Search

This study investigated the ability of hepatitis B core-related antigen (HBcrAg) to predict hepatitis B virus (HBV) relapse in HBeAg-negative patients after cessation of entecavir therapy. A total of 301 HBeAg-negative patients without cirrhosis who had stopped entecavir therapy for at least 12 months were recruited. All patients fulfilled the stopping criteria proposed by the APASL 2012 guidelines. The five-year cumulative rates of virological relapse, clinical relapse and HBsAg loss were 71.6%, 57.3% and 18.7%, respectively. Serum HBsAg at end of treatment (EOT) was an independent predictor of virological relapse, clinical relapse and HBsAg loss; an EOT HBsAg of 150 IU/ml was the optimal cut-off value. The 5-year virological relapse rates for patients with <150 and ≥150 IU/ml HBsAg at EOT were 43.3% and 82.2% (p < 0.001), clinical relapse rates were 32.3% and 66.3% (p < 0.001), and HBsAg loss rates were 46.1% and 5.2% (p < 0.001), respectively. A baseline HBcrAg of 4 IU/ml was the optimal cut-off value for predicting HBV relapse. Among patients with an EOT HBsAg <150 IU/ml, the five-year virological relapse rates for patients with baseline HBcrAg levels ≤4 and >4 log U/ml were 27.9% and 59.1% (p = 0.006) and the clinical relapse rates were 18% and 48.1% (p = 0.014), respectively. EOT HBcrAg was not a significant predictor of virological or clinical relapse after cessation of entecavir. In conclusion, the combination of an EOT HBsAg of 150 IU/ml and baseline HBcrAg of 4 log U/ml can effectively predict the risk of HBV relapse after stopping entecavir therapy., (© 2021 John Wiley & Sons Ltd.)

Data on direct-acting antiviral agent (DAA) treatment for mixed genotype hepatitis C virus (HCV) infection are scant. This study examined the effectiveness of glecaprevir/pibrentasvir (GLE/PIB) and ledipasvir/sofosbuvir (LDV/SOF) for mixed HCV genotype infection in a real-world setting in Taiwan. We analysed the data from all patients with mixed HCV genotype infections treated with GLE/PIB or LDV/SOF from 2017 to 2019 in three Chang Gung Memorial Hospitals in Taiwan. The primary treatment outcome was sustained virologic response 12 weeks after treatment cessation (SVR12). Adverse events (AEs) were also evaluated. A total of 5190 HCV patients received DAA treatment during this time period. Among them, 116 patients (2.2%) had mixed infections of any 2 or 3 genotypes of 1a, 1b, 2, 3 and 6. Fifty-four patients received GLE/PIB and 62 received LDV/SOF. SVR12 rates for LDV/SOF vs GLE/PIB therapy were 96.6% (56/58) vs 100% (51/51) by the per-protocol analysis and 90.3% (56/62) vs 94.4% (51/54) by the evaluable population analysis. Two patients with 1b + 6 and 1b + 2 genotype infections in the LDV/SOF group had relapse. Evaluating the GLE/PIB vs LDV/SOF groups for the most common AEs revealed pruritus (16.7% vs 4.8%), abdominal discomfort (5.6% vs 8%) and fatigue (5.6% vs 4.8%). One patient with AE-related treatment discontinuation presented with liver decompensation after 4-week GLE/PIB therapy. DAA-related significant laboratory abnormalities occurred in two patients with >3× elevated bilirubin level in the GLE/PIB group. GLE/PIB and LDV/SOF are well tolerated and achieve high SVR12 rates for patients with mixed HCV genotype infection., (© 2020 John Wiley & Sons Ltd.)

Elbasvir/grazoprevir with or without ribavirin has excellent efficacy and safety for the treatment of hepatitis C virus (HCV) genotype 1 and 4 patients. The real-world experience has been reported but the detailed analysis of liver and renal adverse effects is lacking. This study evaluated the real-world experience relating to the effectiveness and liver/renal safety of elbasvir/grazoprevir in HCV genotype 1 patients with compensated liver disease. In the four medical centres of Chang Gung Medical System, 350 HCV genotype 1 patients with compensated liver disease who were treated with elbasvir/grazoprevir were enrolled. Clinical characteristics and laboratory data were collected. The effectiveness (sustained virologic response 12 weeks after end of treatment, SVR12) and safety were assessed. A consecutive series of 350 patients with a mean age of 68.8 ± 10.0 years old were enrolled. The majority were treatment-naïve (72.3%), genotype 1b (97.7%) and advanced fibrosis/cirrhosis (94.3%). Seventy-nine (22.6%) had hepatocellular carcinoma and 23 (6.6%) had coinfection with hepatitis B. The effectiveness of SVR12 was 94.6% (95% CI: 92.2%-97.0%) in the full analysis set and 99.1% (95% CI: 98.1%-100.1%) in the per-protocol set. There were two relapses and one nonresponder. Seven (2.0%) patients had adverse events resulting in premature discontinuation of treatment. Five of them were considered drug-related. One was due to autoimmune hepatitis. Contrary to previous reports, around 49% of ALT elevation was observed after 8 weeks, and in two patients was due to hepatitis B flares. As to the renal function during the course of therapy, a minor deterioration of eGFR was observed in patients with baseline eGFR ≥60 mL/min/1.73 m 2 , but not in those with baseline eGFR <60, <60-30 or <30 mL/min/1.73 m 2 . In this real-world data, elbasvir/grazoprevir was effective with few liver/renal adverse effects. One patient developed autoimmune hepatitis., (© 2020 John Wiley & Sons Ltd.)

Renal toxicity of direct-acting antivirals (DAAs) in chronic hepatitis C (CHC) patients has not been well-characterized. The aim of this study was to assess renal safety of DAAs in an Asian CHC patient cohort. Data from CHC patients (n = 1536) treated with DAAs were used in this retrospective study. Serial estimated glomerular filtration rate (eGFR) at pretreatment (1-year prior to treatment), baseline, end of treatment (EOT), and 12 weeks after treatment (SVR 12 ) was evaluated. While a significant decrease in eGFR from baseline to EOT (84.8 → 81.8 mL/min/1.73 m 2 , P < .001) was observed; subsequently, a slight rise at SVR 12 (84.3 mL/min/1.73 m 2 ) was also evident. Changes in eGFR after DAA treatment were similar to those seen in PrOD, DCV/ASV and GZP/EBV regimens, except in the SOF-based regimen wherein eGFR remained unchanged from EOT to SVR 12 , especially in liver transplant recipients. Multivariate analysis revealed that age >65 years (OR = 1.862, P = .011), baseline eGFR ≥ 60 mL/min/1.73 m 2 (OR = 2.684, P = .023), and liver transplant (OR = 3.894, P = .001) were independent risk factors for deteriorating renal function. In conclusion, DAA treatment led to a significant decline in eGFR at EOT but was followed by a slight rise at 12 weeks after treatment. A similar trend was observed with PrOD, DCV/ASV and GZP/EBV, but not in SOF-based regimens. As age >65 years, baseline eGFR ≥ 60 mL/min/1.73 m 2 and liver transplantation are significant risk factors for deterioration in renal function, we strongly advice close monitoring of renal function in these populations., (© 2019 John Wiley & Sons Ltd.)

