Noninvasive methods for the assessment of liver fibrosis: A window open on the future?

Authors

  • Massimo Pinzani M.D., Ph.D., F.R.C.P.

    1. Dipartimento di Medicina Interna Center for Research, Higher Education, and Transfer “DENOThe” Università degli Studi di Firenze/Azienda Ospedaliero Universitaria Careggi (AOUC) Firenze, Italy
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

Vergniol J, Foucher J, Terrebonne E, Bernard PH, le Bail B, Merrouche W, et al. Noninvasive tests for fibrosis and liver stiffness predict 5-year outcomes of patients with chronic hepatitis C. Gastroenterology 2011;140:1970-1979.

Comment

The introduction of noninvasive methodologies for the assessment of liver fibrosis, to be employed as an alternative or in association with liver biopsy for the management of chronic liver disease, particularly chronic hepatitis C virus (HCV) hepatitis, represents a major advancement in hepatology clinical practice.1 The foundations of the currently available noninvasive methods are of the most disparate: from serum markers based on the biology of chronic inflammation and fibrogenesis, to algorithms incorporating common biochemical abnormalities typical of chronic liver diseases (CLD), to modifications of the physical properties of liver tissue (i.e., changes in liver stiffness detected by transient elastography; TE). The availability of noninvasive methods has generated a large debate among hepatologists, especially on the diagnostic accuracy of serum markers, TE, or different combinations in predicting the stage of fibrosis, when compared to liver biopsy. Overall, this competition bears a central contradiction: How can a noninvasive method be equally or even more accurate than liver biopsy when the latter is used as the gold standard? Indeed, noninvasive methods are as inaccurate as liver biopsy for the assessing of fibrosis stage, as intended according to the so-called “semiquantitative” scoring systems (i.e., METAVIR, Ishak's, and others). This is simply because liver biopsy, when used for this purpose, is not a true gold standard because of the following: (1) intrinsic poor representation of the fibrotic changes occurring in the whole organ (an average liver biopsy allows the examination of 1/50,000 of the liver); (2) descriptive and nonquantitative features of the scoring systems; and (3) nonlinearity of the fibrotic evolution of CLD. When performed with optimal standards and at the best of its interpretative potential, liver biopsy can discriminate a liver without fibrosis from one with significant fibrosis (i.e., METAVIR F2) from one with distinctive features of cirrhosis. For exactly the same reasons, liver biopsy cannot have an ideal performance as a prognostic index predicting liver-related outcomes (i.e., death and complications of cirrhosis), including hepatocellular carcinoma, although the presence of significant fibrosis is a clear sign that the fibrogenic process is characterized by a net accumulation of scar tissue and, therefore, is likely to progress toward cirrhosis.

Recently, it has been highlighted that some noninvasive tests may perform better than liver biopsy for the prediction of long-term clinical outcomes in patients with CLD. This possibility was first suggested by the study by Parkes et al.,2 who performed a median 7-year follow-up in 457 patients of the original ELF cohort.3 In addition, the same investigators showed that the prognostic performance of the ELF panel was significantly better than the MELD or Mayo R score for the long-term prediction of liver-related events in patients with primary biliary cirrhosis.4

The study by Vergniol et al., recently published in Gastroenterology,5 adds a great deal of information to the potential usefulness of noninvasive methods as prognostic indicators. Differently from the studies mentioned previously, the work by Vergniol et al. was concentrated on a consecutive cohort of 1,457 patients with chronic HCV at different stages of fibrotic evolution to evaluate the 5-year prognostic value of different noninvasive methods (i.e., liver stiffness measured by TE, FibroTest, APRI, and FIB-4) for predicting survival and liver-related death. Though all methods were able to predict shorter survival times, liver stiffness values and results of FibroTest had the highest 5-year predictive values, which did not change after adjustment for treatment response, patient age, and estimates of necroinflammatory grade. Overall, the results and the conclusions of the study are convincing, and it is quite striking that any of the employed cut-off values for the different noninvasive tests is a better prognostic discriminator than the stage of fibrosis, when the patients are grouped in two groups (i.e., METAVIR F0+F1+F2 versus F3+F4). However, the poor performance of liver biopsy in this context may be an artefact reflecting a crucial defect of the current histopathological staging systems (and METAVIR in particular) in accurately defining advanced fibrosis (i.e., F3, which is a true precirrhotic stage) and the fibrotic evolution of the disease within METAVIR stage F4 (i.e., all cirrhotic patients are F4, irrespective of progressive scarring occurring in the cirrhotic liver). Indeed, the much higher prognostic flexibility of noninvasive methods becomes clearly evident exactly at a stage of the disease (i.e., the cirrhotic stage), when liver biopsy is paralyzed by the conceptual rigidity of the scoring systems. Importantly, although all noninvasive methods were designed and validated using histology as a gold (or best) standard, the results of the study by Vergniol et al.5 highlights the fact that their prognostic power is largely independent of a close correspondence with histological scoring. In other words, increasing cut-off values are able to define different classes of survival and liver-related death within the same histopathological stage.

Overall, the study by Vergniol et al. adds fuel to the sacrificial pyre of liver biopsy, and it is, indeed, a clear signal of the dynamic changes occurring in hepatology practice. It is possible that additional advancements will derive from further validation of noninvasive methods through the comparison with quantitative measures of liver tissue fibrosis obtained by standardized morphometrical analysis, such as the recently proposed collagen proportionate area,6 which has shown an excellent correlation with portal pressure in cirrhotic patients.7

Ancillary