Should Surveillance for Liver Cancer be Modified in Hepatitis C Patients After Treatment-related Cirrhosis Regression?

Roberta D'Ambrosio; Massimo Colombo


Liver International. 2016;36(6):783-790. 

In This Article

How Accurately Can Liver Fibrosis be Staged in Non-viraemic Patients

Liver biopsy has long been the gold standard for staging liver fibrosis in chronic hepatitis C, in an era in which assessment of hepatitis C prognosis was mandatory to select the ideal candidate to receive such a poorly tolerated and effective anti-HCV regimen as interferon. Not surprisingly, therefore, the increasing effectiveness and safety of new all oral antiviral therapies against HCV coupled with cons like poor patient compliance, significant rates of sampling errors and risk of clinical complications typical of invasive procedures, progressively led to abandon liver biopsy as a preferred procedure to stage HCV.[39,40,54–58] In parallel, non-invasive assays of liver fibrosis have gained popularity to evaluate patients with chronic hepatitis C, to the point of being recommended by international societies for staging of both viraemic and non-viraemic hepatitis C patients.[59] However, among non-invasive tests of liver fibrosis, serum markers have been validated in viraemic patients, only, making therefore diagnostic accuracy of these assays in non-viraemic patients, unclear. This was the message of a proof of concept study evaluating serological biomarkers of fibrosis based upon biochemical serum parameters of liver inflammation, owing to the fact that the diagnostic accuracy of such tests can be modified following persistent HCV eradication, once hepatic inflammation is shut down.[60,61] In principle, assessment of liver stiffness by TE offers better chances than serum biomarkers for evaluating residual liver fibrosis in viraemic patients, since this method senses structural remodelling of the liver in addition to hepatic inflammation. Yet, TE was not accurate enough to detect residual fibrosis in 37 patients who underwent concomitant liver biopsy and liver stiffness measurement (LSM) after 48–104 months from the achievement of an SVR: in this study, in fact, TE led to misclassification of five patients (21%) with residual cirrhosis.[41] One possible explanation for these diagnostic discrepancies with TE could be a loss of hepatic stiffness following a reduction in the area of fibrosis consequent to HCV eradication. Indeed, 21% of patients with LSM values below the cut-off of 11.9 kPa, currently adopted to predict cirrhosis in viraemic patients, were found to have residual cirrhosis at histology. In the same cohort of SVR patients, indirect assays like APRI, CDS, FIB-4, FibroQ, Forns Score, GUCI Index, King Score, Lok Index, PLF and a direct test of fibrosis like enhanced liver fibrosis (ELF) failed to separate SVR patients stratified by the degree of residual fibrosis in the liver.[60] More recently, a retrospective study in Japan reported slight different results among a cohort of 115 HCV patients who underwent both liver biopsy and non-invasive tests of fibrosis (Forns Score, APRI, FIB-4) before treatment and 5 years after HCV eradication.[62] The authors found that after an SVR, values of non-invasive tests increased along with the increase in the post-SVR histological fibrosis score (P < 0.0001), with an acceptable accuracy for the identification of F2–F4 or F3–F4 fibrosis. However, the prevalence of pretreatment cirrhosis in the Japanese study was extremely low (2.6%), making it therefore impossible to assess the performance of the tests in the diagnosis of residual cirrhosis. In that study, the accuracy of non-invasive tests to predict F3 and F4 fibrosis in SVR patients was between 29.4% for FIB-4 index and 44.8% for APRI only. In another study, a decrease in serum fibrosis markers YKL-40, TIMP-1, PIIINP and HA was demonstrated in 81 patients with a pretreatment diagnosis of advanced fibrosis 24 weeks after successful antiviral treatment, however without any parallel investigation on liver histology outcome being performed.[63] In a retrospective scrutiny of more than 900 HCV patients undergoing repeated Fibrotest (FT) and TE examinations during a period of 0.5 and 12.2 years, 171 SVR patients showed improved values of both LSM and FT values, including 24 patients (56%) presumed to have cirrhosis.[6] A challenging finding of that study, however, was the increased values of both assays in 12% of patients with a SVR, whereas liver complications like HCC and cholangiocarcinoma (CCK) were observed among four SVR patients, including two with improved FT values, overall suggesting poor accuracy of non-invasive tests in the prognostication of SVR patients.