Crigler-Najjar syndrome (CNs) is an orphan enzymopathy caused by loss-of-function variants in UGT1A1 [1]. The lack of active uridine diphosphate glucuronosyltransferase 1A1, which leads to accumulation of bilirubin conjugates clinically represented by jaundice and severe neurological impairments, is potentially fatal in infancy. The diagnosis involves biochemical and molecular genetic tests for, respectively, unconjugated bilirubin and UGT1A1 variants. No specific pathogenetic treatment for the condition has been proposed apart from transplantation of the liver in CNs type I, the severe form. The patients require constant observation and are on continual bilirubin clearance by plasmapheresis and (or) phototherapy, which is physiologically straining and profoundly affects the lifestyle options in young patients.

The potential of using adeno-associated virus serotype 8 for the treatment of Crigler-Najjar syndrome has been substantiated in preclinical studies in vivo [2, 3]. In 2023, D'Antiga et al. reported successful treatment of Crigler-Najjar syndrome (CNs) type I in adult patients using a gene drug with AAV-mediated delivery [4]. In pediatric patients, the efficiency could (speculatively) interfere with the high rates of growth and physiological renewal of liver cells. Here we describe a sustained clinical response (107 weeks by the time of writing) in a pediatric patient with CNs type I using UGT1A1 coding sequence delivered by two consecutive doses of adenoassociated viral vector serotype 8 (AAV8). This is the first report on AAV therapy in a two-dose delivery mode.

Case description

Vector design and purification

сDNA from a healthy donor with wild-type UGT1A1 sequence was used as a template. The coding sequence was cloned into pAAV-TBG plasmid. The pAAV-TBG-UGT1A1 construct was packaged in HEK293 cells.

The particles were purified in an iodixanol stepwise density gradient (OptiPrep, StemCell Technologies, USA) at 350,000 g, +18 °C for 1.5 hours (Optima-XPN ultracentrifuge; Beckman, USA). The 60% fraction and a half of the 40% iodixanol fraction were collected, purified by dialysis (Spectrum™, Fisher Scientific #0867140; pore size 100 kDa) against (1x) PBS/350 mM NaCl/0.001% Pluronic F-68 buffer at 4°C for 14–18 h, concentrated with Amicon® centrifugal filters, pore size 100 kDa (Millipore Sigma, USA) and additionally sterilized by 0.22 μm syringe filtration. The product, alphaglucuronosyltransferas egene unoparvovec, was quantitated by real-time PCR with plasmid DNA as a reference. All purity indicators were found to be within the reference limits for clinical use; the product contained endotoxin <0.64 EU/ml (Endosafe® LAL; Charles River Endosafe, USA), BSA < 159 ng/ml (BSA ELISA Kit; Wuhan Fine Biotech, PRC), HEK293 residual proteins <2 ng/ml (HEK293 НСР ELISA Kit; Cygnus Technologies, USA) and was Mycoplasma-negative.

Preclinical study

In vitro tests were carried out on HeLa cells (ATCC, USA).

Animal study was authorized by the Institutional Animal Care and Use Committee at the Pirogov University. In vivo tests on C57BL6 mice were performed in three rounds: doseresponse assessment, overdose toxicity assessment and fade-out assessment in adolescence. The control group received identical volume of dilution buffer instead of the drug in all experiments.

Dose-response assessment was performed on 8-month-old mice in groups n(c)=11, n(L)=11, n(M)=11, n(H)=11. Toxicology and biodistribution studies involved monitoring of leukocyte counts and biochemical tests at 3, 4, and 5 weeks postinfusion, and post-mortem assessment of blood parameters and biodistribution of the drug to muscle and viscera on week 6. Immunohistochemistry revealed selective biodistribution of the human UGT1A1 coding sequence to the liver. The hUGT1A1 signal increased incrementally with the dose by both the intensity and the spread into the liver parenchyma from veins of the lobules (fig. 1).

Overdose toxicity assessment of the maximum of potential therapeutic dose was performed by administering a 1.2 × 10¹⁴ vg/kg dose to 8-month-old mice in control and experimental groups (n = 6 each) with the same check points and controlled parameters as above. No significant toxicity effects were detected.

