Primary Biliary Cholangitis—The Changing Biomarker Paradigms for Staging Fibrosis

Abstract

Primary biliary cholangitis (PBC) is an autoimmune-mediated disease characterized by chronic, non-suppurative, small-duct lymphocytic cholangitis. The prognosis largely depends on early disease recognition and treatment. Suboptimal response to first-line therapy (ursodeoxycholic acid) is associated with risk for disease progression. Reliable biomarkers are also required to enhance risk stratification. The traditional gold standard for assessing fibrosis is liver biopsy, but it is invasive and unsuitable for serial evaluations. Hence, trends are towards non-invasive surrogate biomarkers (blood-based and imaging biomarkers respectively) which have a much better safety profile. Blood-based biomarkers include: (i) Fibrosis-4 [Fib-4], (ii) Aspartate Aminotransferase to Platelet Ratio Index [APRI], (iii) Enhanced Liver Fibrosis score [ELF], and (iv) total bile acid to platelet ratio [TPR]. They show much potential but are not particularly sensitive tests. Ultrasound-based imaging biomarkers are increasingly being utilized for liver stiffness measurement (LSM), with vibration-controlled transient elastography (VCTE) emerging as the preferred technique. However, despite its growing popularity, VCTE is limited by technical issues. Hence, currently, none of the non-invasive tests fulfill the prerequisites to be the new gold standard as defined by the FDA. Nonetheless, there may be value to combining LSM with various serum biomarkers such as Fib-4, APRI, as aforementioned. The hope is to create nomograms for predicting liver-related events and decision tree algorithms. Newer studies are investigating microbiota in the gut-liver axis, biomolecules such as nanovesicles/nanofibers, and metabolic reprogramming as it pertains to e.g., proteomics and lipidomics. These approaches hold much promise, and if validated, could significantly change the management of PBC.

1. Introduction

Primary biliary cholangitis (PBC) is a chronic autoimmune liver disease characterized by damage and destruction of interlobular and proximal septal bile ducts [1,2]. It primarily affects middle-aged women with a peak incidence in the 40 to 60-year age group. Despite the strong 9:1 female predominance, recent studies suggest that epidemiologic shifts may be underway with increasing male incidence [1,3,4]. For example, the age-adjusted incidence of PBC in the United States per 1 million person-years for women is 45, and 7 for men [5]. Although PBC occurs worldwide and has been found in all races, there is considerable geographic variation e.g., more cases reported in North America and Europe compared to the Asia-Pacific region or Africa [3,4]. However, unlike other autoimmune liver diseases, it is exceptionally rare in the pediatric age group.

Most patients with PBC initially present without any symptoms and the disease is discovered incidentally through routine bloodwork for other conditions. The abnormal test is often an elevated alkaline phosphatase (ALP), which then triggers other tests, including antimitochondrial antibody (AMA). Not surprisingly, the true incidence of PBC is rising as more screening tests, such as liver chemistry and lipid profiles, are performed in otherwise healthy individuals [1,2]. For those patients who do have symptoms, fatigue is the most common complaint, being documented in approximately 80%, and often rated as the most disabling. The underlying mechanism of the fatigue is poorly understood, and there is no specific treatment for it [6]. Another common symptom is pruritus, which can be severe and widespread, often causing insomnia. Patients may also have abdominal pain, arthralgia, depression and cognitive dysfunction. In addition, they are at a higher risk of having other autoimmune disorders, such as Sjogren’s syndrome, Hashimoto’s thyroiditis and rheumatoid arthritis, which may concurrently present with their own symptoms. With disease progression, features of chronicity start to appear. Cutaneous xanthelasmas may develop, a manifestation that harkens back to the 1940s when the disease was called xanthogranulomatous biliary cirrhosis [7]. Other manifestations of advanced disease (e.g., jaundice, skin hyperpigmentation, edema and bleeding) are basically reflective of portal hypertension, cirrhosis or liver failure.

