The overall benefit associated with sustained virologic response (SVR) was evident for the last 2 decades, including a reduction in the rate of occurrence of hepatocellular carcinoma (HCC) and extra-hepatic manifestations (1).
This evidence has been challenged at the 2016 EASL meeting by the report of an unexpected high rate and aggressiveness of tumor recurrence in patients with apparent complete response after HCC local treatment (2-3) which could be explained by a disruption of immune surveillance, facilitating the emergence of local or metastatic clones. Methodological limitations have been pointed out (4) and we provided prospective evidence from 3 French cohorts indicating no sign of HCC recurrence risk (5). Several retrospective or prospective studies now preclude such a risk and are summarized in a systematic review (6). The meta-analysis of 26 studies (17 with interferon (IFN)-based regimen and 9 with oral DAAs) did not indicate a significant risk of neither HCC occurrence nor recurrence associated with DAA. Similar retrospective results have been reported from the very large US veteran-cohort: among the 62 354 HCV patients without detectable HCC from 1999 to 2015, 35 871 (58%) were treated by INF-only, 4 535 by DAAs + IFN and 21 948 by DAAs only (7). Survival free of HCC by cirrhosis and SVR status was similar across antiviral therapies, arguing against a deleterious role for DAA especially since IFN has antiproliferative properties.
An Italian prospective study including 3 381 patients with a F3-F4 fibrosis score treated by DAA (SVR12 in 97.2% of F3, in 92.7% of F4 Child-Pugh A and 80% of F4 Child-Pugh B/C) did not show an increased HCC risk associated with the DAAs: the HCC incidence of 1.64% person/year (CI95%: 1.18-2.21) was not different than that of the control group (8). Incidence of HCC was related to the severity of the underlying liver disease.
Finally, by using weighting methodologies (IPTW, Cox models), given the heterogeneities of treated and untreated patients (at least in countries with treatment prioritization), we established a decreased risk of de novo HCC in DAA-treated patients (9). Among 9 895 HCV-infected patients (median follow-up of 33 months), DAA treatment was initiated in 7 344 patients while 2 551 patients remained untreated. Death, HCC and decompensated cirrhosis were reported during follow-up in 218, 258 and 106 patients, respectively. Exposure to DAA treatment was associated with an increased risk of HCC and decompensated cirrhosis on the unadjusted Cox model (HR=2·77 (95% CI 2·07-3·71) and (HR=3·83 (2·29-6·42)), respectively. On adjusted multivariable analysis, exposure to DAA was associated with a decrease in all cause-mortality (HR=0·48 (95% CI 0·33-0·70)) and HCC (HR=0·66 (0·46-0·93)) and was no longer associated with decompensated cirrhosis (HR=1·14 (0·57-2·27)). These results are clearly reassuring for DAA treatment being associated with a reduced risk of both hepatic and extra-hepatic mortality and HCC; they underline the need of weighting methodologies to avoid misinterpretations.
In summary, SVR improves the prognosis and quality of life of patients and DAA does not increase the risk of HCC recurrence or occurrence.