Non-invasive diagnosis of cirrhosis and long-term disease monitoring by transient elastography in patients with Wilson disease.

Abstract Background & Aims The value of liver stiffness measurement (LSM) by transient elastography (TE) for non‐invasive fibrosis staging and disease monitoring has not been established in patients with Wilson disease (WD). Methods Liver stiffness measurement by TE and non‐invasive fibrosis scores (APRI, FIB‐4) were analysed from 188 WD patients with liver biopsy (LBX). Longitudinal LSM was performed in 128 (68.1%) patients. Results One hundred and eighty‐eight patients (mean age: 35 ± 14 years, 54.8% women; 27.1% with histological cirrhosis) were studied. Forty‐four[23.4%] patients were recently diagnosed with WD, while 144[76.6%] were previously diagnosed (>1 year between LBX and LSM). Overall, LSM (11.3 vs 6.1 kPa, P < .001), APRI (0.72 vs 0.38, P < .001) and FIB‐4 (1.54 vs 0.89, P < .001) were higher in cirrhotic than in non‐cirrhotic patients. This was even more pronounced in recently diagnosed patients (35.2 kPa vs 6.4 kPa, P < .001). Accuracy for diagnosing cirrhosis at an LSM cut‐off ≥9.9 kPa was better in recently diagnosed (PPV: 74%, NPV: 100%) vs previously diagnosed (PPV: 53%, NPV: 82%) patients. Recently diagnosed patients had higher Area Under the Curve (AUC) for APRI (0.79 vs 0.61) and FIB‐4 (0.84 vs 0.65) than previously diagnosed patients. At APRI <1.5 and FIB‐4 <3.25 cirrhosis was ruled out with a specificity of 93% and 95% respectively. During a median follow‐up of 46 (24‐66) months, only 5.9% (5/85) of non‐cirrhotic WD patients showed progression to cirrhotic LSM values, while 30.8% (4/13) of cirrhotic WD patients showed LSM suggestive of cirrhosis regression. Conclusion TE‐based LSM ≥9.9 kPa accurately identifies cirrhosis in WD patients. Next to TE‐LSM <9.9 kPa, APRI <1.5 and FIB‐4 <3.25 values assist to non‐invasively rule out cirrhosis. LSM remains stable in most non‐cirrhotic patients on WD therapy, while one‐third of cirrhotic patients present clinically relevant decreases in LSM.


| INTRODUC TI ON
Wilson disease (WD) is a genetic disease of impaired biliary copper excretion resulting in accumulation of copper in the liver and extrahepatic tissues. [1][2][3] The key features of WD are all forms of liver disease, neuropsychiatric disturbances, Kayser-Fleischer rings and acute episodes of haemolysis that often associate with acute liver failure. The clinical presentation is highly variable and patients may become symptomatic at any age, 4 and about half of the patients develop cirrhosis. 5 Currently there is an ongoing discussion on how to evaluate the efficacy of new treatments of WD. Similar to Non-alcoholic steatohepatitis (NASH), 6 there are no validated study endpoints in WD.
Due to the various phenotypic presentation validated endpoints for neurologic and hepatic disease are needed. Until recently, liver biopsy was the only tool to stage fibrosis/cirrhosis in any patient with liver disease. Novel non-invasive staging-methods, such as transient elastography (TE) and blood-based fibrosis scores such as APRI 7 or FIB-4 8 have been developed. However, both TE and APRI/ FIB-4 have been mostly validated for fibrosis staging in viral liver disease 7,[9][10][11][12] and in non-alcoholic fatty liver disease. 10,13,14 Data on non-invasive methods to stage fibrosis in WD are scarce. 15 After long discussions with Food and Drug Administration (FDA) and European Medicines Agency (EMA) it became clear that our armamentarium to predict the outcome of liver disease (ie surrogate endpoints) in WD is very limited. One of the proposals was to assess the value of repeated liver stiffness measurements (LSM) by TE. To date, only two small studies 16,17 have investigated TE-based LSM in adult WD patients, however, both included very few patients with cirrhosis. and MSD; advisory boards for Albireo, Falk Pharma GmbH, BiomX, Boehringer-Ingelheim, Genfit, Gilead, Intercept, MSD, Novartis and Phenex. He further received travel grants from Abbvie, Falk, Gilead and Intercept and unrestricted research grants from Albireo, Cymabay, Falk, Gilead, Intercept, MSD and Takeda. He is also a co-inventor of a patent on the medical use of norUDCA. TR: received grant support from Abbvie, Boehringer-Ingelheim, Gilead, MSD, Philips Healthcare and Gore; speaking honoraria from Abbvie, Gilead, Gore, Intercept, Roche and MSD; consulting/ advisory board fee from Abbvie, Bayer, Boehringer-Ingelheim, Gilead, Intercept, MSD and Siemens; and travel support from Boehringer-Ingelheim, Gilead  Therefore the aim of our retrospective international multicentre study was to evaluate the value of TE-based LSM, APRI and FIB-4 as non-invasive tools for fibrosis assessment in patients with liver disease: Firstly, we assessed the diagnostic accuracy of TE-based LSM, APRI and FIB-4 for non-invasive detection of cirrhosis in WD patients undergoing liver biopsy as reference.
Secondly, we evaluated the course of TE-based LSM in WD patients in daily clinical practice and its potential use in prospective phase 3 trials.

