Seroprevalence of hepatitis B and immune response to hepatitis B vaccination in Chinese college students mainly from the rural areas of western China and born before HBV vaccination integrated into expanded program of immunization

Abstract

The prevalence of hepatitis B surface antigen (HBsAg) in a population aged 15 y or older was high in China, but an immunization strategy for this population was unavailable. We investigated the seroprevalence of hepatitis B and immune response to HBV vaccine in Chinese college students (n = 2040 participants), 11.1%, 80.8%, and 8.1% had confirmed, unknown and no HBV vaccination history, respectively. The seropositive rates for HBsAg, anti-HBs sole and anti-HBs plus anti-HBc were 12.6%, 25.7%, and 30.1%, respectively. The HBsAg seropositive rate was significantly lower in participants with confirmed HBV vaccination history than in those with unknown or no vaccination history (5.3%, 13.6%, and 12.6%, respectively, P = 0.0019). The anti-HBs alone seropositive rate was significantly higher in participants with confirmed HBV vaccination history than in those with unknown or no vaccination history (37.6%, 25.3%, and 13.8%, respectively, P < 0.0001). Participants negative for HBsAg, anti-HBs, and anti-HBc at baseline (n = 600) were given three doses of recombinant HBV vaccine (GlaxoSmithKline) at month 0, 1, and 6. Robust immune response was elicited after two and three doses (seroprotective rate: 91.9% and 99.0%, respectively, and geometric mean concentration [GMC]: 95.8 and 742.6 IU/L, respectively). Fourteen months after the third dose, the anti-HBs seroprotective rate of the group remained more than 97%. The seroprotective rates and GMCs did not differ significantly by vaccination history. This study suggested that three doses of 20 μg HBV vaccine were needed for college students negative for HBsAg, anti-HBs, and anti-HBc to ensure high seroprotective rates and concentrations.

Keywords: :

• hepatitis B
• seroprevalence
• hepatitis B vaccine
• immune response
• college students

Introduction

Hepatitis B is a serious infectious disease caused by hepatitis B virus (HBV). HBV is mainly transmitted by blood or blood products, mother-to-child, sexual exposure, and close contact involving open wounds on skins or mucous membranes. Vaccination is by far the most cost-effective way to prevent HBV infection, and in turn, to reduce the incidence of acute and chronic hepatitis B, hepatic cirrhosis, and heptocellular carcinoma.¹

Hepatitis B (HB) prevention has long been focused on newborns, because China is a highly endemic area where hepatitis B mother-to-child transmission is very common. In 1992 the Chinese Ministry of Health recommended HBV vaccine for routine immunization of infants but parents had to pay for vaccinations. China integrated HBV vaccines into the Expanded Program of Immunization (EPI) in 2002, when the cost of HBV vaccines was paid by the government while administration fees of HBV vaccinations (up to $1.10 per dose) had to be charged to the parents. Since 2005, China’s government has required that all infants should be immunized with the HBV vaccine free of charge. According to the national serological surveys, in 1992 and 2006; it was estimated that from 1992 to 2006, the children infected with HBV had been reduced by nearly 80 million, and the children carriers of hepatitis B surface antigen (HBsAg) had been reduced by 19 million. However, the prevalence of HBsAg in adolescents and adults more than 15 y of age was still as high as 8.57%, where the decline was not significant. HBsAg carrier rate showed a rising trend with increasing age. The 2 surveys showed that the carrier rate for the whole population had slightly declined from 9.75% to 7.18%, which was still at a high level.²

In order to reduce the whole population’s prevalence of HBV infection, worldwide experience has confirmed that vaccinations should also be given to children and adolescents to block the horizontal transmission, in addition to routine vaccinations in newborns for interruption of mother-to-child transmission. Taiwan started vaccinations in neonates born to mothers of HBV carriage in 1984 and then gradually expanded the vaccination target population, which achieved good results.³ Currently, Canada, the United States, Australia, Greece, Malta, Germany, France, Belgium, Italy, Spain, and other countries with low prevalence of HBV infection have included adolescents into routine immunization.^4,5

At present, children less than 15 y of age have been regarded as a key group for hepatitis B immunization in China. However, to date, there is no special immunization strategy for the population 15 y of age and older. In this study, we investigated the seroprevalence of hepatitis B and immune response to HBV vaccine among Chinese college students to uncover the need for universal mass vaccination or booster immunization only for students with HBV vaccination.

