Risk stratification and prognostic modelling in primary biliary cholangitis

Abstract

Primary biliary cholangitis (PBC) is a slowly progressive chronic cholestatic liver disease that, in a subgroup of patients, may result in liver failure or death. The definition of specific risk profiles, i.e. risk stratification, is of critical importance for the identification of these subgroups and thereby the targeting of care. Over the last few years large multicentre cohort studies have improved our knowledge regarding factors associated with progressive disease. Stratification based on biochemical response to ursodoxycholic acid provides a readily available measure to identify groups that might benefit from additional therapies to further improve prognosis. In addition, serum total bilirubin and alkaline phosphatase are now considered the most robustly validated biomarkers of long-term outcome in PBC and are used as endpoints in clinical trials. The GLOBE score and UK-PBC risk score enable us to quantify the risk of future events for the individual patient, allowing more individualized risk prediction. In this review, we discuss both established prognostic factors and newly developed tools to estimate prognosis in PBC, highlighting their strengths, limitations and applicability in clinical practice.

Introduction

Primary biliary cholangitis (PBC) is a slowly progressive chronic cholestatic liver disease [1,2]. The disease has an indolent course in the majority. However, there is a subgroup with an impaired survival associated with poor response to treatment with ursodeoxycholic acid (UDCA) [[3], [4], [5], [6], [7], [8], [9], [10], [11]]. For decades, UDCA was the only available drug to treat PBC. Currently, the clinical scenario is changing with new agents becoming available, all with different profiles of efficacy and tolerability. In this scenario, risk stratification is a critical tool to identify patients in need of second-line therapies or patients who can be safely continued on UDCA monotherapy. In addition, risk stratification has an important role in trial design.

Over the past decade, our ability to identify subgroups of patients with a higher chance to have a progressive disease, and to estimate the risk of future adverse events based on readily available clinical parameters has considerably improved. Newly developed prediction models now enable us to quantify the risk of future events for individual patients and provide important tools in clinical practice for patient counselling, timing of diagnostic procedures and therapeutic interventions, and selection of patients for clinical trials. This review discusses factors associated with prognosis in PBC, and in particular focuses on biochemical response criteria and prediction models. Especially, this review provides insights in the application and generalizability of the various models and tools in clinical practice.

PBC is considered a rare disease with incidence rates varying between 0.3-5.8 per 100.000 persons per year and a prevalence ranging between 1.9-40.2 per 100.000 inhabitants [[12], [13], [14]]. PBC predominantly affects middle-aged women, with a female to male ratio of 10:1 [1,2], although recent studies suggest an increasing male prevalence [15]. Diagnosis is suspected in cases of chronically elevated markers of cholestasis and/or symptoms of fatigue and pruritus, which are the most commonly reported symptoms in PBC [16]. However, at present most PBC patients are diagnosed in (early) asymptomatic stages of disease [[17], [18], [19], [20]].

Thus far, PBCs pathogenesis has not completely been elucidated, but is thought to be an interplay of genetic susceptibility and environmental triggers that result in immunoregulatory changes and selective destruction of intrahepatic cholangiocytes [[21], [22], [23]]. This may lead to progressive fibrosis, cirrhosis, and eventually liver failure requiring liver transplantation (LT) or resulting in premature death [1,2].

Although PBC is a slowly progressive disease in the majority of patients, the clinical course may differ greatly between patients and prognosis is largely dependent on development of cirrhotic complications [24]. The introduction of UDCA as mainstay treatment has greatly affected the clinical disease course [16,25]. UDCA has been shown to improve liver biochemistry [26], can delay histological progression [[27], [28], [29]] and impacts on long-term outcome [26,[30], [31], [32]]. At present, most patients are diagnosed at earlier stages of disease and are usually promptly treated with an adequate dose of UDCA [20]. However, despite adequate treatment, patients can remain at risk of developing cirrhosis and associated complications.

An early study showed that approximately 20% of UDCA-treated patients progress to cirrhosis within 4 years after start of treatment [33]. Later, Corpechot et al. showed that annually 7% of UDCA-treated patients progress to extensive fibrosis or cirrhosis [28]. A study with 254 paired biopsies over 655 patient-years showed that the rate of histological progression to cirrhosis under UDCA treatment from fibrosis stage I, II, and III was 4%, 12% and 59% respectively [34]. Median time to cirrhosis in these stages was 25, 20, and 4 years, respectively [34]. A recent cohort study by Trivedi et al. in 511 patients showed a cumulative incidence of cirrhosis development of 40% over a 10-year follow-up period, indicating that nowadays a substantial percentage of patients still develops cirrhosis [35,36]. Nonetheless, the ten-year transplant-free survival of UDCA-treated PBC patients is considered good and is approximately 80% [[4], [5], [6],17,30,31,[37], [38], [39]]. Over recent years the transplantation rates for PBC have decreased in North America and Europe [40,41].