1457. Screening for Hepatitis B Virus (HBV) Infection by Primary Care Physicians in New York City: Are Screening Recommendations for Persons Born in Endemic Countries Being Followed?

H. Pollack; K. Wan; T. Miyoshi; G. Fryer; S. Tawdekar; P. Baker; D. McEwen; C. Weinbaum; S. R. Bialek; R. Low

 

Screening for hepatitis B virus (HBV) infection is recommended for persons born in regions where hepatitis B is endemic. Current physician practices with respect to screening for chronic HBV infection are largely unknown. In this study we examined screening practices for chronic HBV among primary care clinics in an urban hospital system.  De-identified patient information was abstracted from electronic medical records.

 

Aim

We present preliminary results on screening practices for 18,457 patients 19- 49 years of age, who first visited primary care clinics in the NYC Health and Hospital’s South Manhattan Network (including Bellevue Hospital Center) in New York City during 2005-2006. The patient population consisted of Asian/Pacific Islanders (26.1%), Blacks (12.7%), Hispanics (42.3%), Native Americans (0.2%), and Whites (7.0%). 69.6% were foreign born.

 

Among the 18,457 patients, 42.4% (7,826) were screened for HBsAg. 1026 (71.8%) of 1429 persons with elevated ALT were screened for HBsAg compared to 5826 (49.6%) of 11749 with normal ALTs, p<0.01. The screening rate was greatest for those 19-24 years of age (47.7%) and 25-29 years of age (48.5%), decreased among age group 30-34 (44.4%), 35-39(41.3%), 40-44 (37.0%) and least among the eldest group, 45-49 (34.4%), p for trend<0.01. Women were slightly more likely to be tested than men (43.4% vs. 41.3%, p<0.01). A higher rate of APIs were screened (65.6%) than Blacks (38.8%), Hispanics (33.4%), Whites (32.1%) and Native Americans (28.3%), p<0.01. Among patients born in regions with high rates of HBV infection (>=8%), those born in East Asia countries had the highest screening rate (692/1088, 63.6%) followed by Southeast Asia (153/304, 50.3%), and Africa (230/521, 44.1%). Patients born in regions of intermediate HBV endemicity (2-7%) including Eastern Europe, South Asia, Middle East, Central America, South America, and the Caribbean had screening rates of 35.6-40.1%, comparable to rates for patients born in the U.S. (39.5%).

 

The screening rates varied greatly among individual clinics, highest (1040/1810, 57.5%) at community clinics which served large numbers of APIs and lowest (25/212, 11.8%) at a community clinic which served large numbers of Hispanics.

 

Results

This pilot study revealed that screening for HBsAg was correlated with elevated ALT, younger age, Asian race and region of birth with HBV epidemic. However, nearly 50% of patients from countries where HBV infection is endemic, and for whom screening is recommended by the CDC, had not been tested. More rigorous efforts will be necessary to increase compliance for HBV screening in this group of patients in order to decrease long-term morbidity and mortality.

 


1473. Clinical and Laboratory Characteristics of Pregnant Women Chronically Infected with Hepatitis B Virus (HBV)

H. Pollack; K. Wan; T. Miyoshi; C. Peng; N. K. Sodhi; P. Baker; S. Yiu; G. Fryer

 

Most states in the U.S. mandate HBsAg testing of all pregnant women during prenatal care in order to prevent mother-infant-transmission. The clinical and laboratory characteristics of hepatitis B virus (HBV) infection in these women are largely unknown.

 

Methods

In this study, we reviewed the records of all HBV-infected women who received care in Prenatal Clinics at Bellevue Hospital and Gouverneur Healthcare Services in NYC between 2004 and 2006. Analysis was conducted retrospectively, through electronic medical record review of de-identified data, after IRB-approval. Out of 7,354 pregnant women tested for HBV, 352 (4.8%) were HBsAg+. The mean age of HBV-infected women was 27.1±5.3 years with a median age of 26.0 years. 82.1% were Asian/Pacific Islanders (APIs). 336 (95.5%) was tested for ALT, 318 (90.3%) for HBeAg, and 187 (53.1%) for HBV DNA.

 

Women with elevated ALT were more likely to be tested for viral load than those with normal ALT (87.5% vs. 53.4%, p<0.01) as were women who were HBeAg+ compare to those who were HBeAg- (65.3% vs. 47.4%, p<0.01). 39.0% (124/318, 95% CI: 33.7, 44.4) of the women were HBeAg+, and 4.8% (16/336, 95% CI: 2.8, 7.4) had elevated ALT (>2 x ULN). ALT levels ranged from 5 to 207, with a mean of 27 IU/ml. The median viral load was 1.0 X 104 copies / ml and mean viral load 1.8 X 108 copies / ml. 80.7% (151/187, 95% CI: 74.7, 86.0) had detectable virus and 39.6% (74/187, 95% CI: 32.8, 46.7) had a viral load >105 copies/ml.

 

Results

Among the 81 HBeAg+ women who had a viral load measurement, 29 (35.8%, 95% CI: 26.0, 46.5) had a viral load > 5.5 X108 copies/ml and at least 11 (13.6%) had a viral load > 109 copies/ml.