Results

Demography

A total of 2040 participants were recruited including 366 males (17.9%) and 1674 females (82.1%) with a male-to-female ratio of 1: 4.57. All of the participants were born between 1982 and 1992 and aged 17–27 y with a median age of 19.7 y. With respect to ethnicity, 60.8%, 30.8%, and 5.9% of the participants were Han, Zhuang, and others respectively. All participants were born in Guangxi Zhuang Autonomous Region and 70% were from a rural area (Fig. 1).

Figure 1. Study flowchart.

HBV vaccination history

Among the 2040 participants, 226 (11.1%) had a confirmed HBV vaccination history, of whom 215 (10.5%) had completed a serial of three doses and 11 (0.5%) had not. Among the participants, 1648 (80.8%) had an unknown HBV vaccination history and 166 participants (8.1%) had never received any HBV vaccine. Of the 215 participants receiving full doses of vaccination, 65 (30.2%) received their first dose within 24 h after birth, while the rest received their first dose beyond 24 h after birth (Table 1). An amount of 129 (60%) of the 215 participants received a plasma-derived HBV vaccine in their full doses of vaccination before 2000 and the remaining 86 (40%) received recombinant vaccines in their full doses of vaccination after 2000.

Table 1. Baseline seropositive rates of HBsAg and anti-HBc and seroprotective rate of anti-HBs by HB vaccination history

Vaccination history	N (%)	HBsAg	Anti-HBc alone	Anti-HBc+anti-HBs both positive	Anti-HBs alone	All negative	Anti-HBs GMC1(IU/L)	Anti-HBs GMC2(IU/L)
Confirmed	226 (11.1)	12 (5.3, 2.8–9.1)	5 (2.2, 0.7–5.1)	73 (32.3, 26.3–38.8)	85 (37.6, 31.3–44.3)	51 (22.6, 17.3–28.6)	41.3 (27.7–61.8)	49.3 (32.9–74.0)
1 or 2 doses	11 (0.5)	0 (0, 0–28.5)	0 (0, 0–28.5)	8 (72.7, 39.0–94.0)	2 (18.2, 2.3–51.8)	1 (9.1, 0.2–41.3)	137.3 (30.7–614.2)	137.3 (30.7–614.2)
3 doses	215 (10.5)	12 (5.6, 2.9–9.5)	5 (2.3, 0.8–5.3)	65 (30.2, 24.2–36.8)	83 (38.6, 32.1–45.5)	50 (23.2, 17.8–29.5)	38.9 (25.7–58.9)	46.7 (30.6–71.1)
Within 24 hr after birth	65 (3.2)	1 (1.5, 0.04–8.3)	1 (1.5, 0.04–8.3)	15 (23.1, 13.5–35.2)	30 (46.2, 33.7–59.0)	18 (27.7, 17.3–40.2)	27.7 (12.9–59.9)	29.7 (13.7–64.1)
Beyond 24 hr after birth	150 (7.3)	11 (7.3, 3.7–12.7)	4 (2.7, 0.7–6.7)	50 (33.3, 25.9–41.5)	53 (35.3, 27.7–43.5)	32 (21.3, 15.1–28.8)	42.8 (26.0–70.5)	52.5 (31.6–87.2)
Unknown	1648 (80.8)	225 (13.6, 12.0–15.4)	29 (1.8, 1.2–2.5)	486 (29.5, 27.3–31.8)	417 (25.3, 23.2–27.5)	491 (29.8, 27.6–32.1)	14.2 (12.2–16.6)	22.2 (18.8–26.1)
None	166 (8.1)	21 (12.6, 8.0–18.7)	8 (4.8, 2.1–9.3)	56 (33.7, 26.6–41.5)	23 (13.8, 9.0–20.1)	58 (34.9, 27.7–42.7)	11.6 (6.9–19.4)	18.3 (10.6–31.6)
Total	2040 (100.0)	258 (12.6, 11.2–14.2)	42 (2.1, 1.5–2.8)	615 (30.1, 28.2–32.2)	525 (25.7, 23.8–27.7)	600 (29.4, 27.4–31.4)	15.7 (13.7–18.1)	24.0 (20.7–27.8)

