Inflammatory and cardiovascular diseases biomarkers in chronic hepatitis C virus infection: A review

Abstract Hepatitis C virus (HCV) infects 180 million people worldwide and over 4 million people in the United States. HCV infection is a major cause of chronic liver disease and is recognized as a risk factor for clinical cardiovascular disease (CVD). Many studies have shown increased prevalence of cardiac and inflammatory biomarkers in patients with chronic HCV infection (CHC), and though these markers may be used to risk stratify people for cardiac disease in the general population their role in the HCV population is unknown. Patients with CHC have elevated cardiac and inflammatory biomarkers compared to noninfected controls which may play a role in CVD risk stratification. We undertook a systematic review of inflammatory and cardiac biomarkers in people with HCV infection with a focus on the effect of CHC on serum levels of these markers and their utility as predictors of CVD in this population. Medline, EMBASE, and Cochrane databases were searched for relevant articles until June 2019. A total of 2430 results were reviewed with 115 studies included. Our review revealed that HCV infection significantly alters serum levels of markers of inflammation, endothelial function, and cardiac dysfunction prior to HCV treatment, and some of which may change in response to HCV therapy. Current risk stratification tools for development of CVD in the general population may not account for the increased inflammatory markers that appear to be elevated among HCV‐infected patients contributing to increased CVD risk.


| INTRODUCTION
Hepatitis C virus (HCV) is a single stranded RNA virus belonging to the Flaviviridae family. HCV has a global prevalence of 2.5% and infects 180 million people worldwide. 1 In the United States, it is the most common blood borne infection affecting 0.8 persons in every 100 000 and causing significant morbidity and mortality. 2 In 2016, over 2 million Americans had an opioid use disorder with 10% to 20% of those escalating to injection drug use. 3 In this setting the prevalence of HCV has dramatically increased, especially in younger patients, with injection drug use (IDU) now being the primary mode of transmission in the US and a 2-fold increase in the incidence rate of acute HCV infection. 4 Chronic viral diseases, in particular human immunodeficiency virus (HIV), have been strongly linked to the development of clinical cardiac diseases. 5,6 Chronic HCV infection (CHC) has been linked to subclinical and clinical cardiovascular diseases (CVD), such as myocardial infarctions, congestive heart failure, cerebrovascular accident (CVA), and peripheral arterial disease, 7 and proposed mechanisms include chronic inflammation and immune activation driven by HCV infection as well as direct endothelial invasion and dysfunction. Studies have shown increased prevalence of certain cardiac biomarkers associated with increased CVD risk in patients with CHC compared to age-matched uninfected patients. 8 Some of these biomarkers are combined with traditional risk factors to risk stratify persons for cardiac disease in the general population, 9 but their utility in setting of HCV infection is unknown. Therefore, we aim to review the current literature on inflammatory and cardiac biomarkers in patients with HCV infection with a focus on the effect of CHC on serum levels of these markers and their utility as predictors of CVD in this population.

| MATERIALS AND METHODS
We conducted a search based on PRISMA guidelines 10 on Medline, EMBASE, and Cochrane for English language articles published through 14 June, 2019 using the following keywords and mesh terms: Hepatitis C, HCV, hepacivirus, chronic hepatitis C, inflammatory biomarkers, biomarkers and inflammation, biomarkers and inflammation mediators, cardiac biomarker, C-reactive protein, CRP, high sensitivity C-reactive protein, hsCRP, interluekin-6, tumor necrosis factor-alpha, troponin T, troponin I, brain natriuretic peptide, BNP, pro B-type natriuretic peptide, N-terminal pro b-type natriuretic peptide, NT-proBNP cardiovascular diseases, coronary disease, heart failure. All abstracts with the following inclusion criteria were reviewed: human studies in adults with CHC investigating serum levels of biomarkers of interest that included a HCV negative control group, and had full articles available for review. Study designs included randomized clinical trials, prospective, and retrospective observational cohorts, case-control studies, cross-sectional studies, and systemic reviews. Studies of acute hepatitis C and studies in children were excluded. Full-length articles were reviewed by three independent reviewers (A.B., S.M., and M.H.), and any differences in reviewed data from articles were discussed and resolved by these reviewers and S.B. who reviewed these selected articles.
Cardiac biomarkers were categorized into three main groups: biomarkers of inflammation, biomarkers of endothelial function, and biomarkers of cardiac dysfunction.