 

Conclusion

In conclusion, among HBsAg+ pregnant women in this study, 80% had active infection, almost 50% were at risk of significant future complications of chronic HBV infection, and as many as 13% may have had viral loads >109 copies/ml, associated with a higher risk of neonatal HBV prophylaxis failure. This analysis should be valuable for estimating the burden of HBV-related disease and projecting the cost of care among pregnant women with chronic HBV infection. Recommendations for the evaluation and care of women identified by prenatal HBV screening would be of great value for this vulnerable population as the diagnosis of hepatitis B during pregnancy is an opportune time to provide proper evaluation of risk for future HBV-related complications and the need for HBV-specific treatment.

 


147. A Novel Therapeutic HBV Vaccine Induces Potent Surface- and Core-specific Immunogenicity in Mice, Rhesus Macaques and HBV Transgenic Mice

J. D. Marshall; M. L. Gesner; D. S. Heeke; P. Buchmann; E. Martins; B. D. Livingston; K. Melber; G. Van Nest

 

HBV infects approximately 350 million people worldwide, and carriers are at increased risk for cirrhosis and hepatocellular carcinoma. Current HBV therapies do not eradicate HBV and have limited long-term efficacy. We have developed a novel immunotherapeutic vaccine for treatment of chronic HBV infection. The vaccine, composed of HBsAg (HBV surface antigen), HBcAg (HBV core antigen) and a proprietary adjuvant, was designed to induce virus-specific cellular immune responses such as are typically correlated with HBV clearance.

 

Mice immunized with the vaccine exhibited robust cellular and humoral immune responses to both HBcAg and HBsAg. The vaccine induced HBV-specific cell-mediated immunity in both CD4 and CD8 compartments. The total CD8-restricted T cell responses, including antigen-restricted responses directed against both HBcAg and HBsAg (as high 500 SFC/106 cells) were demonstrated using IFN-γ ELISPOT assays. Additionally, CTL lytic activity as measured by Granzyme B release was observed in response to HBsAg and HBcAg peptides.

 

The vaccine also elicited strong anti-HBsAg and anti-HBcAg immunoglobulin responses in both IgG1 and IgG2a isotypes. In rhesus macaques, immunization induced antigen-specific T cells at frequencies as high as 1000 SFC/106 PBMC. These T cells secreted IFN-γ and IL-2 in response to stimulation with either HBcAg or HBsAg peptides in ELISPOT assays. Treated rhesus also exhibited anti-HBsAg IgG levels, greater than 105 mIU/mL, after two immunizations. As a model for chronic hepatitis, we are conducting studies of HBV transgenic mice to examine the effect of this vaccine on viral replication, seroconversion, and generation of HBV-specific T cell responses.

 

Conclusion

Preliminary results indicate that the vaccine is effective in breaking immune tolerance, inducing HBV-specific T cell responses and high anti-HBsAg levels simultaneously with a vigorous multivalent cell-mediated T cell response. These data suggest this dual-antigen immunotherapeutic vaccine may be valuable in the treatment of chronic HBV infection. Currently, this vaccine is being tested in Phase 1 human clinical trials.

 


967. Amino Acid Variability within Hepatitis B Surface Antigen and the Overlapping Reverse Transcriptase Region in HBsAg-Negative/HbcAb-Positive Patients Presenting HBV Reactivation While Undergoing Chemotherapy and/or Stem Cell Transplantation for Cancer

R. Gerolami; P. Colson; P. Borentain; D. Coso; A. Charbonnier; A. Stoppa; T. Auran; E. Bories; A. Motte; M. Henry; D. Botta-Fridlund; C. Tamalet

 

Background/Aim

Hepatitis B virus (HBV) reactivation has been described in HBsAg-/anti-HBc+ patients (pts) undergoing immunosuppressive chemotherapy (CT). We previously reported HBV reactivation in 8.3% of pts (n=7) with HBV serological profile indicating past infected who underwent CT and/or hematopoietic cell transplantation (HCT) for cancer. Baseline anti-HBs titer <100 IU/L was significantly associated with HBV reactivation in these pts. In the present study, we aimed to analyse the amino acid (aa) variability and mutations patterns within HBsAg and the overlapping HBV reverse transcriptase (rt) region of HBV strains from these pts to assess whether HBV reactivation might be associated with specific virological features.

 

Patients/Methods

7 male pts (mean age, 60 years), HBsAg-/anti-HBc+ (HBsAb+/-) prior to CT and/or HCT who presented HBV reactivation between 11/2003 and 12/2005. 4/7 pts received anti-CD20. HBV HBsAg and rt genes were amplified/sequenced using in-house protocols. Newly-diagnosed HBsAg-positive pts (n=51) served as controls.

