941. Mutations in Immunodominant Epitopes of Hepatitis B Virus (HBV) Core Region: Enhancement of Viral Evolution by the Effect of Interferon Therapy

F. Rodriguez-Frias; R. Jardi; M. Buti; M. Schaper; D. Tabernero; R. Esteban

 

Background and Aims

In HBV infection, the host immune system attacks core peptides as the main target. Amino acid (aa) substitutions in immunodominant epitopes of the HBcAg sequence may allow HBV to escape immune surveillance, thereby maintaining infection. The aim of this study was to analyze aa substitution rates in HBcAg immunodominant epitopes in a group of chronic hepatitis B (CHB) patients, throughout IFNa therapy and during a period without this therapy.

 

Patients and Methods

A cohort of 44 CHB patients (19 HBeAg+) (BASELINE group) was studied retrospectively, all cases with a baseline serum sample. Sixteen (1 HBeAg+) of these patients who did not respond to 24 weeks of IFNa therapy (IFN group) were longitudinally studied. Eleven of the 16 patients (1 HBeAg+) and 10 other CHB cases (5 HBeAg+) (untreated group) were longitudinally studied in the absence of antiviral treatment. In the IFN and untreated groups, serum samples were analyzed during therapy and every six months over the non-treatment period. The complete sequence of HBV core region (precore and HBcAg) was determined by direct sequencing. The viral genotype was established by comparison of obtained sequences with consensus sequences deduced from the alignement of 185 complete sequences of the eight HBV genotypes obtained from GenBank. Yearly rates of aa changes in the immunodominant HBcAg epitopes (B cell HBc/HBe1-aa 74 to 84, and Th cell-aa 50 and 69), were calculated in the IFN and untreated groups. In the BASELINE group this changes were recorded in relation to the consensus sequences.

 

Results

Rates of aa changes in the immunodominant HBcAg epitopes are shown in table.

 

Conclusion

The higher rates of aa substitutions suggest enhancement of the evolutive pressure due to IFN therapy, which may allow HBV to escape from immuno-surveillance, resulting in persistent infection. This does not seem to be related to viral genotype or precore variants. In baseline samples the highest variability in Th cell epitope (aa 50- 69) could be related with the absence of the immunotolerogen effect of HBeAg.

This study was supported by a grant from the Spanish Ministery of Health (FISS PI061512).

 

GROUP

Th cell epitope (aa 50-69)

B cell (HBc/HBe1 epitope(aa 74-84)

BASELINE (N=44)

Genotype A (N=22)

1.45

1.23

Genotype D (N=22)

1.36

1.27

Precore variant:

WT (N=20)

1.05 (e)

1.15

Precore variant:

MT (N=24)

1.71 (f)

1.33

IFN (N=16)

4.5 (a)

3.38 (c)

Non-treated (N=21)

0.66 (b)

0.86 (d)

Statistical results: a vs. b P<0.001, c vs. d P<0.001, e vs. f P=0.087

 


953. Demonstration of an Association Between Detection of IgG Antibody Reactivity Towards the C-terminal Region of the preS1 Protein of Hepatitis B Virus and the Capacity to Respond to Interferon Therapy in Chronic Hepatitis B

S. P. Sylvan; U. B. Hellstrom; K. Krogsgaard; M. Lindh

 

Background

The aim of this study was to determine the predictive value of the pre-treatment presence of circulatory antibodies towards a synthetic peptide mimicking the amino acids 94-117 of the preS1 protein of hepatitis B virus (HBV) and the capacity to respond to interferon–alpha (IFN-alpha) treatment.

 

Methods

The anti-preS1 (94-117) antibodies were measured by a peptide-based enzyme-linked immunosorbent assay (ELISA) and the response to INF-alpha therapy was judged by the effect on the viral kinetics as measured by a quantitative polymerase chain reaction (PCR) based assay during the treatment and follow-up.