Background and Aim: Noninvasive fibrosis indices can predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). Modified FIB-4 (mFIB-4) is a promising noninvasive index for predicting liver fibrosis. To investigate the predictive accuracy of several extant noninvasive fibrosis indices, including mFIB-4, for HCC incidence in CHB patients receiving long-term entecavir therapy., Methods: We enrolled 1325 nucleos(t)ide analogue-naïve CHB patients (noncirrhotic 844; cirrhotic 481) treated with entecavir. Baseline clinical features and fibrosis indices were collected and evaluated for predicting HCC risk through univariate and multivariate Cox regression analyses., Results: Of the 1325 patients, 105 (7.9%) developed HCC during a median follow-up period of 4.1 years. Age (hazard ratio [HR], 1.039; 95% confidence interval [CI], 1.020-1.059; P < 0.0001), diabetes mellitus (DM) (HR, 1.902; 95% CI, 1.185-3.052; P = 0.0077), and mFIB-4 (HR, 4.619; 95% CI, 1.810-11.789; P = 0.0014) were independent predictors of HCC in all patients (mFIB-4 ≥ 1.5 for the noncirrhotic cohort; DM and mFIB-4 ≥ 2.0 for the cirrhotic cohort). A combination of mFIB-4 and the DM status stratified the cumulative risk of HCC into three subgroups in all patients (high: mFIB-4 ≥ 1.5/DM; intermediate: mFIB-4 ≥ 1.5/non-DM; and low: mFIB-4 < 1.5, P < 0.0001) and in the cirrhotic cohort (high: mFIB-4 ≥ 2.0/DM; intermediate: mFIB-4 ≥ 2.0/non-DM; and low: mFIB-4 < 2.0, P = 0.0007). An mFIB-4 cutoff value of 1.5 stratified the cumulative risk of HCC in the noncirrhotic cohort (P = 0.015)., Conclusions: The mFIB-4 index alone or in combination with DM is the optimal noninvasive predictor of HCC risk in CHB patients receiving entecavir therapy., (© 2018 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Background and Aim: This study investigated whether hepatitis B surface antigen (HBsAg) could predict hepatitis B virus (HBV) relapse after cessation of entecavir or tenofovir disoproxil fumarate (TDF) prophylaxis for chronic hepatitis B cancer patients who are undergoing chemotherapy., Methods: The study enrolled 122 hepatitis B e-antigen-negative cancer patients who underwent chemotherapy with entecavir or TDF for antiviral prophylaxis and posttreatment follow-up for at least 6 months., Results: Of the 122 patients, 52 and 18 experienced virological and clinical relapse, which had 3-year cumulative incidences of 46.6% and 18.6%, respectively. Multivariate analysis showed that end-of-treatment HBsAg levels and baseline HBV-DNA ≥ 2000 IU/mL were independent predictors of virological relapse. The best HBsAg cutoff value was 500 IU/mL. An end-of-treatment HBsAg of 500 IU/mL was useful for predicting virological relapse in patients with baseline HBV-DNA < 2000 IU/mL (3-year rate: 21.3% vs 46.4%, P = 0.038, in patients with HBsAg < 500 and ≥ 500 IU/mL, respectively), but not in patients with baseline HBV-DNA ≥ 2000 IU/mL. Of the 52 patients who experienced virological relapse, 13 experienced transient virological relapse. Patients with baseline HBV-DNA level < 2000 IU/mL experienced a higher rate of transient virological relapse (42.1% vs 15.2%, P = 0.031). Three patients experienced hepatic decompensation upon alanine aminotransferase flares, and no patient died after timely retreatment. Ten patients experienced posttreatment HBsAg loss, and the 3-year HBsAg loss rate was 30.7% in patients with end-of-treatment HBsAg < 100 IU/mL., Conclusions: The baseline HBV-DNA and end-of-treatment HBsAg levels could predict virological relapse after withdrawal of entecavir and TDF prophylaxis for chemotherapy., (© 2018 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Background and Aim: Commonly used non-invasive fibrosis scores usually included serum transaminase levels in the equations, including Aspartate transaminase to Platelet Ratio Index (APRI) and fibrosis-4 (FIB-4). Transaminases fluctuated significantly in chronic hepatitis B patients with exacerbations, leading to unsteady score values. As such, here, we aim to develop a transaminase-free score suitable for pretherapeutic evaluation of fibrosis stages., Methods: Firstly, 1082 treatment-naïve chronic hepatitis B patients were enrolled and divided into modeling (n = 541) and verification (n = 541) cohorts. Secondly, 265 patients having received liver biopsy, with known Ishak fibrosis stages, were included for independent correlation., Results: Cross-sectional analysis of 1082 patients revealed age-dependent variation of association between virological factors and cirrhosis. A fibrosis score including Anti-hepatitis B e antibody, Basal core promoter (BCP) A1762T mutation, and Platelet count Index (named ABPI) was derived from the modeling cohort. ABPI performed better than APRI and FIB-4 in the verification cohort for identifying cirrhotic patients (comparison of area under the receiver operating characteristic curves: ABPI vs APRI and FIB-4 = 0.785 vs 0.563 [P < 0.001] and 0.700 [P = 0.026], respectively). The performance of ABPI was even better in young (< 40 years old) hepatitis B patients (area under the receiver operating characteristic curves: 0.856 vs 0.402 [P < 0.001] and 0.599 [P = 0.009], respectively). Finally, in the independent cohort of 265 patients with known Ishak fibrosis stages, it was found that ABPI effectively distinguished between Ishak fibrosis stages 3 and > 3 and between 4 and > 4 (P < 0.001 for each)., Conclusions: We developed a transaminase-free fibrosis score (ABPI) utilizing basal core promoter A1762T data, which outperformed APRI and FIB-4., (© 2018 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

