INTRODUCTION
It is estimated that 350 million people globally are chronically infected with hepatitis B virus (HBV) (1) (2–4) (5,6)
Previous data indicate that there is an unfavorable cytokine balance in patients with chronic HBV infection in comparison with those with acute HBV infection (7–9)
PATIENTS AND METHODS
This prospective study was performed from 2000 to 2005 at the referral centers for pediatric hepatology and for infectious medicine at Karolinska University Hospital, Huddinge, respectively.
Patients fulfilling all of the following inclusion criteria were asked to join the study:
Chronic HBV infection (ie, HBsAg detected in serum at least twice, with an interval of at least 6 months in between).
Age 2 to 19 years at inclusion.
No other chronic viral hepatitis detected.
When clinically relevant blood sampling for liver function tests, including alanine aminotransferase (ALT) levels, was performed, 10 mL of extra blood was drawn for study purposes. The genotype, HBV-DNA, and plasma cytokine levels were analyzed at least once in all of the patients. In 1 patient treated with IFN, the latter 2 parameters and the HBcAg-specific cytokine pattern were repeatedly analyzed before, during, and after completion of the treatment (patient 27, Table 1 ). In 3 other patients treated with IFN (patients 13, 25, and 26) and in another 4 patients (patients 12, 23, 24, and 34) serial HBV-DNA levels were analyzed. Finally, ALT levels and T cell proliferation ratios were analyzed on all of the blood sampling occasions in all of the patients.
TABLE 1: Baseline data for all patients with chronic hepatitis B infection
A total of 36 patients (22 boys, 14 girls) were included. The patients were divided into 3 groups according to HBV serology and ALT levels at inclusion:
Group 1: HBeAg positive, anti-HBe negative, and consistently normal ALT (<0.8 μkat/L), 9 patients (2 girls, 7 boys), mean age, 10.6 years (range, 2–19 years).
Group 2: HBeAg positive, anti-HBe negative, and elevated ALT (>0.8 μkat/L) on at least 1 occasion, 18 patients (7 girls, 11 boys), mean age, 14.6 years (range, 11–18 years). Fifteen of the patients had repeatedly elevated ALT levels, whereas 3 had elevated ALT on 1 single occasion.
Group 3: HBeAg negative, anti-HBe positive, regardless of ALT level, 9 patients (5 girls, 4 boys), mean age, 11.9 years (range, 6–18 years). In 7 of 9 patients, previous HBV serological results were available. The first anti-HBe positive test occurred 1 to 6 years (median 3 years) before these 7 patients entered the study. Two patients had ALT levels just above 1 μkat/L; the other 7 had constantly normal ALT levels.
The mean number of study samples per patient was 3.2 (range, 1–15). Four of 9 patients in group 1 and 4 of 18 patients in group 2 had only 1 sample analyzed.
During the study, 4 of the patients in group 2 were treated with IFN-α at a dosage of 5 million units per square meter body surface 3 times per week subcutaneously. The indication for treatment was prolonged ALT elevation and liver biopsy findings consistent with moderate to severe inflammation (ie, in accordance with the international consensus document for the treatment of these patients at the time) (5)
Markers of HBV Infection
Detection of antigens and antibodies specific for the HBV (HBsAg, HBeAg, anti-HBe) were carried out by use of AxSYM (Abbott Laboratories, Chicago, IL). Depending on the time period, HBV-DNA concentrations in serum were measured by the Amplicor HBV monitor test, which is a quantitative assay based on the polymerase chain reaction (PCR), or by the COBAS TaqMan HBV test (Roche Diagnostics Systems, Mannheim, Germany). The concentrations were adjusted to copies per milliliter to enable comparison of the 2 assays. All of the assays were carried out according to the manufacturers' instructions.
HBV Genotyping
HBV DNA was genotyped by sequencing of a part of the HBsAg gene as previously described (10)
Recombinant Proteins and Recall Antigens
Recombinant HBV core (HBcAg) antigen, expressed in Escherichia coli, was kindly supplied by D. Peterson, Virginia Commonwealth University, Richmond, VA. Phytohemagglutinin (PHA) was obtained from Sigma (St. Louis, MO) and was used as a positive control for the proliferation assay at a final concentration of 1 μg/mL.
