Dual cut-offs and beyond: Expanding the role of transient elastography in primary biliary cholangitis

Abstract

Accurate fibrosis staging is pivotal in the management of primary biliary cholangitis (PBC). Although liver biopsy is considered the diagnostic gold standard, its invasiveness limits widespread use in disease monitoring. Chen et al provided strong evidence validating vibration-controlled transient elastography (VCTE) as a noninvasive alternative for identifying advanced fibrosis in Chinese patients with PBC. By establishing dual cut-offs (≤ 10.0 kPa and > 14.5 kPa) with high diagnostic performance, the authors offer a clinically applicable tool that could substantially reduce the need for biopsy. Nevertheless, challenges persist, including the management of intermediate “grey zone” results, technical variability, and the confounding effects of cholestasis and inflammation. The integration of elastography with biochemical and prognostic indices, such as the GLOBE and UK-PBC scores, is essential for individualized care. This editorial discusses the expanding role of VCTE in PBC, current limitations, and future research directions toward standardized, integrated, noninvasive fibrosis assessment, and individualized PBC care.

Key Words: Primary biliary cholangitis; Vibration-controlled transient elastography; Liver stiffness measurement; Prognostic value; Liver fibrosis staging

Core Tip: Transient elastography is transforming the fibrosis evaluation in primary biliary cholangitis (PBC). The dual cut-offs proposed by Chen et al (≤ 10 kPa and > 14.5 kPa) achieved excellent diagnostic accuracy and could eliminate the need for biopsy in most patients. However, interpretation requires attention to confounders such as cholestasis, inflammation, and overlap syndromes. Standardized protocols, cross-modality validation, and integration with clinical scores will define the next frontier of precision staging for PBC.

INTRODUCTION

Primary biliary cholangitis (PBC) is a chronic autoimmune cholestatic liver disease characterized by the progressive destruction of intrahepatic bile ducts and variable fibrosis, leading to cirrhosis, portal hypertension, and hepatic failure[1]. The assessment of fibrosis severity remains central to prognosis, treatment decisions, and surveillance of complications. Liver biopsy is essential for diagnosis and prognosis in select PBC cases, especially when noninvasive markers are inconclusive. Its main advantages are risk stratification and detection of overlap syndromes; however, its invasiveness, sampling error, and limited necessity in typical cases are significant drawbacks that make it unsuitable for repeated fibrosis staging[2]. Over the last decade, noninvasive tools based on serum biomarkers [aspartate aminotransferase to platelet ratio index (APRI), fibrosis-4 (FIB-4), and enhanced liver fibrosis (ELF)] and elastography techniques have redefined hepatology practice[3]. According to recent meta-analyses, vibration-controlled transient elastography (VCTE) (FibroScan^®) has emerged as the most accessible and reproducible modality for quantifying liver stiffness measurement (LSM) in PBC, reflecting the degree of liver fibrosis[4-6]. While widely validated in viral hepatitis and metabolic dysfunction-associated steatotic liver disease, its application in PBC has historically been limited by small sample sizes and a lack of ethnicity-specific cut-offs. The study by Chen et al[7], recently published in World Journal of Gastroenterology, addressed these limitations by validating dual VCTE thresholds for advanced fibrosis (≤ 10 kPa to rule out, > 14.5 kPa to rule in) using a large biopsy-verified cohort and external validation (Figure 1). This represents a major contribution to evidence-based risk stratification in Asian patients with PBC.

Figure 1 Assessment of liver fibrosis in primary biliary cholangitis using vibration-controlled transient elastography according to the results of study by Chen et al[7]. Vibration-controlled transient elastography provides a reliable non-invasive alternative to liver biopsy, showing excellent diagnostic accuracy for advanced fibrosis (area under the receiver operating characteristic curve: 0.93-0.97) and outperforming standard serum-based fibrosis markers. A dual liver stiffness cut-off strategy (low risk ≤ 10.0 kPa; high risk > 14.5 kPa) allows advanced fibrosis to be confidently excluded (negative predictive value 93%) or confirmed (positive predictive value 92%), leaving an indeterminate “grey zone” (10.0-14.5 kPa) where additional evaluation is needed and potentially avoiding up to 92.5% of liver biopsies in patients with primary biliary cholangitis. PBC: Primary biliary cholangitis; VCTE: Vibration-controlled transient elastography; AUROC: Area under the receiver operating characteristic curve; FIB-4: Fibrosis-4; LSM: Liver stiffness measurement.

