The role of transplacental hepatitis B core antibody in the mother-to-infant transmission of hepatitis B virus

Journal ojHeparology 1996; 24: 674679 Printed in Denmark All rights reserved Munksgaard

Copenhagen

Copyrrght 0 European Assocration for the Studv of the Liver 1996 Journal

of Hepatology

ISSN 0168.8278

The role of transplacental hepatitis B core antibody in the mother-toinfant transmission of hepatitis B virus Mei-Hwei Chang, Hong-Yuan Hsu, Li-Min Huang, Ping-Ing Lee, Ho-Hsiung Lin’ and Chin-Yun Lee Departments

of Pediatrics and Obstetrics and ‘Gynecology, College of Medicine, National Taiwan University, Taipei, Taiwan

To investigate the influence of transplacental hepatitis B core antibody (antiHBc) on perinatal hepatitis B virus (HBV) transmission, we studied the anti-HBc titers in 294 mother-neonate pairs. ResuZts: The anti-HBc titer was highest (10J~‘3M*80 to 104*W-W in mothers, 105*‘M*76to 105-52M*98 in infants) in the 200 hepatitis B e antigen (HBeAg) positive hepatitis B surface antigen (HBsAg) carrier mothers and their infants, second highest (1@51M*76 and 1@-68fl*76) in the 60 HBeAg-negative HBsAg carrier mothers and their infants, and lowest (103*“fo~76 and ld.““*83) in the 34 non-carrier mothers and their infants (~~0.05). One hundred and ninety-two infants of HBeAg-positive carrier mothers received hepatitis B humunoglobulin as well as hepatitis B vaccines, and were followed prospectively from birth. Ten infants became HBsAg carriers, and their mothers had significantly lower anti-HBc titers than those of

Aims/Methods:

I

hyperendemic for hepatitis B virus (HBV) infection like Taiwan or other part of Asia, hepatitis B carriers are infected mainly during infancy or early childhood (1,2). Perinatal transmission is an important route of infection, occurring mainly in infants of hepatitis B e antigen (HBeAg)-positive, hepatitis B surface antigen (HBsAg) carrier mothers (3). The mechanism of HBV transmission from mothers to infants remains unclear. The success of immunoprophylaxis is convincing. Most of the HBV transmission from HBeAg-positive mothers to their infants in Asia occurs during the perinatal, rather than N AREAS

Received 27 February; revised 6 September; accepted I4 September 1995

Correspondence: Dr. Mei-Hwei Chang, Department of Pediatrics, National Taiwan University Hospital, Taipei, Taiwan. Fax 886-23938871. Tel. 886-2-3970800, extension 5131. 674

the mothers of 182 infants who did not become carriers Cp=O.O03),while maternal serum hepatitis B virus DNA levels (29.9S3.6 versus 39.9zt58.1 pg/lO ml) did not differ in those two groups (~~-0.25). The same trend was observed in the infants’ anti-HBc titers in those two groups @=0.0006). Conclusions: The association of lower anti-HBc titers in HBeAg-positive carrier mother-infant pairs and the development of carrier status in the infants suggests a positive role of anti-HBc in the modulation of mother-to-infant transmission of HBV. A high maternal anti-HBc level in serum may be a negative predictor of immunoprophylaxis failure in high-risk infants.

Key words: Hepatitis B e antigen; Hepatitis B surface antigen carrier mother; Immunoprophylaxis failure.

the intrauterine period (4,5). But still about 5% or less of infants of HBeAg positive mothers had HBsAg in the serum at birth and became HBsAg carriers in spite of complete immunoprophylaxis. Intrauterine HBV infection, though infrequent, is possible (6) (Li L et al. Lmcet 1986; ii: 872, letter). The role of hepatitis B core antibody (anti-HBc) in modulating the immune response of neonates born to HBeAg-positive HBsAg carrier mothers remains undetermined. Alexander et al. (7) proposed that infants of high-risk mothers are infected in utero, but transplacental anti-HBc inhibits the expression of viral antigen in the fetus, which impedes the clearance of virus by the immune system of the fetus (8,9). Thomas et al. (10) suggested that transfer of maternal IgG anti-HBc across the placenta to the fetus results in either modulation of hepatitis B core antigen (HBcAg) display or of the cell-mediated immune response to this protein, and leads to failure of clear-