 

Results

We found a significant (p<0.001) correlation between the pre-treatment presence of anti-preS1 (94-117) antibodies and a decrease in viral levels on follow-up after the end of IFN-alpha therapy. The combined response of HBV DNA suppression (p<0.001), HBeAg loss (p<0.0001) anti-HBe seroconversion (p<0.005) and ALT normalisation (p<0.01), was also highly associated with the pre-treatment presence of anti-preS1(94-117) antibodies.

 

Conclusions

The positive predictive value (PPV) of anti-preS1 (94-117) in determining a virological response was 83% and the negative predictive value (NPV) was 100%, indicating that in the absence of pre-treatment anti-preS1 reactivity virtually no patient has the capacity to respond to IFN-alpha therapy.

Our findings may help to improve the efficacy of INF-alpha therapy for chronic hepatitis B (CHB) by guiding the selection of patients to treatment and optimising the clinical management of the individual patient.

 


109. In vitro and in vivo Inhibition of HBV Replication by Peptide Aptamers via Disruption of the HBX-Proteasome Interaction

U. Zaid; Z. Zhang; T. Liang

 

The X protein (HBX) of the hepatitis B virus (HBV) is important for productive infection in vivo. Although this viral protein is not essential for viral replication in vitro, HBV negative for HBX replicates less efficiently in transfected cells. Our previous studies have suggested that the interaction between HBX and the proteasome complex may underlie the pleiotropic functions of HBX (Hu et al. JVI 2006; 8-: 1405-1413, Zhang et al. JCI 2001; 108: 1523-1531, Zhang et al. JBC 2000; 257: 15157-15165, Zhang et al. JVI 2004; 78: 4566-4572). We have further demonstrated that inhibition of cellular proteasome activities enhances hepadnavirus replication in an HBX-dependent manner.

 

In this study, we aim to develop potential anti-HBV agents by targeting the functions of HBX using a novel screening process.

 

Methods

A modified yeast two-hybrid disruptor system was developed to screen a randomly generated library of peptide aptamers displayed on the bacterial protein Thioredoxin A (TrxA). By screening for a disruption of the interaction between HBX and the proteasome subunit PSMA7, 367 yeast transformants were isolated from 1.5 x 10^7 independent yeast colonies of the peptide aptamer library. On secondary screen against nonspecific interacting pairs, 23 colonies were confirmed to show specific disruption of the HBX-PSMA7 interaction. The peptide aptamers from these colonies were isolated, sequenced, and cloned into a plasmid expression construct for transfection into HepG2 cells for functional studies. Transcriptional activation assays by HBX demonstrated that most of these peptide aptamers interfered with HBX transactivation by decreasing the reporter activities to 20-80% of the control level. When co-transfected with an HBV replication competent construct, most of the peptide aptamers which inhibited HBX transactivation also suppressed HBV viral replication by 50-80%. Very preliminary in vivo murine studies via intravenous injection of one of the peptides suggest a beneficial effect on inhibition of HBV DNA in HBV transgenic mice. Further studies are required to confirm this observation.

 

Conclusion

Our results demonstrate that selection of random peptide aptamers based on disruption of the HBX-proteasome interaction using a modified yeast two-hybrid system may identify peptide aptamers as potential therapeutic drugs for HBV infection. Also, this system may provide a molecular and structural basis for novel antiviral drug development.

 


962. The Indication and Limitation of Interferon Therapy as the First Line Therapy of Chronic Hepatitis B from the Histological Analysis of 800 Chronic Hepatitis B Patients

M. Shindo; K. Hamada; T. Morikawa; Y. Harano; T. Nakajima; T. Okuno

 

Aim

Current therapy of chronic hepatitis B does not eradicate HBV and has limited long-term efficacy. Thus, careful consideration of the patient’s age, severity of liver disease, likelihood of response, and potential adverse events is needed before treatment is initiated. The role of interferon in the first line therapy is not clearly defined. The degree of liver histology might be related to interferon’s efficacy and limitation. To prove this, we investigated the relationship between the staging score and the interferon response and durability in patients with chronic hepatitis B.