Isolation of Peripheral Blood Mononuclear Cells
Peripheral blood mononuclear cells were isolated by gradient centrifugation by use of Ficoll-Paque (Pharmacia, Uppsala, Sweden) and washed several times in phosphate-buffered saline. The cells were resuspended in RPMI 1640 supplemented with 2 mmol/L L-glutamine, 1 mmol/L sodium pyruvate, 100 U/mL penicillin-streptomycin, and 10% AB-positive normal human sera (complete medium).
Proliferation Assay
Proliferation assays were performed on freshly isolated peripheral blood mononuclear cells as previously described (11) 3 H] thymidine (TDR) (Amersham, UK) for the last 16 hours. The level of [3 H] TDR incorporation was determined in a beta-counter. All of the values were obtained in duplicate or triplicate. A stimulation index (SI) was calculated as the ratio between counts per minute obtained in the presence of antigen to that obtained without antigen.
It was assumed that all of the samples with an SI below 2 were negative. To avoid false-positive results, we calculated the mean of all of the values below 2 and considered only values above this mean value plus 2 standard deviations to be positive (ie, reflecting T cell proliferation). The cutoff value was thereby set at 2.5. For comparison, the T cell proliferation was studied in samples from 5 healthy noninfected adult control individuals. Their mean SI value was 1.33; the mean value plus 2 standard deviations was 2.5.
Only samples with a PHA value above the cutoff were included in the study. For a sample to be regarded as positive, at least 3 wells had to show an SI above 2.5. All of the results of the T cell proliferation assays were presented as peak SI values.
ALT Levels
The ALT levels were analyzed according to hospital routines. The reference value was <0.8 μkat/L.
Cytokine Analysis
Cytokine production was measured in vivo in plasma and in vitro. Luminex technology, which uses carboxylated polystyrene beads with a distinct emitting fluorescence pattern and is coupled covalently with capture antibodies specific for individual cytokines, was used (Biosource, Nivelles, Belgium). In vitro cytokines were detected from the supernatants of 106 peripheral blood mononuclear cells. The cells were recalled alone or with 1 μg/mL of rHBcAg, and cytokines were detected after 24 and 48 hours. Data were presented as means of duplicate samples minus spontaneous cytokine production. The Th1/Th2 Five-Plex Antibody Bead kit was used, with beads coated with antibodies for interleukin (IL)-2, -4, -5, and -10 and IFN-γ. Assays were performed according to the manufacturers' instructions and samples were run in duplicate. Concentrations of individual samples were calculated and expressed as picograms of protein per milliliter in each sample (12)
Statistical Analysis
The Fisher exact test, χ2 test, t test, and correlation test were used for statistical evaluation.
RESULTS
Patient Characteristics
The characteristics of the studied patients at baseline are shown in Table 1 . The mean age at inclusion was 12.9 years (range, 2–19 years). The mean follow-up time was 84 weeks (range, 0–210 weeks). The patients originated from the Middle East (10 patients), southeast Asia (9), Africa (7), India (4), eastern Europe (5), and western Europe (1). Eleven patients had been adopted.
Data on the maternal HBV status were available for 13 patients (Table 1 ). In 6 of them (46%), the mother had a chronic HBV infection. This included 2 children who were infected despite adequate prophylactic measures at birth and during infancy. In 15 of the 23 remaining patients, the mother was not available for testing.
The mean age in group 1 was significantly lower than that in group 2 (P = 0.02, t test) and tended to be lower than that in group 3 (P = 0.07, t test). There was no statistical difference in sex distribution between the 3 groups (P = 0.35, χ2 test).
Distribution of HBV Genotypes
Genotypes A through E were detected. Genotype D was noted in 16 (44%) of the patients, including all of those originating from the Middle East. In patients originating from southeast Asia, both genotypes B and C were detected. A majority (13 of 18 patients) in group 2 had genotype A or D, and 6 of 7 patients with genotype A belonged to group 2.