Data are presented as participant number or value (proportion of participants and/or 95%CI). N, Number of participants with different HB vaccination history; %, proportion of participants with different HB vaccination history among total participants; 95%, 95% confidence interval; GMC1, calculated on all the participants with specific vaccination history; GMC2, calculated on the participants other than HBsAg positive. For HBsAg seropositive rate: P = 0.0019 among the participants with confirmed, unknown or none vaccination history; P = 0.0004, 0.0097, 0.7192 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively; within 24 h vs. beyond 24 h: P = 0.1123. For anti-HBs alone seroprotective rate: P < 0.0001 among the participants with confirmed, unknown or none vaccination history; P < 0.0001, < 0.0001, = 0.0010 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively; within 24 h vs. beyond 24 h: P = 0.1345. For GMC1: P < 0.0001 among the participants with confirmed, unknown or none vaccination history; P < 0.0001, = 0.0002, = 0.5083 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively; within 24 h vs. beyond 24 h: P = 0.3482. For GMC2: P = 0.0014 among the participants with confirmed, unknown or none vaccination history; P = 0.0006, 0.0035, 0.4855 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively; within 24 h vs. beyond 24 h: P = 0.2172. For multiple comparison, Bonferroni adjustment was used with α’ = α/month = 0.05/3 = 0.0167.

Baseline seroprevalence

The baseline seroprevalence for HBsAg, anti-HBs, and anti-hepatitis B core antigen (anti-HBc) is shown in Table 1.

The HBsAg seropositive rate in participants with confirmed HBV vaccination was significantly lower than in those without HB vaccination history (P = 0.0097) or with unknown HB vaccination history (P = 0.0004). The anti-HBs alone seroprotective rate was significantly higher in participants with confirmed HBV vaccination than in those without HBV vaccination (P < 0.0001) or with unknown HBV vaccination history (P < 0.0001). Anti-HBc prevalence was significantly lower in participants with confirmed or unknown HBV vaccination than in those without HBV vaccination history (both P < 0.01). There was no statistical difference between those with confirmed HBV vaccination and with unknown HB vaccination history (P > 0.05).

Of the 215 participants with confirmed HBV vaccination history, 65 participants received their first dose within 24 h after birth and had a lower HBsAg seropositive rate than those receiving their first dose beyond 24 h after birth, although there was no statistical difference (1.5% vs.7.3%, P = 0.1123).

Immune response to HBV vaccine in participants seronegative for HBsAg, anti-HBs, and anti-HBc at baseline

The 600 participants seronegative for HBsAg, anti-HBs, and anti-HBc at baseline were given a HBV vaccine, of whom 414, 407, 406, and 399 participants had post-vaccination serological results available at month 1, 6, 7, and 20, respectively, and were included in immunologic analysis (Fig. 2; Table 2). Regardless of the participants HBV vaccination histories, one dose induced mild immune response while robust immune response was elicited with anti-HBs seroprotective rate of more than 87% after 2 doses (month 6) and more than 98% after three doses (month 7). Fourteen months after the third dose, the anti-HBs seroprotective rates remained more than 97%. At each time point after HBV vaccination, the seroprotective rates and geometric mean concentrations (GMCs) did not show statistical difference among participants having confirmed, none or unknown HBV vaccination history. The anti-HBs response distribution as expressed with proportion of participants having anti-HBs concentration <10 IU/L, 10–100 IU/L, ≥100 IU/L among participants with different HBV vaccination history, was statistically different at month 7 (P = 0.0079) but not at month 1, 6, and 20 (P = 0.6333–0.9266). After 1, 2, and three doses, 21.7%, 51.1%, and 96.1% of participants had anti-HBs concentration more than 100 IU/L. Even 14 mo after the third dose, 81.0% of participants had anti-HBs concentrations more than 100 IU/L.

Figure 2. Anti-HBs seroprotective rate after HB vaccination given at month 0, 1, and 6 among participants seronegative for HBsAg, anti-HBs, and anti-HBc at baseline.

