Comment on “Synthetic human ABCB4 mRNA therapy rescues severe liver disease phenotype in a BALB/c.Abcb4^-/- mouse model of PFIC3”

We have read with interest the paper by Wei et al. and we thank the authors for their reference to our 2019 study on adeno-associated virus (AAV)-mediated gene therapy correction of progressive familial intrahepatic cholestasis type 3 (PFIC3) in a clinically relevant mouse model. Their results utilizing lipid nanoparticles (LNP) to deliver functional human ABCB4 mRNA to hepatocytes of BALB/c.Abcb4^-/- mice and the therapeutic effect achieved in this severe PFIC3 mouse model with a high degree of fibrosis were quite remarkable. However, they framed their conclusions with respect to our previous study based on improper interpretations of several key aspects of our results. First, they did not consider our results when claiming they identified for the first time a ‘minimum’ of clinically meaningful restoration of hepatic phosphatidylcholine (PC) output, which was 10-42% of normal levels, i.e. de novo phenotypic ABCB4 enzymatic activity that resulted in a therapeutic effect. They stated that our results showed that a bile PC restoration of 70–100% was necessary for a therapeutic effect. In reality, our data pinpointed a threshold of around 4,000 μM of PC concentration in bile, which corresponded to 12–13% of the levels we measured in healthy wild-type mice (Fig. 1), which was clearly shown in our paper. This is substantially less than the 70-100% they claimed we reported and actually shows that their results corroborated our findings much more closely than was indicated. The identification of a minimum level of phenotypic restoration needed to revert or prevent this disease is an important detail in the daunting task of translating minimum therapeutic doses from preclinical models to human patients with PFIC3. There are many confounding variables in this process, including different delivery efficiencies depending on the species or genetic background and inherent differences in disease phenotype between mice and humans. Thus, the more complete the understanding of the therapeutic process in the preclinical models, the better they will serve for transitioning therapies into the clinic.

Second, the studies by Wei et al. were performed in 4-week-old mice, which showed that their approach was distinct in its ability to successfully revert fibrosis in juvenile mice. To substantiate their claims, they incorrectly cited our paper as having used older 7-week-old mice. In fact, our paper reported data showing an AAV-mediated therapeutic effect (including prevention of fibrosis) in both 5-week-old and 2-week-old mice. The misreporting of our results allowed them to imply that mRNA gene therapy holds an advantage over AAV treatment, which lies in being able to exert a durable therapeutic effect in very young study animals. Indeed, very young animals, like paediatric patients, represent a difficult challenge for AAV-mediated gene therapy because of a dilution effect due to the predominantly non-integrative nature of AAV genomes and the growing liver of immature recipients of treatment. One approach to circumvent the problem of the growing liver, as proposed by Wei et al., would be the continual periodic readministration of mRNA-mediated therapy. This, due to the high volume of treatments (potentially, up to 10 or 20 a year) would exert a high burden on patients both financially and in terms of quality of life, as well as require safety evaluations for long-term treatments with LNP-mRNA.

However, under appropriate conditions, the feasibility of readministering AAV is well established,^{,  ,}  and our most recent research has examined this hypothesis by testing the immuno-tolerogenic properties of rapamycin-loaded nanoparticles (ImmTOR) when coadministered with AAV and its ability to allow for subsequent AAV readministration to treat PFIC3 (Weber et al. manuscript in preparation). Indeed, the ability to readminister a gene therapy, regardless of its modality, will be of high importance for treating inborn errors of metabolism, most of which present and require treatment in early childhood while the liver is still growing. To this end, any advancement in developing novel treatment strategies, such as coadministering immunomodulators with gene therapy, in order to improve long-term treatment outcomes in paediatric patients are of utmost importance.