Systemic sclerosis and primary biliary cholangitis: An overlapping entity?

Abstract

Systemic sclerosis (SSc) is a complex autoimmune disease that may lead to skin and internal organ fibrosis. Based on skin involvement, two subsets of the disease are recognized (limited cutaneous SSc and diffuse cutaneous SSc). The new 2013 American College of Rheumatology/European League against Rheumatism classification criteria allow to identify SSc patients at the early stage of the disease that allows new research avenues. The aetiology of the disease is still unknown, but it has an important autoimmune basis and its association with other autoimmune diseases has been reproducibly reported. Among them, primary biliary cholangitis is considered the most common liver disease in SSc. The aim of this review is to provide an overview on recent findings about SSc associated to primary biliary cholangitis. Although the aetiology of the two diseases is still unknown, data suggest that these two disorders share the expression of fibrogenic cytokines, involved both in generation and function of T lymphocytes subpopulation (Th17 cells) and regulatory T lymphocytes. In addition, the relationships between SSc and primary biliary cholangitis may be closer as suggested by the presence of primary biliary cholangitis–specific antibodies in SSc patients and vice versa. Recent findings confirm a prevalence of overt primary biliary cholangitis in about 2% of SSc population, in particular in patients with limited cutaneous SSc and positive anticentromere antibodies. The prevalence increases if also patients with only primary biliary cholangitis–specific antibodies are considered. Data regarding SSc prevalence in primary biliary cholangitis patients have also been recently clarified. Altogether, stimulating results are moving the field forward regarding the relationships of these two autoimmune and fibrotic disorders that may belong to an overlapping entity.

Introduction

Systemic sclerosis (SSc) is an autoimmune, multisystem and chronic connective tissue disease (CTD) leading to fibrosis of skin and internal organs. Its prevalence ranges from 50 to 300 cases per million and the incidence from 2 to 23 per million with a higher frequency in women. The aetiology of the disease is still unknown, and different mechanisms seem to be involved in the development of the disorder, including genetic predisposition, molecular mimicry, environmental agents and an impairment of regulatory T cells. ¹ The disease is a ‘microvasculopathy’ with presence of autoantibodies and fibroblast dysfunction from which the interplay provokes an increased tissue deposition of extracellular matrix with the fibrosis of skin and internal organs.^2,3 From the clinical perspective, the 1980 American Rheumatology Association (ARA) classification criteria for years have been the more used criteria, although they demonstrated a low sensitivity to recognize SSc in the early stages. ⁴ For this reason, LeRoy et al. ⁵ proposed new criteria, including antibodies (Abs), nailfold videocapillaroscopy (NVC) and clinical features, also recognizing two cutaneous subsets of the disease: limited cutaneous SSc (lcSSc) and diffuse cutaneous SSc (dcSSc), based on skin involvement and clinical features. Although, the sub-classification of SSc in different subsets or clusters remains a challenge for rheumatologists, ⁶ in the recent years, more attention was posed on SSc-specific Abs and on NVC in order to try to identify patients in an ‘early phase’ of SSc. ⁷ In 2013, the American College of Rheumatology (ACR) and the European League against Rheumatism (EULAR) developed the new classification criteria for SSc, including clinical, laboratory (the three specific Abs, anticentromere -ACA-, anti-topoisomerase -TopoI- and anti-RNA polymerase III -anti-RNAP-) and instrumental features. Feasible to use in clinical practice and more sensitive and specific of the previous ones, the 2013 ACR/EULAR classification criteria allow to recognize both SSc patients in the early phase of the disease both those in the late and fibrotic stage. ⁸ The new criteria are definitively adopted by the community with so far already more than 800 PubMed citations, 4 years after their release.

