Fig. 8.1
The development of science and technology greatly improved the practice of primary biliary cholangitis. The scientific achievements based on basic research and clinical investigation have significantly changed the practice of PBC involving its acknowledgment, pathogenesis, diagnosis, and therapy
8.1.2 Clinical Presentation of PBC
Many asymptomatic patients will develop symptomatic liver disease within 5 years of diagnosis; however, a third could remain symptom-free for many years. Although nonspecific, fatigue is the most common symptom of PBC, and more than 40% reported moderate-to-severe symptoms. Related to long-standing cholestasis, pruritus seems to be the most typical complaint and is reported by 20–70% of patients. A reduction in bone density is common in patients with PBC, with features of osteopenia (33%) and, less frequently, osteoporosis (11%). Hypercholesterolemia, typically caused by a rise in HDL cholesterol, is common in patients with PBC but does not increase cardiovascular risk or cause early signs of atherosclerosis. So, the statins may usually not be necessary for those patients.
Because there may exist liver inflammation, some PBC patients will experience right upper quadrant abdominal pain, where the liver is located. Like other liver diseases, cirrhosis is the final stage of PBC, with symptoms of liver dysfunction (ascites, jaundice, spider angiomata, etc.) and portal hypertension (splenomegaly, esophageal varices, caput medusa, etc.). However, a unique feature of PBC is that about 6% of patients with early-stage disease have varices before the onset of cirrhosis. Further study showed that esophageal varices develop in about a third of patients with stages III–IV disease over a median of 5–6 years; roughly half of these patients will have a bleeding event. The 3-year survival rate after initial variceal bleed is about 50%.
Hepatocellular carcinoma (HCC) and additional autoimmune diseases could also be observed in PBC patients. The most frequently associated autoimmune disease for PBC is Sicca syndrome. It is also known as Sjogren’s syndrome (SS) that classically combines dry eyes, dry mouth, and another disease of connective tissue such as rheumatoid arthritis, lupus, scleroderma, or polymyositis. Schirmer’s test is highly specific for the diagnosis of SS, and all PBC patients should undergo this easy test at the outpatient clinic. The association between SS and PBC strongly suggested a paradigm for a common immunopathogenesis.
About 10–20% of the PBC cases are associated with thyroid dysfunction (TD). Among thyroid disorders, PBC has been described in association with Hashimoto’s thyroiditis, hypothyroidism, euthyroid goiter, and, less commonly, Grave’s disease. Until now, no predictor of thyroid disease in PBC or other chronic liver disease has been identified, except possibly end-stage liver disease and treatment of chronic hepatitis C with interferon. Nonetheless, thyroid hormones have an extensive interrelationship with liver function, and prompt diagnosis and management of TD in patients with chronic liver disease can substantially improve the quality of their life.
8.2 Lessons from the Epidemiology and Natural History of PBC
PBC is a rare disease, but can be found in many if not most populations. Several large population-based studies have suggested that the prevalence of antimitochondrial antibodies (AMAs) is higher than the prevalence of diagnosed PBC, suggesting that there is a large undiagnosed cohort of PBC patients or that there is a substantial number of individuals with AMA but not clinically significant PBC. As with other autoimmune diseases, a change in the geoepidemiology of PBC is likely as countries become increasingly developed. In addition, the clinical symptoms and natural history in PBC patients are heterogeneous spanning from asymptomatic and nonprogressive to rapid progression to cirrhosis. These subgroups of PBC patients are increasingly being recognized, and specific criteria for phenotypes are being developed to better understand the underlying mechanisms leading to more severe disease.
8.2.1 The Epidemiology of PBC
The prevalence of PBC varies both on a regional and an international level. This can be explained, in part, by differences in clinical practice and case-finding activity. It is likely, however, that substantive geographical differences exist both in terms of genetic susceptibility and environmental factors that potentially trigger the disease in genetically susceptible individuals [2].
Triger DR et al. reported in 1984 the first multinational survey of the clinical and epidemiological aspects of PBC involving ten countries in Western Europe with a population of over 24 million. The results showed a female to male sex ratio of 10:1 with the ratio skewed more in later stages of disease (11.4:1 in stages III and IV compared to 6.5:1 in stages I and II). In addition, the prevalence of PBC showed a marked variation from center to center. Further, the annual incidence of PBC was increasing [3]. In 1985, Löfgren et al. estimated the incidence of AMA-positive PBC patients in Sweden during the period 1976–1983 at 140 per million [4]. Subsequent studies reported prevalence rates per 100,000 persons in Australia (1.9), Japan (2.7–5.4), the United Kingdom (20.0–25.1), the USA (40.0), and Finland (18.0). Interestingly, in 2010, Liu et al. reported the prevalence of PBC from southern China at 49.2 cases per 100,00 population [5]. Due to the very high prevalence of hepatitis B in China, which is responsible for the majority of liver disease in this area, the diagnosis of PBC is likely to have been previously underreported. A recent systematic review including 24 studies published between 1972 and 2007 found that the incidence rates ranged from 0.33 to 5.8 per 100,000 persons/year and prevalence rates ranged from 1.91 to 40.2 per 100,000 persons [6].
Whether PBC is increasing in prevalence remains controversial. In 1999, James et al. found an increase in prevalence of PBC in Northern England from 2 to 33.5 cases per 100,000 persons between 1987 and 1994 [7]. Similarly, Myers et al. found an increase in prevalence of PBC in Canada from 10.0 to 22.7 cases per 100,000 persons between 1996 and 2002 [8]. In addition, the systematic review of PBC by Boonstra et al. found that an increasing prevalence of PBC was found in most studies [6].
The true prevalence of PBC however remains unknown due to lack of case-finding studies. These limitations include a largely asymptomatic stage that is likely to escape presentation to a medical provider and the lack of knowledge or awareness of PBC in the general medical community preventing timely diagnosis even when a patient presents for a medical evaluation. Several studies of AMA prevalence in serum collected during routine health checks or blood donation suggest that the prevalence of AMA-positive cases is 430 to 1000 per 100,000 persons [9–13]. Whether those AMA-positive cases have undiagnosed PBC or preclinical PBC or will never develop PBC is not clear. A study of 29 cases with AMA-positive serum but normal liver tests demonstrated that 75% of the cases developed clinical criteria for PBC after a median follow-up of 18 years [14], suggesting that the true prevalence of PBC is much greater than measured by case-finding studies. Further studies with strict case-finding protocols are needed to find the true incidence and prevalence of PBC.
8.2.2 Changing Natural History
Early studies of PBC reported that less than 20% of cases were asymptomatic [15]. However, following the identification of the AMA antigen and the development of a diagnostic test, the corresponding percentage increased to 60% [15, 16]. The course of PBC disease falls within four phases: (i) a preclinical phase characterized by AMA reactivity but normal liver tests; (ii) an asymptomatic phase, which can last up to 20 years; (iii) a symptomatic phase, in which the patient remains anicteric or mildly jaundiced, lasting for up to 5–10 years; and (iv) a short-lasting preterminal phase characterized by a severe jaundice [17–19]. Without treatment, PBC is progressive even in those without symptoms. Several studies have documented that a quarter or more of asymptomatic patients will develop symptoms and have a decreased survival [20–23].
With the introduction of UDCA treatment, earlier identification, and likely with the identification of less severe cases of PBC, the natural history of the disease appears to have improved. Before the use of UDCA, the 10-year transplant-free survival of PBC was 50–70% in patients who were asymptomatic, and for symptomatic patients, the median transplant-free survival was 5–8 years [24]. Following the introduction of UDCA, the natural history of PBC appears to have changed dramatically. Evidence for this comes from large patient cohorts which demonstrate that survival among patients that respond to UDCA and life expectancy is not different compared to a match control population [25–27]. In addition, the number and rate of liver transplantations done for PBC has decreased significantly over the past decade [28]. However, for the one third of PBC patients who are nonresponders to UDCA, the disease appears to progress more rapidly and is associated with a reduced transplant-free survival [29].