 

Materials and Methods

A total of 800 Japanese genotype C treatment naive chronic hepatitis B were studied. All of them had liver biopsy before the treatment. Seven hundred patients were HBeAg positive and the remaining 100 were HBeAg negative. All patients had elevated ALT levels (>2x UNL) and increased HBVDNA (>106 ) levels. Among these 800 patients, 269 patients with HBeAg positive chronic hepatitis B were treated with interferon alfa therapy ( Sumiferon, 6 MU/d, TIW, 4 mo). The patients were stratified by age into 5 year groups beginning with 15-20 y/o until 70-75 y/o. All liver tissue specimens were obtained by needle biopsy. The grade and stage were based on international standard criteria and METAVIR. The relapse after seroconversion was defined as that HBeAg became positive again or that HBeAg remained negative but HBVDNA and ALT elevated for one year after the therapy.

 

Result

1) The staging scores for each group were generally increased with age from 1.51 (15-20) to 2.75 (61-65), and the degree of fibrosis increased significantly especially after 30 y/o (mean, 2.2) every year at the rate of 0.04-0.06 until 45 y/o (mean, 2.54). 2) Among 269 HBeAg positive patients treated with interferon, 89 (30%) seroconverted to anti-HBe, and among them 74 (83%) had staging score either 1 or 2, and durability was 80-90%. While those with stage 3 or 4 had 50% durability. In nonresponders, the staging score was significantly higher than seroconverters (p=0.0001).

 

Conclusions

1)     The current study showed that progression of fibrosis appeared to increase significantly especially after 30 year old until 45 year old at the rate of 0.04 to 0.06 every year.

2)      Interferon appeared to be effective in patients who had less than staging score 2 with 80-90% durability, while in patients with staging score 3 or 4, interferon did not seem to be similarly effective and the durability appeared to be poor.

Thus the degree of fibrosis may be one of the important factors to indicate and limit interferon efficacy for the treatment of chronic hepatitis B.

 


965. Sustained HBeAg and HBsAg Loss after Long-term Follow-up of HBeAg-Positive Patients Treated with Peginterferon alpha-2b

E. H. Buster; H. J. Flink; Y. Cakaloglu; C. Simon; J. Trojan; F. Tabak; T. M. So; V. S. Feinman; T. Mach; U. Akarca; W. C. Tielemans; H. J. van Vuuren; B. E. Hansen; H. L. Janssen

 

We evaluated the long-term sustainability of response and clinical outcome in patients treated with PEG-IFN α-2b alone or in combination with lamivudine for 52 weeks.

 

Methods

All 266 HBeAg positive chronic hepatitis B virus (HBV) infected patients from 41 centers enrolled in the randomized HBV99-01 study were offered participation in a long-term follow-up (LTFU) study. Initial response was defined as HBeAg negativity at the end of the initial study (26 weeks post-treatment; week 78). For the LTFU study, patients had one additional visit after the initial study (mean interval 3.0 ± 0.8 years). For analysis, retreatment was considered as non-response.

 

Results

172 patients (65%) from 28 centers (68%) were enrolled in the LTFU study. 91 patients (53%) received PEG-IFN α-2b alone and 81 (47%) its combination with lamivudine. Response rates at different time points are shown in the table. At LTFU, HBeAg loss was durable in 52 of 64 initial responders (81%), of whom 67% had HBV DNA <10,000 copies/ml and 65% had normal ALT. 14 initial responders (21%) and 67 of 108 (62%) non-responders were retreated (p<0.001). HBsAg was negative in 19 initial responders (30%). Sustained HBeAg negativity was observed in 96%, 86%, 67% and 77% of initial responders with genotype A, B, C and D infection, respectively (A vs. C, p=0.04; A vs. D, p=0.07). Serum HBV DNA <10,000 copies/ml was observed in 81%, 29%, 44% and 35% of initial responders with genotype A-D, respectively (A vs. B or D, p<0.02). HBsAg negativity was observed in 58% of genotype A infected initial responders compared to 14%, 0% and 6% of those with genotype B, C or D (A vs. C or D, p<0.004). At LTFU, 3 non-responders had died and 1 developed hepatocellular carcinoma; decompensated liver disease was observed in 1 responder and none underwent liver transplantation.

