血清干扰素诱导蛋白10水平预测乙肝S抗原血清学清除慢性乙

StephenW

Serum interferon-inducible protein 10 levels predict hepatitis B s antigen seroclearance in patients with chronic hepatitis B

    G. L.-H. Wong1,2,3, H. L.-Y. Chan1,2,3, H.-Y. Chan1,2,3, C.-H. Tse1,2,3, A. M.-L. Chim1,2, A. O.-S. Lo1,2 andV. W.-S. Wong1,2,3,*

Article first published online: 3 NOV 2015

DOI: 10.1111/apt.13447

© 2015 John Wiley & Sons Ltd

Issue
Alimentary Pharmacology & Therapeutics
Alimentary Pharmacology & Therapeutics

Volume 43, Issue 1, pages 145–153, January 2016


Summary
Background

Hepatitis B s antigen (HBsAg) seroclearance is regarded as the optimal virological end-point.
Aim

To investigate the dynamic changes in serum cytokine levels around the time of HBsAg seroclearance.
Methods

This was a case–control study. Consecutive patients with chronic hepatitis B (CHB) who lost HBsAg were matched with those remained positive for HBsAg with same age, gender, HBeAg status and presence of cirrhosis in 1:2 ratio. Relevant serum cytokines [interleukin (IL)-2, IL-3, IL-4, IL-7, IL-9, IL-10, IL-12, IL-15, IL-21 interferon-γ, tumour necrosis factor-α (TNF-α), granulocyte macrophage colony-stimulating factor (GM-CSF) and interferon-inducible protein 10 (IP-10)] were assayed at the time (Year 0) and 3 years before (Year −3) HBsAg seroclearance.
Results

Seventy-one and 142 CHB patients who did and did not achieve HBsAg seroclearance were included. Mean age was 48 ± 11 years; 76% were male, 20% had positive HBeAg, 99 (46%) patients received anti-viral therapy, and mean baseline HBV DNA was 3.78 ± 2.28 log IU/mL vs. 4.36 ± 2.13 log IU/mL respectively (P = 0.05). In those who achieved HBsAg seroclearance, serum IL-15 and GM-CSF levels decreased significantly from Year −3 to Year 0 (P = 0.017 and 0.05 respectively). When compared to controls, only serum IP-10 level was significantly lower at Year 0 than at Year −3 in patients with HBsAg seroclearance. Lower serum IP-10 level at Year 0 was the only factor associated with HBsAg seroclearance. There was no correlation between serum IP-10 and HBsAg levels around the time of HBsAg seroclearance.
Conclusion

Lower serum IP-10 level at Year 0 was the only factor associated with HBsAg seroclearance.


Discussion

This was the one of the largest series of CHB patients who developed HBsAg seroclearance and had their serum dynamic cytokine profiles studied in detail. Despite an extensive profiling of their serum cytokines, we could only identify one of 13 potential cytokines associated with HBsAg seroclearance, the important therapeutic end-point. The initial drop followed subsequent rise of serum IP-10 level around the time of HBsAg seroclearance implied host immune profile was altered in a complex manner. On the other hand, we could not demonstrate any correlation between serum IP-10 and HBsAg levels around the time of HBsAg seroclearance.

Despite the potential role of a few cytokines, namely IL-10, IL-12, IL-21 and IP-10, on immune control on HBV based on some surrogate virological markers (HBeAg seroconversion and drop in HBsAg level) illustrated in previous reports,[16-18] only serum IP-10 level was found to be modestly associated with HBsAg seroclearance. We postulated that different types of host immune response take place at different phases of CHB. While IL-10, IL-12, IL-21 have important role at the immune clearance phase and hence is associated with HBeAg seroconversion; they may not be as important in the process of HBsAg seroclearance.

A recent large-scale Taiwanese study based on the REVEAL-HBV cohort demonstrated that serum IL-9 level was associated with HBsAg seroclearance, particularly in patients with undetectable HBV DNA.[32] The data from this study have only been published in abstract form. In contrast, the serum IL-9 level was not different between case and control and at different time points from HBsAg seroclearance in our study.

