Reversibility of some oxidative stress markers in chronic hepatitis C patients after receiving direct-acting antiviral agents

Abstract

Introduction

Hepatitis C (HCV) is associated with extra-hepatic involvment, morbidity as well as metabolic changes. Whether these might be reversible if sustained virologic response (SVR) is achieved by direct-acting antiviral (DAA) therapy remains unknown.

Methods

Chronic hepatitis C (CHC) individuals receiving DAA treatment with SVR were compared to those who underwent spontaneous clearance (SC) of HCV infection at the 2-year follow-up. Plasma oxidative stress markers (oxidized low-density lipoprotein (oxLDL), 8-hydroxy-2′-deoxyguanosine (8-OHdG), malondialdehyde (MDA) and ischemia-modified albumin (IMA)) as well as progression of liver fibrosis were evaluated.

Results

Compared to SC individuals, those in the CHC group exhibited at baseline higher levels of oxLDL, 8-OHdG and IMA but not of MDA. In the SC group, 8-OHdG levels were elevated at 2-year post-SVR (p = 0.0409), while the DAA-treated CHC group showed decrease in oxLDL (p < 0.0001) and 8-OHdG (p = 0.0255) levels, approaching those of the SC group, but increased MDA (p = 0.0055) levels. Additionally, oxLDL levels were positively correlated with liver stiffness measurements at SVR (p = 0.017) and at 1 year post- SVR (p = 0.002).

Conclusions

Plasma oxLDL showed post-SVR normalization after clearance of HCV viremia with DAAs and was associated with levels of hepatic fibrosis.

1. Introduction

Currently 58 million persons are chronically infected with hepatitis C (HCV) worldwide, with a global prevalence of 0.7% and about 1.5 million new cases per year.¹ Chronic hepatitis C (CHC) individuals might develop hepatic inflammation, cirrhosis and cancer and, concomitantly, metabolic disturbances leading to metabolic syndrome (MetS), type II diabetes and steatosis.² Well-tolerated direct-acting antiviral (DAA) therapy for CHC individuals can achieve viremia clearance and sustained virological response (SVR), with cure rates of about 95% when using recommended regimens for 8–12 weeks³; however, the post-SVR benefits over the years have not been fully determined.

HCV infection disrupts the host redox homeostasis, which is reflected by chronic oxidative stress in the face of virus-induced reactive oxygen species (ROS) overload, with excess oxidants contributing to DNA, protein and lipid damage, thereby inducing liver inflammation, pathogenesis and disease progression.^4,5 Unlike the rare elimination of chronic hepatitis B, HCV is characterized by viral persistence that can be eradicated by DAAs. It is therefore now important to assess CHC individuals regarding the potential reversibility of such aspects as impaired immunity⁶, and metabolic^7,8 and redox^9,10 homeostasis following viral clearance. Within 1 year of HCV clearance by DAAs, previous results have shown the restoration of host innate and cellular function⁶ as well as improvement of metabolic^8,10, 11, 12, 13 and redox⁹ parameters but not all impaired markers were fully restored.^6,7,10,14

However, since HCV infection may profoundly affect the cellular oxidative status in those individuals during decades of chronic inflammation, a short duration after the removal of viremia may not be long enough to correct the redox disturbance. Therefore, in the present study, the reversibility of the systemic oxidative stress status was examined in CHC participants receiving DAAs 2 years after SVR and compared with treatment-naïve patients who underwent spontaneous clearance (SC) as controls. We have assessed plasma biomarkers for DNA oxidative damage [8-hydroxy-2′-deoxyguanosine, (8-OHdG)], lipids [oxidized low-density-lipoprotein (oxLDL], malondialdehyde (MDA)) and protein [(ischemia-modified albumin (IMA)], as well as their relationship with liver fibrosis progression.

2. Materials and methods

2.1. Patients

Forty CHC participants who had achieved SVR with DAA treatment were recruited as the study group, and 14 SC participants as control group. All participants had not had previous HCV treatment, including those in the CHC group who had HCV antibody as well as detectable HCV RNA at least twice 6 months apart, and those in the SC group who had HCV antibody but negative HCV RNA viremia at recruitment. Another 132 CHC participants who had achieved SVR with DAA treatment (50 men and 82 women, median age: 63 years, range: 29–87 years) were recruited for further confirmation of results. Plasma samples were collected at baseline (BL), SVR, 1 year after SVR (SVR-1Y), and 2 years after SVR (SVR-2Y) in the DAA-treated group, whereas samples in the SC group were collected at baseline and at 1 and 2 years follow-up (SC-BL, SC-1Y and SC-2Y). Serum samples were stored at −80 °C until analysis. All participants were enrolled at the National Cheng Kung University Hospital (NCKU), Tainan, Taiwan and each signed an informed consent form. The study was approved by the local Institutional Review Board.

