Severe Liver Fibrosis and Association with Plasma Inammatory Biomarkers among HIV/HCV Coinfected Patients in China :a Cross-Sectional Study

Background: Immune dysregulation among HIV/HCV co-infected patients with impaired liver function is common. Thus, this study aimed to evaluate the association of liver brosis with microbial translocation and related inammation among HIV/HCV co-infected patients. Methods: This cross-sectional study involved 343 HIV/HCV co-infected patients who received cART. All patients had current blood biochemical testing data. We measured sCD14 and 27 serum cytokines concentrations using the Hycult Biotech sCD14 ELISA kit and Bio-plex Human Cytokine 27-plex Assay, respectively. We compared the concentrations of each marker between severe liver brosis and mild liver brosis. Odds ratios (ORs) and 95% condence intervals (95%CIs) for the association of each marker with severe liver brosis were estimated using logistic regression. Results: Of the 343 HIV/HCV coinfect-ed patients enrolled, 188 (54.8%) had severe liver brosis (FIB-4 >3.25). Patients with higher FIB-4 score (>3.25vs. ≤ 3.25) had higher plasma level of IL-1β, IL-6, IL-7, IL-9, IL-12, IL-15, IL-17, GM-CSF, IFN-γ, TNF-α, IL-4, IL-10, IL-13, BasicFGF and MCP-1. Multivariate logistic regression analysis showed that increased plasma level of IL-1β, IL-6, IL-7, IL-12, IL-17, GM- CSF, IFN-γ, IL-4, IL -10, MCP-1, Eotaxin, BasicFGF and sCD14 were linked to severe liver brosis in our study. Conclusions: Severe liver brosis are associated with increased microbial translocation plasma inammatory biomarkers among HIV/HCV co-infected patients. a on BioPlex®200 27-plex protein protein–1 macrophage inammatory MIP–1b, T-cell expressed and secreted (RANTES); and growth factors Eotaxin, broblast FGF), growth factor and vascular growth factor and the more


Background
Human immunode ciency virus (HIV) and Hepatitis C virus (HCV) co-infection is a worldwide serious public health concern because of increased mortality and disease burden [1,2]. Compared to HCV monoinfected patients, HIV/HCV co-infected patients have more frequent and accelerated progression to brosis, cirrhosis and end-stage liver disease (ESLD) [3]. Differ from western countries, direct-acting antivirals (DAAs) even peg-interferon plus ribavirin (PR) in China especially rural area cannot be widely used because of the high price [32].So the burden of disease caused by HIV/HCV may be heavier in China. Pathogenesis of progressive liver brosis in HIV/HCV co-infected is complex and multifactorial [4], and is mainly related to persistent HCV infection with altered cellular immunity as well as immune activation with elevated pro-in ammatory and pro brogenic cytokines [5]. More speci cally, HIV infection induces immune suppression and leads to CD4 T-cell depletion, causes persistent innate and acquired immune activation, and directly stimulates Kupffer cells and hepatic stellate cells (HSCs) to secrete either probrotic cytokines or type 1 collagen through a C-C chemokine receptor-5 (CCR5)-dependent pathway [5]. In addition, the persistence of HCV infection due to weakened CD4 and CD8T cell responses caused by HIV[6] contributes to increased rates of liver brosis[7-9].
Moreover, microbial translocation (MT), which is thought to be associated with loss of mucosal barrier function and increased intestinal permeability secondary to immune dysregulation and/or alterations in the intestinal microbiome, has often been observed among HIV-infected individuals [10]. Blood microbial components including peptidoglycan, lipoteichoic acid, lipopolysaccharide (LPS) and agellin may directly promote liver brosis in HIV/HCV co-infected patients by stimulation of HSCs and/or Kupffer cells with LPS binding protein (LBP) and soluble CD14 (sCD14). These microbial components may enhance local hepatic in ammatory immune responses and activation-induced liver cell death [4].
Thus, the above-mentioned mechanisms jointly emphasize the important role played by in ammation in liver brosis among HIV/HCV co-infected patients. Cirrhosis-associated dysregulation of immune responses is re ected by increased production and elevated serum levels of pro-in ammatory cytokines and upregulated expression of cell activation markers [11]. Kupffer cells generate IL-1, IL-6, IL12, IL-18 and also release anti-in ammatory cytokines, including IL-10 after stimulation by LPS [12]. Upregulation of IL-4, IL13, TGF-1 and platelet-derived growth factor (PDGF) was observed during brogenesis, while among HIV-infected patients, gp120 may induce HSC accumulation by secretion of monocyte chemoattractant protein-1 (MCP-1) by HSCs [13]. Furthermore, HIV suppression by combination antiretroviral therapy (cART) among HIV/HCV co-infected patients may decrease in ammation and immune activation and slows down the progression of liver disease [14], whereas the levels of plasma in ammatory biomarkers remain abnormal in many individuals [15]. However, the in ammatory pro le among HIV/HCV co-infected patients with different levels of liver brosis has not been well established.
Up to now, studies of the role of microbial translocation and plasma in ammatory biomarkers in liver brosis are limited by small sample sizes and small number of examined biomarkers [16][17][18]. To ll this gap, the present study aimed primarily to examine the association of liver brosis with microbial translocation and related in ammation among HIV/HCV co-infected patients in the era of cART.

