Comparative analysis of hepatitis B virus infections in blood donors born before and after the implementation of universal HBV vaccination in southern China

In China, the vaccinated blood donors have rapidly increased by recent years, which may impact blood safety. The true prevalence of HBV between vaccinated blood donors and non‐vaccinated blood donors should be explored.

frequency of vaccine escape mutations M133L (32.4%) and E164G in S region of genotype B strains and substitution L175S (40.9%) related to vaccine escape in S region of genotype C strains were identified.
Conclusion: The universal HBV vaccination program markedly reduces the risk of HBV infection in blood donors, and provides a significant guarantee for the safety of blood transfusion. Several important mutations detected related vaccine escape and notable mutations needed further investigated. HBV infection continues to be a major public health concern worldwide despite the availability of an effective vaccine and potent antiviral treatments. 1 One-third of the world's population is predicted to have been infected by HBV. 2 The WHO estimates that 257 million people are currently living with a HBV infection and approximately 885 000 deaths were cause by HBV-associated complications in 2015. 3 In China, nearly 50% of the Chinese population has a history of HBV infection. 4 To curb hepatitis B epidemic, the universal HBV vaccination has been implemented by Chinese government from 1992 and resulted in HBsAg carrier reduction significant from 10% to <1% in children group over two decades. 5  However, previous data suggested that HBV vaccine made of genotype A2 recombinant protein might not be fully efficient when used in other genotype prevalent areas. 10 To investigate the true epidemiology of HBV infection and the effectiveness of the vaccination program for blood safety, a large group of Chinese blood donors, including presumed vaccinated or non-vaccinated recruited by SZBC, were tested and identified for HBV by various serologic and molecular assays. This study was intended to examine the effectiveness of the vaccination program and explore the potential threat of vaccination failures to the blood supply.

| MATERIAL AND METHODS
A total of 26 318 blood donors were enrolled in this study from Jan 2016 to Jun 2016. The presumed vaccinated and non-vaccinated blood donor populations were separated into two groups, according to being born before or after 1 Jan, 1992. Blood donors born after 1 Jan, 1992, were designated as vaccinated donors. Blood donors born before 1 Jan, 1992, were selected as non-vaccinated donors. There 6421 (24.4%) donors for vaccinated groups and 19 897 (75.6%) donors for the non-vaccinated groups were available in this study. The pre-donation questionnaire, rapid pre-donation testing and dual ELISA assays for routine screening are described as previous study. 7 Blood samples with HBsAg reactivity were re-tested in duplicate using at least one ELISA assay. Samples were determined to be HBsAg ELISA+ if results were reactive in any assay in re-testing. The donors were defined as first-time blood donors who gave blood for the first time, while donors were defined as repeat-donors who donated blood more than once at the SZBC. Multiple Procleix ultrio plus assay (Grifols diagnostic solutions, Inc. and hogic) was used to detect HBV (limit of detection (LOD):3.4 IU/ml), HCV and HIV-1 genomes in all donors. Individual reactive samples were further tested with a discriminatory Procleix Ultrio plus test to identify the virus responsible for NAT reactivity (HBV, HCV, or HIV-1) as the manufacturer recommends.
If necessary, MPX2.0 ID HBV NAT (LOD: 2.3 IU/ml) was used as an alternative HBV NAT assay for further identification. Serum and aliquots of the index-retrieved frozen plasma unit for HBsAg ELISA+ and/ or NAT initial reactive were collected for additional determination.

| HBV DNA confirmation
A 200-2500 μl HBsAg ELISA + and / or NAT initial reactive samples were extracted for HBV DNA by HighPure Viral Nucleic Acid Large Volume Kits(Roche Diagnostics Gmbh, Mannheim, Germany). The basic core promoter / pre-core (BCP / PC) and S region were amplified by qPCR (LOD:5 IU/ml) and nested PCRs (LOD:10 IU/ml) for further detection. 7,8 As mentioned above, the samples reacted by any two of the five NAT methods were confirmed to be DNA positive.
Donors that tested HBsAg+ and HBV DNA+ with anti-HBc+ were designed as chronic hepatitis infections (CHB), donors that tested HBsAg-and HBV DNA+ with anti-HBc+ and/ or anti-HBs+ were OBIs, and some donors with HBsAg-, anti-HBc-, anti-HBs-but HBV DNA+ were designed as serological window period (WP). 6

| HBV DNA sequencing, genotyping and comparison
To confirm HBV genotypes, the amplified PCR products obtained from the BCP/PC (295 bp) and the S regions (495 bp) were sent to Shanghai Invitrogen Co., Ltd. (Guangzhou, China) for sequencing. HBV genotype determination was performed by phylogenetic analysis using the MEGA7.0 program. The neighbour-joining method based on Kimura 2-parameter mode and complete deletion for gaps with 1000 bootstrap replications was used. 11 Bootstrap values of 70% or greater were considered significant. Amino acid sequences isolated from HBsAg+ blood donors from China, Thailand and Malaysia were used as reference sequences. 12  HBsAg ELISA-and NAT-(see Figure 1).  (Table 1, Table 2); however, no significant differences in the distribution of HBV genotype, and the median of anti-HBs titers between the two groups were found (p > 0.05).  (Table 3)

