and cap. The cap transcript is spliced to generate the three proteins VP1, VP2, and VP3, which at a ratio of 1:1:10 form the outer capsid of the virion. Multiple serotypes of AAV, which differ in use of cell entry receptors and tropism, have been isolated from cell lines,¹⁷ humans,^18
–20 nonhuman primates,²¹ and other species.^22,23 AAV infects most primates, including humans during childhood, and many adults carry AAV-specific CD8⁺ and CD4⁺ T cells^24,25 as well as AAV neutralizing antibodies.^26,27 Infections occur only in the context of a helper virus, most commonly adenovirus, and are not known to cause disease. A recent report that linked AAV to hepatocellular carcinoma²⁸ was dismissed by experts in the field,^29,30 and according to the authors was not meant to establish a link between rAAV gene transfer and increased risk of liver cancer.³¹ Upon infection, AAV persists either episomally or upon integration into the host cell genome.^32
–34 Reintroduction of a helper virus or application of a genotoxic reagent can reactivate AAV replication.³²

AAV Vectors

The genome of AAV vectors is completely gutted, which means that all the viral sequences but for the ITRs are provided during production in trans. The transgene expression cassette is inserted between the ITRs. The rep and cap genes together with essential helper genes are provided in trans during vector production.

Viral gene therapy vectors have been generated from numerous AAV serotypes. They show distinct tropism in experimental animals.^35,36 Initial studies in mice showed that, for example, rAAV2 and 8 vectors have high tropism for skeletal and heart muscle, rAAV4, 5, 6 and 9 vectors preferentially transduce lung while rAAV7, 8, and 9 vectors are best suited for transduction of liver.³⁷ Mouse data may not necessarily predict rAAV vector tropism in humans, as was shown recently in mice with humanized livers.^38,39 In these mice, rAAV8 and 9 vectors showed higher transduction of mouse hepatocytes compared with a capsid modified AAV3 vector, which in turn was superior in transducing human hepatocytes.³⁹

Vectors have been bioengineered to increase transduction, change vector tropism, or reduce vector degradation. Chimeric vectors were generated by DNA shuffling.⁴⁰ Mutation of exposed tyrosine residues, which upon phosphorylation target the vector for ubiquination and degradation, have resulted in enhanced nuclear transport and increased transgene product expression.^41,42 The AAV vector genome has been modified to improve transgene product expression or reduce innate immune responses by 5′-C-phosphate-G-3′ (CpG)-depletion.⁴³ The vector genome can be single stranded or it can be designed into a self-complementary, double-stranded DNA, which reduces the vector's packaging capacity from about 4.8 kb down to about 2.4 kb but allows for more rapid transgene product expression.^44,45

Immune Responses to AAV and AAV Vectors

Innate responses

Innate immune responses are a prerequisite for stimulation of primary T-cell responses. The inflammatory response of the innate immune system activates specialized antigen-presenting cells (i.e., dendritic cells), causing upregulation of co-stimulatory molecules. Co-stimulation is essential to allow for full activation of T cells once their receptors engage the antigen, which is displayed in the form of peptides on major histocompatibility (MHC) antigens. Display of antigen by non-activated dendritic cells fails to induce an effector T-cell response but rather causes T-cell unresponsiveness. It was initially thought that AAV and rAAV vectors fail to induce an inflammatory response. Subsequent studies showed that CpG sequences present in the vectors' genome trigger an innate response upon engagement to Toll-like receptor (TLR)-9.^46
–48 This response is more pronounced with double-stranded rAAV vectors.⁴⁸ A crucial role for the rAAV vectors' CpG-content in driving transgene product- and capsid-specific T-cell responses was further confirmed by preclinical trials with a highly immunogenic rAAV vector called rAAVrh32.33. This vector mimics in mice the destructive immune responses toward AAV observed in human clinical trial participants of rAAV-mediated, liver-directed gene transfer. TLR-9 signaling was identified as a critical element in rAAVrh32.33-associated immunoreactivity causing loss of transgene expression following muscle gene transfer. rAAVrh32.33 vectors from which CpG-motifs had been depleted, except for some of those in the ITRs, escaped innate and adaptive immune responses, and exhibited persistent transgene expression in mice following muscle gene transfer.⁴³ In addition, human liver cells, specifically Kupffer cells and liver sinusoidal endothelial cells, were shown to mount an inflammatory reaction due to interactions between rAAV vectors and TLR-2.⁴⁹

In summary, although innate immune responses to rAAV vectors are rather weak when compared with the potent responses induced by, for example, adenovirus vector, they are sufficient to support activation of adaptive responses.

Adaptive immune responses

Responses to natural infection

AAV acquired by natural infections induces adaptive immune responses, including CD4⁺ and CD8⁺ T cells and neutralizing antibodies. Even low titers of AAV neutralizing antibodies, which to some serotypes and in some geographic regions can be found in up to 50% of the adult human population,^26,27 strongly reduce transduction when rAAV vectors are injected systemically.⁵⁰ This can be overcome in part by adding empty AAV capsid to the vector preparation as decoys.⁵¹ AAV capsid-specific T cells, which are highly cross-reactive between different serotypes,^52,53 are even more prevalent. By multicolor flow cytometry, 50% of human adults residing in the United States have circulating AAV capsid-specific CD8⁺ and CD4⁺ T cells, which largely represent central memory T cells, although more activated effector or effector memory T cells can also be detected.^24,54 The presence of activated AAV capsid-specific T cells in the blood of humans may either reflect repeated infections or occasional boosts of T-cell responses by AAV reactivation. With more sensitive methods such Tetramer-Associated Magnetic Enrichment, AAV capsid-specific CD8⁺ T cells could be detected in all healthy adults at frequencies of about 1 in 10⁴–10⁶ of all circulating CD8⁺ T cells.²⁵

