Non-Invasive Biomarkers for Hepatic Fibrosis

With great advancements in the therapeutic modalities used for the treatment of chronic liver diseases, the accurate assessment of liver fibrosis is a vital need for successful individualized management of disease activity in patients. The lack of accurate, reproducible and easily applied methods for fibrosis assessment has been the major limitation in both the clinical management and for research in liver diseases. However, the problem of the development of biomarkers capable of non-invasive staging of fibrosis in the liver is difficult due to the fact that the process of fibrogenesis is a component of the normal healing response to injury, invasion by pathogens, and many other etiologic factors. Current non-invasive methods range from serum biomarker assays to advanced imaging techniques such as transient elastography and magnetic resonance imaging (MRI). This review provides a systematic overview of these techniques, as well as both direct and indirect biomarkers based approaches used to stage fibrosis and covers recent developments in this rapidly advancing area. INTRODUCTION Liver fibrosis is defined as the building up of excessive amount of extracellular matrix, also known as scar tissue, in the liver parenchyma. While reviewing fibrosis as a component of the pathogenesis of a disease, it is important to remember that the process of fibrogenesis is also a component of the normal healing response to injury, invasion by pathogens, and many other etiologic factors. In the liver, this healing process normally involves the recruitment of immune and/or inflammatory cells to the site of injury, in order to counteract the effects of a damaging agent, secretion of extracellular matrix proteins, reorganization of the extracellular matrix and possible regeneration of new hepatic tissue. However, when the damage to the liver is chronic, excess fibrous connective tissue accumulates due to an imbalance between the production and dissolution of the matrix proteins. The process of excessive fibrogenesis taking place over an extensive period of time eventually distorts the normal parenchymal structure of the liver and impairs its normal hepatic function, leading to different stages of fibrosis and cirrhosis. As chronic liver disease progress, hepatic fibrosis is accompanied by the formation of septae and nodules that intervene with the blood flow from the portal area of liver, producing portal hypertension and an abnormal angio-architecture which is a distinctive feature of cirrhosis. As previously noted, liver fibrosis is generally accompanied by stress or injury to the liver parenchyma. This stress is exemplified by subsequent activation of the immune system accompanied by increased levels of certain cytokines and growth factors, which are the primary players in the development of fibrogenesis. It is important to realize that recent evidence suggest that liver fibrosis and even early cirrhosis can be reversible [1, 2], and this reversal is probably related to the suppression of the fibrotic response. THE BIOLOGY OF LIVER FIBROSIS: The most important cellular player in the production of the extracellular matrix is the myofibroblast. A wide array of cells of different origins can be converted into myofibroblasts and contribute to the production of fibrous tissue (Figure 1), including portal myofibroblasts and bone marrow-derived mesenchymal stem cells. Some epithelial cells including hepatocytes and biliary epithelial cells (cholangiocytes) can be activated to function as myofibroblasts through the process of Epithelial-Mesenchymal Transition (EMT) [3]. However, the predominant liver cell types that differentiate into activated myofibroblasts are quiescent hepatic stellate cells (HSC), also known as Ito cells or perisinusoidal cells. HSCs are located in the space of Disse and are the storage site for Vitamin A or retinoid [4]. The process of fibrogenesis is triggered as a response to damage to the liver by hepatotoxic substances and involve a variety of non-HSC types of cells. For example, hepatocytes can respond to this damage in multiple ways, including production of reactive oxygen species (ROS) and apoptosis, while the resident liver macrophages called Kupffer cells elicit a massive immune response resulting in the recruitment of other inflammatory cells to the site of injury [5]. The recruitment of additional inflammatory cells is usually achieved via the adhesive interaction between the selectin molecules in the endothelium of the blood vessels and the leukocytes. Attracted to the chemokines produced by the Kupffer cells, the leukocytes exit out of the vasculature towards the injury site and contribute to the release of additional pro-inflammatory and pro-fibrotic mediators, including cytokines such as tumor necrosis factor alpha (TNF-α) and various interleukins. Reactive oxygen and nitrogen species, proteases, and lipid metabolites such as prostaglandins and thromboxane are also released [6]. As a result of this response, quiescent HSCs lose their retinoid function and are converted to activated myofibroblasts [7]. These cells, in turn, contribute to the chemotaxis of leukocytes as well as their own chemotaxis through the production of chemokines