PLoS One. 2016 Dec 9;11(12):e0166543. doi: 10.1371/journal.pone.0166543. eCollection 2016.
Factors Predicting HBsAg Seroclearance and Alanine Transaminase Elevation in HBeAg-Negative Hepatitis B Virus-Infected Patients with Persistently Normal Liver Function.Chien TL1, Wang JH2, Kee KM2, Chen CH2, Hung CH2, Lu SN2. Author information
1Division of Hepato-Gastroenterology, Department of Internal Medicine, Antai Medical Care Cooperation Antai Tian-Sheng Memorial Hospital, Donggang Township, Pingtung County, Taiwan.
2Division of Hepato-Gastroenterology, Department of Internal Medicine, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Kaohsiung City, Taiwan.
AbstractBACKGROUND: A certain proportion of hepatitis B virus (HBV)-infected patients with persistently normal alanine transaminase (ALT) levels have significant fibrosis. Using liver stiffness measurements (Fibroscan®) and laboratory data, including serum ALT, quantitative HBsAg (qHBsAg), and HBV DNA, we attempted to predict the natural histories of these patients.
METHODS: Non-cirrhotic HBeAg-negative chronic hepatitis B patients with persistently normal ALT were followed up prospectively with the end points of HBsAg seroclearance and ALT elevation above the upper limit of normal. The factors that were predictive of the end points were identified.
RESULTS: A total of 235 patients with an average age of 48.1 +/- 10.7 years were followed up for 7 years. Eight patients (3.4%) lost HBsAg, and 15 patients (6.4%) experienced ALT elevation. The overall cumulative HBsAg seroclearances were 0.4%, 1.3% and 2.3% at years 1, 3 and 5, respectively. Regarding HBsAg seroclearance, the qHBsAg (< 30 IU/ml) cutoff resulted in a hazard ratio (HR) of 19.6 with a 95% confidence interval (CI) of 2.2-166.7 (P = 0.008). The baseline ALT level (odd ratio (OR) 1.075, 95% CI 1.020-1.132, P = 0.006) and a qHBsAg above 1000 IU/ml (3.7, 1.1-12.4, P = 0.032) were associated with ALT elevation. Limited to men, the baseline liver stiffness (1.6, 1.0-2.5, P = 0.031) and a qHBsAg above 1000 IU/ml (10.4, 2.1-52.4, P = 0.004) were factors that were independently associated with ALT elevation.
CONCLUSION: A low qHBsAg level predicted HBsAg clearance. Baseline ALT and a qHBsAg above 1000 IU/ml were independent predictive factors for ALT elevation. Among the men, the independent predictive factors for ALT elevation were qHBsAg and liver stiffness.
HBsAg seroclearance is a landmark in the treatment of chronic HBV infection because hepatitis B is mainly immune-mediated, and the elimination of HBsAg is considered to be an indicator of the effective treatment of HBV infection. A lower HBsAg level may indicate better host immune control of HBV duplication and a decreased level of intrahepatic covalently closed circular DNA [10]. The annual HBsAg seroclearance rate of hepatitis B patients is between 0.5% and 2.26% per year depending on enrollment in treatment [11–13]. In a study conducted in Taiwan regarding spontaneous HBsAg seroconversion, the average follow-up period of hepatitis B patients was 7.4 years, and 18 patients of 3,000 cleared HBsAg with an annual rate of 0.6%. At 1 year after HBeAg seroconversion, low serum levels of HBsAg predicted the elimination of HBsAg in patients with genotype B or C infections [14]. A Korean study demonstrated that the overall annual HBsAg seroclearance rate was 1.8%, but the annual HBsAg seroclearance rate of the older-aged group was higher, i.e., the annual seroclearance rate was 2.69% among patients aged over 60, whereas the annual seroclearance rate was 1.91% among the patients aged between 40 and 59. The cumulative probabilities of HBsAg seroclearance were 1.2%, 3.8% and 8.7% for patients who were followed for 1, 3 and 5 years, respectively [15]. The annual HBsAg seroclearance rate in our study was 0.48%, i.e., 8 patients exhibited HBsAg seroclearance in the 1645 person-years study, and the rate was 0.3%, i.e., 5 patients exhibited HBsAg seroclearance in the 1624 person-years study. The overall cumulative HBsAg seroclearance rates were 0.4%, 1.3% and 2.3% for the patients who were followed up for 1, 3 and 5 years, respectively. In the present study, the HBsAg seroclearance rate was lower than that in the aforementioned Korean study. In the Korean study, the patients were all genotype C, and approximately 20% of them were treated with an antiviral agent or interferon; however, in our study, the patients were all treatment-naïve, and genotype B was predominant.
