Improvement of liver disease and inflammation in patients with advanced HCV-related cirrhosis after antiviral therapy


 Eradication of hepatitis C virus (HCV) promotes an improvement in liver disease and deactivation of the immune system. Here, we aimed to evaluate the changes in liver disease scores and plasma biomarkers following HCV clearance with direct-acting antivirals (DAAs) in patients with advanced HCV-related cirrhosis. We performed a prospective study of patients with advanced cirrhosis (liver decompensation or liver stiffness measurement [LSM]≥25 kPa or hepatic liver pressure gradient [HVPG]≥10 mmHg or Child-Pugh-Turcotte (CTP)≥7) who received DAA therapy. Variables were assessed at baseline and 48 weeks after HCV treatment completion. Statistical analyses were performed employing Generalized Linear Mixed Models (GLMM). In this study, 62 patients (12 HCV-monoinfected and 50 HIV/HCV-coinfected) who achieved SVR were included in the analysis. We found significant decreases in LSM, HVPG, LBP, IP-10, IL-8, IL-1β, IL-18, IL-1RA, OPG, sVCAM-1, sICAM-1, TNFR-I, PAI-1, and VEF-A during follow-up. Variations in most outcome measures were similar as for HCV-monoinfected and HIV/HCV-coinfected patients. We found significant positive associations between changes in LBP, IP-10, MCP-1, IL-8, IL-1β, IL-18, IL-6, OPG, sVCAM-1, sICAM-1, and PAI-1 and LSM values after HCV clearance. We found variations in LBP, IL-8, IL-6, IL-18, OPG, PAI-1, and D-dimer were positively associated with changes in HVPG values; and variations in IP-10, IL-1β, IL-6, TNF-α, IL-1RA, OPG, and sICAM-1 were related to changes in CTP score. In conclusion, the eradication of HCV with all-oral DAAs promoted a decrease in the severity of advanced cirrhosis and plasma biomarkers (inflammation, coagulopathy, and angiogenesis), which were positively associated with each other.

The human immunode ciency virus (HIV) co-infects a large number of patients infected with HCV since both viruses share the same routes of transmission [21]. HIV infection accelerates the course of CHC, resulting in higher rates of cirrhosis, end-stage liver disease, and death when compared to HCVmonoinfected patients [22]. In HIV/HCV-coinfected patients, HIV infection increases a series of negative factors such as HCV replication, HCV-induced hepatic in ammation, hepatocyte apoptosis, and microbial translocation, while it also leads to a deterioration of the speci c immune responses against HCV [23,24]. Antiretroviral therapy (ART) delays the progression of brosis and reduces clinical complications among HIV/HCV-coinfected patients, but despite suppressive ART, these patients still have abnormally high levels of plasma biomarkers related to bacterial translocation, immune activation, in ammation and coagulation [24], which are related to higher morbidity and mortality [25]. Similarly, DAA treatments against HCV infection also achieve elevated SVR rates [26,27] and delay CHC progression in HIV/HCVcoinfected patients [28,29]. Several reports have also shown a signi cant decrease in plasma biomarkers related to in ammation and immune activation after SVR with all-oral DAAs in HIV/HCV-coinfected patients [18,[30][31][32][33][34]. However, DAA therapy alone does not completely block uncontrolled in ammation and liver injury, particularly with advanced liver disease [35]. Thus, some cirrhotic patients who achieve SVR with DAAs remain at risk of cirrhosis progression and developing hepatocellular carcinoma [2]. Despite many reports that have shown signi cant decreases in liver disease scores and plasma biomarkers after SVR with DAAs therapy, there still is a lack of consistency in the characteristics of these studies, such as the time points after completion of HCV treatment and statistical analyses used, as well as the biomarkers analyzed and identi ed to be statistically different and liver disease stage of the patients included.
In this study, we assessed changes in liver disease scores (LSM, HVPG, and CTP) and plasma biomarkers related to in ammation, coagulopathy, and angiogenesis in patients with advanced HCV-related cirrhosis following SVR with all-oral DAAs, as well as the association between the changes in the two set of markers during the follow-up.