Several studies reproducibly reported that SSc may be associated to different autoimmune disease, not only rheumatic diseases as Sjögren’s Syndrome (SjS) or Systemic Lupus Erythematosus (SLE) but also specific-organ disorders as thyroiditis. ⁹ Furthermore, it is known the possible association between SSc and autoimmune liver diseases, as autoimmune hepatitis (AIH), primary sclerosing cholangitis (PSC) and with a higher frequency primary biliary cholangitis (PBC), previously known as primary biliary cirrhosis. ¹⁰ The first association of PBC with SSc (lcSSc) was described by Murray-Lyon et al., ¹¹ afterwards other case reports and case series were described and PBC is still recognized as the most common autoimmune liver disorder in SSc patients. ¹² PBC is an autoimmune cholestatic liver disease with a female predominance (9:1), with a prevalence ranging from 7 to 402 per million and an annual incidence of PBC from 10 to 49 case per million. The disease is characterized by a progressive destruction of the intrahepatic bile ducts that may lead to cholestatic disease. The disease is characterized by circulating antimitochondrial Abs (AMA) in 80% to 96.5% of patients. In asymptomatic patients, these Abs are predictive for the future development of PBC. The other Abs characteristic of PBC are anti-sp100 and anti-gp210.^13,14 In Table 1, PBC-specific Abs are reported. The diagnosis of PBC is based on clinical features, abnormal liver enzymes persisting for more than 6 months and the presence of PBC-specific Abs. The diagnosis may be confirmed by liver biopsy, and it is considered as probable when two of the three above items are present and definite when all three items are present. ¹⁴ The absence of AMA or other Abs in patients with a cholestatic disease makes liver biopsy a fundamental tool for the diagnosis or the exclusion of PBC. The aetiology of PBC is not completely known, and different factors seem to be involved in the development of PBC, as genetic, infectious and environmental/chemical agents. In PBC, the main findings at liver biopsy are inflammation, fibrosis and bile duct damage, and according to their degree, different histological stages are recognized. These findings reflect another aspect of the pathogenesis, that is, the chronic inflammation leading to duct destruction and to a fibrogenic response promoting a fibroproliferative response to cholestatic injury. ¹⁴ In PBC population, the association with other autoimmune disorders is reported, and SjS and SSc seem to be the most common. ²¹ The association with PBC is estimated in approximately 12.3% of patients with rheumatic diseases. ²²

Table 1.

More frequent PBC-specific Abs and their epidemiological and clinical characteristics.

Autoantibodies	PBC
AMA	Prevalence in PBC patients: 90%; prevalence in general population: 0.16% and 1% 16 Predictors for the development of PBC
ACA	Often present in patients with PBC and associated PBC (in particular lcSSc) A study reported that ACA in PBC patients may represent a risk factor in the development of portal hypertension
Antinuclear pore Abs (gp210)	Prevalence in PBC: about 25% 17 These Abs may predict the progression of hepatic involvement as consistently high levels of these Abs may be associated to progression and end-stage hepatic failure18,19
Antinuclear dot Abs (anti-sp100)	Prevalence: about 25% Laboratory clinical feature for the diagnosis of PBC in AMA-negative patients 20 These Abs may be present in other autoimmune diseases (as SLE and SSc)

PBC: primary biliary cholangitis; Abs: antibodies; AMA: antimitochondrial Abs; ACA: anticentromere Abs; lcSSc: limited cutaneous SSc ; SLE: systemic lupus erythematosus.

The aim of this review is to provide an overview on recent findings about SSc-PBC and discuss the possibility to consider it as an overlap syndrome.

A common aetiology?

Although no common aetiology has been established, the two diseases seem to share an abnormal mesenchymal cell activation and an autoimmune basis. As previously mentioned, genetic predisposition seems to play an important role in the pathogenesis of SSc. ¹ The hypothesis is that SSc and PBC share pathogenetic pathways that still need to be elucidated in particular to understand the common immune and fibrogenic or fibroproliferative pathways. Some genetic factors have been reported in different autoimmune disorders as disease susceptibility genes, although not considered ‘disease-specific’ genes. HLA-DRB1, DQA1, DRB1, signal transducer and activation of transcription 4 (STAT4) and interferon factor 5 (IRF5) are included in these susceptibility genes that may be shared by SSc and PBC. ²³ In fact, PBC and SSc are characterized by an increase in fibrogenic cytokines involved in the production and function of T helper (Th)-17 cells and regulatory T cells (Treg), as transforming growth factor (TGF)-β and interleukin (IL)-6. ²⁴ Recently, it has been highlighted that Th17 and Treg generation and function in PBC and SSc have some interesting similarities. IL1, IL6 and TGF-β induce fibroblast activation and are elevated in SSc sera and tissues. ²⁵ In addition, some data showed that levels of IL6 may correlate with the extension of skin involvement and that TGF-β is more increased in the early stages of SSc. ²⁶ As mentioned above, these cytokines have a role in activation and function of Th17 cells.