Fade-out assessment in adolescence was assessed on 5-week-old mice (n = 24) received tail vein 1.2 × 10¹³ vg/kg drug injections. At 2 weeks post-infusion and then every 3 weeks until the end of experiment (at the age of 19 weeks), groups of mice (n = 6, 4, 4, 4, 6) were euthanized with livers taken for qPCR analysis of viral genome copies and quantitative protein detection by immunohistochemistry (liver sections stained with hUGT1A1-specific antibody). Quantitative PCR showed a 2-fold decrease in the vector DNA content every 6 weeks (4-fold over the experiment). Quantitative protein analysis showed that hUGT1A1 levels increased up to 5 weeks post-infusion and remained stable until the end of the experiment.

Patient

A girl with CNs type I, confirmed by UGT1A1 gene sequencing and manifested as severe unconjugated hyperbilirubinemia, under continuous observation at the Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology, was maintained on 12 h daily phototherapy, which afforded a neat confinement of indirect bilirubin levels within the toxicity limit of 300–350 μmol/l.

The AAV therapy was approved by Ethics Committee at the Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology on Dec 01 2022, Protocol #12 for infusion I, and on Jul 20 2023, Protocol #7 for infusion II. Application for clinical trial phase I–II No. 13571. The patient's legal representatives provided voluntary informed consents for the study and for each infusion of the drug.

Treatment and outcome

The gene therapy was initiated at an age of 7 years 5 months. The patient received two intravenous infusions of alphaglucu ronosyltransferasegene unoparvovec: 6 × 10¹² viral genomes per kg body weight (vg/kg) on week 0 and 1.2 × 10¹³ vg/kg on week 27 (fig. 2). The safety was monitored by transaminase levels and patient-reported well-being. The efficacy was assessed by reduction in serum bilirubin levels. A chart of indirect bilirubin levels over 3.5 years, including an extensive time length preceding the treatment, is given in fig. 2. The phototherapy regimen was continuously adjusted in accordance with the indirect bilirubin levels. Both infusions were accompanied by prednisolone support to mitigate immune reactions: the patient received prednisolone daily, in oral doses reduced stepwise from 1 mg/kg to 0 in the course of 8 and 2 weeks since the infusion, respectively (fig. 2). In addition, on day of infusion II, the patient received 250 mg (10 mg/kg) methylprednisolone intravenously.

No neutralizing antibodies to AAV8 were detectable before infusion I. The positive immunological reaction with AAV8 epitopes persisted post-infusion.

Infusion I reduced the serum bilirubin 2-fold, the minimum achieved throughout weeks 2–9. As the daily phototherapy doses were gradually reduced to 4 h by week 8, serum bilirubin concentrations rebound to ~250 μmol/l starting from week 10. The clinical decision on infusion II was intended at complete elimination of phototherapy. Infusion II on week 27 produced no toxic effects and afforded a further ~20% reduction in serum bilirubin concentrations by week 29. Considering the positive dynamics, the phototherapy support was discontinued, but resumed 2 weeks later at the same limited dose of 4 h, as serum bilirubin concentrations rapidly increased to ~400 μmol/l. The re-initiated phototherapy provided rapid correction of serum bilirubin concentrations to ~250 μmol/l. The liver enzyme profiles remained normal throughout the observation period. At the time of writing (week 97), the patient is stable on phototherapy administered 4 h daily, without systemic complaints.

Case discussion

The treatment produced no toxic side effects. Although complete discontinuation of the phototherapy proved unfeasible at this stage, the treatment afforded a lasting 3-fold reduction in daily exposure compared with initial values.

CONCLUSION

The use of liver-specific gene therapy in pediatric patients is associated with accelerated loss of the restored function, apparently due to the high physiological regeneration capacity and liver growth. The opportunity of safe repeated administration of AAV8 vector allows satisfactory control of the critical biochemical indicator.

The Gene Therapy for Crigler-Najjar Syndrome Type I is currently in a Phase 1/2 trial (https://clinicaltrials.gov/study/NCT06641154) and could be applicable to other patients aged 3 months to 10 years.