8.2.3 Special Groups of PBC Patients
8.2.3.1 PBC in Young Women
PBC is infrequently diagnosed in patients less than 25 years of age [30], and the youngest PBC patient reported is a 12-year-old girl [31]. PBC patients presenting at an earlier age are more likely to experience significant symptoms, impaired quality of life, less likely to respond to therapy, and potentially to pursue a more aggressive disease course. Such patients should be the focus for novel therapy to improve outcome. Although male patients showed only a weak age-associated UDCA response rate pattern, women showed a clear age-related effect, with age again being an independent predictor of UDCA response on multivariate analysis. Women presenting younger than the age of 45 (the age of equivalent likelihood of response between the groups) were significantly less likely than either men or older women at presentation to respond to UDCA. On the contrary, women presenting older than age 70 had a greater than 90% chance of responding to UDCA [32].
8.2.3.2 PBC in Men
Although typically PBC affects middle-aged women, the disease may also occur in men who make up approximately 10% of most cohorts reported, and some evidence suggests that they have more severe disease. Rubel and colleagues first reported that compared to women, men with PBC had similar AMA titers, but higher serum alkaline phosphatase activities and lower frequency of piecemeal necrosis on liver histology [33]. Similar to other chronic liver diseases, men with PBC have also been reported to have a higher risk of hepatocellular carcinoma compared to women [34]. More recently, a large cohort from the United Kingdom that included 221 men found that PBC in men was less responsive to UDCA treatment regardless of age [32, 35].
Several reasons have been proposed to explain these sex-based differences in PBC. Firstly, in men, biochemical abnormalities might be attributed to alcohol. Secondly, men report fewer symptoms and, thus, may be less likely to seek medical attention than female patients [36, 37]. Finally, the rarity of PBC in men may prevent the diagnosis of PBC.
8.2.3.3 PBC Overlap with Other Autoimmune Diseases
Additional autoimmune diseases involving the liver or extrahepatic organs are commonly observed in patients with PBC [38]. Within liver disorders, the term “overlap syndrome” is used to define the coexistence of autoimmune hepatitis (AIH) and another hepatic autoimmune condition, namely, PBC or primary sclerosing cholangitis (PSC) [39]. However, overlapping PBC and PSC have been described in only a few cases. The prevalence of the PBC/AIH overlap syndrome has ranged from 2% to 20% in series published since 1998. The variable diagnostic criteria used and the rarity of the condition are likely to have led to the wide variation observed. Currently, the criteria for diagnosing overlap syndrome remain controversial. The two criteria most often used and validated in the literature come from the Paris study group [40] and the International Autoimmune Hepatitis Group (IAIHG) [41]. By analyzing the efficacy of Paris criteria and the revised and the simplified IAIHG scores in diagnosing PBC/AIH overlap syndrome, Paris criteria appeared to diagnose overlap syndrome better than the IAIHG scores [42].
The European Association for the Study of the Liver (EASL) guidelines recommend combined therapy with UDCA and corticosteroids for patients with PBC/AIH overlap syndrome. An alternative approach is to start with UDCA alone, adding corticosteroids if there is not an adequate biochemical response within 3 months. Steroid-sparing agents should be considered in patients requiring long-term immunosuppression [43].
More than 84% of PBC patients have been reported to exhibit features of at least one extrahepatic autoimmune disease, which may include rheumatologic, endocrine, gastrointestinal, pulmonary, or dermatological conditions. Sometime during the clinical course, evidence of two or more extrahepatic autoimmune diseases can be found in nearly 40% of PBC cases [38, 44]. Thyroid dysfunction is frequently (25%) associated with PBC, often predating its diagnosis [45], and xerostomia and/or keratoconjunctivitis sicca, commonly referred to as the sicca syndrome, is seen in up to 70% of patients [46]. The treatment is symptomatic and consists of the use of artificial saliva preparations with a neutral pH that contain a balance of electrolytes corresponding to normal saliva. The frequency of other extrahepatic autoimmune diseases associating with PBC varies with different type of diseases, including scleroderma (8%) [47], rheumatoid arthritis (1.8–5.6%) [48, 49], and systemic lupus erythematosus (2.7–7.5%) [50]. The prevalence of celiac disease in PBC patients has been reported to be as high as 11% in Northern Ireland, but in other populations, no associations have been found [51–53]. This may be related to true differences in prevalence or differences in the criteria used for diagnosis, particularly the requirement for small bowel biopsy rather than simply serologic testing [51–53].
8.2.3.4 PBC and Cancers
Although there is a risk of hepatocellular carcinoma (HCC) in patients with PBC, relative to other causes of chronic liver disease such as chronic viral hepatitis, the risk is low. In a study of 212 Greek PBC patients with a median follow-up of 6 years, the cumulative HCC incidence at 5 years was 1% and occurred only in those with stage IV disease [54]. In a separate UK study, compared to a control cohort, the hazard ratio for liver cancer mortality was found to be quite high (8.52; 95% CI 3.18–23.06), but the incidence was only 19 per 10,000 person years [55–57]. In Asians, the rates have also been found to be low with a cumulative incidence of 2.4% over a median follow-up of 58 months in Japan [58] and a 5-year cumulative incidence of 2.6% reported from China [59–60]. Several risk factors for those HCC patients have been investigated, including advanced histological stage (stage IV), sex, age, presence of portal hypertension, history of alcohol intake, smoking, prior infection with hepatitis B virus, and lack of response to UDCA [58–63]. Despite this overall low rate of HCC, the AASLD Practice Guidelines recommend regular screening for HCC with cross-sectional imaging with or without alpha-fetoprotein at 6–12-month intervals in PBC patients with cirrhosis [24].
Several extrahepatic malignancies (EMs) have also been suggested to be associated with PBC, but this remains controversial. The most studied cancer aside from HCC has been breast cancer. Early studies found that the incidence of breast cancer was four times the incidence expected from a comparable normal population [64, 65]. However, further studies could not confirm these findings [54, 61, 66]. A systematic review and meta-analysis on PBC and cancer risk, which considered approximately 16,300 PBC patients, concluded that PBC was not associated with extrahepatic malignancies, including breast cancer [67].
8.2.3.5 PBC and Pregnancy
Although a large proportion of the female patients with PBC are diagnosed in the postmenopausal period, up to a quarter of PBC patients may be of childbearing age [32], and nearly 15% may become pregnant after their liver disease is recognized or be diagnosed with PBC in the peripartum period [68]. Early studies noted that PBC patients often have a history of miscarriages, limited fertility, and hysterectomies before the onset of the disease [69] and that they have higher rates of maternal and fetal complications [70–75]. More recent studies however suggest a much better outcome in pregnancies, likely due to the inclusion of early-stage disease in these later studies [76, 77]. Compared to controls, PBC cases have been found to have more pregnancies [78], and two case series totally over 300 pregnancies found them to be mainly uneventful [68, 79]. Specifically, liver chemistries remained stable in 70% of patients throughout pregnancy; no adverse maternal events were observed during pregnancy or postpartum; and only 6% developed progressive disease following delivery [79]. Based upon these data, it appears that early-stage, non-cirrhotic PBC patients do not have an increased risk of miscarriage or preterm delivery and that most women with PBC maintain a stable disease during pregnancy, but up to 60% of them will develop postpartum biochemical flares. Although UDCA appears to be safe during pregnancy and breastfeeding, formal evaluation of its safety in this setting has not been established [80].
8.3 Understanding the Science of PBC Pathogenesis
The development of PBC is believed to entail the action of environmental stressors in genetically susceptible individuals, as with most other autoimmune diseases. An important role in PBC pathogenesis has been also attributed to toxic bile acids and the disruption of protective mechanisms, which under normal conditions protect biliary epithelial cells from the harmful effects of bile acids [81–83]. However, the etiology of this disease remains enigmatic despite much recent effort, reflecting the true complexity underlying autoimmunity and the difficulties inherent to understanding and testing how environmental factors interact with the genetic background to result in disease.
The greatest insights into PBC pathogenesis have come from the unique autoimmune attack on the mitochondrial proteins to which the AMA autoantibodies are directed. The immunodominant epitopes recognized by AMA are all mapped within the lipoyl domains of PDC-E2 and the other related target antigens [84]. High-resolution structural analysis and modeling studies of the PDC-E2 lipoyl domains from both prokaryotes and eukaryotes demonstrate that lipoic acid is covalently attached to the ε group of lysine (K) via an amide bond and is prominently displayed on the outer surface of PDC-E2 [85]. The change in conformation and the existence of multiple conformations of the lipoyl domain during reductive acylation are important in catalyzing the acyl transfer, but render PDC-E2 susceptible to aberrant chemical modifications. Indeed, many small molecule lipoyl mimics conjugated to PDC-E2 demonstrate specific reactivity to AMA often at levels higher than the native PDC-E2 [86–88]. Further, tolerance to lipoic acid and recapitulation of a PBC-like cholangitis can be induced by immunization of animals with xenobiotics [87–93].