| Patients
One hundred and eighty-eight WD patients from three tertiary  The following data were recorded: Phenotypic presentation (hepatic, neurologic), type and duration of treatment, standard laboratory data to calculate APRI 7 and FIB-4 8 and endoscopic data (available in 49 of 51 cirrhotic patients, no gastroscopy was performed in non-cirrhotic patients). Follow-up LSM was recorded in all patients if available and the same quality criteria as for the baseline TE were used (see below). Decreasing LSM was defined as reduction by at least −10% and/or −1 kPa, stable LSM as values ±10% and increasing LSM as an increase of at least +10% and/or +1 kPa from BL-to last LSM. Furthermore the proportion of patients whose last LSM measurement overstepped respectively fell below the threshold of 9.9 kPa are reported (threshold was derived by the study data).
Transaminase levels and platelet count to calculate APRI and FIB-4 were recorded in recently diagnosed patients during follow-up.

| Liver histology
Liver biopsies (LBX) were performed at the time of diagnosis or histological findings were obtained from the explanted liver. In many patients the date of biopsy predated the first elastography. Biopsies were read by the local hepatopathologist (FW, TL, KL) as part of the clinical routine. For this study, we classified biopsy results from the written report as: normal, steatosis without significant fibrosis, fibrosis (F1-F3) and cirrhosis.

| Statistics
Continuous variables were reported as mean ± standard deviation  was used for all statistical analysis.

| Ethics
This study was approved by the local ethics committees and performed in accordance to the current version of Helsinki Declaration.
Since this was a retrospective study, no informed consent was needed. Further main patient characteristics are listed in Table 1 and Figure 1.  Table 2) and the AUC for detecting cirrhosis in any patient that was classified as "previously diagnosed" was 0.70 (95% CI: 0.60-0.80, P < .001; PPV: 53%, NPV: 82%, Sensitivity: 46%, Specificity: 86%). When all patients with transaminase levels (ALT or AST) ≥2x ULN were excluded from analysis diagnostic accuracy stayed stable or declined marginally (Table   S4). LSM was significantly higher in cirrhotic vs non-cirrhotic patients

| Factors influencing LSM in WD patients
Binary logistic regression analysis was used to highlight factors that directly influence LSM in patients with WD (  (Table 3). (44.4%) with LSM ≥9.9 kPa decreased below 9.9 kPa ( Figure 4A) at last LSM.

TA B L E 3
Determinants of LSM values ≥ 9.9 kPa in patients with WD (univariable and multivariable binary regression analyses)

| D ISCUSS I ON
There is a need for non-invasive parameters to monitor the out-  (Table 4, Tables S1 and S2, Figure 4)  Accordingly, long-term follow-up data found no survival difference in adequately treated WD patients compared to the general population 23 and death-rates and liver-related mortality were low. [24][25][26][27][28] Similarly, studies evaluating TE pre-and post-HCV therapy, also showed a marked decrease of TE values in patients who achieved sustained virologic response. 29 The availability of non-invasive markers is not only useful to longterm monitoring of patients, but will become essential for the design of treatment trials.

S U PP O RTI N G I N FO R M ATI O N
Additional supporting information may be found online in the Supporting Information section.