The etiology is thought to be multifactorial, with genetic susceptibility, environmental triggers and epigenetic factors being implicated [8,9,10]. The genetic predisposition is inferred from the strong prevalence of the disease in first-degree relatives, particularly daughters of women with PBC [11,12]. A high degree of disease concordance in monozygotic twins has also been observed [13]. A number of human leukocyte antigen (HLA) allele associations have been reported with PBC, the most prominent being DQA1*0401, DQB1*0301, DQB1*0402, DQB1*0602, DRB1*08, DRB1*0803, DRB1*11, and DRB1*1101; some of these alleles may overlap with other autoimmune diseases [14,15,16]. The putative environmental triggers include various xenobiotics such as bacteria, drugs, plant constituents, cosmetics, industrial chemicals, food additives, fragrances, toxic waste, hair dyes and cigarette smoking. It is thought that, in genetically susceptible individuals, these environmental triggers induce an autoimmune reaction to an intracytoplasmic antigen that is manifested by the presence of AMAs. The AMAs target the lipoyl domain on the E2 subunit of the pyruvate dehydrogenase complex (PDC-E2) located on the inner mitochondrial membrane.

2. Cholangiocyte Pathophysiology

Cholangiocytes are exposed to very high concentrations of hydrophobic bile acids, which have detergent effects on the cell membrane. These bile acids are sensed by receptors located on the cholangiocyte membrane, leading to an increased secretion of chloride and bicarbonate, and, consequently, to stabilization of the biliary bicarbonate umbrella. The process is mostly regulated by the bicarbonate-chloride exchanger protein called anion exchanger 2 (AE2). Studies suggest that altered intracellular pH regulation, mediated by AE2 downregulation, predisposes cholangiocytes to damage by hydrophobic bile acids [10,16,17,18,19]. This leads to senescence, dysregulated autophagy, defective clearance of apoptotic debris, and expression of intact PDC-E2 on their cell membrane, potentially enabling an autoantibody-mediated immune attack [10,16,20]. Thus, according to this hypothesis, bile duct injury is initiated by increased bile acid toxicity, which then creates a cascade that leads to autoimmunity.

3. Autoantibodies

AMAs are detected in >90% of PBC patients, and being rare outside PBC, they are quite specific. Hence, they are regarded as the serologic hallmark of PBC [21,22]. However, they are not the only autoantibodies found in this condition. Antinuclear antibodies (ANAs) can be detected in 30–50% of PBC patients. Amongst the ANAs, anti-gp210 and anti-sp100 are highly specific for PBC, making them particularly useful in the diagnosis of PBC patients who are negative for AMA [23,24,25]. Furthermore, while AMA is not associated with disease progression, these ANAs are associated with disease severity and confer a worse prognosis compared to conventional PBC patients [23,24,25]. Two different PBC disease progression patterns have been observed with ANAs: anti-gp210 is associated with hepatic failure type progression whereas anti-centromere antibodies are associated with portal hypertension [23,24,25,26]. Within the context of autoimmune liver disease, the ANA positivity due to anti-double-stranded (ds-DNA) antibodies is considered a specific serological marker of the overlap syndrome AIH-PBC [27,28].

4. The Treatment Response

Ursodeoxycholic acid (UDCA), which modulates bile acid metabolism, is an FDA-approved first-line treatment for PBC and the current standard-of-care [1,2,4]. When administered in the early stages of the disease, UDCA can improve biochemical tests, lessen symptoms, slow progression, and potentially delay end-stage liver disease. However, up to 40% of patients have a suboptimal response and are identified as high-risk i.e., more likely to progress to cirrhosis and liver failure unless transplantation is carried out. This raises the question about the definition of treatment response to UDCA. While there is currently no consensus, the common denominator relates to the degree of improvement in biochemical surrogates of cholestasis, particularly as measured by ALP and/or bilirubin, AST or GGT. The duration of treatment also varies from 6 to 24 months depending on the protocol e.g., the Rochester, Paris, Barcelona, Toronto or Rotterdam protocol. Recent guidelines recommend additional medications such as FXR agonists (e.g., obeticholic acid), PPAR ligands (e.g., bezafibrate) or other novel and repurposed drugs for use as combination therapies in the high-risk patients [1,2,4]. Nevertheless, it can take a long time to determine the treatment response. Thus, reliable biomarkers that can identify high-risk patients early are urgently needed for refining therapy strategies [1,2,4,28,29,30]. The liver biopsy is one of the time-honored biomarkers that has been used to assess such PBC patients (

Table 1. Summary table of PBC tissue biomarkers highlighting key features.