In addition to AMAs, autoreactive CD8+ and CD4+ T cells to PDC-E2 are present in peripheral blood and enriched in the liver of patients with primary biliary cholangitis [94]. Reduced regulatory T cells [95], raised levels of serum polyclonal IgM [96], and hyperresponsiveness to CpG (cytosine-phosphate-guanine dinucleotide motif) [96] enhanced natural killer cell [97, 98], and monocyte responses [99] are all features found in primary biliary cholangitis.
A central conundrum to PBC pathogenesis has been how tolerance to AMAs is once lost to a ubiquitous autoantigen does such a highly specific injury of biliary epithelial cells ensue. Indeed, the answer appears to be linked to the unique processes of biliary epithelial cell apoptosis [100–105]. Unlike other cell types, the E2 component of the pyruvate dehydrogenase complex remains intact in bile duct cells after apoptosis, thus probably retaining its immunogenicity [100]. This enzyme is also found within apoptotic blebs and is accessible to AMAs and local antigen-presenting cells [101]. In vitro studies have shown an intense and specific immune response when macrophages from patients with primary biliary cholangitis are combined with apoptotic blebs of biliary epithelial cells and AMAs [105]. However, recurrence of primary biliary cholangitis after liver transplantation suggests that this occurrence is not an intrinsic defect of bile duct cells of affected individuals but is a feature of biliary epithelia in general, not seen in other epithelial cells.
8.3.1 Genetic Factors
The heritability of PBC has been clearly established in a series of 16 twin pairs, 8 of which were monozygotic and demonstrated a 63% concordance rates compared to 0% in the dizygotic twin pairs [106]. The prevalence of PBC in first-degree relatives of PBC patients has been also reported to be 4–9%
Although specific human leukocyte antigen (HLA) haplotypes have been associated with PBC, the strength of the association is less than that of other classic autoimmune diseases and is primarily limited to class II alleles. In addition, the associated haplotypes vary significantly across populations [107]. In populations of European origin, PBC has consistently been associated with the DRB1*08:01-DQA1*04:01-DQB1*04:02 risk haplotype [108] and the protective haplotype DRB1*11:01-DQA*05:01-DQB1*03:01 [109]. In Japan, DRB1*08:03-DQB1*06:01 and DRB1*04:05-DQB1*04:01 haplotypes appear to confer susceptibility to PBC, while conversely, DRB1*13:02-DQB1*06:04 and DRB1*11:01-DQB1*03:01 haplotypes are protective [110]. In China, the DRB1*08:03-DQB1*06:01 was also found to be a risk allele along with DRB1*07:01-DQB1*02:02, while the DQB1*03:01 allele and DRB1*12:02-DQB1*03:01 haplotype were protective [111]. Recent genome-wide association studies (GWAS) from North America, Italy, the United Kingdom, and Japan have confirmed that the HLA remains the most strongly associated PBC susceptibility locus with the peak of association lying between DQB1 and DQB2 [112–115].
In addition to the HLA locus, GWAS have identified 27 non-HLA risk loci for PBC including 2q32 (STAT1, STAT4), 2q33 (CTLA-4), 3q25 (IL12A, SCHIP1), 7q32 (IRF5, TNPO3), 11q23 (CXCR5), 12p13 (TNFRSF1A, LTBR), 16p13.13 (SOCS1, CLEC16A), 17q12 (IKZF3), and 19q13.3 (SPIB) [114, 116, 117].
Epigenetic modulation has also been suggested to play a role in PBC with significantly lower levels of DNA methylation of the CD40L promoter in CD4+ T cells, which is inversely correlated with levels of serum IgM in PBC patients [118]. Other DNA modifications include X chromosome inactivation [119], partial and variable methylation of CpG in CLIC2 and PIN4 [120], and changes in microRNA expression [121].
8.3.2 Environmental Factors
Although genetic factors play an important role in the development of complex autoimmune diseases, it is apparent that environmental exposures are equally important. Environmental exposures, including infectious and noninfectious agents, have been linked to PBC with most suggesting a mechanism of molecular mimicry leading to loss of tolerance to PDC-E2 [122]. Multiple studies have reported that PBC patients have more frequent urinary tract infections (UTI) compared with controls and that the infections primarily precede the development of PBC [123, 124]. Several peptides and proteins derived from bacteria frequently causing UTIs including E. coli, Novosphingobium aromaticivorans [125, 126], and Pseudomonas aeruginosa [127] cross-react with PDC-E2 antibodies and activate T-cell clones from PBC patients. Other specific infectious agents implicated in PBC pathogenesis include herpes simplex virus [128], mouse mammary tumor virus [129, 130], Epstein-Barr virus [131, 132], and Saccharomyces cerevisiae [125, 127, 133].
In addition to infection factors, xenobiotics have been suggested to play a crucial role in PBC pathogenesis due to the central role of the liver in metabolizing chemicals that in turn may modify cellular proteins to form neo-antigens. As noted above, substantial evidence exists to support the hypothesis that xenobiotic-induced and/or oxidative modification of the lipoic acid on mitochondrial autoantigens can lead to loss of tolerance [90, 91]. In clinical practice, AMA has been observed in subjects with acute liver failure from acetaminophen (APAP), presumably due to the oxidation of the lipoic acid into a neo-antigen [134]. Other environmental factors such as vitamin D, heavy metals, smoking, and nail polish have all been reported to be associated with PBC [135].
8.3.3 Innate and Adaptive Immune Responses
Defects in immune regulation that govern components of both innate and adaptive immunity contribute to the induction and abnormal perpetuation of the immune responses in PBC. From the first cloning of PDC-E2 in 1987 [136], a clearer picture of the immune mechanisms has been developed (Fig. 8.2) [137].
Fig. 8.2
Innate and adaptive immunity in primary biliary cholangitis patients (This figure is cited from Wang et al. [137]). (1) Microorganism proteins, xenobiotics, and apoptosis of biliary epithelial cells (BEC) can be recognized and endocytosed by antigen-presenting cells (APCs), which subsequently activate innate immune cells such as Toll-like receptors (TLRs), DCs, macrophages, natural killer (NK) and natural killer T (NKT) cells, and others. (2) After being processed by APCs, potentially T-cell immunogenic peptides were generated and presented to uncommitted T helper (Th0) lymphocytes and CD8 T cells. (3) Activated Th0 cells then differentiate into Th1, Th2, Tfh, and Th17 cells. Furthermore, Th1 cells secrete cytokines such as interleukin-2 (IL-2) and interferon-γ, which stimulate development of cytotoxic T lymphocytes (CTL). Th2 cells or Tfh cells secrete IL-4, IL-10, IL-13, or IL-21 and may stimulate autoantibody (e.g., AMA) production by B lymphocytes. Finally, CTL (autoreactive CD8þ T), B lymphocytes, Th17, autoreactive CD4 +T, NK, and NKT infiltrate and gather around the small bile duct and participate in the development of autoimmunity. (4) Simultaneously, the number and function of immunosuppressive cells (Treg, Breg, Tr1, and CD8 Treg) decrease significantly, which indirectly promotes overactivation of immune responsiveness
8.3.3.1 Humoral Immunity
Although high titers of serum AMA can be detected in up to 95% of patients with PBC, there is no correlation between the level of serum AMA and the severity of PBC leading to query the significance of the humoral immune response to PDC-E2. However, in vitro experiments have demonstrated the unique ability of AMA to induce a large cytokine release upon co-culture of peripheral blood mononuclear cells and biliary epithelial cells [105]. In addition and in contrast to AMA, PBC-specific antinuclear antibodies are also detected in nearly 50% of patients with PBC, and their presence does correlate with a more severe disease. Yet, whether or how those autoantibodies directly participate in immunopathogenesis of PBC remains unclear.
