- Email: [email protected]
Clinical and virological outcomes of hepatitis B virus inactive carriers and gray zone patients: a comparative analysis using current EASL and AASLD Guidelines criteria
POSTER PRESENTATIONS levels were quantified in baseline samples of 366 HBeAg positive CHB patients who participated in 2 global trials and had not received antiviral treatment in the preceding 6 months. Results: Of all patients, 274 (75%) were male. The mean age was 33 (SD 11) years, and genotypes A/B/C/D/other were present in 20/13/ 30/24/2%. The mean serum level of HBV RNA, HBV DNA, qHBsAg, qHBeAg, and ALT were 6.5 (SD 1.2) log c/mL, 8.1 (SD 1.1) log IU/mL, 4.3 (SD 0.7) log IU/mL, 2.3 (SD 0.9) log IU/mL, and 3.8 (SD 3.4) × ULN, respectively. In one patient, flRNA was below LLD (28-year-old female, genotype D). Overall, levels of HBV RNA correlated strongly with HBV DNA (r = 0.72, p < 0.001), moderately with qHBsAg (r = 0.54, p < 0.001) and qHBeAg (r = 0.41, p < 0.001), and weakly with ALT (r = 0.22, p < 0.001). Correlations however were HBV genotype dependent (Figure A), showing strongest correlation with HBV DNA for genotype A, strongest correlation with qHBsAg for genotype B and C, and a weak correlation with qHBsAg for genotype D. By multivariable linear regression, genotype ( p < 0.001) and any BCP and/or PC mutation ( p < 0.001) were associated with HBV RNA level. The estimated marginal mean (EMM) for wildtype HBV was 6.8 (SE 0.2) log c/mL, which was comparable to the level in presence of PC mutation (EMM 6.8 [SE 0.2] log c/mL, p = 1.00), but higher than HBV RNA level in presence of BCP mutation, either alone (6.0 [SE 0.2], p = 0.001) or in combination with PC mutation (6.5 [SE 0.1], p = 0.001). Figure B shows estimated marginal means for HBV genotypes.
Methods: We analyzed 123 antiHBe+ non treated patients (HBVDNA < 20000 IU/ml and ALT < 2ULN at inclusion and during followup) with 2 valid TE separated with a 3 year interval. 53% female, age 44 ± 11 years, most infected by genotypes A/D. At inclusion, 28% had HBV-DNA < 2000 IU/ml and HBsAg < 1000 IU/ml; during the followup, 49% remained with HBV-DNA < 2000 IU/ml and 80% had persistently normal ALT. 23% had steatosis and 4% diabetes. The influence on TE value of demographic, clinical and virological variables was analyzed. Results: Mean basal TE was 5.6 ± 1.6 kPa and at 3 years 5.8 ± 1,8 kPa ( p = 0.24). TE value increased in 63(1.3 ± 1.1 kPa), did not change in 6 and decreased in 54(1.1 ± 0.9 kPa). In 28% the increase was ≥1 kPa. Comparing patients with decrease in TE (n = 54) and those without change or increase (n = 69), there were no differences in any of the analyzed variables. Comparing patients with an increase ≥1 kPa (n = 35) with those with no change or decrease (n = 60), the former had higher AST levels (27.8 ± 8,9 vs.22.7 ± 6,9; p = 0.054), without differences in the rest of the variables. TE value was ≤6,5 kPa in 73% in the initial study and in 76% in the second one. Comparing patients with TE ≤ 6.5 and >6.5 in the second TE, the former had higher basal levels of HBsAg (3.4 ± 1.0 vs 2.9 ± 1.0 log10 IU/ml, p = 0.03), higher levels of basal HBV-DNA (2795 ± 5371 vs 836 ± 1862 IU/ml; p = 0.05) and of highest HBV-DNA during the follow-up (8669 ± 13467 vs 3295 ± 5902; p = 0.03), while those with TE > 6.5 had higher AST levels (29.9 ± 9.1 vs. 23.8 ± 9.1; p = 0.027) and BMI (27.9 ± 5.4 vs. 25.5 ± 5.3; p = 0.038). Diabetes was more frequent among those with TE > 6.5 kpa (10% vs. 2.1%;p = 0.09), as was steatosis (36% vs. 20%; p = 0.08). There were no differences in sex, age, basal ALT, highest ALT, platelets or proportion of patients with persistently normal ALTor with HBV-DNA < 2000 IU/ml. Conclusions: TE values in patients with CHB antiHBe+ patients no candidates to antiviral treatment have little variations over a three years period. Parameters linked to HBV activity seem to have no influence on TE increase. Nevertheless, parameters as AST level and BMI are related with higher TE values, suggesting that metabolic rather than virological factors induce fibrosis progression in these patients. THU-149 Clinical and virological outcomes of hepatitis B virus inactive carriers and gray zone patients: a comparative analysis using current EASL and AASLD Guidelines criteria M.S. Bonacci1, S. Lens1, Z. Mariño1, M.C. Lodoño1, S.P. del Pulgar1, J. Costa1, J.M. Sanchez-Tapias1, X. Forns1. 1Liver Unit, Hospital Clinic Barcelona, Barcelona, Spain E-mail: [email protected]
Conclusions: Serum full length HBV RNA levels in HBeAg positive CHB are independently associated with genotype and presence of HBV basal core promoter mutation. In addition, correlations of HBV RNA with other serum markers are genotype dependent. The effect of these factors should be taken into consideration when studying the role of HBV RNA in the HBV life cycle and as a biomarker for treatment response in HBeAg positive CHB. THU-148 Transient elastography evolution in non treated patients with chronic hepatitis B (CHB) infection M.L. González-Diéguez1, M.A. de Jorge1, C. Alvarez-Navascues1, C. Rodríguez-Escaja1, E. Rubio1, A. Castaño1, V. Cadahia1, M. Varela1, M. Rodriguez1. 1Hepatology Unit, Hospital Universitario Central de Asturias, Oviedo, Spain E-mail: [email protected] Background and Aims: TE has shown usefulness in predicting histological fibrosis in patients with CHB; a TE value <6,5 kPa has been proposed to exclude significant fibrosis (Papatheodoridis GV, et al. JVH 2014). To know TE evolution in patients with antiHBe + CHB not candidates for treatment S254
Background and Aims: Distinction between the inactive HBV carrier (IC) state and progressive chronic hepatitis B may be difficult due to fluctuations of ALT and HBV-DNA levels which lead to the so-called “gray zone” [(GZ)HBV-DNA <2.000 IU/mL and minimally elevated ALT] with uncertain prognosis. The scenario is further complicated by different definitions of IC state by the EASL and AASLD current Guidelines. We aimed to assess the natural history of these patients considering both definitions of IC. Methods: We analyzed the outcome of a series of chronic HBV patients fulfilling the AASLD or EASL criteria for IC and GZ visited at our Liver Unit since 1985. Previously treated or patients with cirrhosis or other causes of liver disease were excluded Results: 197 patients were included. Considering the highest ALT value observed during the first year of follow-up (FU), patients were separated into four groups (Table). Importantly, 82 (60%) ICs according to EASL would have been classified as GZs by AASLD. There were no differences between groups concerning demographic characteristics. After a median FU of 95 (52–217) months, outcomes did not differ across groups. Overall, the 10-year probability of at least one HBV-DNA peak >2.000 UI/mL was 20% and >20.000 IU/mL in 5%. At the end of FU HBV-DNA was 2.000–20.000 IU/mL [median ALT 27 (23–36 IU)] in around 10% but only two subjects progressed to CHB
Journal of Hepatology 2017 vol. 66 | S95–S332