and cytokines such as monocyte chemotactic protein-1 (MCP-1) [8]. As a result, activated HSCs start expressing the Platelet Derived Growth Factor (PDGF) receptor and Transforming Growth Factor (TGF) receptor. TGF-β is the central mediator of fibrogenesis, while PGDF stimulates proliferation of the HSCs. Importantly, activation of HSCs is associated with a gradual replacement of the basement membrane-like extracellular matrix (ECM) within the space of Disse by the collagen rich fibers [7] and the production of fibrous bands [8]. In advanced stages of fibrosis, the liver contains approximately six times more ECM components than normal, including collagens (I, III, and IV), fibronectin, undulin, elastin, laminin, hyaluronan, and proteoglycans [8]. The distribution of the fibrous material in the liver depends on the nature of the injury [8]. In chronic viral hepatitis and chronic cholestatic disorders, the fibrotic tissue is initially located around portal tracts, while in alcohol-induced liver disease (ALD) it first appears in the pericentral and perisinusoidal areas [9]. The portal area of the liver is centered at the hepatic artery, bile duct, and the portal vein embedded in the connective tissue matrix that normally contains a number of the fibroblasts surrounding the bile duct basal membrane. When liver injury is initiated at the portal field, the first layer of hepatocytes are liable to immediate destruction leading to an enlargement of the portal field and rapid activation of the portal fibroblasts [10]. Availability of the portal fibroblasts for myofibroblastic conversion makes the portal area especially susceptible to fibrotic changes. TYPES OF LIVER FIBROSIS: Liver Fibrosis can be subdivided into two categories: Congenital Liver Fibrosis and Acquired Liver Fibrosis. Congenital Hepatic Fibrosis (CHF) is a rare type of liver disease, usually a developmental disorder of the porto-biliary system, specifically, malformation of the ductal plate with an excess number of immature embryonic duct structures, leading to periportal fibrosis as well as cholangitis, portal hypertension and associated complications [11]. Additionally, CHF is associated with ciliopathies or disorders of the primary cilia [12]. In these conditions, the degree of the hepatic fibrosis may vary, but the main cause of morbidity and mortality remains the failure of other organ systems [11]. Acquired fibrosis may result from the action of a number of pathogenic factors and toxic exposures such as long-term excessive alcohol consumption, cholestasis, autoimmune liver diseases, iron or copper overload, chronic viral hepatitis, the presence of non-alcoholic fatty liver disease (NAFLD) and other etiologic factors. These factors may work separately or in combination with each other to produce cumulative effects. This review will concentrate on acquired liver fibrosis and biomarkers that are being developed to quantify and stage it. CAUSES OF ACQUIRED LIVER FIBROSIS: The process of hepatic fibrogenesis is an example of the universal type of cellular response to a broad spectrum of chronic liver injury. The main etiologic factors responsible for liver fibrosis are chronic viral hepatitis, metabolic syndrome-related fatty liver disease, excessive alcohol consumption, and autoimmune liver diseases (including cholestatic liver disease) as well as iron and copper over load. All of these etiologic pathways can produce mediators eliciting the inflammatory response and, eventually, initiating fibrogenesis. In the following paragraphs, we will discuss a few of the most common causes of hepatic fibrosis. Alcoholic Liver Disease: Excessive and chronic alcohol consumption is an important causal factor of liver fibrosis and cirrhosis. This is primarily due to the fact that the hepatocytes are the primary site for alcohol metabolism. The process of the breakdown of ethanol produces two profibrotic agents, acetaldehyde and reactive oxygen species (ROS). Acetaldehyde can directly upregulate the transcription of collagen I [13]. It can also indirectly contribute to the process of fibrogenesis by upregulating the synthesis of transforming growth factor-beta 1 (TGF-β1). However, the hepatic stellate cells do not express alcohol dehydrogenase enzyme, therefore, it is likely that damaging alcohol derivative acetaldehyde originates in hepatocytes and enters collagen and TGF-β1-producing HSCs from an outside [14]. Similarly, ROS generated during alcohol metabolism in hepatocytes can also be taken up by HSCs to activate collagen production [15]. Chronic ethanol exposure sensitizes HSCs to various pro-inflammatory factors and elicits the production of inflammatory mediators that contribute to the fibrotic changes in the liver in fashion similar to that described for fibrogenesis in other types of chronic liver diseases. Non-alcoholic Fatty Liver Diseases (NAFLD): From the spectrum of NAFLD, non-alcoholic steatohepatitis (NASH) can often be accompanied by liver fibrosis. NAFLD and its subtype of NASH are usually seen in individuals with metabolic syndrome (MS) or its components such as obesity, type-2 diabetes (DM), dyslipidemia, and insulin resistance. In general, development of NASH is augmented by the production of mediators such as free fatty