In PBC patients, the liver biopsy may show a progressive intrahepatic bile duct destruction, and at liver tissue immunohistochemistry, a higher concentration of Th17 cells has been found.^25,27,28 In epithelial cells, the expression of IL17 receptors may lead to respond to IL17 with the secretion of IL6, IL1β and IL23, cytokines that are involved in the maintenance of liver inflammation. ²⁹ As recently showed by in vitro studies and works on animal models, also the IL12 seems to play an important role in the development of PBC, inducing the production of pro-inflammatory Th1-type cytokines. In SSc and in PBC, abnormalities in CD8+ Treg subset are found and in SSc also the CD4+ and CD5+ Treg subpopulations may be involved. ²⁵ Moreover, SSc-PBC patients may have clonally expanded CD8+ T cells that express a receptor, which has been speculated to be involved in the pathogeneses of the diseases. ²³ The fact that shared genetic bases have to be considered in SSc and PBC susceptibility has been demonstrated by data from a recent study in which PBC susceptibility single-nucleotide polymorphisms (SNPs) were genotyped in SSc patients. This analysis reported that in addition to the already known STAT4 and IRF5 variants, NF-κB and PLCL2 PBC variants were also associated for SSc together with also IRF8 locus. ³⁰

Prevalence of PBC in SSc, clinical features and prevalence of specific Abs

The prevalence of PBC in SSc varies according to the different studies but may be estimated at about 2% to 3%. However, the prevalence of PBC-specific Abs in SSc population may increase considerably. PBC is more frequent in lcSSc than in dcSSc and in SSc patients with ACA positivity. Akimoto et al. ³¹ compared two populations – lcSSc-PBC and lcSSc alone – and showed that ACA positivity was more prevalent in SSc-PBC patients who also had a higher frequency of calcinosis and telangiectasia. In an SSc population, AMA were detected by indirect immunofluorescence (IIF) on Hep-2 cell substrate and enzyme-linked immunosorbent assay (ELISA) using an M2-enhanced (MIT3) assay that contains three epitopes recognized by AMA. ³² The authors investigated also by ELISA anti-sp100 and anti-gp210 Abs. In this SSc population, the prevalence of PBC was of 2%: the diagnosis was made satisfying two or more of the following criteria characterized by AMA positivity, increased hepatic enzymes and liver biopsy. They observed that SSc/PBC patients had a higher frequency of limited skin involvement, although the distribution of lcSSc was the same in the two populations (SSc alone vs SSc/PBC). All SSc/PBC patients were ACA positive, and no difference was found in clinical features or frequency of internal organ involvement when compared to SSc patients without ACA. In addition, ACA and TopoI were associated to an increased alkaline phosphatase levels differently than in anti-RNAP patients. This study also suggested that the combination of AMA and anti-sp100 Abs may allow to increase the sensitivity for PBC detection compared to the use of AMA only. Similar data are reported by the recent study of Imura-Kumada et al., ³³ where the authors investigated the prevalence of PBC and of PBC-specific Abs in 225 Japanese patients with SSc (106 with lcSSc and 119 with dcSSc) fulfilling the 1980 ARA criteria or with at least three of the CREST features (calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia) and ACA positive. AMA, anti-sp100 and anti-gp210 Abs were tested by ELISA: 16.4% of patients were positive with AMA (30 with lcSSc and 7 with dcSSc) and of those, 59.5% were diagnosed with PBC. Anti-sp100 was found in 13/37 patients (three of them also had AMA), and in this case the majority of subjects presented a dcSSc (7 vs 6). Three patients