Toward non-invasive assessment strategies in autoimmune hepatitis

Abstract

In this article, we comment on the article by Peta et al. This study evaluates the diagnostic performance of FibroTest-Actitest, transient elastography, and the fibrosis-4 index against a histological reference. Using the Obuchowski measure, the authors demonstrate that FibroTest and vibration-controlled transient elastography outperform the fibrosis-4 index in detecting fibrosis. Additionally, Actitest offers superior estimation of inflammatory activity compared to conventional biomarkers. Assessing liver fibrosis is crucial for managing autoimmune hepatitis (AIH), yet reliance on invasive liver biopsy remains higher than in other liver diseases. This is partly due to more complex diagnostic criteria for AIH, the lack of standardized scoring for non-invasive testing, and the presence of inflammation, which can lead to falsely elevated results with non-invasive tests. A Bayesian latent class model further supports the reliability of these non-invasive tests, highlighting their potential to complement biopsy, particularly for long-term disease monitoring. These findings underscore the importance of non-invasive diagnostics in optimizing AIH management.

Key Words: Autoimmune hepatitis; Non-invasive test; Liver biopsy; FibroTest; Actitest; Transient elastography; Fibrosis-4 index

Core Tip: This article reviews a study comparing non-invasive liver fibrosis tests in autoimmune hepatitis (AIH). FibroTest, transient elastography, and Actitest showed better accuracy than the fibrosis-4 index, with Actitest excelling in detecting inflammation. Although liver biopsy remains common in AIH due to diagnostic complexity and inflammation’s impact on non-invasive tests, a Bayesian model supports their reliability. The findings emphasize the value of non-invasive methods in improving long-term AIH monitoring and complementing liver biopsy.

TO THE EDITOR

Autoimmune hepatitis (AIH) is a chronic inflammatory liver disease that, if left unchecked, can progress to fibrosis, cirrhosis, and liver failure. Accurate assessment of liver fibrosis and inflammatory activity is essential for disease management, guiding treatment decisions, and predicting long-term outcomes. Traditionally, liver biopsy has been the gold standard for staging fibrosis and grading inflammation in AIH[1,2]. However, the invasive nature of biopsy, along with its risks, costs, and potential sampling variability, necessitates the development of reliable non-invasive alternatives[3].

A recent study by Peta et al[4] sought to evaluate the diagnostic performance of three widely used non-invasive tools-FibroTest-Actitest, transient elastography (VCTE), and the fibrosis-4 index-against a histological reference in patients with AIH. The results highlight the growing potential of non-invasive assessments in clinical practice, particularly in reducing dependence on repeated biopsies for disease monitoring.

A NEW ERA OF NON-INVASIVE DIAGNOSTICS

The study employed the Obuchowski measure to compare these tests’ efficacy in detecting liver fibrosis. Findings indicate that both FibroTest and vibration-controlled VCTE outperform the fibrosis-4 index in fibrosis detection[4]. FibroTest, a blood-based composite biomarker, demonstrated particularly promising results, suggesting its applicability in AIH, where fibrosis assessment remains a challenge[5,6].

Beyond fibrosis assessment, Actitest - a component of FibroTest designed to evaluate inflammatory activity - exhibited superior diagnostic capability compared to conventional markers such as alanine aminotransferase and immunoglobulin G[4]. Given the fluctuating nature of liver enzymes in AIH, Actitest offers a more stable and reliable measure of inflammatory activity, improving disease monitoring and treatment response evaluation.

STATISTICAL VALIDATION: THE BAYESIAN LATENT CLASS MODEL

To ensure the robustness of their findings, the researchers applied a Bayesian latent class model, a sophisticated statistical approach that integrates multiple diagnostic test results without relying on a perfect reference standard[4]. This analysis confirmed the strong performance of FibroTest and VCTE in detecting fibrosis, and of Actitest in assessing inflammatory activity. Specifically, FibroTest (cutoff > 0.48) and VCTE (cutoff ≥ 8 kPa) demonstrated sensitivities of approximately 73% and 81% and specificities of 86% and 84%, respectively, for significant fibrosis (≥ F2). For advanced fibrosis (≥ F3), VCTE maintained high sensitivity (84%) while FibroTest showed strong specificity (86%). Inflammatory activity (≥ A2) was best captured by Actitest (cutoff > 0.52), which achieved nearly 97% sensitivity and 99% specificity, notably outperforming alanine aminotransferase (cutoff ≥ 50), which offered comparable sensitivity but substantially lower specificity. These data not only strengthen confidence in the clinical utility of non-invasive diagnostics but also illustrate their potential to reduce reliance on liver biopsy, particularly for long-term disease monitoring. Taken together, these findings reinforce the advancing role of non-invasive tools in shaping the future of AIH management.

IMPLICATIONS FOR CLINICAL PRACTICE

The findings from Peta et al[4] have significant implications for AIH management. As clinicians strive to balance diagnostic accuracy with patient safety, non-invasive tools represent a promising adjunct, particularly for long-term monitoring where repeated biopsies may be impractical an unethical[3]. With FibroTest and VCTE demonstrating strong diagnostic performance, these tools can be integrated into AIH care protocols to complement histological evaluation, potentially reducing the frequency of invasive procedures while maintaining reliable disease assessment.