HBV-DNA Levels
At inclusion, 8 of 9 patients in group 1 had HBV-DNA levels higher than 40 million copies per milliliter, and the remaining patient had 26 million copies per milliliter. In group 2, 13 of 18 patients had higher than 40 million copies per milliliter, whereas the remaining 5 patients had 15,000 to 29 million copies per milliliter. In group 3, 3 of 9 patients had undetectable levels, whereas the remaining 6 patients had 733 to 27,000 copies per milliliter. The viremia level in 1 patient in group 3 (patient 30) with elevated ALT was 1100 copies per milliliter, whereas in the other (patient 33) it was below the detection limit.
Plasma Cytokine Levels at Inclusion
The plasma cytokine levels were generally low, regardless of group. Markedly increased levels of all cytokines were seen in only 1 patient. The concomitant analysis for T cell proliferation showed an SI level clearly above 2.5.
Follow-up HBV Serology
At the end of the study period, all 9 patients in group 1 and all 9 patients in group 3 remained in the same group, respectively (Fig. 1 ). Four of 18 patients (22%) in group 2 cleared the HBeAg; this included 2 of the 4 patients treated with IFN (patients 26 and 27). Patient 27 also cleared the HBsAg.
FIG. 1: Intergroup changes in the number of patients during the course of the study.
T Cell Proliferation
Patients with at least 1 sample with detectable T cell proliferation (ie, SI >2.5) were seen in all 3 groups (Fig. 2 , Table 2 ). The percentage of such patients varied from 55% in group 1 to 89% in group 2 (P = 0.07, Fisher exact test). A significant difference was noted when the number of patients with detectable T cell proliferation in group 1 was compared with that in groups 2 and 3 combined (P = 0.013, Fisher exact test). In patients with only 1 sample analyzed, SI >2.5 was noted in 2 of 4 patients in group 1 and in 3 of 4 patients in group 2, respectively.
FIG. 2: SI and ALT values for all patients in groups 1, 2, and 3. The dotted vertical line defines the cutoff level for positive SI (2.5), indicating T cell proliferation. Two outlier ALT values from group 2 (11 and 15, respectively) were omitted.
TABLE 2: Number of patients in each group who had SI value >2.5 on at least 1 occasion or always had SI value <2.5
In group 2, the proliferative responses in the 5 patients with HBV-DNA levels below 40 million copies per milliliter were compared with those in the 13 patients with HBV-DNA >40 million copies per milliliter. Among those with lower levels of viremia, 4 of 5 patients had detectable T cell proliferation. However, the patient with the lowest HBV-DNA level had no detectable proliferation. For comparison, all 13 patients in group 2 with HBV-DNA >40 million copies per milliliter had detectable T cell proliferation.
In group 3, the single patient without detectable T cell proliferation had a viremia level of 1800 copies per milliliter. The number of patients with and without detectable T cell proliferation did not differ with respect to genotypes.
In Figure 2 , all of the SI and ALT values for each patient subgroup are described. A tendency for positive correlations was noted in group 1 (r = 0.10; P = 0.37) and group 3 (r = 0.24; P = 0.13). By contrast, in the subset of patients of group 2 who seroconverted, a tendency for negative correlations was seen. Thus, for spontaneous seroconverters the r value was −0.27 (P = 0.25), and for IFN-treated seroconverters it was −0.21 (P = 0.16). For the group 2 patients in whom no seroconversion was detected, no correlation was seen between SI and ALT values. This included both treated (r = −0.09; P = 0.39) and untreated (r = 0.03; P = 0.44) patients.
Longitudinal Follow-up in Individual Patients
IFN-treated Patients
In samples from patient 27, the SI, ALT, and HBV-DNA levels were high before treatment was initiated, whereas the plasma cytokine levels were generally low (Fig. 3 ). During the first part of the treatment, increased ALT levels and a concomitant but transient reduction in HBV-DNA levels were seen. At the same time, the plasma IFN-γ levels increased significantly, whereas the other cytokines remained low. During the second part of the treatment (ie, when the proper dosage was given), increased ALT levels and a concomitant and profound reduction of HBV-DNA levels were noticed. The SI values also tended to increase. A decrease in IFN-γ levels was paralleled by an elevation in all of the other cytokine levels. At the end of treatment, the ALT, SI, and HBV-DNA levels were low, and seroconversions for HBsAg and HBeAg were noticed. However, pretreatment levels of cytokines were not recorded until 21 weeks after treatment ceased.
