Background: Gaps remain in the detection of nucleic acid test (NAT) yield and occult hepatitis B virus (HBV) infection (OBI) by current HBV surface antigen (HBsAg) assays. The lack of detection may be due to HBsAg levels below current assay detection limits, mutations affecting HBsAg assays or HBsAg levels, or the masking of HBsAg by antibody to HBsAg (anti-HBs). In this study, we evaluate the incremental detection of NAT yield and OBI from five diverse geographic areas by an improved sensitivity HBsAg assay and characterize the samples relative to the viral load, anti-HBs status, and PreS1–S2–S mutations. Included is a comparison population with HBV DNA levels comparable to OBI, but with readily detectable HBsAg (High Surface–Low DNA, HSLD). Methods: A total of 347 samples collected from the USA, South Africa, Spain, Cameroon, Vietnam, and Cote D’Ivoire representing NAT yield (HBsAg(−), antibody to HBV core antigen (anti-HBc)(−), HBV DNA(+), N = 131), OBI (HBsAg(−), anti-HBc(+), HBV DNA(+), N = 188), and HSLD (HBsAg(+), anti-HBc(+), HBV DNA(+), N = 28) were tested with ARCHITECT HBsAg NEXT (HBsAgNx) (sensitivity 0.005 IU/mL). The sequencing of the PreS1–S2–S genes from a subset of 177 samples was performed to determine the genotype and assess amino acid variability, particularly in anti-HBs(+) samples. Results: HBsAgNx detected 44/131 (33.6%) NAT yield and 42/188 (22.3%) OBI samples. Mean HBV DNA levels for NAT yield and OBI samples were lower in HBsAgNx(−) (50.3 and 25.9 IU/mL) than in HBsAgNx(+) samples (384.1 and 139.5 IU/mL). Anti-HBs ≥ 10 mIU/mL was present in 28.6% HBsAgNx(+) and 45.2% HBsAgNx(−) OBI, and in 3.6% HSLD samples. The genotypes were A1, A2, B, C, D, E, F, and H. There was no significant difference between HBsAgNx(−) and HBsAgNx(+) in the proportion of samples harboring substitutions or in the mean number of substitutions per sample in PreS1, PreS2, or S for the NAT yield or OBI (p range: 0.1231 to >0.9999). A total of 21/27 (77.8%) of HBsAgNx(+) OBI carried S escape mutations, insertions, or stop codons. HSLD had more PreS1 and fewer S substitutions compared to both HBsAgNx(−) and HBsAgNx(+) OBI. Mutations/deletions associated with impaired HBsAg secretion were observed in the OBI group. Conclusions: HBsAgNx provides the improved detection of NAT yield and OBI samples. Samples that remain undetected by HBsAgNx have exceptionally low HBsAg levels below the assay detection limit, likely due to low viremia or the suppression of HBsAg expression by host and viral factors.

Hepatitis B virus (HBV) is endemic throughout Africa, but its prevalence in the Democratic Republic of the Congo (DRC) is incompletely understood. We used dried blood spot (DBS) samples from the 2013 to 2014 Demographic and Health Survey in the DRC to measure the prevalence of HBV using the Abbott ARCHITECT HBV surface antigen (HBsAg) qualitative assay. We then attempted to sequence and genotype HBsAg-positive samples. The weighted national prevalence of HBV was 3.3% (95% CI: 1.8–4.7%), with a prevalence of 2.2% (95% CI: 0.3–4.1%) among children. Hepatitis B virus cases occurred countrywide and across age strata. Genotype E predominated (60%), and we found a unique cluster of genotype A isolates (30%). In conclusion, DBS-based HBsAg testing from a nationally representative survey found that HBV is common and widely distributed among Congolese adults and children. The distribution of cases across ages suggests ongoing transmission and underscores the need for additional interventions to prevent HBV infection.