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

Cirrhosis Regression: Facts or Myths?

In the last decades, combination therapy of interferon with the oral analogue ribavirin (RBV) in the context of personalized therapeutic algorithms led to the foundation of effective treatment of hepatitis C patients, resulting in appreciable survival benefits for those who achieved an SVR, including patients with advanced fibrosis and cirrhosis.[8] Most clinical benefits observed in SVR patients were the consequence of the arrest of fibrosis progression while in more than one patient regression of a pre-existing cirrhosis could be documented.[15,38,46–52] This was the clear message of a combined analysis of four randomized trials, where cirrhosis regression was shown to occur in several SVR patients as early as within 2 years of post-treatment follow-up.[48] A few years later, a small multicenter study in Paris provided conclusive evidence that clinical benefits of an SVR in patients with cirrhosis were the consequence of cirrhosis regression by combining the systematic analysis of liver tissue with the occurrence of HCC, variceal bleeding, liver-related deaths and listing to liver transplantation: adverse outcomes, in fact, were observed only in patients with histologically documented residual cirrhosis, not in cirrhosis regressors.[15] It should be noted, however, that studies on cirrhosis regression are quite difficult to interpret, mainly owing to disparities between studies in the time lag separating pre- and post-SVR biopsies and the inherent suboptimal accuracy of liver biopsy itself to diagnose cirrhosis, even when tissue cores of several centimetres of length are available.[39] In the end, accuracy of fibrosis detection by liver biopsy is not challenged by sampling errors only, but it may be flawed by intra-observer variation as shown by a study of patients with HCV in whom topography of sampling contributed to differences in staging of fibrosis too.[40] It is no surprise, therefore, that histological studies based on the comparison of paired liver biopsies provided evidence of fibrosis and/or cirrhosis regression at different rates ranging between 24% and 100%,[15,38,46–53] with seven studies reporting any degree of cirrhosis regression (Table 1). While interpretation of liver fibrosis regression in SVR patients can further be biased by the retrospective design of investigations and the exclusion of cirrhotic patients lacking a pretreatment histological assessment, there is enough evidence to say that cirrhosis regression in SVR patients did occur independently of the type of interferon regimen administered. An important by-product of the histological studies in these patients was the demonstration of a reduction in total collagen content observed in the livers of SVR patients, including those with residual cirrhosis. In a study of 38 patients with compensated cirrhosis of any HCV genotype, who were treated with Pegylated interferon and ribavirin, cirrhosis regression was histologically documented in 61% of the patients who underwent a second liver biopsy 48–108 months after the achievement of an SVR, and the total collagen content of the liver was significantly reduced following viral eradication in both regressors and non regressors.[38] While remodelling of the extracellular matrix in the fibrotic livers may take years to complete after the achievement of an SVR, this same process may have practical clinical implications, particularly as far as non-invasive assessment of liver fibrosis with transient elastography (TE) is concerned, since liver stiffness appears to be greatly influenced by the pattern of residual fibrosis in viraemic patients and in both non-viraemic regressors and non-regressors.[41]