| DISCUSSION
China is endemic for hepatitis B, and the residual risk of transfusion transmission is significantly higher for HBV (1:17501) than for HIV-1 (1:903498) and HCV (1:59588). 22 HBV NAT has been preliminarily introduced in some major blood centres in China since 2003, and the HBV NAT yield ranged between 1:1000 to 1:10000. 23 In China, neonatal vaccination resulted in a decrease of HBV incidence. 5 However, vaccination may also favour the development of escape mutants and neutralising anti-HBs antibodies level decrease over time in vaccinated people who may become susceptible to HBV infection. 24  study, algorithms of confirmatory testing and supplemental testings are adopted to avoid false-positive results and identified the lowvirus-load donations utmostly as many investigations. 23 Confirmed positivity was based on multiple assay reactivity or sequences generated as accurate as possible. These measurements would give a better comparison between vaccinated blood donors and non-vaccinated blood donors for the prevalence of HBV, HBsAg and OBI. In this study, 242 reactive by NAT and/or HBsAg ELISA samples from 26 318 candidate blood donors were investigated, 23 initial HBsAgpositive and 27 HBV DNA results could not be confirmed by further testing. In total, of 192 confirmed HBV infected blood donors, 61 (31.3%) cases were HBsAg-/DNA+, and 131 (60.1%) were HBsAg+/DNA+ due to adoption of large volume DNA extraction assay. Therefore, the confirmed HBsAg positive rate is 0.50%, which coincides with the report. 25 Of 54 NRR samples, 30 donations were clarified HBsAg-/DNA+, a higher NAT yield rate was confirmed compared with the multi-regional study 26 and Hongkong study. 27 The true interdiction HBV DNA positive rate by Ultrio Plus ID-NAT screening in combination with HBsAg was 0.73%, higher than true HBsAg positive rate(0.50%, p = 0.001), and the true OBI yield was 1:453 (58/26318), nearly had two-fold increase compared with a previous study in Shenzhen 6 due to application of ID NAT. For The Ultrio Plus assay has used a target enhancer reagent, which helps to disrupt viral particles and exposes more single-stranded DNA for the capture probe. This assay modification increased the proportion of OBI yield at least more twofold than the Ultrio assay. 27 Moreover, in the NAT NRR donations, 31.5% were got sequences, and 38% were qPCR positive due to shorter length of the primer, overall, 55% were identified HBV DNA positive in which 100% were low-virus-load OBI, gained another half of OBI cases. In our study with those reported in other regions, the differences of the rates varied considerably depending on HBV epidemiology, the proportion of repeat or the first time donors, NAT sensitivity, and pooling strategy used; for example, 1:624 in Xiamen China, 28 1:3471 in Hongkong in China, 27 1:894 in Taiwan in China, 29 1:4232 in Thailand 30 and 1:770000 in Germany. 31 During the 5 months of study, HBV prevalence in vaccinated donors aged 18-24 years was lower than in non-vaccinated donors aged 25-60 (p < 0.05). Although there was no difference in genotype distribution between OBI infected vaccinated and non-vaccinated individuals, OBI yield rates were also confirmed lower than in non-vaccinated donors (p < 0.05). This difference might be related to the increasing cumulative HBV exposure with the ages, and vaccinerelated protection would be the definite cause. The OBIs with anti-HBs in present study suggested that OBIs occur primarily in individuals who have recovered from the infection but are unable to develop an effective immune control. 32 Furthermore, among OBI samples, the percentage of those carrying anti-HBs in vaccinated OBI blood donors is lower than the non-vaccinated OBI blood donors (p < 0.05), suggested that lower level or less of anti-HBs were insufficiently protected and are susceptible to infection associated with breakthrough or occult HBV infections, even when vaccinated at birth. Interestingly, in HBsAg positive vaccinated blood donors, the titers are much higher than non-vaccinated blood donors (p < 0.005), because the infected vaccinated blood donors are no response or warning off vaccines, HBsAg is secreted more than in usual after infection.
HBV with HBsAg escape mutants are rare but potentially highly infectious and pathogenic, particularly in immune-compromised recipients. 33 The prevalence of the well-known neutralisation escape mutation G145R in the HBV envelope protein was as high as 22% in American blood donors. 34 However, in our study, the G145R mutation was not found in genotype B; only three cases harboured G145A/E/R individually were found in genotype C donations. Furthermore, compared to G145R mutation, several mutations within and out of MHR occurred at high frequency such as M134L and L175S. Surprisingly, the isolates from the CHBs non-vaccinated genotype B group also showed high variability in their S gene sequences in comparison with CHBs vaccinated genotype B group. However, the frequency of M133L (52.4% vs. 16.9%) and E164G (14.3% vs. 3.1%) associated with escape from vaccine-induced immunity was observed higher significantly in CHBs vaccinated group than in the CHBs non-vaccinated group (p < 0.05). This is because antibodies induced by the current vaccine may not recognise changes in the surface antigen as a result of mutation. In the genotype C non-vaccinated group, lots of mutations out of MHR such as Q34K, N40S, T47V, P49H, S55F, L175S, V177A were detected. These mutations are associated with major histocompatibility complex (MHC) class I-restricted cytotoxic CD8+ lymphocytes (CTLs) epitopes, and it has been experimentally proved that adaptive immune response mediated by CTLs is necessary for controlling HBV infection. This may be because mutations in CTL epitopes can evade cellular immunity and contribute to persistence, and are potentially responsible for vaccine breakthrough infection and HBsAg undetectability. 1 Interestingly, two membrane-embedded C-terminus mutations L175S, V177A were observed at high frequency in genotype C donors in the present study and proved tightly to correlate with OBI, and powerfully to affect HBsAg detection. 19 In conclusion, the prevalence of HBV+ and OBI in vaccinated donors is lower than in non-vaccinated donors (p < 0.05), suggested that the universal HBV vaccination program markedly reduces the risk of HBV infection in blood donors. Furthermore, there is a high frequency of mutations in the MHR and out of MHR of the HBV S gene, which may cause vaccine escape, diagnosis failure and failure in HBIg therapy problem and highlights the need for more studies into the prevalence of mutants.