 

Results

HBV genotype was D and C in 5 and 2 cases, respectively. Mean number of substitutions/100 aa was significantly higher vs controls within HBsAg major hydrophilic region (MHR) (8.,6 vs 2.1; p<0,01), its “a” determinant (15.7 vs 2.5; p=0.003), and the HBsAg C-terminal region (7.5 vs 3.8; p=0,054). Substitutions D144A/V and G145R/A previously associated with altered S antigenicity were more frequent (from 2 pts each) than in controls (p=0.001 and p=0.004, respectively). Among genotype D sequences, substitutions at positions s116, s126, s134, and s144 within the MHR were significantly more frequent than in controls (p<0.05). No rt drug resistance mutation was detected at time of reactivation. However, within rt sub-domains B-C-D that overlap HBsAg C-terminal region and where drug resistance mutations occur, the mean number of substitution/100 aa showed a tendency to be higher in the reactivation group (3.0 vs 1.3; p=0.081), and a significantly higher proportion of sequences harbored >1 substitutions (P<0.05). Mutations rtR/W153Q, known to restore HBV replication of lamivudine (LAM) resistant strains and corresponding to substitutions sG145R, were found in HBV from 1 pt. In 1 pt, LAM-resistance was detected 1 month after LAM introduction.

 

Conclusions

Our data show an increased aa variability within HBsAg and HBV RT associated with HBV reactivation in AgHBs-/anti-HBc-positive patients undergoing CT and/or HCT. Whether or not these virological patterns may play a role in the pathogenesis and frequency of HBV reactivation or might impact the response to antiviral treatments needs further investigations.

 


970. Liver Biopsy: Still Essential in the Management of Chronic Hepatitis B

P. Kennedy; L. Jeyalingam; H. Lee; R. Goldin; P. Karayiannis; J. Main; M. R. Thursz; A. Brown; B. Smith; H. C. Thomas

 

Background and Aims

In the absence of an established role for non-invasive markers of fibrosis, liver biopsy remains the gold standard for the assessment and staging of disease in chronic hepatitis B virus (HBV). With growing numbers of chronic HBV patients attending hepatology clinics, the indications and timing of liver biopsy needs to be clearly defined. The aim of this study was to formally review all biopsies undertaken for a primary diagnosis of HBV in a tertiary referral centre and to assess how histology correlates with disease profile, in particular HBV DNA & ALT. Factors associated with more advanced disease or progression of disease on repeat biopsy were sought.

 

Patients and Methods

Between 1995 and 2007, 212 chronic HBV patients underwent liver biopsy. This comprised of 76 HBeAg+ patients (Group 1), 9 of whom underwent repeat biopsy and 136 anti-HBe+ patients (Group 2), 31 of whom underwent repeat biopsy. Two specialist liver histopathologists reviewed all biopsy material and determined Ishak fibrosis scores. A fibrosis score of 0-2 was designated mild disease, 3-4 moderate and 5-6 severe. Quantitative HBV DNA and ALT values [> or < twice upper limit of normal (2ULN)] at the time of biopsy were collected.

 

Results

Fibrosis score correlated with age at biopsy (p<0.01). In Group 1, 43 patients (57%) (M:F 29:14) had mild disease on histology: median age at biopsy was 30 years (range 16-64); 33 patients (43%) (M:F 27:6) had moderate/severe disease on biopsy with a median age at biopsy 37 years (range 16-73); 4/14 (29%) patients with severe disease had HBV DNA >107 and normal ALT were, therefore, considered immuno-tolerant. In Group 2, 71 patients (52%) had mild disease on biopsy, (M:F 50:21) median age at biopsy 36 years (range 18-70). The remaining 48% (65/136) had moderate/severe disease (M:F 50:15), median age at biopsy 47 years (range 23-73); 27/65 (42%) with moderate/severe fibrosis had both HBV DNA<105 and ALT <2ULN.

 

Repeat biopsy group Rate of fibrosis change in untreated patients undergoing serial biopsies was highly variable, ranging between improvements of fibrosis score of 0.44 points/year versus progression of one point per year (mean 0.22 points/year). Rate of progression did not correlate with eAg status, HBV DNA or ALT level. However, there was a trend towards faster progression in those with a BMI>25.

 

Conclusions

This study highlights the importance of liver biopsy in the diagnosis and management of chronic HBV. It emphasizes the inadequacy of quantitative HBV DNA and ALT values in reflecting underlying liver damage. We conclude that any decision not to offer antiviral therapy should be based on histological findings.

 


975. Intrahepatic and Serum Markers of HBV Replication and Their Relationship to Serum HBeAg Titers: Implications for the Use of Quantitative HBeAg Testing as a Predictive Tool for Treatment Outcome

A. Thompson; M. Littlejohn; A. Ayres; S. Lim; G. K. Lau; K. Visvanathan; S. D. Bowden; P. V. Desmond; S. Locarnini

 

Background

HBeAg seroconversion is one of the major therapeutic endpoints used when treating HBeAg-positive chronic hepatitis B (CH-B). Data from clinical trials suggests a potential role for quantitative HBeAg titres in guiding therapy. However, a commercial assay is not available and HBeAg titres have not been well-characterized in vivo. The relationship of HBeAg titre to viral load (VL) and cccDNA in vivo remains unresolved; interpretation may be confounded by the presence of the basal core promoter (BCP) or precore (PC) variants in the quasispecies pool. Although shown to reduce/abolish HBeAg production in vitro, their impact on HBeAg titre in vivo has not been defined.

