nature.com

Preclinical efficacy and safety of AAVrh10-based plakophilin-2 gene therapy (LX2020) as a treatment for arrhythmogenic…

Abstract

Plakophilin-2 (PKP2) mutations cause fatal genetic heart disease and arrhythmogenic cardiomyopathy (ACM) with primary effects on the right ventricle (RV). Adeno-associated virus (AAV)-PKP2 gene therapy shows promise as a therapeutic strategy but lacks long-term data and guidelines on minimal effective doses in animal studies for treating RV deficits, arrhythmia burden, and improving survival when administered during disease settings, which are most relevant to clinical trials. Using AAVrh10, known for its preferential cardiac gene expression at lower doses, we show minimal doses required for efficacy for AAVrh10.PKP2 (LX2020) to rescue cardiac (molecular and especially RV) deficits, arrhythmia burden and survival in PKP2 ACM mice, suggesting its potential to reverse late-stage pathology. Safety assessments in non-human primates revealed no adverse events. These data support LX2020 as a viable treatment for PKP2 ACM patients.

Introduction

Arrhythmogenic cardiomyopathy (ACM) is a genetic heart disease characterized by mutations in desmosomal genes, with plakophilin-2 (PKP2) being a mostly right heart-driven disease and frequently mutated gene (70%), leading to sudden death or heart failure, for which effective treatments are lacking1. Previous studies have shown the potential of adeno-associated virus-PKP2 (AAV-PKP2) gene therapy in mitigating PKP2 loss and ACM-related cardiac deficits in animal and human-based models1,2,3,4. While promising, there are limited long-term data and assessment of minimal effective doses in this context in animal models for treating primary RV deficits, arrhythmia burden and improving survival when administered during disease settings, which are most relevant to human clinical trials.

Results

LX2020 administration showed efficacy on cardiac deficits in PKP2 ACM models

We employed AAVrh10, which exhibits preferential cardiac gene expression at lower doses5,6 to deliver relatively low doses of human PKP2 to evaluate its efficacy in PKP2 (Homozygous [Hom] and Heterozygous [Het]) mice, harboring a prevalent human PKP2 splice site mutation1. Severe PKP2 Hom mice display molecular (desmosomal and connexin 43) and arrhythmogenic deficits as early as postnatal day 2 in the absence of gross cardiac morphological deficits, which transition to severe structural abnormalities that mimic cardiac deficits observed in ACM patients at late disease stages1. On the other hand, the mild PKP2 Het mice mirror ACM patient genetics and display cardiac desmosomal/cell junction molecular deficits without severe cardiac deficits1, highlighting the spectrum of ACM disease observed in patients.

To assess the efficacy, the AAVrh10 vector containing the cardiac-specific chicken troponin T promoter (cTNT) and human PKP2 (AAVrh10.hPKP2, aka LX2020) was retro-orbitally injected at various doses into diseased 3-week-old PKP2 Hom mice (Fig. 1a). Post-dosing assessments at 12 weeks revealed a dose-dependent increase in cardiac hPKP2 protein expression that resulted in the significant rescue of cardiac cell–cell junctional protein deficits, compared to vehicle-treated (Fig. 1b, c). A substantial reduction in premature ventricular contractions (PVCs) burden was also seen compared to vehicle-injected group via surface electrocardiogram analyses (Fig. 1d). Cardiac magnetic resonance imaging demonstrated significant dose-dependent improvements in left and right ventricular dimensions and ejection fraction, compared to vehicles (Fig. 1e). Histological analysis revealed abrogation of the cardiac histopathology deficits (especially right ventricular wall thinning) in LX2020 treated groups, with mice at the 6E13 gc/kg dose resembling wild-type controls at 12 weeks post-dosing (Fig. 1f), highlighting a potential for reversal of disease pathology at late stages. These striking cardiac outcomes culminated in a significant extension in lifespan in LX2020 treated groups (Fig. 1g). LX2020 showed a significant aversion to premature death at 6E13 gc/kg, but also at 2E13 gc/kg (Fig. 1g), highlighting long-term efficacy in relation to vehicle-treated disease controls as cardiac (RV) function, histology and survival at these timepoints outcompete 84% of vehicle-treated diseased controls that exhibit a shortened lifespan at this same time point. Furthermore, there was equal representation of male and female mice within each dosing group to account for potential sex-related differences. Analyses of cardiac proteins and function, as well as survival, revealed no significant differences between male and female mice within each group, suggesting that sex did not influence the observed phenotypes and outcomes.