Supporting evidence for a role of AMA in PBC include the higher numbers of B cells spontaneously producing disease-specific autoantibodies in blood and liver tissue of PBC patients compared to controls and the correlation of the frequency of PDC-specific B lymphocytes in peripheral blood lymphocyte infiltration in the liver [138, 139]. In addition, the proportion of CD19+CD69+ activated B cells is markedly higher in the liver than in peripheral blood of PBC patients, and the number of AMA-producing cells is five times greater in the liver than in peripheral blood [140]. Further, findings of high levels of autoantigen-specific peripheral plasmablasts indicating recent activation of naive or memory B cells and a continuous and robust activation along with the presence of CXCR7+ CCR10low PDC-E2-specific antibody-secreting cells suggest a mechanistic basis for the migration of circulating antigen-specific plasmablasts to the mucosal epithelial [141].
8.3.3.2 Autoreactive T Cells
CD4+ and CD8+ T cells are present in portal tracts and around damaged bile ducts in PBC, strongly supporting their role in the development of biliary damage. Using six HLA-DRB4*01:01-restricted autoreactive CD4+ T cells from four different patients with PBC, Shimoda et al. mapped all six clones to peptide residues 163–176 (GDLLAEIETDKATI) of PDC-E2, which corresponds to the inner lipoyl domain [142]. Further study showed that those PDC-E2-specific autoreactive CD4 T cells are present in peripheral blood and liver with a 100–150-fold increase in the number of PDC-E2-specific CD4+ T cells in the hilar lymph nodes and liver compared to peripheral blood in patients with PBC [143]. Similar results were noted with HLA class I (HLA-A0201) restricted autoreactive CD8+ T cells mapping to residues 159–167 of PDC-E2 in close vicinity to the epitopes recognized by CD4 T cells as well as by AMA [127]. Further, a tenfold higher frequency of PDC-E2 159–167-specific CD8+ T cells was found in the liver compared with blood. Moreover, the frequency of precursor of PDC-E2-specific autoreactive CD8 T cells was significantly higher in early- compared to late-stage disease [144].
8.3.3.3 T Helper Cells and Regulatory T Cells
In addition to the classical CD4+ T helper cells, including Th1 and Th2 that were reported to be abnormal in over a decade ago [145, 146], many new classes of helper T cells have been noted to be abnormal in PBC. Several PBC animal models have impaired T regulatory (Treg) functions including disruption of TGF-β signaling in T cells and targeted deletion of CD25 and Foxp3 [147–149]. In human PBC, FoxP3+ Tregs can be identified in the lymphoid infiltrates localized to portal tracts, and a significantly lower proportion of circulating CD4 + CD25high Tregs are observed in PBC patients and their family members compared to healthy controls [95]. Th17 cells have been noted to accumulate around damaged bile ducts in liver tissue both from a mouse model and PBC patients [150], and follicular T helper cells in PBC have been noted to be increased and have a functional activation including IL-21 production and the ability to promote B-cell maturation and autoantibody production [151].
8.3.3.4 Monocytes
The finding of macrophages among the infiltrate of the chronic nonsuppurative destructive cholangitis classically seen in PBC was noted several decades ago [152]. Subsequently, TLR3 has been noted to be highly expressed on those macrophages and to promote the expression of type I interferons in early-stage PBC [153]. In addition, peripheral blood monocytes in PBC express higher levels of TLR4 and secrete more proinflammatory cytokines in response to infectious stimuli including lipopolysaccharide (LPS) [99, 154]. At the same time, the level of RP105, which is involved in the negative regulation of TLR4 signaling, is decreased in PBC monocytes [154].
8.3.3.5 Dendritic Cells
Localization and activation of dendritic cells have been noticed in the abnormal immune pathogenesis of PBC long time ago. By using S100 protein, dendritic cells were first identified inside the basement membrane between biliary epithelial cells of septal bile ducts in early-stage PBC [155]. Then, high restricted distribution of CD83-positive-activated DC was also found in the liver from patients with primary biliary cholangitis [156].
As antigen-presenting cells, their function in activation of the autoreactive T cells was investigated. In 2001, Akbar et al. firstly demonstrated peripheral blood T cells from PBC and showed PDC-specific proliferation when cultured with PDC-pulsed DCs [157]. In 2002, Kita et al. also found that by pulsing DC with full-length recombinant PDC-E2 protein, PDC-E2-specific cytotoxic T lymphocytes (CTL) could be generated, which indicated that CTL activation could be augmented by immune complexes cross presented by DC [158]. In addition, a phenotype of DC2 with reduced expression of HLA D2 and CD123 in PBC was considered relevant to the breakdown of tolerance to self-antigen [159]. Recently, Langerhans cells (LCs), another subtype of DC, were found existing around or within biliary epithelial layers and closely associated with the periductal cytokine milieu in patients with PBC [160].
8.3.3.6 NK and NKT Cells
With increased numbers in the peripheral blood [161] and higher frequency of cytotoxic activity [97], NK cells have been involved in the pathogenesis of PBC. A higher frequency of CD56dim/CD16pos hepatic NK cells with cytotoxic activity against autologous biliary epithelial cells was present within the liver of PBC patients, which may reflect the breakdown of NK cell immune tolerance [162]. The latest study found that NK cell-mediated innate immune responses are likely critical at the initial stage of PBC, but also facilitate and maintain the chronic cytopathic effect of autoantigen-specific T cells, essential for progression of disease.
Natural killer T (NKT) cells are a regulatory T-cell lineage that has a range of immune activities. Their correlation role with PBC patients was firstly reported by Kita et al. in 2002 [98]. Then, through a series of mice model experiments [125, 163, 164], NKT cells are found to be involved in disease exacerbations. In 2014, Aso-Ishimoto et al. further confirmed that NKT cells were significantly decreased in the liver of patients with early PBC, but increased in advanced PBC, which suggest that activated NKT cells may contribute to the biliary epithelial cell death resulting in the progression of PBC [165]. At last, by depletion both NK and NKT cells in 2-OA-BSA mouse model, Shimoda et al. found that there is a marked suppression of AMA and cytokine production from autoreactive T cells [95, 98, 127, 138–151, 158, 166–170].
8.3.4 Animal Models
The successfully constructing several murine models with manifest characteristic clinical features of human PBC, the pathogenesis of PBC, were further investigated in earlier stages and at more detailed levels, although the diverse clinical courses and the complexity of the immunological mechanisms of PBC cannot be fully recapitulated by a single animal model [171]. Until now, the PBC models could arbitrary be divided into genetically relative spontaneous models, environmental (xenobiotic and infection) triggered models, and autoimmunity relative models (Table 8.1) [176].
Table 8.1
The characteristic of different factor-induced mouse models of PBC
Genetically relative models | Environmental triggered models | Autoimmunity relative models | ||||||
|---|---|---|---|---|---|---|---|---|
NOD.ABD | Ae2a,b−/− | 2-OA-BSA | E. coli | N. aromaticivorans | dnTGFβRII | IL-2Ra−/− | Scurfy | |
Classification | Spontaneous | Spontaneous | Induced | Induced | Induced | Spontaneous | Spontaneous | Spontaneous |
Background | NOD.c3c4 | ? | C57BL/6 or NOD.1101 | NOD.B6-Idd10/Idd18 | NOD1101 | C57BL/6 | C57BL/6 | C57BL/6 |
Gender differences | Female dominant | − | − | − | − | − | − | − |
Serum biochemistry | − | Bile stasis | Bile stasis | ? | ? | Bile stasis | − | − |
AMA | 50–60% | 40–80% | 100% | 100% | 100% | 100% | 100% | 100% |
ANA | 40–50% | ? | ? | ? | ? | 100% | 80% | ? |
Ig | IgM+++, IgG+++ | IgA+++, IgG++ | ? | ? | ? | IgA+++, IgG+ | IgA+++, IgG+++ | IgM+++, IgA++ IgG +++ |
Portal lymphocytic infiltration | +++ | +~+++ | +~++ | + | + | +++ | +++ | ++~+++ |
Bile duct destruction | + | +~+++ | + | + | + | +~++ | +~++ | +~++ |
Granuloma | + | − | + | − | − | − | − | − |
Eosinophilia | + | + | − | − | − | − | − | + |
Liver fibrosis | 30% | − | -~± | − | − | − | − | − |
Commons | Also develop common biliary dilation and biliary epithelial cell proliferation | Late onset, difficult to breed | Proof of molecular mimics could lead to breach tolerance, late onset, difficult to breed | High light the importance of microbial infections in breach tolerance | High light the importance of microbial infections in breach tolerance | Spontaneously develop inflammatory bowel disease | Severe anemia, inflammatory bowel disease, short life span | Short life span |
Ref | [174] | [175] | [125] | [147] | [148] | [149] | ||
8.3.4.1 Genetically Relative Spontaneous Models
NOD.c3c4 mice is double-congenic mouse strain, which the Idd-resistant alleles from B10 and B6 mice were replaced on chromosomes 3 and 4, in nonobese diabetic (NOD) mice, respectively. In 2004, Koarada et al. [172] first discovered the PBC-like characteristics in the NOD.c3c4 mice, with spontaneous lymphocyte (CD3+, CD4+, and CD8+ T cells and PDCA1+ dendritic cells) infiltrations around the bile duct, the appearance of AMA and ANA (PDC-E2 positivity, 56% for 9–10 weeks; ANA positivity, 80–90% for 20–25 weeks), and biliary destruction [173]. Furthermore, the lack of apoptosis in bile duct epithelium due to decreased expression of Fas antigen is likely responsible for the development of autoimmunity in this model [177].