Given that there is no evidence of liver cirrhosis or HCV/HDV super infection and the subject’s age is <50 years at the time of the HBsAg loss, there is a minimal risk of HCC development [16]. The REVEAL-HBV study demonstrated that HBsAg seroclearance is significantly associated with increased age, a low serum HBV DNA level (100000 vs ≧ 100000 copies/mL) and a high body mass index (30 vs 30 kg/m2) [17]. In the natural course of chronic HBV infection, a lower HBV DNA level has been reported to be important in determining the subsequent elimination of HBsAg, and the HBV DNA levels in the sera of 95.8% of patients have been found to be undetectable before the elimination of HBsAg [18]. The HBsAg level was a better indicator than the HBV DNA level regarding the prediction of HBsAg elimination in patients who were followed for 6 years. Comparison of the patients with HBsAg levels ≧ 1000 IU/ml with those with HBsAg levels that were 100~999 and < 100 IU/ml revealed that the hazard ratios were 4.4 (95% CI, 1.1–17) and 24.3 (8.7–67.5), respectively, which indicated that the HBsAg loss rate of the latter group was higher. [14]. In a longitudinal study of 117 patients performed in Hong Kong, during the natural course of chronic hepatitis B, a HBsAg reduction > 1 log10 IU/ml was associated with greater viral suppression [19]. Our study revealed that the baseline HBV DNA level had nothing to do with HBsAg seroclearance, and only the qHBsAg was associated with HBsAg seroclearance; at the cutoff of qHBsAg 30 IU/ml, the HR was 19.230, and 95% CI was 2.2–166.7. There was no significant relationship between genotype and HBsAg seroclearance in our study. This study suggested the existence of an uncoupled immune effect on HBsAg production and viral duplication. However, a low qHBsAg level might suggest a more complete immune clearance than a low HBV DNA level.
At present, there is little data available for predicting spontaneous ALT elevation above the ULN in asymptomatic chronic HBV-infected patients with persistently normal ALT. In a recent study performed in India with patients with HBeAg-negative chronic hepatitis B who were yet asymptomatic and had normal ALT levels, the factors that were found to be predictive of the occurrence of spontaneous ALT flares were the male gender, the presence of precore mutants and an age equal to or above 30 years old at baseline. ALT flares were defined as elevations in the serum ALT level to above 2 times the ULN in combination with an HBV DNA level ≧105 copies/ml or a 100-fold rise in the HBV DNA from the previous baseline level. The rate of ALT flares was approximately 4.3% per year. The examined patients were predominantly genotypes D and A [20]. In a prospective study performed in Canada, ALT elevation was defined as a change from a normal ALT (ALT ≦ 40 IU/ml) to an elevated ALT (ALT > 40 IU/ml). This study followed 37 HBeAg-negative patients with normal ALT levels at baseline for a median of 3 years (0.67–4 years), and their baseline HBV DNA levels were found to be highly predictive of future ALT elevations above the ULN [21]. In contrast, a study of chronic hepatitis B among Asian American patients found that the ALT levels were normal (genotype B or C), and there were no strong associations of ALT flares with any of the assessed clinical factors, which included an age ≧ 50, gender, ALT at biopsy ≧ 1/2 ULN, fibrosis stage > 1, inflammation grade > 1, HBV DNA ≧ 100000 IU/ml or ≧ 20000 IU/ml, HBeAg positivity, HBV PC mutation, HBV BCP mutation, all combinations of PC and BCP mutations, and genotypes B or C. However, this study of Asian Americans did not test qHBsAg [22]. In our study, the annual rate of spontaneous ALT elevation was 0.9%, and the qHBsAg played a more important role in predicting ALT elevations. According to the medical care guide of the Taiwan National Health Insurance Administration, in the patient group with ALT elevations above the ULN, 7 of 15 patients (46.7%) would receive nucleoside/nucleotide analogues.
Arena et al. demonstrated that there is a strong association between liver stiffness and serum aminotransferase level [23]. In CHB patients with ALT flares, liver stiffness increases and returns to normal levels after 6 months [24]. A recent meta-analysis demonstrated that the FibroTest has good accuracy in terms of the identification of HBV-associated significant fibrosis and cirrhosis [25]. Regarding the prediction of ALT elevations above the ULN, the patients were further divided into male and female groups because different etiologies may be present in the different genders. However, there were only 3 females, so it was not possible to statistically analyze the females, and we only analyzed the male patients in the subgroup analysis. Baseline liver stiffness was correlated with future ALT elevation in our cohort. Across all patients, liver stiffness was only significant in the univariate analysis. However, in the subgroup analysis of the male patients, liver stiffness was significant in both univariate and multivariate analyses. We also analyzed the fibrosis-4 score (FIB-4) and AST to platelet ratio index (APRI). However, there was no significant relationship between these scores and ALT elevation.