One of the puzzles of AAV-mediated gene transfer was that it resulted in stable gene expression in nonhuman primates, but using very similar vectors elicited a destructive T-cell response in humans. The hypothesis that this was due to reactivation of memory CD8⁺ T cells in humans was obviously at odds with the sustained gene expression in nonhuman primates, as they are also naturally infected with AAV as juveniles and in fact have higher frequencies of AAV capsid-specific CD8⁺ and CD4⁺ T cells in the blood compared with humans. Nevertheless, in nonhuman primates, most AAV capsid-specific T cells belong to the highly activated effector cell subset.⁵⁴ As already mentioned, AAVs acquired by natural infections persist and may become reactivated in the presence of a helper virus. Nonhuman primates are more permissive to a high level of adenovirus persistence and constantly shed substantial amounts of this virus.⁵⁵ This suggests that the high levels of persisting adenovirus may also facilitate continued replication of AAV, which may maintain high frequencies of activated T cells. Highly activated T cells respond differently to an external boost than the more resting central memory T cells in humans.⁵⁶

Responses to AAV vectors

rAAV-mediated gene transfer causes a rise in AAV neutralizing antibodies.⁹ In some but not all patients, this is even observed if vectors are given to immune-privileged sites such as the central nervous system (CNS),^57
–59 presumably reflecting that some vectors inevitably leak into the vasculature. In humans, as was first shown with a rAAV2-FIX vector, high doses increase frequencies of AAV capsid-specific CD4⁺ and CD8⁺ T cells, which in some,^9,60 but not all, cases⁶¹ was linked to loss of transgene product expression.

The Debate

The hypothesis that rAAV vector-mediated gene transfer elicits recall T-cell responses that can eliminate transduced cells was initially met with substantial skepticism. One argument was that AAV vectors achieved sustained gene transfer in preclinical trials and could thus not possibly induce T-cell responses. However, no one had really looked, and nor had anyone at the time been compelled to analyze human T-cell responses to a virus that does not cause a human disease. Some rAAV vectors such as those based on AAV2 were shown to elicit capsid-specific CD8⁺ T cells in mice, but developing mouse models that recapitulated the clinical finding turned out to be far from straightforward.^62

–66 Others argued and provided evidence in nonhuman primate models that only rAAV vectors based on serotypes, such as AAV2, that bind heparan sulfate proteoglycan (HSPG) would induce capsid-specific T-cell responses, while others, such as those based on AAV8, which does not interact with HSPG, would remain immunologically silent.⁶⁷ A subsequent clinical trial with a rAAV8 vector disproved this concept,⁶⁰ since four out of six subjects infused with the highest vector dose developed transaminase elevation, increases in AAV8 capsid-specific circulating T cells, and a drop in FIX levels.⁶⁰ Some suggested that contaminations with rep/cap plasmid used in vector production elicited the response, but considering the rAAV vector purification protocol, this was highly unlikely.⁶⁸ For immunologists, two aspects of the human T-cell responses to AAV capsid following gene transfer were puzzling. First, the capsid antigens are provided in trans. That means only antigen present in the capsid at the time of injection can elicit an immune response and that amount is limited and declines over time. The AAV capsid degrades rather slowly so that the number of epitopic peptides displayed by MHC molecules on transduced cells would be very low at any given time. Nevertheless, T cells, once activated, are exquisitely sensitive. Fewer than 10 copies of MHC-peptide complexes are needed to trigger a response.⁶⁹ Second, particular antigen readily triggers CD4⁺ T- and B-cell responses, but CD8⁺ T cells are best induced by de novo synthesized polypeptides that due to incorrect synthesis or faulty folding are degraded by a proteasome complex in the cytoplasm. The resulting peptides are then actively transported into the endoplasmic reticulum where they bind to MHC class I molecules. Peptide binding stabilizes the MHC class I molecules and allows for their transfer to the cell surface where they can be recognized by CD8⁺ T cells. A process called cross-presentation permits the induction of CD8⁺ T cells to viruses that do not readily infect antigen-presenting cells.⁷⁰ In this process, antigen that is taken up by dendritic cells is diverted out of the endosomes into the cytoplasm where it is then treated like an endogenously synthesized faulty polypeptide. Several studies provided evidence that indeed the capsid antigens of AAV vectors elicit responses upon cross-presentation.^71,72

Clinical trials for AAV-FIX transfer failed to show an increase in transgene product-specific T cells.¹⁵ The argument was made that cryptic epitopes within the F9 gene triggered a T-cell response in the rAAV2-FIX trial.⁷³ This T-cell response would not have been detected using peptides representing the sequence of FIX, but nevertheless could have caused the decline in FIX expression. Again, this idea could be laid to rest experimentally by testing for T cells to peptides that would have been encoded by an alternate open reading frame—such responses were not detected.⁶⁰

The Exceptions

Not all AAV vector recipients enrolled into clinical trials that addressed a number of diseases showed increases in AAV capsid-specific circulating T cells, or in some cases, such increases failed to cause a loss of transgene expression.^60,61 The first sustained clinical benefits upon AAV-mediated gene transfer were reported for treatment of Leber's congenital amaurosis by replacement of the RPE65 gene^10,74,75 and then choroideremia by replacement of the REP gene.⁷⁶ In both cases, rAAV vectors were transferred to the subretinal space of the eye, and vector doses were well below those that were needed to elicit a T-cell response upon systemic transfer. Similarly, injection of rAAV vectors intraparenchymally into the CNS was not linked to destructive T-cell responses, as evidenced by a number of clinical trials. One was designed for treatment of Batten's disease, a lysosomal storage disorder in which tripetidyl peptidase 1 was delivered by AAV2 or AAVrh10 vectors.⁵⁹ In another trial, Canavan disease, which causes degeneration of myelin and progressive brain atrophy, was treated with rAAV2 vectors to replace amyloacylase 2.⁵⁸ In adults with Parkinson's disease rAAV-2 vectors were used to deliver neutrophin,⁷⁷ aromatic amino acid decarboxylase (AADC),⁷⁸ or glutamic acid decarboxylase (GAD).⁷⁹ The CNS is an immune-privileged site, which lacks traditional antigen-presenting cells and does not mount a robust inflammatory response upon introduction of antigens. Antigen delivered to the CNS thus in general fails to elicit an adaptive immune response. Further protection is provided by the blood–brain barrier, which shields the CNS from the systemic immune system and prevents entrance of antibodies or resting lymphocytes, although it can readily be crossed by activated T cells. Induction of AAV capsid-specific antibodies following rAAV gene transfer into the brain suggests that some vector may have leaked from the injection site into the blood vessels. The amount that in the end became accessible to the immune system would most likely not be enough for induction of T-cell recall responses. Nevertheless, potential increases in AAV capsid-specific T-cell responses upon injection of rAAV vectors to immune-privileged sites should be monitored to avoid equating lack of testing with lack of a T-cell response.