 

Conclusions

HBeAg response to PEG-IFN α-2b ± lamivudine is durable in the majority of patients and is associated with an increase in HBsAg loss. This study further emphasizes the importance of HBV genotype in PEG-IFN therapy, with high rates of sustained virological response in genotype A-infected patients. PEG-IFN should therefore particularly be considered as first line therapy in genotype A infected HBeAg positive patients.

 

 

Week 32

Week 52

Week 78

LTFU

 

PEG

PEG

+

LAM

PEG

PEG

+

LAM

PEG

PEG

+

LAM

PEG

PEG

+

LAM

HBeAg negative

21%

35%*

34%

47%*

39%

36%

37%

36%

HBV DNA <10,000

14%

69%*

19%

70%*

17%

20%

25%

31%

HBV DNA <400

9%

32%*

9%

40%*

6%

11%

13%

26%*

Normal ALT

25%

40%*

30%

52%*

34%

37%

28%

33%

HBsAg negative

ND

ND

7%

9%

6%

9%

8%

15%

*p<0.05 PEG-IFN α-2b + lamivudine (PEG LAM) vs. PEG-IFN α-2b (PEG); ND = not done

 


974. Not HBeAg-status But HBV Genotype Predicts Response to Interferon in Chronic hepatitis B

A. Erhardt; A. Ludwig; M. Brunetto; M. Popescu; F. van Bömmel; M. Siemer; O. Adams; D. Haussinger

 

Background and Aims

In contrast to HBeAg-positive hepatitis B, HBeAg-negative hepatitis B is considered to be associated with a poor sustained virological response towards interferon therapy. The present study investigated the influence of HBV genotype on the presumed HBeAg-dependence of IFN response.

 

Patients and Methods

399 patients with chronic replicative hepatitis B infected by HBV genotype A (n=141) or HBV genotype D (n=258) were treated with standard interferon-alfa or pegylated interferon-alfa for 4-12 months. HBV genotype was determined by direct sequencing of the HBV X or S gene. 44.4% of the patients had HBeAg-negative hepatitis B and 55.6% HBeAg-positive hepatitis B.

 

Results

Sustained virologic response to standard interferon-alfa therapy was lower in HBeAg negative compared to HBeAg-positive patients (22.1% vs 35.6%; p<0.003). Prevalence of HBV genotype A was higher in HBeAg–positive hepatitis compared to HBeAg-negative hepatitis (59.3% vs.16.2%; p<0.0001) and opposite to the prevalence of HBV Genotype D (40.7% in HBeAg-positive hepatitis vs. 83.8% in HBeAg-negative hepatitis). SVR was lower for HBV genotype D patients (19.8%) compared to HBV genotype A patients (43.3%; p<0.001). However, there was no difference in SVR between HBeAg-positive and HBeAg-negative patients of HBV genotype A (44.8% vs 38.9%;p=n.s.) neither for HBeAg-positive and HBeAg-negative patients of HBV genotype D (18.5% vs. 21.1%: p=n.s.). Multivariate logistic regression performed for a representative cohort of patients (n=208) identified HBV genotype A vs. D (p<0.002; 95%CI:1.48-5,43) and HBV DNA (2 x ULN) nor HBeAg-status as independent predictive parameters of IFN response.

 

Conclusions

The present data demonstrate that SVR to IFN in hepatitis B was not affected by HBeAg status but by HBV genotype. Therefore genotype dependence of IFN response and geographic variability of HBV genotype prevalence might explain the divergent SR to IFN between HBeAg-positive and HBeAg-negative patients. This should be taken into account for delineation of antiviral therapy in chronic hepatitis B.

 


978. Twenty-four Weeks of Therapy with Peginterferon alfa-2a Is Similar to 48 weeks of Therapy in Patients with HBeAg-positive Chronic Hepatitis B and Good Predictors of Response

A. Gadano; L. Rezzonico; O. Galdame; B. Frider; J. Solari; A. Villamil; P. Casciato; M. Reig; J. Bandi; A. Alessio

 

Background

Forty-eight weeks of therapy with Peginterferon alfa-2a has demonstrated to be effective in about one third of patients with HBeAg-positive chronic hepatitis B. Although the recommended treatment duration for these patients is 48 weeks, there is no enough data supporting 48 weeks of therapy over 24 weeks of therapy. Treatment might be shortened particularly in patients with good predictors of response.

