Non-invasive systems for staging compensated cirrhosis

The lack of a true gold standard 

The introduction and evaluation of different non-invasive measures for assessing disease progression in CLD is based on the key limitation of using liver biopsy as a reference standard. Indeed, several limitations of liver biopsy, and particularly the fact that a single bioptic cylinder is representative of no more than 1/50.000 of the whole liver, make a fair comparison with a serum marker or LSM rather difficult – this also considering that standard histopathological analysis may have the same difficulty of non-invasive estimates in discriminating adjacent stages of fibrosis (i.e., F2 vs. F1 or F3 vs. F2) [1], [2]. To minimise these limitations, it is absolutely important that histopathological staging is assessed with all the proposed recommendations with regard to the size of the biopsy and the number of portal tracts analysed [1], [3], [4] and that the non-invasive methodology (blood test, LSM, etc.) is tested within a reasonable time from the liver biopsy, ideally within 24h and certainly not within 3–6months, as often reported.

Measurement of the HVPG, currently employed for the evaluation of portal hypertension, has been suggested as a reliable end-point to assess the therapeutic benefit of antiviral therapy in patients with advanced hepatic fibrosis due to chronic HCV infection [5], [6], [7], [8], [9]. In the absence of significant fibrotic evolution, HVPG does not exceed 5mmHg, whereas a gradient of more than 5mmHg is always associated with significant fibrosis. Therefore, when considering treatment of patients with advanced fibrosis with an HVPG in the range between 5 and 10mmHg, the measurement of HVPG could provide relevant indications of improvement, stabilisation or worsening within the stage of compensated cirrhosis. In summary, it appears that the major limitation of HVPG measurement relies on logistics: it is expensive, requires a dedicated setting and very experienced operators, and hence is available only in specialised centres.

Biochemical markers 

A large scientific and commercial investment has been made in the past 10years to develop serum markers that are able to predict the fibrotic stage of CLD. Among the proposed markers, some reflect alterations in hepatic function but do not directly reflect extra-cellular matrix (ECM) metabolism: ‘indirect markers’. Others are directly linked to the modifications in ECM turnover occurring during fibrogenesis: ‘direct markers’ [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]. In most of the studies published so far, the area under the curve (AUC) for the receiver operating characteristic (ROC) curve is employed as a measure of test performance, with optimal values being as close to 1.0 as possible. Nevertheless, the reported median AUC in differentiating mild/no fibrosis and significant fibrosis in validation populations is about 0.77, which is far from high diagnostic accuracy. However, all tests show an improved performance when the end-point is to differentiate cirrhosis/non-cirrhosis with median AUC in validation sets of approximately 0.87 [28].

In general, it is very important to note that, although direct, indirect and combined serum marker systems measure rather different biomarkers, they are all characterised by an AUC for the ROC clustering around 0.85. As already mentioned, it is likely that the explanation of this diagnostic equivalency lies in the inaccuracy of liver biopsy as a reference standard either in absolute terms or relative to the lack of adequate standards in the validation studies performed so far [29].

A better definition of serum tests could derive from their evaluation in prospective studies employing a combination of different tests. Along these lines, Sebastiani et al. reported that a stepwise combination of different algorithms (e.g., APRI, Fibrotest and Forns’ index) in cohorts of patients with chronic HCV or HBV infection may reduce the need for liver biopsy in 50–70% and 50–80% of cases, respectively [30], [31]. Moreover, Leroy et al. prospectively compared six non-invasive scores for the diagnosis of liver fibrosis in chronic HCV hepatitis patients [32]. They found that the best combination (including MP3, Fibrotest and APRI) could select one-third of patients with either absence of significant fibrosis or presence of advanced fibrosis with more than 90% certainty.

In summary, at the present stage of development, we can conclude that the diagnostic accuracy of systems employing serum biomarkers has been proved useful for the detection (or exclusion) of significant fibrosis or cirrhosis mainly in patients with chronic HCV infection. However, it is rather clear that these tests may reduce but not eliminate the need of liver biopsy and that the platelet count per se allows the exclusion of cirrhosis with a fairly similar degree of accuracy [33], [34].

Transient elastography 

In the past 5years, a non-invasive medical device based on transient elastography (Fibroscan®, Echosense, Paris, France) has received increasing interest. This system has been proposed for the measurement of liver stiffness, considered as a direct consequence of the fibrotic evolution of CLD [35]. As a result, several studies aimed at evaluating the clinical usefulness and diagnostic accuracy of LSM have been published in the past 2years [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46]. The results of the two pioneering studies performed in cohorts of patients with chronic hepatitis C suggested that this system could be useful in assessing the presence of significant fibrosis (i.e., ⩾F2) and in suggesting the presence of cirrhosis [36], [44].