Type of Biomarker	Key Features	Comments
Liver biopsy	Traditional gold standard	Invasive; therefore, suboptimal for serial evaluations
(i) Percutaneous	Currently most common procurement method	Less acceptable to patients due to pain and other complications
(ii) EUSB	Less pain and shorter recovery time compared to percutaneous biopsy. Greater accessibility to multiple sites on both right and left liver regardless of body habitus [31,32]	More acceptable to patients. Gaining momentum and becoming more popular. Includes real-time imaging during procedure. Sample size comparable to percutaneous biopsy. Concurrent EUS portal pressure gradient measurement [33,34,35,36].
(iii) Transjugular	Useful in patients with coagulopathies, ascites, liver transplants, extreme obesity, fulminant liver failure, vascular tumor.	Biopsy samples often small, fragmented and suboptimal. Allows hemodynamic evaluations of hepatic and portal venous systems.

Abbreviations: EUSB: Endoscopic ultrasound biopsy.

).

5. Liver Biopsy

In the proper clinical context, a diagnosis of PBC can be established on the basis of any 2 of 3 criteria based on traditional biomarkers: (i) persistent unexplained biochemical cholestasis, (ii) positive anti-mitochondrial antibodies, and (iii) compatible liver histology (Figure 1) [1,4]. Though liver biopsy is not mandatory for the diagnosis, strong consideration should be given to procuring it under these circumstances: (i) in patients without PBC-specific autoantibodies, (ii) for the PBC-AIH variant since these patients can benefit from immunosuppressive therapy, (iii) to identify concurrent liver disorders such as metabolic dysfunction-associated steatotic liver disease (MASLD), and (iv) for risk stratification in PBC after diagnosis [2,4,37].

The pathognomonic histologic features of PBC can be found in the early stages (Figure 2) but the typical lesion can be difficult to identify in advanced disease where fibrotic scars can obscure interlobular bile ducts making them hardly discernible [38,39]. As a result, the value of staging core biopsies has been somewhat downgraded due to the uneven distribution of the pathology within the liver as a whole. Nonetheless, the disease has traditionally been classified into 4 histologic stages in an attempt to understand its progression and natural history [40,41].

6. Traditional Histologic Staging

Over the years, many histologic staging systems have been developed. However, the most well-known were based on the works of Scheuer and Ludwig [40,41]. They each proposed their own system but, by and large, the classifications are broadly similar, namely:

• Stage 1: Chronic non-suppurative cholangitis with or without the florid duct lesion (Figure 2)
• Stage 2: Bile duct proliferation/interface hepatitis (Figure 3)
• Stage 3: Septal fibrosis/bridging necrosis (Figure 4 and Figure 5)
• Stage 4: Cirrhosis (Figure 4 and Figure 5)

Though this classification is conceptually useful, it assumes that liver histology worsens sequentially and that there is a sequential progression from stage 1 to stage 4. However, there is a marked individual variation in the rate of disease progression. In addition, staging has inherent difficulties that must be recognized. For example, it is common to see several stages in one biopsy specimen or liver explant, and at times, all 4 [42,43,44]. One could take the position that staging should best be based on the most advanced histology, but this is not without problems. Thus, a small percutaneous core biopsy may miss an advanced lesion. On the other hand, a subcapsular biopsy may exaggerate the pathologic stage. Sampling variability may also explain the suboptimal correlation between pathologic stage and symptoms. Visualization of the liver at laparoscopy may be the most accurate way to recognize cirrhosis, but unfortunately it is neither practical nor desirable in most cases. Notwithstanding these problems, liver histology remains a crucial resource for hepatologists in evaluating disease severity. For example, specific pathologic features such as interface hepatitis, ductopenia, cholestasis, and degree of fibrosis have management and prognostic implications [2,4,22]. At the same time, it should be recognized that major pathophysiologic concepts and treatments have been introduced since the visionary work of Scheuer and Ludwig in staging PBC in the 1960s and 70s. The question now is whether newer approaches can improve on the historical classifications [40,41].