Treatment of α1 antitrypsin (AAT) deficiency with a rAAV1 vector resulted in a detectable increase in AAV capsid-specific T cells and in one patient in a transient response to the transgene product.⁸⁰ Unlike in the hemophilia B trial, there was no relationship between vector dose and magnitude of the AAV capsid-specific T-cell response in the AAT trial. In some patients, onset of the AAV capsid-specific T-cell response, which was initially seen 1 month after vector administration, was associated with an increase in serum creatine kinase and a drop in circulating AAT. Nevertheless, although the AAV capsid-specific T-cell response was remarkably sustained and remained detectable for at least another 2 months, AAT levels did not decline to baseline. In a follow-up study, the group reported persistent AAT expression in muscle after 1 year, accompanied by a cellular infiltrate that contained ∼10% natural regulatory T cells which may have prevented destruction of the transduced myofibers.⁸¹

It is possible that T cells induced by intramuscular injection of rAAV vectors are fundamentally different from those induced by systemic injection. In most clinical trials, AAV capsid-specific T-cell responses to rAAV gene transfer are tracked by an ELISpot assay, which screens for T cells that produce interferon-(IFN)-γ in response to peptide pools that reflect the AAV capsid amino acid sequence. This assay does not give insight into other key functions of AAV capsid-specific T cells. Thus, it is not known if AAV capsid-specific T cells induced by rAAV vectors given by different routes are functionally similar or distinct. It is also possible that hepatocytes targeted for treatment of hemophilia B more easily become victims to lytic AAV capsid-specific CD8⁺ T cells than muscle cells do. In order to be targeted for CD8⁺ T lymphocyte-mediated cytolysis, rAAV-transduced cells must express sufficient amounts of AAV capsid-derived peptides displayed by MHC class I molecules on the cell surface. Low levels of MHC class I molecules or delayed degradation of the AAV capsid may prevent effective recognition of transduced cells by CD8⁺ T cells. Healthy muscle cells express only low amounts of MHC class I antigens,⁸² although, and this is unknown, they may increase MHC class l expression upon local rAAV vector injection due to interferons released by inflammatory cells.⁸³ T-cell responses can be subverted by the expression of co-inhibitory ligands such as PD-L1, which is induced on muscle cells in response to inflammation.⁸⁴ In addition, the transgene product may influence AAV capsid-specific T-cell responses. AAT is known to be anti-inflammatory, it delays allograft rejection, and it increases activation of regulatory T cells,^85
–87 all of which may have prevented the loss of AAT-producing muscle cells.

The Consequences

Monitoring of immune responses

As an immediate result of the informative but in the end unsuccessful rAAV2-FIX trial,⁹ other investigators started to monitor AAV capsid- and transgene product-specific T-cell responses after rAAV gene transfer. Results were fairly consistent—patients who received low doses of rAAV vectors failed to show detectable increases in AAV capsid-specific T cells, which were seen in individuals treated systemically or intramuscularly with doses above ∼5 × 10¹¹ vg/kg. As already mentioned, in most trials, T-cell responses were measured by an ELISpot assay for IFN-γ. This assay is very robust and can be easily standardized, but it has some disadvantages. First, it does not distinguish between CD4⁺ and CD8⁺ T-cell responses, although this can be achieved by depleting or enriching for one of these subsets. Second, it requires high numbers of cells, so it is not practical for use in small infants who do not tolerate large bleeds. Third, it underestimates T-cell responses by missing T cells that produce effector functions other than IFN-γ. Elucidating additional functions, such as the production of the lytic enzymes perforin and granzyme B, would provide essential information on the T cell's ability to eliminate their target cells. Furthermore, ELISpot assays do not provide any information on T-cell fitness such as their ability to proliferate or their activation/exhaustion status. Some of this information can be gained by the use of multicolor flow cytometry, which allows the analysis of up to 16 parameters on any given cell.⁸⁸ Even more attractive would be the use of mass cytometry, which combines time-of-flight mass spectrometry with the detection of rare metal-labeled antibodies. This technique, which unlike flow cytometry is not hampered by spectral overlap of fluorochromes, can analyze simultaneously >40 different parameters on a single cell⁸⁹ and might thus be ideal to track T-cell responses in rAAV gene transfer recipients.

Use of alternative AAV serotypes

T and B cells are highly cross-reactive among most AAV serotypes, as was initially shown in mice⁵² and then with human PBMCs.⁵³ An exception is AAV5, as the sequence of this serotypes is divergent from that of AAV2 and other serotypes that have been used in the clinic.⁹⁰ The prevalence of pre-existing neutralizing antibodies to AAV5 is comparatively low.⁹¹ rAAV5 vectors were thus used to overcome pre-existing neutralizing antibodies in a clinical trial for treatment of acute intermittent porphyria.⁹² Vectors expressing pre-uroporphyrinogen synthase (PBGD) under a hepatocyte-specific promoter were given intravenously at doses ranging from 5 × 10¹¹ vg/kg to 1.8 × 10¹³ vg/kg. In spite of these high doses, patients failed to show increases in AAV capsid- or transgene product-specific CD8⁺ T cells. Clinical benefits were overall modest and not related to the vector dose. Subsequent studies showed that rAAV5 has poor tropism for human liver cells, at least in a humanized mouse model,³⁹ which may explain the lack of more pronounced clinical benefits. Can a lack of hepatocyte transduction explain the absence of an AAV capsid-specific T-cell response? It is assumed—and there is some experimental evidence from animal studies—that the AAV capsid is presented by dendritic cells or other antigen-presenting cells to the immune system,⁶⁴ and this should not be affected by the rAAV vector's tropism for hepatocytes. The use of an AAV vector to which most humans lack pre-existing immunity may thus indeed have prevented a T-cell recall response and thus immune-mediated destruction of transduced cells.