Table 2. Anti-HBs response after HB vaccination given at month 0, 1, and 6 among participants seronegative for HBsAg, anti-HBs, and anti-HBc at baseline

Time after dosing	vaccination history	<10 IU/L	10–100 IU/L	≥100 IU/L	GMC (IU/L)
Month 1	All (n = 414)	287 (69.3, 64.6–73.7)	37 (8.9, 6.4–12.1)	90 (21.7, 17.9–26.0)	4.4 (3.3–5.9)
	None (n = 43)	33 (76.7, 61.4–88.2)	3 (7.0, 1.5–19.1)	7 (16.3, 6.8–30.7)	3.3 (1.5–7.5)
	Confirmed (n = 41)	25 (61.0, 44.5–75.8)	5 (12.2, 4.1–26.2)	11 (26.8, 14.2–42.9)	8.4 (3.0–23.3)
	Unknown (n = 330)	229 (69.4, 64.1–74.3)	29 (8.8, 6.0–12.4)	72 (21.8, 17.5–26.7)	4.3 (3.1–5.9)
	P value	P1 = 0.6333	P = 0.3029
Month 6	All (n = 407)	33 (8.1, 5.6–11.2)	166 (40.8, 36.0–45.7)	208 (51.1, 46.1–56.1)	95.8 (81.7–112.5)
	None (n = 41)	5 (12.2, 4.1–26.2)	16 (39.0, 24.2–55.5)	20 (48.8, 32.9–64.9)	85.4 (48.7–149.9)
	Confirmed (n = 40)	4 (10.0, 2.8–23.7)	16 (40.0, 24.9–56.7)	20 (50.0, 33.8–66.2)	96.5 (58.8–158.2)
	Unknown (n = 326)	24 (7.4, 4.8–10.8)	134 (41.1, 35.7–46.7)	168 (51.5, 46.0–57.1)	97.1 (81.2–116.2)
	P value	P2 = 0.7910	P = 0.8945
Month 7	All (n = 406)	4 (1.0, 0.3–2.5)	12 (3.0, 1.5–5.1)	390 (96.1, 93.7–97.7)	742.6 (680.5–810.2)
	None (n = 41)	0 (0, 0.0–8.6)	5 (12.2, 4.1–26.2)	36 (87.8, 73.8–95.9)	611.6 (446.6–837.6)
	Confirmed (n = 41)	0 (0, 0.0–8.6)	1 (2.4, 0.1–12.9)	40 (97.6, 87.1–99.9)	815.0 (677.6–980.4)
	Unknown (n = 324)	4 (1.2, 0.3–3.1)	6 (1.9, 0.7–4.0)	314 (96.9, 94.4–98.5)	752.1 (680.8–830.9)
	P value	P > 0.9999	P = 0.0066	P = 0.0325	P = 0.2957
Month 20	All (n = 399)	10 (2.5, 1.2–4.6)	66 (16.5, 13.0–20.6)	323 (81.0, 76.7–84.7)	308.1 (268.2–354.0)
	None (n = 32)	0 (0, 0.0–10.9)	6 (18.8, 7.2–36.4)	26 (81.3, 63.6–92.8)	444.8 (293.5–674.1)
	Confirmed (n = 37)	0 (0, 0.0–9.5)	6 (16.2, 6.2–32.0)	31 (83.8, 68.0–93.8)	325.4 (211.4–500.9)
	Unknown (n = 330)	10 (3.0, 1.5–5.5)	54 (16.4, 12.5–20.8)	266 (80.6, 75.9–84.7)	295.6 (252.8–345.5)
	P value	P3 = 0.9266	P = 0.2852

Data are presented as participant number or value (proportion of participants and/or 95%CI). P1, P2, P3: comparison of anti-HBs distribution among the participants with confirmed, unknown or none vaccination history. At month 7 for 10–100 IU/L: P = 0.5689, 0.2012, 0.0039 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively. At month 7 for ≥100 IU/L: P > 0.9999, = 0.2012, = 0.0180 for confirmed vs. unknown, confirmed vs. none, unknown vs. none, respectively. For multiple comparison, Bonferroni adjustment was used with α = α/month = 0.05/3 = 0.0167.

Response to additional three doses of HBV vaccine in non-responders

Among the 330 participants with unknown HBV vaccination history and negative for all the three serological markers, four non-responders after the initial three doses of the HBV vaccine were administrated an additional three doses from month 20 onward at the regimen of month 0, 1, and 6, and achieved seroprotective anti-HBs. One participant had high anti-HBs response (3385.5 IU/L), one had middle response (306.1 IU/L), and another 2 participants had low response (17.5 IU/L and 64.4 IU/L).