7. Newer Histologic Classification

A more recent histologic classification of PBC has been proposed by Japanese investigators [45,46,47]. The main difference from the classical systems is that it separates histologic staging from grading. In essence, the fixed lesions (fibrosis, bile duct loss and deposition of copper-binding proteins) (Figure 5) are used for staging while the necroinflammatory lesions (cholangitis activity and hepatitis activity) are used for grading. Each of the 3 fixed lesions is semiquantitatively assessed on a 4-point score (0–3), and the sum of the scores is computed into an overall stage (1–4). The grading of each of 2 necroinflammatory criteria for activity is also assessed on a 4-point score 0–3 [45,46,47,48].

Just as with the Scheuer and Ludwig systems, the newer approach correlates well with clinical and laboratory findings [45,46,47]. Over and above that, the grading of the necroinflammatory activity may help to identify those patients who would benefit from budesomide treatment in combination with UDCA [39,47]. Overall, while this classification is meritorious, it is quite complex and more prone to interobserver variability due to the multiplicity of variables involved. This makes it difficult to use in the day-to-day reporting of liver biopsies [39,44]. Furthermore, the various criteria for staging and grading respectively are based on a composite score that may reflect different patterns of injury. For example, stage 4 in this system does not necessarily equate to the cirrhosis of the traditional classifications, and this can be problematic from a clinical viewpoint [39]. Hence, there is an ongoing impetus aimed at developing other biomarkers to stratify patient risk according to the degree of fibrosis. Unlike liver biopsy, which can potentially cause complications such as bleeding, perforation and pneumothorax, non-invasive liver disease assessments (NILDA) have a much better safety profile (

Table 2. Summary table of PBC blood-based biomarkers showing key features.

Type of Biomarker	Key Features	Comments
Blood-based	Simple, inexpensive, easily used in the clinical setting	Anticipated increase in utilization in line with trends for molecular testing
(i) Persistent elevation of ALP > ×1.5 ULN for >24 weeks.	Often the first clue of PBC. Discovered incidentally through routine blood tests for other conditions.	Concurrent elevations of GGT and ALP substantiate that the elevated ALP is of hepatobiliary origin and increase prognostic value of ALP measurement [49]
(ii) Raised AMA titers	Most specific serologic marker amongst >60 autoantibodies; present in 90% of patients; often antedate clinical signs by months or years. Multiple AMA subtypes but M2 is the most specific	Titer of AMAs is not associated with disease progression or the patient’s clinical course [21]
(iii) ANA: specifically -Anti-sp100 and anti-gp210	Helpful in diagnosis of AMA-negative PBC	Associated with more severe PBC phenotypes
(iv) Hyperbilirubinemia	Often a late manifestation	Generally indicative of poor prognosis

Abbreviations: ALP: Alkaline phosphatase. AMA: anti-mitochondrial antibodies. ANA: anti-nuclear antibodies. GGT: Gamma-glutamyl transferase. ULN: Upper limit of normal.

).

8. Non-Invasive Liver Disease Assessments

8.1. Serum Biomarkers

Since the degree of liver fibrosis is the critical determinant in the development of liver-related events, various NILDAs have been developed to predict fibrosis (Figure 6).

Of these, serum biomarkers offer much appeal since they are generally simple and inexpensive and can be easily used in the clinical setting (Table 2). They can be combined with other parameters to create liver scores such as: (i) Fibrosis-4 score (Fib-4), (ii) Aspartate Aminotransferase to Platelet Ratio Index (APRI), (iii) Enhanced Liver Fibrosis (ELF) score (iv) Hyaluronic acid and total bile acid to platelet ratio (TPR) [50,51,52,53,54]. However, a number of studies have found that they are not particularly sensitive or reproducible in assessing the fibrosis. Furthermore, caution is required in interpreting the results since they can be affected by the patient’s systemic condition [49,50,51,52].