 

Aim

To compare the efficacy of 48 weeks vs 24 weeks of therapy with Peginterferon alfa-2a, in patients with chronic hepatitis B who had good predictors of response.

 

Patients and Methods

Nineteen patients with high baseline ALT levels (> 3 ULN) and low viral load (HBV DNA < 109 cp/ml) were treated with Peginterferon alfa-2a 180 mcg/week, during 48 weeks. Virological, biochemical and serological responses were compared with those obtained in 16 patients with similar baseline characteristics treated with Peginterferon alfa-2a for 24 weeks. All patients had a follow-up period of 24 weeks after end of therapy.

 

Results

At end of follow-up, HBeAg seroconversion was observed in 7/19 (36.8 %) of patients treated for 48 weeks and in 6/16 (37.5 %) of patients treated for 24 weeks (p=ns). Patients treated for 48 weeks evidenced a significantly higher decrease in HBV DNA at the end of therapy than patients treated for 24 weeks (-4.8 logs vs -3.6 logs respectively, p<0.05). However, the % of patients with HBV DNA < 100.000 cp/ml was similar in both groups at the end of follow up (42.1 % vs 43.7 %, ns). No significant differences between both groups were observed regarding ALT normalization, HBsAg loss or seroconversion. Serious adverse events occurred in 1 patient from each group.

 

Conclusion

The results from this study indicate that 24 weeks of therapy with Peginterferon alfa-2a is similar to 48 weeks therapy in patients with HBeAg positive chronic hepatitis B who have good predictors of response.

 


979. Long-term Follow-up of HBsAg Clearance in Patients with HBeAg-negative CHB Treated with Peginterferon alfa-2a: Increase in HBsAg Clearance Rate from 3% Six Months Post-treatment to 8% after Three Years

P. Marcellin; M. Brunetto; F. Bonino; G. K. Lau; P. Farci; C. Yurdaydin; T. Piratvisuth; K. Luo; Y. Wang; S. J. Hadziyannis; E. Wolf; M. Popescu

 

Clearance of HBsAg in patients with CHB is associated with long-term clinical outcome and improved survival. It can be induced by treatment with interferon-based therapy but is extremely rare following treatment with direct antivirals.

 

Methods

The HBsAg clearance rates 24 weeks post-treatment in patients with HBeAg-negative CHB who were treated for 48 weeks treatment with peginterferon alfa-2a, peginterferon alfa-2a plus lamiviudine or lamivudine alone in a multinational randomized trial were 3%, 4% and 0%, respectively. All patients included in this trial were offered entry into an observational study of long-term response 3 years post-treatment. The proportion of peginterferon alfa-2a-treated patients losing HBsAg increased from 3% 24 weeks post-treatment to 8% at 3 years; 9 patients treated with peginterferon alfa-2a monotherapy cleared HBsAg and 9 in the combination therapy group. Loss of HBsAg was observed across all the major HBV genotypes. Eight of the 18 patients developed anti-HBs antibodies.

 

Results

We analyzed the quantitative on-treatment HBsAg levels using the Architect assay (Abbott Diagnostics) for 198 patients treated with peginterferon alfa-2a +/- lamivudine who were included in the long-term follow-up study for whom 3-year post-treatment follow-up data were available. An HBsAg level <10 IU at week 48 was signficantly associated with HBsAg loss 3 years post-treatment (relative risk [RR] 22.8; 95% CI 8–64.9; p<0.0001). Of the 23 patients with HBsAg <10 IU/mL at week 48, 12 (52%) had cleared HBsAg by year 3 vs 4/171 patients (2.3%) with HBsAg >10 IU/mL at the end of treatment. An on-treatment reduction of >2 log IU/mL from pre-treatment level to week 48 was significantly associated with HBsAg clearance at year 3 (RR 14.6; 95% CI 5.5–38.5; p<0.0001). Of the 26 patients with an on-treatment HBsAg decline of >2 log IU/mL, 11 (42.3%) had HBsAg at year 3 compared with only 5 of 172 patients (2.9%) with an on-treatment HBsAg decline <2 log IU/mL. Despite the fact that at the end of the 48-week treatment period patients who cleared HBsAg had HBV DNA suppressed to levels <400 cp/mL, an HBV DNA level <400 cp/mL was not predictive of HBsAg clearance 3 years post-treatment.