Response to HBV vaccine in participants seropositive for anti-HBc alone

In addition to the participants negative for all the three serological markers, the 42 students with anti-HBc positive alone were also given three doses of HBV vaccine. The anti-HBs response is shown in Figure 3. Sixty-six percent (28/42) of participants had seroprotective anti-HBs and 28.6% (12/42) had higher than 100 IU/L anti-HBs one month after the first dose of the HBV vaccine. After 2 doses, 85.7% (36/42) of participants had seropretective response and 45.2% (19/42) participants had anti-HBs with higher than 100 IU/L. Thirty-eight subjects (90.5%) had seroprotective response after three doses, of which 34 (89.5%) participants had anti-HBs concentrations higher than 100 IU/L. The GMCs at month 1, 6, and 7 were 19.3 (9.0–41.5), 67.9 (38.2–120.4), and 328.4 (188.8–571.1), respectively. There were four non-responders and an additional three doses of HBV vaccine were given to them. One non-responder negative for HBV DNA achieved a low level of anti-HBs concentration (84.5 IU/L). The remaining three non-responders who were positive for HBV DNA did not develop seroprotective anti-HBs.

Figure 3. Anti-HBs response after HB vaccination given at month 0, 1, and 6 among participants seropositive for anti-HBc alone at baseline.

Discussion

According to the seroepidemiologic survey conducted in China in 2006, the prevalence rates of HBsAg and anti-HBs in the population aged 20 to 25 y were 12% and 46%, respectively, highlighting the importance of having an immunization strategy of HBV vaccines for this population.² Most of college students enjoy participating in sports activities which are associated with an increased risk of accidental injuries. They are in a period of sexual activity and may engage in unprotected sexual activities. In addition, they are in close contact with each other due to crowded living, by which they may share daily necessities such as toothbrushes, razors, etc. According to the Technical Guide for Adult Hepatitis B Immunization in China, college students are at high risk for HBV.⁶ Previous studies showed that the prevalence rates of HBsAg in Chinese college students were 4% to 13%.^7-10 Therefore, immunizing college students with HBV vaccines should be considered as an important part of the immunization strategy to lower the prevalence of HBsAg.

The HBV vaccination histories in children were not routinely recorded in China before 1992. Most studies obtained HBV vaccination histories by questionnaires, which could introduce recall bias. In order to get the HBV vaccination history as exact as possible in this study, the record of HBV vaccination was retrieved through the Centre for Disease Control and Prevention (CDC) system. There was no significant difference in the HBsAg prevalence between participants who had definitely not received hepatitis B vaccination and those with an unknown history of vaccination, possibly because most of the students with an unknown vaccination history might not have been vaccinated or failed to complete the full-course immunization. HBV vaccination was not recommended before 1992 in China. So the HBV vaccine coverage was very low, especially in rural and remote areas. Guangxi is located in a poor and remote area of China. According to a national survey conducted in 1999 in 31 provinces of China, the HBV vaccine coverage rates in children younger than 3 y of age in different provinces or autonomous regions of China were uneven. Four provinces had the coverage ≥90%, 9 provinces 75–89%, 8 provinces 50–74%, and 10 provinces <50% (including Guangxi).¹¹

Our study found that the HBsAg prevalence rate (12.6%) in the college students studied was higher, and anti-HBs prevalence (25.7%) was lower as compared with the results of the national serosurvey conducted in 2006 (7.2% and 50.1%, respectively).² Although the HBsAg prevalence in participants with confirmed HBV vaccination history was up to 5.3%, it was significantly lower than that in participants without HBV vaccination history (12.6%) and those with unknown vaccination history (13.6%). Seventy (32.5%) participants who had a confirmed history of vaccination, with three doses, were positive for anti-HBc, in comparison with 13/81 (16.0%) in Qian’s study.¹² This difference was likely because of the failure in timely HBV vaccination with the first dose. As pointed out in the WHO’s position paper, perinatal or early childhood transmission was the main cause of chronic HBV infection throughout the world. Thus, the first dose of hepatitis B vaccine should be administered after birth as soon as possible (within 24 h).¹³ Delay in the birth dose resulted in an increased risk of HBV infection.¹⁴ In this study, among the 215 participants who had confirmed HBV vaccination of three doses, 65 (30%) and 150 (70%) participants received the first dose within and beyond 24 h after birth, respectively. One out of the 65 participants (1.5%) and 11 out of the 150 participants (7.3%) were positive for HBsAg. Although there was no statistically significant difference (P = 0.1123), the HBsAg prevalence was lower in the participants receiving the first dose within 24 h after birth than beyond 24 h after birth. Therefore, the rate of timely vaccination with the first dose was very important to prevent HBV infection. In addition to the timeliness of the first dose, the study in Thailand found the risk of infants born to mothers with chronic HBV infection becoming chronically infected was associated with a delayed second dose of HBV vaccine, even if the first dose had been immunized promptly.¹⁵ In this study, this issue could not be analyzed due to the small sample size of participants with delayed second dose of HBV vaccine. Incomplete HBV vaccination also resulted in increased risk of chronic HBV infection.¹⁶ Our study found 11 out of the 226 participants (4.9%) with confirmed HBV vaccination histories had not completed their three doses of HBV vaccine. Although the 11 participants did not become chronic HBV infection, 8 of them (72.7%) had anti-HBc positive, indicating a natural infection with HBV. For the participants with three doses of HBV vaccination, only 32.5% (70/215) had anti-HBc marker.