8.2. Metabolic Reprogramming in Primary Biliary Cholangitis

Traditional markers sometimes fail to correlate with the extent of liver injury or disease progression in PBC. This underscores the desire for alternative biomarkers e.g., those involved in metabolic reprogramming that can more accurately reflect disease dynamics and guide therapeutic decisions [55,56,57]. An emerging area of research in this regard are complement proteins, involving both classical and non-classical pathways. Proteomics has shown that complement protein signatures such as CR1, C1QA, C1QL2, C7, and C9, are considerably elevated in PBC patients, and show high accuracy in differentiating PBC from non-PBC individuals [58]. In some instances, they significantly outperform traditional liver function markers such as FIB-4, ELF, or TPR [40,41,42,43,44]. Furthermore, these complement proteins are strongly associated with PBC onset and adverse outcomes such as cirrhosis and liver failure, potentially making them useful tools for early diagnosis, risk stratification and outcome prediction [58].

The metabolic reprogramming does not only involve proteomics but also lipid and carbohydrate pathways. For example, studies have shown distinct lipidome remodeling in PBC with alterations in polyunsaturated fatty acid, sphingolipids and serum HDL-cholesterol levels, and augmented fatty acid β-oxidation [55,59]. Furthermore, PBC patients showed high levels of glucose-6-phosphate and purines, as well as a reduction in pyruvate, suggesting impaired glycolysis and increased purines biosynthesis [55]. All in all, targeting specific metabolic reprogramming pathways may offer potential targets for therapeutic intervention in PBC.

8.3. Non-Coding RNAs in Primary Biliary Cirrhosis

Non-coding RNAs (ncRNAs) have emerged as promising predictive and prognostic biomarkers in PBC due to their key regulatory roles in autoimmunity [60,61,62]. They encompass 3 principal subtypes: microRNAs, long non-coding RNAs, and circular RNAs [63]. Collectively, they orchestrate gene expression networks by interacting with DNA, RNA, and proteins, thereby influencing transcriptional programs, translational efficiency, and post-translational modifications. Of particular relevance to PBC, ncRNAs can be encapsulated and transported within exosomes to generate stable exosomal ncRNAs that mediate intercellular communication and pathophysiological signaling across tissues [64,65].

Although most studies show that plasma ncRNAs expression profiles are heterogenous across the PBC continuum, emerging evidence suggests that certain ncRNAs (e.g., MiR-126-3p and exosomal lncRNA H19) can potentially complement AMAs as diagnostic biomarkers and/or predictive markers [60,61]. Furthermore, recent work has demonstrated the differential expression of miR-122-5p in serum exosomes of PBC patients and healthy controls by high-throughput sequencing. It also showed that hepatic stellate cell-derived exosomal miR-122-5p can regulate the expression of inflammatory factors in cholangiocytes through the p38 MAPK signaling pathway [62]. Put together, this recent work on ncRNAs and exosomes holds much promise for biomarkers in the pathophysiology of PBC.

8.4. Genetic Biomarkers

Since the pathogenesis of PBC invokes genetic susceptibility, extensive investigations aimed at the mechanistic basis of risk-associated variants have been conducted. Studies have shown that a number of genes such as TXNIP, CD44, and ENTPD1 are upregulated in high-risk PBC. Another one is CDKN1a; its upregulation correlated with significantly increased expression of the senescence marker p21^WAF1/Cip [66]. Other investigators also found that high-risk PBC was characterised by upregulation of genes linked to T-cell activation, apoptosis, INF-γ signalling and leukocyte migration. In contrast, the genes linked to the complement pathway were downregulated. Related to these developments, tumor necrosis factor alpha-induced protein was associated with treatment response, and as such, could provide new insights into disease progression [67]. Hence there are promising biomarkers for monitoring disease progression and risk stratification of PBC [29]. In addition, Genetic analysis has a prospective role to play from a diagnostic viewpoint, especially in AMA-negative patients who have reservations about invasive liver histologic testing. Candidate genes such as ITGAL have potential to substantiate the diagnosis in PBC [68]. However, since the genetic architecture of PBC is complex, involving both HLA and non-HLA risk loci, further studies are required to analyze the polygenic traits involved [69].

