Pache.co Information Technologies

In recent years, the safety and efficacy of AAV as a vector for gene therapy have been validated by clinical trials of AAV-mediated liver-targeted therapy for hemophilia A and B [36,37,38]; consequently, in vivo clinical applications of AAV-based vectors for liver-targeted gene therapy are rapidly increasing. Liver-targeted gene therapies for FD, namely FLT190 [39] and ST-920 [40], are both in phase 1/2 clinical research, and both have achieved good safety and efficacy. Therefore, we attempted to design a highly efficient AAV2/8-hGLA to establish a proof-of-concept for the clinical development of liver-targeted gene therapy for FD so as to improve the therapeutic effect and reduce the dose of AAV vector, thereby reducing the cost of gene therapy.

We validated the therapeutic effect of liver-targeted AAV2/8-hGLA in FD mice, which was produced using the HEK293 scientific research-scale manufacturing process. The results showed that α-Gal A activity significantly increased in FD mouse plasma and tissue. Indeed, the α-Gal A activity in the FD mouse livers was significantly higher than that in the FD group under the treatment of 0.75E + 12 vg/kg AAV2/8-hGLA (p = 0.0001), which resulted in a large amount of α-Gal A being secreted into plasma. The α-Gal A activity in the 0.75E + 12 vg/kg AAV2/8-hGLA–treated FD mouse plasma was approximately 53-fold higher than that of the FD group. The above results showed that AAV2/8-hGLA treatment results in high expression of α-Gal A in the livers of FD mice. The entire experiment was performed in the absence of immunosuppression, and following a single tail vein injection, the plasma enzyme activity in FD mice significantly increased in a dose-dependent manner, reaching a plateau in the second week and then maintaining sustained and stable plasma enzyme activity for 38 weeks.

After 12 weeks of AAV2/8-hGLA treatment, we found no significant difference in the plasma levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in FD mice treated with different doses of AAV2/8-hGLA (data not shown). The highest dose of AAV2/8-hGLA was 5E + 12 vg/kg, but no adverse reaction was observed in this group, confirming the safety of AAV2/8-hGLA. Compared with the FD group, at the dose of 0.75E + 12 vg/kg, the content and enzyme activity of α-Gal A in the tissues of FD mice were significantly increased in a dose-dependent manner. Reports in the literature show that as long as 5%–10% of α-Gal A enzyme activity remains, Gb3 and Lyso-Gb3 storage substrates can be effectively removed [41], indicating that the α-Gal A expressed in the AAV2/8-hGLA–treated FD mouse liver is effectively secreted into the plasma and taken up in FD mouse tissues, especially by the heart and kidney. Studies conducted earlier have shown that female FD mice exhibit reduced efficiency of transduction and transgene expression of AAV in comparison to males [25, 32, 39]. The effect may be caused by sex hormones, however, other studies pointed out that, unlike mice, androgens did not affect the transgene expression of AAV in nonhuman primates (NHPs) [42]. Thus, it is still necessary to determine if there exists a disparity in AAV2/8-hGLA transgenic expression between male and female FD mice. Although Gla KO mice have metabolomics profiles of FD, they do not exhibit the abnormal behavioral phenotype during our entire experiment cycle (from weeks 12 to 24). Despite this, a variety of preclinical pharmacokinetic and pharmacodynamic studies have been conducted on such models to evaluate enzyme replacement, substrate reduction, and gene therapy strategies for FD [43,44,45,46].

FD occurs as a result of the increased accumulation of Gb3 and Lyso-Gb3 in lysosomes, which leads to various complications [1]. Indeed, the storage content of substrates in patients is the current gold standard for the diagnosis and treatment of FD [47]. Our study showed that the content of Gb3 and Lyso-Gb3 in the plasma, urine, and tissues (heart, kidney, liver, and spleen) of FD mice treated with AAV2/8-hGLA was significantly reduced and tended to normalize compared with the FD group. Previous studies have shown that cardiac and renal failure are the most common and life-threatening complications in the end stage of most patients with FD disease; the main reason for the renal complications observed in FD is the poor clearance of Gb3, while the reason for the cardiac complications is the accumulation of Lyso-Gb3 [48]. Therefore, it is important to pay special attention to the clearance of renal Gb3 and cardiac Lyso-Gb3 when developing new therapies. Previous studies on FD have shown that poor clearance of substrate storage in the kidney is an important factor restricting the therapeutic effect on FD. However, our study showed that in FD mice treated with 0.75E + 12 vg/kg AAV2/8-hGLA, the residual Gb3 storage substrate in urine and kidney was < 10%, and the residual Lyso-Gb3 in heart was < 5%, indicating that AAV2/8-hGLA at a low dose had a good therapeutic effect in FD. And not only that the AAV2/8-hGLA treatment resulted in abundant α-Gal A presence in cardiomyocytes, renal tubular cells, and podocytes of FD mice. This could be taken as an indication that the Gb3 and Lyso-Gb3 could be efficiently cleared in these cells.