Nevertheless, there are some caveats. In another hemophilia B trial, patients were given an rAAV5 vector-expressing FIX under a liver-specific promoter. One patient showed an elevation in alanine aminotransferase (ALT) and was treated with prednisolone. One patient showed at one time point only a rise in AAV capsid-specific T cells. In four out of five patients, treatment effected sustained FIX expression between 3.1% and 6.7% of normal, and patients could discontinue FIX prophylaxis.⁹³ These results show that at least one patient mounted a T-cell response to rAAV5 and thus presumably had pre-existing immunity that could have been induced by a different yet cross-reactive serotype.

An AAV5 vector expressing FVIII was in addition used in a clinical trial in hemophilia A. In a Phase 1/2 trial, vector was given at doses ranging from 6 × 10¹² vg/kg to 6 × 10¹³ vg/kg. At the highest dose, which resulted in an average FVIII activity of about 160 IU/dL, moderate increases in ALT were observed, and patients were treated with steroids. Currently, no information is available from this trial on AAV5 capsid- or transgene product-specific T-cell responses.⁹⁴

Overall, these results indicate that T-cell responses to capsid antigens of a rare AAV serotype, such as 5, might indeed be less common than to other serotypes, such as AAV2 or 8. Nevertheless, investigators should be strongly encouraged to check for AAV capsid-specific T-cell responses in human rAAV5 recipients so that one can base this conclusion on a more comprehensive database.

Use of CpG-modified vectors

In preclinical studies, activation of innate responses through CpG-triggered TLR-9 signaling was essential to stimulate AAV capsid-specific CD8⁺ T cells,⁴³ indicating that depletion of CpG motifs in rAAV vectors for human subjects was prudent. In the first successful clinical trial for hemophilia B that achieved sustained therapeutic levels of FIX, a self-complementary rAAV8-FIX vector that had been modified to reduce the vector genome's CpG-content was given systemically.⁹⁵ At high doses, the treatments restored FIX to levels that significantly reduced spontaneous bleeds. The six patients who received the highest rAAV vector dose also achieved the highest peak levels of circulating FIX (>10% of normal). Following gene transfer, four of these patients developed increases in AAV-capsid-specific T-cell responses that were accompanied by increases in transaminases and decreases in FIX at approximately 8–10 weeks after treatment. Magnitude, kinetics, and dose-dependency of AAV capsid-specific T-cell responses were similar to those in the earlier trial with a single-stranded rAAV2-FIX vector. These results fail to support the notion that a reduction in the rAAV vectors' CpG-content will eliminate recall of AAV capsid-specific T-cell responses. The discrepancy in the effect of modulating the CpG content of the AAV vectors' genome between experimental animals and humans may reflect differences in innate sensing of pathogen-associated molecular patterns between different species. Alternatively, or in addition, the recall T-cell responses in humans may be less dependent on inflammatory responses than the primary T-cell responses that were assessed in experimental animals to compare CpG-containing with CpG-depleted rAAV vectors.

Use of immunosuppressive drugs

Numerous drugs are available to suppress unwanted immune responses such as those causing auto-immunity or organ rejection following transplantation. Several clinical rAAV gene transfer trials used immunosuppressive drugs with varied success. In the above-mentioned trial with the self-complementary rAAV8-FIX vector,^60,95 four out of six patients in the high-dose group (2 × 10¹² vg/kg) showed increases in ALT levels between 7 and 10 weeks after vector infusion. Patients with ALT increases were immediately placed on tapering doses of prednisolone, which rapidly reduced ALT levels back to normal. ALT increases were not observed in the low (2 × 10¹¹ vg/kg) or intermediate (6 × 10¹¹ vg/kg) dose groups. AAV capsid-specific T-cell responses were not detected in the low-dose group but were detected in both the intermediate- and high-dose groups. Elevations in AAV capsid-specific T cells were thus not linked to ALT increases in all patients. In some patients, frequencies of AAV capsid-specific T cells remained elevated for >1 year, while the transaminase increases were more transient. Although circulating FIX decreased over time, most patients maintained levels above baseline and reported reduced reliance on factor replacement therapy. It should be noted that one patient who never showed any increases in ALT or AAV capsid-specific T cells also showed very stable FIX levels.

In a subsequent trial conducted by the Children's Hospital of Philadelphia (CHOP) with a codon-optimized single-stranded AAV8-FIX vector (NCT01620801), three subjects were infused at doses of 1–2 × 10¹² vg/kg. One subject, a 67-year-old male, showed increases in liver enzymes and AAV-capsid specific T cells by day 42 after transfusion. He responded well initially to a tapering course of steroids, and his factor usage was reduced from 50 infusions in the year prior to gene transfer to one infusion in the first year after vector administration. Two younger individuals developed robust T-cell responses by day 30 following treatment. In spite of a tapering course of steroids, the FIX levels decreased to baseline by day 60.

In a trial in which a rAAV1 vector expressing a naturally occurring variant of LPL was used intramuscularly at 1 × 10¹² vg/kg to treat individuals with lipoprotein lipase deficiency, all human subjects were treated with immunosuppressants starting before vector administration. Specifically, they received cyclosporine A and mycophenolate mofetil for 12 weeks and a bolus injection of methylprednisolone 30 min before injection of the rAAV vector. All patients developed transient increases in AAV1 capsid-specific CD8⁺ T cells following gene transfer. In one patient, this response was not detected until 52 weeks after injection of the rAAV1 vector. Muscle biopsies taken within a few weeks after transfer showed cellular infiltrates composed of macrophages, T and B cells. In one patient, who showed evidence of a polymyositis, CD8⁺ T cells were positive for granzyme B and Fas-ligand; in the other patients, they were negative for these markers. Lack of lytic factors in infiltrating CD8⁺ T cells was linked to the presence of regulatory FoxP3⁺CD4⁺ T cells. Subjects showed long-term expression of the transgene product and an improvement in plasma lipoprotein clearance.⁹⁶

Steroids were also used in the BAX 225 trial, which used a self-complementary rAAV8 vector expressing a high specific activity, codon-optimized F9 gene. The first two patients of the highest dose group (3 × 10¹² vg/kg) showed initially robust FIX expression of ∼60% of normal but then developed increases in ALT levels accompanied by immune responses. Although both patients were treated with prednisone, FIX levels continued to decline.⁹⁷

Overall, these results show that immunosuppression administered with AAV vectors appears to be safe and has value in dampening T-cell responses to AAV capsid and thus preserving transgene product expression. Unfortunately, immunosuppression does not completely block AAV capsid-specific T-cell responses and is not successful in all patients, which may in part reflect timing of treatment in relation to reactivation of AAV capsid-specific T-cell responses. Other aspects of vector design and manufacture may also affect response to immunosuppression. Additional studies are needed to optimize the immunosuppressive regimens for AAV-mediated gene transfer and to assess further their effect not just on kinetics and magnitude of T-cell responses, but also on other characteristics such as the T cells' lytic potential, their proliferation capacity, and their differentiation status.