 

Conclusion

In conclusion, the ability to induce HBsAg clearance, an outcome that is associated with long-term improvement in survival, supports the use of peginterferon alfa-2a as a first-line treatment for patients with HBeAg-negative CHB. Potent HBV DNA suppression appears to be required, but is not sufficient, for subsequent clearance of HBsAg. The ability of interferon-based therapy to induce HBsAg clearance may be associated with its dual immunomodulatory and antiviral mode of action.

 


990. Reduction in Serum HBsAg Level in Patients with Chronic Hepatitis B Infected with Genotype D Induced by (pegylated) Interferon alfa-2a Alone or in Combination with Nucleos(t)ide Analogs: A Long-term Single Center Cohort Study

M. Brunetto; F. Moriconi; D. Cavallone; F. Oliveri; A. M. Maina; P. Ciccorossi; P. Columbatto; B. Coco; G. Moscato; F. Bonino 

 

In patients with CHB undergoing antiviral therapy, clearance of serum HBsAg appears to be associated with better long-term clinical outcome. In clinical practice quantification of serum HBsAg might provide a valuable tool to identify the best therapeutic strategy for individual patients. We quantified serum HBsAg (Architect, Abbott Diagnostics) in a total of 80 consecutive HBV genotype D patients treated at our chronic viral hepatitis reference center. The patients received interferon [IFN] alfa-2a (recombinant or pegylated) monotherapy (11 patients), PEG-IFNalfa-2a + lamivudine [LMV] (24 patients), LMV or adefovir [ADV] monotherapy (6 patients) or LMV + ADV (39 patients). Of the 80 patients, 45 were resistant to LMV; we treated 39 of these with LMV + ADV and 6 with with LMV + PEG-IFNalfa-2a). Baseline characteristics such as gender, HBeAg, viral load and ALT level were not significantly different between the 2 groups of patients receiving (PEG)-IFN +/- nucleos(t)ide analogs (NA) or NAs alone. Patients treated with NAs had lower HBsAg baseline levels (mean 3.39 vs 3.77 log IU/mL, p=0.025) and were older (mean 57 vs 45 years, p<0.001). Patients were followed up for a mean duration of 40 months (24–84 months). The overall serum HBsAg delta variations (log decline from baseline to last observation) according to treatment group are reported in the Table. Almost all patients treated with NAs alone had no or minimal reduction in HBsAg level and none of these cleared HBsAg during follow-up. In contrast, the majority of patients who received interferon-based therapy showed a reduction in HBsAg level of at least 0.5 log and 4 patients achieved HBsAg clearance.

 

Conclusion

In conclusion, IFNalfa-2a (conventional or pegylated) appears to be most effective in suppressing serum HBsAg independently of baseline host and virologic features, also in patients who are resistant to LMV therapy. Quantification of serum HBsAg could provide a useful tool to predict the long-term outcome of interferon-based therapy and its ability to induce HBsAg clearance in CHB patients infected with genotype D.

 

Reduction in HBsAg level (log)

<0.5

n (%)

>=0.5

n (%)

>=1

n (%)

>=2

n (%)

>=3

n (%)

HBsAg loss

n (%)

(PEG)-IFN +/- NA

n=35

17 (48.6%)

18 (51.4%)

8 (22.9%)

8 (22.9%)

5 (14.3%)

4* (11.4%)

NA alone

n=45

33 (73.3%)

12 (26.7%)

4 (8.9%)

1 (2.2%)

0

0

Total

n=80

50

30

12

9

5

4

*1 patient cleared HBsAg during PEG-IFNalfa-2a rescue therapy for LMV resistance.