Serum IP-10 level was previously found to be significantly associated with patient age (r = 0.23, P = 0.001) and correlated strongly with baseline ALT (r = 0.45, P < 0.001) in a cohort of HBeAg-positive CHB patients who received peginterferon therapy.[18] An interesting note was the relatively low median baseline level of IP-10 level (158 pg/mL), compared to 477–702 pg/mL at different time points in our cohort. This might infer that the IP-10 level was initially not as high at immune clearance phase and at younger age, then it would be increased with age, subsequently further increased until a certain period of time, probably a couple of years, prior HBsAg seroclearance.

The interesting V-shape change in serum IP-10 levels around the time of HBsAg seroclearance might be partly supported by the recent observation from a small Chinese cohort CHB patients receiving telbivudine.[33] The investigators observed that serum IP-10 levels decreased during telbivudine treatment. Even more interestingly, the decline of IP-10 was higher among virological responders.[33] Whether this was related to a feedback mechanism as there was much reduced in the viral load among virological responders, such that the host immune response as reflected by serum IP-10 level might have already been dampened.

We failed to demonstrate any correlation between serum IP-10 and HBsAg levels around the time of HBsAg seroclearance, which was different to the observation from a recent study showing serum IP-10 level predicts HBsAg decline in patient receiving entecavir.[15] However, patients in this European study had rather high HBsAg levels (median 3.51 log10 IU/mL) to start with. This implied these patients were still far from HBsAg seroclearance. The interval of 3 years before HBsAg seroclearance might have been too long to demonstrate any true correlation, as profound immune response happened closer to the time of HBsAg seroclearance. So whether the role of IP-10 was different at various stages prior HBsAg seroclearance remained to be defined.

IP-10, also named as chemokine ligand 10 (CXCL10) is a chemotactic CXC chemokine that is secreted by hepatocytes and hepatic sinusoidal endothelium in patients with hepatitis.[34] Serum IP-10 level at commencement of therapy was found to correlate with viral load and ALT level in CHB patients.[35] But our cohort was different, as our patients would have passed the immune clearance phase as in the previous study. Therefore, the HBV DNA was relatively low and ALT level was mostly normal in our cohort.

Serum IP-10 level has an important role not just in CHB patients, but also in those with chronic hepatitis C (CHC). The interesting bit was the trend of serum IP-10 level and virological response run an opposite direction in these two types of chronic viral hepatitis. High baseline serum IP-10 level predicted worse outcomes in CHC patients, e.g. non-responders to therapy; whereas low baseline serum IP-10 level predicted early and sustained virological responses.[36] But these observations do not necessarily mean host immune response acts in an inverse manner in CHB and CHC. In fact, serum IP-10 levels declined more significantly among CHC patients who would have sustained virological response. Our cohort also demonstrated a significant drop of serum IP-10 levels in CHB patients impending HBsAg seroclearance. Recent study showed specific immune response fails to control viral replication, the infected hepatocytes would secrete IFN-gamma-induced chemokines (e.g. CXCL9, CXCL10 and CXCL11) leading to migration of nonspecific mononuclear cells into the liver.[37] These mononuclear cells are in fact unable to control infection but result in sustained necroinflammation and hence liver damage.[38] On the other hand, inhibition of these chemokines limits nonspecific cell migration and hence reduces the inflammation.[39]

Our study has the strength of a respectable number of subjects with HBsAg seroclearance, very stringent matching criteria of control subjects, which facilitated the detection of even relatively subtle difference in changes of cytokines. We also checked a comprehensive panel of cytokines. Nonetheless, our study also has a few limitations. The incomplete serum collection in some time points precluded the analysis of serial serum IP-10 and HBsAg levels in all case and control subjects. Inclusion of both treated and untreated subjects might introduce confounding effect from anti-viral treatment-induced immune response. Patients who had HBsAg seroclearance before our systematic collection of serum sample were not included in this analysis, which have led to a smaller sample size. Although cases and controls had been matched in several important clinical parameters (age, gender, HBeAg status and presence of cirrhosis), there might still be unmeasured confounders.

In conclusion, serum IP-10 level was associated with HBsAg seroclearance, the optimal virological end-point in CHB. The initial drop followed subsequent rise of serum IP-10 level around the time of HBsAg seroclearance implied host immune profile was altered in a complex instead of a simple, linear manner around that time.