| RESULTS
The search performed in June 2019 yielded 1255 results on Medline, 1100 results on EMBASE, and 75 results on Cochrane giving a total of 2430 results. After duplicates were removed, 2394 references remained for review. Twenty-four additional articles were added after performing ancestry and bibliography searches of all relevant articles, meta-analyses, and systematic reviews. On review of titles and abstracts, 2156 were removed as they were found to be not relevant to our review. Of the remaining 262, 147 were removed because they did not meet eligibility criteria: 66 lacked a seronegative control group, 39 did not examine biomarkers of interest, 27 described only expression of and not serum levels of biomarkers, eight had relevant data missing, four were in a pediatric population, and three described levels of biomarkers following stimulation. Ultimately, 115 studies were included in our review ( Figure S1).

| Biomarkers of inflammation
Biomarkers of inflammation are commonly used to assess CVD risk in the general population. Fifty-six studies evaluating the effect of HCV infection on inflammatory biomarkers were reviewed (Table 1 and   Table S1) Biomarkers reviewed included interleukin-6 (IL-6), interleukin-10 (IL-10), tumor necrosis factor-alpha (TNF-α), TNF-α receptors (TNFR), soluble CD163 (sCD163), and soluble CD14 (sCD14). IL-10 is an anti-inflammatory cytokine that regulates the production of proinflammatory cytokines 11 and down regulates the expression of adhesion molecules, 12 and through these mechanisms may have anti-atherosclerotic properties. 13 The imbalance between anti-and pro-inflammatory cytokines is thought to be critical in the pathogenesis of plaque formation and destabilization, though the results of clinical studies in angina patients remain inconclusive. 14,15 The same imbalance resulting in increased IL-10 levels may be central to the persistence of HCV infection in CHC. 16 IL-6 is a proinflammatory cytokine that promotes activation and proliferation of lymphocytes and induction of hepatic acute phase proteins. 17 Like IL-10, IL-6 has been linked to the development of atherosclerosis and serum levels have been correlated with cardiovascular disease and mortality. 18 TNF-α is a proinflammatory cytokine secreted by activated monocytes and macrophages in response to various infections. TNF-α stimulates the release of acute phase proteins in the liver leading to lymphocyte and endothelial activation, 19 and exerts its action through the binding of cellular TNF-α receptor-1 (TNFR-1) and TNF-R2. 20 In healthy subjects, increased serum levels of TNF-alpha have predicted CVD risk, 21 and represented an independent risk factor for reduced event-free survival. 21 TNF-dependent processes are up regulated and activated in CHC, and TNF-α mRNA is ubiquitously expressed in hepatocytes, Kupffer cells, and infiltrating mononuclear cells at higher levels in CHC patients than in healthy controls. 22 31 Falasca et al 32 Helaly et al 33 Khan et al 34 Migita et al 36 Zekri et al 37 Malaguarnera et al 38 Costantini et al 39 Capone et al 40 Oyanagi et al 41 Lapinski et al 42 Lecube et al 43 Sandler et al 44a Grungreiff et al 46a Cotler et al 48a Hung et al 49 Antonelli et al 50 Shive et al 52 Cua et al 53 Grungreiff et al 54 Lee et al 84 Mishra et al 87 Sousa et al 91 Tsui et al 35 Zuwała-Jagiełło et al 57 Mourtzikou et al 58 Han et al 85 31 Falasca et al 32 Helaly et al 33 Khan et al 34 Tsui et al 35 Zekri et al 37 Lecube et al 43 Hung et al 49 Cua et al 53 Zuwała-Jagiełło et al 57 Mourtzikou et al 58 Akcam et al 59  C-reactive protein (CRP) is an acute phase protein produced predominantly by hepatocytes, and influenced by IL-6 and TNF-α. 24 Studies have shown significant association between increased CRP levels and underlying atherosclerosis, the risk of recurrent cardiovascular events among patients with established disease, and the incidence of first cardiovascular events among individuals at risk for atherosclerosis. CRP has been validated as one of the tools for assessment of CVD risk in the general population, 25 and shown to correlate with survival and mortality in both non-CHC 26 and cirrhotic patients. 27 Furthermore, monocyte/macrophage activation markers of immune activation such as soluble CD163 (sCD163) 28 and soluble CD14 (sCD14) 29  In a cohort of 981 patients with coronary heart disease enrolled in the Heart and Soul study, HCV seropositivity was associated with changes in levels of CRP, TNF-α, IL-6 increased Framingham risk scores, 31 and hospitalizations due to clinical cardiac failure and death. 35 However, other studies have reported different findings. These studies were smaller in size, 92 and though unable to demonstrate differences between HCV groups and healthy controls they demonstrated increased serum levels of inflammatory markers with advanced HCV liver disease, 91,93 suggesting an evolution of inflammatory changes and cytokine imbalance in CHC as liver disease progresses. 94 Other studies have also demonstrated increased hepatic expression and serum levels of IL-10, 80 35 Adinolfi et al demonstrated increased pro-inflammatory cytokines levels were associated with a significantly higher prevalence of carotid atherosclerosis in HCV-infected patients compared to controls. 100 Similarly, Alyan et al found coronary artery disease (CAD) severity scores were significantly higher among CHC patients than among HCV negative controls. In the study, both HCV, CRP, and fibrinogen were significantly correlated to severity of CAD. 102 Finally, many studies investigated the effect of HCV therapy on inflammatory markers and the majority demonstrated that HCV therapy significantly altered levels of inflammatory levels and/or predicted treatment success. 22,[46][47][48]62,65,70,71,[78][79][80]89,96,105 Taken the together, the preponderance of data shows a clear modulation of the immune system, an imbalance of pro-and antiinflammatory biomarkers with a shift towards a proinflammatory state, and increased serum inflammatory levels among CHC-infected patients. However, possibly due to blunted hepatocyte response and decreased production of CRP, CHC patients may score lower on CVD risk assessment models which rely heavily on CRP values.

