Current therapies for chronic hepatitis B (CHB) include nucleos(t)ide analogues (NAs) and interferon (IFN), but their relative efficacy as monotherapy or in combination has not been examined systematically for HBsAg loss (functional cure). Hence, we systematically reviewed the evidence for HBsAg loss in CHB patients treated with IFN, NA or the combination. We searched PubMed, EMBASE and abstracts from EASL, Asia Pacific Association for study of the Liver and American Association for the Study of Liver Disease for randomized controlled trials of CHB patients, comparing NA, IFN or the combination. The Cochrane Risk of Bias tool v2.0 and GRADE method were used. Analyses were stratified by NA genetic barrier, cirrhosis, type of combination therapy, HBeAg, treatment naivety, IFN dosage/duration and outcome duration. Sensitivity analysis was performed for selected strata, and HBsAg loss was measured at the end‐of‐study (EOS), end‐of‐treatment (EOT) or end‐of‐follow‐up (EOF). Effects were reported as risk differences (RD) with 95% confidence intervals (CI) using a random‐effects model. Forty‐five studies were included, all with low risk of bias. For HBsAg loss at EOS, when comparing combination vs IFN, RD = 1%, 95%CI‐1%, 2%; combination vs NA, RD = 5%, 95%CI 3%,7%; IFN vs NA, RD = 3%, 95%CI 2%,5%. Subgroup analysis showed a significant effect of standard IFN dose vs nonstandard; IFN duration ≥48 weeks vs <48 weeks, and loss of efficacy >2 years of follow‐up. Similar findings were seen in HBsAg seroconversion, but only three studies reported HBsAg seroreversion. In conclusion, IFN monotherapy/combination had a small but significant increase in HBsAg loss over NA, associated with standard dose of IFN and ≥48 weeks of therapy, although this effect faded over time.