8.5. Imaging Biomarkers

Imaging biomarkers are steadily gaining ground in PBC management since they can be used to assess changes in morphology, texture, or perfusion [70]. From the viewpoint of assessing fibrosis, elastography has taken an increasingly prominent role because it is quite amenable to evaluating liver stiffness measurement (LSM) (

Table 3. Summary table of PBC imaging biomarkers highlighting key features.

Type of Biomarker	Key Features	Comments
Imaging	Uses LSM as a surrogate marker for fibrosis. Encompasses much larger area compared to tissue biopsy; hence, more representative	Non-invasive. Minimal risk compared to tissue biopsy
(i) VCTE	Generally cost effective; More accessible.	Considerable intra- and inter-observer reproducibility. Accuracy limited by multiple variables e.g., age, body habitus, BMI, ascites.
(ii) MRE	Higher diagnostic accuracy. High equipment cost. Higher level training and expertise.	Better at assessing more advanced fibrosis (stage 3 and 4) than earlier stages

Abbreviations: BMI: Body mass index. LSM: Liver stiffness measurement. MRE: Magnetic resonance elastography. VCTE: Vibration-controlled transient elastography.

). It can be implemented on ultrasound or magnetic resonance elastography (MRE) [70,71]. Various ultrasound-based methods are currently available such as vibration-controlled transient elastography (VCTE), point shear wave elastography and two-dimensional shear wave elastography [72]. Of these, VCTE appears to have emerged as the preferred technique [70,71,72,73,74]. Although MRE has higher accuracy than VCTE, it is more expensive, requires more equipment and training, and therefore tends to be restricted to higher level institutions (Table 3). In contrast, VCTE is more widely available and can even be performed at the point-of-care in most health care centres [73,75] (Figure 7).

Using VCTE, several studies have established thresholds for distinguishing between low-, medium- and high-risk patients respectively, based on whether LSM values are below 8 kPa (low-risk) (≤F1) or above 15 kPa (high-risk) (F4) or in between (medium-risk F2/F3) [76]. However, the utility of LSM as a prognostic tool in PBC is evolving, especially with respect to the cut-offs e.g., the respective values in other studies are <6.5 kPa and >11 kPa [1]. The differences in cutoffs can be due to any of several factors: (i) different definitions of low-, medium- and high risk depending on the study e.g., F0–F2, F3, and F4 respectively (or ≤F1, F2–F3, F4) (ii) the cut-offs can be influenced by the type of equipment, probe size, calibration, clinical practice guidelines and training depending on geographic location, (iii) histologic staging system that is used for validation e.g., Scheuer, Ludwig, Metavir, Brunt, Kleiner or Japanese system), (iv) age and gender [76,77,78,79,80]. There are ongoing efforts to standardize guidelines, but universally accepted guidelines have not yet been established.

Despite the growing popularity of VCTE, its utility is tempered by a number of limitations e.g., (i) technical: inter- and intra-patient variability, (ii) confounding factors such as extreme body habitus, acute inflammation, congestive cardiac failure, ascites, biliary obstruction, amyloidosis, and recent food intake, (iii) suboptimal performance in early disease (stages 1–2) [81,82]. In addition, studies have shown that a significant proportion of patients with initially high LSM can have normal LSM at a subsequent assessment, with nil or mild fibrosis on liver biopsy. Overall, it is estimated that the rate of unreliable or failed results is 10–15%. Hence patients with abnormal LSM scores should consider repeated testing before embarking on liver biopsy or treatment [70,71,81,82]

Taken together, it would appear that LSM assessed by VCTE improves prediction of survival beyond what is offered by biochemical parameters. Since a significant proportion (40%) of patients with PBC are at medium or high risk of poor clinical outcome, LSM can potentially be regarded as a useful prognostic marker and surrogate endpoint in routine care and clinical trials.

9. Variants of Autoimmune Liver Diseases

While the majority of PBC cases conform to the classic disease description, a small proportion have concurrent features of autoimmune hepatitis (AIH). These cases are classified as AIH-PBC variants (overlap syndrome). Their recognition is important because they mostly respond to immunosuppressive therapy. Likewise, AIH can have features resembling primary sclerosing cholangitis (PSC), the AIH-PSC variant. This highlights the importance of recognizing the 3 main entities comprising autoimmune liver disease i.e., AIH, PBC and PSC (

Table 4. Summary table of the 3 main autoimmune liver diseases.