StephenW

血清干扰素诱导蛋白10水平预测乙肝S抗原血清学清除慢性乙型肝炎

    G. L.-H. Wong1,2,3，H L.-Y. Chan1,2,3，H.-Y. Chan1,2,3，C.-H. Tse1,2,3，A M.-L. Chim1,2，A O.-S. Lo1,2和V。 W.-S. Wong1,2,3，*

文章首次在网上公布：2015年11月3号

DOI：10.1111 / apt.13447

©2015年约翰·威利父子有限公司

问题
消化系统药理学和治疗
消化系统药理学和治疗

第43卷第1期，页145-153，2016年一月


概要
背景

肝炎BS抗原（HBsAg）血清清除被视为最佳病毒学终点。
目的

为了调查各地的HBsAg血清清除的时间血清细胞因子水平的动态变化。
方法

这是一项病例对照研究。连续治疗慢性乙型肝炎（CHB）谁失去了乙肝表面抗原进行匹配与保持HBsAg阳性与1相同的年龄，性别，HBeAg状态和存在肝硬化：2的比例。有关血清细胞因子[白介素（IL）-2，IL-3，IL-4，IL-7，IL-9，IL-10，IL-12，IL-15，IL-21干扰素γ，肿瘤坏死因子α（TNF-α），粒细胞巨噬细胞集落刺激因子（GM-CSF）和干扰素诱导蛋白10（IP-10）]进行测定的时间（年0）和前3年（公元-3）的HBsAg血清清除。
结果

七十一和142例慢性乙型肝炎患者谁做的，并没有达到乙肝表面抗原血清学清除都包括在内。平均年龄为48±11岁; 76％为男性，20％的人HBeAg阳性，99（46％）患者接受抗病毒治疗，平均基线HBV DNA为3.78±2.28日志IU / mL相对4.36±2.13分别登录国际单位/毫升（P = 0.05 ）。在那些谁取得了乙肝表面抗原血清学清除，血清中IL-15和GM-CSF水平显著的年份（分别为P = 0.017和0.05）下降-3至0年。当与对照组相比，只有血清IP-10水平显著降低在年0比年度-3患者的HBsAg血清清除。在0年降低血清IP-10水平与乙肝表面抗原血清学清除相关联的唯一因素。有各地的HBsAg血清清除的时间血清IP-10和HBsAg水平无相关性。
结论

在0年降低血清IP-10水平与乙肝表面抗原血清学清除相关联的唯一因素。


讨论

这是最大的系列CHB患者谁开发的HBsAg血清清除，并有他们的血清中的动态细胞因子水平进行了详细研究中的一个。尽管他们的血清细胞因子的一个广泛的分析，我们只能确定用HBsAg血清清除，重要的治疗终点相关联的13潜在的细胞因子之一。最初的下降，随后围绕乙肝表面抗原血清学清除暗示宿主免疫配置文件被修改以复杂的方式时血清IP-10级以后的上升。另一方面，我们不能证明周围的HBsAg血清清除的时间的血清的IP-10和HBsAg水平之间的任何相关性。

尽管有一些细胞因子，也就是IL-10，IL-12，IL-21和IP-10的基础上，以前的报告说明了一些替代病毒学指标（HBeAg血清转换和下降的HBsAg水平）的潜在作用，对免疫控制乙肝病毒[16-18]只有血清IP-10级被认为是谦虚与乙肝表面抗原血清学清除有关。我们假定，不同类型的宿主免疫反应发生在慢性乙型肝炎的不同阶段。而IL-10，IL-12，IL-21具有在免疫清除期的重要作用，因此与血清转换相关联;它们可能不会在HBsAg的血清清除的过程中同样重要。

基于所述REVEAL - HBV队列最近大规模台湾研究表明，血清中IL-9水平用HBsAg血清清除相关联的，特别是在患者的检测不到HBV DNA中。[32]本研究的数据只被发表在抽象的形式。与此相反，血清中IL-9水平并非病例和对照之间，并在从HBsAg的血清清除在我们的研究中不同的时间点不同。

血清IP-10级以前发现与患者年龄相关（r = 0.23，P = 0.001），与基线ALT强相关（r = 0.45，P <0.001），谁收到HBeAg阳性慢性乙型肝炎患者的队列来显著相关聚乙二醇干扰素疗法。[18]一种有趣的说明是IP-10级（158皮克/毫升）的相对低的值基线水平相比，477-702皮克/毫升的在我们的队列不同的时间点。这可能会推断，IP-10级，最初并不高，在免疫清除期和低龄化，那么这将是随着年龄的增加，随后进一步增加，直到一段时间，大概一两年之前，乙肝表面抗原血清学清除。