In a recent report, Fraquelli et al. have provided important information for the establishment of practical guidelines in the use of LSM as a diagnostic tool in CLD [45]. Firstly, LSM is characterised by high intra- and inter-observer agreement (0.98), which likely makes this system the most reproducible among those currently available. Importantly, the inter-observer, and significantly the intra-observer, agreements are influenced by variables such as body mass index (BMI) (particularly when ⩾28) and hepatic steatosis, and therefore LSM should be used cautiously as a surrogate of liver biopsy for assessing liver fibrosis in the presence of these clinical features. This study also strongly suggests that factors other than fibrosis may have an impact on the results of LSM. Indeed, the extent of necro-inflammatory activity influences LSM with a steady increase of LSM values in parallel with the degree of histological activity. This possibility is also emphasised by a recent report showing significant variations in liver stiffness occurring during ALT flares in patients with chronic viral hepatitis [47] and confirmed by the recent results obtained by our group in patients with acute viral hepatitis [48]. In principle, it is increasingly evident that the use of LSM will allow a broader diagnostic flexibility and possibility of integration with other diagnostic methodologies for the clinical assessment of patients with advanced fibrosis and cirrhosis.

Non-invasive staging in patients with advanced fibrosis/cirrhosis 

By and large, it is rather clear that serum markers, LSM as well as the standard imaging techniques perform quite well in advanced fibrosis and cirrhosis and, accordingly, these systems could be proposed for the clinical work-out of advanced CLD. Indeed, there is a need to differentiate advanced fibrosis (i.e., ⩾F3–F4) and beyond, that is, when a patient is in a clinical stage defined as ‘compensated cirrhosis’ and, subsequently, when clinically significant portal hypertension and relative complications develop.

Carrion et al. [37] reported that LSM reflects both the extent of fibrosis and the elevation of portal pressure in terms of HVPG in recurrent hepatitis C after liver transplantation. Recent results from our group have confirmed an excellent correlation between LSM and HPVG values <10mmHg, whereas this correlation becomes poor for values ⩾12mmHg [49]. In this context, transient elastography could be useful in differentiating stages of progressive cirrhotic evolution of liver tissue once the histological stage of cirrhosis has been reached. In some studies, this possibility has been extended to the prediction of the most common complication of cirrhosis [50], [51]. However, the poor correlation existing between LSM and HVPG values ⩾12mmHg makes the possibility of predicting the presence and the size of oesophageal varices with a sufficient degree of accuracy rather unlikely [49].

In conclusion, some major considerations arise from the experience accumulated so far. First, all non-invasive methodologies are characterised by a sufficient-to-excellent diagnostic accuracy for the detection (or exclusion) of advanced fibrosis and cirrhosis, and none is able to allow a step-wise follow-up of the fibrogenic evolution of CLD according to the existing histopathological staging systems. In other words, owing to the absence of a true gold standard, the definition of a 90% diagnostic accuracy remains a goal for the future. In addition, none of the currently available tests has a well-defined prognostic value such as the prediction of decompensation or death. Second, owing to the ‘spectrum bias’ and the possible causes of discordance with the histopathological assessment, the applicability of the different proposed cut-off values in clinical practice is presently hazardous. Third, the combination of two unrelated non-invasive tests (i.e., one biochemical test with LSM) may provide a useful system for the initial assessment of fibrosis evolution in patients with CLD. Fig. 1 illustrates a flowchart for chronic HCV hepatitis, which is proposed to stimulate a constructive debate. The definition of intermediate stages of fibrosis (i.e., F2 and F3) as a ‘grey area’ highlights the current difficulty in the clinical use of the proposed non-invasive tests. Nevertheless, it is increasingly evident that a rational and prudent use of the proposed methodologies will reduce the need for liver biopsy in a significant percentage of patients and represents a diagnostic advantage.

View Large Image Download to PowerPoint

Fig. 1. Proposed flowchart for the non-invasive assessment of fibrosis evolution in chronic HCV patients. Abbreviations: NITs, non-invasive tests (including serum biomarkers and transient elastography); HCC, hepatocellular carcinoma; ALT, alanine aminotransferase.