Dose-sparing through vector modifications

One of the most consistent observations across numerous trials that tested for cellular immune responses in patients who received rAAV vectors either intravenously or intramuscularly was that AAV capsid-specific T cells were only recalled at high vector doses. Improving transduction rates through capsid modifications, levels of transgene product expression through enhanced transcription or superior biological activity of the transgene product, for example by using the gain-of-function FIX Padua mutant,⁹⁸ may all allow for reduction in rAAV vector dose to levels that escape immune detection. The Padua variant carries a single point mutation, R338L, that results in around a four- to eightfold higher specific activity compared with wild-type FIX, as measured in standard coagulation assays.

In a hemophilia B trial that used a bioengineered hepatotropic capsid expressing FIX-Padua, nine hemophilia B subjects infused at a dose of 5 × 10¹¹ vg/kg all developed therapeutic FIX levels ranging from 25% to 35% of normal. Two required immunosuppression with a tapering course of steroids following development of rising transaminases and falling FIX levels. Larger numbers of subjects must be studied to determine whether lower doses consistently result in a lower frequency of immune responses.¹⁶

Clinical trials have used rAAV vectors produced by a baculovirus expression system in Sf9 cells.⁹⁹ rAAV particles from insect cells compared with those from mammalian cells have lower levels of VP1,¹⁰⁰ the capsid protein that is essential to allow AAV's escape from endosomes.¹⁰¹ Baculovirus-derived rAAVs thus have to be used at higher doses to achieve the transduction rates of rAAVs from mammalian cells, which may increase their immunogenicity. Improved production procedures developed for baculovirus-derived rAAV5 may address this potential problem.¹⁰²

Summary

rAAV-mediated gene transfer has made tremendous progress in recent years and has by now generated long-lasting FIX expression in several individuals with hemophilia B. T-cell responses to AAV capsid still present a challenge, although there is ample experimental evidence that they can be circumvented. Immunosuppression is one way, but not only does it have side effects, but its ability to prevent loss of transgene product expression reliably has been capricious. Better and more reliable drug regimens have to be developed, and screening procedures to allow for their timely use have to be improved. The use of vectors such as rAAV5 to which humans lack pre-existing immunity is an option, but its usefulness for specific applications will be dictated by its tropism. Furthermore, it should be pointed out that although AAV5 is highly divergent from most other AAV serotypes that have been used as vectors, there is still substantial homology. Considering the diversity of HLA class I molecules, it is thus likely that a subset of humans will encounter T-cell epitopes on the AAV5 capsid that cross-react with those of more common AAV strains. Improving transduction rates and the biological activity of the therapeutic transgene product is currently the most elegant option, as it allows for the use of rAAV doses that are below the threshold needed to reactivate T-cell responses.

Footnotes

Author Disclosure

H.C.J.E. is the recipient of sponsored research contracts from Pfizer and Spark. K.A.H. is an employee of and holds equity in Spark Therapeutics.

References

Flotte

, Carter

. Adeno-associated virus vectors for gene therapy. Gene Ther, 1995; 2:357–362.

Walter

, High

. Gene therapy for the hemophilias. Adv Vet Med, 1997; 40:119–134.

Liu

, Mingozzi

, Rodriguéz-Colôn

, et al. Therapeutic levels of factor IX expression using a muscle-specific promoter and adeno-associated virus serotype 1 vector. Hum Gene Ther, 2004; 15:783–792.

Fisher

, Jooss

, Alston

, et al. Recombinant adeno-associated virus for muscle directed gene therapy. Nat Med, 1997; 3:306–312.

Niemeyer

, Herzog

, Mount

, et al. Long-term correction of inhibitor-prone hemophilia B dogs treated with liver-directed AAV2-mediated factor IX gene therapy. Blood, 2009; 113:797–806.

Marcos-Contreras

, Smith

, Bellinger

, et al. Sustained correction of FVII deficiency in dogs using AAV-mediated expression of zymogen FVII. Blood, 2016; 127:565–571.

Gao

, Lu

, Sun

, Johnston

, et al. High-level transgene expression in nonhuman primate liver with novel adeno-associated virus serotypes containing self-complementary genomes. J Virol, 2006; 80:6192–6194.

Wang

, Bell

, Lin

, et al. AAV8-mediated hepatic gene transfer in infant rhesus monkeys (Macaca mulatta). Mol Ther, 2011; 19:2012–2020.

Manno

, Pierce

, Arruda

, et al. Successful transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by the host immune response. Nat Med, 2006; 12:342–347.

10.

Bennett

, Wellman

, Marshall

, et al. Safety and durability of effect of contralateral-eye administration of AAV2 gene therapy in patients with childhood-onset blindness caused by RPE65 mutations: a follow-on phase 1 trial. Lancet, 2016; 388:661–672.

11.

Pierce

, Bennett

. The status of RPE65 gene therapy trials: safety and efficacy. Cold Spring Harb Perspect Med, 2015; 5:a017285.

12.

Leone

, Janson

, Bilaniuk

, et al. Aspartoacylase gene transfer to the mammalian central nervous system with therapeutic implications for Canavan disease. Ann Neurol, 2000; 48:27–38.

13.

Kaplitt

, Feigin

, Tang

, et al. Safety and tolerability of gene therapy with an adeno-associated virus (AAV) borne GAD gene for Parkinson's disease: an open label, Phase I trial. Lancet, 2007; 369:2097–2105.

14.

Vendomele

, Khebizi

, Mormin

, et al. Anti-transgene cellular immune responses can be induced by subretinal gene transfer with rAAV in a dose-dependent manner. ASCGT meeting 2016 Session Immunological Aspects of Gene Therapy II: AAV Vectors, Abstract 694.

15.