qHBsAg provides not only a very useful means to identify the HBV-related cause of liver damage in HBsAg carriers but also it is the most important predictor of clinically relevant outcomes associated with HBV-induced liver disease. In a prospective study performed in France, reactivation of hepatitis among HBeAg-negative asymptomatic patients could be predicted by a combination of HBsAg > 1000 IU/ml and HBV-DNA > 200 IU/ml (92% sensitivity and 96% negative predictive value) [26]. Tseng TC et al. showed that HBsAg, ALT and age could be used as predictors for hepatocellular carcinoma in HBeAg negative patients with HBV DNA < 2000 IU/ml. This study revealed higher risk of HCC in patients with HBsAg ≥1000 IU/ml compared with those <1000 IU/ml (adjusted hazard ratio 13.7, 95% confidence interval: 4.8–39.3) [27]. Another prospective study from Korea also demonstrated that combined HBsAg and HBV DNA (> 850 IU/ml and 850 IU/ml, respectively) could predict the reactivation of HBV with 84.6% diagnostic accuracy [28]. One recent study showed that combined HBV DNA with qHBsAg or liver stiffness measurement could correctly identify inactive carrier at a single time point [29]. Our study emphasizes the influence of single baseline ALT, qHBsAg and liver stiffness measurement on natural history of asymptomatic hepatitis B infected patients. However, our study has some limitations. This study conducted only on a small sample size of population with relatively short duration of follow-up. We used the same level of ALT normal cutoff value (40 IU/ml) for both genders (2016 American Association for the Study of Liver Disease Guideline: normal ALT levels are <30 IU/ml for males and <10 IU/ml for female) because we use the same normal ALT cutoff value in real-world clinical practice in Taiwan.
In conclusion, our study demonstrated that a single baseline qHBsAg is an important factor for future ALT elevations above the ULN and qHBsAg seroclearance. Very few studies have used baseline liver stiffness to predict future ALT elevations. Currently, no simple and readily available markers are able to accurately predict biochemical elevations in HBV-infected patients. Using the qHBsAg, liver stiffness and HBV DNA, we have demonstrated a new perspective for predicting the natural course of HBeAg-negative patients with persistently normal liver function.作者: StephenW 时间: 2016-12-11 07:40
Arena et al。表明肝硬度与血清氨基转移酶水平之间存在强相关性[23]。在CHB患者ALT发烧,肝硬度增加,并在6个月后恢复正常水平[24]。最近的荟萃分析表明,FibroTest在鉴定HBV相关的显着肝纤维化和肝硬化方面具有良好的准确性[25]。关于ULN上方的ALT升高的预测,将患者进一步分为男性和女性组,因为在不同性别中可能存在不同的病因。然而,只有3个女性,因此不可能对女性进行统计分析,我们只在亚组分析中分析了男性患者。基线肝硬度与我们队列中未来ALT升高相关。在所有患者中,肝硬度仅在单变量分析中显着。然而,在男性患者的亚组分析中,肝硬度在单变量和多变量分析中都是显着的。我们还分析了纤维化-4评分(FIB-4)和AST与血小板比率指数(APRI)。然而,这些分数与ALT升高之间没有显着的相关性。
qHBsAg不仅提供了一种非常有用的方法来鉴定HBsAg携带者中HBV相关的肝损伤原因,而且它是与HBV诱导的肝病相关的临床相关结果的最重要的预测因子。在法国进行的前瞻性研究中,通过HBsAg> 1000 IU / ml和HBV-DNA> 200 IU / ml(92%敏感性和96%阴性预测值)的组合可以预测HBeAg阴性无症状患者中肝炎的再激活, [26]。 Tseng TC et al。 HBsAg,ALT和年龄可用作HBeAg阴性患者HBV DNA <2000 IU / ml的肝细胞癌的预测因子。这项研究显示HBsAg≥1000IU / ml患者的HCC风险高于1000 IU / ml(校正的风险比为13.7,95%置信区间:4.8-39.3)[27]。来自韩国的另一项前瞻性研究也表明,HBsAg和HBV DNA的合并(> 850 IU / ml和850 IU / ml)可以预测HBV的再激活率为84.6%[28]。最近的一项研究表明,联合HBV DNA与qHBsAg或肝硬度测量可以在单个时间点正确识别不活动载体[29]。我们的研究强调单基线ALT,qHBsAg和肝硬度测量对无症状乙型肝炎感染患者的自然史的影响。然而,我们的研究有一些限制。本研究仅对样本量较小的人群进行,其随访时间相对较短。我们对两种性别使用相同水平的ALT正常截断值(40IU / ml)(2016美国肝病研究协会指南:对于男性,正常ALT水平<30 IU / ml,对于女性,正常ALT水平<10 IU / ),因为我们在台湾现实临床实践中使用相同的正常ALT临界值。