2.2. Laboratory evaluations

We performed blood sampling to test for plasma triglycerides (TG), cholesterol (Chol), high-density lipoprotein (HDL), low-density lipoprotein (LDL), oxLDL, 8-OHdG, MDA and IMA to evaluate atherogenic risks and oxidative stress. Biochemical parameters in plasma or serum samples were extracted from the automated routine laboratories of the NCKU Hospital. Homeostasis model assessment (HOMA)-insulin resistance (IR) and HOMA-beta values were also calculated.¹⁵ The oxLDL (Cusabio, TX), 8-OHdG (Invitrogen, MA), MDA (Cloud-Clone Corp, CN) and IMA (Cloud-Clone Corp) levels were quantitated by the corresponding competitive ELISA kits, and the final absorbance measured by a microplate spectrophotometer (Thermo Fisher, MA). A four-parameter logistic curve was used to fit the competitive ELISA reading values. Liver stiffness measurement (LSM) was performed by transient elastography (FibroScan®, Echosens, France). The presence of medical conditions including diabetes, hypertension and chronic kidney disease was recorded (Supplementary Table 1).

2.3. Data and statistical analysis

Distributions of continuous variables were described as median (range). Differences between groups were evaluated with the Mann-Whitney U test and independent sample t-test or between time points within a group by the paired t-test. The Spearman correlation test was used to analyze correlations between continuous variables. Dynamic differences between the DAA-treated CHC and SC groups were evaluated with a linear mixed-model analysis and represented as the mean ± S.E.M. Statistical tests were conducted using SPSS software (version 19.0; SPSS, Inc., Chicago, IL). A two-tailed p value of <0.05 was regarded as significant.

3. Results

3.1. Baseline characteristics of the DAA-treated CHC patients versus SC participants

The study group included 40 CHC participants (13 men and 27 women, median age: 65 years, range 41–87 years) with equal number of cases of HCV genotypes 1 and 2 in the control group [14 SC participants (3 men and 11 women, median age: 65 years, range 45–74 years)]. Baseline characteristics of the DAA-treated CHC and SC participants are shown in the Supplementary Table 1. In addition to HCV viremia and abnormal liver function test results (ALT, AST, γ-GT, albumin and FIB-4 score), the DAA-treated CHC group showed distinctive glycemic and lipid parameters, including for insulin, HOMA-IR, HOMA-beta and TG, compared to the SC group (Supplementary Table 1). The presence of diabetes, hypertension and chronic kidney disease was comparable between the two groups.

3.2. Dynamic changes in circulating markers for atherogenic risks and oxidative stress

We looked at plasma levels of oxLDL, 8-OHdG, MDA and IMA to assess the oxidative stress in CHC individuals following DAA-related viremia clearance and in the SC ones. In the SC group, plasma levels were comparable for oxLDL, 8-OHdG, MDA and IMA measured at 1-year intervals over two years from either SC-BL to SC-1Y or SC-1Y to SC-2Y. Only at the 2-year follow-up was 8-OHdG increased from SC-BL to SC-2Y by approximately 10% (p = 0.0409), but levels of oxLDL, MDA and IMA were unchanged (Fig. 1a–d). In the DAA-treated group, percentage changes compared to baseline showed a decrease in 8-OHdG at SVR by 10% (p = 0.0255) which was maintained for 2 years until SVR-2Y. Moreover, the oxLDL level was unchanged at SVR but decreased from SVR to SVR-2Y by 30% (p < 0.0001), while MDA (p = 0.0055) increased by 10% at SVR-2Y and IMA remained constant (Fig. 2a–d).

Fig. 1.

Percentage changes in oxidative stress markers in the spontaneous clearance (SC) controls over two years.

Plasma levels of (a) 8-hydroxy-2′-deoxyguanosine (8-OHdG) (b) oxidized low-density-lipoprotein (oxLDL) (c) malondialdehyde (MDA), and (d) ischemia-modified albumin (IMA) were quantified in SC participants (n = 14) at baseline (SC-BL) and at 1 year (SC–1Y) and 2 year (SC–2Y) follow-up. Data is shown as a percentage relative to baseline, expressed as the median with range and compared by the paired t-test between each two time-points.

Fig. 2.

Percentage changes in oxidative stress markers in direct-acting antiviral (DAA)-treated chronic hepatitis C (CHC) patients after sustained virological response (SVR) for two years.