血清IP-10左右的HBsAg血清清除的时间级别有趣的V形变化可能是由最近的观察从一个小的中国人群乙肝患者接受替比夫定可部分支持。[33]研究人员发现，在血清IP-10水平下降替比夫定治疗。甚至更有趣的是，为IP-10的下降病毒学应答中更高。[33]这是否是有关的反馈机制，因为在病毒负荷大大降低病毒学应答，使得宿主免疫反应中所反映的血清IP-10级可能已经受到打击。

我们未能证明预测的HBsAg下降病人接受恩替卡韦周围的HBsAg血清清除的时间，这是不同的观察从最近的研究中显示出血清的IP-10水平的血清的IP-10和HBsAg水平之间的任何相关性。[15]然而，患者在这个欧洲研究中有相当高的HBsAg水平（中位数3.51日志10 IU / mL）的开始。这意味着这些患者还远远没有乙肝表面抗原血清学清除。 3年的HBsAg血清清除之前的时间间隔可能是太长展示任何真正的相关性，如深刻免疫应答发生接近的HBsAg血清清除的时间。因此，无论IP-10的作用是在不同阶段不同的HBsAg前转阴仍有待确定。

IP-10，也称为趋化因子配体10（CXCL10）是趋化CXC趋化因子是受肝细胞和肝窦内皮肝炎患者分泌的。在开始的治疗[34]血清的IP-10水平发现与病毒慢性乙型肝炎患者负荷和ALT水平。[35]但是，我们的研究是不同的，因为我们的病人会通过免疫清除期在以往的研究中。因此，HBV DNA是相对较低，ALT水平大多正常在我们的队列。

血清IP-10级已经不单单是在慢性乙肝患者，而且在那些患有慢性丙型肝炎（CHC）具有重要作用。我们感兴趣的是血清IP-10级和病毒学应答的走势运行在这两种类型的慢性病毒性肝炎相反的方向。高基线血清IP-10级别的预测更糟糕的结果丙肝患者，例如：非应答者，以治疗;而低基线血清的IP-10水平预测早期和持续的病毒学反应。[36]但是，这些观察不一定意味着宿主的免疫应答作用在慢性乙型肝炎和CHC以相反方式。事实上，血清IP-10水平之间的丙肝患者谁也持续病毒学应答更显著下降。我们的研究也证明慢性乙型肝炎患者的HBsAg即将转阴一个显著下降血清IP-10级。最近的研究显示特异性免疫应答无法得到控制病毒复制，被感染的肝细胞会分泌的IFN-γ诱导的趋化因子（例如CXCL9，CXCL10和CXCL11）导致非特异性的单核细胞到肝脏的迁移。[37]这些单核细胞是在事实上无法控制感染，但导致持续的坏死性炎症，因此肝损伤。[38]在另一方面，抑制这些趋化因子的限制非特异性细胞迁移，因而降低了炎症反应。[39]

我们的研究具有了患有乙肝表面抗原血清学清除，对照组的非常严格的匹配条件，这有利于即使是相对细微差别的检测细胞因子的变化可敬号码的实力。我们还检查细胞因子的综合面板。然而，我们的研究也有一些局限性。在某些时间点的血清中不完全排除收藏系列血清IP-10和HBsAg水平的分析，在所有的情况下，与对照组。的夹杂物都处理和未处理的受试者可能引入从抗病毒治疗引起的免疫反应的混杂影响。例谁的HBsAg血清清除我们的系统的血清样品的收集前并没有包括在该分析中，这些都导致了更小的样本大小。虽然病例和对照已经匹配在几个重要的临床指标（年龄，性别，HBeAg状态和肝硬化的存在），仍有可能不可测量的干扰因素。

总之，血清IP-10水平与乙肝表面抗原血清学清除，最佳的病毒学终点慢性乙肝有关。最初的下降，随后的血清IP-10级左右的HBsAg血清清除暗示宿主免疫配置文件被修改在一个复杂的，而不是在那个时候一个简单的，线性的方式时间以后上升。