DIAGNOSTIC PERFORMANCE AND CLINICAL UTILITY OF DUAL CUT-OFFS

Chen et al[7] evaluated VCTE in > 200 Chinese PBC patients with histological correlation. The area under the receiver operating characteristic curve (AUROC) for detecting advanced fibrosis (≥ F3) was 0.93, and the proposed dual cut-off approach allowed confident classification in > 90% of cases while maintaining sensitivity and specificity > 85%. Patients with stiffness below 10 kPa had a very low probability of advanced fibrosis, whereas those with stiffness above 14.5 kPa could be diagnosed non-invasively with high accuracy. This “rule-in/rule-out” framework mirrors earlier European data by Corpechot et al[8], Corpechot et al[9] and Cristoferi et al[10]; however, the present study extends its generalizability by including a large Asian sample and an external validation cohort. These data were confirmed by using the models combining LSM and alkaline phosphatase values which were developed using Asian retrospective PBC cohorts. The AUROC values for two different models exceed 0.90 and was better than any other non-invasive markers[11]. Importantly, the results of Chen et al[7] align with recent meta-analyses, which confirmed pooled AUROCs of 0.90-0.94 for advanced fibrosis and 0.93-0.96 for cirrhosis detection by VCTE in PBC (Table 1)[4-6]. Clinically, the dual cut-off approach can reduce unnecessary biopsies, guide therapy escalation, and stratify surveillance intensity. When combined with biochemical remission status, LSM helps refine the risk beyond conventional serology-based scores.

Table 1 Diagnostic accuracy of vibration-controlled transient elastography for advanced fibrosis/cirrhosis vs liver histology.

Ref.	Reported AUROC/Sens/spec	Cut-offs (kPa)
Cristoferi et al[10]	AUROC: AF: 0.89	AF > 11.0
Osman et al[39]	AUROC: F4: 0.94	F4: 14.4
Zhang et al[37]	AUROC: F4: 0.90	F4: 13.2
Chen et al[4]1	AUROC: SF: 0.93; AF: 0.93; F4: 0.91	AF: 9.6-10.7; F4: 14.4-16.9
Manzo-Francisco et al[22]1	Sens/spec: F3: 68%/92%; F4: 90%/94%; AUROC: F3: 0.91; F4: 0.97	Mean cut-off: F3: 9.28; F4: 15.2
An et al[5]1	AUROC: SF: 0.87; AF: 0.89; cirrhosis: 0.99	AF: 7.5-17.9; cirrhosis: 11.4-25.1
Marcos Carrasco et al[6]1	Sens/spec: ≥ F2: 0.76/0.93; ≥ F3: 0.88/0.90; = F4: 0.91/0.95	NA

¹Meta-analysis.

Sens: Sensitivity; Spec: Specificity; AUROC: Area under the receiver operating characteristic curve; AF: Advanced fibrosis; SF: Significant fibrosis; NA: Not applicable.

Despite its strengths, transient elastography faces several interpretive challenges. Approximately 10% of patients fall within the “grey zone” (10-14.5 kPa), where overlap between adjacent fibrosis stages precludes clear categorization. In the study by Chen et al[7] 14.4% of patients from both the derivation and validation cohorts fell within the grey zone, and in 27.5% of them, Ludwig stage IV was found by histology. In these cases, a multimodal approach using LSM with serum biomarkers (ELF and APRI) or imaging features (spleen size and portal flow) is usually recommended. Recently, spleen stiffness measurement (SSM) with the latest FibroScan models was introduced, which may help predict decompensation in patients with PBC. Thus, patients with LSM ≥ 10 kPa but SSM ≤ 40 kPa remain compensated during 32 months of follow-up[12].