Tramplacental

hepatitis B core antibody

TABLE 1 294 mother-infant pairs were divided into three groups according to the maternal seral HBsAg and HBeAg on the day of delivery No. of pairs

Maternal status

Follow-up

HBIG Before@

HBsAg (+), HBeAg (+) HBsAg (+), HBeAg (-) HBsAg (-), HBeAg (-)

200 60 34

23 60 34

Total no. of pairs

294

117

After@

Yes

No

177 _ _

192 49 22

8 11 12

117

263

31

@ Blood of infants taken before, or after HBIG injection.

ante of HBV-infected cells by cytotoxic T cells sensitized to HBcAg. However, several authors questioned the role of anti-HBc in the perinatal transmission of HBV (London WT et al. Z_uncet 1986; i: 1037-8, letter; Panda SK et al. Lancet 1986; ii: 919-920, letter). In order to further investigate the role of anti-HBc in perinatal transmission of HBV, the anti-HBc titers were studied in 294 pairs of mothers and infants, who were then followed longitudinally for their outcome following immunoprophylaxis.

Subjects and Methods Two hundred and ninety-four mothers and their neonates were recruited into this study with the consent of the parents. This protocol was approved by the National Science Council of the Republic of China. The mothers were first screened for HBsAg and HBeAg during the second trimester. The second maternal blood was taken on the day of delivery. The first blood of the infants was taken within 24 h after birth. These 294 pairs of mothers and infants were divided into three groups according to the maternal serum HBsAg and HBeAg status on the day of delivery (Table 1). Group I: Maternal serum HBsAg(+),

HBeAg(+)

Two hundred mother-infant pairs were enrolled in this group. Hepatitis B immunoglobulin (HBIG) in a dose of 0.5 ml was given to all the infants in group 1 in the delivery room within 24 h after birth. The first blood samples of the infants were drawn within 24 h after birth, before the injection of HBIG in 23 infants and after HBIG in 177 infants. Although the time sequence of blood sampling and HBIG injection was different, all HBIG injections were given immediately after birth in the delivery room. The timing of HBIG injection thus should not influence the outcome of HBV immunoprophylaxis. No cord blood was used. One hundred and thirty-nine infants received HBV vaccines 1, 5, and 9 weeks and 12

months after birth, and 61 infants received HBV vaccines at 1 and 5 weeks, and 6 months of age. According to our previous study, there was no difference in the efficacy of three doses or four doses of HBV vaccination (11). HBV markers (HBsAg, anti-HBs, antiHBc, HBeAg, anti-HBe) in the 10 infants who became HBsAg carriers were followed up at 2, 3, 6, and 12 months of age in seven, at 2 and 12 months of age in two, and at 12 months of age in one. Nine infants were then followed every 6-12 months thereafter, and one was lost to follow-up. Anti-HBc level on day 1 was not measured in one infant who was HBsAg seropositive on the first day of life because of inadequate serum, but maternal anti-HBc level was measured. This infant became an HBsAg carrier during follow-up. Since her follow-up sera were drawn at 2, 3, 6, and 12 months of age and every 6 to 12 months after 12 months of age as in the nine other infants who were found to be carriers, we included this infant in the study. Group 2: Maternal serum HBsAg(+),

HBeAg(-)

There were 60 mother-infant pairs in this group. All the blood samples were drawn within 24 h after delivery. All the infants received HBV vaccines at 1, 5, and 9 weeks, and 12 months of age. No HBIG was given. Group 3: Maternal serum HBsAg(-),

HBeAg(-)

Forty HBsAg-negative mothers and their infants were recruited, but only the 34 mothers who were anti-HBc positive, suggesting past HBV infection, were included in this study. All the infants received HBV vaccines at 1,5, and 9 weeks, and 12 months of age. Follow-up serum HBsAg was checked at the 13th to 14th month of age in 192 infants in group 1, 49 infants in Group 2, and 22 infants in Group 3 to determine whether or not they had become HBsAg carriers. Anti-HBc was also studied by radioimmunoassay at the 13th to 14th month. 675

M.-H. Chang et al. TABLE 2 Seral geometric mean anti-HBc titers (Log 10 values of reciprocal titers) in neonates and mothers of different maternal HBsAg and HBeAg status Maternal HBsAg & HBeAg HBsAg(+) HBeAg(+) Before HBIG” After HBIGe HBsAg(+) HBeAg(-) HBsAg(-) HBeAg(-)

Case no.

