Recent reports from Europe have questioned whether IFN-free DAA therapy is associated with an increase in recurrent HCC in people with cirrhosis, and whether recurrent tumours have an altered, more aggressive clinical course. A Spanish study identified 58 patients with a complete radiological response to prior HCC therapy who were subsequently treated with DAAs (SVR in 97%). The median time between HCC response and start of DAA therapy was 11 months. Among the 58 patients, 16 (28%) experienced recurrence within a median of 3.5 months from the start of DAA treatment. While there was no control group in this study, the recurrence rate is higher than that reported by a previous meta-analysis of IFN-based therapy, which showed a significant decrease in the risk of HCC recurrence following SVR (odds ratio, 0.22). The Spanish investigators were suitably cautious in their data interpretation and reported these patients so that others could compare with their experience. Subsequently, an Italian study of 344 patients with HCV and cirrhosis treated with DAAs (SVR in 91%) observed HCC recurrence in 17 of 59 patients (29%) with a history of HCC, and de novo HCC in nine of 285 patients (3%) with no history of HCC, in the 24 weeks after antiviral treatment. The authors also hypothesised that DAAs may accelerate HCC recurrence. In contrast, a French collaborative group has reported outcomes from three prospectively studied cohorts, including 346 HCV patients with treated HCC and 314 patients who underwent liver transplantation for HCC. Over a median follow-up of 20 months, there was no difference in HCC recurrence rates according to DAA treatment. In the transplant group, all of whom were DAA-treated, HCC recurred in 2.2%, compared with a historical recurrence rate of 8%–20% in the first 2 years after transplantation. Therefore, whether or not DAA therapy influences HCC recurrence in people with cirrhosis is unclear based on current published data.
Incident HCC is a separate issue, but emerging data are reassuring. A large prospective registry is following 1067 people with cirrhosis who have been treated with sofosbuvir-containing regimens. A preliminary analysis of people followed for a median of 85 weeks from the end of treatment observed a rate of incident HCC of 0.50 per 100 patient-years in 663 people with compensated cirrhosis. Similarly, in a prospective Italian cohort, among 2007 patients with cirrhosis followed for a median of 301 days, the rate of incident HCC was 1.63 per 100 patient-years. The data compare favourably to rates of incident HCC reported among people with compensated cirrhosis who achieved SVR after treatment with IFN-based therapy (1.39–1.82 per 100 patient-years). In a prospective study of 406 patients with cirrhosis, most with Child–Pugh B+ liver disease, who were treated with DAAs through the English Expanded Access Programme, the SVR rate was 81%. Compared with a control group of 261 untreated patients with cirrhosis followed for 6 months, there was no difference in HCC incidence between the DAA-treated and untreated groups, or between those who achieved SVR and those who did not.
Therefore, we strongly recommend DAA therapy for all individuals with advanced liver disease who do not have a history of HCC. We recommend continuing to offer DAA therapy to patients with advanced liver disease and previous HCC, after informed discussion of potential risks. These people should all be enrolled in HCC screening programs. HCV treatment should not suspend HCC screening. We recommend a liver ultrasound within 1 month before DAA therapy for all individuals with cirrhosis to ensure that HCC screening remains up to date during the treatment and follow-up period. Importantly, there are no data to suggest that HCC risk may be increased in people with no cirrhosis. We do not recommend HCC screening for people with no cirrhosis who are treated for HCV.