Several biological confounders can influence LSM, independent of fibrosis. It has been shown that cholestasis, reflected by elevated bilirubin/alkaline phosphatase, can transiently increase stiffness through biliary pressure and edema and influence the staging of fibrosis and prediction of esophageal varices[13]. Liver inflammation, particularly with high alanine aminotransferase (ALT) elevation, may cause reversible stiffness overestimation, which is typical in untreated autoimmune hepatitis[14]. PBC-autoimmune hepatitis overlap syndrome is characterized by prominent necroinflammatory activity that may elevate LSM beyond histologic fibrosis. However, a study of 70 patients with overlap syndrome showed that LSM was superior to non-invasive tests (FIB-4 and APRI) in detecting advanced fibrosis compared to liver biopsy, with impressive AUROCs of 0.91 to 0.96 for F3 and F4 stages of liver fibrosis[15]. The prevalence of liver steatosis in patients with PBC varies significantly (from 20% to 90%)[16-19] likely due to the size of the study cohorts, variations in diagnostic methods, and differences in patient populations. There are also contradictory conclusions about the role of liver steatosis in the long-term outcomes of patients with PBCs. Thus, a retrospective study of 363 Chinese patients with PBC showed that non-alcoholic fatty liver disease does not significantly impact the biochemical response to ursodeoxycholic acid but may improve the degree of liver fibrosis and long-term prognosis[19]. In contrast, a retrospective Spanish study of 129 PBC patients found that concomitant metabolic dysfunction-associated steatotic liver disease significantly worsened the prognosis of PBC[18]. Nevertheless, the only study that addressed the relationship between steatosis determined as the controlled attenuation parameter (CAP) score and liver stiffness in PBC patients showed that higher CAP values (P = 0.02, odds ratio = 3.11) were independently associated with higher liver stiffness values and appeared to impact liver stiffness[17]. Therefore, LSM interpretation in patients with PBC must consider the biochemical context, disease activity, and treatment status, as re-measurement after biochemical stabilization may also improve the accuracy of LSM.

Technical aspects such as the fasting state, appropriate probe selection (M vs XL types of probe), and adherence to quality criteria (≥ 10 valid LSM, interquartile range/median ≤ 30%) are also important[3]. Quality-controlled acquisition and transparent reporting are essential, as vendor-specific algorithms and inter-observer differences complicate cross-study comparability. Recently, equipment for transient elastography from different vendors has overestimated liver stiffness compared to FibroScan[20], or its development is in the early stages with the first evaluation reports published[21]. Novel liver stiffness cut-off values need to be defined for this equipment based on the results of well-planned comparative studies. The growing literature on elastography underscores the methodological variability. Heterogeneity in study design partly explains the divergent cut-offs reported worldwide[9,10,22]. Moreover, referral bias toward tertiary centers may have inflated the diagnostic accuracy estimates. Community-based validation and inclusion of early-stage patients are needed for real-world applicability. The study by Chen et al[7] is the first large study of the real-world performance of transient elastography in assessing advanced fibrosis in Chinese patients with PBC. Multicenter clinical trials registries may enable pooled analyses stratified by sex, age, body mass index, and disease duration that can modestly influence LSM independent of fibrosis. Such analyses may help establish better universal cut-offs for different stages of liver fibrosis in patients with PBC.

