In the era of IFN based regimen, main concerns for HCV treatment were increased SVR and monitoring of liver-related events for non-virologic responders. The development of IFN-free regimen has made it easier to achieve SVR with a simplified method having a good safety profile.17 Hence, the focus has shifted to the improvement in hepatic fibrosis after viral response, which could help us figure out who will have the remained risk.18 Yet, prospective data regarding the long-term outcome and the potential fibrosis regression after DAA treatment are insufficient.
Another important issue associated with DAA treatment is whether more HCC occurs after SVR induced by DAA.19,20 Recent studies have suggested that SVR by DAA can lower the risk for HCC development.21,22 In addition, pre-DAA albumin, post-DAA LS value, and post-DAA albumin are independent predictors for HCC development.17
Main findings of the present study were: 1) regression of fibrosis continued up to 144 weeks after starting DCV/ASV: 2) regression of fibrosis by more than 30% was observed in 47.7 % and 63.4 % of patients at 48 and 96 week from the initiation of DAA; 3) the degree of fibrotic change was consistently larger in the DAA group than in the peg-IFN/RBV group before and after PSM; and 4) both groups showed similar HCC occurrence after SVR until 144 weeks.
Several studies have evaluated a change of liver fibrosis after IFN based regimen using paired biopsies. Lower baseline fibrotic stage, younger age, lower viral load, lower BMI, and higher baseline ALT level have been suggested as factors associated with larger improvement in liver histology.4,5 In a meta-analysis with three randomized control trials, about one third of cirrhotic patients showed decrease in fibrosis score at 24 weeks after the end of treatment.4 In another study, the reversal of cirrhosis was observed in 75 (49 %) of 153 patients.5
This study showed results consistent with previous studies on DAA treatment. They reported that significant regression was observed in 40–46% at post-SVR12. In addition, about 31–38% of cirrhotic patients according to LS value before treatment were re-categorized as non-cirrhotic after SVR.23–26
We followed up LS measurement and serum markers for a longer duration compared to previous studies. The LS value showed significant reductions at 3 years even after starting the treatment. A recent study on HIV/HCV coinfected patients has reported that LS measured using TE shows a gradual decline at about 4 years after starting the treatment.27 Another study with long-term LS measurement assessment for 5 years has reported that LS value shows a plateau decline until 5 years for patients after receiving IFN with DAA treatment.28 When further long-term data of our cohort become available, they could help us determine potential markers for fibrosis regression.
To the best of our knowledge, this is the first study to compare the extent of fibrosis improvement between DAA and peg-IFN + RBV regimen using matched analysis. However, there were incurable imbalances in cohorts between the two groups. The major difference within cohorts was attributed to a broader therapeutic window for HCV treatment in the DAA group. We acknowledge that there are discrepancies even after matching. To overcome this weak point, we evaluated a fibrotic alteration as a gap of the LS value and a proportional change. The DAA group showed larger reduction consistently. In a previous meta-analysis29, the absolute decrease in LS between before and at the end of the treatment was greater in DAA group than in the IFN-based treatment group. Considering that cirrhosis was more common in DAA treated patients in this meta-analysis, a proportional change of LS value rather than an absolute difference from baseline would be more appropriate to compare. They suggested that the larger magnitude of decline in LS might have resulted from characteristics of DAA, which can eliminate virus more quickly, leading to rapid declines of inflammation and fibrogenesis. Accumulation of robust evidence is necessary to support this assumption.
Several previous studies have demonstrated that higher LS value, alcohol intake, type 2 diabetes, and absence of thrombocytopenia could be factors associating with significant regression.23,30,31 In our study, however, neither baseline LS value, alcohol intake, type 2 diabetes, nor thrombocytopenia was related to a significant regression. However, suggested factors were quite different among various studies. This should be clarified with further studies.
This study has some limitations. In 2015, when this study was planned and started, the only available DAA was DCV/ASV. Now in the era of pan-genomic DAA, DCV/ASV is no longer a preferred option for HCV treatment. However, the major concern of this study is how much liver stiffness has improved since the use of DAA, not specific drugs. Further prospective studies including various widely used regimens are required. Included patients of our study were restricted to genotype 1b infection. Although a previous study with a small number of subjects has shown that significant fibrosis regression after DAA is associated with genotype 1, it is currently unclear whether the improvement in liver fibrosis is associated with viral genotype.30 Recent studies which investigated the alteration of liver stiffness after DAA treatment, there was a rapid decrease in LS within a few weeks after DAA treatment.27,30 We did not check LS at the end of DAA treatment because it was assumed that fibrosis regression occurred gradually over time rather than immediately after drug use. Although many studies have demonstrated that liver stiffness measured by TE is well correlated with fibrosis stage, suggested cut-off values have discrepancies according to studies.10 Thus, the suggested liver fibrosis stage used in our study requires attention in its interpretation. In addition, the number of cohort was not large enough to evaluate the association between the change of liver stiffness and predictive factors. Finally, this study was not a controlled trial between DAA and peg-IFN based regimen because a direct comparison was not feasible ethically or time wise. Recently, there has been an increased interest in the risk of HCC development after DAA treatment.18,32 The incidence of HCC in our cohort was not significantly different from that in the peg-IFN group. It was not sufficient for identifying a relationship with the extent of regression.
In conclusion, eradication of HCV with DCV/ASV resulted in a continuous improvement of liver stiffness over time. SVR induced by DAA treatment was strongly associated with a greater regression of fibrosis. DAA resulted in a comparable occurrence of HCC to peg-IFN/RBV treatment. Further studies are needed to confirm our results and identify the group of patients who need careful monitoring after SVR.