The plasma levels of (a) 8-OHdG, (b) oxLDL, (c) MDA, and (d) IMA were quantified in CHC individuals (n = 40) receiving DAA treatment at BL, SVR, SVR-1Y, and SVR-2Y. Data is shown as a percentage relative to baseline, expressed as the median with range and compared by the paired t-test between each two time-points. SVR: sustained virological response. SVR-1Y: SVR at 1 year; SVR-2Y: SVR at 2 years.

3.3. Reversibility of oxLDL in DAA-treated participants after HCV clearance for two years

The DAA-treated CHC participants had cleared HCV viremia at SVR, which was defined as the starting time point for the HCV spontaneous clearance group at baseline (SC-BL). Comparisons showed higher levels of oxLDL (p < 0.0001, Fig. 3a), 8-OHdG (p < 0.05, Fig. 3b) and IMA (p < 0.01, Fig. 3c) but not of MDA (Fig. 3d) in the DAA group. Taking into consideration the combined 1- and 2-year intervals by the linear mixed-model, results revealed significant differences in oxLDL (p < 0.0001, Fig. 3a) and IMA (p = 0.0017, Fig. 3c) levels between the DAA and SC groups. Results for oxLDL further revealed comparable levels between SVR-2Y and SC-2Y (Fig. 3a), suggesting that excess ROS might affect the circulating oxLDL levels which displayed reversibility after 2 years of HCV viral clearance by DAAs.

Fig. 3.

Comparisons of the dynamics of oxidative stress markers in direct-acting antiviral (DAA)-treated CHC individuals and spontaneous clearance (SC) controls.

The dynamics of plasma (a) oxLDL, (b) 8-OHdG, (c) IMA, and (d) MDA in SC (n = 14) and DAA-treated CHC individuals (n = 40) was evaluated by linear mixed-model analysis. Levels are expressed as the mean ± SEM and compared by the Mann-Whitney U test between the DAA-treated and SC groups, *, p < 0.05; **, p < 0.01; ****, p < 0.0001. SVR: sustained virological response. SVR-1Y: SVR at 1 year; SVR-2Y: SVR at 2 years.

3.4. Correlations between liver stiffness and oxidative stress markers in DAA-treated CHC individuals

To confirm the long-term effects of DAAs on oxLDL modulation after SVR, the second group consisting of DAA-treated 132 CHC individuals (Supplementary Table 2) was further studied. The results confirmed that oxLDL levels gradually declined during the 2-year interval post-SVR following DAA treatment (Fig. 4a and b).

Fig. 4.

Confirmation of the reversibility of plasma oxidized low-density lipoprotein (oxLDL) in the second group of direct-acting antiviral (DAA)-treated chronic hepatitis C (CHC) individuals.

(a) Plasma oxLDL levels in the second CHC group (n = 132) were examined at baseline, SVR, SVR-1Y, and SVR-2Y. Data is shown as a percentage relative to baseline, expressed as the median with range and compared by the paired t-test between each two time points. (b) The dynamics of plasma oxLDL in SC (n = 14) and DAA-treated CHC individuals (n = 132) was evaluated by linear mixed-model analysis. Levels are expressed as the mean ± SEM and were compared by the Mann-Whitney U test between the DAA-treated and SC groups, *, p < 0.05; **, p < 0.01; ****, p < 0.0001. SVR: sustained virological response. SVR-1Y: SVR at 1 year; SVR-2Y: SVR at 2 years; SC-BL: spontaneous clearance baseline.

We next assessed, the therapeutic effect of DAA on liver fibrotic progression using liver stiffness measurements (LSM). The LSM was compared with the one at baseline, revealing declines in LSM values at SVR and further declines at SVR-1Y and SVR-2Y (Fig. 5). Furthermore, the relationship between LSM and oxLDL when examined, showed positive correlations at SVR (rho = 0.208, p = 0.017) and SVR-1Y (rho = 0.274, p = 0.002) (Table 1).

Fig. 5.

Improvement of liver stiffness measurements (LSM) in direct-acting antiviral (DAA)-treated chronic hepatitis C (CHC) individuals.

The LSM score of CHC individuals (n = 132) was determined and shown as a percentage relative to baseline, expressed as the median with range and compared by the paired t-test between each of the two time-points. LSM: liver stiffness measurement. SVR: sustained virological response; SVR-1Y: SVR at 1 year; SVR-2Y: SVR at 2 years.

Table 1.