23 177 60 34

Geometric mean anti-HBc Mothers

Neonates

5.13iO.8oabs 5.36~!~0.97~ 4.51M.76” 3.11fo.76”

5.13M.76deh 5.52j~O.98’~ 4.681t0.82df 3.24M.83ef

Seral HBV DNA of Mothers (pg/lO Fl serum)

39.98f58.32 -

Kruskal-Wallis test and Tukey’s test for multiple comparison, a,b,c*d.e*f: ~~0.05, ssh. . p>O. 1, *n= 176, because one infant had inadequate serum for anti-HBc determination; @before HBIG: blood of infants drawn before HBIG injection, after HBIG: blood of infants drawn after HBIG injection.

Hepatitis B markers of the serum (1) HBsAg, HBeAg, anti-HBe, anti-HBs and IgM anti-HBc and follow-up total anti-HBc were checked

by radioimmunoassay Chicago, IL, USA).

(Abbott Laboratories, North

(2) Anti-HBc titer: Passive hemagglutination testing was performed using a commercial kit (Corecell) manufactured by Green Cross Corporation (Osaka, Japan). The serum was first diluted 32 times followed by serial 1:2 dilution and put into separate wells. An equal volume of sensitized RBC was added to each well. After vibration, the mixture was kept at room temperature for 2-3 h, and the results were read. Each serum was tested in duplicate. Geometric mean titers of anti-HBc were calculated. For those infants who became HBsAg carriers, anti-HBc was rechecked by radioimmunoassay (Abbott Laboratories), and the sera drawn during long-term follow-up were also tested by radioimmunoassay. (3) Serum HBV DNA determination: Maternal serum HBV DNA was detected by the dot hybridization method as by Scotto et al. (12) with minor modifications. The purified HBV probe, separated from the plasmid, was prepared from a head-to-tail dimer of HBV DNA (subtype adw) cloned in the EcoRI site of plasmid vector pBR 322. The specific activity of the probe was 1-4~10~ cprn/g. Each specimen was tested twice.

Maternal course of gestation and delivery The records of gestation and delivery were reviewed

in detail by an experienced obstetrician (HHL) who did not know the outcome in the infants. Any history of amniocentesis, choriovillous sampling, or threatened abortion was specified. 676

Statistics

The Wilcoxon Mann-Whitney rank-sum test or Student t-test was used to compare differences between two groups. The Kruskal-Wallis test was used to compare differences among the three groups. If differences among the three groups were significant, Tukey’s test was used for multiple comparison. To delineate the influence of anti-HBc on the infants’ outcome, anti-HBc titers were arbitrarily divided into high (>105.42),intermediate (105.‘2-105.42),and low (c~O~.‘~)levels. The Mantel-Haenszel chi-square test was used to test the trend association between antiHBc levels and the development of chronic HBsAg carriers.

Results Comparison of serum anti-HBc titers in the mothers and neonates of different maternal HBsAg and HBeAg status The geometric mean anti-HBc titer in the infants of

mothers with positive HBsAg and HBeAg was the highest, that in the infants of mothers with positive HBsAg and negative HBeAg was second highest, and that in the infants of negative maternal HBsAg was the lowest (Table 2). The differences in geometric means among the three groups were statistically significant. The mean anti-HBc titers in the mothers of the above three groups were very close to the infant titers and showed a similar trend of differences. Comparing the anti-HBc levels in sera of Group I infants taken before and after HBIG injection showed no statistically significant difference (p>O. 1) (Table 2). The relationship between the outcome of the neonates and the anti-HBc titers of the mother-infant pairs in Group 1 [maternal HBsAg(+), HBeAg(+)]

One hundred ninety-two of the 200 infants of HBeAg positive HBsAg carrier mothers (group 1) were fol-

Tramplacental hepatitis B core antibody TABLE 3 The relationship between the outcome in the infants and (1) the geometric mean anti-HBc titers in the serum of the infants and their mothers, and (2) the seral HBV DNA levels of the mothers with positive HBsAg and HBeAg Outcome in the infant

Geometric mean anti-HBc Mothers Infants Maternal seral HBV DNA(pg/lO ~1)