had anti-gp201 positivity (two with lcSSc and one with dcSSc, of them two also presented AMA positivity). Interestingly, out of 177 patients without PBC-specific Abs, 13 had high serum levels of biliary enzymes without evidence of other liver diseases ((hepatitis C virus (HCV), hepatitis B virus (HBV), drug-induced liver disease or cholelithiasis). Five patients were subjected to liver biopsy with histopathological findings compatible with PBC. The study also included a multivariate analysis on AMA and ACA that seemed to be independently associated with PBC in this population of SSc patients suggesting that these Abs may represent an increased risk in the development of PBC in SSc. Even regarding AMA positivity in SSc patients, E Wielosz et al. recently reported that out of the 86 SSc enrolled patients, 11 presented AMA positivity and 10 PBC. The AMA Abs and the PBC prevalence were higher in lcSSc patients if compared to dcSSc ones. In addition, SSc patients with AMA Abs presented an overlap syndrome with more than one CTD more frequently than SSc subjects AMA negative. In addition, ACA Abs were more frequent in patients with AMA positivity than in those without AMA. ³⁴ In this study, no differences in SSc was observed in the two groups (AMA positive vs AMA negative), conversely to the study of Cavazzana et al. ³⁵ that showed a less frequency of telangiectasia and calcinosis in patients with AMA positivity if compared to subjects without AMA. This latter article indicated a prevalence of PBC-specific Abs in SSc similar to that reported by previous studies (21%) and confirmed the high prevalence of ACA in SSc patients with specific-PBC Abs. Similarly, GL Norman et al. investigating the prevalence of AMA, anti-gp210 and anti-sp100 in an SSc population composed by 52 patients (33 lcSSc and 19 dcSSc) reported that 15% of subjects presented PBC-specific Abs. In this study, no difference was indicated in clinical features and SSc subsets in the two groups (SSc vs SSc with PBC Abs), and patients with AMA, anti-gp210 or anti-sp100 did not have increased liver enzymes. ³⁶ The higher prevalence of liver Abs in SSc patients when compared to healthy controls has been reported also by another recent study that enrolled 63 SSc patients and 100 controls, all investigated for AMA, liver–kidney–microsomal (LKM1) and anti-smooth muscle antibodies (SMA). The prevalence of these Abs was of 14.3%, in particular 9.5% of SSc patients presented AMA positivity, of them four had lcSSc, two had a form characterized by overlap features and none had dcSSc. No patients with liver Abs presented abnormal liver enzymes or an impaired liver function. ³⁷ See Table 2 for a summary of exposed data.

Table 2.

PBC and PBC-specific antibodies prevalence in SSc.

Study	Number of SSc patients enrolled in the study	PBC prevalence in all studied SSc population	Differences of PBC prevalence in lcSSc and dcSSc	PBC-specific Abs in all studied SSc population	Differences of PBC-specific Abs prevalence in lcSSc and dcSSc
Assassi et al. 32	817	2% (16/817)	13/16 lcSSc 3/16 dcSSc	AMA (MIT3) 7% (56/817) sp100 (ELISA) 3% (26/817) gp210 (ELISA) 4% (3/817)
Imura-Kumada et al. 33	225	15% (35/225)		AMA 16.4% (37/225) sp100 5.8% (13/225) gp210 1.3% (3/225)
Wielosz et al. 34	86	12% (10/86)	9/10 lcSSc 1/10 dcSSc	AMA 13% (11/86)	10/11 lcSSc 1/11 dcSSc
Cavazzana et al. 35	201			MIT3 17.9% (36/201) sp100 2.5% (5/201) gp210 0.5% (1/201)	lcSSc in 80% (32/40)
GL Norman et al. 36	52			MIT3 7/52 sp100 1/52	No significant differences
TL Skare et al. 37	63			AMA 9.52% (6/63)

PBC: primary biliary cholangitis; lcSSc: limited cutaneous SSc; dcSSc: diffuse cutaneous SSc; Abs: antibodies; AMA: antimitochondrial Abs; ELISA: enzyme-linked immunosorbent assay.