PROGNOSTIC AND LONGITUDINAL VALUE OF VCTE IN PBC

Beyond static staging, longitudinal VCTE measurements provide robust and independent prognostic information for patients with PBC, enabling risk stratification and prediction of clinical outcomes. Large multicenter studies have validated that both baseline and serial LSM using VCTE are strong, independent predictors of poor clinical outcomes in PBC, such as liver-related complications, transplantation, or death. LSM improves prognostic accuracy beyond established biochemical response criteria and fibrosis scores, with cut-offs of 10 kPa and 15 kPa optimally separating low-, medium-, and high-risk groups, respectively. Approximately 40% of patients fall into the medium- or high-risk categories based on these thresholds. The predictive value of LSM remains stable over time, supporting its use as a surrogate endpoint in clinical trials and for routine monitoring[8,23]. A “rule-of-five” gradient (10 kPa, 15 kPa, 20 kPa, and 25 kPa) has been proposed to dynamically stratify risk[24]. Importantly, LSM retains prognostic value independent of the biochemical response. In patients who achieve biochemical remission but have persistently high LSM, the risk of disease progression remains substantial[25]. Conversely, fibrosis regression under treatment is reflected by declining LSM values.

Longitudinal changes in LSM are clinically meaningful: A significant increase ≥ + 20% or any increase to ≥ 15 kPa is associated with a substantially higher risk of adverse outcomes, whereas a significant decrease (> - 20% or any decrease to < 10 kPa) is linked to a reduced risk. Thus, monitoring LSM dynamics over 6 months, 12 months, or 24 months can guide prognosis and therapeutic decisions[12,23,26]. Stable LSM and SSM over time are associated with a low risk of decompensation, whereas increases in these parameters predict disease progression[12]. Combining LSM with SSM and platelet count further refines risk stratification, particularly for portal hypertension-related complications. An SSM ≤ 40 kPa can rule out high-risk esophageal varices, reducing unnecessary endoscopies. The combination of LSM and SSM is particularly useful for identifying patients at risk of liver decompensation[12,27]. When integrated with GLOBE and UK-PBC scores, LSM improves the discrimination of transplant-free survival[8,28]. This reinforces VCTE as not only a diagnostic but also a dynamic prognostic biomarker in PBC.

ACCESS TO VCTE

VCTE has transformed non-invasive fibrosis assessment in chronic liver disease; however, access for patients with PBC remains incomplete and lags behind the historical experience in viral hepatitis. The VCTE was first developed and extensively validated in chronic hepatitis C and B, and early guideline recommendations, device deployment, and reimbursement pathways were largely organized around staging viral hepatitis for treatment decisions and public health programs. In contrast, current PBC practice guidelines still base the initiation and reimbursement of second-line therapies almost exclusively on biochemical response criteria, whereas liver stiffness is primarily recognized as a prognostic marker rather than a gatekeeper for drug prescription[3]. Recent outcome studies showing that VCTE markedly refines risk prediction and identifies patients who derive the greatest survival benefit from deep biochemical response have reinforced its potential role in optimizing the timing of second-line therapy; however, these data have not yet been translated into VCTE-driven prescribing rules[8]. Real-world data clearly illustrate access gaps. In a large French-Belgian cohort, LSMs at the time of PBC diagnosis were available for 45% of patients[29]. In Spanish cohorts treated with obeticholic acid or fibrates, VCTE data were available in only a quarter of the patients[30]. In Germany, the PBC registry reports VCTE results in 61% of patients overall, with higher coverage in academic than in non-academic centers (66% vs 50%), and only approximately 140 FibroScan devices are available nationwide, highlighting capacity constraints even within high-income countries[31]. While individual patients with PBC who reach hepatology centers often have theoretical access to VCTE, the historical viral hepatitis focus, lack of VCTE-linked prescribing or reimbursement criteria, and heterogeneous insurance coverage collectively translate into lower and more inconsistent VCTE use in PBC than in chronic viral hepatitis, excluding VCTE monitoring despite comparable prognostic value. Moreover, only approximately half (53.5%) of patients with PBC in the United States, irrespective of therapy use, were most recently treated for PBC by a specialist[32], limiting access to VCTE monitoring. According to the data of a large tertiary center of more than 9000 patients during 8 years, VCTE found advanced liver fibrosis in 22% of patients with cholestatic liver disease, but 34% showed no fibrosis[33]; therefore, subsequent VCTE monitoring in the majority of patients with PBC will be determined by local guidelines, insurance policies, and healthcare settings.