In China, recombinant HBV vaccine totally replaced plasma-derived HBV vaccine after 2000. In this study, all students were administrated with plasma-derived HBV vaccine in their childhood. A lot of studies have demonstrated that both types of HBV vaccines performed very well with respect to safety, immunogenicity and protective efficacy.^17-19 Therefore, the reason of relatively low seroprotective rates of anti-HBs alone in this study population may account for the lower HBV vaccine coverage rates but not for the usage of plasma-derived HBV vaccines. Currently, international public health authorities did not recommend a booster for immune competent people after primary immunization because of the robust anamnestic response to HBV in most people for many years after priming, even when antibodies were undetectable upon exposure.^20-23 Nevertheless, some studies in Taiwan have revealed the absence of an anamnestic anti-HBs response in approximately 25% adolescents after one booster dose.^24-26 Chronic HBV infections were reported after vaccine-induced protecting antibodies had disappeared.²⁷ This absence of immune memory may raise concerns about the need of booster dose for high-risk groups.

High HBsAg prevalence, low anti-HBs seroprotective rates, low concentration at baseline, and low HBV vaccination coverage rates in the study population indicated the necessity of immunization with HBV vaccines for college students. How to immunize this population should consider the following factors: primary HBV vaccination history, the condition of primary HBV vaccinations (such as HBV vaccination timing, the doses and the type of HBV vaccine), and the anti-HBs titer.²⁸ Taking into account nearly 81% of the participants having unknown vaccination history, it was less practical to figure out an immunization strategy on the basis of previous vaccination history. Therefore, we investigated the anti-HBs response in the participants negative for HBsAg, anti-HBs, and anti-HBc and with different HBV vaccination histories. About 30% of the participants were negative for HBsAg, anti-HBs, and anti-HBc, and were susceptible for HBV infection; they should be deemed the target population of HBV immunization. We found that among students with none, confirmed or unknown HBV vaccination history, anti-HBs GMCs, and seroprotective rates were not different after one, two, or three doses of HBV vaccine. One dose of HBV vaccine only induced 30% anti-HBs seroprotective rate. After two doses, 90% of participants produced protective anti-HBs but the anti-HBs GMC was less than 100 IU/L. Three doses induced more than 95% seroprotective rate and the GMC was 742.6 IU/L. The results were similar to the study conducted in Taiwan college students where at least two doses were needed for at-risk youths who received complete plasma-derived HBV vaccine in neonatal and infant periods but were seronegative for HBsAg, anti-HBc, and anti-HBs in adolescence to ensure higher than 90% of anti-HBs seroprotection rates.²⁹ However, anti-HBs of 10–99 IU/L was considered as low response. Some studies suggested that all individuals with anti-HBs concentration below 100 IU/L should receive a booster dose every 2 y.³⁰ Therefore, we recommended three doses for the college students negative for HBsAg, anti-HBs, and anti-HBc to obtain high anti-HBs seroprotective rate and concentration. Another study in Taiwan also indicated that including three doses of the booster vaccination including three doses may be essential for adolescents negative for anti-HBs, HBsAg, and anti-HBc to maintain anti-HBs concentration and reduce HBsAg prevalence.³¹ One study showed that high concentrations of anti-HBs could provide long-lasting protection against HBV infection.³² In this study, all participants were recruited from one medical college in the university, which resulted in enrolling a higher proportion of female students (82.4%). A predominantly female sample may increase anti-HBs response since it has been observed that being male was associated with an increased risk for nonresponse or a low response to HBV vaccination.³³