Fig. 1: Efficacy of AAVrh10.PKP2 (LX2020) in PKP2 ACM models.

figure 1

a Schemata of groups and doses administered to PKP2 Hom mice. 4 treatment conditions (Low dose, Mid dose, High dose and Vehicle) were administrated to PKP2 Hom mice retro-orbitally at approximately 3 weeks of age (n = 12 per group, mixed sex). Protein expression, arrhythmias, cardiac function (MRI), cardiac histopathology, and survival were analyzed after 12 weeks post-injection. Representative western blot (b) and quantitation (c) in WT, PKP2 Hom dosing groups. Two-way ANOVA with Tukey’s multiple comparison test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n = 9 for WT and n = 12 for each dosing group. Molecular weight (kDa) labeled on the right side of the gel. d The number of mice that displayed PVCs (gray bars) or normal rhythm (white bars) in different dosing groups compared with WT. e Quantification of the left (LV) and right (RV) ventricle end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) using MRI in WT and PKP2 Hom mice. n = 9 for WT, n = 1 for Veh, n = 3 for no treatment, n = 4 for low dose, n = 10 for mid and high dose. Two-way ANOVA with Tukey’s multiple comparison test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. f Representative cardiac histology images (H&E and Masson Trichrome stains) from all groups in the study. Scale bar, 1 mm. g Kaplan–Meier survival curves of WT, all dosing groups in PKP2 Hom mice (***P < 0.001, ****P < 0.0001). h Schemata of dosing groups in PKP2 Het mice. 4 treatment conditions were administrated retro-orbitally to PKP2 Het mice at approximately 8 weeks of age (mixed sex). Protein expressions were analyzed after 8 weeks post-dosing. Representative Western blot (i) and quantitation (j) in WT, PKP2 Het groups. Molecular weight (kDa) labeled on the right side of the gel. Two-way ANOVA with Tukey’s multiple comparison test. *P < 0.05, **P < 0.01, ****P < 0.0001. n = 4 for WT, n = 12 for low dose, n = 8 for veh, mid and high dose groups.

Full size image

To examine the effects of LX2020 in PKP2 Het mice (which emulate autosomal dominant genetics of ACM patients), the same doses (as in PKP2 Hom) were administered (Fig. 1h) to 8-week-old PKP2 Het mice. A dose-dependent increase in cardiac hPKP2 protein expression (Fig. 1i, j) was seen that translated into rescue of desmoplakin and desmoglein-2 cardiac cell–cell junction protein deficits at 2E13 gc/kg with more pronounced effects on N-cadherin and connexin 43 deficits at 6E13 gc/kg (Fig. 1i, j), highlighting the correlative impact of PKP2 dose on PKP2 expression and downstream molecular deficits. Given the genotypic closeness of PKP2 Het mice to patients, these data pave the way for more immediate insights into cardiac biopsy evaluations in future clinical trials.

Collectively, LX2020 administration in diseased conditions of both severe and mild PKP2 mice yielded efficacy at relatively lower doses to previously reported1,2,3,4, as seen by robust expression of protein and reversal of the cardiac phenotype associated with PKP2 at 2E13–6E13 gc/kg doses. These data also identify the minimally and maximally effective therapeutic doses that are clinically relevant for ACM patients.

