1 INTRODUCTION

Hepatitis B virus (HBV), a partially double-stranded DNA virus classified in the genus Orthohepadnavirus, can cause acute and chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma. The World Health Organization estimates that 296 million people were living with chronic hepatitis B infection in 2019, with 1.5 million new infections and 820 000 deaths each year.¹ In Japan, the current number of HBV carriers has been estimated as approximately 1.1 million.^2-5

Routes of transmission for HBV include mother-to-child (vertical) transmission, horizontal transmission during infancy and early childhood, sexual transmission primarily after adolescence, and blood transfusion. Transfusion-transmitted HBV infection (TT-HBV) largely occurs via two types of donation: window period (WP) donation and donation by individuals with occult HBV infection (OBI), defined as HBV DNA positivity without detectable HBV surface antigen (HBsAg) in peripheral blood.^{6, 7}

The Japanese Red Cross (JRC) blood service is the national blood establishment in Japan and annually accepts approximately 5 million blood donations, of which approximately 93% are repeat donors. The JRC has achieved a high level of safety for blood products regarding HBV due to a testing algorithm using sensitive serological testing and nucleic acid amplification testing (NAT) with an individual donation format (ID-NAT). Nevertheless, several cases of TT-HBV have been reported within the past decade. Most such cases occurred after the transfusion of WP-related blood.⁸ On the other hand, we recently encountered a TT-HBV case in which the HBV infection phase of the implicated repeat blood donor was unknown; this was either a vaccine-breakthrough case or an anti-hepatitis B core (anti-HBc) negative OBI.⁹ We thus needed to revisit HBV prevalence among blood donors and determine detailed occurrence rates for HBV infection in each infection phase.

This study was conducted as a retrospective cohort analysis using donor records saved in the JRC database. The population of repeat blood donors with a continued donation history represents a good target for a study investigating the occurrence of disease acquisition. We looked at the incidence of HBV infection among blood donors who were all negative for HBV DNA, HBsAg, and anti-HBc and were followed for HBV DNA-positive conversion. We verified the frequencies of infectants in each infection phase including acute HBV infection, OBI, and HBV re-exposure, focusing on the influence of anti-HBs.

2 MATERIALS AND METHODS

3 RESULTS

4 DISCUSSION

We conducted this retrospective cohort study to look at HBV DNA conversions and changes in serological markers among a large sample of repeat blood donors and addressed which infection phase and what number of new infections are detectable with sensitive NAT, mainly focusing on the ability of anti-HBs to protect against HBV infection.

An anti-HBs titer ≥ 10 mIU/mL has long been considered protective against HBV infection for at least 20 years and more likely for life.^13-15 That value has also been used as the cutoff for anti-HBs testing. Our results reinforced this position, with all 115 HBV DNA- and HBsAg-positive cases found solely within the group showing preconversion anti-HBs titers < 10 mIU/mL. This held true for 72 sole DNA conversion cases without HBsAg, anti-HBc, or anti-HBs conversion (corresponding to pre-HBsAg WP donation), and the 15 HBV infectants presumed to be within the post-HBsAg viremic phase represented by both HBV DNA and anti-HBc positivity. Maximum preconversion anti-HBs titer allowing for the 202 new infections was 8.1 mIU/mL. Notably, the mean preconversion anti-HBs titer among these infectants was 0.63 mIU/mL, with 99% (201 of 202 cases) showing titers below 5.0 mIU/mL.

Note that the infection-protective ability of anti-HBs titer ≥ 10 mIU/mL could apply only to otherwise healthy individuals such as blood donors. In other populations such as immunocompromised patients, this anti-HBs level may not be sufficient to protect patients against HBV reactivation; HBV reactivation has been reported to occur more frequently in patients with anti-HBs between 10 and 100 mIU/mL than in those with > 100 mIU/mL.^{16, 17}

Although the group of 15 donors with anti-HBc plus HBV DNA conversion described above was considered to have experienced new infection within a short viremic period just after HBsAg disappearance, the possibility cannot be excluded that these donors were instead HBV vaccinees with no or low titer anti-HBs who were exposed to HBV, resulting in the establishment of infection as represented by positive anti-HBc and HBV DNA (vaccine breakthrough).

