Author names and institutions: Jennifer Cathcart, Ashis Mukhopadhya, Hepatology department, Aberdeen Royal Infirmary, Scotland

Learning Points

1. An individual approach to management and early referral to liver transplant, if needed.
2. To identify those at greatest risk of disease progression, it is recommended that all patients have individualised risk stratification using biochemical response indices following a year of UDCA therapy.
3. Asking about symptoms, documenting presence or absence, and offering treatments in outpatient follow-up are important to improve quality of life.

Introduction

Primary Biliary Cholangitis (PBC), formerly called primary biliary cirrhosis, is an autoimmune disease leading to the destruction of bile ducts, subsequent cholestasis and progressive fibrosis that leads to cirrhosis over time. Approximately 20,000 people in the UK are affected. Most are women and >50 years old when diagnosed. [1] The majority have a slow progression to cirrhosis but an important subset of patients have rapidly progressive disease leading to liver failure requiring liver transplantation.

Presentation and Diagnosis

Figure 1. Diagnostic pathway. AMA: anti-mitochondrial antibodies. PBC specific ANA: includes anti-nuclear-rim, anti-nuclear-dot or anti-centromere antibodies found on immunofluorescence or anti-gp210 or sp-100 by ELISA 

Most cases are asymptomatic and detected after abnormal cholestatic liver function tests, namely a raised Alkaline Phosphatase (ALP) but also Gamma-Glutamyl Transferase (GGT). [2] A polygonal elevation of IgM could be suggestive of PBC. A raised ALP for over 6 months and the presence of anti-mitochondrial antibodies (AMA; >1 in 40) or highly specific antinuclear antibodies (ANA) is diagnostic of PBC. [2,3] PBC-specific ANA includes anti-nuclear-rim, anti-nuclear-dot or anti-centromere antibodies found on immunofluorescence or anti-gp210 or sp-100 detectable by ELISA. Centres vary in offering immunofluorescence or ELISA for detecting PBC specific ANA but there is no superiority between tests [2]. ANA antibodies are detectable in approximately 50% of subjects with PBC. [4] Imaging is used to exclude other biliary and infiltrative conditions; ultrasound alone is normally sufficient. This is typically normal. In seronegative patients, MRCP +/- EUS can be helpful. MRCP can exclude large duct primary sclerosing cholangitis (PSC) and EUS can exclude distal biliary disease. Liver biopsy is reserved only for those with unclear aetiology, diagnosis overlap syndrome and to diagnose true sero-negative disease. It should be noted that AMA can react in patients with metabolic liver disease and a liver biopsy may be needed identify the dominant disease. Small duct PSC can be excluded by liver biopsy if there is a high index of suspicion. True overlap with autoimmune hepatitis is rare and liver biopsy is often needed to make the diagnosis. Liver pathology shows lymphocytic infiltrate in portal tracts with granulomatous destruction and loss of medium sized interlobular bile ducts. If liver biopsies do not show any cause for cholestasis, then consider genetic studies e.g. BRIC, ABCB4, ABCB11 etc.

Variant presentations

• AMA-positive with normal LFTS: The clinical significance of AMA-positive patients with normal liver biochemistry is currently unclear. Most will go on to develop PBC symptoms and biochemical signs but progress less to complications, albeit in limited studies. [2] The British Society of Gastroenterology (BSG) guidance suggests that neither biopsy nor UDCA therapy is recommended in this group. They state that follow-up with annual LFTS can be done in primary care (following initial assessment in the secondary setting). If tests become abnormal, patients should either be re-referred to secondary care or have UDCA commenced if the abnormality in ALP is for longer than 6 months.
• AMA-negative PBC: The majority of patients who are AMA-negative will be positive for the PBC-specific nuclear antibodies [2]. However, true autoantibody negative PBC exists although it is rare and requires biopsy diagnosis.
• PBC/AIH overlap syndrome:Affects a small minority of patients with PBC and is diagnosed via biopsy.

Management

Management is three-fold;

• Treatment and risk stratification according to treatment response
• Surveillance of disease including staging of fibrosis
• Symptom management

Treatment and risk stratification according to treatment response

Whilst there is no cure for PBC, treatment aims to prevent or delay disease progression. However, treatment options are limited.