Recent data suggested that SSc patients with PBC may progress to liver cirrhosis; however, in this population mortality is more commonly caused by SSc organ involvement and consequent complications than to PBC that seemed to be less aggressive in overlap syndrome.^10,23 In accordance to these data, also the above-mentioned study of Rigamonti et al. ³⁸ in which PBC patients were compared to PBC-SSc subjects reported a lower liver progression in this last group when compared to PBC alone population.

Prevalence of SSc in PBC, clinical features and prevalence of specific Abs

Regarding the prevalence of SSc in patients with PBC, many studies investigated the frequency of the rheumatic disorder and of its specific Abs. A recent study confirmed the association between PBC and SSc and tried to identify if extrahepatic autoimmune (EHA) conditions may influence the incidence of major complications lied to PBC. Authors showed that EHA conditions seem to not have any impact on endstage liver disease complications. The majority of enrolled patients were AMA positive; however, the group of patients with EHA conditions was comparable for autoantibodies (AMA, antinuclear antibodies (ANA), SMA, LKM), histological stage and clinical features if compared to subject without EHA conditions. Other important data from the study were the fact that the prevalence of EHA conditions was stable during the follow-up (8.0 ± 6.9 years) and that females were more prevalent in the group PBC and EHA conditions (to remark the fact that 339/361 patients were females). ³⁹ In addition, a recent review investigated the EHA conditions associated to PBC confirming that more than 70% of patients with PBC presented extrahepatic manifestations, of them the more common were thyroid diseases and SjS. SSc was confirmed to be associated to PBC with a prevalence ranging from 1.4% to 12.3%, in particular the lcSSc subset. ⁴⁰ Tovoli et al. investigated the prevalence of early SSc in PBC patients compared to a control group composed by subjects with a chronic liver disease other than PBC. Early SSc was defined by the presence of Raynaud’s phenomenon (RP) associated to scleroderma-specific Abs and NVC scleroderma pattern. Out of 80 PBC patients, more than 18% of these presented an early SSc. In addition, also the single SSc features (RP, specific SSc-Abs, videocapillaroscopy pattern) were more present in PBC patients than in the control ones. Some subjects of the control group presented RP (5.6%), however never together with other signs of SSc. ⁴¹ A recent study published in this year investigated the prevalence of SSc in a population of PBC patients using the new 2013 ACR/EULAR classification criteria. The authors examined 100 PBC patients founding a higher prevalence of SSc using the new ACR/EULAR criteria if compared to that reported with the older ones (1980 ARA) (17% vs 1%). These data suggested that PBC patients should be screened for the presence of SSc investigating for clinical, laboratory and instrumental features of the disease in order to early recognize the rheumatic disease, perhaps underestimated in the past with the old criteria. ⁴² The frequency of CTDs has been evaluated also in another recent Chinese study in which 322 patients with diagnosis of PBC according to the criteria of the American Association for the Study of Liver Diseases have been enrolled. 87% of patients had ANA positivity and nine subjects (2.8%) were classified as having SSc according to the 1980 ARA criteria – these patients presented a significant higher incidence of interstitial lung disease (ILD) if compared to the others. The frequency shown in this study was lower than that reported in other ones, probably due to the fact that in China many SSc patients with a cutaneous involvement referred to dermatologists. ⁴³ Interestingly, M Bektas et al. recently investigated the presence of manometric alteration in PBC patients and in a control group composed by patients with functional dyspepsia. Data from this study revealed that oesophageal dysmotilities were detected only in the PBC population with a prevalence of 45.9% (17/37 patients). Knowing the frequent gastroesophageal involvement in some CTDs, in particular SSc and SjS, the same study reported that out of 17 patients with manometric alterations, 8 also presented SSc. ⁴⁴ Therefore, the study seems to suggest to investigate for SSc or other CTD patients with PBC symptomatic for heartburn, dysphagia and/or epigastric pain, as gastroesophageal involvement is very frequent in these rheumatic disorders. In addition, two recent articles investigated the role and the significance of ACA in patients with PBC. These Abs are characteristic of SSc; however, they are present also in other autoimmune diseases as SjS, SLE and PBC. In the study of S Mandai et al., the authors enrolled 37 patients, of those 32% presented ACA positivity showing no differences in sex, blood pressure, body mass index (BMI) and liver function tests in the two groups (ACA+ vs ACA–). Only the estimated glomerular filtration rate (eGFR) was significantly different in the two populations, as ACA patients had a lower value and a more decline during the follow-up independently of PBC progression. ⁴⁵ Recently, a article reported the clinical characteristics of PBC patients with ACA highlighting that they were more likely to develop portal hypertension.^10,46 Studies reported that ACA positivity could be found not only in SSc-PBC but also in subjects with PBC alone without the evidence of a coexisting SSc. ACA might be considered as predictor of a CTD development, in particular of SSc, ²² but it still remains to be clarified whether ACA are a marker of a future development of SSc or if they identify a specific subset of PBC. ¹³ Regarding ACA positivity, it is known that the major centromere proteins detected by ACA are represented by CenpA, CenpB, CenpC and CenpH. Recently, CenpI was found in 25.8% of SSc/CTD subjects and 62.5% of them also presented PBC-specific Abs. According with these results, anti-CenpI may be a marker for a concomitant autoimmune liver disease in SSc patients. ⁴⁷