 


991. High Rates of HBsAg Seroconversion in Chronic Hepatitis B Patients Responding to Interferon Therapy: a Long-term Follow-up Study

R. Moucari; A. Korevaar; T. Asselah; O. Lada; N. Boyer; M. Martinot-Peignoux; P. Bedossa; P. Marcellin

 

Background and Aim

Loss of HBsAg with seroconversion to anti-HBs is the most desired treatment endpoint in chronic hepatitis B (CHB), and can be considered a complete response indicating resolution of hepatitis B. The difficulty is that HBsAg loss is rare, occurring spontaneously in less than 1% per year in inactive carriers and in patients receiving a one-year course of nucleos(t)ide analogue therapy. HBsAg loss is more frequent with interferon based therapy, occurring in up to 8% of patients three years post-treatment. The aim of this study was to assess the long term HBsAg response in a cohort of HBeAg-positive CHB patients treated with interferon.

 

Patients and Methods

All consecutive patients with HBeAg-positive CHB treated with conventional interferon between 1987 and 2000 were retrospectively evaluated. There were 97 patients. Sustained virological response (SVR) was defined as persistent HBeAg seroconversion and undetectable HBV DNA in serum 24 months after treatment discontinuation and during the follow-up period. Cumulative incidence of HBsAg seroconversion was assessed during the follow-up period and compared between patients with and without SVR.

 

Results

The baseline characteristics of the study population were: male gender (88%), mean age (39±13 years), mean ALT level 4.5±3.5 ULN, mean serum HBV DNA level 7.3±0.9 log10 copies/ml. Liver histology showed moderate-severe necroinflammation (METAVIR A2-A3) in 42%, and severe fibrosis (METAVIR F3-F4) in 44 % of patients.

 

SVR was achieved in 25 patients (26%). At multivariate analysis, SVR was independently associated with age > 45 years (p= 0.008, OR=4.6) and serum HBV DNA level < 6 log10 copies/ml (p=0.002, OR=9.4).

 

During a median follow-up period of 11 years (4-20), HBsAg seroconversion developed in 28 patients (29%), and was significantly more frequent in patients who achieved SVR than in those who failed interferon treatment (64% vs. 17%, p<0.0001).

 

The cumulative incidence of HBsAg seroconversion was significantly higher in patients who developed SVR compared with those who failed interferon treatment during the whole follow-up period (40% vs. 8% at year 5, and 60% vs. 18% at year 10 and 80% vs. 30% at year 15).

 

Five patients developed hepatocellular carcinoma (HCC) after a median period of 7 years (5-12) post interferon treatment. None of these patients had developed HBsAg seroconversion (p=0.02).

 

Conclusion

Patients with HBeAg-positive CHB achieved a 26% SVR rate following treatment with conventional interferon. Long-term follow-up showed that SVR was associated with a very high rate of HBsAg seroconversion (60% at 10 years) and a significant decrease in the incidence of HCC.

 


992. HBsAg Seroconversion: Different Kinetics of Serum HBsAg Level Decrease in Inactive Carriers and Chronic Hepatitis B Patients Treated with Interferon

R. Moucari; V. Mackiewicz; O. Lada; A. Devergne; N. Boyer; T. Asselah; M. Martinot-Peignoux; M. Vidaud; M. Nicolas-Chanoine; P. Marcellin

 

Background and Aim

HBsAg is used typically as a qualitative serological marker for diagnosing an ongoing HBV infection. HBsAg clearance is the most desired treatment endpoint in hepatitis B and considered as resolution of disease. Recently, a quantitative, fully automated chemiluminescent microparticle immunoassay (Abbott Architect HBsAg QT) has become available for the quantitation of HBsAg at a wide range of concentrations. The aim of this study was to assess the kinetics of HBsAg level decrease in serum of patients who have cleared HBsAg spontaneously or following interferon therapy.