This study investigates the incidences and predictors of hepatitis B virus (HBV) relapse after tenofovir disoproxil fumarate (TDF) therapy in hepatitis B e antigen (HBeAg)-positive and -negative patients. We retrospectively recruited 143 chronic hepatitis B (CHB) patients without cirrhosis (39 HBeAg-positive and 104 HBeAg-negative patients) who were previously treated with TDF and had post-treatment follow-up for at least 6 months (median: 55, IQR 36-85 weeks). All the patients fulfilled the stopping criteria of APASL 2012. The virological and clinical relapse rates at 104 weeks in HBeAg-positive patients were 66.6% and 59.1%, while they were 72.3% and 55.9%, respectively, in HBeAg-negative patients. Cox regression analysis revealed that the higher end-of-treatment HBsAg levels were an independent factor of virological relapse in HBeAg-positive and HBeAg-negative patients. The end-of-treatment HBsAg levels of 200 (area under the receiver operating characteristic (AUROC): 0.624) and 80 IU/mL (AUROC: 0.959) were the optimal values for predicting HBV relapse in HBeAg-positive and HBeAg-negative patients, respectively. The virological relapse rate at 78 weeks was 14.3% and 19.6% in HBeAg-positive and HBeAg-negative patients who achieved HBsAg ≤200 IU/mL and HBsAg ≤80 IU/mL, respectively. Two patients experienced hepatic decompensation upon hepatitis flares, and no patient died after timely retreatment. Seven patients experienced off-therapy HBsAg loss. The cumulative rates of HBsAg loss at 104 weeks were 45.5% and 59.3% in patients with end-of-treatment HBsAg ≤80 IU/mL and ≤50 IU/mL, respectively. In conclusions, the end-of-treatment HBsAg levels were a useful marker for predicting HBV relapse in HBeAg-positive and HBeAg-negative CHB patients., (© 2017 John Wiley & Sons Ltd.)

Background and Aim: This multinational (Taiwan, South Korea, Russia) phase 3 study evaluated the all-oral, ribavirin-free, fixed-dose combination (DCV-TRIO) of daclatasvir (NS5A inhibitor) 30 mg, asunaprevir (NS3 inhibitor) 200 mg, and beclabuvir (NS5B inhibitor) 75 mg, in patients with chronic hepatitis C virus genotype-1 infection, with or without compensated cirrhosis., Methods: UNITY-4 (NCT02170727) was an open-label, two-cohort study in which 169 patients, treatment-naive (n = 138) or treatment-experienced (n = 31), received twice-daily DCV-TRIO for 12 weeks with 24 weeks of post-treatment follow-up. The primary efficacy end point was sustained virologic response at post-treatment week 12 (SVR12) in treatment-naive patients., Results: Eighty-eight (52%) patients were men, 81 (48%) Taiwanese, 78 (46%) Korean, and 10 (6%) Russian; 23 (14%) had compensated cirrhosis, and 52 (31%) were IL28B (rs1297860) non-CC genotype. Baseline resistance-associated NS5A polymorphisms (L31 and/or Y93) were detected in 25/165 (15%) patients with available genotype-1 sequencing data. SVR12 was achieved by 98.6% (136/138; 95% confidence interval: 94.9-99.8%) of treatment-naive and 100% (31/31; 95% confidence interval: 88.8-100%) of treatment-experienced patients. Both virologic failures were found to be infected with hepatitis C virus genotype-6g; 100% SVR12 was observed for genotype-1a (n = 8) and genotype-1b (n = 157). Two patients experienced serious adverse events. Eight (5%) patients experienced reversible grade 3/4 alanine aminotransferase or aspartate aminotransferase elevations, leading to discontinuation in four (2%); all achieved SVR12. There were no grade 3/4 total bilirubin increases and no deaths., Conclusions: Twelve weeks of DCV-TRIO was well tolerated and provided 100% SVR12 in treatment-naive and treatment-experienced patients with genotype-1 infection, with or without cirrhosis, including those with baseline NS5A-L31 or NS5A-Y93 resistance-associated substitutions., (© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Background and Aim: The Albumin-Bilirubin (ALBI) grade is a new index to assess objectively liver function and prognosis in patients with hepatocellular carcinoma (HCC). This study aimed to elucidate the application of ALBI grade in baseline and sorafenib-end in advanced HCC patients who received sorafenib., Methods: A total of 415 consecutive advanced HCC patients in Child-Pugh A received sorafenib in our hospital. Sorafenib was terminated when radiologic tumor progression or clinical liver function deterioration (LD) occurred in the reassessment bimonthly. Patients who failed with sorafenib monotherapy were retrospectively analyzed., Results: A total of 260 (62.6%) patients were enrolled, including 98 (37.7%) ALBI grade I and 162 (62.3%) grade II in baseline. More patients in ALBI grade II stopped sorafenib because of LD than in grade I (33.3% vs 14.3%, P < 0.001). Those who in baseline ALBI grade I had a superior overall survival than in grade II (8.5 months vs 4.4 months, P = 0.003). Cox regression analysis confirmed that baseline ALBI grade II (P < 0.001) and ALBI grade increase during treatment (P < 0.001) strongly contributed to the mortality of HCC patients who received sorafenib. After sorafenib failure, those with post-sorafenib treatment had a better post-sorafenib survival than those without (9.3 vs 1.6 months, P < 0.001). Logistic regression analysis indicated that sorafenib-end ALBI grade and LD occurrence were the only two predictors of post-sorafenib treatment after sorafenib failure., Conclusions: In clinical practice, we firstly demonstrated that not only ALBI grade in baseline but also ALBI grade change during treatment could predict the prognosis of advanced HCC patients who received sorafenib., (© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

We investigated the incidence and predictors of post-treatment hepatitis B virus (HBV) relapse and hepatitis B surface antigen (HBsAg) loss. After cessation of nucleoside analogue (NA) treatment in hepatitis B e antigen (HBeAg)-negative patients with cirrhosis. The rates of HBsAg loss and hepatocellular carcinoma (HCC) development in HBeAg-negative patients with cirrhosis who continued NA treatment were compared with those who discontinued treatment. Patients with compensated cirrhosis who had discontinued NA treatment for at least 12 months (discontinuing group; n=73) and patients who continued entecavir treatment for at least 4 years (continuing group; n=158) were recruited. Serum HBsAg levels were analysed at the end of treatment (discontinuing group) or at 2.5-3 years of treatment (continuing group). In the discontinuing group, the 6-year cumulative incidence of post-treatment virological relapse and HBsAg loss were 56.3% and 46.7%, respectively. The end-of-treatment HBsAg level of 300 IU/mL was a cut-off value for subsequent post-treatment HBsAg loss and sustained response. In the continuing group, HBsAg loss occurred in five of 158 patients. Cox regression analysis showed that HBsAg levels in the discontinuing group were independent predictors for HBsAg loss in all patients and 104 propensity score (PS)-matched patients. There was no significant difference in HCC development between the groups in all patients and 104 PS-matched patients. Two patients experienced post-treatment alanine aminotransferase flare with hepatic decompensation, and neither of them died after retreatment. In conclusion, HBeAg-negative patients with cirrhosis who discontinued NA treatment might have a higher rate of HBsAg loss and their risk of developing HCC did not increase compared with those who continued entecavir treatment., (© 2017 John Wiley & Sons Ltd.)