Another consideration is that the HBsAg-negative and HBV DNA-positive group could have included infections by HBsAg mutant strains,^{18, 19} which could alter the assignment of infection phase. The sensitivity of HBsAg assay is also critical for assignment; recent improvements in HBsAg assay sensitivity could comparatively decrease the number of cases assigned as OBI. Higher assay sensitivity for HBsAg could also assign a proportion of resolved HBV infection to chronic infection or reactivation,^{20, 21} necessitating a rethinking of the criteria for OBI.

One donor (Case 11 in Table 2) showed a decrease in anti-HBs titer from 13.4 to 3.4 mIU/mL, then acquired HBV infection 14 months later. That case represented a new infection because of the anti-HBc- and anti-HBs-negativity at preconversion. However, another interpretation would be HBV vaccine breakthrough if the positive anti-HBs at enrollment was considered.²² Although such a discussion would be influenced by the timing of blood testing before conversion, the necessity of boosting HBV vaccination would have to be discussed so that anti-HBs titer can be maintained at a certain level to protect vaccinees against HBV infection,^22-27 if a decrease in anti-HBs titer indeed increases vulnerability to HBV infection.

Infection phases of new HBV infection in this study included HBsAg-positive-, pre-HBsAg-, and post-HBsAg viremic phase.²⁸ The total incidence of such new HBV infection among anti-HBc-negative individuals was thus calculated as 2.38 per 100 000 PY. The incidence was increased by 76% by including incidences during the pre- and post-HBsAg phase as compared with the incidence obtained by evaluating HBsAg conversion alone. Note that this incidence does not include overt HB hepatitis because individuals with overt illness do not visit donation sites and those with a recent history of hepatitis are rejected for blood donation. Moreover, donors who had recovered from HB hepatitis and cleared HBV during the study period, typically represented as anti-HBc-conversion without detectable HBV DNA, were not included in this study.

In addition to the 202 new infections described above, we found 30 anti-HBc-negative OBIs (incidence: 0.35 per 100 000 PY; Tables 1 and 3). Preconversion serostatus for these donors was anti-HBc-negative and anti-HBs-positive, suggesting that these individuals comprised either HBV vaccinees or those with past natural HBV infection who had lost anti-HBc. Anti-HBc is usually considered to be sustained lifelong or at least longer than anti-HBs, but sometimes decreases faster than anti-HBs, resulting in anti-HBs-only individuals with resolved HBV infection.

Seven of the donors with anti-HBc-negative OBI (Cases 24–30 in Table 3) showed a > 4-fold increase in anti-HBs titer without anti-HBc conversion at DNA conversion, indicating a secondary antibody response against HBV contact. Mean age of these individuals was 35 years, comparable with new infections (37.9 years) and much younger than donors with anti-HBc-positive OBI (55.0 years). Preconversion anti-HBs titers ranged from 0.1 to 18.1 mIU/mL. Even vaccinees without detectable anti-HBs are well documented to have the ability to mount secondary responses against HBV exposure.^29-32 Notably, two of the seven cases showed elevated ALT (350 and 162 IU/L) on HBV DNA conversion, which might have been related to liver injury caused by HBV infection. Liver injury is known to be mainly associated with exacerbated anti-HBV cellular immune response. Although a secondary response in vaccinees was suggested for these seven cases, these cases are presented as anti-HBc-negative OBI with immune response because of the lack of a clear demonstration of vaccination history for those donors. The other 23 donors (Cases 1–23 in Table 3) turned positive for HBV DNA in the virtual absence of both anti-HBc conversion and anti-HBs titer increase. We deemed that at DNA conversion, those donors were temporarily viremic with pre-existing anti-HBs neutralizing contact with HBV without any significant change in anti-HBs titer. Deng et al. reported that anti-HBc-nonreactive anti-HBs-detectable viremic blood donors are mostly vaccinees exposed to HBV who later develop aborted infection.³³ Interestingly, stable anti-HBs level over DNA conversion was observed mostly in those with high anti-HBs (Cases 1–23, anti-HBs ≥ 10 mIU/mL), whereas vigorous immune response was mostly observed in the group with low anti-HBs (Cases 24–30, anti-HBs < 10 mIU/mL) (Table 3). High titers of pre-existing anti-HBs may be able to confront HBV as is. Alternatively, only young individuals among the 30 donors may have shown measurable increases in anti-HBs titer.