LX2020 administrations were safe and well-tolerant in non-human primates (NHP)

In order to assess the safety profile of LX2020, we intravenously administered 4 groups of immunosuppressed healthy cynomolgus non-human primates (NHP) with a single dose of LX2020 at 2E13, 6E13, and 1E14 gc/kg, including vehicle as a control (Fig. 2a). At 12 weeks post-dosing, our studies revealed no adverse events associated with LX2020 administration on critical parameters such as cardiac dimensions, function, and electrocardiograms, even at the highest dose of 1E14 vg/kg (Fig. 2b–d). Additionally, no adverse effects were observed on clinical pathology, cardiac injury biomarkers, histopathology and immunogenicity (Fig. 2e). Dose-dependent biodistribution of LX2020 in various regions of the heart, including the disease-relevant right and left ventricles, were seen at these efficacious doses (Fig. 2f), concomitant with increases in hPKP2 mRNA expression (Fig. 2g). Although LX2020 administration showed vector biodistribution in multiple tissues outside the heart including the liver for example, this did not result in hPKP2 expression in the liver of NHPs, further reinforcing the LX2020 vector design which targeted hPKP2 to cardiomyocytes (Fig. 2h, Supplementary Fig. 1). Importantly, when subtracting for endogenous cardiac NHP PKP2 amounts in the vehicle-treated group, we show that LX2020 administration in NHPs results in an increase of 30, 52, and 32 ng/mg in cardiac hPKP2 in low, mid and high dose groups, respectively. These amounts will supersede or closely reach healthy cardiac hPKP2 protein amounts, as ACM patients are anticipated to be in the range of ~40–50% of normal healthy human hPKP2, which translates to 24–30 ng/mg, as healthy cardiac hPKP2 levels were quantified to be 61 ng/mg (n = 4). These data underscore the potential for this approach to achieve therapeutic levels of hPKP2 expression with all tested doses.

Fig. 2: Safety profile analysis of AAVrh10.PKP2 (LX2020) in non-human primates.

figure 2

a Schemata of dosing groups in NHPs (cynomolgus monkeys) following IV administration. N = 3–4 for each group, mixed sex. All NHPs were euthanized after 12 weeks, and clinical pathological assays, immunology, cardiac function, protein expression, and vector distribution were evaluated. b and (c) Representative M-Mode echocardiography images and statistical analysis of cardiac function from all NHP dosing groups at 12 weeks showed no significant difference. d Cardiac electrocardiogram showed no significant difference between NHP dosing groups at different time points. e Table of safety evaluations indicate that there is no abnormalities found in NHP dosing groups. f Biodistribution of LX2020 across the NHP hearts as quantified by vector copy number (VCN) analysis. g Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) analysis of PKP2 mRNA levels within different compartments of NHP hearts following treatment with AAV-rh10 PKP2. h Biodistribution of LX2020 across the NHP livers as quantified by VCN (left) and PKP2 protein expression analysis (right). f–h Two-way ANOVA with Tukey’s multiple comparison test. *P < 0.05, ***P < 0.001, ****P < 0.0001.

Full size image

Discussion

In summary, these data support the long-term efficacy of LX2020 in restoring cardiac hPKP2 expression and rescuing severe and mild ACM disease, while being safe and well tolerated. Future studies focused on the mechanism of how hPKP2 RNA and protein restoration could scaffold and restore protein expression of other cell–cell junction proteins and trigger reversal of pathology should be investigated. Our data highlight the LX2020 dose of 2E13 gc/kg as the minimally effective dose and, thus, its selection as the starting clinical dose to be tested in PKP2-ACM patients (NCT06109181). Given that multiple PKP2 clinical trials, which include LX2020, will be commencing using different vectors and starting doses (NCT06109181, NCT05885412, NCT06228924), balancing clinical safety alongside an impact on RV-related arrhythmias, histopathology, and function will be key metrics to circumvent the risks and showcase the long-term efficacy of gene therapy approaches for ACM.

Methods

Animals

Mouse lines harboring heterozygous or homozygous PKP2 IVS10-1 G > C mutation were previously generated using CRISPR-Cas9 mediated methods1. All mouse procedures were in full compliance with the ethical guidelines approved by the University of California, San Diego Institutional Animal Care and Use Committee and carried out in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Mice were euthanized according to ethical guidelines by first being anesthetized by isoflurane, followed by cervical dislocation to ensure confirmation of death. NHPs were humanely euthanized by first being sedated via intravenous administration of sodium pentobarbitone followed by a testing facility standard operating procedure approved method of exsanguination via the femoral vein. The NHP safety study reported in this study was carried out in compliance with GLP as per 21 Code of Federal Regulations Part 58.