Anti-HBc acts as a serum marker for HBV infection. There should be an anamnestic response of the anti-HBs titers equal to or higher than 10 IU/L after one dose of the HBV vaccine in a population with isolated positive anti-HBc due to prior infection and immunity after one dose of the HBV vaccine.³⁴ In this study, 28 (66.7%) of 42 participants with isolated positive anti-HBc had ≥10 IU/L anti-HBs after one dose of HBV vaccine, which indicated that 66.7% participants with isolated positive anti-HBc had true anti-HBc positive and had already eliminated HBV in their body. Ten (23.8%) of 42 participants with isolated positive anti-HBc had no seroprotective anti-HBs after one dose of HBV vaccine but achieved ≥10 IU/L anti-HBs after full course of the HBV vaccine. Thus it was deduced that 23.8% (10/42) participants may have had a false anti-HBc positive and had never been infected with HBV. The observation was consistent with the previous reports of unspecificity problem concerning the Abbott Architect anti-HBc.^35,36 Ollier et al. reported an unspecificity problem concerning the Abbott ARCHITECT^® old anti-HBc test (Abbott ARCHITECT^® anti-HBc, Abbott Laboratories).³⁵ Seiskari et al. studied the specificity of ARCHITECT^® new anti-HBc II assay, and concluded that the samples with very low reactivity may be adequate to confirm, especially if the other HBV markers or clinical findings do not support the diagnosis.³⁶ Four of (9.5%) of 42 participants with isolated positive anti-HBc could not induce seroprotective anti-HBs after three doses of the HBV vaccine. Three of them were HBV DNA positive and regarded as occult HBV infection. The fourth participant with negative HBV DNA achieved a low level of anti-HBs concentration (84.5 IU/L) after additional three doses of HBV vaccine and his anti-HBc positivity also may be regarded as a false positive and he was susceptible to HBV infection. Through the immune response of these participants with isolated positive anti-HBc to HBV vaccine, we found that 26.2% (11/42) participants should be considered susceptible to a HBV infection. Currently, there is a lack of consensus on whether those having positive anti-HBc alone and waned anti-HBs concentrations should receive a HBV vaccination. Su et al. recommended a single booster dose of HBV vaccine for participants with isolated positive anti-HBc who had been fully vaccinated with a HBV vaccine as infants.³⁷ The present study found that three doses of HBV vaccine could induce a higher protection rate and anti-HBs concentration in students with isolated positive anti-HBc compared with one dose of the HBV vaccine.

One of the strengths of our study was that it was the first study to analyze seroprevalence of hepatitis B and immune response to HBV vaccination in Chinese college students based on the confirmed HB vaccination history. Another strength was conducting the study on the immune response for isolated positive anti-HBc participants after HBV vaccination. It was demonstrated that the isolated anti-HBc positive subjects may be divided into three groups according to their immune responses to the HBV vaccine: (1) occult HBV infection (no immune response to three doses of HBV vaccine), (2) true anti-HBc positive with resolved HBV infection (anamnestic response to HBV vaccination), and (3) false anti-HBc positive without HBV infection (primary response to HBV vaccination). A predominantly female sample and the lack of HBV infection status of participants’ mothers were thought as limitation in our study.

In summary, immunization with three doses of 20 µg hepatitis B vaccine for college students negative for HBsAg, anti-HBs, and anti-HBc was recommended to ensure high anti-HBs seroprotective rate and concentration. However, to immunize all college students or only those negative for HBsAg, anti-HBs, and anti-HBc warranted more studies. The cost-effectiveness analysis may facilitate the development of a reasonable hepatitis B immunization strategy for college students in China.