Background and Aim: Pegylated-interferon-alpha plus ribavirin is the current standard-of-care regimen for treating chronic hepatitis C virus (HCV) infection in Taiwan; however, interferon-based regimens can be poorly tolerated. The interferon-free, two-drug, fixed-dose combination tablet ledipasvir/sofosbuvir is approved in Europe, the USA, and Japan for treating chronic genotype 1 HCV infection. Little is known about its efficacy/safety in Taiwanese patients., Methods: In this multicenter, open-label, phase 3b (NCT02021656) study, 85 Taiwanese patients (n = 42, treatment-naïve; n = 43, treatment-experienced) with chronic genotype 1 HCV infection (±compensated cirrhosis) received 12 weeks of ledipasvir/sofosbuvir fixed-dose combination tablet. The primary efficacy end point was the proportion of patients with sustained virologic response 12 weeks after treatment discontinuation (SVR12). Safety and pharmacokinetic data were collected., Results: The overall SVR12 rate was 98% (83/85), with 100% (42/42) and 95% (41/43) of treatment-naïve and treatment-experienced patients, respectively, achieving SVR12. There were no on-treatment virologic failures. One patient relapsed after treatment discontinuation; one patient withdrew consent on day 2. The most common treatment-emergent adverse event (AE) was headache (14%, 12/85). There was one grade 3 AE (small cell lung cancer unrelated to ledipasvir/sofosbuvir), no grade 4 AEs, and four grade 3-4 laboratory abnormalities. Only the patient with small cell lung cancer prematurely discontinued treatment. Two patients reported three serious AEs; none was considered related to ledipasvir/sofosbuvir., Conclusions: Data from this phase 3b study suggest that 12 weeks of once-daily treatment with the interferon-free, ribavirin-free regimen ledipasvir/sofosbuvir is effective and well-tolerated in Taiwanese patients with chronic genotype 1 HCV infection, irrespective of treatment history., (© 2016 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Background and Aims: This study compared the efficacy and safety of tenofovir disoproxil fumarate (TDF) up to 3 years of innucleos(t)ide analog (NA)-naïve and NA-experienced chronic hepatitis B (CHB) patients., Methods: Tenofovir disoproxil fumarate-treated NA-naïve and NA-experienced CHB patients were retrospectively analyzed., Results: After 3 years of TDF therapy, 97.7%, 71%, and 45.5% NA-naïve patients achieved a virological response, alanine aminotransferase normalization, and hepatitis B e antigen seroconversion, respectively. Compared with NA-naïve patients, NA-experienced patients without drug resistance and infected with lamivudine/telbivudine-resistant mutants showed similar results. In contrast, patients previously infected with adefovir-resistant mutants and with a suboptimal entecavir response showed significantly lower rates of virological response and hepatitis B e antigen loss/seroconverion than NA-naïve patients. Mean estimated glomerular filtration rate markedly reduced within 12 months of TDF therapy; however, it did not decrease significantly during 12-36 months of treatment. Diabetes mellitus was an independent predictor of a ≥ 0.5 mg/dL increase above baseline in serum creatinine level, and age, hypertension, diabetes mellitus, and baseline creatinine level were independent factors for > 20% decline in estimated glomerular filtration rate from baseline. Liver stiffness measurements improved significantly, but bone mineral density did not change significantly during treatment. Hepatocellular carcinoma incidence was low at 36 months. Age of > 60 years, cirrhosis, a low baseline platelet count and a high α-fetoprotein level at 12 months were significant predictors of hepatocellular carcinoma development., Conclusions: Tenofovir disoproxil fumarate is effective and safe for NA-naïve and NA-experienced CHB patients and should be used cautiously in patients with comorbidities because of a renal dysfunction risk., (© 2016 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Background and Aims: Previous studies have indicated that lamivudine-induced hepatitis B e antigen (HBeAg) seroconversion may not be durable in the Asian population. We investigated the useful predictors of post-treatment hepatitis B virus (HBV) relapse in patients with nucleos(t)ide analogue (NA)-induced HBeAg loss/seroconversion., Methods: A total of 157 non-cirrhotic patients with NA-induced HBeAg loss/seroconversion (78, lamivudine; 68, entecavir; 11, telbivudine) were retrospectively analyzed. All patients had at least 12 months of post-treatment follow-up and consolidation therapy duration., Results: The cumulative rate of post-treatment HBV relapse at 5 years was 57.1%. Multivariate analysis revealed that age and baseline hepatitis B surface antigen (HBsAg) levels independently predicted post-treatment HBV relapse. The post-treatment HBV relapse rate was significantly higher in patients aged > 40 years than in those < 40 years (P < 0.001). A baseline HBsAg level of 2000 IU/mL was the optimal cut-off value for predicting post-treatment HBV relapse (P = 0.002). The post-treatment HBV relapse risk further increased with the presence of both risk factors (age ≥ 40 years and baseline HBsAg level ≥ 2000 IU/mL; P < 0.001). A prolonged consolidation therapy period of ≥ 18 or 24 months had no positive effect on sustained viral suppression. There was no significant difference in post-treatment HBV relapse rates between patients with lamivudine- and entecavir-induced HBeAg loss/seroconversion during the off-treatment follow-up (P = 0.31)., Conclusion: The combination of an age of 40 years and a baseline HBsAg level of 2000 IU/mL was a useful marker for predicting post-treatment HBV relapse in patients with NA-induced HBeAg loss/seroconversion., (© 2014 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd.)