Adeno-associated virus manufacture and injections

LX2020 manufacturing and titering was performed by SignaGen Laboratories. LX2020 administrations were performed at 3 weeks-old PKP2 Hom/ 8 weeks-old Het mice by using a 31 gauge needle and syringe via retro-orbitally at a titer of low (6E12 gc/kg), mid (2E13 gc/kg), and high (6E13 gc/kg) respectively. A vehicle group of mice was injected with formula buffer. Retro-orbital injections were used in mice as they offered an efficient and reliable intravenous delivery method that resulted in a lower incidence of extravasation as well as offered less stress and procedural times to tail vein injections. For AAV dosing in NHPs, an immunosuppressant, methylprednisolone, was administered intramuscularly on a weekly basis starting on Day −1 prior to LX2020 administration and continuing until necropsy (40 mg/animal for Day −1 and Weeks 1–10, 20 mg/animal for Weeks 11, and 10 mg/animal for Weeks 12 and 13). Single dose of 2E13, 6E13, and 1E14 gc/kg LX2020 or vehicle (AAV formulation buffer solution) was administered into a healthy cynomolgus monkey by a 5 min IV infusion.

Western blot analysis

Heart total protein lysates were extracted from PKP2 Hom/Het mice as previously described1. Cardiac protein were detected by primary antibodies including plakophilin-2 (rabbit, 1:1000, Sigma-Aldrich, catalog #HPA014314), desmoplakin (mouse, 1:1000, Bio-Rad, catalog#2722-504), desmoglein-2 (mouse, 1:1000, Fitzgerald, catalog #10R-D106a), plakoglobin (JUP) (goat, 1:1000, Sigma-Aldrich, catalog #SAB2500802), connexin 43 (rabbit, 1:8000, Sigma-Aldrich, catalog #C6219, N-cadherin (rabbit, 1:1000, Abcam, catalog #ab76057) and β-actin (mouse, 1:2000, Santa Cruz Biotechnology, catalog #sc-47778) used as a loading control. Blots were imaged in the ChemiDoc Touch Imaging system (Bio-Rad) and analyzed by Image J (NIH).

Surface ECG and premature ventricular beats (PVC)

PKP2 Hom mice were anesthetized with 5% isoflurane (Piramal, #6679401710) for 10–15 s and maintained at 1.5% isoflurane during the whole recording. Needle electrodes (30 gauge) were inserted subcutaneously into the left/right forearm and left leg. Electrical cardiac activity was recorded for 5 min. For PVC analysis, ectopic beats were identified manually over the entire 5 min recording by using LabChart 8 software.

Magnetic resonance imaging

Cardiac MRI on PKP2 Hom mice was performed on a 7T horizontal bore MR scanner (Bruker) as previously described1. For cardiac MRI image analyses, we manually segmented two-dimensional endocardial contours, short-axis continuous slices spanning entire ventricles will be analyzed with Segment v3.0 (Medviso). Functional parameters will be quantified with a summation of disks method to generate end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) for both left and right ventricles.

Histological analysis

All the mouse hearts were collected and fixed with 4% paraformaldehyde (Thermo Fisher, #J61899.AK). Fixed hearts were embedded in paraffin cut at 5 μm thickness (cross-section). Sections were stained with hematoxylin and eosin (Sigma-Aldrich, #HT1079-1SET) and Masson’s trichrome (Sigma–Aldrich, #HT15) stains according to the manufacturer’s instructions. Images were acquired with the Olympus VS200 slide scanner at the UCSD microscopy core (NIH (P30 NS047101).

NHP (cynomolgus monkeys) safety studies

Cardiovascular safety pharmacology was evaluated in a GLP toxicity study in healthy cynomolgus monkeys administered a single dose of 2.0 E13, 6.0E13, or 1.0E14 gc/kg LX2020 or vehicle (AAV formulation buffer solution). Blood draws were taken pre-dosing and every 4 weeks after dosing. All NHPs were euthanized after 12 weeks post-injection. Cardiac function measured by echocardiography, surface ECG, clinical pathological assays, immunology, and biodistribution (VCN and mRNA) were evaluated at pre-dose and weeks 4, 8, and 12 post-injections. All the studies were performed and analyzed by Charles Rivers Laboratories.