Notably, previous preclinical evaluations of AAV1- and AAV2-mediated gene therapy approaches in FD mice have shown suboptimal substrate clearance in key target tissues, particularly in the kidney [30,31,32]. Although the use of AAV8 vectors significantly improved transgene expression, normalization of renal Gb3 was only achieved when FD mice were treated at a young age (1 month old), before the onset of significant disease pathology [33]. The reason for this result may be related to the immune response, and the systemic expression of α-galactosidase using a constitutive promoter induces a strong immune response at low levels. The capsid, its genome, and transgene products are the main potentially immunogenic components of AAV vectors [49]. Other host-dependent and vector-dependent factors can modulate overall vector immunogenicity [49]. However, previous studies have shown that in the context of liver-directed gene transfer, transgene immunogenicity does not appear to be a problem compared with other tissues. Since the initial observation that mice expressing human factor IX in the liver are immune to transgenic products, several studies using AAV vectors in small and large animal models of genetic disease have shown that expression of the antigen in hepatocytes can promote strong antigen-specific immune tolerance [50,51,52]. Despite the success of AAV2/8-hGLA in clearing glycosphingolipids from peripheral organs, the enzyme generated is not expected to cross the blood–brain barrier (BBB). In the ensuing plan, we will develop a modified α-Gal A protein with a brain-targeting peptide in the future to achieve BBB transport and alleviate brain pathology and behavioral deficits [53].

Based on the results of previous studies, the infection efficiency and expression elements of the viral vector are the main factors affecting the therapeutic effect of gene therapy. In line with this, we aimed at improving the design of the transgene expression cassette to further enhance the AAV2/8-hGLA–mediated α-Gal A expression in plasma and tissues. As the GLA cDNA sequence used in AAV2/8-hGLA was optimized, further efforts have focused on engineering the AAV serotype to minimize the immune response elicited by the AAV vector. In vitro studies have demonstrated that AAV plasmid can effectively transfect liver cell and expression α-Gal A. In vivo studies have demonstrated that AAV2/8-hGLA can effectively transfect FD mouse livers, resulting in the expression and secretion of biologically active α-Gal A with high efficiency. Maximizing the efficacy of therapeutic AAV vectors is critical for clinical applications. Indeed, previous studies have shown that transduction of hepatocytes is more challenging in large animals, such as non-human primates (NHPs), compared with mice, with a 50- to 100-fold reduction in transduction efficiency [42, 54]. Furthermore, a more potent vector would allow to use lower vector doses, potentially allowing the vector to circumvent or minimize the anti-AAV capsid immune response associated with the administration of higher vector doses [54]. Notably, previous studies in transgenic mice overexpressing α-Gal A have demonstrated the safety of elevated systemic α-Gal A activity levels of up to 155- and 44-fold in the plasma and liver, respectively [55].

AAV2/8-hGLA (a safe and effective gene therapy for liver-specific expression) has several advantages over ERT (the current standard of care for FD): ① Liver-mediated AAV gene therapy has the potential to induce immune tolerance to the therapeutic target protein, while the formation of neutralizing antibodies against recombinases remains a limiting factor for the therapeutic effect of ERT. ② Gene therapy may become a “one-time treatment, life-long cure” treatment method for patients with FD, while ERT requires life-long treatment of patients, infusions of 2–4 h every 2 weeks, resulting in physical and psychological burdens for patients [19, 20, 56]. Furthermore, in mouse models of hemophilia B and Pompe disease, AAV-mediated gene therapy using preexisting antibodies against the respective therapeutic proteins resulted in sustained reduction or elimination of antibodies [43, 57, 58]. If true in humans, this would be a significant advantage, as it would allow the use of AAV gene therapy to treat patients with FD with preexisting α-Gal A antibodies (generated by ERT). The successful application of the AAV2/8 vector in FD mice has laid a theoretical foundation for the use of AAV2/8-mediated liver-specific expression gene therapy to treat LSDs. We will create a LSD gene therapy platform that we expect to reduce the dose and treatment cost of AAV vectors, thereby bringing hope for the realization of a one-time treatment for patients with LSDs.