VCTE ALTERNATIVES

Alternative imaging modalities, such as magnetic resonance elastography (MRE) and shear-wave elastography (SWE) (ARFI or 2 dimensional-SWE), offer additional insights into tissue mechanics[34,35]. MRE provides three-dimensional stiffness mapping with broader sampling, and meta-analyses suggest slightly higher AUROCs (0.93) for advanced fibrosis detection[36]. However, MRE is costly, time-consuming, and less accessible, which limits its routine use. ARFI and 2 dimensional-SWE are ultrasound-based techniques that are integrated into conventional scanners. Their diagnostic accuracy in PBC appears comparable to that of VCTE (AUROC approximately 0.89-0.92)[37,38], although device-specific calibration and operator dependency remain limitations. However, head-to-head comparisons of these techniques are limited. A multicenter study by Osman et al[39] demonstrated no significant difference between VCTE and MRE in identifying ≥ F3 fibrosis, emphasizing VCTE’s practicality given its portability and cost-effectiveness. However, cross-modality validation using harmonized thresholds and identical histological endpoints remains a priority. Emerging evidence supports the use of SSM as a complementary parameter for portal hypertension[40]. The Baveno VII criteria advocate the combination of LSM and SSM to noninvasively predict high-risk varices, an approach that could be extended to PBC once validated.

FUTURE RESEARCH DIRECTIONS

A global consensus on PBC-specific VCTE thresholds remains a priority. Device calibration across vendors, correction for probe-type differences, and adjustment for bilirubin or ALT levels may help refine the cut-off accuracy. Future algorithms should integrate VCTE with serological (ELF, FibroTest) and clinical scores (GLOBE, UK-PBC) into composite indices to improve diagnostic accuracy. Machine learning approaches may enable the individualized prediction of outcomes based on combined metrics. Large prospective trials comparing VCTE, MRE, and ARFI within the same cohort and histological reference are required. Such designs could elucidate modality-specific strengths, especially in the “grey zone”. In PBC-autoimmune hepatitis overlap or coexisting metabolic dysfunction-associated steatohepatitis, inflammation confounds stiffness interpretation; however, dynamic monitoring post-therapy or with biochemical normalization should be systematically studied. Coupling liver and spleen stiffness can noninvasively predict clinically significant portal hypertension. Integration into follow-up algorithms may reduce the reliance on endoscopy, aligning with the Baveno VII recommendations[24]. Cost-effectiveness analyses comparing elastography-based algorithms with traditional biopsies or serial imaging are needed, especially in resource-limited settings. Real-world implementation studies should evaluate the training, throughput, and patient acceptability. The concept of dual VCTE cut-offs represents an elegant synthesis of diagnostic accuracy and clinical pragmatism. However, as PBC management evolves toward precision medicine, elastography should not be viewed as an isolated test but as a component of an integrated ecosystem encompassing biochemical, imaging, and molecular data. Emerging “digital hepatology” initiatives leverage artificial intelligence to interpret LSM trajectories, predict decompensation, and optimize surveillance, and the integration of VCTE into electronic decision-support tools could transform longitudinal PBC care pathways, enabling proactive rather than reactive management.

CONCLUSION

The results of the study by Chen et al[7] consolidate VCTE as the cornerstone of non-invasive fibrosis assessment in Chinese patients with PBC. Their dual cut-off strategy (≤ 10 kPa and > 14.5 kPa) demonstrated robust diagnostic performance and practical applicability across diverse settings. Nevertheless, residual challenges, including intermediate “grey zone” interpretation, confounding by cholestasis and inflammation, and inter-vendor variability, underscore the need for continued refinement. Future research should focus on standardizing data acquisition, validating multimodal algorithms, and establishing longitudinal prognostic models. As evidence accumulates, VCTE is poised to move “beyond” dual cut-offs toward a central role in precision hepatology, integrating structural, biochemical, and functional data to guide individualized PBC treatment.