The possibility cannot be excluded that the DNA conversion has been caused by the infection by HBsAg mutant strain as the current hepatitis B vaccines are incapable of preventing infection by such strains. Although sequencing a determinant may determine whether the infection was caused by wild-type or mutant strain, it was infeasible because of nonavailability of relevant samples.

Another explanation for the 23 anti-HBc-negative DNA converters is that they represented OBI just showing occasional DNA positivity as a facet of fluctuating viremia. In line with this idea, immune response at HBV exposure was absent for these cases. Notably, the mean age for these 23 cases was 59 years, much higher than that of the 7 donors with apparent immune response described above (35 years). In elderly individuals, HBV may intermittently spill over from the infected liver to the peripheral blood or undergo mild reactivation beyond the host defense mechanisms.

Clinical outcomes for these 30 cases were unknown; some may have ended up with temporary viremia without any clinical sequel or with “aborted” infection without developing anti-HBc.^{33, 34} Others may have developed overt hepatitis afterward. Anti-HBc-negative OBI may thus include vaccinees in the midst of immune response, status before vaccine breakthrough, abortive HBV infection, and OBI with temporary viremia reflecting fluctuating viremia. Although vaccine breakthrough may be diagnosed for a vaccinee with only positivity for anti-HBs and HBV DNA, HBV infection should not be determined for such cases at least until subsequent anti-HBc development is confirmed.^{10, 35, 36} However, even the development of anti-HBc does not necessarily indicate vaccine breakthrough since a considerable proportion of vaccinees may experience self-limited HBV infection without anti-HBc production.³⁷

We identified 11 anti-HBc-positive OBIs in this study.³⁸ These donors, mostly older individuals (mean age, 55 years), showed stable anti-HBc S/CO and anti-HBs titers over HBV DNA conversion. All showed stable ALT levels at < 35 IU/L. DNA positivity in these cases thus seems to reflect fluctuating low viremia, not true reinfection.^39-41 The incidence of anti-HBc-positive OBI among anti-HBc-positive population is 51 times as high as that of DNA conversion among anti-HBc-negative population.

Limitations of this study included the fact that the target population comprised blood donors who themselves felt healthy. Individuals who had a present illness or a recent history of hepatitis were ineligible to donate blood. These conditions would inevitably have resulted in underestimated incidences in this study. Vaccination history or clinical course after HBV DNA conversion for each donor were unavailable. Viral load in HBV DNA-positive samples was also not available. Determination of the infection phase for each donor depends on the sensitivity and cutoff of anti-HBc, anti-HBs, or HBsAg of the testing system used, so different results might be brought about when another system is employed.

In conclusion, new HBV infection was found only in the population with low anti-HBs level (< 10 mIU/mL). The incidence of new HBV infection was 2.38 per 100 000 PY among donors all negative for HBV DNA, HBsAg, and anti-HBc, a result increased by probing HBV DNA conversion. The population of blood donors is heterogeneous regarding HBV infection. In addition to acute infection and anti-HBc-positive OBI, blood donors show statuses including anti-HBc-negative OBI potentially comprising vaccinees in the midst of immune response, status before vaccine breakthrough, abortive HBV infection, and OBI with temporary viremia reflecting fluctuating viremia.

AUTHOR CONTRIBUTIONS

Masahiro Satake conceived the study, analyzed the data, and wrote the manuscript. Masaya Sugiyama, Masashi Mizokami, and Junko Tanaka critically reviewed and revised the content. Junko Tanaka edited the manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS APPROVAL

This study was approved by the institutional ethics committee of the Japanese Red Cross (approval no. 2018-049-1).