 

Aims

1) To define the normal range of HBeAg titres seen in patients with HBeAg-positive CH-B in the immunoelimination phase of disease. 2) To correlate HBeAg titres with VL, liver cccDNA, viral genotype and the presence of PC/BCP mutations

 

Methods

A high throughput commercial HBeAg ELISA kit (Architect, Abbott Laboratories, Ill) was optimised for use as a quantitative assay using the Paul-Ehrlich (PE) reference standard. HBeAg titre was correlated with VL in all patients. Detailed virological characterization was performed in a subset; this included genotype, BCP/PC sequence, and intrahepatic cccDNA (Werle-Lapostolle, Gastro, 2004).

 

Results

The assay was validated using a cohort of 85 untreated HBeAg-positive patients in the immunoelimination phase of disease. Sequencing was performed in 30 patients. Intrahepatic cccDNA was measured in 17 patients. The linear range of the assay was 0.5 – 90 PE IU/ml. The observed HBeAg titres followed a log-normal distribution (median 1509, range 1–14236 PE IU/ml). Median VL was 3.43x108 IU/ml. Median cccDNA was 3.65 copies/Geq. VL correlated strongly with cccDNA (r =.81, p <.05). HBeAg titre correlated modestly with VL (r =.50, p-value <.001) and cccDNA (r =.48, p-value =.05). The lack of a stronger correlation was attributed to the presence of BCP and PC variants in the quasipspecies pool (dominant virus WT/WT in 17/30, BCP/WT in 9/30, and WT/PC in 4/30; median HBeAg titres 3095, 209 and 149 respectively, p <.05 vs WT). The effect of the BCP/PC variants was independent of VL. HBeAg titre did not differ between genotypes A - D.

 

Conclusion

We have optimized a commercial HBeAg assay for use as a sensitive and accurate quantitative tool. HBeAg titers were log-normal distributed and modestly correlated with VL and cccDNA, which was due to the presence of BCP/PC variants reducing HBeAg titre independent of VL. BCP/PC sequencing should be incorporated into research protocols evaluating the utility of HBeAg titers in predicting treatment outcome.

 


984. Management of Chronic Hepatitis B Virus (HBV) Infection by Primary Care Physicians in Urban Hospitals and Clinics in New York City

H. Pollack; K. Wan; T. Miyoshi; S. Tawdekar; P. Baker; D. McEwen; C. Weinbaum; S. R. Bialek; G. Fryer; R. Low

 

Current primary care physician practices with respect to the management of persons with chronic HBV infection are largely unknown and comprehensive recommendations directed to primary care physicians are lacking. In this pilot study we examined chronic hepatitis B disease management practices in primary care settings in New York City HHC South Manhattan Network Hospitals and clinics.

 

Analysis was conducted, retrospectively, through electronic medical record review. We present preliminary results for patients 19-49 years of age who first visited primary care clinics in affiliated facilities during January 2005-December 2006. For the purposes of this analysis, chronic HBV infection was defined as a positive test for hepatitis B surface antigen (HBsAg). The minimum adequate evaluation for chronic HBV infection was defined as testing for HBeAg, VL and ALT. Among the 18,457 patients, 42.4% (7,826) were screened for HBsAg. 893 were HBsAg+. The mean age of HBV-infected patients was 31.8 ±7.4 years, 57.8% were men. The HBV-infected patients were overwhelmingly APIs (83.9%), followed by Blacks (7.8%), Hispanics (3.6%), and Whites (0.9%). 97.9% of HBV-infected persons were tested for ALT, 71.0% for HBeAg, 68.2% for viral load, and 63.6% for AFP. The rates of testing for HCVAb, HAVAb, HDVAb and HIV were 52.0%, 47.3%, 3.8% and 22.8%, respectively. Approximately one third of the patients (37.8%) received an abdominal ultrasound exam, 9.9% had a CT scan and 0.6% had a MRI. 564 patients (63.2%) had HBeAg, VL and ALT measurements and 306 patients (34.3%) had all three measurements and an ultrasound exam. 83.1% of those with abnormal ALTs had HBeAg and VL tests, higher than 58.5% among those with normal ALTs, p<0.01. Management practices did not differ by patient sex (p=0.51), race/ethnicity (p=0.07), age (p=0.18) or country of birth (p=0.82). Patients at hospital facilities were more likely to receive the minimum adequate evaluation (68.6%, and 64.9%, respectively), compared to patients at community clinics (45.5%, and 29.3%, respectively), p<0.01.

 

Conclusion

In conclusion, the management of patients with chronic HBV infection by primary care physicians in this study was suboptimal according to practice guidelines established by the AASLD. Additional education and training of primary care physicians in the proper management of these patients needs to be implemented and evaluated in order to decrease the long-term morbidity and mortality of persons chronically infected with HBV.

 


988. Screening for Hepatitis B in Chemotherapy Patients: Survey of Current Oncology Practices

T. Tran; M. Oh; F. Poordad; P. Martin

 

Background

Hepatitis B virus (HBV) reactivation occurs in up to 78% of patients receiving cytotoxic chemotherapy for non-hepatic malignancies. Consequences of reactivation include hepatic dysfunction, jaundice, and fulminant hepatic failure. Patients with hepatic failure due to reactivation are generally precluded from liver transplantation secondary to underlying malignancy and antiviral therapy may be ineffective in reducing liver injury during chemotherapy induced flares. Current recommendations to prevent chemotherapy induced flares include screening patients at risk for HBV prior to immunosuppressive therapy and providing a short course of lamivudine in patients with chronic HBV as pre-emptive therapy against reactivation. Current practice among oncologists is unknown in regard to screening and prophylaxis strategies for HBV.