FIG. 3: Biochemical and virologic parameters, T cell proliferation, and cytokine levels in a patient (no. 27) treated with IFN. * Indicates single value above maximum range of the method.
When the concomitant HBcAg-specific cytokine production was analyzed, the IL-10 level was generally high, with additional peaks early in the first phase of treatment and in the middle of the second phase of treatment. The corresponding IFN-γ levels were initially low, but a peak was seen early in the second phase of treatment.
In patient 25, SI values indicating T cell proliferation were noted during and after completion of the IFN treatment (Fig. 4 ). At treatment cessation and at follow-up 19 weeks later, a tendency to form anti-HBe antibodies was noticed. However, HBeAg was not cleared. Unfortunately, no samples at later follow-up were available. By contrast, patient 26, whose SI pattern was similar to that in patient 25, had cleared HBeAg at follow-up 28 weeks later. Patient 13 showed a transient response with SI values >2.5 during his 7-week course of IFN treatment. No response was seen on repeated follow-up samples after treatment cessation.
FIG. 4: Biochemical and virological parameters and T cell proliferation in 3 patients (a–c) treated with IFN, 2 patients (d, e) with spontaneous HBeAg seroconversion, 1 patient (f) without seroconversion, and 1 patient (g) who had already achieved seroconversion.
Nonspecific T cell reactivity, as measured by stimulation with PHA, occurred before, during, and after IFN treatment in all 4 patients. In 1 of them (patient 25), it was enhanced during treatment.
Untreated Patients
Two patients who seroconverted spontaneously are described in Figure 4 . Patient 23 was a more clear-cut example of HBeAg seroconversion, inasmuch as for patient 24 this had already started before the study inclusion. Patient 12, who also belonged to group 2, did not seroconvert (Fig. 4 ). Repeated SI values were below 2.5; however, on 1 occasion it was elevated above 2.5, and this coincided with a transient decrease in the HBV-DNA level. For comparison, the longitudinal pattern of 1 patient from group 3 was added in Figure 4 (patient 34). SI levels above 2.5 were noted on 3 of 4 occasions.
DISCUSSION
The present study showed that HBcAg-specific T cell proliferation occurred over time in a majority of the investigated pediatric patients with chronic HBV infection. Three different groups were studied: HBeAg-positive patients with normal ALT (group 1), HBeAg-positive patients with elevated ALT (group 2), and anti-HBe-positive patients (group 3). The percentage of patients with any significant SI level was higher in groups 2 and 3 than in group 1. The result for group 2 was not surprising when the elevated ALT levels and the results of previous adult studies were considered (8) (13,14)
No patient in group 3 had high levels of viremia, and only 2 of 9 had slightly elevated ALT levels, whereas the T cell proliferation pattern was almost universal. Thus, it is less likely that the existence of precore mutants can explain the results concerning T cell proliferation in this group.
Almost half of the patients in group 1 had any SI value above 2.5. Considering the constantly normal ALT levels and lack of HBeAg seroconversion over time in this group, this is somewhat unexpected. It should be noted that the total number of samples analyzed in this group was rather small. However, comparison with the data from group 2 suggests that repeated rather than single SI values above 2.5 may possibly be associated with elevated ALT levels.
Data from other studies suggest that the T cell response in chronically infected patients may be more vigorous if the viremia level is lower, and a threshold level of 107 copies per milliliter has been suggested (15,16)
In other studies using similar techniques to evaluate the virus-specific T cell response, the cutoff level for the SI has varied between 2 and 4 (8,15,17,18)
We included the detection of genotypes in the study because recent data on both adults and children indicate a difference in the HBeAg seroconversion rates and also in the long-term outcome when genotype B is compared with genotype C in patients from southeast Asia (19,20) (21–23)
Weak or no correlations between SI and ALT levels were seen. Most probably this reflects the fact that the T cell response, as illustrated by the SI level, occurs before the ALT elevation. However, the pattern of correlation seemed to be different when the 3 groups were compared. These differences may reflect the lack of a vigorous enough T cell response in group 1 (weak positive correlation) compared with group 2 (weak negative correlation).