POSTER PRESENTATIONS and needed therapy (Table). No patient developed hepatocellular carcinoma nor significant liver fibrosis [liver stiffness 4.9 kPa (4–5.7)]. The rate of HBsAg loss at 10 years was 20% and not different between groups. The number of HBV-DNA peaks correlated inversely with the probability of HBsAg loss (HR = 0.25 [0.1–0.6] p = 0.03). Classification at baseline IC EASL
IC AASLD
ALT < 40 IU ALT < 19 IU (F) Classification at the end of follow-up, n (%) IC EASL GZ EASL IC AASLD GZ AASLD HBsAg loss Chronic hepatitis
(n = 137) 98 (71,5) 16 (11.5) – – 23 (17) –
ALT < 30 IU (M) (n = 55) – – 30 (55) 19 (34.5) 6 (11) –
GZ EASL ALT (40 to 80 IU)
(n = 60)
GZ AASLD ALT 19 to 38 IU (F) ALT 30 to 60 IU (M) (n = 142)
38 (63) 8 (14) – – 12 (20) 2 (3)
– – 59 (41.6) 52 (36.6) 29 (20.4) 2 (1.4)
Conclusions: Independently of the criteria used for diagnosis, and despite relatively frequent peaks of HBV-DNA during FU, ICs and GZ patients have an excellent long-term prognosis and a relatively high rate of HBsAg loss. The use of a restrictive IC definition (AASLD guidelines) does not seem to add a benefit regarding prognosis and may increase the need for liver biopsies as well as more follow-up visits. THU-150 Liver histologic spectrum in hepatitis B virus patients co-infected with the human immunodeficiency virus R. Sterling1, D.K. Kleiner2, W.S. Abdus3, M. Khalili4, M. Ghany5, D.K. Wong6, R.T. Chung7, M. Jain8, M. Lisker-Melman9 and Hepatitis B Research Network – HBV/HIV Cohort Study. 1GastroenterologyHepatology-Nutrition, Virginia Commonwealth University, Richmond; 2 Laboratory of Pathology, National Institutes of Health, Bethesda; 3 Statistics, University of Pittsburgh, Pittsburgh; 4Gastroenterology, University of California at San Francisco, San Francisco; 5Liver Section, National Institutes of Health, Bethesda, United States; 6Centre Liver Disease, University Health Network, Toronto, Canada; 7 Gastroenterology, Massachusetts General Hospital, Boston; 8 Gastroenterology, UT Southwestern, Dallas; 9Gastroenterology, Washington University School of Medicine, St. Louis, United States E-mail: [email protected] Background and Aims: The HBV/HIV Cohort is a prospective study of the Hepatitis B Research Network (HBRN) working to define the clinical, virologic, serologic and histologic characteristics of a cohort of HBV/HIV patients in North America on combination antiretroviral therapy (cART). We report the histologic spectrum at entry into the cohort. Methods: Adult HIV/HBsAg (≥6 months) participants from 8 HBRN clinical sites were included. Patients with decompensated cirrhosis, HCC and HCV-RNA were excluded. Liver biopsy (36 months of entry) was assessed by a central pathology group for inflammation, fibrosis, steatosis (Ishak and HALT-C scores) and HBcAg and HBsAg staining and pattern of distribution. Demographic data, diabetes mellitus (DM), hypertension (HTN), lipodystrophy (LD; stage 0–3), cART, HIV stage (1–4), alcoholism (AUDIT > 7), liver enzymes, HIV-RNA, CD4, HBV serology, HBV-DNA and anti-HDV were collected. Those with and without significant fibrosis (Ishak 3–6) and steatosis (>5%) were compared. Results: 124 patients on cART were enrolled: age 48.9, 93% male, 40% Caucasian, 47% Black, DM 10%, HTN 34%, alcoholism 8% and BMI 26.5. CD4 567 (38–1395), 25% HIV-RNA detectable, 12% LD, HIV stages 1–4 in 68%, 17%, 6.5%, 7.5% respectively. Mean ALT 35 (range 8–153), HBeAg positive 55%, and detectable HBV-DNA 43%. Six % were “Incomplete HBV suppressors” (HBV-DNA with negative HIV-RNA).