Methods

Participants

All freshmen in one college in Liuzhou city of Guangxi Zhuang Autonomous Region except for those with acute febrile diseases or born outside of this province were enrolled in September 2009, and tested for baseline HBsAg, anti-HBs, and anti-HBc for the seroprevalence study. Those participants negative for all the three markers or positive for anti-HBc only were enrolled for the HBV vaccination study and given intramuscularly three doses of HBV vaccine (Engerix-B, recombinant hepatitis B surface antigen, 20 µg/mL/vial, GlaxoSmithKline) at month 0, 1, and 6, then tested for anti-HBs at month 1, 6, 7, and 20. Students having any of the following conditions were excluded: acute illness, immune compromised conditions, renal insufficiency, pregnancy, and allergic history to HB vaccine or yeast. Non-responders were defined as having an anti-HBs below 10 IU/L one month after three doses of HBV vaccine,³⁸ and were administrated,³⁸ an additional three doses of HBV vaccine according to the schedule of month 0, 1, and 6. Records on HBV vaccination history were retrieved from the Guangxi CDC.

This study was conducted in accordance with Good Clinical Practice and the Helsinki Declaration. Written informed consent was obtained from all the participants and their parents. All relevant documents were approved by the ethical review committee of Guangxi CDC.

Serological assay

HBsAg, anti-HBs, and anti-HBc were determined by chemiluminescent microparticle immunoassay (CMIA) using the Architect i2000SR analyzer (Abbott Diagnostic). Serum HBsAg and anti-HBs were determined quantitatively, while anti-HBc was determined qualitatively. HBsAg was considered positive if the concentration ≥0.05 IU/L. Anti-HBs ≥10 IU/L were considered seroprotective, ≥10 IU/L and <100 IU/L were considered low response, and anti-HBs concentrations more than 100 IU/L were medium or high response.³⁸ The upper detection limits for HBsAg and anti-HBs were 250 and 1000 IU/L, respectively. When calculating the GMC, a concentration of 0.1 IU/L was used for those lower than 0.1 IU/L and of 1000 IU/L for those higher than 1000 IU/L. Anti-HBc was interpreted with the ratio of the sample signal to cutoff ratio (S/CO). S/CO ≥1.0 was deemed anti-HBc positive per manufacturer’s specification. For anti-HBc alone, double testing was implemented. HBsAg was detected using a HBsAg ad/ay serotype reference panel. Sensitivity results calculated by linear regression ranged from 0.15 to 0.29 ng/mL and were estimated to be 99.52% (418/420) with a 95% confidence interval of 98.29% to 99.94%. Anti-HBs had a sensitivity of 97.53% (594/609) with a 95% confidence range of 95.57% to 98.62%. Anti-HBc had a sensitivity of 100% (406/406) with a 95% confidence interval range of 99.10–100%.

Statistical analysis

The results on seroprevalence and immune response to HBV vaccination were presented with point estimates and two-sided 95% confidence intervals (CI). The outcomes for anti-HBs were expressed as seroprotective rates and GMC. GMC1 were calculated on all the participants with specific vaccination history and GMC2 were calculated on the participants other than HBsAg positive. Chi-square tests or Fisher exact tests were used for analyzing categorical data when relevant. Student t-tests or analysis of variance (ANOVAs) were used for analyzing continuous data when relevant. The statistical analysis was conducted by an independent statistician using SAS version 9.2. Hypothesis testing was two-sided with α value of 0.05. Bonferroni adjustments were used for multiple comparisons, by which the α’s value was adjusted by comparison time.

Abbreviation:
ANOVA	=	analysis of variance
anti-HBc	=	antibody against hepatitis B core antigen
Anti-HBs	=	antibody against hepatitis B surface antigen
CDC	=	Centre for Disease Control and Prevention
CI	=	confidence interval
CMIA	=	chemiluminescent microparticle immunoassay
GMC	=	geometric mean concentration
HB	=	hepatitis B
HBcAg	=	hepatitis B core antigen
HBsAg	=	hepatitis B surface antigen
HBV	=	hepatitis B virus
S/CO	=	sample signal-to-cutoff ratio

Disclosure of Potential Conflicts of Interest

All authors declare no conflict of interest.

Role of the Funding Source

This study was supported by Major Science and Technology Special Project of China twelfth Five-year Plan (2012ZX10002001).

Acknowledgments

We thank the students for participating in this study; the study nurses, and other staff members of Guangxi University of Science and Technology. The authors thank Dr Lisa Nicholson, Dr Yin Peng, and Dr Yuan-Zheng Qiu for critically proofreading this manuscript. And also the authors thank Dr Qian Zhang and Dr Ming-Huan Zheng for the support of the statistical analyses in the study.

10.4161/hv.26311