The Cl−/HCO3 − anion exchanger 2 (AE2) plays an important role in acid-base transport and export of biliary bicarbonate, which are involved in intracellular pH regulation. In 2008, AE2-knockdown mice (Ae2a, b −/− mice) were noticed to be another genetic relative PBC mouse model [174]. AMA and increased levels of IgM, IgG, and alkaline phosphatase were observed in Ae2a,b−/− mice, and histologically, 30% of the mice showed infiltration of CD4+ and CD8+ T cells in the portal areas and around the damaged bile duct. However, the disadvantage of this model is that many mice show no changes and are difficult to breed.
8.3.4.2 Environmental Factors Triggered Models
Xenobiotic Trigged Models
In 2007, Leung and colleagues immunized groups of guinea pigs with 6-bromohexanoate (6-BH)-conjugated BSA and found that immunized guinea pigs not only developed AMA responses similar to human PBC but also developed autoimmune cholangitis after 18 months [89]. 2-octynoic acid (2-OA) is found in several cosmetic products including nail polish, and their frequent use among women may contribute to the female predominance of PBC. Exposing NOD.1101 [93] or C57BL/6 [92] to 2-octynoic acid-conjugated BSA (2OA-BSA) could induce AMA and PBC-like liver lesion. The 2OA-BSA PBC model was used not only in studying the mechanisms of cytokines [178] and NK/NKT cells [164, 166], but also has been used in preclinical trials, including depleting B cells by two different monoclonal antibodies (CD20 and CD79) [179] and CTLA4-based therapy on cholangitis by using CTLA4-Ig [180].
Infectious Trigged Models
In 2014, Wang et al. demonstrated that NOD.B6-Idd10/Idd18 mice infected with E. coli developed AMA and severe cholangitis [175]. It has been reported that there are six E. coli peptide sequences that mimic the human PDC-E2 autoepitope with 6–8 identical amino acid residues [181], which may account for the E. coli-induced anti-PDC-E2 response in the NOD.B6-Idd10/Idd18 mice. In 2008, Mattner et al. reported that mice (common mouse strains C57BL/6, NOD, SJL) infected with N. aromaticivorans could also construct a model with PBC-like features [125]. N. aromaticivorans exhibits molecular homology with the PDC-E2 epitope and may potentiate the breakdown of self-tolerance in PBC patients [182].
8.3.4.3 Autoimmunity Relative Models
TGF-β receptor II is critical for signal transduction of TGF-β, which regulates the activation of lymphocytes [183]. In 2006, Oertelt S et al. firstly reported that under the direction of the CD4 promoter, a transgenic mouse directly expressing a dominant-negative form of TGF-β receptor type II (dnTGFβRII) mimics several key phenotypic features of human PBC [147]. Using this model, the pathology roles of NKT [44], CD4+, CD8+ T [184, 185], B cells [186], and lots of cytokines (such as IL-12, IL-23, and IL-17) [187–189] were extensively investigated. In addition, a preclinical experiment of B-cell depletion therapeutic was also done in dnTGFβRII mice, which indicated the potential usage of anti-CD20 in early PBC, but, at the same time, raised a cautionary note regarding the use of anti-CD20 in inflammatory bowel disease [186].
The IL-2 and Foxp3 signal pathways both play an important role for differentiation of Treg cells. In 2006, Wakabayashi et al. firstly reported that chronic nonsuppurative destructive cholangitis (CNSDC) features were found in IL-2Rα−/− c57BL/6 mice [148]. In 2009, the scurfy mice, in which the function of Tregs is abolished, were also reported to show similar pathogenesis and clinical feature to PBC patients [149]. However, the short life span makes these two model less attractive.
8.4 Progress in the Diagnosis of PBC
When initially described, PBC was believed to be a rare autoimmune disease with nearly all patients diagnosed at an advanced stage based upon the classic signs and symptoms of biliary cirrhosis with pruritus, jaundice, and xanthelasma. However, advances in diagnostic testing, particularly in autoantibody detection, have significantly changed the ability to confidently diagnose PBC (Fig. 8.3). In fact, no other autoimmune condition has autoantibodies with such specificity and sensitivity as those found in PBC. Indeed, the diagnosis of PBC requires at least two of three criteria including the presence of AMA, persistent elevation of serum alkaline phosphatase, and liver histology consistent with PBC [24].
Fig. 8.3
History of the diagnosis of primary biliary cholangitis. At the beginning, PBC were mainly diagnosed in advanced stage depending on classical signs and symptoms (such as pruritus, jaundice, and xanthomatosis). With the development of science (discovering PBC-specific autoantibodies, learning about the characteristic of liver histology, and development of more sensitive and accurate detection methods), the PBC patients were diagnosed in much early stage. Right now, PBC is diagnosed provided two of the following three criteria are satisfied: (1) AMA titer > 1:40, (2) alkaline phosphatase (AP) > 1.5 times the upper limit of normal for > 24 week, and (3) compatible liver histology
8.4.1 Autoantibodies
Since Ian Mackay’s first discovery of complement fixing antibodies in PBC [190], more than 60 distinct autoantibodies have been identified in PBC, but only a few are specific to PBC and useful for diagnosis. In addition, some may also assist in the assessment of disease severity, clinical phenotype, and long-term prognosis [191] (Table 8.2).
