Research in context
Evidence before this study
We searched PubMed for reports published in any language between Jan 1, 2012, and Jan 1, 2018, with the MESH terms (“hepatitis C, chronic/drug therapy” OR “hepatitis C, chronic/therapy”) AND (“hepatitis C, chronic/mortality” OR “hepatitis C, chronic/complications”). We searched for evidence of randomised trials or observational studies assessing the risk of mortality, liver cancer, and the complications of liver disease after antiviral treatment with direct-acting antivirals in patients with chronic hepatitis C virus (HCV) infection. We identified a 2017 review of 138 randomised trials assessing the effects of 51 different direct-acting antivirals, indicating that use of these drugs increased the proportion achieving a sustained virological response. However, the review did not reach any conclusions on clinical effects. We only found one additional retrospective cohort study reporting a significant decrease in all-cause mortality in patients receiving either paritaprevir, ritonavir, ombitasvir, and dasabuvir or sofosbuvir and ledipasvir, compared with untreated patients. This study did not report on liver-related mortality or liver-related events (eg, liver cancer or liver decompensation). No existing prospective study has examined the benefits and harms of direct-acting antivirals in chronic HCV infection on liver-related clinical outcomes using time-to-event analyses and by comparison of treated and untreated patients.
Added value of the study
To our knowledge, the ANRS CO22 Hepather cohort study is the first prospective longitudinal study to investigate clinical outcomes associated with direct-acting antiviral treatment in patients with chronic HCV infection, by comparing patients treated with direct-acting antivirals with those untreated with these drugs, irrespective of the status of sustained virological response, with careful control of confounding and indication biases. The adjusted multivariable analyses show that direct-acting antiviral treatment is associated with a rapid decrease in all-cause mortality and the incidence of hepatocellular carcinoma, and that these inverse associations are stronger in patients with cirrhosis.
Implications of all the available evidence
For ethical reasons, a trial with an untreated control arm cannot be undertaken to confirm these findings. We encourage other researchers to do similar comparisons of patients treated with direct-acting antivirals and untreated patients, using existing observational databases. Nevertheless, our results support urgent treatment of patients with advanced liver disease and extension of the follow-up of treated patients with less severe disease to assess the long-term clinical effect of direct-acting antiviral treatment.