Baseline histology was available in 68 participants: significant periportal and portal fibrosis (score ≥2) 15% and 22% respectively; bridging fibrosis/cirrhosis 28%; lobular inflammation (score ≥2) in 34%; steatosis 26%, steatohepatitis 9%. 62% and 75% had positive staining for HBcAg and HBsAg respectively. The % of HBcAg and HBsAg distribution in hepatocytes (none, <10, 10–50, 51–90, and >90) was 38, 41, 12, 4 and 4 for HBcAg and 25, 56, 13, 3 and 3 for HBsAg. The % of HBcAg distribution was nuclear 21, cytoplasmic 12 and mixed 33. The % HBsAg distribution was inclusion-like 62, granular-cytoplasmic 96, membranous 22. No associations were found between demographic, clinical or laboratory parameters with or without significant fibrosis. Steatosis was associated with DM (P = 0.07), increased ALT (P < 0.001) and BMI (P = 0.06). HBcAg staining was associated with serum HBeAg (OR 17; P < 0.0001) and undetectable HBV-DNA (OR 0.14; P = 0.003). HBsAg staining was only associated with serum HBeAg (OR 6, P = 0.01) Conclusions: Marked inflammation and advanced fibrosis are common despite good HBV and HIV suppression among HBV/HIV co-infected patients on cART. THU-151 Validation of the PAGE-B model in Asian chronic hepatitis B patients receiving entecavir or tenofovir M.N. Kim1, S.G. Hwang1, K.S. Rim1, B.K. Kim2, J.Y. Park2, D.Y. Kim2, S.H. Ahn2, K.-H. Han2, S.U. Kim2. 1Department of Internal Medicine, CHA Bundang Medical Center, CHA University School of Medicine, Seongnam-si; 2Department of Internal Medicine, Yonsei University College of Medicine, Seoul, South Korea E-mail: [email protected] Background and Aims: A new hepatocellular carcinoma (HCC) risk prediction model, PAGE-B, which includes age, gender, and platelet count as constituent variables, has recently been proposed in Caucasian chronic hepatitis B (CHB) patients. We validated the PAGE-B model and compared its accuracy with that of conventional risk prediction models in Asian CHB patients. Methods: A new hepatocellular carcinoma (HCC) risk prediction model, PAGE-B, which includes age, gender, and platelet count as constituent variables, has recently been proposed in Caucasian chronic hepatitis B (CHB) patients. We validated the PAGE-B model and compared its accuracy with that of conventional risk prediction models in Asian CHB patients. Results: A total of 1,092 CHB patients (668 men, 61.2%) were selected between August 2006 and January 2015. The mean age was 48 years. During the follow-up period (median, 43.6 months), 36 (3.3%) patients developed HCC. Older age [hazard ratio (HR) = 1.069], male gender (HR = 3.054), cirrhosis (HR = 4.306), and a lower platelet count (HR = 0.991) were independent predictors of HCC development. PAGE-B showed similar area under receiver operating characteristic curves (AUROCs) to GAG-HCC and CU-HCC at 3 years (0.777 vs. 0.793 and 0.743, respectively; all P > 0.05) and 5 years (0.799 vs. 0.803 and 0.744, respectively; all P > 0.05), whereas the AUROCs of PAGE-B were significantly higher than those of REACH-B (0.602 at 3 years and 0.572 at 5 years, P < 0.05). When cirrhosis was incorporated into PAGE-B (modified PAGE-B), its predictive performance became significantly higher than that of PAGE-B at 5 years (P < 0.05). Conclusions: Our study demonstrated that PAGE-B is applicable to Asian CHB patients and has similar accuracy to conventional risk prediction models. Further studies are warranted to establish strategies for HCC surveillance using PAGE-B in CHB patients. THU-152 Outcome in chronic hepatitis delta: differences between African and non-African patients M. Spaan1, I. Carey1, B. Wang1, D. Shang1, M. Horner1, M. Bruce1, G. Dusheiko1, K. Agarwal1. 1Institute of Liver Studies, King’s College Hospital, London, United Kingdom E-mail: [email protected]
Journal of Hepatology 2017 vol. 66 | S95–S332
S255