HBsAg carriers

Non-carriers

4.49k0.94a (n=lO) 4.48k0.95b (n=9)‘ 29.9f23.6C (log=1.38f0.3)d

5.35f1.03a (n=182) 5.53j~O.96~ (n=182) 39.5k58.1C (log=1.15M.72)d

Wilcoxon Mann-Whitney rank-sum test, a: p=O.O03, b: p=O.O006,‘: ~90.25, d.. Student’s t-test, pXI.1. ‘: n=9 because one infant who was HBsAg positive at birth did not have adequate serum for anti-HBc measurement.

lowed for more than 14 months. Ten infants became HBsAg carriers, while 182 infants remained HBsAg negative and developed anti-HBs. Serum HBsAg was first detected to be positive at birth in five infants, at 2 months of age in two infants, and negative in three infants who were found to be HBsAg positive at 12 months of age. Infants who became HBsAg carriers had significantly lower maternal and neonatal antiHBc titers than those who did not become HBsAg carriers during follow up (Table 3, ~~~0.001). IgM antiHBc was tested in four of the five infants who had been HBsAg positive since birth, and all were negative. Three of the 10 infants lost anti-HBc transiently at 20 months, 24 months and 4 to 6.5 years of age, and regained anti-HBc later at 36, 36 months and 7 years of age, respectively. The former one was HBsAg seropositive at birth and the latter two were HBsAg seropositive later during infancy. One was first found to be HBsAg seropositive at birth, and the other two were found to be HBsAg seropositive later during infancy. Thus no definite relation regarding intrauterine infection and the absence of anti-HBc could be demonstrated. None of the 85 infants with maternal anti-HBc titers >105.42 became HBsAg carriers, while four (7%) of the 61 infants with maternal anti-HBc titer between 105~12-10s~42,and five (11%) of the 46 infants with maternal anti-HBc titers <105.12 became HBsAg carriers (pO.l). The anti-HBc titers in their neonates showed the same trend.

Among the five infants who were HBsAg seropositive on day 1 and whose blood was drawn after HBIG injection, anti-HBs titer was ~10 mIU/ml in two and cl0 mIU/ml in three infants on day 1. Anti-HBs titer declined to ~10 mIU/ml in all five infants at 3 months of age, and remained negative thereafter. Among the other five infants who became HBsAg positive later during infancy, two infants whose blood was drawn after HBIG injection at birth had positive anti-HBs on day 1. One became anti-HBs negative at 3 months old, and the other at 6 months old. The remaining three whose blood was drawn before HBIG injection were anti-HBs negative on day 1 and thereafter. None of the infants who were followed in Groups 2 and 3 became HBsAg carriers (Table 1). The relationship between the outcome in the infants and the seral HBV DNA levels of the mothers with positive seral HBsAg and HBeAg Maternal HBV DNA levels in sera were not different between those whose infants became HBsAg carriers and those whose infants did not become carriers (Table 3). One of the mothers whose infant became an HBsAg carrier had received amniocentesis in the 15th week of gestation. The gestational and delivery histories of the other nine carrier infants were uneventful.

Discussion It is important to note that in HBeAg(+) carrier mother-infant pairs, a significantly lower geometric mean titer of anti-HBc was demonstrated in mothers and their neonates who became HBsAg carriers than in those mothers and their neonates who were successfully protected by immunoprophylaxis. The maternal HBV DNA concentration of these two groups was not different. A maternal anti-HBc titer greater than 105.42is a good negative predictive index for immunoprophylaxis failure. Maternal anti-HBc very likely has some protective effect against the highly infectious HBeAg-positive maternal blood, so that the neonates of those mothers with lower antiHBc titers became HBsAg carriers even with immunoprophylaxis. Immunization with HBcAg in complete Freud’s adjuvant had been shown to protect chimpanzees from HBV infection (13). Lower anti-HBc titers in the ten HBeAg-positive carrier mothers whose infants became HBsAg carriers may be due to the decreased production of antiHBc by the maternal immune system. Because the HBV DNA level was not different, relatively greater tolerance to HBcAg is likely to be related to the maternal immune system itself, rather than to the 677

M.-H. Chang et al.