Patients
We carried out a multicenter prospective study of 62 patients with advanced HCV-related cirrhosis who started anti-HCV therapy with all-oral DAAs, 12 HCV-monoinfected patients (HCV-group) and 50 HIV/HCVcoinfected patients (HIV/HCV-group). Patients were recruited at four tertiary referral hospitals in Madrid (Spain) between January 2015 and June 2016 (ESCORIAL study; see Appendix). This study was approved by the Research Ethics Committee of the Instituto de Salud Carlos III (CEI PI 41_2014) and was conducted according to the Declaration of Helsinki. All participants gave their written consent before enrollment.
The inclusion criteria were: 1) active HCV infection at baseline de ned by detectable plasma HCV-RNA; 2) advanced cirrhosis (prior history of liver decompensation (ascites, bleeding esophageal varices, hepatic encephalopathy), or LSM ³25 kPa, or HVPG ³10 mmHg or CTP ³7); 3) starting anti-HCV therapy with alloral DAAs (without interferon (IFN) or ribavirin) and achieving a SVR de ned as an undetectable HCV-RNA load 12 weeks after nalization of anti-HCV therapy.
From 97 patients included in the ESCORIAL study and with plasma sample at baseline, 35 were lost to follow-up (2 deaths, 6 abandonment of the study, 2 virologic failure, and 25 unavailable at the end of follow-up (48 weeks after completing HCV therapy)). Therefore, 62 patients were included in the study: 12 HCV-monoinfected patients (HCV-group) and 50 HIV/HCV-coinfected patients (HIV/HCV-group).

Clinical data
Epidemiological and clinical data were collected through an online form, which ful lled the con dentiality requirements. These data were monitored to verify that all the information in the database matched the patients' records.
The LSM was carried out by trained operators of transient elastography (FibroScan®, Echosens, Paris, France), as we previously described [36]. LSM values were reported in kilopascals (kPa) and range from 2.5 to 75 kPa. The CTP score was calculated from ve factors (total bilirubin, albumin, international normalized ratio, ascites, and encephalopathy) and range from 5 to 15 points. Hemodynamic studies were performed after overnight fasting under light sedation with intravenous midazolam, as we previously described [37]. The HVPG was measured as the difference between wedged hepatic venous pressure and free hepatic venous pressure in millimeters of mercury (mmHg).
Statistical analysis Stata 15.0 (StataCorp, Texas, USA) and SPSS 22.0 (SPSS INC, Chicago, IL, USA) were used to perform the statistical analyses. All p-values were two-tailed and were corrected for multiple testing to reduce the risk of a spurious result by using the false discovery rate with Benjamini and Hochberg (q-values). The statistical signi cance was de ned as p≤0.05 or q≤0.05, as appropriate.
Categorical variables were analyzed by the Chi-square test or Fisher's exact test, as required. Continuous variables were analyzed by Mann-Whitney U test (independent groups). Moreover, we used Generalized Linear Mixed Models (GLMM) with a gamma distribution (log-link) to evaluate repeated measurements.
Gamma distribution in GLMM is recommended for modeling skewed continuous outcomes. The log-link was applied only to the outcome measure or dependent variable (y), which were the severity scores of liver cirrhosis and plasma biomarkers values. Our model only included two factors or independent variables (x), group (HIV/HCV-group vs HCV-group) that is the between-subject effect and time (basal vs nal) that is the within-subject factor time. The interaction between group and time was taken into account. This analysis gives us the estimation of average increase (Δx= nal value (x2) -baseline value (x1)) of each biomarker in each of the two study groups. GLMM models were also used to analyze the association between the change in plasma biomarkers (Δx) and the changes in severity scores of liver disease (Δy) during the follow-up. This analysis gives us the regression coe cient (β), which indicates the size and direction of the effect, according to your positive or negative sign.
Finally, GLMM was used to evaluate whether the change in outcome measures (liver disease scores and plasma biomarkers values) during the follow-up was similar between the HIV/HCV group and HCV group (impact of HIV infection). In this case, GLMM was adjusted by the most relevant patient's characteristics at baseline: gender, age, alcohol consumer, intravenous drug user, previous IFN therapy, HCV genotype, log 10 HCV RNA, and statins. This analysis gives us an estimate of the difference between the increments for each group (Δx). Table 1 shows the epidemiological and clinical characteristics of the 62 patients with advanced HCVrelated cirrhosis (12 HCV-monoinfected and 50 HIV/HCV-coinfected patients). The HIV/HCV-group were younger (p-value=0.002) and had less percentage of patients previously treated with IFN therapy (p-value=0.021), but had a higher percentage of intravenous drug users (p-value<0.001). All HIV/HCVcoinfected patients were on ART and plasma HIV viral load was undetectable (<50 copies/mL). None of the participants were actively using injection drugs.