 

Patients and Methods

22 HBV-infected patients who have cleared HBsAg have been retrospectively studied. Five patients were inactive carriers while 17 have been successfully treated with interferon. HBsAg clearance occurred after a median period of 6 years (2-14) after treatment discontinuation. Serum samples collected at yearly interval up to four years prior to HBsAg clearance and adequately kept frozen were retrospectively analysed. Samples were diluted 1:20 and 1:500 with the Architect HBsAg diluent in order to expand the upper limit of the dynamic range up to 125000 IU/ml.

 

Results

Inactive carriers had significantly lower HBsAg levels than interferon-treated patients (p<0.05),4 years (1.6 ± 0.4 vs. 2.7 ± 0.8 log10 IU/ml), 3 years (1.3 ± 0.4 vs. 2.4 ± 0.9 log10 IU/ml), 2 years (0.8 ± 0.7 vs. 1.9 ± 0.9 log10 IU/ml) and 1 year (0.3 ± 0.5 vs. 0.9 ± 0.8 log10 IU/ml)before HBsAg clearance.

 

Inactive carriers and interferon-treated patients had similar kinetics of decrease of HBsAg level in serum 4 years (0.5 ± 0.2 vs. 0.4 ± 0.3 log10 IU/ml) and 3 years (0.7 ± 0.3 vs. 0.7 ± 0.6 log10 IU/ml) prior to HBsAg clearance. However, interferon-treated patients showed a more rapid decline in HBsAg serum level compared to inactive carriers in the last 2 years prior to HBsAg clearance (p<0.05): year 2 (0.8 ± 0.5 vs. 0.5 ± 0.2 log10 IU/ml)and year 1 (0.9 ± 0.8 vs. 0.3 ± 0.5 log10 IU/ml)(figure).

 

Conclusion

These preliminary data suggest that Abbott Architect HBsAg QT is an effective assay for HBsAg quantitation in serum at wide range of concentrations. These data also suggest that interferon treatment accelerates HBsAg clearance. Further studies are ongoing to explore the clinical relevance of HBsAg quantitation for the monitoring of patients treated either with interferon or analogues.

 


1002. Predicting Sustained HBeAg Loss after Treatment with Peginterferon alpha-2b: Development and Validation of a Practical Model

E. H. Buster; B. E. Hansen; S. Zeuzem; S. W. Schalm; E. W. Steyerberg; H. L. Janssen

 

We developed and validated a model based on readily available factors for the prediction of response to peginterferon in individual patients.

 

Methods

Patients randomized to PEG-IFN α-2b 100μg alone (n=136) or in combination with lamivudine 100mg (n=130) for 52 weeks were analyzed. Univariate logistic regression analysis, followed by backward stepwise selection was used to identify predictors of HBeAg loss at 26 weeks post-treatment among the variables age, sex, HBV genotype (A-D), serum HBV DNA, ALT, GGT, treatment allocation and previous antiviral therapy. Interactions between variables and nonlinear relationships were investigated. Discrimination was quantified by the area under the receiver-operating characteristic curve (AUC). Bootstrap sampling was used for internal validation to reduce overfit bias.

 

Results

Twenty-three patients were excluded because of missing values or infection with HBV genotype other than A-D, leaving 233 patients for analysis. HBV genotype, serum HBV DNA, GGT and previous IFN therapy were found to be independent predictors of HBeAg loss (p<0.05). Since the influence of ALT was dependent on HBV genotype, an interaction between these variables was included (p=0.03). A multivariable model based on the above mentioned variables had adequate discriminative ability (AUC 0.73). After bootstrapping, the discriminative ability of the model was found to be somewhat lower (AUC 0.69). A nomogram was generated from the logistic regression formula (figure). The probability of HBeAg loss is calculated by drawing a vertical line from each of the 4 variable axis to the top Points axis. The sum of these 4 points is put on the Total score axis, from which a vertical line is drawn to the Chance of HBeAg loss axis. Conclusion: A multivariable model based on HBV genotype, serum HBV DNA, ALT, GGT and previous IFN therapy provides an adequate prediction of PEG-IFN induced HBeAg loss and will be a useful tool to guide treatment choice of PEG-IFN vs. nucleos(t)ide analogues in individual patients.