acids, high glucose, and adipocytokines. Furthermore, these factors are also known to be involved in the development of hepatic fibrosis. To date, the pathogenesis of NASH-related liver fibrosis is not entirely well understood [8]. Evidence provided by numerous studies links obesity, insulin resistance and the progression of fibrosis together in one vicious circle [16]. For example, the well-known adipokine, leptin is produced proportionally to the mass of the visceral adipose compartment. It has previously been shown that leptin augments fibrogenesis by stimulating phagocytic activity and cytokine secretion by Kupffer cells and macrophages [17] as well as the proliferative and ROS generating activities of the endothelial cells [18]. Another adipokine, resistin, exerts proinflammatory actions in HSC by increasing the expression of both MCP-1 and interleukin-8 (IL8) as well as activating the transcription factor, NFkB [19]. From examples mentioned above, one can derived that initial stages of the pathogenesis of liver fibrosis associated with NAFLD depends primarily on the soluble factors produced by excessive visceral adipose and on an skewed distribution of the soluble fat particles in the bloodstream. Cholestatic Liver Diseases: Cholestasis (reduced bile duct excretion) is another well-known cause of liver fibrosis. Cholestasis triggers the proliferation of the cholangiocyte lining of the intrahepatic and extrahepatic bile duct systems through a complex regulatory milieu that involves both autocrine and paracrine factors [20]. The activation of biliary proliferation is known as ductular reaction. Proliferating bile duct epithelial cells produce the profibrogenic connective tissue growth factor (CTGF) that stimulates myofibroblast generation through EMT and collagen deposition [21]. The primary players in the fibrotic reaction to cholestasis are inflammatory response propagated by neutrophils and resulting from that oxidative stress. Furthermore, inhibition of ROS during cholestasis reduces fibrosis. Chronic Viral Hepatitis: Chronic viral infections such as hepatitis B (HBV) or hepatitis C (HCV) viruses pose an important risk for the development of liver fibrosis. The process of fibrogenesis related to chronic viral infection follows the same path as fibrosis related to other etiologic factors. Nevertheless, this process is compounded by additional virus-specific factors. In chronic HCV infection, the viral core, NS5 and NS3 proteins have been demonstrated to initiate a cascade of molecular events that can eventually lead to fibrosis. HCV proteins appear to affect both lipid accumulation and degradation, with the consequent disruption of the normal process of lipid compartmentalization and metabolism, skewing towards ROS production. In the case of HBV infection, studies have shown that the X protein of HBV (HBx) directly induces TGF-β secretion by hepatocytes and, thus, contributes to the paracrine activation of HSC's [22]. Interestingly, the superinfection of hepatitis delta virus (HDV) in patients who are chronic carriers of HBV, can also accelerate the progression of fibrosis [7]. The deleterious effect of this type of viral infection rests on the fact that, despite the massive inflammatory response in the liver, the viral particles cannot be cleared. However, the drastic rise in the release of cytokines, particularly, TNF-α, IL-1-α and -β, IL-2, IL-6, and IL-8, is strong enough to activate the myofibroblasts and induce fibrogenesis [23]. Both HIV-HBV and HIV-HCV coinfected patients are at increased risk for progression of their liver disease as compared to patients who are mono-infected with HCV or HBV [23]. In the case of HIV-HCV co-infection, the proposed mechanism involves enhanced induction of the production of ROS that occurs in an NFαB-dependent fashion [24]. The general mechanism of the fibrogenesis in chronic viral hepatitis is less clear than that in nonviral chronic diseases. Most likely, the pathogenesis is multifactorial as it involves a combination of both viral and host-specific factors, including oxidative stress, hepatic steatosis, increased iron stores, and increased rate of hepatocyte apoptosis, under the pressure of the viral proteins and viral replication. With the advancements new regimens to treat patients with chronic liver diseases, the accurate assessment of liver fibrosis has become important for individualized management these patients. The lack of accurate, reproducible and easily applied methods for assessment of hepatic fibrosis has been the major limitation for both the clinical management and research in liver diseases. The following paragraphs summarize the current modalities used for quantifying and staging hepatic fibrosis. Liver biopsy scoring techniques: For the past 50 years liver biopsy has been considered to be the gold standard for staging of liver fibrosis. Liver biopsy allows physicians to obtain information not only on fibrosis, but also on many other liver injuring processes, such as inflammation, necrosis, steatosis, hepatic deposits of iron or copper. A number of these pathologic factors are can help determine the diagnosis of chronic liver disease. However, many recent studies clearly highlight several crucial drawbacks of liver