In patient 27 the effects of the unintentional and biphasic treatment regimen were studied in detail. It was noted that the suboptimal dosage during the first treatment phase actually coincided with repeatedly high SI levels and HBcAg-specific IL-10 production, whereas plasma cytokine levels remained low and HBV-DNA levels were unchanged. Clearly, a different pattern was noted when the appropriate dosage was applied during the second phase of the treatment. This second pattern seemed to have been highly effective with regard to T cell proliferation; cytokine production, including that of HBcAg-specific IFN-γ; and in the end also with regard to treatment response on serology and HBV-DNA levels. Although the dosage of IFN was inadequate during the first 16 weeks, it prolonged the duration of immune activation. This may actually have influenced the treatment result positively.
The data on the IFN-treated patients presented in Figure 4 further suggest that the therapy may be important in maintaining the T cell immune response over time. The duration of treatment seems to be important. Thus, in all 3 patients (patients 13, 25, and 26) the pretreatment SI levels indicated proliferation. In patient 26 proliferation was withheld throughout and after the treatment period, and HBeAg seroconversion was achieved. In patient 13 the abortive treatment course was associated with a transiently increased SI level but no signs of effect on HBeAg or on the HBV-DNA level. Finally, in patient 25 a pattern similar to that in patient 26 was seen. The HBV-DNA levels did indeed fall, and on the last follow-up a borderline level of anti-HBe was detected.
It was noted that less intense proliferation patterns were seen in the 2 patients who seroconverted spontaneously than in those who were treated with IFN and seroconverted. This observation may indirectly suggest that the effect of the IFN treatment is to enhance a preexisting immune response against the virus; however, it may be suggested that this observation was secondary to a nonspecific T cell reactivity associated with the treatment. Such an enhanced reactivity during treatment was seen in only 1 of 4 patients.
When patient 23 was compared with those who were successfully treated with IFN, a similar reduction in HBV-DNA levels over time was seen. In fact, recent reports on the effects of IFN therapy in children with chronic HBV infection show that in the long run the HBeAg seroconversion rate was unaffected by the treatment (24) (25)
It should be noted that the results from the presently studied immune mechanisms give only a glimpse of the obviously complex events that are responsible for maintaining the chronicity of the HBV infection. Furthermore, the patients represented a select group because they were referred to a tertiary unit. This is reflected in the disproportionate number of patients in group 2 compared with those in groups 1 and 3. Thus, the present study gives limited information on the population-based natural history of chronic HBV infection in children. However, few similar studies are available that use modern techniques to detect virus-specific T cell proliferation on a longitudinal basis in a relatively large cohort of clinically, and virologically well-defined, pediatric patients.
We conclude that HBV-specific T cell responses are detected in a majority of children and adolescents with ongoing chronic infection. The application of the presently used techniques would be of interest in the study both of HBV patients who receive chemotherapy and of previously chronically infected patients who seem to have cleared the virus. Our data may suggest that a more vigorous immune response over time is associated with a better chance of HBeAg seroconversion.
Acknowledgments
The authors thank Lena Hallberg and Marit Bjon-Holm for handling the study patients and their blood samples with great care.
REFERENCES
1. Slowik MK, Jhaveri R. Hepatitis B and C viruses in infants and young children. Semin Pediatr Infect Dis 2005; 16:296–305.
2. Arakawa K, Tsuda F, Takahashi K,
et al . Maternofetal transmission of IgG-bound hepatitis B e antigen. Pediatr Res 1982; 16:247–250.
3. Milich DR, Jones JE, Hughes JL,
et al . Is a function of the secreted hepatitis B e antigen to induce immunologic tolerance in utero? Proc Natl Acad Sci U S A 1990; 87:6599–6603.
4. Wang JS, Zhu QR. Infection of the fetus with hepatitis B e antigen via the placenta. Lancet 2000; 355:989.
5. Jara P, Bortolotti F. Interferon-alpha treatment of chronic hepatitis B in childhood: a consensus advice based on experience in European children. J Pediatr Gastroenterol Nutr 1999; 29:163–170.
6. Sokal EM, Kelly DA, Mizerski J,
et al . Long-term lamivudine therapy for children with HBeAg-positive chronic hepatitis B. Hepatology 2006; 43:225–232.
