Efficacy of antiviral therapy in adults with chronic hepatitis B according to baseline hepatitis B virus DNA and alanine aminotransferase concentrations: a systematic review and meta-analysis

Background In the 2015 WHO guidelines for chronic hepatitis B (CHB), antiviral therapy was recommended for individuals with cirrhosis and individuals without cirrhosis with persistently elevated alanine aminotransferase (ALT) concentrations and hepatitis B virus (HBV) DNA >20 000 IU/mL, whereas treatment was deferred for individuals with persistently normal ALT concentrations and HBV DNA <2000 IU/mL. To inform 2024 WHO guidelines on CHB and the potential expansion of treatment threshold recommendations, we conducted two linked systematic reviews and meta-analyses; this analysis examines the efficacy of antiviral therapy in adults with non-cirrhotic CHB according to baseline HBV DNA and ALT concentrations. Methods In this systematic review and meta-analysis, we searched PubMed, Embase, Web of Science, and the Cochrane Library for randomised controlled trials (RCTs) and non-randomised (prospective or retrospective) confounder-controlled cohort studies of antiviral therapy versus placebo or no treatment in people with CHB, published in any language between Jan 1, 2000, and Feb 6, 2023. We also reviewed the reference lists of included studies and systematic reviews to identify RCTs published before 2000. Eligible studies reported baseline HBV DNA and ALT concentrations of adults with CHB, had less than 30% of participants with cirrhosis at baseline, and reported on at least one of our predefined outcomes. We excluded studies focused exclusively on pregnant women, individuals co-infected with HIV, hepatitis C virus, or hepatitis D virus, or individuals with primary conditions other than CHB, and studies that included participants who had received anti-HBV therapy in the 6 months before study enrolment. We extracted aggregate data to examine clinical outcomes (ie, hepatocellular carcinoma, cirrhosis, all-cause mortality, and liver-related mortality) and intermediate outcomes (ie, liver fibrosis, liver necroinflammation, ALT normalisation, HBsAg and HBeAg seroclearance and seroconversion, and HBV DNA suppression) stratified by baseline HBV DNA concentration (<2000 IU/mL, 2000–19 999 IU/mL, 20 000–199 999 IU/mL, 200 000–1 999 999 IU/mL, 2 000 000–19 999 999 IU/mL, and ≥20 000 000 IU/mL) and ALT (less than the upper limit of normal [ULN], 1·0–1·9 × ULN, and ≥2·0 × ULN). We used random-effects meta-analysis to pool unadjusted risk ratios (RRs) from RCTs and adjusted or unadjusted hazard ratios (aHRs or HRs) or unadjusted RRs for non-randomised studies. We estimated the number needed to treat (NNT) to prevent one case of hepatocellular carcinoma with nucleoside or nucleotide (nucleos[t]ide) analogue treatment. The study was registered with PROSPERO (CRD42023437560). Findings Of 13 224 articles screened, 24 met the inclusion criteria, including 16 studies on nucleos(t)ide analogues (12 RCTs and four non-randomised studies) and eight studies on interferon alfa-2-based therapy (four RCTs and four non-randomised studies). In adults with HBV DNA concentrations ≥20 000 IU/mL or elevated ALT (ie, ≥1·0 × ULN) at baseline, nucleos(t)ide analogue therapy was associated with a reduced risk of hepatocellular carcinoma (in non-randomised studies) and improvements in multiple intermediate outcomes (ie, liver fibrosis, necroinflammation, ALT normalisation, HBV DNA suppression, and HBeAg seroclearance and seroconversion), with certainty of evidence ranging from very low to high. In adults with baseline HBV DNA concentrations <20 000 IU/mL, nucleos(t)ide analogue therapy had no statistically significant effect on the risk of hepatocellular carcinoma, and no other outcomes were evaluated. From non-randomised studies, the aHRs for the risk of hepatocellular carcinoma with nucleos(t)ide analogue therapy were 0·72 (95% CI 0·43–1·20) for HBV DNA <2000 IU/mL, 0·45 (0·14–1·47) for 2000–19 999 IU/mL, and 0·39 (0·29–0·54; I 2 =0·0%) for ≥20 000 IU/mL. For adults with normal baseline ALT, nucleos(t)ide analogue therapy showed some efficacy for HBV DNA suppression (RR 31·50, 95% CI 2·02–492·36), but no efficacy for the remaining outcomes. Overall, higher certainty of evidence was seen with higher baseline HBV DNA or ALT concentrations. The between-study heterogeneity for pooled estimates varied across outcomes ( I 2 range 0·0–82·7%) but was low for most analyses (median I 2 =0·0%, IQR 0·0–25·5). Of the 16 RCTs, six were rated to have a low risk of bias, four were rated intermediate, and six as high. Of the eight non-randomised studies, four each were rated as fair and poor quality, respectively. The estimated NNT for nucleos(t)ide analogue therapy to prevent one case of hepatocellular carcinoma was 149 for baseline HBV DNA <2000 IU/mL over a median treatment duration of 12·0 years (IQR 4·1–26·2), 45 for HBV DNA 2000–19 999 IU/mL over 10·6 years (3·7–24·0), and 15 for HBV DNA ≥20 000 IU/mL over 13·6 years (6·7–22·1). Interpretation Evidence supports the efficacy of nucleos(t)ide analogue therapy in adults with HBV DNA ≥20 000 IU/mL or elevated ALT. However, the efficacy of nucleos(t)ide analogues remains uncertain in adults with HBV DNA <20 000 IU/mL or normal ALT. Future research should prioritise establishing treatment benefits in these groups, especially in high-burden settings in sub-Saharan Africa. Funding WHO.