1. The first line treatment is Ursodeoxycholic Acid (UCDA) at a dose of 13mg-15mg/kg/day lifelong. The number needed to treat (NNT) for UCDA is 11. [5] Risk assessment after 1 year is recommended to be performed as per guidelines with any recognised biochemical strategy risk score. [2] The different scoring models are listed below in Table 1. If non-response, we suggest that a discussion at an MDT or tertiary centre takes place for escalating treatment.
2. Second line treatment is Obeticholic Acid (OCA) to be started after intolerance or insufficient response to UCDA. The starting dose is 5mg/day increased to 10mg/day at 6 months. Treatment with OCA is associated with a dose-dependent exacerbation in pruritus leading to treatment discontinuation in 1–10% of patients. [2] Additional safety concerns exist with the use of OCA, these are hepatic decompensation and hepatic failure. In the pivotal POISE trial, patients treated with OCA showed decreases in alkaline phosphate that differed significantly from placebo. [6] However, following real-life clinical use of OCA there were a small number of case reports of hepatic toxicity, decompensation and death following OCA treatment in patients with cirrhosis. [7] Furthermore a small in retrospective cohort study involving 21 patients on OCA with compensated cirrhosis showed increased decompensation but not liver-related mortality or transplantation. [8] However a study by Perez et al, comparing patients treated with OCA in the POISE trial with non-OCA treated patients from the GLOBAL PBC real-world database, found that OCA demonstrated a statistically significant 58% reduction in relative risk of hepatic decompensation, liver transplant or death compared to real life patients not treated with OCA. [9]. Furthermore, in the post-Hoc analysis of the POISE trial, there was no safety concerns in those treated with OCA with cirrhosis. [10] Whilst evidence of harm with OCA must be taken into consideration, it remains difficult to distinguish a drug-induced effect from disease progression in the patients with liver disease. As case reports were mostly from patients with advanced liver disease, the Food and Drug Administration (FDA) subsequently restricted its use in 2021 in patients with advanced cirrhosis. [7] Such restrictions are not in place in the UK. However, the BSG guidance does recommend, as per the drug label, the OCA dose is reduced in patients with cirrhosis. They recommend 5mg weekly initially (with a maximum dose of 10mg twice weekly) in Child Pugh B or C liver disease. In patients with Child Pugh A with evidence of portal hypertension consideration should be given to dose reduction. In addition, they recommend frequent monitoring (e.g. repeat blood tests monthly at outset) and, at any development of decompensation or progression, OCA is stopped or the dose adjusted. [2]
3. Additional off-label medications or clinical trials can be considered after discussion at an MDT or tertiary centre. Fibrates and budesonide have been used but large trials are needed. In one randomised control trial published in the New England Journal of Medicine with 100 patients, treatment with bezafibrate in addition to ursodeoxycholic acid had significant biochemical response rate to that of the placebo.[11] Several clinical trials are currently active for PBC and patients should be informed and enrolled as appropriate.

Prognostic models in PBC

Table 1 lists the range of scoring models that are used in PBC to predict response to treatment and prognosis to death or liver transplant. There is not one scoring system that is favoured in the recent BSG guideline.

Table 1 list of PBC scoring models

Surveillance of disease including staging of Fibrosis

The intensity of follow-up varies between patients and should be individualised. Non-invasive fibrosis staging at baseline and on treatment should be performed for to identify those at high risk. BSG guidelines recommend serum blood tests looking for high bilirubin, platelets <150 or biochemical activity despite treatment, liver ultrasound to look for overt cirrhosis, and transient elastography to identify the stage of fibrosis. [2] Table 2 shows suggested threshold cut-offs for FibroScan from studies. [12] Liver biopsy is not recommended for staging. This is because liver tissue can have patchy pathology with more severe disease missed on biopsy. There are reports of all stages of disease from 1 to 4 (cirrhosis) being found on explanted livers. [13].

Table 2 – Fibrosis cut-offs

Frequency of follow-up

In the BSG PBC guideline, in those who are low risk (without cirrhosis, ALP <1.67 x ULN and normal bilirubin after one year of treatment with UDCA) it is recommended for annual LFTS and documented repeated risk assessment every 3 years. [2] The BSG guidelines state that low risk patients can be followed up in primary care if appropriate shared pathways exist [2].  However, a younger age at presentation and male gender, are risk factors for progressive disease and may warrant closer follow-up. The Scoring models in table 1 help to divide patients into high or low risk groups. There are no suggested time intervals for high-risk groups in the BSG guidance or the European association for the study of the liver (EASL) guidance but the national institute for clinical excellence (NICE) recommends 6 monthly surveillance ultrasound for hepatocellular carcinoma in all adults with cirrhosis. [14] Patients with hepatic decompensation, intractable pruritus and elevated UKELD score or elevation of bilirubin over 50umol/l should be referred for transplant assessment. Fatigue in isolation does not respond well to liver transplant and is not an indication. [2]

Screening for varices and complications for portal hypertension

Roughly, a third of patients with advanced disease develop varices over approximately 5 years. [2] All patients should be considered for endoscopic screening but a number of risk scoring systems are available which may decrease the number of endoscopies required. These are the Baveno VI guidelines (please note that PBC patients are not well represented in studies) and the Newcastle Varices Score. The BSG guidelines do highlight that patients with PBC can develop varices because of non-cirrhotic portal hypertension although these cases are uncommon. In terms of other portal hypertension-related complications, patients with PBC appear to develop hepatic encephalopathy at a smaller frequency than in other liver diseases. [2] However, when present should be managed as per conventional management.

Active management of symptoms

Symptoms are important to ask specifically in outpatient clinics due to the impact of quality of life. However, the severity of symptoms does not correlate with the stage of the disease. [2]

• Fatigue, 50% of patients will have fatigue and 20% will experience fatigue that impacts on their quality of life. [2] This is difficult to treat and is subjective. There are no specific evidence-based treatments. Consider treating other medical causes of fatigue such as depression, sleep apnoea, anaemia and thyroid disease as these can be effectively treated. Fatigue in isolation is not an indication for transplant. All patients can benefit from patient support groups such as the PBC foundation. [1]
• Itching, up to 75% of patients experience pruritus. [15] Excluding bile duct obstruction is needed. Intractable pruritus unresponsive to medical therapy is an indicator for transplant and consideration for available clinical trials.

Table 3 Treatment for Itch [16]

• Sjogren’s syndrome is common in PBC. Patients can also have ‘sicca complex’ but not Sjogren’s disease. Treatment is with artificial tears and saliva.
• Bone assessment for osteoporosis and to prevent fractures should be done in all patients. The FRAX or Qfracture scores can be performed and subsequent DXA scan if initiated. Treatment and follow-up should be according to national guidelines (standard is 80% assessment within the last 5 years). [2]

Conclusion

For several years the pharmacological treatment for primary biliary cholangitis hasn’t changed but management has improved as our understanding of disease progression grows. An individual approach to management to include assessing risk factors and response to treatment and early referral to liver transplant if needed is key.