Methods: We analyzed 123 antiHBe+ non treated patients (HBVDNA < 20000 IU/ml and ALT < 2ULN at inclusion and during followup) with 2 valid TE separated with a 3 year interval. 53% female, age 44 ± 11 years, most infected by genotypes A/D. At inclusion, 28% had HBV-DNA < 2000 IU/ml and HBsAg < 1000 IU/ml; during the followup, 49% remained with HBV-DNA < 2000 IU/ml and 80% had persistently normal ALT. 23% had steatosis and 4% diabetes. The influence on TE value of demographic, clinical and virological variables was analyzed. Results: Mean basal TE was 5.6 ± 1.6 kPa and at 3 years 5.8 ± 1,8 kPa ( p = 0.24). TE value increased in 63(1.3 ± 1.1 kPa), did not change in 6 and decreased in 54(1.1 ± 0.9 kPa). In 28% the increase was ≥1 kPa. Comparing patients with decrease in TE (n = 54) and those without change or increase (n = 69), there were no differences in any of the analyzed variables. Comparing patients with an increase ≥1 kPa (n = 35) with those with no change or decrease (n = 60), the former had higher AST levels (27.8 ± 8,9 vs.22.7 ± 6,9; p = 0.054), without differences in the rest of the variables. TE value was ≤6,5 kPa in 73% in the initial study and in 76% in the second one. Comparing patients with TE ≤ 6.5 and >6.5 in the second TE, the former had higher basal levels of HBsAg (3.4 ± 1.0 vs 2.9 ± 1.0 log10 IU/ml, p = 0.03), higher levels of basal HBV-DNA (2795 ± 5371 vs 836 ± 1862 IU/ml; p = 0.05) and of highest HBV-DNA during the follow-up (8669 ± 13467 vs 3295 ± 5902; p = 0.03), while those with TE > 6.5 had higher AST levels (29.9 ± 9.1 vs. 23.8 ± 9.1; p = 0.027) and BMI (27.9 ± 5.4 vs. 25.5 ± 5.3; p = 0.038). Diabetes was more frequent among those with TE > 6.5 kpa (10% vs. 2.1%;p = 0.09), as was steatosis (36% vs. 20%; p = 0.08). There were no differences in sex, age, basal ALT, highest ALT, platelets or proportion of patients with persistently normal ALTor with HBV-DNA < 2000 IU/ml. Conclusions: TE values in patients with CHB antiHBe+ patients no candidates to antiviral treatment have little variations over a three years period. Parameters linked to HBV activity seem to have no influence on TE increase. Nevertheless, parameters as AST level and BMI are related with higher TE values, suggesting that metabolic rather than virological factors induce fibrosis progression in these patients. THU-149 Clinical and virological outcomes of hepatitis B virus inactive carriers and gray zone patients: a comparative analysis using current EASL and AASLD Guidelines criteria M.S. Bonacci1, S. Lens1, Z. Mariño1, M.C. Lodoño1, S.P. del Pulgar1, J. Costa1, J.M. Sanchez-Tapias1, X. Forns1. 1Liver Unit, Hospital Clinic Barcelona, Barcelona, Spain E-mail: [email protected]
Conclusions: Serum full length HBV RNA levels in HBeAg positive CHB are independently associated with genotype and presence of HBV basal core promoter mutation. In addition, correlations of HBV RNA with other serum markers are genotype dependent. The effect of these factors should be taken into consideration when studying the role of HBV RNA in the HBV life cycle and as a biomarker for treatment response in HBeAg positive CHB. THU-148 Transient elastography evolution in non treated patients with chronic hepatitis B (CHB) infection M.L. González-Diéguez1, M.A. de Jorge1, C. Alvarez-Navascues1, C. Rodríguez-Escaja1, E. Rubio1, A. Castaño1, V. Cadahia1, M. Varela1, M. Rodriguez1. 1Hepatology Unit, Hospital Universitario Central de Asturias, Oviedo, Spain E-mail: [email protected] Background and Aims: TE has shown usefulness in predicting histological fibrosis in patients with CHB; a TE value <6,5 kPa has been proposed to exclude significant fibrosis (Papatheodoridis GV, et al. JVH 2014). To know TE evolution in patients with antiHBe + CHB not candidates for treatment S254
Background and Aims: Distinction between the inactive HBV carrier (IC) state and progressive chronic hepatitis B may be difficult due to fluctuations of ALT and HBV-DNA levels which lead to the so-called “gray zone” [(GZ)HBV-DNA <2.000 IU/mL and minimally elevated ALT] with uncertain prognosis. The scenario is further complicated by different definitions of IC state by the EASL and AASLD current Guidelines. We aimed to assess the natural history of these patients considering both definitions of IC. Methods: We analyzed the outcome of a series of chronic HBV patients fulfilling the AASLD or EASL criteria for IC and GZ visited at our Liver Unit since 1985. Previously treated or patients with cirrhosis or other causes of liver disease were excluded Results: 197 patients were included. Considering the highest ALT value observed during the first year of follow-up (FU), patients were separated into four groups (Table). Importantly, 82 (60%) ICs according to EASL would have been classified as GZs by AASLD. There were no differences between groups concerning demographic characteristics. After a median FU of 95 (52–217) months, outcomes did not differ across groups. Overall, the 10-year probability of at least one HBV-DNA peak >2.000 UI/mL was 20% and >20.000 IU/mL in 5%. At the end of FU HBV-DNA was 2.000–20.000 IU/mL [median ALT 27 (23–36 IU)] in around 10% but only two subjects progressed to CHB