Study sample
The present cross-sectional study was conducted in were identi able during the study period from April to October in 2016. Of them, 390 (84.2%) were receiving cART and gave informed consent to participate in the present study. Forty-seven patients were further excluded from the study due to missing data on biochemical measures about aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count (PLT) that are necessary to evaluate liver brosis status. Thus, a total of 343 HIV/HCV co-infected patients were included in the nal analysis. These 343 participants had no signi cant differences with the 120 identi able yet excluded HIV/HCV co-infected patients in the distribution of age, sex, marital status, HIV transmission route, HBsAg serostatus, baseline CD4 cells count, years on cART, current HCV RNA and HIV RNA levels except for ethnicity (Table S1).

Data Extraction
Demographical and clinical epidemiological data were extracted from the CRIMS. The data included age, sex, marital status, ethnicity, HIV transmission route, date of cART initiation, antiretroviral regimen, and CD4 cell counts at cART initiation and follow-up visits.

Blood testing
Biochemical tests for liver brosis assessment

HCV RNA Quanti cation and genotyping
Plasma HCV viral RNA was extracted (Roche diagnostic products (Shanghai) Co., Ltd., China) and quanti ed by a real-time polymerase chain reaction (RT-PCR) technique using commercially available kits for the quanti cation of HCV RNA (PCR-Fluorescent Probing, PG Biotech Ltd., Shenzhen, China). The limit of detection was 500 copies/ml, and the linear range of HCV RNA quanti cation was from 1.0 × 10 3 to 5.0 × 10 7 copies/ml. Ampli cation was completed by a nested PCR with E1-or NS5B-speci c primers. Splicing, proofreading, and aligning sample sequences were performed using ChromasPro 1.5 and BioEdit7.0.9.0 software. HCV genotype reference sequences were retrieved from the HCV database (http://hcv.lanl.gov/content/sequence/HCV/ToolsOutline.html). The phylogenetic tree was established by the Neighbor-joining method of MEGA 7.0 software. The Bootstrap repeat detection value was set to 1000 times, and the HCV gene subtype was further determined according to the phylogenetic tree[25].
Measurements less than the lower limit of quanti cation (LLOQ) were assigned a value of half the LLOQ, and measurements more than the upper limit of quanti cation (ULOQ) were assigned a value of twice the ULOQ for each marker.

Statistical analysis
Group comparisons were assessed using chi-square test or Fisher's exact test for categorical variables and t-test or Mann-Whitney U test for continuous variable. Log10-transformation were conducted for variables of plasma cytokines levels for further statistical and regression analyses. A multiple logistic regression analysis with adjustment for age, sex, ethnicity, current HIV RNA, current HCV RNA, current CD4 cell counts, years since cART and ART regimen type was undertaken to explore the correlation of liver brosis with the plasma level of each of the twenty-seven cytokines and sCD14. The odd ratio (OR) and 95% con dence interval (95%CI) represent the risk of liver brosis per one log-unit change in plasma cytokine concentration. Spearman correlations were computed to explore associations between the plasma in ammatory biomarkers. Statistical signi cance was de ned as p <0.05 and Bonferroni p <0.002 (0.05/27≈0.002) for multiple comparison adjustment [26]. All statistical analyses were performed using R software (version 3.3.2).

Demographical and clinical characteristics
The median (IQR) age at baseline was 35.4 (31.3-39.3) years old. The majority of the participants were male, injection drug users (IDUs), seronegative for hepatitis B surface antigen (HBsAg), detectable for plasma HCV RNA but undetectable for plasma HIV RNA. Around half of the participants were non-Han ethnicities, i.e., ethnic minorities (54.2%), currently married (51.9%), and of HCV subtype 3 (47%) ( Table  1).

Prevalence of Severe Liver brosis
Among all participants, 75 (21.9%) had no or mild liver brosis, 80 (23.3%) had intermediate liver brosis, and 188 (54.8%) had severe liver brosis. Compared to patients without severe liver brosis, those with severe liver brosis were signi cantly different in age, education, cART regimen, and years on cART (Table  1).

Correlations of in ammatory biomarkers and their associations with severe liver brosis
Most of the 27 cytokines were signi cantly correlated with each other with the exception of RANTES, which was not correlated with most of the other cytokines ( Figure 2). The correlation coe cients are shown as a supplement in Table S2.
Given the high correlation between the 27 cytokines, separate multiple logistic regression models adjusting for potential confounders were performed to examine the association of sever liver brosis with the each of the 27 cytokines and sCD14 levels. Severe liver brosis was shown to be associated with higher levels of eleven out of the twenty-seven cytokines with a signi cance level of 0.05, only Basic FGF was found to be positively and signi cantly associated with severe liver brosis after Bonferroni correction for multiple comparisons (aOR = 1.82; 95%CI: 1.26-2.66; p = 0.002). Furthermore, plasma sCD14 was signi cantly associated with severe liver brosis (aOR = 1.14; 95%CI: 1.01-1.30; p = 0.048. Figure 3).