	PBC	AIH	PSC
Demographics	Female predominance (90%). Median age 50 years. Exceptionally rare in children.	Affects all ages including children and all populations. Female 75% [83]	Male predominance (70%). Median age 30 years. Does occur in children but prevalence 20% lower than in adults [84,85]
Associated diseases	Sjogren’s syndrome Hashimoto’s thyroiditis, scleroderma, Raynaud’s disease, and celiac disease [57,86].	Thyroiditis, arthritis, Sjogren syndrome, vitiligo, glomerulonephritis	Strong association with IBD (70%) particularly UC, Pancreatitis, Retroperitoneal fibrosis
Anatomic/histologic site	Predominantly interlobar and proximal septal bile ducts	Liver parenchyma	Predominantly extrahepatic bile ducts, or both extra- and intrahepatic. Small duct PSC is rare (<5% of patients)
Characteristic histology	Nonsuppurative destructive cholangitis; florid duct lesion	Portal lymphoplasmacytosis, interface hepatitis, necrosis, hepatocyte rosettes, emperipolesis	Multifocal biliary strictures; concentric periductal fibrosis.
Conventional blood-based biomarkers	Chronic cholestatic biochemical profile: elevated ALP. Presence of AMA is serological hallmark of disease. IgM often high; helps to distinguish it from AIH [44]	ALT activity is often purported as the biochemical hallmark of AIH. IgG levels typically high, while IgM levels are low [44,87].	Elevated ALP is characteristic, along with GGT. No diagnostic serum autoantibody test [88]. Various non-specific autoantibodies (ANA, ASMA, anti-cardiolipin, Rheumatoid factor).
Serology	>90% AMA-positive 50% ANA-positive 40% ANCA-positive	ANA, ASMA, ALKM1, ALC1A, ASLA > 20, Absence of viral hepatitis [89,90]	0–5% AMA-positive 5% ANA-positive >60% ANCA-positive but not disease specific. Therefore of limited value to establish the diagnosis
Imaging	VCTE, MRE	Inflammation is potential confounder for LSM. Thus, limited role e.g., assessment of complications such cirrhosis and HCC.	ERCP or MRCP are the gold standard for diagnosis [44]
Treatment	UDCA	Responds favourably to immunosuppressive treatment e.g., Corticosteroids, azathioprine	No effective medical therapies exist for PSC
AIH (overlap) variants. Paris criteria [91,92]	About 10–20% of PBC patients may present AIH	AIH-PBC more common than AIH-PSC	PSC-AIH variant may reflect a PSC subphenotype
Complications Clinical course	cirrhosis, HCC	Acute liver failure, cirrhosis, HCC	Strong association with cholangiocarcinoma (10–20% of cases); colorectal cancer, cirrhosis, HCC

Abbreviations: AIH: Autoimmune hepatitis. ALC1A: anti-liver cytosol type 1 antibody. ALKM1: anti-liver kidney microsomal antibody type 1. ANCA: Anti-neutrophil cytoplasmic antibody. ASLA: Anti soluble liver antigen. ERCP: Endoscopic retrograde cholangiopancreatography. HCC: Hepatocellular carcinoma. IBD: Idiopathic inflammatory bowel disease. MRCP: Magnetic resonance cholangiopancreatography. PSC: Primary sclerosing cholangitis. UC: Ulcerative colitis. VCTE: Vibration-controlled transient elastography.

).

Due to the inherent difficulties of establishing the correct diagnosis for a suspected variant syndrome, such patients should receive a complete work-up with liver histology, serology and imaging. For the majority of patients with either AIH or PBC, the focus is on early treatment i.e., immunosuppressants and UDCA respectively, since this can significantly change outcomes. The relative efficacy of these treatments has driven ongoing interest in clinical trials for additional regimens including novel and repurposed drugs [1,91,92]. Unfortunately, to date, no established medical therapies exist for PSC. Hence, for PSC, the emphasis appears to be directed at diagnostic and prognostic markers that can provide further insights into disease pathogenesis and effective therapies [93].