 

Aims

To assess the practice of oncologists in screening and management of HBV prior to initiating chemotherapy.

 

Methods

A survey was sent to AMA registered oncologists in the United States assessing practice demographics and HBV screening practices. Fischer’s exact test was used for analysis.

 

Results

Responses were obtained from 265 individuals: 38% (n=101) from an office-based practice, 55% (n=145) from a university/academic/research institution, and 6% (n=17) from other practice setting. Office-based physicians were less likely to screen for HBV prior to chemotherapy (p<0.001). Years in practice varied: 51% with <5 years (n=136), 29% with 5-15 years (n=77), and 18% with >15 years (n=49). No difference was found in screening practices based on years of experience (p=NS). Surveyors screen for HBV as follows: never- 20% (n=54), abnormal liver tests- 30% (n=80), risk factors or history of hepatitis- 38% (n=101), randomly- 11% (n=28), always- 13% (n=35). Most oncologists used HB surface antigen as a screening test (76%), in conjunction with surface antibody (64%), and core antibody (46%). In pts with HBV, 75% (n= 200) of oncologists refer to specialists and 7% (n=18) initiate therapy themselves, while 15% (n=39) do not refer or initiate therapy, most of whom are in an office setting (p=0.02).

 

Conclusions

Twenty percent of oncologists never screen for HBV prior to initiating chemotherapy, and fewer than 40% screen even if patients have risk factors or a history of hepatitis. Screening practices were not influenced by years in practice, however office-based physicians were less likely to screen, treat, or refer to a specialist prior to chemotherapy. Greater education and awareness regarding risk of HBV reactivation is needed for clinicians treating patients with immunosuppressive therapies.

 


993. High Prevalence of Significant Histologic Disease in Patients with Chronic Hepatitis B and Normal ALT

M. H. Nguyen; H. N. Trinh; R. T. Garcia; J. Phan; G. H. Nguyen; G. Weiss; H. Nguyen; K. Nguyen; E. B. Keeffe

 

Purpose

Current guidelines recommend therapy for CHB with elevated ALT. However, many patients presenting with “normal” ALT may not have persistently normal levels on f/u. Our goal was to define the spectrum of histologic findings in such patients.

 

Methods

We performed a retrospective cohort study of all patients with active CHB (HBV DNA≥10000 copies/mL) & ALTUNL. Mild steatosis was seen in 28% and moderate in 11% but this finding was not a significant predictor. The overall prevalence of significant histology was 43.4% & increased with age (Figure): 31% in those with persistently normal ALT and 63% in those without. On univariate analysis, predictors of significant histology were age, weight, HBeAg negativity, & abnormal ALT on f/u. On multivariate analysis, predictors were elevated ALT (OR=5.1, p<0.0001) and age (OR=5.2, p=0.037 for age 35-49; OR=11.2, p=0.005 for age 50-64; & OR=28.0, p=0.007 for age≥65).

 

Conclusion

Only 62% of patients with normal ALT at evaluation have persistently normal levels on follow-up, a third of whom can still have significant histology. Older age starting at 35 is the strongest independent predictors of significant histology. Asian patients with normal ALT should be followed closely and histologic evaluation should be considered if patients are ≥ 35 or if ALT becomes elevated.

 

 


1001. Clinical Features of Lymphoma Chemotherapy Associated Hepatitis Virus B Reactivation.

Q. Xie; D. Shi; H. Gui; H. Wang; N. Jia

 

Objective

Investigate the incidence of Hepatitis Virus B reactivation and its related severity of hepatitis after the Non-Hodgson’s Lymphoma (NHL) patients had been administered with chemotherapy, recognize the significance of detecting HBV-DNA and employing prophylactic antiviral therapy.

 

Methods

Retrospective study was conducted on 252 NHL patients who had received chemotherapeutics, among which 166cases with HBV serologic test negative, 86 cases with HBV positive ( 60 with HBsAg positive, 26 with HBsAg negative and anti HBc-IgG positive).

 

Results

After these 252 NHL patients received chemotherapy, 56 cases occurred hepatic biochemical abnormality, among which 11 with HBV negative (6.6%), 45 with HBV positive (52.3%), there was a statistically significant difference between these 2 groups (P<0.01). In addition, within the HBV positive group, 39 cases with HBsAg positive (65.0%) and 6 cases with HBsAg negative and anti HBc-IgG positive (23.1%) . There was also a statistical difference between these 2 groups (P<0.05). The prevalence of HBV reactivation in patients with HBsAg positive was much higher when compared with those with HBsAg negative and anti HBc-IgG positive (53.3% vs 7.7%, P<0.01). 32 among 39 patients with HBsAg positive occurred liver function abnormal within three cycles of chemotherapy, nevertheless 4 among 6 patients with HBsAg negative and anti HBc-IgG positive occurred liver function abnormal following five cycles of chemotherapy.