Mingozzi

, Maus

, Hui

, et al. CD8(+) T-cell responses to adeno-associated virus capsid in humans. Nat Med, 2007; 13:419–422.

16.

George

, Sullivan

, Teitel

, et al. Preliminary results of a Phase 1/2 trial of SPK-9001, a hyperactive FIX variant delivered by a novel capsid, demonstrate consistent factor IX activity levels at the lowest dose cohort. WHF 2016, Late-Breaking Abstract. Haemophilia, 2016; 22:151–152.

17.

Dutta

. Isolation and characterization of an adenovirus and isolation of its adenovirus-associated virus in cell culture from foals with respiratory tract disease. Am J Vet Res, 1975; 36:247–250.

18.

Grossman

, Mendelson

, Brok-Simoni

, et al. Detection of adeno-associated virus type 2 in human peripheral blood cells. Gen Virol, 1992; 73:961–966.

19.

Walz

, Anisi

, Schlehofer

, et al. Detection of infectious adeno-associated virus particles in human cervical biopsies. Virology, 1998; 247:97–105.

20.

Chen

, Jensen

, Schnepp

, et al. Molecular characterization of adeno-associated viruses infecting children. J Virol, 2005; 79:14781–14792.

21.

Gao

, Alvira

, Somanathan

, et al. Adeno-associated viruses undergo substantial evolution in primates during natural infections. Proc Natl Acad Sci U S A, 2003; 100:6081–6086.

22.

Lochrie

, Tatsuno

, Arbetman

, et al. Adeno-associated virus (AAV) capsid genes isolated from rat and mouse liver genomic DNA define two new AAV species distantly related to AAV-5. Virology, 2006; 353:68–82.

23.

Dutheil

, Yoon-Robarts

, Ward

, et al. Characterization of the mouse adeno-associated virus AAVS1 ortholog. J Virol, 2004; 78:8917–8921.

24.

Parzych

, Li

, Yin

, et al. Effects of immunosuppression on circulating adeno-associated virus capsid-specific T cells in humans. Hum Gene Ther, 2013; 24:431–442.

25.

Vandamme

, Devaux

, Jaulin

, et al. Phenotypic and functional characterisation of human anti-AAV CD8+ T cells using MHC class I tetramer-associated magnetic enrichment. ASCGT meeting, 2016, Session Immunological Aspects of Gene Therapy II: AAV Vectors 697.

26.

Calcedo

, Vandenberghe

, Gao

, et al. Worldwide epidemiology of neutralizing antibodies to adeno-associated viruses. J Infect Dis, 2009; 199:381–390.

27.

Boutin

, Monteilhet

, Veron

, et al. Prevalence of serum IgG and neutralizing factors against adeno-associated virus (AAV) types 1, 2, 5, 6, 8, and 9 in the healthy population: implications for gene therapy using AAV vectors. Hum Gene Ther, 2010; 21:704–712.

28.

Nault

, Datta

, Imbeaud

, et al. Recurrent AAV2-related insertional mutagenesis in human hepatocellular carcinomas. Nat Genet, 2015; 47:1187–1193.

29.

Gil-Farina

, Fronza

, Kaeppel

, et al. Recombinant AAV integration is not associated with hepatic genotoxicity in nonhuman primates and patients. Mol Ther, 2016; 24:1100–1105.

30.

Nault

, Mami

, La Bella

, et al. Wild-type AAV insertions in hepatocellular carcinoma do not inform debate over genotoxicity risk of vectorized AAV. Mol Ther, 2016; 24:660–661.

31.

Schmidt

, Gil-Farina

, Büning

. Reply to “Wild-type AAV insertions in hepatocellular carcinoma do not inform debate over genotoxicity risk of vectorized AAV.”. Mol Ther, 2016; 24:661–662.

32.

Afione

, Conrad

, Kearns

, et al. In vivo model of adeno-associated virus vector persistence and rescue. J Virol, 1996; 70:3235–3241.

33.

Berns

, Linden

. The cryptic life style of adeno-associated virus. Bioessays, 1995; 17:237–245.

34.

Nakai

, Montini

, Fuess

, et al. AAV serotype 2 vectors preferentially integrate into active genes in mice. Nat Genet, 2003; 34:297–302.

35.

Burger

, Gorbatyuk

, Velardo

, et al. Recombinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. Mol Ther, 2004; 10:302–317.

36.

Srivastava

. In vivo tissue-tropism of adeno-associated viral vectors. Curr Opin Virol, 2016; 21:75–80.

37.

Zincarelli

, Soltys

, Rengo

, et al. Analysis of AAV serotypes 1–9 mediated gene expression and tropism in mice after systemic injection. Mol Ther, 2008; 16:1073–1080.

38.

Lisowski

, Dane

, Chu

, et al. Selection and evaluation of clinically relevant AAV variants in a xenograft liver model. Nature, 2014; 506:382–386.

39.

Vercauteren

, Hoffman

, Zolotukhin

, et al. Superior in vivo transduction of human hepatocytes using engineered AAV3 capsid. Mol Ther, 2016; 24:1042–1049.

40.

Grimm

, Lee

, Wang

, et al. In vitro and in vivo gene therapy vector evolution via multispecies interbreeding and retargeting of adeno-associated viruses. J Virol, 2008; 82:5887–5911.

41.

Zhong

, Li

, Mah

, et al. Next generation of adeno-associated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proc Natl Acad Sci U S A, 2008; 105:7827–7832.

42.

Martini

, Silva

, Ferreira

, et al. Tyrosine mutation in AAV9 capsid improves gene transfer to the mouse lung. Cell Physiol Biochem, 2016; 39:544–553.

43.

Faust

, Bell

, Cutler

, et al. CpG-depleted adeno-associated virus vectors evade immune detection. J Clin Invest, 2013; 123:2994–3001.

44.

McCarty

, Monahan

, Samulski

. Self-complementary recombinant adeno-associated virus (scAAV) vectors promote efficient transduction independently of DNA synthesis. Gene Ther, 2001; 8:1248–1254.

45.

Nathwani

, Gray

, Ng

, et al. Self-complementary adeno-associated virus vectors containing a novel liver-specific human factor IX expression cassette enable highly efficient transduction of murine and nonhuman primate liver. Blood, 2006; 10:2653–2661.

46.