Table 8.2
Autoantibodies detected in primary biliary cholangitis
IIF pattern | Autoantibody | Autoantigen properties | Sensitivity | Specificity | Clinical associations | Ref |
|---|---|---|---|---|---|---|
MIT | AMA | Mitochondrial | 90–95% | High | # diagnostic value # prevalence in 0.16%–1% of general population # not associated with disease severity and treatment effect (except for AMA-IgA) # no clinical different between AMA-positive and AMA negative patients | |
Anti-PDC-E2 | Outer and inner lipoyl domain | 80–90% | ||||
Anti-PDC-E3BP | Not reported | 10% | ||||
Anti-PDC-E1a | TTP binding and phosphorylation site | 5–25% | ||||
Anti-OGDC-E2 | lipoyl domain | 20–60% | ||||
Anti-BCOADC-E2 | lipoyl domain | 50–80% | ||||
NE | Anti-gp210 | Integral glycoprotein of the nuclear pore | 20–40% | Very high | # diagnostic value for AMA-negative patients # associated with disease severity and treatment effect | |
Anti-P62 | Glycoprotein of the nuclear pore | 10–30% | High (also detected in SjS) | # diagnostic value for AMA-negative patients # presence related to the progressive or advanced stage of PBC # indicate marked inflammatory infiltrates in liver biopsy | ||
Anti-lamin B receptor (LER) | A protein integral to the inner nuclear membrane with a nucleoplasmic | 2–6% | High | # diagnostic value for AMA-negative patients # clinical significance is nuclear | ||
MND | Anti-SP100 | Nuclear protein antigen | 20–40% | High (also detected in SLE, pSS) | # diagnostic value for AMA-negative patients # associated with disease faster progression | |
Anti-PML | A protein fused with the retinoic acid rector-a | 15–20% | High | No report | ||
Anti-SP140 | Promyelocytic leukemia protein nuclear body components | 15% | High | # diagnostic value for AMA-negative patients # no association was found in any clinical feature | [203] | |
Anti-SUMO-1, 2 | Small ubiquitin-related modifiers | 2–6% | High | # diagnostic value for AMA-negative patients | [204] | |
CENP | Anti-centromere A,B,C | Major centromere proteins | 10–30% | Not high (also detected in SSc) | # predictive value for progression to portal hypertension for PBC |
In 1965, antimitochondrial antibodies (AMA) in PBC sera were first recognized by Walker and his colleagues [207] using indirect immunofluorescence, and subsequently the 74 kD mitochondrial autoantigen, PDC-E2, was cloned and sequenced in 1987 [136, 208]. The epitopes recognized by AMA are often referred to as M2 antigens for historical reasons but more accurately include the lipoylated domains of the E2 and E3 binding protein (E3BP) components of the pyruvate dehydrogenase complex (PDC-E2) and the E2 components of the 2-oxoglutarate dehydrogenase (OADC-E2) and branched-chain 2-oxo-acid dehydrogenase (BCOADC-E2) complexes. Accordingly, the discovery of those autoantigens greatly changed AMA detection, moving from indirect immunofluorescence (IIF) of rat kidney, stomach, liver, or HEp-2 cells to immunoblotting of purified or recombinant mitochondrial antigens and finally to sensitive enzyme-linked immunoassay (ELISA) systems and bead assays using recombinant proteins co-expressing the immunodominant epitopes of PDC-E2, BCOADC-E2, and OACD-E2 [9, 84, 209, 210]. With improvements in the sensitivity of assays for detecting AMAs, up to 95% of PBC patients can now be classified as AMA-positive compared to 80% with older technologies [24].
In addition to AMA, PBC-specific antinuclear autoantibodies (ANA) with membranous/rim-like and multiple nuclear dots (MND) IIF patterns are also detected in approximately 30–50% of PBC patients [211–213]. These antigens include components of the nuclear envelope pore complexes (gp210 and p62), which correspond to a dotted nuclear envelope or rim-like/membranous pattern; lamin A, B, and C and lamin B receptor, which correspond to a smooth nuclear envelope pattern; sp100, PML proteins, sp140, and small ubiquitin-related modifiers (SUMOs), which correspond to a multiple nuclear dot (MND) pattern; and centromere A, B, and C proteins, which correspond to a centromere (CENP) pattern [191]. The detection of these PBC-specific ANA can be used to confirm the diagnosis of PBC in AMA-negative cases, thus increasing the sensitivity of serological tests in diagnosing PBC. Approximately half of AMA-negative PBC patients are positive for at least one of the three PBC-specific ANAs, including anti-gp210, anti-SP100, or antipromyelocytic leukemia (PML) antibodies [9, 214].
Although AMA titers do not change over time and are not associated with disease severity or progression [215], some PBC-specific ANAs have been reported to associate with a more severe disease course [194, 211]. Antibodies to gp210, which are highly specific for PBC and detected in 20–40% of PBC patients [192, 193, 216], have in particular been associated with disease severity [194, 205, 217–221]. In fact, serum titers of anti-gp210 antibodies change from negative to positive or vice versa depending on disease activity or stage progression. Anti-sp100 and anti-PML antibodies are also highly specific for PBC, with a prevalence of 20–40% and 15–20%, respectively [199, 200, 202, 222], and have also been reported to associate with disease severity and poor prognosis [199, 200, 202].
8.4.2 Liver Biochemistry Changes
Presently, more than half of patients diagnosed with PBC are asymptomatic and only suspected based upon routine liver tests. They generally attract medical attention by findings of elevated serum alkaline phosphatase. In addition, serum alkaline phosphatase is used as a measure of disease activity and response to treatment in PBC. Although there is a linear correlation between serum bile acids and alkaline phosphatase in PBC, there are several limitations to the use of serum alkaline phosphatase that should be considered. Four distinct alkaline phosphatase isoforms are present in humans, namely, intestinal, placental, placenta-like, and liver/bone/kidney, and the liver alkaline phosphatase is found in the serum of patients with cholestatic forms of liver disease including PBC. However, serum alkaline phosphatase activity can be influenced by genetic variants, pregnancy, bone growth, age, and medications [223].
In addition to alkaline phosphatase, elevated levels of serum γ-glutamyl transferase and aminotransferases are usually observed in PBC. Increased serum levels of (conjugated) bilirubin as well as alterations in prothrombin time and serum albumin are usually late phenomena in PBC similar to other cirrhotic states. However, serum bilirubin is a strong and independent predictor of survival with a high impact on all established models of prognosis.
Serum IgM value is often increased in cholestasis, particularly in PBC [224]. The combination of a high serum IgM together with a very high ALP and a normal or only moderately increased serum bilirubin concentration strongly suggests PBC. Increased awareness of the condition and the increasing availability of diagnostic tools, in particular serological testing, have led to a more frequent and earlier diagnosis of PBC.
8.4.3 Liver Histology
The histological changes of PBC are well defined and traditionally based on Ludwig’s four classification stages, including stage I, portal inflammation with granulomatous destruction of the bile ducts with or without granulomas; stage II, periportal hepatitis and bile duct proliferation; stage III, fibrous septa or bridging necrosis; and stage IV, cirrhosis [225]. Notably, the liver is not uniformly involved, and features of all four stages of PBC can be found in one biopsy specimen. The role of liver biopsy in patient management currently remains controversial, and most patients do not undergo liver biopsy as it is not required for diagnosis in most cases and does not alter management. However, there may be some potential prognostic value in obtaining a liver biopsy as the presence of piecemeal necrosis has been associated with a decrease in transplant-free survival [226–230]. In addition, liver biopsy should also be performed if there is concern for an overlap with autoimmune hepatitis.
8.4.4 Novel Diagnosis Methods
In rare cases, the diagnosis of PBC remains challenging when there is an absence of AMA or PBC-specific ANA in a patient with other features suggestive of PBC. In addition, development of less expensive diagnostic tools would be welcomed in clinical practice. In 2014, Tan Y et al. reported a panel of microRNAs (hsa-miR-122-5p, hsa-miR-141-3p, and hsa-miR-26b-5p) that showed a high diagnostic accuracy for PBC but was significantly less accurate compared to AMA [231]. More recently, anti-KLHL12 and anti-HK1 antibodies added to AMA and ANA serological assays significantly improved efficacy in the clinical detection and diagnosis of PBC, especially for AMA-negative subjects [232].
8.5 Current Therapies for PBC Patients
PBC therapies are largely targeted to the liver pathology in order to prevent or reverse bile duct injury, subsequent liver fibrosis and liver-related complications, and death. Secondarily, these effects would be expected to prevent the consequences of chronic cholestasis, including pruritus, fatigue, osteoporosis, etc.
8.5.1 Ursodeoxycholic Acid (UDCA)
Despite major breakthroughs in the understanding of PBC, there remains only one agent approved by the US Food and Drug Administration for the treatment of PBC, ursodeoxycholic acid (UDCA). UDCA dosed at 13–15 mg/kg/day is recommended by the practice guidelines of both the European and the American Associations for the Study of Liver Diseases for PBC patients [233–236].
UDCA may act through multiple mechanisms including the stimulation of hepatocellular and ductular secretions, cytoprotection against bile acid-induced injury, anti-apoptotic effects, and further anti-inflammatory and immunomodulatory effects. Hydrophobic bile acids damage cell membranes and that conjugates of UDCA counteract [237] and modulate the composition of mixed phospholipid-rich micelles in bile [238, 239]. UDCA also stimulates biliary secretion of bile acids [240, 241] by enhancing the expression of transporter proteins [242–244] and targeting the insertion of transporter molecules into the hepatocyte canalicular membrane [245–248] (Figs. 8.4 and 8.5). In addition, accumulation of hydrophobic bile acids in hepatocytes can lead to apoptosis and subsequent inflammation and liver fibrosis. Anti-apoptotic effects of UDCA in vitro and in vivo in the rat [251, 252] and in human hepatocytes [253] are associated with a reduction in mitochondrial membrane permeability to ions and mitochondrial cytochrome c release [254]. UDCA diminishes Fas ligand-induced apoptosis in mouse hepatocytes [255] and protects rat hepatocytes against bile acid-induced apoptosis by preventing bile acid-induced, c-Jun N-terminal kinase-dependent Fas trafficking to the plasma membrane [256]. Further actions of UDCA include the suppression of IL-2, IL-4, and IFN-y from activated T cells, immunoglobulin production from stimulated B cells [257–259], IL-1 secretion from monocytes/macrophages [260], and suppression of IFN-γ-induced MHC class II expression [261–263] (Fig. 8.6).