7. Ferrari C, Penna A, Bertoletti A,
et al . Cellular immune response to hepatitis B virus-encoded antigens in acute and chronic hepatitis B virus infection. J Immunol 1990; 145:3442–3449.
8. Vingerhoets J, Michielsen P, Vanham G,
et al . HBV-specific lymphoproliferative and cytokine responses in patients with chronic hepatitis B. J Hepatol 1998; 28:8–16.
9. Szkaradkiewicz A, Jopek A, Wysocki J,
et al . HBcAg-specific cytokine production by CD4 T lymphocytes of children with acute and chronic hepatitis B. Virus Res 2003; 97:127–133.
10. Bjornsdottir TB, Stanzeit B, Sallberg M,
et al . Changing prevalence of hepatitis B virus genotypes in Iceland. J Med Virol 2005; 77:481–485.
11. Hultgren C, Weiland O, Milich DR,
et al . Cell-mediated immune responses and loss of hepatitis B e-antigen (HBeAg) during successful lamivudine and famciclovir combination therapy for chronic replicating hepatitis B virus infection. Clin Infect Dis 1999; 29:1575–1577.
12. Kellar KL, Kalwar RR, Dubois KA,
et al . Multiplexed fluorescent bead-based immunoassays for quantitation of human cytokines in serum and culture supernatants. Cytometry 2001; 45:27–36.
13. Yeo W, Chan PK, Zhong S,
et al . Frequency of hepatitis B virus reactivation in cancer patients undergoing cytotoxic chemotherapy: a prospective study of 626 patients with identification of risk factors. J Med Virol 2000; 62:299–307.
14. Kawatani T, Suou T, Tajima F,
et al . Incidence of hepatitis virus infection and severe liver dysfunction in patients receiving chemotherapy for hematologic malignancies. Eur J Haematol 2001; 67:45–50.
15. Boni C, Bertoletti A, Penna A,
et al . Lamivudine treatment can restore T cell responsiveness in chronic hepatitis B. J Clin Invest 1998; 102:968–975.
16. Webster GJ, Reignat S, Brown D,
et al . Longitudinal analysis of CD8+ T cells specific for structural and nonstructural hepatitis B virus proteins in patients with chronic hepatitis B: implications for immunotherapy. J Virol 2004; 78:5707–5719.
17. Hsu HY, Chang MH, Ni YH,
et al . Cytokine release of peripheral blood mononuclear cells in children with chronic hepatitis B virus infection. J Pediatr Gastroenterol Nutr 1999; 29:540–545.
18. Avanzini MA, Belloni C, De Silvestri A,
et al . Antigen-specific T cell response in infants after recombinant hepatitis B virus vaccination at birth: evaluation of T helper lymphocyte diversity. Clin Immunol 2003; 107:122–128.
19. Kao JH, Chen PJ, Lai MY,
et al . Hepatitis B virus genotypes and spontaneous hepatitis B e antigen seroconversion in Taiwanese hepatitis B carriers. J Med Virol 2004; 72:363–369.
20. Ni YH, Chang MH, Wang KJ,
et al . Clinical relevance of hepatitis B virus genotype in children with chronic infection and hepatocellular carcinoma. Gastroenterology 2004; 127:1733–1738.
21. Chang MH. Natural history of hepatitis B virus infection in children. J Gastroenterol Hepatol 2000; 15(Suppl):E16–E19.
22. Bortolotti F, Jara P, Crivellaro C,
et al . Outcome of chronic hepatitis B in Caucasian children during a 20-year observation period. J Hepatol 1998; 29:184–190.
23. Ruiz-Moreno M. Chronic hepatitis B in children: natural history and treatment. J Hepatol 1993; 17(Suppl 3):S64–S66.
24. Vo Thi Diem H, Bourgois A, Bontems P,
et al . Chronic hepatitis B infection: long-term comparison of children receiving interferon alpha and untreated controls. J Pediatr Gastroenterol Nutr 2005; 40:141–145.
25. Bortolotti F, Jara P, Barbera C,
et al . Long-term effect of alpha interferon in children with chronic hepatitis B. Gut 2000; 46:715–718.