Zhu

, Huang

, Yang

. The TLR9-MyD88 pathway is critical for adaptive immune responses to adeno-associated virus gene therapy vectors in mice. J Clin Invest, 2009; 119:2388–2398.

47.

Rogers

, Suzuki

, Zolotukhin

, et al. Unique roles of TLR9- and MyD88-dependent and -independent pathways in adaptive immune responses to AAV-mediated gene transfer. J Innate Immun, 2015; 7:302–314.

48.

Rogers

, Martino

, Aslanidi

, et al. Innate immune responses to AAV vectors. Front Microbiol, 2011; 2:194.

49.

Hösel

, Broxtermann

, Janicki

, et al. Toll-like receptor 2-mediated innate immune response in human nonparenchymal liver cells toward adeno-associated viral vectors. Hepatology, 2012; 55:287–297.

50.

Murphy

, Li

, Zhou

, et al. Prolonged susceptibility to antibody-mediated neutralization for adeno-associated vectors targeted to the liver. Mol Ther, 2008; 16:138–145.

51.

Mingozzi

, Anguela

, Pavani

, et al. Overcoming preexisting humoral immunity to AAV using capsid decoys. Sci Transl Med, 2013; 5:194ra92.

52.

Sabatino

, Mingozzi

, Hui

, et al. Identification of mouse AAV capsid-specific CD8+ T cell epitopes. Mol Ther, 2005; 12:1023–1033.

53.

Hui

, Edmonson

, Podsakoff

, et al. AAV capsid CD8+ T-cell epitopes are highly conserved across AAV serotypes. Mol Ther Methods Clin Dev, 2015; 2:15029.

54.

, Lasaro

, Jia

, et al. Capsid-specific T-cell responses to natural infections with adeno-associated viruses in humans differ from those of nonhuman primates. Mol Ther, 2011; 19:2021–2030.

55.

Roy

, Vandenberghe

, Kryazhimskiy

, et al. Isolation and characterization of adenoviruses persistently shed from the gastrointestinal tract of non-human primates. PLoS Pathog, 2009; 5:e1000503.

56.

Wherry

. T cell exhaustion. Nat Immunol, 2011; 12:492–499.

57.

Sondhi

, Johnson

, Purpura

, et al. Long-term expression and safety of administration of AAVrh.10hCLN2 to the brain of rats and nonhuman primates for the treatment of late infantile neuronal ceroid lipofuscinosis. Hum Gene Ther Methods, 2012; 23:324–335.

58.

McPhee

, Janson

, Li

, et al. Immune responses to AAV in a Phase I study for Canavan disease. J Gene Med, 2006; 8:577–588.

59.

Worgall

, Sondhi

, Hackett

, et al. Treatment of late infantile neuronal ceroid lipofuscinosis by CNS administration of a serotype 2 adeno-associated virus expressing CLN2 cDNA. Hum Gene Ther, 2008; 19:463–474.

60.

Nathwani

, Reiss

, Tuddenham

, et al. Long-term safety and efficacy of factor IX gene therapy in hemophilia B. N Engl J Med, 2014; 371:1994–2004.

61.

Flotte

, Trapnell

, Humphries

, et al. Phase 2 clinical trial of a recombinant adeno-associated viral vector expressing α1-antitrypsin: interim results. Hum Gene Ther, 2011; 22:1239–1247.

62.

, Murphy

, Giles-Davis

, et al. Pre-existing AAV capsid-specific CD8+ T cells are unable to eliminate AAV-transduced hepatocytes. Mol Ther, 2007; 15:792–800.

63.

, Lin

, Giles-Davis

, et al. A preclinical animal model to assess the effect of pre-existing immunity on AAV-mediated gene transfer. Mol Ther, 2009; 17:1215–1224.

64.

, Tuyishime

, Wu

, et al. Adeno-associated virus vectors serotype 2 induce prolonged proliferation of capsid-specific CD8+ T cells in mice. Mol Ther, 2011; 19:536–546.

65.

, Hirsch

, Asokan

, et al. Adeno-associated virus type 2 (AAV2) capsid-specific cytotoxic T lymphocytes eliminate only vector-transduced cells coexpressing the AAV2 capsid in vivo. J Virol, 2007; 81:7540–7547.

66.

Martino

, Basner-Tschakarjan

, Markusic

, et al. Engineered AAV vector minimizes in vivo targeting of transduced hepatocytes by capsid-specific CD8+ T cells. Blood, 2013; 121:2224–2233.

67.

Vandenberghe

, Wang

, Somanathan

, et al. Heparin binding directs activation of T cells against adeno-associated virus serotype 2 capsid. Nat Med, 2006; 12:967–971.

68.

Hauck

, Murphy

, Smith

, et al. Undetectable transcription of cap in a clinical AAV vector: implications for preformed capsid in immune responses. Mol Ther, 2009; 17:144–152.

69.

Sykulev

, Joo

, Vturina

, et al. Evidence that a single peptide–MHC complex on a target cell can elicit a cytolytic T cell response. Immunity, 1996; 4:565–571.

70.

Blander

. The comings and goings of MHC class I molecules herald a new dawn in cross-presentation. Immunol Rev, 2016; 272:65–79.

71.

Wang

, Figueredo

, Calcedo

, et al. Cross-presentation of adeno-associated virus serotype 2 capsids activates cytotoxic T cells but does not render hepatocytes effective cytolytic targets. Hum Gene Ther, 2007; 18:185–194.

72.

, He

, Nicolson

, Hirsch

, et al. Adeno-associated virus capsid antigen presentation is dependent on endosomal escape. J Clin Invest, 2013; 123:1390–1401.

73.

, Goudy

, Hirsch

, et al. Cellular immune response to cryptic epitopes during therapeutic gene transfer. Proc Natl Acad Sci U S A, 2009; 106:10770–10774.

74.

Bainbridge

, Smith

, Barker

, et al. Effect of gene therapy on visual function in Leber's congenital amaurosis. N Engl J Med, 2008; 358:2231–2239.

75.

Maguire

, High

, Auricchio

, et al. Age-dependent effects of RPE65 gene therapy for Leber's congenital amaurosis: a Phase 1 dose-escalation trial. Lancet, 2009; 374:1597–1605.

76.