Background and Aim: Hepatocellular carcinoma (HCC) diagnosis could be made with one typical imaging study in a cirrhotic liver by the guideline of the American Association for the Study of Liver Diseases (AASLD) in 2010. Patients with hepatitis B who may not have fully developed cirrhosis could be applied. We aim to retrospectively analyze and validate the diagnostic power of the 2010 guideline in an HCC endemic area (Taiwan)., Methods: From January 2006 to December 2010, a total of 648 patients with liver tumor post-surgical resection were reviewed. The fibrotic scores were verified by METAVIR score 4. Among the 648 patients, 569 (87.8%) were HCC patients. Hepatitis B accounts for 54.5%, hepatitis C 21.9%, hepatitis B + C 2.8%, and non-hepatitis B or C 20.7% of patients. Two hundred eighty-eight of 648 (44%) patients were with cirrhotic liver., Results: The diagnostic sensitivity, specificity, positive predictive value (PPV), negative predictive value, and accuracy of the 2010 AASLD guideline f are 99.1%, 36.7%, 91.9%, 85.3%, and 91.5%, respectively. Cirrhotic liver exhibited a higher PPV (P < 0.001) but lower specificity (P = 0.0479) than non-cirrhotic liver. In both cirrhotic and non-cirrhotic condition, no difference existed in patients with hepatitis B or hepatitis C (P > 0.05)., Conclusions: Similar sensitivity of HCC diagnosis existed between cirrhotic and non-cirrhotic liver, and across different fibrotic stages. But cirrhotic liver exhibited a higher PPV. Hepatitis B or C has no decisive effect in HCC diagnosis., (© 2014 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd.)

We investigated the antiviral effect of entecavir in nucleos(t)ide analogue (NA)-naïve and NA-experienced chronic hepatitis B patients without virological response (VR, HBV DNA < 300 copies/mL) at week 24 or 48. A total of 369 NA-naïve and 181 NA-experienced patients treated with entecavir monotherapy were analysed. Of the 369 NA-naïve patients, 34 did not achieve VR at week 48. Of them, patients with HBV DNA ≤ 2000 copies/mL at week 48 achieved a higher VR rate than those with HBV DNA >2000 copies/mL (18/23 vs 3/11, P = 0.004). Two naïve patients with HBV DNA >2000 copies/mL developed entecavir- or lamivudine-resistant mutants. In 98 lamivudine-experienced patients without ever having lamivudine resistance, most patients with VR (72/72) and partial VR (300-10(4) copies/mL; 20/23) at week 24 or VR at week 48 (89/91) could maintain or achieve VR after prolonged therapy. In 75 patients with prior resistance to lamivudine, prolonged entecavir therapy led to low VR rate in those without VR at week 24 (13/45) or 48 (4/34) and high entecavir-resistance rate in those with or without VR at week 24 (6/30 with and 23/45 without) and 48 (8/41 with and 21/34 without). VR at week 48 was an independent predictor (HR 0.14, 95% CI 0.06-0.33) for entecavir-resistant mutant development among the 75 patients with prior lamivudine-resistant mutants. In conclusion, prolonged entecavir treatment resulted in a poor response in naïve patients with HBV DNA >2000 copies/mL at week 48 and patients with prior lamivudine-resistant mutants without VR at week 24 or 48., (© 2014 John Wiley & Sons Ltd.)

Background and Aim: There is limited data on the efficacy and outcome of telbivudine (LdT) therapy in patients with chronic hepatitis B and compensated cirrhosis. We evaluated LdT as first-line therapy in these patients and compared with those treated with entecavir (ETV)., Methods: We consecutively enrolled 88 chronic hepatitis B patients with compensated cirrhosis primarily treated with LdT at least for 2 years or less than 2 years but developed resistance, and evaluated the efficacy and clinical outcomes. Meanwhile, we matched a control group who treated with ETV for comparison., Results: In LdT group, alanine aminotransferase normalization (65.8%), hepatitis B e antigen seroconversion (39.8%), hepatitis B virus (HBV) DNA undetectablility (71.6%), and virologic resistance (23.9%) were noted after 2 years treatment. Compared with ETV group, there were significant difference in HBV DNA undetectablility (P < 0.001) and virologic resistance (P < 0.001). In addition, the decline of serum hepatitis B surface antigen levels, hepatocellular carcinoma development, mortality, disease progression, and the change of renal function were similar. Cox regression analysis showed that pretreatment low albumin level and high model for end-stage liver disease scores were risk factors for disease progression., Conclusions: These results indicated that although LdT and ETV are similar in clinical outcomes for patients with HBV-related compensated cirrhosis, LdT still had lower HBV undetectablility and higher resistant rate after 2 years treatment, which was a challenge for being as first-line therapy in these patients who need lifelong therapy.

Background and Aim: Patients with gastroesophageal reflux disease (GERD) are often advised to avoid large meals, based on their complaints of increased symptoms after eating too much, and epidemiological evidence of a link between high volume intake and the presence of GERD. However, the precise effects of meal volume on gastroesophageal reflux have not been well studied. We aimed to clarify the effect of meal volume on acid regurgitation and symptoms in patients with GERD., Methods: Fifteen patients (10 female, 5 male; mean 54 ± 10 years old) with GERD were studied twice each in random order, during 24 h ambulatory pH monitoring. On one day, they consumed a 600 mL liquid test meal three times (breakfast, lunch, and dinner), and on the other, they consumed a 300 mL test meal six times (breakfast, snack, lunch, snack, dinner, and snack). Gastric fundus and antral areas and antral contractions were measured by transabdominal ultrasound. Symptoms were recorded using questionnaires., Results: During the 600 mL regimen, there were more reflux episodes (17 ± 4 vs 10 ± 2, P = 0.03) and a greater total acid reflux time (12.5 ± 5.9% vs 5.5 ± 3.6%; P = 0.045) than the 300 mL regimen. Both the cross-sectional area of the gastric fundus (P = 0.024) and the number of antral contractions (P = 0.014) were greater for the 600 mL regimen., Conclusions: Larger meals are associated with distension of the gastric fundus and an increase in gastroesophageal reflux when compared with smaller, more frequent meals.

Background and Aims: Oral nucleos(t)ide analogs (NAs) are effective in suppressing hepatitis B virus (HBV) replication in treatment naïve chronic hepatitis B (CHB) patients. However, little is known about the treatment modification and adherence on such patients with prolonged NA treatment., Methods: In this multicenter observational study, a total of 600 NA-naïve Taiwanese CHB patients aged 16 years and older were enrolled. The 600 patients were retrospectively identified by their NA treatment history from August 2008 to July 2009; this cohort was prospectively followed up over 3 years. During the 3-year period, incidence of treatment modifications, reasons for modification, and rate of adherence were evaluated., Results: Among the 583 evaluable patients, the initial NA treatment included entecavir (ETV) in 468 patients, telbivudine (LdT) in 67, and lamivudine (LVD) in 48. During the 3-year treatment, 9.0% of ETV-treated patients, 38.8% of LdT-treated patients, and 54.2% of LVD-treated patients had treatment modification. The main reasons for treatment modification were fulfilling stopping criteria in the ETV group (40.5%) and virological breakthrough in both the LdT (61.5%) and LVD (46.2%) groups. The proportion of patients with adherence rate (> 90%) at year 3 was 90.8% in the ETV group, 83.9% in the LdT group, and 83.9% in the LVD group., Conclusions: Treatment-naïve CHB patients with a 3-year ETV treatment in Taiwan have the lower likelihood of treatment modification and better rate of adherence compared with those with LdT or LVD treatment., (© 2013 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd.)