Fig. 8.4
Potential mechanisms and sites of action of ursodeoxycholic acid in cholestatic liver diseases and primary biliary cholangitis. Several mechanisms could contribute to the beneficial effect of ursodeoxycholic acid (UDCA) under various cholestatic conditions: (1) by stimulating impaired hepatobiliary secretion, (2) via inhibition of bile acid-induced hepatocyte (and cholangiocyte?) apoptosis, (3) by decreasing bile cytotoxicity, and (4) via stimulation of cholangiocellular calcium-dependent chloride/bicarbonate anion secretion. The relative contribution of each of these mechanisms to the anti-cholestatic action of UDCA is unknown (This figure is cited from Beuers [249])
Fig. 8.5
UDCA conjugates act as posttranscriptional secretagogues in experimental cholestasis. Experimental model of TUDCA-induced stimulation of hepatocellular secretion. TUDCA is taken up into the hepatocyte by the Na+-taurocholate cotransporting polypeptide (Ntcp) and stimulates apical vesicular exocytosis and insertion of key canalicular transporters such as the conjugate export pump Mrp2 and the bile salt export pump, Bsep, via Ca2+– and PKCα-dependent mechanisms or via activation of p38MAPK and Ras-, Raf-, Erk-1/2-dependent mechanisms. TLCA is the most potent cholestatic agent among the major human bile acids. TLCA is a potent signaling molecule which elevates hepatocellular [Ca2+]i without stimulation of Ca2+ influx, selectively translocates nPKCε to canalicular membranes and activates membrane-bound PKC, and induces retrieval of key apical transporters such as the bile acid export pump, Bsep/Abcb11, from canalicular membranes of hepatocytes (This figure is cited from Wimmer et al. [250] and Gustav Paumgartner and Ulrich Beuers [239])
Fig. 8.6
Basic and clinical research made UDCA the only treatment for primary biliary cholangitis. Through clinical research, UDCA could dramatically improve liver biochemistries, significantly delay the progression, and prolong survival free of transplantation for PBC patients. Basic research revealed a series of mechanisms responsible for UDCA’s treatment effects: (1) stimulate hepatocellular and ductular secretions, (2) anti-apoptotic effects, (3) decrease of bile cytotoxicity, (4) immune modulation. Finally, both the basic and clinical research supporting each other made UDCA, dosage at 13–15 mg/kg/day, to be the only therapy for PBC approved by the US Food and Drug Administration and recommended by the practice guidelines of both the European and the American associations
The first open-label trial of UDCA at a dose of 12 to 15 mg/kg/d in patients with PBC successfully demonstrated a dramatic improvement in liver biochemistries with few adverse effects [264]. In 1991, the first multicenter, double-blind randomized controlled trial of UDCA further validated the usefulness of UDCA as a therapeutic agent for PBC [265] followed by a series of clinical trials that found a similar favorable effect on improvement of biochemical cholestasis [266–268]. However, when UDCA has been stopped, there is a prompt rebound of serum biochemical values to pretreatment levels indicating that long-term, perhaps lifelong, treatment is likely to be necessary [266, 267, 269–272].
Because PBC is a slowly progressive and rare disease, establishing benefits of any therapy in terms of histologic or clinical outcomes has been challenging. Nevertheless, histologic improvement with UDCA has been demonstrated with long-term UDCA treatment delaying histological progression [57]. In addition, a Markov model of a randomized, double-blinded, placebo-controlled trial of UDCA found that UDCA therapy was associated with a fivefold lower progression rate from early-stage disease to extensive fibrosis or cirrhosis (7% per year under UDCA vs. 34% per year under placebo), but was not associated with a significant difference in regression rates [273]. Further, analysis of paired liver biopsy specimens from several studies with a time interval of 36 months between biopsies indicated that UDCA had the greatest effects on the subgroup of patients with baseline stages I and II [274].
The effect of UDCA on survival has been more difficult to establish. The first long-term clinical trial involving 145 patients with biopsy-proved PBC randomized to receive UDCA or placebo for 2 years followed by an open-label trial for two additional years showed that UDCA therapy reduces the need for liver transplantation [275]. Several additional clinical trials and meta-analyses have been performed with most, though not all [276–279], concluding that UDCA improves transplant-free survival among patients with later-stage disease [280]. In addition, several large cohorts have demonstrated significantly better survival among those treated with UDCA compared to predicted survival using models of historical controls [25–27, 281–284].
Although UDCA appears to be effective in many if not most cases of PBC, there remains a group of patients that continues to progress. The early recognition of PBC patients with a predicted poor long-term outcome is a key issue for patient management and effective design of clinical trials. The biochemical response to UDCA is an independent predictive factor for death and liver transplantation and has been recommended as a study endpoint in clinical trials where clinical endpoints are deemed unfeasible. Thus, it is necessary and important to identify patients with an incomplete biochemical response as soon as possible.
Several criteria have been reported to define an “incomplete” biochemical response to UDCA. Pares et al. from Barcelona defined response to UDCA treatment by an alkaline phosphatase decrease greater than 40% of baseline values or normal levels after 1 year of treatment and successfully predicted the survival for UDCA treatment patients [27]. Several other studies have shown that the biochemical response to UDCA assessed at 1 year or 6 months can predict the long-term outcome with UDCA responders having a survival similar to that estimated for the matched control population [25, 26, 282–285]. More recent studies of large cohorts have compared these criteria and developed new models, but in general they all conclude that the main predictors are serum levels of alkaline phosphatase and bilirubin [32, 286–288] (Table 8.3). In addition to biochemical response, other predictors of outcome may capture additional information related to outcomes. The AST/platelet ratio index (APRI), which is an indicator of advanced liver fibrosis, has been reported to predict adverse events in patients with PBC independent of UDCA response [289].
Table 8.3
Biochemical criteria for the responses to ursodeoxycholic acid (UDCA) in primary biliary cholangitis
Criteria | Issue date | Author | Evaluation time | Biochemical response to UDCA | Ref |
|---|---|---|---|---|---|
Barcelona | 2006 | Pares A, et al. | After 1 year | Decreased of ALP by > 40%, or ALP normalization | [27] |
Paris I | 2008 | Corpechot C, et al. | After 1 year | ALP ≤3 x ULN, AST≤2 x ULN, and serum bilirubin ≤1 mg/dL | [25] |
Rotterdam | 2009 | Kuiper E.M, et al. | After 1 year | Normalization of abnormal serum bilirubin and /or albumin | [26] |
Ehime | 2009 | Azemoto | After 6 months | Normalization of GGT level or reduction rate of GGT above the ULN> 70% | [285] |
Toronto | 2010 | Kumagi T, et al. | After 2 year | ALP <1.67 × ULN | [282] |
Paris II | 2011 | Corpechot C, et al. | After 1 year | ALP < 1.5 × ULN or AST<1.5 × ULN or Bilirubin ≤1 mg/dL | [283] |
Beijing | 2013 | Zhang L.N, et al. | After 6 months | ALP ≤3 × ULN, decreased bilirubin, increased albumin | [284] |
8.5.2 Liver Transplantation
Liver transplantation for PBC patients has some of the best survival data compared to any other indication with the 1-, 5-, and 10-year survival rates of 86%, 80%, and 72%, respectively [290]. Until the 1990s, PBC was the most common indication for liver transplantation in the USA accounting for up to 55% of transplanted patients. Currently, PBC accounts for less than 5% of all transplantations in the USA and has steadily decreased, likely reflecting the impact of the widespread use of UDCA.