Moreover, the superior performance of Actitest in estimating inflammatory activity highlights its potential to guide immunosuppressive therapy adjustments[4]. By providing a more consistent measure of disease activity, it may facilitate timely treatment modifications and ultimately improve patient outcomes. Importantly, these findings align with evolving international perspectives. The 2019 American Association for the Study of Liver Diseases practice guidance emphasizes liver biopsy as the diagnostic gold standard in AIH but acknowledges the need for less invasive tools for disease monitoring and treatment response assessment[2]. Similarly, the 2015 European Association for the Study of the Liver clinical practice guidelines recognize the limitations of non-invasive tests in AIH compared to viral hepatitis, emphasizing the need for further validation[1]. The study by Peta et al[4] helps address this gap by providing statistical validation and demonstrating that tools such as FibroTest and VCTE can deliver diagnostic performance comparable to biopsy in specific contexts. While these results do not negate the role of biopsy in initial diagnosis, they suggest an important adjunctive role for non-invasive assessments.

THE FUTURE OF AIH DIAGNOSTICS

The evolution of non-invasive diagnostic strategies marks a paradigm shift in AIH management. While liver biopsy remains a valuable tool in certain scenarios[1,2], this study further reinforces the growing confidence in non-invasive alternatives. As research continues to refine these methodologies, the integration of FibroTest, VCTE, and other emerging biomarkers could enhance standard care for AIH patients by offering safer, more patient-friendly options that complement the information provided by liver biopsy[5,6]. In an era in which precision medicine is gaining traction, leveraging non-invasive tools can revolutionize how we diagnose and manage chronic liver diseases[3]. By embracing these advancements, clinicians can move beyond biopsy, enabling a future where AIH care is more accessible, efficient, and patient-centered.

STRENGTHS AND LIMITATIONS

A principal strength of this study is its novel application of FibroTest, Actitest, and vibration-controlled VCTE to AIH, an area where non-invasive tests have not been thoroughly validated[3,4]. By using the Obuchowski measure, the study offers a comprehensive assessment that includes all pairwise comparisons across different stages of fibrosis and degrees of inflammatory activity, providing a more detailed analysis than standard area under the curve of the receiver operating characteristic-based approaches. Additionally, the use of a Bayesian latent class model addresses recognized limitations of liver biopsy and accounts for the absence of a universally accepted gold standard[4]. The blinded acquisition of VCTE data and adherence to standardized measurement criteria further strengthen the internal validity of the findings. Notably, the study also shows that these non-invasive tests perform comparably in AIH to their established use in other chronic liver diseases, such as viral hepatitis and metabolic dysfunction-associated steatotic liver disease, supporting their potential applicability across liver disease types[5,6]. However, the distinct immunologic and inflammatory characteristics of AIH may influence the diagnostic performance of these tools in ways that differ from metabolic or viral liver conditions, highlighting the importance of disease-specific validation.

Nonetheless, several limitations should be acknowledged. The single-center design and relatively modest sample size, although larger than in many AIH-focused investigations, limit the generalizability of the findings and reduce statistical power-particularly within underrepresented subgroups such as cirrhotic patients or those with overlap syndromes like AIH-primary biliary cholangitis[2]. The median biopsy length of 12 mm, shorter than the recommended 15 mm in AIH, could affect the accuracy of histological evaluation, and the use of a single morphologist introduces the potential for observer bias, especially when grading inflammatory activity, which is inherently variable[1]. Additionally, the diagnostic cut-offs for FibroTest, Actitest, and VCTE were adapted from studies in viral hepatitis and metabolic dysfunction-associated steatotic liver disease, as no AIH-specific thresholds currently exist[5]. This lack of disease-specific calibration remains a key translational barrier, as thresholds derived from other liver diseases may not fully reflect the unique inflammatory and fibrotic patterns seen in AIH. Without dedicated validation, there is a risk of misclassification, particularly when interpreting borderline or indeterminate results. Future research should prioritize establishing AIH-specific benchmarks to improve the precision and clinical applicability of these non-invasive tools.

The study also did not adjust for potential confounding factors, such as concurrent autoimmune conditions, immunosuppressive therapy, or inflammation-related effects on VCTE, all of which may influence diagnostic performance[6]. While the Bayesian latent class model helps address some limitations of liver biopsy, it relies on key assumptions-such as conditional independence between tests and accurate model specification-that may not always hold in a heterogeneous AIH population. If these assumptions are violated, the validity of the model’s estimates can be compromised. It is important to keep these statistical limitations in mind when interpreting the findings and considering their clinical application. Prospective, multicenter studies with standardized protocols and repeated measures are needed to validate these results and guide the broader integration of non-invasive diagnostics into AIH management.

As a final point, while the study offers promising insights, its applicability to the full spectrum of AIH presentations warrants caution. AIH is a heterogeneous condition, and patients with overlap syndromes, such as AIH-primary biliary cholangitis, were underrepresented in this cohort. Applying non-invasive tools to these subtypes presents unique challenges, as their distinct pathophysiological features may affect diagnostic accuracy. Further research is needed to validate and calibrate non-invasive assessments across the diverse clinical presentations of AIH.

CONCLUSION

Non-invasive diagnostic tools such as FibroTest, VCTE, and Actitest are proving to be reliable alternatives to liver biopsy in the assessment of AIH-related fibrosis and inflammation. Their clinical application can significantly enhance patient care by reducing the need for invasive procedures while ensuring accurate disease monitoring and treatment adjustments.