Recent studies have indicated that amino acid (aa) substitutions in the core region and NS5A interferon sensitivity-determining region (ISDR) of hepatitis C virus (HCV) as well as genetic polymorphisms in the interleukin-28B (IL-28B) locus affect the outcome of interferon (IFN)-based therapies. We aimed to investigate the role of these factors on response to peginterferon plus ribavirin in a prospective study of response-guided therapy. The aa sequences in core region and ISDR and rs12979860 genotypes were analysed in 115 HCV-1 patients. The treatment was 24 weeks for patients achieving a rapid virological response (RVR), 48 weeks for those with an early virological response (EVR) and early terminated in those without an EVR. A sustained virological response (SVR) was achieved in 82% of 34 RVR patients, 45% of 74 EVR patients and 0% of seven non-EVR patients. Logistic regression analysis showed that ISDR mutation (≥2) [odds ratio(OR): 6.024], double core 70/91 mutations (OR: 0.136), and platelet counts≥15×10(4) /μL (OR: 3.119) were independent pretreatment factors associated with SVR. Apart from rs12979860 CC genotype, low viral load and ISDR mutation (≥2) were significant factors predictive of RVR. Combination of rs12979860 genotype and baseline viral characteristics (viral load and core/ISDR mutations) could predict RVR and SVR with positive predictive value of 100% and 91%, and negative predictive value of 80% and 54%, respectively. In conclusion, pretreatment screening rs12979860 genotype and aa substitutions in the core region and ISDR could help identifying patients who are good candidates for peginterferon plus ribavirin therapy., (© 2013 John Wiley & Sons Ltd.)

Aims: HuR is an RNA-binding protein that post-transcriptionally modulates the expression of various target genes involved in carcinogenesis, such as CCNA2, which encodes cyclin A. The aim of this study was to evaluate the significance of HuR expression and subcellular localization in a large cohort of gastrointestinal stromal tumours (GISTs)., Methods and Results: HuR immunostaining was assessable for nuclear and cytoplasmic expression in 341 cases on tissue microarrays of primary GISTs, of which 318, 296 and 193 cases were also characterized for Ki67 labelling, cyclin A immunoexpression, and KIT and PDGFRA receptor tyrosine kinase (RTK) genotypes, respectively. The results of HuR nuclear and cytoplasmic expression were correlated with disease-free survival (DFS) and clinicopathological, immunohistochemical and RTK genotypic variables. HuR cytoplasmic expression was present in 42% of primary GISTs, and was significantly related to epithelioid histology, larger tumour size, NIH risk category, and nuclear expression of Ki67 and cyclin A. Importantly, HuR cytoplasmic expression (P < 0.001) and cyclin A overexpression (P < 0.001) were strongly associated with worse DFS. Both variables remained independently predictive of adverse outcome [P = 0.020 and risk ratio (RR) 2.605 for cytoplasmic HuR; P = 0.026 and RR 2.763 for cyclin A]., Conclusions: HuR cytoplasmic expression not only correlates with adverse prognosticators and cyclin A overexpression, but also independently predicts worse DFS, indicating a causative role in conferring tumour aggressiveness., (© 2013 John Wiley & Sons Ltd.)

Background and Aim: Transient elastography (TE) has been useful in esophageal varices (EV) diagnosis for chronic hepatitis C patients. In the present study, we evaluate the usefulness of TE and simple blood markers in the EV diagnosis of patients with hepatitis B virus (HBV)-related cirrhosis, prospectively., Methods: Consecutive patients with compensated cirrhosis and positive HBV surface antigen were enrolled, prospectively. At enrollment, the aspartate aminotransferase (AST) to alanine aminotransferase ratio (AAR) and the AST to platelet ratio index (APRI) were recorded, and TE was performed. Two experienced endoscopists assessed EV independently. High-risk EV was defined as small size with a red color sign, and medium or large in size. The diagnostic performances, optimal cut-offs, and the validities of TE, APRI, platelet count (PLT), and AAR in EV diagnosis were assessed., Results: A total of 126 patients (male/female: 93/33; mean age: 54.5 years) with reliable TE results were analyzed. There was good agreement between two endoscopists in assessing the presence of EV and high-risk EV (kappa value: 0.82 and 0.96). Forty-eight (38.1%) patients had EV (small: 35; high risk: 13). There was correlation between TE result and EV size (r = 0.515, P < 0.001). TE, APRI, and PLT were similar; however, superior to AAR in the diagnostic accuracies for EV and high-risk EV. In high-risk EV prediction, the negative predictive value (NPV) was 97%, 98%, and 98%, with cut-offs of 21 kPa, 1.24, and 110 (× 10(9) /L) for TE, APRI, and PLT, respectively., Conclusions: For compensated patients with HBV-related cirrhosis, TE, APRI, and PLT are useful in excluding high-risk EV with high NPV., (© 2012 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd.)

Background and Aims: The American Association for the Study of Liver Disease issued guidelines that proposed that hepatocellular carcinoma (HCC) can be diagnosed if a mass is larger than 2 cm in a cirrhotic liver and shows typical features of HCC at triphasic liver computed tomography (CT) or dynamic magnetic resonance imaging (MRI). In non-cirrhotic livers, the criteria were not applicable. The aim of the present study was to retrospectively analyze the sensitivity of imaging by samples of definite HCC postoperatively and test their application to diagnose HCC in non-cirrhotic livers., Methods: From January 2006 to November 2008, a total of 343 pathologically-diagnosed HCC patients via surgical resection were reviewed. Among the 343 patients, 204 patients had undergone liver CT examination, and 80 patients underwent MRI examination; serum α-fetoprotein had been checked for all 343 patients prior to operation. The diagnostic sensitivity of HCC by imaging was evaluated and compared in patients with/without cirrhosis by ultrasound and histology., Results: The diagnostic sensitivity of HCC by single imaging was approximately 65-80% (liver CT or MRI). A higher sensitivity of HCC diagnosis was found in patients with ultrasound-diagnosed cirrhosis than non-cirrhosis, but the difference in sensitivity disappeared after histologically-cirrhotic validation. The results indicated that regardless of the presence or absence of cirrhosis (histology), a typical vascular pattern could diagnose HCC with equally high sensitivity., Conclusions: We provide evidence that the sensitivity of HCC diagnosis by imaging is not influenced by the cirrhotic background. Further study is needed to validate the specificity and accuracy., (© 2011 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd.)