Journal of Hepatology 2017 vol. 66 | S95–S332
POSTER PRESENTATIONS and needed therapy (Table). No patient developed hepatocellular carcinoma nor significant liver fibrosis [liver stiffness 4.9 kPa (4–5.7)]. The rate of HBsAg loss at 10 years was 20% and not different between groups. The number of HBV-DNA peaks correlated inversely with the probability of HBsAg loss (HR = 0.25 [0.1–0.6] p = 0.03). Classification at baseline IC EASL
IC AASLD
ALT < 40 IU ALT < 19 IU (F) Classification at the end of follow-up, n (%) IC EASL GZ EASL IC AASLD GZ AASLD HBsAg loss Chronic hepatitis
(n = 137) 98 (71,5) 16 (11.5) – – 23 (17) –
ALT < 30 IU (M) (n = 55) – – 30 (55) 19 (34.5) 6 (11) –
GZ EASL ALT (40 to 80 IU)
(n = 60)
GZ AASLD ALT 19 to 38 IU (F) ALT 30 to 60 IU (M) (n = 142)
38 (63) 8 (14) – – 12 (20) 2 (3)
– – 59 (41.6) 52 (36.6) 29 (20.4) 2 (1.4)
Conclusions: Independently of the criteria used for diagnosis, and despite relatively frequent peaks of HBV-DNA during FU, ICs and GZ patients have an excellent long-term prognosis and a relatively high rate of HBsAg loss. The use of a restrictive IC definition (AASLD guidelines) does not seem to add a benefit regarding prognosis and may increase the need for liver biopsies as well as more follow-up visits. THU-150 Liver histologic spectrum in hepatitis B virus patients co-infected with the human immunodeficiency virus R. Sterling1, D.K. Kleiner2, W.S. Abdus3, M. Khalili4, M. Ghany5, D.K. Wong6, R.T. Chung7, M. Jain8, M. Lisker-Melman9 and Hepatitis B Research Network – HBV/HIV Cohort Study. 1GastroenterologyHepatology-Nutrition, Virginia Commonwealth University, Richmond; 2 Laboratory of Pathology, National Institutes of Health, Bethesda; 3 Statistics, University of Pittsburgh, Pittsburgh; 4Gastroenterology, University of California at San Francisco, San Francisco; 5Liver Section, National Institutes of Health, Bethesda, United States; 6Centre Liver Disease, University Health Network, Toronto, Canada; 7 Gastroenterology, Massachusetts General Hospital, Boston; 8 Gastroenterology, UT Southwestern, Dallas; 9Gastroenterology, Washington University School of Medicine, St. Louis, United States E-mail: [email protected] Background and Aims: The HBV/HIV Cohort is a prospective study of the Hepatitis B Research Network (HBRN) working to define the clinical, virologic, serologic and histologic characteristics of a cohort of HBV/HIV patients in North America on combination antiretroviral therapy (cART). We report the histologic spectrum at entry into the cohort. Methods: Adult HIV/HBsAg (≥6 months) participants from 8 HBRN clinical sites were included. Patients with decompensated cirrhosis, HCC and HCV-RNA were excluded. Liver biopsy (36 months of entry) was assessed by a central pathology group for inflammation, fibrosis, steatosis (Ishak and HALT-C scores) and HBcAg and HBsAg staining and pattern of distribution. Demographic data, diabetes mellitus (DM), hypertension (HTN), lipodystrophy (LD; stage 0–3), cART, HIV stage (1–4), alcoholism (AUDIT > 7), liver enzymes, HIV-RNA, CD4, HBV serology, HBV-DNA and anti-HDV were collected. Those with and without significant fibrosis (Ishak 3–6) and steatosis (>5%) were compared. Results: 124 patients on cART were enrolled: age 48.9, 93% male, 40% Caucasian, 47% Black, DM 10%, HTN 34%, alcoholism 8% and BMI 26.5. CD4 567 (38–1395), 25% HIV-RNA detectable, 12% LD, HIV stages 1–4 in 68%, 17%, 6.5%, 7.5% respectively. Mean ALT 35 (range 8–153), HBeAg positive 55%, and detectable HBV-DNA 43%. Six % were “Incomplete HBV suppressors” (HBV-DNA with negative HIV-RNA).