Results
Baseline values of the outcome measures Table 2 shows the baseline values of the analyzed markers in this study. Regarding liver disease, the HIV/HCV-group had a lower percentage of patients with CTP≥7 (p-value=0.014) and lower values of sCD14 (p-value=0.001), FABP-2 (p-value=0.009), and sRANKL (p-value=0.029) than HCV-monoinfected patients. However, only the difference in sCD14 remained statistically signi cant after an adjustment for multiple comparisons (q-value=0.029).
We used an adjusted GLMM to assess whether there were differences in the change of outcome measures during the follow-up between the two study groups (HIV/HCV-group vs HCV-group). In this case, signi cant differences between groups were found in CTP (p-value=0.002) and TNFR-I (p-value=0.012), but these differences disappeared after correcting for multiple testing (see Additional File 4). This indicates that changes during the follow-up in outcome measures were very similar between the HIV/HCV-group and the HCV-group for most markers.

Discussion
In this study, the major ndings in patients with advanced HCV-related cirrhosis were: i) HCV eradication with all-oral DAAs promoted an improvement in liver disease scores and plasma biomarkers (in ammation, coagulopathy, and angiogenesis); ii) HIV infection showed a slight impact on the evolution of outcome measures analyzed (liver disease scores and plasma biomarkers), since the evolution of the two study groups (HIV/HCV-coinfected and HCV-monoinfected patients) was similar; and iii) the decreases in plasma biomarkers (in ammation and coagulopathy) were associated with decreases in liver disease severity scores. Therefore, our data suggest HCV elimination with all-oral DAAs was linked to a signi cant improvement in liver disease scores and plasma biomarkers related to in ammation, coagulopathy, and angiogenesis in patients with advanced cirrhosis, with there being a relation between both changes. Our results could shed light on the evolution of patients with advanced HCV-related cirrhosis after HCV eradication with all-oral DAAs, irrespective of HIV infection.
Both HIV and HCV infection and advanced cirrhosis promote in ammation, immune activation, and dysfunction of the immune system, which are all linked to a greater severity of liver disease and the development of comorbidities [2][3][4]. Bacterial translocation in patients with advanced HCV-related cirrhosis is a key step in the pathogenesis of spontaneous bacterial peritonitis and bacteremia as well as a main factor that triggers the immune activation and in ammation, which in turn promotes hemodynamic changes and the development of decompensated cirrhosis [4]. Endothelial dysfunction is also promoted by in ammation and is linked to the progression of CHC and the development of cardiovascular events [39]. Moreover, hepatocytes release most of the proteins in the blood, where plasma levels can be altered by CHC progression, promoting an increase in thrombotic risk [40].
Coagulopathy is related to increased risk of disease progression and death in HIV-infected patients [41] and patients with advanced liver disease [42]. In a recent article on HIV/HCV-coinfected patients, we showed that plasma biomarkers of bacterial translocation, in ammation, endothelial dysfunction, and coagulopathy increased with the increase of liver brosis severity, particularly in patients with LSM ≥ 40 KPa [36]. However, the eradication of HCV with antiviral therapy can stop this pathological process of the liver, at least almost entirely [35]. Additionally, the decrease in in ammation biomarker levels could also indicate a lower risk of developing comorbidities in cirrhotic patients who achieved SVR with DAA therapy [2,25].
In our study overall, HCV clearance after DAA therapy promoted a signi cant improvement in severity scores of liver cirrhosis and many plasma biomarkers linked to in ammation (bacterial translocation, in ammatory response, and endothelial dysfunction) and coagulopathy. Our data are in concordance with a large number of previous studies that found a signi cant decrease in liver disease scores of HIV/HCV-coinfected patients [6,7,32,[43][44][45] and HCV-monoinfected patients [6][7][8][9][10][11][12][13][14] after HCV eradiation with DAA therapy; and also in plasma biomarkers of HIV/HCV-coinfected patients [18,[30][31][32][33][34] and HCV-monoinfected patients [15][16][17][18][19][20]. However, there is an important lack of consistency in these previous publications regarding the plasma biomarkers and liver severity scores evaluated, time-points