10. Future Directions

In the past decade, significant progress has been made in our understanding of the pathogenesis, diagnosis and prognostic biomarkers of PBC. An emerging area of research in this regard is metabolic reprogramming in proteomics, purine biosynthesis, glycolysis, and fatty acid metabolism. In particular, further exploration of complement dysregulation pathways could offer opportunities for early diagnosis, risk stratification, and personalized therapeutic strategies to improve patient outcomes [55,56,57,58]. However, since the majority of the biomarker studies were typically carried out on small groups of patients, this limits statistical power and affects the robustness and generalizability of the conclusions reached. Therefore, one of the key challenges is to conduct validation studies with large patient cohorts. Upon validation, diagnostic and prognostic biomarkers could also be used as surrogate endpoints in clinical trials, especially for patients with high-risk PBC.

Another area of continuing active investigation in PBC is the gut-microbiome and gut-liver axis. Recent studies suggest that microbiota dysbiosis has an adverse effect on treatment in PBC [94,95,96]. These studies report that the gut flora is less diverse in PBC patients compared to the general population. Furthermore, there is loss of certain protective bacteria (e.g., Bacteroides) coupled with enrichment in others (e.g., Pseudomonadota) [94,95,96]. However, further research is required to elucidate the interaction of microbiota and metabolite signatures in PBC patients.

A separate question that needs to be addressed is whether established fibrosis (F4) can be reversed in a clinically discernible manner. To date, this remains a formidable challenge. Hence, current management approaches are largely palliative with the goal of slowing disease progression, minimizing symptoms and preventing complications. However, recent approaches using biomolecules such as fabricated supramolecular carrier-free bilirubin nanofibers, hyaluronic acid–bilirubin nanoparticles or nano-bioconjugated vesicles, though still experimental, provide new perspectives for the treatment of liver fibrosis [97,98,99,100].

In summary, although UDCA has revolutionized treatment, alternative and adjunctive treatments are an area of active investigation. Since novel drug targets are based on complex pathophysiology, the hope is that future studies will elucidate additional pathophysiologic features, including the aforementioned metabolic reprogramming, gut-liver microbiome axis and biomolecules.

11. Conclusions

Liver biopsy is the traditional gold standard for evaluating fibrosis, the main predictor of liver-related morbidity and mortality [56,69]. However, since it is an invasive procedure, it has limited utility especially for diseases like PBC that require serial assessments [70]. According to the US Food and Drug Administration, to replace liver histology as a surrogate endpoint in clinical trials, a number of guiding principles have to be fulfilled before a NIT can be ratified for that specific purpose. First prerequisite pertains to the context of utilization, which defines the intended use of the NIT and the decisions that will emanate from the results i.e., the biomarker qualification pathway in which the FDA approves the use of the NIT. In this particular case, the intended use would be to replace liver biopsy in the assessment of fibrosis for diagnosis, disease monitoring, treatment response and prognostic purposes. The second important determinant of the NIT is the analytical robustness of the tool or assay to be used, particularly with respect to accuracy and reproducibility. The third key component is clinical performance i.e., the ability of the NIT to accurately reflect a biological phenomenon, e.g., fibrosis. Lastly, the NIT should demonstrate clinical utility i.e., the benefit to the patient in terms of outperforming what is currently available [58,71]. Currently, VCTE it does not appear to fulfill all these prerequisites to be the new gold standard. However, there may be value in combining it with various serum biomarkers to create nomograms for routine clinical practice [72,73,76]. Indeed, similar nomograms incorporating LSM have been constructed for predicting liver-related events in other hepatic diseases e.g., AGILE 3+, AGILE 4 and FibroScan-aspartate aminotransferase (FAST) scores in MASLD [58,73,74]. The serum biomarkers that can be co-opted into this predictive consortium of NITs include Fib-4, APRI, ELF, BHI and TPR. The hope is to create nomograms for predicting liver-related events and decision tree algorithms [40]. Indeed, studies have shown that combinations of NITs can improve the accuracy of detecting patterns of liver involvement in patients with or suspected of having various liver diseases e.g., MASLD [69,70].