BACKGROUND:Functional cure of chronic hepatitis B requires a sustained loss of HBsAg and hepatitis B virus (HBV) DNA, which is unlikely with current therapies. We aimed to investigate the safety, antiviral activity, and anti-HBV immune responses of the N-acetylgalactosamine-conjugated, small interfering RNA imdusiran, a therapeutic strategy designed to reduce viral antigens on the path to achieving immune control. METHODS:AB-729-001 was a phase 1a/b trial conducted at 11 centres and hospitals: three in Australia, one in New Zealand, three in Thailand, one in Hong Kong, two in South Korea, and one in Moldova. In part 1, which had a double-blinded, single ascending dose design consisting of four sequential dose groups, healthy individuals aged 18-45 years were randomly assigned (2:1), using fixed block randomisation (block number of two for 1:1 active:placebo sentinel dosing, followed by block number of four for 3:1 active:placebo for the next four participants), into three equal-sized cohorts and received subcutaneous imdusiran or placebo as a single dose of 60 mg, 180 mg, or 360 mg. Participants and investigators were masked to treatment assignment until part 1 was complete. Parts 2 and 3 were open-label and enrolled individuals aged 18-65 years with chronic hepatitis B, who were HBeAg-positive or HBeAg-negative and were taking or not taking nucleoside or nucleotide (nucleos[t]ide) analogue therapy (continued during the imdusiran treatment period), and who had HBsAg concentrations of at least 250 international units per mL, no clinically significant abnormalities on liver ultrasound, and an absence of cirrhosis. These participants received subcutaneous imdusiran as a single dose of 60 mg, 90 mg, or 180 mg (part 2) or as repeat doses of 60 mg or 90 mg every 4, 8, or 12 weeks for up to 48 weeks (part 3). The primary endpoint for all parts of the study was to evaluate the frequency and severity of treatment-emergent adverse events, discontinuations due to adverse events, and laboratory abnormalities following the administration of single doses to healthy participants (evaluated until day 29 after dosing) and of single and multiple doses of imdusiran to participants with chronic hepatitis B (evaluated until week 48 of follow-up and until month 36 of follow-up for participants who discontinued nucleos[t]ide analogue therapy after completing imdusiran treatment and meeting eligibility criteria). All individuals who received at least one dose of study drug comprised the safety population and were included in safety analyses. This study was registered with the Australia New Zealand Clinical Trials Registry-ACTRN12619000954123 (part 1), ACTRN12619001197123 (part 2), and ACTRN12620000295943 (part 3)-and is complete. FINDINGS:Between July 16 and Aug 30, 2019, 18 individuals (18 [100%] male) were enrolled in part 1 and randomly assigned (12 [67%] to imdusiran and six [33%] to placebo). Between Sept 3, 2019, and June 11, 2020, of 43 individuals assessed for eligibility, 22 (51%) were enrolled in part 2, of whom 15 (68%) were male and seven (32%) were female. 78 individuals were assessed for eligibility in part 3 between April 30, 2020, and June 26, 2021, of whom 43 (55%) were enrolled (26 [60%] male and 17 [40%] female). All 83 enrolled participants received at least one dose of imdusiran or placebo and were included in the safety analyses. Across all parts, no participants discontinued imdusiran owing to an adverse event and no dose-related trends in reported treatment-emergent adverse events were observed across groups. There were no deaths. The most commonly reported treatment-emergent adverse events occurring in two or more participants in part 1 were an increase in alanine aminotransferase concentration (n=2; both reported as treatment-related), dizziness (n=2), medical device site reaction (n=3), headache (n=2), and oropharyngeal pain (n=2). The most commonly reported treatment-emergent adverse events in part 2 were transient injection-site pain (n=5; all reported as treatment-related), headache (n=4; two treatment-related), an increase in alanine aminotransferase (n=3; two treatment-related) and aspartate aminotransferase (n=2; one treatment-related) concentrations, and dizziness (n=2). The most commonly reported treatment-emergent adverse events in part 3 were coronavirus infection (n=18), injection-site pain (n=8; all of which were reported to be treatment-related), injection-site erythema (n=4; all treatment-related), headache (n=4), upper respiratory tract infection (n=4), pyrexia (n=4), fatigue (n=3; one treatment-related), and injection-site bruising (n=3; all treatment-related). All were reported as grade 1. No clinically relevant changes in clinical laboratory, vital signs, and electrocardiogram values over time were noted in any part of the study. INTERPRETATION:Single and multiple doses of imdusiran were safe and well tolerated in healthy individuals and in individuals with chronic hepatitis B, supporting drug development of imdusiran as a future treatment targeting functional cure for chronic hepatitis B. FUNDING:Arbutus Biopharma.

HBV capsid assembly modulators (CAMs) target the core protein and inhibit pregenomic RNA encapsidation and viral replication. HBV CAMs also interfere with cccDNA formation during de novo infection, which in turn suppresses transcription and production of HBV antigens. In this report, we describe the antiviral activities of AB-836, a potent and highly selective HBV CAM. AB-836 inhibited viral replication (EC50 = 0.010 μM) in HepDE19 cells, and cccDNA formation (EC50 = 0.18 μM) and HBsAg production (EC50 = 0.20 μM) in HepG2-NTCP cells during de novo infection. AB-836 showed broad genotype coverage, remained active against variants resistant to nucleos(t)ide analogs, and demonstrated improved antiviral potency against core variants resistant to other CAMs. AB-836 also mediated potent inhibition of HBV replication in a hydrodynamic injection mouse model, reducing both serum and liver HBV DNA. In a Phase 1 clinical study, 28 days of once-daily AB-836 oral dosing at 50, 100, and 200 mg resulted in mean serum HBV DNA declines of 2.57, 3.04, and 3.55 log10 IU/mL from baseline, respectively. Neither on-treatment viral rebound nor the emergence of viral resistance was observed during the 28-day treatment period. Furthermore, HBV DNA sequence analysis of baseline samples from the Phase 1 study revealed that 51.4% of the chronic hepatitis B participants contained at least one core polymorphism within the CAM-binding pocket, suggesting that genetic variations exist at this site. While AB-836 was discontinued due to clinical safety findings, data from the preclinical and clinical studies could help inform future optimization of HBV CAMs.