Background and Aims: This study investigated outcome predictors in hepatitis-B-e-antigen (HBeAg)-positive chronic hepatitis B patients treated with peginterferon alfa-2a., Methods: A total of 88 HBeAg-positive patients receiving peginterferon alfa-2a for 6 months and followed up for at least 24 weeks were prospectively analyzed. Precore and core promoter genes of hepatitis B virus (HBV) were sequenced from the serial serum samples of 88 patients., Results: After 24 weeks of follow up, 38.6% and 28.4% of patients achieved HBeAg clearance and combined response, respectively. Multivariate analysis disclosed that pretreatment HBeAg sample to cut-off (S/Co) ratio ≤ 200, alanine aminotransferase > 200 IU/mL, HBV genotype B and T1846 were independent factors for HBeAg clearance, and HBeAg S/Co ratio ≤ 200 and HBV genotype B were major determinants for combined response. HBeAg S/Co ratio ≤ 10 at week 12 of therapy was the useful factor for treatment response and had a greater power (P = 0.012) to predict HBeAg clearance than HBV DNA. Patients with HBeAg clearance had a higher frequency of A1896 mutation at baseline and during therapy than those without HBeAg clearance, and the frequency of A1896 decreased during treatment. During follow up, delayed HBeAg seroconversion and reactivation of HBV after HBeAg clearance were observed in eight non-responders and 20 patients with HBeAg clearance, respectively. HBV genotype B was a significant factor to predict both responses., Conclusions: Pretreatment HBeAg S/Co ratio ≤ 200 and HBV genotype B were major determinants for treatment response to peginterferon. Genotype-B-infected patients had higher probability of delayed HBeAg clearance and sustained response. Rapid decrease of HBeAg titer was useful on treatment predictor., (© 2011 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd.)

Aim: The aim of the present study was to assess the changes of liver stiffness (LS) and its associated factors in patients with chronic hepatitis C virus infection (HCV) after interferon (IFN)-based therapy., Methods: Patients with chronic HCV who had previously undergone at least 20 weeks of IFN-based therapy were enrolled. The patients' initial LS measurement was taken at the time of enrollment, and a second LS measurement was made after an interval of at least 38 weeks. LS measurement was carried out with FibroScan, and changes of LS and its associated factors were analyzed., Results: One hundred and forty-four patients, including 95 sustained virological response (SVR) patients and 49 non-sustained virological response (NSVR) patients, were enrolled. There was a significant decrease of LS among SVR patients (median, 0.6; P < 0.001). NSVR patients showed an increase of LS (median, 0.8; P = 0.557). For SVR patients, a high initial LS was the predictive factor of a rapid reduction of LS values. However, advanced fibrosis stage before therapy, higher body mass index (BMI) and longer time remission were predictive factors for slow reduction of LS values., Conclusions: LS decreases in sustained responders following IFN-based therapy in patients with chronic HCV. Advanced fibrosis, higher BMI, longer time for remission and lower initial LS value are predictive factors for a slow improvement of LS in sustained responders.

Background and Aim: To assess the significance of adequate alpha-fetoprotein decrease in monitoring the treatment effects of radiofrequency ablation (RFA) in hepatocellular carcinoma (HCC) patients., Methods: A total of 72 RFA treatments in 54 HCC patients were analyzed. The favorable alpha-fetoprotein decrease was defined as the alpha-fetoprotein half-life of less than 7 days. The efficacy of the ablation response is assessed by standard imaging modality, a computed tomography scan 1 month after RFA. We assessed the correlation between different alpha-fetoprotein decreases and treatment outcomes by standard imaging modality., Results: Of the 72 therapies, 15 (21%) were favorable alpha-fetoprotein decreases. Fifty-one (71%) therapies showed concordant results through standard image modality and alpha-fetoprotein decrease, including 14 (27%) therapies with a complete radiological response and favorable alpha-fetoprotein decrease, and the remaining 37 (73%) therapies with an incomplete radiological response and unfavorable alpha-fetoprotein decrease. The accuracy was 70.8% by using alpha-fetoprotein decrease in the detection treatment response based on a complete radiological response. Among the 34 therapies with a complete radiological response, 14 therapies with a favorable alpha-fetoprotein decrease had a better disease-free survival curve than 20 therapies with an unfavorable alpha-fetoprotein decrease (P = 0.003). Only one case had a favorable alpha-fetoprotein decrease, but incomplete radiological response, with massive necrosis, with the exception of a small residual tumor., Conclusions: A favorable alpha-fetoprotein decrease has better predictive power for disease-free survival than for an unfavorable alpha-fetoprotein decrease. HCC patients after RFA with an unfavorable alpha-fetoprotein decrease should be considered to have undergone incomplete treatment, despite the complete response by standard image modality post-RFA.

Background and Aim: Hepatocellular carcinoma (HCC) tends to metastasize to extrahepatic organs. Stomach involvement has been seldom reported and has always been considered as direct invasion. This study aims to propose a possible existing pathway for the hematogenous metastasis of HCC to the stomach., Methods: Only seven cases with stomach involvement were found from 8267 HCC patients registered at our hospital between 2000 and 2007. Their laboratory data, the findings of computed tomography and upper endoscopy, therapeutic procedures, such as esophageal variceal banding ligation (EVL), and transhepatic arterial embolization (TAE) were further studied., Results: All seven patients were male. Liver cirrhosis was found in six patients (6/7 = 85.7%), HCC with portal vein thrombosis (PVT) in six patients (6/7 = 85.7%), splenomegaly in five patients (5/7 = 71.4%) and esophageal varices in five patients (5/7 = 71.4%). Six patients underwent TAE and one patient underwent EVL before the development of HCC in the stomach. Four patients had HCC at the cardia, one patient at the anterior wall of the high body and two patients at the greater curvature of the high body, far away from the original HCC. Six patients eventually developed distant metastasis. HCC with gastric metastasis developed 53-126 days after TAE in five patients and 74 days after EVL in one patient., Conclusions: When cirrhotic patients with portal hypertension have HCC with PVT, a hematogenous pathway can exist for gastric metastasis of tumor thrombi involving hepatofugal flow to the stomach after TAE or EVL apart from the major pathway of direct invasion.