 

Conclusions

NHL patients with HBV positive have a higher incidence of hepatic biochemical abnormality than those with HBV negative. Patients with HBsAg positive developed abnormal liver function easily than those with HBsAg negative and anti HBc-IgG positive. HBV reactivation usually occurs on patients with HBsAg positive and especially those with higher viral load. HBV reactivation delayed on the NHL patients with HBsAg negative and anti HBc-IgG positive or even with anti HBsAb positive. Our results suggest that the importance of HBV viral load and liver function surveillance routinely during chemotherapy and necessity of prophylactic antiviral therapy to patients with hepatitis B carriers at the onset of chemotherapy.

 


1008. Inflammation and Fibrosis in Patients with Chronic Hepatitis B Despite Low Viral Load at the Time of Liver Biopsy

M. Gambarin-Gelwan; E. B. Grossman; T. A. Hahambis; R. K. Yantiss; B. R. Edlin; I. M. Jacobson

 

Background

Three hundred and fifty million people worldwide have chronic hepatitis B (CHB). Current treatment guidelines recommend starting treatment for patients with elevated liver enzymes and hepatitis B (HBV) DNA ≥100,000 copies/ml in HBeAg+ or ≥10,000 copies/ml in HBeAg- patients. However, the threshold viral load (VL) level associated with significant liver fibrosis is unknown.

 

Objective

To compare the degree of fibrosis and inflammation among CHB patients with low-level vs. high-level viremia.

 

Methods

CHB patients with a liver biopsy from January 1995 to May 2007 and contemporaneous ALT and HBV DNA levels were identified by electronic chart review. Patients with co-existing liver disease (other than NAFLD) or HIV infection were excluded. Normal ALT was defined as ≤30 U/L in men and ≤19 U/L in women. Low-level viremia (LLV) was defined as HBV DNA <100,000 copies/mL in HBeAg+ and <10,000 copies/mL in HBeAg– patients. Liver biopsies were evaluated for fibrosis and necroinflammation according to the modified Scheuer scoring system.

 

Results

Of 53 patients (mean age 40.3y; range 23-69y; 58% male) who met our inclusion criteria, 34 (64%) were HBeAg-. Of these, 9 (26%) had LLV (78% female), including 4 with HBV DNA <1,000 copies/ml (range 185-799 copies/ml). HBV DNA was ≥100,000 copies/mL and ALT was elevated in all HBeAg+ patients. Inflammation was mild in 89% (8/9) of the HBeAg– patients with LLV, 84% (21/25) of HBeAg- patients with high-level viremia (HLV), and 79% (15/19) of HBeAg+ patients. Of 9 HBeAg- patients with LLV, 5(56%) had at least stage 2 fibrosis (including one patient with incomplete cirrhosis), as compared with 52% (13/25) of HBeAg- patients with HBV DNA ≥10,000 copies/ml and 53% (10/19) of HBeAg+ patients. Of the 4 patients with HBV DNA <1,000 copies/mL, 2 (50%) had stage 2 fibrosis. Even among the 5 patients with LLV and normal ALT, 2 (40%) had at least stage 2 fibrosis, including one with incomplete cirrhosis. Stage 2 fibrosis was no more common among patients with elevated ALT and high VL [47% (9/19) of HBeAg- and 53% (10/19) of HBeAg+ patients] for whom treatment is recommended than in patients with normal ALT or low VL [60%(9/15) (all HBeAg-)].

 

Conclusion

Neither HBeAg status nor viral load predicted more severe fibrosis or necroinflammation in our patients. In HBeAg- patients with low-level viremia, a significant proportion had at least stage 2/4 fibrosis, including one with incomplete cirrhosis. More data are needed on the natural history of CHB in patients with low viral loads. For now, liver biopsy should be considered in the evaluation of HBeAg- patients with low-level viremia including those with normal ALT.

 


901. FibroTest® (FT) and ActiTest® (AT) Accurately Predict Risk of Liver Decompensation and Death in Patients with Chronic Hepatitis B (CHB)

T. Poynard; Y. Ngo; Y. Benhamou; M. Munteanu; V. Thibault; P. Lebray; D. Thabut; R. Morra; D. Messous; F. Imbert-Bismut; V. Ratziu

 

Background

FT, and AT, noninvasive methods of measuring biomarkers of liver fibrosis and necroinflammatory activity, are an alternative to liver biopsy for determining the severity of CHB. We assessed the 4-year prognostic value of FTAT staging for predicting cirrhosis decompensation and survival.

 

Methods

Five hundred and four consecutive HBV patients with baseline FTAT and HBV DNA measurements were retrospectively analyzed. FT-AT, fibrosis stage and activity grades were assessed simultaneously in 72 patients. Disease classification at baseline was 115 patients with severe fibrosis (FibroTest >0.58), 126 with moderate fibrosis (FibroTest 0.32– 0.58), and 263 with no or minimal fibrosis (FibroTest <0.32).

 

Results

The area under the ROC curve (AUROCs) for the diagnosis of advanced fibrosis (severe or moderate) was 0.86 (0.05) m(SE). In patients with severe fibrosis, 4-year survival without HBV complications was 63% [95% confidence interval (CI), 53%– 72%; 39 complications], and survival was 77% (95% CI, 68%–86%; 22 deaths). Survival rates were higher in patients with moderate fibrosis, [98% (95% CI, 95%–100%; 4 complications; P <0.001) and 100% (no death; P <0.001) for survival without HBV complications and overall survival, respectively], and in patients with minimal fibrosis [99.6% (95% CI, 99%–100%; 2 complications; P <0.001vs severe) and 100% (no death; P <0.001 vs severe), respectively].