Conclusion

The association of SSc and PBC has been described since 1970, and a prevalence of PBC in SSc population has been found to be about 2%. The fact that the prevalence of PBC is higher in SSc than in the general population suggests to study the liver function and to test AMA Abs in SSc population.

The aetiology of SSc and PBC remains still unknown; however, the pathogenesis of these two autoimmune diseases includes environmental, infectious, genetic and epigenetic factors that together may lead the induction of the development of the disorders. The two diseases are both characterized by a fibrogenic response that in SSc patients leads to skin and internal organ fibrosis and in PBC to bile duct fibrosis. As highlighted by recent findings, both diseases are characterized by the presence of profibrotic cytokines, in particular TGF-β and IL6, that seem to have a role in the mesenchymal cell activation. These cytokines are also responsible for the generation and function of Th17 cells and Treg, and the role of Th17 cells in SSc pathogenesis has been suggested by different studies. In addition, recent findings support the importance of these immune cells also in the PBC pathogenesis.

The association between SSc and PBC is also indicated by the presence of PBC-specific Abs in SSc patients and of SSc-specific Abs in PBC population. ACA are characteristic of SSc; however, they can be found both in SSc-PBC patients and in patients with only PBC. Patients with PBC and ACA positivity, but without diagnosis of SSc, have to be carefully screened for the presence of SSc clinical symptoms and of its potential internal organs’ complications. Regarding the proteins detected by ACA, recent findings reported that CenpI may be considered as a marker of a concomitant autoimmune liver disease in SSc population. In addition, despite these data may suggest a possible pathogenic role of ACA Abs, further studies are needed to confirm this hypothesis. In addition, the fact that liver progression was lower in patients with SSc-PBC if compared to PBC population may suggest that it could be an overlapping entity with some specificities as compared to PBC or SSc alone.

Data from this review seem to suggest that the association of SSc and PBC may identify a group of patients with a milder systemic disease, ¹⁵ which is of importance for the clinician managing such patients. Patients with PBC and SSc present less frequently PBC complications as portal hypertension. On the other hand, data regarding SSc patients showed that this population is more frequently affected by an lcSSc or by a milder disease characterized by ACA positivity and without major organ involvement as pulmonary hypertension or interstitial lung disease.

The role of liver biopsy in SSc patients with PBC-specific Abs positivity still does not seem completely clear; probably it should be reserved to patients with Abs positivity but without signs of liver dysfunction or to those with increase in cholestasis enzymes but without PBC-specific Abs. However, a strong collaboration with a gastroenterologist or with a hepatologist is recommended.

Altogether, the association between SSc and PBC, ‘two autoimmune and profibrotic diseases’, is very well known today. However, more efforts are needed to better understand the pathogenesis and pathophysiology of SSc-PBC to recognize early the patients, and to better understand this ‘overlapping autoimmune disorder’ and its possible evolution in the years. Future investigations and studies should also investigate about possible therapeutic strategies in these patients.