Correlations of liver fibrosis and oxidative status of direct acting antiviral cured chronic hepatitis C individuals at different time-points.

	oxLDL (mU/mL) vs. LSM (kPa)
	Baseline	SVR	SVR1Y	SVR2Y
All recruitment (n = 132)	0.124 (0.174)	0.208 (0.017*)	0.274 (0.002*)	0.012 (0.896)

Data was expressed as Spearman's correlation coefficient (p value). Sustained virological response (SVR). SVR-1Y: SVR at 1 year; SVR-2Y: SVR at 2 years.

4. Discussion

Physiological ROS generation occurs during redox reactions, which produces essential cellular signaling¹⁶; however, many viral infections that modulate ROS metabolism may cause oxidative stress, due to a high oxidative state, and boost a vicious cycle of inflammatory and metabolic distress.¹⁷ In CHC individuals, excess ROS represents one of the major contributing factors towards HCV pathogenesis, including steatosis, fibrosis, cirrhosis, and hepatocellular carcinoma.¹⁸ ROS are difficult to detect as free radicals because of their very short lifespan and low nanomolar concentrations, when considering superoxide anions (O2˙−), hydroxyl radicals (·OH), or hydrogen peroxide (H₂O₂). Since ROS actively react with DNA, lipids and proteins to produce end-products, these biomarkers can be quantitatively applied to assess the degree of oxidative stress. Accordingly, we examined the alterations of DNA-, lipid- and protein-ROS biomarkers in plasma samples of DAA-treated CHC individuals post-SVR during the 2-year follow-up period post-SVR, including 8-OHdG, oxLDL, MDA, and IMA. Results showed declines in 8-OHdG at SVR and oxLDL at SVR-2Y, a slight increase in MDA at SVR-2Y and a consistent level of IMA, indicating the decoupled modulation of various oxidative molecules during the post-therapeutic period. Notably, recovery of oxLDL did not occur until SVR-2Y, suggesting that ROS overproduction might last for several years following HCV clearance in SVR patients receiving DAA treatment, and that plasma level of oxLDL might be a marker that provides useful information during long-term follow-up.

Accumulation of ROS directly induces inflammation in hepatocytes, in concert with hepatic stellate cells and fibroblasts, to produce deposits of extracellular matrix during the development of hepatic fibrosis.¹⁹ Previously, DAA-treated CHC individuals have shown a reduction in liver stiffness,^20,21 which was confirmed in CHC patients in the current study, as evidenced by a serial decrease in LSM at SVR (p < 0.001), SVR-1Y (p < 0.001) and SVR-2Y (p = 0.001) (Fig. 2) in comparison with baseline. This finding implies that HCV infection could lead to a direct relationship between liver stiffness and LDL oxidation in CHC individuals and that reducing ROS affects pro-fibrogenesis. Therefore, CHC individuals may show an improvement in their post-treatment levels of blood oxidative stress parameters and liver fibrogenesis upon HCV eradication.

Although both oxLDL and MDA represent oxidized lipids that play a role in non-alcoholic fatty liver disease and atherosclerosis²², only the former responded to the excessive exposure of free radicals in CHC individuals and was shown to be an elevated blood biomarker prior to treatment as compared to the SC group (Fig. 3a), (Fig. 3d). OxLDL originates from circulating LDL particles that have undergone peroxidation or obtained oxidative products generated at other locations, while the formation of oxLDL might involve nonenzymatic modifications, such as metal ions, ROS, and enzymatic modifications, such as lipoxygenase and myeloperoxidase.^23,24 Accordingly, the consequences of the alterations between prooxidant and antioxidant homeostasis contribute to the increasing levels of blood oxLDL, which has been helpful for the prediction of cardiovascular disease in the past decades.²⁵ In addition to serving as a high-risk biomarker, oxLDL plays a crucial role in atherogenesis by directly inducing inflammation through interaction with the vascular wall and macrophages.^26,27 Similarly, accumulation of oxLDL in the wall of the hepatic portal vein may trigger inflammation and promote hepatic fibrosis²⁸, in support of the positive correlations between liver stiffness and oxLDL in our CHC group (Table 1). Taken together, cure of chronic HCV infection by DAAs can confer viral eradication but there may still remain some metabolic abnormalities. Nevertheless, the reversibility of oxLDL by DAA treatment might benefit CHC individuals by preventing liver fibrosis progression and cardiovascular disease.

In conclusion,we have shown that DAA treatment reduces circulating oxLDL in CHC individuals 2 years post--SVR, which may indicate prolonged beneficial effects in terms of liver disease progression and cardiovascular risk after achieving SVR in this population.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.