antigen overload. Recently, a study revealed a selective immune defect for HBcAg in a chronic HBsAg carrier with anti-HBc seronegativity (14). It is likely that a similar selective immune defect specific for HBcAg, resulting in lower level of anti-HBc production, may be present in those index mothers whose infants became carriers. Their infants may also hereditarily have the same immune defect, and thus were infected intrauterinely, or display a lower ability to handle the few HBV which escaped from the HBIG neutralization, and thus became carriers. In this study, three of the ten infants who failed to respond to immunoprophylaxis lost anti-HBc during follow-up, suggesting the failure of anti-HBc production during the subsequent follow-up period. We have previously reported ten HBsAg carrier children with absence or delayed appearance of anti-HBc (15). Lee et al. (16) also demonstrated 12 anti-HBc negative carrier infants of HBsAg carrier mothers. All those infants with immune tolerance to HBcAg failed to respond to immunoprophylaxis given at birth. Yamada et al. (17) studied anti-HBc IgG titers in infants of HBeAg-positive, HBsAg carrier mothers in the first month of life. They found that the mean antiHBc titer (27.7”.5) in seven infants who became HBsAg positive within the first month of life was significantly lower than that (210.M’.6)in ten infants successfully protected by immunoprophylaxis. Although the case number was small, the trend was similar to our study. Those two independent studies had similar results supporting the significantly lower anti-HBc in those whom immunoprophylaxis failed to protect. Five infants were HBsAg seropositive (neonates’ blood, not cord blood) at birth and were persistently HBsAg positive during follow-up although HBIG and HBV vaccines were given. This indicates that all the five infants were very likely infected in z&em. Our results did not demonstrate a difference of maternal HBV DNA in infants who became HBsAg carriers compared to those who were protected by HBV immunoprophylaxis, although Burk et al. showed that mother-to-infant transmission is linked to maternal virus load (18). Our results do not contradict theirs. The population they studied consisted of neonates of HBsAg carrier mothers not immunoprophylacted by HBIG or HBV vaccines. Our study population consisted of neonates of HBeAg-positive HBsAg carrier mothers immunized by HBIG and HBV vaccines. We thus investigated maternal HBV DNA levels in the infants of immunoprophylaxis failure, while Burk et al. investigated levels in infants with perinatally transmitted HBV without immunization. Ip et al. reported that in those who did not receive 678

HBIG, infants of mothers with serum HBV DNA levels of 5 pg/ml or above were at a significantly higher risk of becoming HBV carriers than those whose mothers had HBV DNA levels below 5 pg/ml (19). However, this difference was not significant in those given HBIG. Their finding may help to explain the different observations in our study and that of Burk et al. In this study, we have demonstrated that anti-HBc titers tend to be highest in mothers with active viral replication reflected by HBeAg in the serum. HBeAg-negative HBsAg carrier mothers had higher anti-HBc titers than non-carrier mothers who had recovered from a previous HBV infection. Before HBIG injection, the anti-HBc titers in neonates were almost the same as titers in their mothers, supporting the transplacental origin of the neonatal anti-HBc. Continuous exposure to HBcAg determinant of immune cells of HBeAg-positive carrier mothers results in a higher anti-HBc level than in HBeAgnegative carrier mothers. On the other hand, most of our non-carrier mothers had serological evidence of past HBV infection. Vento et al. (20) have noted that after recovery from acute hepatitis B, T cells remain unresponsive to HBcAg, in contrast to observations in chronic carriers in whom there is T-cell reactivity to HBcAg. Therefore, lack of help from HBcAg-specific T-cells would lead to the lower production of serum anti-HBc in non-carrier mothers. In conclusion, the association of lower anti-HBc in neonates of HBeAg-positive HBsAg carrier mothers and an outcome of HBsAg carrier status suggests a positive role of anti-HBc in the modulation of mother-to-infant transmission of HBV. Lack of protection by immunoprophylaxis is related to lower anti-HBc levels in mothers and/or infants but not related to maternal viral load, suggesting a possible immune defect in the multistep generation of antiHBc in those mother-infant pairs.

Acknowledgements This work was supported by a grant from the National Science Council of the Republic of China (NSC-78-0419-B002-52).

References 1. Sung JL, Chen DS, Lai MY, Yu JY, Wang TH, Wang CY, Lee CY, Chen SH, Ko TM. Epidemiological study on hepatitis B virus infection in Taiwan. Chin J Gastroenterol 1984; 1: l-9. 2. Hsu HY, Chang MH, Chen DS, Lee CY, Sung IL. Baseline seroepidemiology of hepatitis B virus infection in children in Taipei, 1984. A study just before mass hepatitis B vaccination program in Taiwan. J Med Virol 1986; 18: 301-7. 3. Stevens CE, Beasley RP, Tsui J, Lee WC. Vertical transmis-

Transplacental hepatitis B core antibody

4.