Baseline histology was available in 68 participants: significant periportal and portal fibrosis (score ≥2) 15% and 22% respectively; bridging fibrosis/cirrhosis 28%; lobular inflammation (score ≥2) in 34%; steatosis 26%, steatohepatitis 9%. 62% and 75% had positive staining for HBcAg and HBsAg respectively. The % of HBcAg and HBsAg distribution in hepatocytes (none, <10, 10–50, 51–90, and >90) was 38, 41, 12, 4 and 4 for HBcAg and 25, 56, 13, 3 and 3 for HBsAg. The % of HBcAg distribution was nuclear 21, cytoplasmic 12 and mixed 33. The % HBsAg distribution was inclusion-like 62, granular-cytoplasmic 96, membranous 22. No associations were found between demographic, clinical or laboratory parameters with or without significant fibrosis. Steatosis was associated with DM (P = 0.07), increased ALT (P < 0.001) and BMI (P = 0.06). HBcAg staining was associated with serum HBeAg (OR 17; P < 0.0001) and undetectable HBV-DNA (OR 0.14; P = 0.003). HBsAg staining was only associated with serum HBeAg (OR 6, P = 0.01) Conclusions: Marked inflammation and advanced fibrosis are common despite good HBV and HIV suppression among HBV/HIV co-infected patients on cART. THU-151 Validation of the PAGE-B model in Asian chronic hepatitis B patients receiving entecavir or tenofovir M.N. Kim1, S.G. Hwang1, K.S. Rim1, B.K. Kim2, J.Y. Park2, D.Y. Kim2, S.H. Ahn2, K.-H. Han2, S.U. Kim2. 1Department of Internal Medicine, CHA Bundang Medical Center, CHA University School of Medicine, Seongnam-si; 2Department of Internal Medicine, Yonsei University College of Medicine, Seoul, South Korea E-mail: [email protected] Background and Aims: A new hepatocellular carcinoma (HCC) risk prediction model, PAGE-B, which includes age, gender, and platelet count as constituent variables, has recently been proposed in Caucasian chronic hepatitis B (CHB) patients. We validated the PAGE-B model and compared its accuracy with that of conventional risk prediction models in Asian CHB patients. Methods: A new hepatocellular carcinoma (HCC) risk prediction model, PAGE-B, which includes age, gender, and platelet count as constituent variables, has recently been proposed in Caucasian chronic hepatitis B (CHB) patients. We validated the PAGE-B model and compared its accuracy with that of conventional risk prediction models in Asian CHB patients. Results: A total of 1,092 CHB patients (668 men, 61.2%) were selected between August 2006 and January 2015. The mean age was 48 years. During the follow-up period (median, 43.6 months), 36 (3.3%) patients developed HCC. Older age [hazard ratio (HR) = 1.069], male gender (HR = 3.054), cirrhosis (HR = 4.306), and a lower platelet count (HR = 0.991) were independent predictors of HCC development. PAGE-B showed similar area under receiver operating characteristic curves (AUROCs) to GAG-HCC and CU-HCC at 3 years (0.777 vs. 0.793 and 0.743, respectively; all P > 0.05) and 5 years (0.799 vs. 0.803 and 0.744, respectively; all P > 0.05), whereas the AUROCs of PAGE-B were significantly higher than those of REACH-B (0.602 at 3 years and 0.572 at 5 years, P < 0.05). When cirrhosis was incorporated into PAGE-B (modified PAGE-B), its predictive performance became significantly higher than that of PAGE-B at 5 years (P < 0.05). Conclusions: Our study demonstrated that PAGE-B is applicable to Asian CHB patients and has similar accuracy to conventional risk prediction models. Further studies are warranted to establish strategies for HCC surveillance using PAGE-B in CHB patients. THU-152 Outcome in chronic hepatitis delta: differences between African and non-African patients M. Spaan1, I. Carey1, B. Wang1, D. Shang1, M. Horner1, M. Bruce1, G. Dusheiko1, K. Agarwal1. 1Institute of Liver Studies, King’s College Hospital, London, United Kingdom E-mail: [email protected]
Journal of Hepatology 2017 vol. 66 | S95–S332
S255