 

FibroTest was a better predictor than HBV DNA quantification for HBV complications, with AUROC= 0.91 (95% CI, 0.89–0.96) vs 0.57 (95% CI, 0.85– 0.94; P=0.01), respectively; it was also a better predictor for deaths: AUROC= 0.95 (95% CI, 0.91– 0.97) vs 0.57 (95% CI, 0.42–0.69; P <0.001), respectively. FT was a similar predictor to biopsy staging for HBV complications and deaths, with both AUROCs= 0.99 (95% CI, 0.90– 0.998) vs 0.97 (95% CI, 0.93–0.99; P =0.32), respectively. In Cox multivariate analyses the prognostic value of FT was still significant (P <0.001) after taking into account baseline viral load (P=0.005), gender (P=0.04), age, HBeAg status, treatment, alcohol consumption, ethnic origin, AT and HIV coinfection. An algorithm enabled an AUROC=0.93 (0.89-0.96) (P<0.001) to be obtained for the prediction of death or complications.

 

Conclusion

FibroTest has a 4-year prognostic value similar to that of liver biopsy in patients with chronic hepatitis B and can be combined with baseline viral load to improve prediction for death and complications. This study was supported by IDENIX PHARMACEUTICAL.

 


1457. Screening for Hepatitis B Virus (HBV) Infection by Primary Care Physicians in New York City: Are Screening Recommendations for Persons Born in Endemic Countries Being Followed?

H. Pollack; K. Wan; T. Miyoshi; G. Fryer; S. Tawdekar; P. Baker; D. McEwen; C. Weinbaum; S. R. Bialek; R. Low

 

Screening for hepatitis B virus (HBV) infection is recommended for persons born in regions where hepatitis B is endemic. Current physician practices with respect to screening for chronic HBV infection are largely unknown. In this study we examined screening practices for chronic HBV among primary care clinics in an urban hospital system.  De-identified patient information was abstracted from electronic medical records.

 

Aim

We present preliminary results on screening practices for 18,457 patients 19- 49 years of age, who first visited primary care clinics in the NYC Health and Hospital’s South Manhattan Network (including Bellevue Hospital Center) in New York City during 2005-2006. The patient population consisted of Asian/Pacific Islanders (26.1%), Blacks (12.7%), Hispanics (42.3%), Native Americans (0.2%), and Whites (7.0%). 69.6% were foreign born.

 

Among the 18,457 patients, 42.4% (7,826) were screened for HBsAg. 1026 (71.8%) of 1429 persons with elevated ALT were screened for HBsAg compared to 5826 (49.6%) of 11749 with normal ALTs, p<0.01. The screening rate was greatest for those 19-24 years of age (47.7%) and 25-29 years of age (48.5%), decreased among age group 30-34 (44.4%), 35-39(41.3%), 40-44 (37.0%) and least among the eldest group, 45-49 (34.4%), p for trend<0.01. Women were slightly more likely to be tested than men (43.4% vs. 41.3%, p<0.01). A higher rate of APIs were screened (65.6%) than Blacks (38.8%), Hispanics (33.4%), Whites (32.1%) and Native Americans (28.3%), p<0.01. Among patients born in regions with high rates of HBV infection (>=8%), those born in East Asia countries had the highest screening rate (692/1088, 63.6%) followed by Southeast Asia (153/304, 50.3%), and Africa (230/521, 44.1%). Patients born in regions of intermediate HBV endemicity (2-7%) including Eastern Europe, South Asia, Middle East, Central America, South America, and the Caribbean had screening rates of 35.6-40.1%, comparable to rates for patients born in the U.S. (39.5%).

 

The screening rates varied greatly among individual clinics, highest (1040/1810, 57.5%) at community clinics which served large numbers of APIs and lowest (25/212, 11.8%) at a community clinic which served large numbers of Hispanics.

 

Results

This pilot study revealed that screening for HBsAg was correlated with elevated ALT, younger age, Asian race and region of birth with HBV epidemic. However, nearly 50% of patients from countries where HBV infection is endemic, and for whom screening is recommended by the CDC, had not been tested. More rigorous efforts will be necessary to increase compliance for HBV screening in this group of patients in order to decrease long-term morbidity and mortality.

 


1473. Clinical and Laboratory Characteristics of Pregnant Women Chronically Infected with Hepatitis B Virus (HBV)

H. Pollack; K. Wan; T. Miyoshi; C. Peng; N. K. Sodhi; P. Baker; S. Yiu; G. Fryer

 

Most states in the U.S. mandate HBsAg testing of all pregnant women during prenatal care in order to prevent mother-infant-transmission. The clinical and laboratory characteristics of hepatitis B virus (HBV) infection in these women are largely unknown.