Statistical analysis

GraphPad Prism 10 was used for analyses in this study, all data showed as means ± standard deviation. Significance was determined by one-way or two-way ANOVA with Tukey post-hoc multiple comparison tests; detailed p-value and methods were described in each figure. For Kaplan-Meier survival analysis, significance was assessed by the log-rank test. A p-value < 0.05 was considered statistically significant.

Data availability

All the reagents and other supporting data are available to other researchers from the corresponding authors upon reasonable request.

Abbreviations

AAV:

adeno-associated viruses

AAV9:

adeno-associated virus vector of serotype 9

AAVrh10:

adeno-associated virus vector of serotype rh10

AAVrh10.hPKP2:

adeno-associated virus vector of serotype rh10 harboring human PKP2

ACM:

arrhythmogenic cardiomyopathy

cTnT:

cardiac troponin T

LX2020:

company code for AAVrh10.hPKP2

NHP:

non-human primates

MRI:

magnetic resonance imaging

PKP2:

plakophilin-2

PKP2 Het:

plakophilin-2 heterozygous mutant knock-in mice

PKP2 Hom:

plakophilin-2 homozygous mutant knock-in mice

References

Bradford, W. H. et al. Plakophilin 2 gene therapy prevents and rescues arrhythmogenic right ventricular cardiomyopathy in a mouse model harboring patient genetics. Nat. Cardiovasc. Res. 2, 1246–1261 (2023).

ArticleCASPubMedPubMed CentralGoogle Scholar

Kyriakopoulou, E. et al. Therapeutic efficacy of AAV-mediated restoration of PKP2 in arrhythmogenic cardiomyopathy. Nat. Cardiovasc. Res. 2, 1262–1276 (2023).

ArticleCASPubMedPubMed CentralGoogle Scholar

Wu, I. et al. AAV9:PKP2 improves heart function and survival in a Pkp2-deficient mouse model of arrhythmogenic right ventricular cardiomyopathy. Commun. Med. 4, 38 (2024).

ArticleCASPubMedPubMed CentralGoogle Scholar

Van Opbergen, C. J. M. et al. AAV-mediated delivery of Plakophilin-2a arrests progression of arrhythmogenic right ventricular cardiomyopathy in murine hearts: preclinical evidence supporting gene therapy in humans. Circ. Genom. Precis. Med. 17, e004305 (2024).

PubMedPubMed CentralGoogle Scholar

De, B. P. et al. High levels of persistent expression of alpha1-antitrypsin mediated by the nonhuman primate serotype rh.10 adeno-associated virus despite preexisting immunity to common human adeno-associated viruses. Mol. Ther. 13, 67–76 (2006).

ArticleCASPubMedGoogle Scholar

Munoz-Zuluaga, C. Identification of safe and effective intravenous dose of AAVrh.10hFXN to treat the cardiac manifestations of Friedreich’s ataxia. Hum. Gene Ther. 34, 605–615 (2023).

Download references

Acknowledgements

We would like to acknowledge Curia and Fuji Laboratories for providing the LX2020 virus and Sonia Gutierrez Granados and Ken Law for overseeing drug manufacturing and analytical analysis. We would like to also acknowledge Charles Rivers Laboratories for performing the NHP GLP toxicology study. We would like to acknowledge the following funding sources: (i) National Institutes of Health grants HL142251 and HL162369) (F.S.) and (ii) LEXEO Therapeutics Inc. (F.S.). The University of California San Diego Neuroscience Microscopy Shared Facility is supported by a grant from the NIH (P30 NS047101).

Author information

Authors and Affiliations

Department of Medicine, University of California San Diego, La Jolla, CA, USA

Jing Zhang, Erika Joana Gutierrez-Lara, Aryanne Do, Lena Nguyen & Farah Sheikh

LEXEO Therapeutics Inc, New York, NY, USA

Anju Nair, Nithya Selvan, Tim Fenn, Eric Adler & Richie Khanna

Authors

Jing Zhang

View author publications