Discussion
In the present study, 54.8% of HIV/HCV co-infected patients had severe liver brosis or cirrhosis. The prevalence was higher than studies conducted in western countries which range from 22.9% to 40.3% Many studies have shown that in ammation plays an important role in the development of liver brosis among HIV/HCV co-infected patients [36,37]. In this regard, levels of cytokines, chemokines and growth factors closely related to in ammation were altered. We found that patients with cirrhosis who had FIB-4 >3.25 showed a signi cantly higher plasma level of various in ammatory makers (IL-1β, IL-6, IL-7, IL-9, IL-12, IL-15, IL-17, GM-CSF, IFN-γ, TNF-α, IL-4, IL-10, IL-13, BasicFGF and MCP-1). Although the median plasma level of sCD14 was higher in the FIB-4 >3.25 group than FIB-4 ≤3.25, there was no signi cant difference.
Chronic in ammation and brosis are inextricably linked, and the interactions between immune effector cells, local broblasts and tissue macrophages at sites of scar formation determine the outcome of liver injury. With improved understanding of the processes that govern in ammation and brosis, it has become clear that both the adaptive and innate immune systems are involved in the regulation of brosis [38].
In our study, multivariate logistic regression analysis shows that increased plasma levels of IL-1β, IL-6, IL-7, IL-12, IL-17, GM-CSF, IFN-γ, IL-4, IL -10, MCP-1, Eotaxin, BasicFGF and sCD14 are linked to severe liver brosis. Our ndings are in concordance with previous studies that show plasma levels of proin ammatory, including TNF-a, IL-6, IL-8, IL-12, are pro brogenic and induce liver damage [39,40]. In addition, these studies show that cirrhotic patients display increased production of IFN-γ, shows a positive correlation between IFN-γ levels and liver brosis. The association between IL-17 and brosis is also controversial. Sheila et al found an inverse correlation between IL-17 and aminotransferase-to-platelet ratio index (APRI) [45]; to the contrary, Meng et al found that IL-17 could exacerbate liver brosis [46]. In our study, we found that increased IL-17 concentration is associated with higher FIB-4 score. Regarding IL-2, a previous study found a negative correlation of IL-2 with cirrhosis [47]. However, our data shows that IL-2 levels increase as brosis increases.
Anti-in ammatory cytokines, including IL-4 IL-10 and IL-13, are pro brogenic cytokines, and their levels increase in coinfected patients [48,49], which is in accordance with our ndings.
Chemokines may promote in ammation through the recruitment of lymphocytes to the liver parenchyma in chronic hepatitis C virus (HCV) infection [50]. MIP-1a is a pro brogenic chemokine. Tazi, et al found that cirrhotic monocytes spontaneously produce chemokines (MCP-1) unlike normal monocytes[51]. Although increased MIP-1a increased the risk of developing cirrhosis in our study, it was not statistically signi cant. Many studies have shown that plasma IP-10 levels increased as liver brosis increased in HCV or HIV / HCV co-infected [33, 52, 53, 54], but our data did not nd this association.
As for now, studies on the involvement of growth factors other than TGF-β1 in liver brosis are rare. In our study, multivariate logistic models indicate that increased BasicFGF and Eotaxin levels persist as predictors for liver cirrhosis.

Limitations
This study has several limitations. Firstly, due to cross-sectional design, this study fails to establish the direction of causality, e.g., in ammation may be a cause or response to liver cirrhosis. Secondly, we did not include factors known to in uence the risk of liver brosis that could alter levels of immune/in ammatory markers such as alcohol abuse. Thirdly, we did not set a control group of HCVmonoinfected or HIV-monoinfected patients, and we are not able to know information of possible differential biomarkers from them.

Conclusions
In summary, severe liver brosis are associated with increased microbial translocation plasma in ammatory biomarkers among HIV/HCV co-infected patients. Our ndings provide preliminary evidence that immune/in ammatory markers may be useful predictors of liver brosis. Further research using longitudinal design is warranted to determine the causal relationship between liver brosis and in ammation, in order to look for the optimal care and management of HIV/HCV co-infected patients in China.  data was presented as median and interquartile range (IQR), with μg/mL as the unit for sCD14 and pg/mL for all other cytokines. Figure 1 Comparison of log10 transformed plasma concentration of tested biomarkers between participants with different FIB-4 score. *p<0.05, **p<0.01, and ***p<0.001. All the p-values were calculated by Mann-Whitney tests.

Figure 2
Spearman correlations between plasma in ammatory biomarkers. "x" represents p≥0.05. Blue and red indicate that the two variables were positively and negatively correlated, respectively. The darker the color (the larger the circle), the greater the correlation of the variables.
Page 22/23 Figure 3 adjusted Odds ratios and 95% con dence intervals for the associations between plasma cytokine levels (log10 transformed) and serve liver brosis among HIV/HCV co-infected patients. Each variable was assessed in separate multivariable logistic regression adjusting for age, sex, ethnicity, current HIV RNA, current HCV RNA, current CD4 count, years on cART, and ART regimen.