 

Methods

In this study, we reviewed the records of all HBV-infected women who received care in Prenatal Clinics at Bellevue Hospital and Gouverneur Healthcare Services in NYC between 2004 and 2006. Analysis was conducted retrospectively, through electronic medical record review of de-identified data, after IRB-approval. Out of 7,354 pregnant women tested for HBV, 352 (4.8%) were HBsAg+. The mean age of HBV-infected women was 27.1±5.3 years with a median age of 26.0 years. 82.1% were Asian/Pacific Islanders (APIs). 336 (95.5%) was tested for ALT, 318 (90.3%) for HBeAg, and 187 (53.1%) for HBV DNA.

 

Women with elevated ALT were more likely to be tested for viral load than those with normal ALT (87.5% vs. 53.4%, p<0.01) as were women who were HBeAg+ compare to those who were HBeAg- (65.3% vs. 47.4%, p<0.01). 39.0% (124/318, 95% CI: 33.7, 44.4) of the women were HBeAg+, and 4.8% (16/336, 95% CI: 2.8, 7.4) had elevated ALT (>2 x ULN). ALT levels ranged from 5 to 207, with a mean of 27 IU/ml. The median viral load was 1.0 X 104 copies / ml and mean viral load 1.8 X 108 copies / ml. 80.7% (151/187, 95% CI: 74.7, 86.0) had detectable virus and 39.6% (74/187, 95% CI: 32.8, 46.7) had a viral load >105 copies/ml.

 

Results

Among the 81 HBeAg+ women who had a viral load measurement, 29 (35.8%, 95% CI: 26.0, 46.5) had a viral load > 5.5 X108 copies/ml and at least 11 (13.6%) had a viral load > 109 copies/ml.

 

Conclusion

In conclusion, among HBsAg+ pregnant women in this study, 80% had active infection, almost 50% were at risk of significant future complications of chronic HBV infection, and as many as 13% may have had viral loads >109 copies/ml, associated with a higher risk of neonatal HBV prophylaxis failure. This analysis should be valuable for estimating the burden of HBV-related disease and projecting the cost of care among pregnant women with chronic HBV infection. Recommendations for the evaluation and care of women identified by prenatal HBV screening would be of great value for this vulnerable population as the diagnosis of hepatitis B during pregnancy is an opportune time to provide proper evaluation of risk for future HBV-related complications and the need for HBV-specific treatment.

 


1481. Diagnostic Accuracy of Blood Tests of Liver Fibrosis in Chronic Hepatitis B: Comparison with Hepatitis C

V. Leroy; N. Sturm; M. Hilleret; R. Patrick; C. Trocmé; F. Patrice; J. Zarski

 

Several blood tests are currently used for the non-invasive evaluation of liver fibrosis. They have been mainly described and validated in chronic hepatitis C (CHC). Their diagnostic accuracy is however poorly documented in patients with chronic hepatitis B (CHB). The aim of this study was to describe the diagnostic performance of a panel of blood tests of fibrosis in CHB compared to CHC.

 

Methods

Of patients seen in our center for a pre-therapeutic liver biopsy between 2000 and 2007, 510 were recruited. They included 255 CHC patients and 255 CHB (monoinfected) patients, matched on the stage of fibrosis. Blood tests (Fibrotest, Hepascore, Fibrometer and MP3) were assessed on frozen serums collected the day of the biopsy. Histological lesions were staged according to the METAVIR system. Perisinusoidal fibrosis was noted. Areas of fibrosis were quantified by morphometry in 100 patients (20 per stage of fibrosis).

 

Results

CHC were significantly older (47 vs 40 years, p<0.01), had lower ALT levels (73 vs 93 UI/ml, p<0.005) and higher GGT levels (91 vs 53 UI/ml, p<0.01) than CHB patients. Other characteristics including alcohol consumption, Metavir activity and biopsy length (median 23 mm) were similar. Fibrosis stages were distributed as follow : F0 n=72, F1 n=192, F2 n=132, F3 n=54, F4 n=56. Diagnostic accuracies of blood tests for significant fibrosis (F0F1 vs F2F3F4) were compared between CHC and CHB and were as follow (AUROCs) : Fibrometer : 0.81 vs 0.82, Fibrotest : 0.81 vs 0.78, MP3 : 0.80 vs 0.76 Hepascore : 0.79 vs 0.77 . For the diagnosis of extensive fibrosis the best result was oberved for Fibrometer (0.89 vs 0.89). Test performance profiles were also evaluated. The rate of misclassification was significantly higher in CHB compared to CHC in patients with early stages of fibrosis (F0F1F2) (41% vs 30%, p<0.01 for Fibrotest). This was especially true for discriminating F1 vs F2. Morphometric analysis showed a significant correlation between area and stage of fibrosis (r=0.82,p<0.001), with the notable exeption of F1 vs F2 (4.1 vs 4.0%, NS). Comparisons between CHC and CHB for F0, F1 and F2 stages showed greater areas of fibrosis in CHC (on average 2-fold, p<0.001), that was explained in part by more pronunced perisinusoidal fibrosis in CHC. No difference was observed between CHC and CHB for F3F4 stages.

 

Conclusion

Our results show that in CHB blood tests of fibrosis have good global diagnostic accuracies especially for the diagnosis of extensive fibrosis. However, test performance profiles are altered in early stages of fibrosis (F0F1F2), a result explained by a different distribution of liver fibrosis in CHB compared to CHC.