| Biomarkers of endothelial function
One of the early sentinel events in the development of atherosclerosis is endothelial dysfunction which results in an increased interaction between circulating leukocytes and the endothelium. 106 This increased interaction and recruitment of circulating leukocytes is mediated by cellular adhesion molecules, and circulating forms of these molecules such as vascular cell adhesion molecule-1 (VCAM-1), endothelial-leukocyte adhesion molecule-1 (E-selectin), intercellular adhesion molecule-1 (ICAM-1), and soluble selectin (sP-selectin, sEselectin) have been found to predict endothelial dysfunction variably. 107 Elevated levels of these markers have been correlated with increased risk of cardiovascular mortality in numerous studies in the general population. 8,[107][108][109][110][111][112][113] In CHC patients, these adhesion molecules are expressed on sinusoidal cells and may be under the regulatory control of TNF-alpha. 63,114 Six studies investigated the serum levels of markers of endothelial function among HCV patients (  47 More data is needed to determine whether these markers may be useful in the CVD risk assessment in the HCV patient population, as they are in the general population.

| Biomarkers of cardiac dysfunction
Brain natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP) are natriuretic hormones that are primarily released by the ventricles of the heart. Plasma BNP provides prognostic information in patients with chronic heart failure and in those with asymptomatic or symptomatic left ventricular dysfunction. 115,116 Plasma NT-proBNP has been shown to independently predict long-term risk of death due to congestive heart failure. 117,118 Both markers have been established as reliable diagnostic and prognostic cardiac biomarkers that correlate with both symptoms of CHF and the severity of systolic and diastolic CHF in the general population. 119,120 In addition, these markers have been found to correlate with the degree of circulatory dysfunction in cirrhotic patients. 121 Troponin T (TnT) and Troponin I (TnI) are cardiac proteins which regulate the calcium mediated interactions between actin and myosin 122 released into the serum after myocardial injury. They have demonstrated prognostic value in a wide array of CVD, especially those associated with ischemic myocardial injury. 123,124 Nine studies evaluated serum levels of these biomarkers in CHCinfected patients. (Table 3 and Table S3) Among CHC patients, elevated levels of BNP, NT-proBNP, 50,51,67,99,[125][126][127][128][129] TnT, and TnI 103,128 have been observed compared to healthy controls. ( Table 4 and Table S4) The majority of studies have found that HCV co-infection further increases the serum levels of IL-6, 133-138 IL-10, 139 and sTNFRI 138 but decreases levels of CRP or hsCRP 133,134,136,[140][141][142] irrespective of liver function. 133 Shah et al found that CRP levels fell with increasing IL-6 levels suggesting attenuation of the CRP-related IL-6 response. 134 Similar to HCV mono-infected populations, these markers have been found to increase with progression of HCV liver  50 Antonelli et al 67 Che et al 99 Antonelli et al 125 Okada et al 126 Wang et al 127 Matsumori et al 128 Che However, there were no studies that evaluated the change in these inflammatory markers with HCV therapy among co-infected patients.