biopsy, including variable accessibility, high cost, sampling errors and inaccuracy due to interand intra-observer variability of pathologic interpretations [25]. In addition, there is small but important risk of liver biopsy-associated morbidity and mortality, with pain and hypotension as the most frequent complications and intraperitoneal bleeding and injury to the biliary system as the most serious complications. Studies reveal that the risk for hospitalization after liver biopsy is 1-5%, the risk for severe complications is 0.57%, and mortality rates vary from 0.009% to 0.12% [26, 27]. Because of these reasons, some patients may opt to forgo liver biopsy and may not know the stage of their liver disease with important prognostic implications. The history of the fibrosis scoring systems dates back to 1981 when the histological features of chronic hepatitis were evaluated for potential importance in determining its prognosis by Knodell and colleagues [28]. The Ishak score, or revised Knodell system, has primarily been applied to chronic hepatitis B and C. It considers grading and staging as two separate items; liver fibrosis is classified as: 0 = absent, 1-2 = mild, 3-4 = moderate and 5-6 = severe/cirrhosis. The first three axes of Knodell HAI (Histologic Activity Index) relate to the necroinflammatory grade of the disease while the fourth feature assesses the stage of the disease by evaluating the degree of fibrous portal tract expansion, fibrous portal-portal linking, portal-central fibrous bridges, and the formation of fibrous septa and parenchymal nodules [29]. This grading system has been subsequently modified by other pathologists [30, 31]. There are some limitations of HAI index, in particular, related to the interobserver variation [32]. Another limitation of the Ishak/Knodell fibrosis score is its nonlinearity as it includes scores 0, 1, 3, and 4. The Metavir scoring system was designed specifically for patients with hepatitis C using a sum of experience-based opinions of 10 pathologists augmented by subsequent stepwise discriminant analysis [33]. The scoring uses both grading and staging systems as it includes two separate scores, one for necroinflammatory grade (A for activity) and another for the stage of fibrosis (F). The grade is a number based on the degree of inflammation, which is usually scored from 0-4, with A0 being no activity and A3 to A4 considered severe activity. Determining the amount of inflammation is important because it can correlate with hepatic fibrosis. The degree of activity is assessed by the integration of the severity of both (periportal) necrosis and lobular necrosis as described in a simple algorithm [34]. The fibrosis score (F) is defined as: F0 = no scarring, F1 = portal fibrosis without septa, F2 = portal fibrosis with rare septa, F3 = numerous septa without cirrhosis and F4 = cirrhosis or advanced scarring of the liver [35]. The intraand inter-observer variability of Metavir seems to be improved [36]. The main advantage of the Metavir score for hepatitis C is its relative simplicity, its focus on necroinflammatory lesions, and its increased sensitivity in the fibrosis score due to the addition of one extra fibrosis evaluating level. However, the limitations of the Knodell score also apply to the Metavir score as it retains the semi quantitative and categorical nature of fibrosis staging. Use of the liver biopsy scoring systems often varies among different pathology laboratories, which makes score comparisons among patients from different centers rather difficult. Built-in sampling error problems associated with accepted scoring systems requires the need to design studies with extremely large sample sizes [37]. In addition to staging hepatic fibrosis for viral hepatitis, three pathologic criteria have been used for patients with NAFLD. Of these, the original classification of NAFLD subtypes was developed to histologically categorize NAFLD into 4 subtypes): type 1 NAFLD = steatosis alone; type 2 NAFLD = steatosis with lobular inflammation only; type 3 NAFLD = steatosis with hepatocellular ballooning; or type 4 NAFLD = steatosis with Malloy-Denk bodies or fibrosis. According to these criteria, types 3 and 4 NAFLD were considered to be NASH. Subsequently, Brunt’s criteria was developed to grade NASH and used for clinical research in patients with NAFLD. According to Brunt’s criteria, liver biopsy with at least fat and lobular inflammation is graded as mild (grade 1), moderate (grade 2) or marked (grade 3) NASH. More recently, the NAFLD Activity Score (NAS) was developed to provide a pathologic numerical score for patients who most likely have NASH. Elements of NAS and the stage of fibrosis provides separate scores for steatosis (0-3), hepatocellular ballooning (0-2), lobular inflammation (0-3) and fibrosis (0-4). Accordingly, NAS is the sum of the first three features with most patients with NASH having a NAS score of >5. Fibrosis, according to Brunt and NAS is scored from 0 to 4 (grade 0 = none; 1 = centrilobular/perisinusoidal; 2 = centrilobular plus periportal; 3 = bridging; 4 = cirrhosis) [38]. These pathologic protocols for NAFLD suffer from a lack of data assessing their inter-observer variability as well as their inability to predict liver-related