Cereso

, Pequignot

, Robert

, et al. Proof of concept for AAV2/5-mediated gene therapy in iPSC-derived retinal pigment epithelium of a choroideremia patient. Mol Ther Methods Clin Dev, 2014; 1:14011.

77.

Davidson

, Stein

, Heth

, et al. Recombinant adeno-associated virus type 2, 4, and 5 vectors: transduction of variant cell types and regions in the mammalian central nervous system. Proc Natl Acad Sci U S A, 2000; 97:3428–3432.

78.

Eberling

, Jagust

, Christine

, et al. Results from a phase I safety trial of hAADC gene therapy for Parkinson disease. Neurology, 2008; 70:1980–1983.

79.

LeWitt

, Rezai

, Leehey

, et al. AAV2-GAD gene therapy for advanced Parkinson's disease: a double-blind, sham-surgery controlled, randomised trial. Lancet Neurol, 2011; 10:309–319.

80.

Brantly

, Chulay

, Wang

, et al. Sustained transgene expression despite T lymphocyte responses in a clinical trial of rAAV1-AAT gene therapy. Proc Natl Acad Sci USA, 2009; 106:16363–16368.

81.

Mueller

, Chulay

, Trapnell

, et al. Human Treg responses allow sustained recombinant adeno-associated virus-mediated transgene expression. J Clin Invest. 2013; 123:5310–5318.

82.

Daar

, Fuggle

, Fabre

, et al. The detailed distribution of HLA-A, B, C antigens in normal human organs. Transplantation, 1984; 38:287–292.

83.

Nennesmo

, Olsson

, Ljungdahl

, et al. Interferon-gamma-like immunoreactivity and T-cell marker expression in rat skeletal muscle fibres. Brain Res, 1989; 504:306–310.

84.

Wiendl

, Mitsdoerffer

, Schneider

, et al. Human muscle cells express a B7-related molecule, B7–H1, with strong negative immune regulatory potential: a novel mechanism of counterbalancing the immune attack in idiopathic inflammatory myopathies. FASEB J, 2003; 17:1892–1894.

85.

Lewis

, Mizrahi

, Toledano

, et al. Alpha1-antitrypsin monotherapy induces immune tolerance during islet allograft transplantation in mice. Proc Natl Acad Sci U S A, 2008; 105:16236–16241.

86.

Lewis

, Shapiro

, Bowers

, et al. Alpha1-antitrypsin monotherapy prolongs islet allograft survival in mice. Proc Natl Acad Sci U S A, 2005; 102:12153–12158.

87.

Shahaf

, Moser

, Ozeri

, et al. α-1-antitrypsin gene delivery reduces inflammation, increases T-regulatory cell population size and prevents islet allograft rejection. Mol Med, 2011; 171000–171011.

88.

Mittag

, Tarnok

. Recent advances in cytometry applications: preclinical, clinical, and cell biology. Methods Cell Biol, 2011; 103:1–20.

89.

Leipold

, Newell

, Maecker

, et al. Phenotyping of human PBMCs using mass cytometry. Methods Mol Biol, 2015; 1343:81–95.

90.

Hewitt

, Li

, Gray

, et al. Reducing the risk of adeno-associated virus (AAV) vector mobilization with AAV type 5 vectors. J Virol, 2009; 83:3919–3929.

91.

Halbert

, Miller

, Mcnamara

, et al. 2006. Prevalence of neutralizing antibodies against adeno-associated virus (AAV) types 2, 5, and 6 in cystic fibrosis and normal populations: implications for gene therapy using AAV vectors. Hum Gene Ther, 2006; 17:440–447.

92.

D'Avola

, López-Franco

, Sangro

, et al. Phase I open label liver-directed gene therapy clinical trial for acute intermittent porphyria. J Hepatol, 2016; 65:776–783.

93.

Miesbach

, Klamroth

, Schutgens

, et al. Updated results from a dose escalating study in adult patients with haemophilia B treated with AMT-060 (AAV5-hFIX) gene therapy. WHF 2016, Late-Breaking Abstract. Haemophilia, 2016; 22:151–152.

94.

Pasi

, Rangarajan

, Pierce

, et al. Interim results of an open-label, Phase 1/2 study of BMN 270, an AAV5-FVIII gene transfer in severe hemophilia A, WHF 2016, Late-Breaking Abstract. Haemophilia, 2016; 22:151–152.

95.

Nathwani

, Tuddenham

, Rangarajan

, et al. Adenovirus-associated virus vector-mediated gene transfer in hemophilia B. N Engl J Med, 2011; 365:2357–2365.

96.

Ferreira

, Twisk

, Kwikkers

, et al. Immune responses to intramuscular administration of alipogene tiparvovec (AAV1-LPL(S447X)) in a Phase II clinical trial of lipoprotein lipase deficiency gene therapy. Hum Gene Ther, 2014; 25:180–188.

97.

Monahan

, Walsh

, Powell

, et al. Update on a Phase 1/2 open-label trial of BAX335, an adeno-associated virus 8 (AAV8) vector-based gene therapy program for hemophilia B. International Society on Thrombosis and Haemostasis 2015 Congress, June 24, 2015, Toronto, Canada, Abstract #LB101.

98.

Crudele

, Finn

, Siner

, et al. AAV liver expression of FIX-Padua prevents and eradicates FIX inhibitor without increasing thrombogenicity in hemophilia B dogs and mice. Blood, 2015; 125:1553–1561.

99.

Smith

, Levy

, Kotin

. A simplified baculovirus-AAV expression vector system coupled with one-step affinity purification yields high-titer rAAV stocks from insect cells. Mol Ther, 2009; 17:1888–1896.

100.

Urabe

, Nakakura

, Xin

, et al. Scalable generation of high-titer recombinant adeno-associated virus type 5 in insect cells. J Virol, 2006; 80:1874–1885.

101.

Venkatakrishnan

, Yarbrough

, Domsic

, et al. Structure and dynamics of adeno-associated virus serotype 1 VP1-unique N-terminal domain and its role in capsid trafficking. J Virol, 2013; 87:4974–4984.

102.

Mietzsch

, Casteleyn

, Weger

, et al. OneBac 2.0: Sf9 cell lines for production of AAV5 vectors with enhanced infectivity and minimal encapsidation of foreign DNA. Hum Gene Ther, 2015; 26:688–697.