Following liver transplantation, PBC recurs in 21–37% of patients at 10 years after liver transplantation [291] and in 43% at 15 years with the median time to recurrence of 3–5.5 years [292, 293]. Diagnosis of recurrent PBC (rPBC) is based on the liver histopathology and meeting all diagnostic criteria including PBC as the indication for liver transplantation; histopathology of the graft is suggestive for rPBC, including epithelioid granulomas, mononuclear inflammatory infiltrate, formation of lymphoid aggregates, and bile duct damage; and other causes of graft failure are excluded [294]. Recurrence of PBC has been associated with donor and recipient age [295], cold and warm ischemia times [296], number of HLA mismatches [297, 298], and immunosuppressive regimen including tacrolimus [299] and azathioprine [300]. Recently, male-to-female sex mismatch did not appear to yield a direct negative impact on outcomes following liver transplantation for PBC, but it did appear to aggravate the negative effects of prolonged cold ischemia and blood transfusions [301]. UDCA treatment of liver transplant recipients may improve liver function test results, but it does not impact patient or graft survival [302].
8.5.3 Symptom Therapy
As previously reported, fatigue and pruritus remain the most common symptoms reported by patients with PBC. In addition, patients with fatigue and pruritus at onset are more likely to progress to cirrhosis and are less likely to respond to UDCA [303]. Currently, there are no approved therapies available for these indications, and the data available indicates that off-label use of therapies has limited efficacy for these indications.
8.5.3.1 Fatigue
Fatigue occurs in up to 70% of patients [304, 305]. It is associated with excessive daytime sleepiness and autonomic dysfunction [306, 307]. Severity is measured according to the fatigue impact scale (FIS) [308] or PBC-40 (PBC-40 question profile) [309]. The pathogenesis of fatigue in liver disease is still poorly defined, but it appears to involve both peripheral and central nervous system components [306, 307]. Observational data found some benefit from modafinil in the patients with prominent daytime somnolence [310, 311] with side effects including insomnia, nausea, nervousness, and headaches.
8.5.3.2 Pruritus
Pruritus from cholestasis is mostly generalized, associated with scratching, sometimes violent, and sleep deprivation. Intensity of pruritus is one of the most unbearable symptoms, but not correlated with liver disease severity. Patients with pruritus from liver disease do not have primary pruritic skin lesions, and the lesions are usually secondary to scratching.
Although the exact mediators of cholestatic pruritus remain to be elucidated, two main mechanisms have been proposed: activation of G protein-coupled bile acid receptor 1 (GPBAR1; also known as Takeda G protein-coupled receptor 5, TGR5) and activation of the autotaxin (ATX, also known as ectonucleotide pyrophosphatase/phosphodiesterase 2) [312, 313] (Fig. 8.7). In contrast to other pruritogen candidates, such as bile salts, endogenous opioids, histamine, and serotonin serum, levels of ATX have been reported to correlate with itch intensity. Notably, ATX serum level mirrors treatment response from therapeutic interventions, such as colesevelam, rifampicin, nasobiliary drainage, or MARS® treatment. The beneficial antipruritic action of rifampicin may be explained, at least partly, by PXR-dependent transcription inhibition of ATX expression. Targeting the ATX pathway offers considerable hope for novel therapy for PBC and other cholestatic pruritus [315].
Fig. 8.7
The itch neuron. Schematic representation of receptors and neurotransmitters that may play a role in itch transmission. Pruritogens may bind to their receptors on unmyelinated itch nerve endings in the skin. This may involve established receptors such as histamine receptors, PAR2, IL-31 receptor, and others but also newly discovered receptors, such as those from the transient receptor potential (TRP family, TRPV1, and TRPA1) or from the Mas gene-related (Mrg) family, that are activated by endogenous and exogenous small molecule pruritogens. Itch may also be initiated or potentiated by activation of LPA receptors. Synaptic transmission between the primary and secondary itch neurons may be mediated by Nppb, while transmission between the secondary and tertiary neuron may involve Grp. Finally, at the level of the spinal cord itch transmission is dampened by pain signals. CNS central nervous system, DRG dorsal root ganglia, Grp gastrin-releasing peptide, LPA lysophosphatidic acid, LPC lysophosphatidylcholine, Mrg Mas-related G protein coupled, Nppb natriuretic polypeptide b, Trp transient receptor potential cation channel, TGR5 G protein coupled bile acid receptor (This figure is cited from Beuers [314])
Many therapeutic measures have been used, among which cholestyramine and colestipol are the most common. These drugs are nonabsorbable basic polystyrenes that bind anions in the gut lumen, including bile acids and other substances, blocking their absorption [316–318]. Others, like rifampicin [319, 320], naltrexone [321, 322], sertraline [323, 324], propofol, methyltestosterone, ondansetron, gabapentin, S-adenosylmethionine, ultraviolet light exposure, and plasmapheresis, have been tried for relieving itching associated with PBC, but none have been assessed in a formal manner (Fig. 8.8). Moreover, there is no evidence that standard topical therapies for pruritus are effective in patients with PBC [325].
Fig. 8.8
Therapeutic targets in pruritus of cholestasis. Clinical observations indicate that potential pruritogens (1) accumulate in the systemic circulation (as indicated by relief of severe pruritus after treatment with plasmapheresis or albumin dialysis), (2) are secreted into bile (as proven by rapid relief of severe pruritus after nasobiliary drainage or suggested attenuation of pruritus after administration of cholestyramine), (3) are (biotrans-)formed in the liver and/or gut (as indicated by effective treatment with the potent PXR agonist, rifampicin), and (4) affect the endogenous opioidergic and serotoninergic system (as supported by the alleviating effects of naltrexone or sertraline). Recent evidence indicates that LPA formed by ATX represents a long-sought trigger of unmyelinated itch neuron endings. A biliary factor “X” which might stimulate ATX formation remains to be unraveled (This figure is cited from Beuers et al. [314])
8.5.3.3 Osteoporosis
PBC patients have a 20–44% prevalence of osteoporosis. The prevalence increases with disease progression, and up to 80% of patients with cirrhosis have osteoporosis. The risk factors for osteoporosis in PBC are similar to those in the general population and include old age, female gender, smoking, excessive alcohol consumption, underweight physique (body mass index < 19.0 kg/m2 in adults), early menopause (< 45 years of age), positive family history of osteoporosis, and corticosteroid therapy [326]. The detailed mechanism of osteoporosis is unclear. For the prevention and treatment of osteoporosis, good nutrition is recommended, as is the suppression of the risk factors for osteoporosis. Supplements of calcium and vitamin D, or the dose required to maintain normal levels, should be provided. Particular care should be taken with patients receiving resins because their administration may reduce the intestinal absorption of vitamin D [327]. Although calcium and vitamin D supplements are recommended, there are no data confirming the efficacy of these supplements in preventing bone loss in PBC patients. Bisphosphonates, such as pamidronate, ibandronate, or zoledronic acid to stabilize bone mineral density in PBC patients, appear to be safe and effective [328].
8.6 Scientifically Based Novel Therapeutic Targets for PBC Patients
8.6.1 Bile Acid-Based Therapies
8.6.1.1 24-norursodeoxycholic Acid (nor-UDCA)
nor-UDCA is a side chain shortened UDCA derivate, which lacks a methylene group resulting in a relative resistance to amidation with taurine or glycine compared with UDCA. nor-UDCA is passively absorbed from cholangiocytes and undergoes “cholehepatic shunting” instead of a full enterohepatic cycle. By induction of a HCO3 −-rich hypercholeresis [329–331], nor-UDCA could counteract intrinsic bile acid toxicity [332] and reinforce the “biliary HCO3 − umbrella” [333]. Moreover, nor-UDCA has anti-lipotoxic, antiproliferative, antifibrotic, as well as anti-inflammatory effects which may complement stimulation of HCO3 − secretion with bile acid detoxification and induction of alternative export via overflow systems at the basolateral membrane [334, 335].
Related posts:
Biliary Atresia: From Pathology to Treatment
The Clinical Burden of Biliary Disease: A Global Perspective
Animal Models of Biliary Disease: Current Approaches and Limitations
Conclusions and Future Opportunities
Gallstone Disease: Scientific Understanding and Future Treatment
Primary Sclerosing Cholangitis (PSC): Current Concepts in Biology and Strategies for New Therapy
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