NS5A and E2 proteins of the hepatitis C virus (HCV) have the potential to repress protein kinase R (PKR) that exerts a tumour suppressor function. We investigated the relationship between amino acid variations in the NS5A-PKR-binding domain and E2-PKR-eIF2alpha phosphorylation homology domain (PePHD) region and the development of hepatocellular carcinoma (HCC) in chronic HCV-1b patients. In a cross-sectional, hospital-based setting, we compared the amino acid sequences of NS5A-PKR-binding domain and E2-PePHD in the sera of 104 chronic hepatitis, 44 cirrhosis and 96 HCC patients. The nucleotide sequences were inferred by direct sequencing of the amplified HCV products and deduced amino acid were compared with the sequence of HCV-J. By univariate analysis, old age, lower viral load, fewer amino acid substitutions in the NS5A-PKR-binding domain (codons 2209-2274) and the interferon sensitivity-determining region (ISDR; codons 2209-2248), and wild-type amino acid at codon 2209 and codon 2240 was significantly correlated with HCC, whereas substitutions in the E2-PePHD was not. Patients with a mutated-type (> or = 4) NS5A-ISDR had a lower prevalence of HCC than those with intermediate or wild type (P < 0.05). Based on stepwise logistic regression analysis, age [odds ratio (OR): 1.132, P < 0.001], viral load (OR: 0.305, P < 0.001) and mutated-type ISDR (OR: 0.137, P = 0.001) were independently associated with HCC. In conclusion, NS5A-ISDR variations may play an important role in the development of HCV-related HCC.

Background and Aim: There are limited reports on esophageal motility pressures in aged patients with achalasia and these are inconclusive. The aim of the present retrospective study was to understand the changes of esophageal motility in aged achalasia patients among the Taiwan population., Methods: Manometric studies of 49 patients with achalasia had been performed through January 1998 to June 2005. The findings of lower esophageal sphincter (LES) basal and residual pressures and esophageal body contraction amplitudes were calculated and compared between the older and younger patient groups at different age cut-offs., Results: Higher basal LES pressure increased significantly from the cut-off age of 65 years (i.e. patients over 65 had significantly higher basal LES pressure than younger patients: 37.0 +/- 4.19 mmHg vs 30.0 +/- 1.32 mmHg, P = 0.045). With patients > or =70 years old, it was more obvious (46.0 +/- 3.7 mmHg vs 29.6 +/- 1.2 mmHg, P = 0.001). Beginning at the cut-off age of 55, the LES residual pressure was significantly higher in older patients than those who were younger (14.0 +/- 11.06 mmHg vs 11.1 +/- 0.6 mmHg, P = 0.017). LES residual pressure is more significant in the older groups. A linear correlation between age and residual LES pressures (r = 0.383) was found. No differences were found in esophageal contraction pressure., Conclusions: Older achalasia patients in Taiwan have higher basal LES pressures, with a linear correlation between age and residual LES pressures. Age has no influence on esophageal contraction pressure.

We assessed the efficacy of interferon (IFN) alpha-2b plus ribavirin therapy in patients with hepatitis C virus (HCV)-related cirrhosis, and elucidated the risk factors for the development of hepatocellular carcinoma (HCC) to determine whether these therapies might reduce the incidence of HCC. One hundred and thirty-two HCV-cirrhotic patients receiving IFN alpha-2b (3 or 5 MU thrice weekly) and oral ribavirin (1,000-1,200 mg/day) for 24 or 48 weeks were analysed. Cumulative incidence of HCC was estimated by the Kaplan-Meier method. The prognostic relevance of clinical variables and HCC occurrence was evaluated by univariate analysis with the log-rank test and by multivariate Cox's regression analysis. A total of 116 patients completed the treatment and 73 (55%) achieved a sustained virological response (SVR). Stepwise logistic regression analysis showed that nongenotype 1b (P < 0.001) and low viral load (P = 0.018) were independent variables of SVR. During a median follow-up period of 37 (12-63) months, HCC developed in 11 patients with non-SVR and five with SVR (P = 0.0178), whereas there was no difference between those with transient biochemical response and nonresponse (P = 0.5970). The Kaplan-Meier method also showed that old age (>or=60 years) (P = 0.0034) and genotype 1b (P = 0.0104) were associated with HCC occurrence. Using Cox's regression analysis, non-SVR (odds ratio = 3.521, P = 0.036), male (odds ratio = 6.269, P = 0.011) and old age (odds ratio = 3.076, P = 0.049) were independent significant risk factors contributing to HCC development. Our results suggest that achieving SVR by IFN alpha-2b plus ribavirin therapy may decrease the incidence of HCC in patients with HCV-related cirrhosis.

Nonstructural 5A (NS5A) and the second envelope (E2) proteins of hepatitis C virus (HCV) have the potential to block interferon (IFN)-induced RNA-dependent protein kinase (PKR) and may therefore interfere with the response to IFN therapy, but controversy still exists regarding the relevance of this. This study aimed to assess whether mutations in these regions correlated with the response to combination therapy, IFN and ribavirin. Pretreatment parameters were analysed in 57 HCV-1b patients who had received IFN-alpha2b (3 or 5 MU three times weekly) and ribavirin (800-1200 mg per day) for 24 weeks. The amino acid sequences of the NS5A and PKR-eIF2alpha phosphorylation homology domain (E2-PePHD) were deduced from the corresponding coding sequence, which were determinated by direct sequencing of the HCV genome amplified by the polymerase chain reaction. Twenty (36%) patients achieved a sustained virological response (SVR). The mean number of amino acid substitutions in the NS5A-PKR binding domain (2209-2274), interferon sensitivity-determining region (ISDR) (2209-2248), and E2-PePHD sequence (659-670) in patients with and without SVR were 4.53 +/- 3.31 vs 2.83 +/- 1.78 (P = 0.094), 2.45 +/- 2.74 vs 1.03 +/- 1.32 (P = 0.042) and 0.25 +/- 0.70 vs 0.03 +/- 0.17 (P = 0.109), respectively. Patients with a mutant-type (>/= 4) NS5A-ISDR had a higher rate of SVR (six of nine, 67%) than those with wild-type (five of 22, 23%) (P = 0.038). Stepwise multiple logistic regression analysis of the factors (age, gender, viral load, cirrhosis rate, IFN dosage and amino acid substitutions) revealed that the mutation in NS5A-ISDR (>/= 4 vs < 4) was the only independent variable of treatment outcome. Our study showed that NS5A-ISDR mutations were correlated with the SVR to combination therapy in chronic HCV-1b patients in Taiwan.