5.

6.

7.

8. 9.

10. 11.

12.

sion of hepatitis B in Taiwan. N Engl J Med 1975; 292: 7714. Beasley RP, Hwang LY, Lee GCY, Lan CC, Roan CH, Hwang FY, Chen CL. Prevention of perinatally transmitted hepatitis B immunoglobulin and hepatitis B vaccine. Lancet 1983; ii 1099-102. Lo KJ, Tsai YT, Lee SD, et al. Immunoprophylaxis of infection with hepatitis B virus in infants born to hepatitis B surface antigen-positive carrier mothers. J Infect Dis 1985; 152: 817-22. Steven CE, Taylor PE, Tong MJ, Toy PT, Vyas GN. Hepatitis B vaccine: an overview. In: Vyas GN, Dienstag JL, Hoofnagle JH, eds. Viral Hepatitis and Liver Disease. New York: Grune and Stratton, 1984: 275-92. Alexander GJM, Eddleston ALE Does maternal antibody to core antigen prevent recognition of transplacental transmission of hepatitis B virus infection? Lancet 1986; i: 296-7. Thomas HC, Scully LJ. Antiviral therapy and hepatitis B virus infection. Br Med Bull 1985; 41: 374-80. Pignatell M, Waters J, Lever AML, Iwarson S, Gerety R, Thomas HC. Cytotoxic T-cell responses to the nucleocapsid proteins of HBV in chronic hepatitis. J Hepatol 1987; 4: 1521. Thomas HC, Lever AML. Has immunology become important to the hepatologist? Progr Liver Dis 1986; 8: 179-89. Lee CY, Huang LM, Chang MH, Hsu CY, Safary A, Sung JL. The protective efficacy of recombinant hepatitis B vaccine in newborn infants of hepatitis B e antigen-positive-hepatitis B surface antigen carrier mothers. Pediatr Infect Dis J 1992; 10: 299-303. Scotto J, Hadchouel M, Hery C, Yvart J, Tiollais P, Brechot C. Detection of hepatitis B virus DNA in serum by a simple

13.

14.

15.

16.

17.

18.

19.

20.

spot hybridization technique: comparison with results for other viral markers. Hepatology 1983; 3: 279-84. Iwarson S, Tabor E, Thomas HC, Snoy P, Gerety RJ. Protection against hepatitis B virus infection by immunization with hepatitis B core antigen. Gastroenterology 1985; 88: 763-7. Lee JH, Paglieronic TG, Holland PV, Zeldis JB. Chronic hepatitis B virus infection in an HBc antibody-nonreactive blood donor: variant virus or defective immune response? Hepatology 1992; 16: 24-30. Ni YH, Hsu HY, Chang MH, Chen DS, Lee CY. Absence or delayed appearance of hepatitis B core antibody in chronic hepatitis B surface antigen carrier children. J Hepatol 1993; 17: 150-4. Lee SD, Lo KJ, Tsai YT, Wu JC, Wu TC. HBsAg carrier infants with serum anti-HBc negativity. Hepatology 1989; 9: 102-4. Yamada K, Natsume H, Somiya K, Hirata Y, Sugiyama K, Wada Y, Mizokami M. The role of anti-HBc IgG in transplacental transmission of HBV. Igagu No Ayumi 1987; 142: 4434l. Burk RD, Hwang LY, Ho GYF, Shafritz DA, Beasley RP. Outcome of perinatal hepatitis B virus exposure is dependent on maternal virus load. J Infect Dis 1994; 170: 1418-23. Ip HMH, Lelie PN, Wong VCW, Kuhns MC, Reesink HW. Prevention of hepatitis B virus carrier state in infants according to maternal serum levels of HBV DNA. Lancet 1989; i: 406-10. Vento S, Rondanelli KG, Ranieri S. O’Brien CJ, Williams R, Eddleston AL. Prospective study of cellular immunity to hepatitis B virus antigens from the early incubation phase of acute hepatitis B. Lancet 1987; ii: 119-22.

679