Author information

¹Nuffield Department of Medicine, University of Oxford, Oxford, UK; Department of Infectious Diseases and Microbiology, Oxford University Hospitals NHS Foundation Trust, UK. Electronic address: sheila.lumley@trinity.ox.ac.uk.

²The Francis Crick Institute, 1 Midland Road, London, UK.

³Parasites and Microbes, Wellcome Sanger Institute, Hinxton, UK.

⁴The Francis Crick Institute, 1 Midland Road, London, UK; Africa Health Research Institute, Durban, SA.

⁵University of KwaZulu Natal, Nelson R Mandela Medical School,Durban, SA.

⁶Nuffield Department of Medicine, University of Oxford, Oxford, UK; Department of Infectious Diseases and Microbiology, Oxford University Hospitals NHS Foundation Trust, UK; KEMRI-Wellcome Trust, Kilifi, Kenya.

⁷KEMRI-Wellcome Trust, Kilifi, Kenya.

⁸Nuffield Department of Medicine, University of Oxford, Oxford, UK.

⁹Bodleian Health Care Libraries, University of Oxford, UK.

¹⁰Institut Pasteur, Université Paris Cité, Unité d'Épidémiologie des Maladies Émergentes, Paris, France.

¹¹The Francis Crick Institute, 1 Midland Road, London, UK; Department of Infectious Diseases, University College London Hospitals, London, UK; Division of Infection and Immunity, University College London, Gower Street, London, UK.

Abstract

Background: As nucleos/tide analogue (NA) therapy (e.g. entecavir and tenofovir) for chronic Hepatitis B virus (HBV) infection becomes more widely indicated and available, understanding drug resistance is essential. A systematic review to quantify resistance to these agents has not previously been undertaken.

Methods: We performed a systematic review and random-effects meta-analysis to estimate the risk of HBV resistance to entecavir and tenofovir. We searched nine databases up to 29-Aug-23. We included studies of HBV infection featuring >10 individuals, written in English, reporting treatment ≥48 weeks, with assessment of HBV resistance based on viral sequence data. Data were analysed according to prior exposure history to NA, and choice of NA agent. Analyses were performed in R.

Findings: 62 studies involving a total of 12,358 participants were included. For entecavir, in treatment-naive individuals (22 studies; 4326 individuals), resistance increased over time to 0.9 % at ≥5 years (95 %CI 0.1-2.3 %), and resistance was increased in NA-experienced individuals (18 studies; 1112 individuals), to 20.1 % (95 %CI 1.6-50.1 %) at ≥5 years. For tenofovir, pooled resistance risk was 0.0 % at all time points, whether previously NA naive (11 studies; 3778 individuals) or experienced (19 studies; 2059 individuals). There was a lack of consistent definitions, poor global representation and insufficient metadata to support subgroup analysis.

Interpretation: We have generated the first pooled estimates of HBV entecavir and tenofovir resistance over time. HBV resistance to entecavir in treatment-experienced groups in particular may represent a clinical and public health challenge. To date, tenofovir appears to have an excellent resistance profile, but due to data gaps, we caution that existing studies under-estimate the true real-world risk of resistance. Robust prospective data collection is crucial to reduce health inequities and reduce blind-spots in surveillance as treatment is rolled out more widely.