used to take data or samples after the end of HCV treatment, statistical analysis used, and liver brosis stages included. In our study, a high number of markers were found to be signi cant despite the small sample size, and others might be masked due to limited statistical power. In addition and notedly, other factors provided robustness to our study, such as its prospective design (repeated measures), the evaluation of changes in markers at 48 weeks after treatment completion (medium-term), the higher suitability of the statistical analysis (GLMM), and the fact that all the patients had advanced cirrhosis. Our analysis of VEGF-A, which plays a key role in liver cancer angiogenesis [46], revealed an interesting observation. Previous reports did not nd any change in plasma VEGF-A values [43,47] or showed a signi cant increase [48] after HCV eradication with DAA therapy. However, our data suggest a discontinuation in the process of angiogenesis and a shift towards an "antiangiogenic" pro le, since we observed a signi cant decrease in VEGF-A levels after HCV therapy with DAAs therapy.
Previously published studies do not usually have a control group of HCV-monoinfected patients, or if they do, they do not analyze the impact of HIV infection on the evolution of plasma biomarkers and liver disease scores after HCV eradication with all-oral DAAs [18,[30][31][32][33][34]. In our study, baseline values of outcome measures (plasma biomarkers and liver disease scores) and their subsequent evolution after DAA therapy were very similar in HIV/HCV-coinfected and HCV-monoinfected patients. Therefore, HIV infection only showed a slight impact on the outcome measures in our cohort of patients with advanced HCV-related cirrhosis, both at baseline and during follow-up after HCV eradication with AAD. This fact is remarkable because HIV infection promotes a faster progression of chronic hepatitis C [22] and higher plasma levels of markers linked to bacterial translocation, immune activation, in ammation and coagulation, despite suppressive ART [24]. It is probable that the absence of relevant differences between groups may be due to the fact that all of our patients had advanced stages of cirrhosis, where CAID is usually found, characterized by elevated immune activation, in ammation, and dysregulation of the immune system [4]. Additionally, Corma-Gómez et al. [49] have recently reported that HIV coinfection is not related to a higher risk for developing liver complications in HCV-infected patients with advanced brosis, who achieved SVR with IFN-free regimens. Malin et al [50] observed that the LSM regression after successful DAA therapy did not differ in patients with HCV/HIV coinfection and those with HCV monoinfection.
The relationship between plasma biomarkers (bacterial translocation, in ammatory response, endothelial dysfunction, coagulopathy, and angiogenesis) and liver disease scores (LSM, HVPG, and CTP) has been scarcely explored in patients with advanced cirrhosis after HCV eradication with DAAs. Laursen et al. [20] reported that the levels of macrophage activation markers (sCD163 and soluble mannose receptor) correlated with LSM values during follow-up in HCV-monoinfected patients. Kostadinova et al. [32] showed that an improvement in LSM values correlated with a decrease in sCD14 levels after IFN-free HCV therapy. Additionally, plasma levels of sCD163, IL-6, and Mac2-binding protein correlated with changes in AST level and APRI score. In our study, the decreases in several plasma biomarkers levels linked to bacterial translocation, in ammatory response, endothelial dysfunction, and coagulopathy were directly associated with the decreases in liver disease scores (LSM, HVPG, and CTP) after HCV eradication with DAAs. Therefore, our data suggest a relation between improvement in liver cirrhosis and resolution of in ammation after HCV eradication with all-oral DAAs.

Conclusions
In conclusion, eradication of HCV with all-oral DAAs in patients with advanced HCV-related cirrhosis promoted an improvement in the severity of advanced cirrhosis and plasma biomarkers (in ammation, coagulopathy, and angiogenesis), in both HIV/HCV-coinfected and HCV-monoinfected patients. Changes in these plasma biomarkers related to in ammation were positively associated with an improvement in liver disease scores.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analyzed during the current study may be available from the corresponding author upon reasonable request.

Competing interests
The authors declare that they have no competing interests.