Human Gene Therapy - April 2023 - 274

274
WESTHAUS ET AL.
Keywords: liver gene therapy, AAV, nonhuman primate, xenograft model, hepatocyte
INTRODUCTION
RECOMBINANT ADENO-ASSOCIATED VIRAL (rAAV) vectors are
versatile delivery tools composed of a single-stranded
DNA genome flanked by 145-bp inverted terminal repeats,
packaged within an icosahedral protein capsid. The adenoassociated
virus (AAV), from which this vector system
was derived, is a nonpathogenic helper-dependent parvovirus
with multiple naturally occurring serotypes, including
the prototypical serotype 2 (AAV2).1,2 AAV2 was the
first variant to be vectorized and is the best understood
serotype, still used in many studies to date.3,4 The structure
and amino acid sequence of the nonenvelopedAAV capsid
is the main determinant of tropism.5 Therefore, modifying
the viral capsid has been used as a strategy to target specific
cell types and organs for therapeutic applications.6
Clinical success of gene therapy trials using AAV vectors
has led to the authorization of products for three indications
to date: RPE65-associated retinal dystrophy
(AAV2 capsid; Luxturna),7 spinal muscular atrophy
(AAV9 capsid; Zolgensma),8 and lipoprotein lipase deficiency
(AAV1 capsid, Glybera; no longer available).9
Although these AAV-based products target different organs
(the eye, the central nervous system, or the muscles,
respectively), therapies targeting disorders of other organs,
such as the liver, have not reached market approval to date.
The liver is an important clinical target for gene therapies
because of its key role in metabolism and homeostasis.
Most of the experience in liver gene transfer using
AAV vectors has been obtained in clinical trials for two
coagulation disorders: hemophilia A and B.10,11 Data from
these trials have been encouraging and showed a relatively
good safety profile. However, clinical studies conducted to
date pointed out several challenges that need to be overcome
to facilitate approval of therapeutic products for
these diseases and expanding AAVs as therapeutics for
other liver disorders. These challenges include the activation
of the immune system following vector administration,
unexpectedly low efficiency of targeting human
hepatocytes in vivo, and liver toxicity associated with the
administration of high vector doses.12
The development of immune responses toward the
therapeutic vector, resulting in the reduction of transgene
expression, was first observed in a phase I/II study that
utilized AAV2 to express Factor IX to treat hemophilia
B.13 AAV2 is endemic to the human population and thus
many patients who could benefit from AAV2-based therapeutics
have developed immunity to this serotype during
their lifetimes, including six of seven patients enrolled in
the first systemic clinical AAV2 study. Elevation of liver
transaminases and CD8+ T cells against AAV were detected
following vector administration through the hepatic
artery.13,14
Increase in Factor IX levels was only detected in two
patients and it was transient, a striking contrast to data
obtained in preclinical studies in mice and nonhuman
primates (NHPs), which showed long-term transgene expression.15
Although disappointing overall, the study
confirmed the relative safety of rAAV-mediated liver gene
transfer.16 Critically, however, the activation of the immune
system was not observed in any of the studies performed
in animal models, highlighting the limitations of
the model systems used to develop and validate new AAV
therapeutics before clinical implementation.
Subsequent liver-targeted trials utilized other serotypes,
such as AAV817 and AAV5,18 both selected based
on preclinical data in mice and NHPs.19,20 In a pivotal trial
sponsored by St. Jude Children's Research Hospital, a
self-complementary AAV8 vector encoding Factor IX was
administered to seronegative patients at three different
doses.21 Although long-term clinical efficacy was
achieved, an early increase in liver transaminases was
observed in the high-dose cohort.22 Immune adverse
events in liver clinical trials can generally be controlled by
the administration of corticosteroids to prevent elimination
of transduced cells and thus therapeutic transgene
expression. Nevertheless, prevalence of neutralizing antibodies
(NAbs) reduces the pool of patients that may be
able to benefit from novel experimental therapies.
However, clinical studies suggest that anti-AAV5 antibodies
do not preclude successful liver transduction with
AAV5-based vectors.23 This vector serotype has also been
used to deliver Factor IX at relatively high vector doses
(2 · 1013 vector genomes [vg]/kg) with no significant T
cell-mediated inflammation,24,25 although it must be
considered that this serotype is also less efficient than
others at transducing hepatocytes in animal models.19
Despite these current limitations, it remains clear that
vector serotype selection plays a crucial role in obtaining
the desired clinical outcomes. AAV serotypes used in
clinical studies are selected based on data generated in
preclinical, frequently murine, models. However, clinical
data obtained with AAV2, AAV5, and AAV8 clearly indicate
that AAV liver tropism, which can be species specific,20
can differ significantly between the murine or NHP
preclinical models and human patients.11
One way to overcome this issue has been using ''humanized''
models to functionally evaluate the existing
AAV variants for their ability to transduce human hepatocytes
in vivo. The same models can also be used to
identify novel capsids with improved liver tropism. To this
end, bioengineered capsids with high tropism toward human
hepatocytes, such as AAV-LK0326 and AAVNP59,27
Il2rg-/-
(FRG)28 mice demonstrating the validity of this
were developed in xenografted Fah-/-/Rag2-/-/
Human Gene Therapy - April 2023

Table of Contents for the Digital Edition of Human Gene Therapy - April 2023