Forrester et al assessed lipid profiles and CRP levels in HCV mono-
infected, HIV/HCV co-infected, and healthy controls and found no significant differences in CRP or lipid levels among the different groups. 144 Their results conflicted with previous studies that found decreased lipid levels in co-infected patients. 142 No studies evaluated the change in these inflammatory markers with subclinical and/or clinical CVD among HIV/HCV co-infected patients. among co-infected patients, which correlated with disease progression 145 and treatment response, 145 decreased with therapy, 145,147 predicted liver related mortality 143 and were significantly associated with subclinical cardiovascular disease. 148 Finally, one study reported significantly elevated BNP elevated levels among HIV/HCV coinfected patients compared to HIV mono-infected patients. 152 Therefore, the totality of the data suggests that while HIV monoinfection has a significant influence on serum cardiovascular biomarkers, HCV co-infection may amplify this effect and further increase CVD risk. In addition, it may be possible to identify this increased risk in the HIV/HCV co-infected population through measurement of select inflammatory and endothelial biomarkers.

| CHC patients with other inflammatory comorbidities
The additive effect of comorbidities such as diabetes mellitus 153 and end-stage renal disease (ESRD) on the levels of inflammatory biomarkers may be important as well, since these markers may be elevated de novo in these specific patient populations. 154 (Table 5 and Table S5).

TNF-α and IL-6 have been elevated in ESRD patients with and
without HCV compared to healthy controls. 55  Increased levels of IL-6, IL-10, TNF-α, and sTNFR in HCV-infected patients with complications such as cryoglobulinemia, Sjogren syndrome, lymphoproliferative diseases, hemophilia, and HCV related arthritis have been reported compared to CHC-infected patients without such complications or to healthy controls. 45,51,56,[164][165][166][167][168][169][170] Therefore, special consideration may be needed when interpreting serum inflammatory markers in HCV populations who have DM, ESRD, and/or concomitant immune complications. In these populations with HCV infection and inflammatory conditions, there we no studies evaluating the change of these markers with HCV treatment, and none evaluated the association of these markers to subclinical and/or clinical CVD.

| DISCUSSION
We found that the preponderance of data suggested that HCV infec- In addition, lipid levels, which are a component of many CVD risk stratification tools are known to be decreased in CHC patients, 173  HCV-infected and HCV uninfected patients, 100,174 and there is a growing concern that CVD risk assessment using current risk scoring tools may be suboptimal for patients with heightened inflammatory states leading to an underestimation of risk in these patients. 175 CRP is an established cardiac biomarker for increased CVD risk and has been included in traditional risk assessment algorithms. Treatment strategies based on hsCRP levels have resulted in reduced CVD events, 176 and hsCRP levels are used to re-classify patients with intermediate ASCVD risk currently. 25 However, CRP levels were decreased in HCV patients compared to uninfected patients in many 35,98,99,157,163 but not all studies. 34  In conclusion, inflammatory pathways are fundamental to the pathogenesis of atherosclerosis and the development of cardiac events, and CHC infection has been shown to heighten the state of inflammation in these patients. Current risk stratification tools for development of CVD in the general population do not account for many of the inflammatory markers that appear to be elevated among HCV-infected patients, and therefore there is concern that the CVD risk in these patients may be underestimated. Given the burden of HCV infection both nationally and internationally, there is an urgent need to evaluate additional cardiac biomarkers that may potentially identify or better discriminate CVD risk among HCV-infected patients more accurately than those in the current standard risk assessment tools. Further prospective studies are needed to confirm the predictive value of these biomarkers of interest in this patient population.

ACKNOWLEDGMENTS
Authors would like to thank Research Librarian Jill E. Foust for her assistance with conducting the literature search.