How Do Direct-Acting Antivirals for Hepatitis C Affect the Risk for Hepatocellular Carcinoma?

Sheila L. Eswaran, MD, MS; Nancy Reau, MD


August 15, 2017

Weighing the Evidence

Hepatocellular carcinoma (HCC), which has an annual incidence rate of 1%-4%,[1] is a potential complication of hepatitis C virus (HCV) cirrhosis owing to a multistep carcinogenesis pathway. Previous HCV treatment with interferon was associated with poor cure rates and was limited to individuals with early fibrosis or well-compensated cirrhosis. Such therapy, however, also led to a decrease in HCC after a sustained viral response (SVR).[2,3] Although remarkable progress has occurred in curing HCV in the current direct-acting antiviral (DAA) era, there is controversy regarding the associated risk of patients developing either incident (new, de novo) or recurrent HCC after DAA treatment. Additional data also suggest a lower SVR rate in patients with HCC.

No Difference in HCC Risk

Given the interferon-era data, the assumption was that HCV cure via DAA treatment would also translate to a reduced incidence of HCC. This was challenged by a 2016 publication from Conti and colleagues[4] that retrospectively compared people with cirrhosis treated with DAAs. The authors looked at 285 patients without HCC and 59 with HCC (previously treated and under complete response prior to HCV therapy).[4] An ultrasound was performed at the end of treatment as well as 12 and 24 weeks later. At the final follow-up, HCC was detected in nine patients without previous HCC (3.16%; 95% confidence interval [CI], 1.45-5.90) and 17 with previous HCC (28.81%; 95% CI, 17.76-42.07). Given that these rates of de novo and recurrent cancers are similar to the historical data[5] in patients with HCV cirrhosis, the study authors concluded that SVR with DAAs does not reduce the occurrence of HCC.

Higher HCC Risk

Research conducted in 2016 and 2017 by Reig and colleagues,[6,7,8] however, concluded that there is a high rate of HCC recurrence following previous complete response to resection, ablation, or transarterial chemoembolization after treatment with DAAs. In a multicenter, retrospective evaluation of 77 patients with HCC under complete response before starting DAAs, the team noted a recurrence rate of 27.3% (21/77) with a median follow-up from the start of DAA therapy of 8.2 months. The study was limited by the brief follow-up, small sample size, and short duration of recurrence, suggesting that recurrent tumors were developing at initiation of HCV DAA treatment but were not causally related. The major criticism of this position is that patients may have small tumors before treatment that are only detected after they complete 3 months of therapy. Therefore, it is imperative that both researchers and clinicians complete a rigorous assessment to ensure that no residual tumor remains prior to DAA therapy.

There are other experiences published as letters and small retrospective datasets[9,10] that support an increased incidence of recurrent HCC after complete response, but there are no robust data showing an increased risk for incident tumors.

The mechanism by which patients with cirrhosis potentially experience a higher HCC rate may be related to DAA therapy's ability to reduce the tumor-specific immune response, specifically of HCC-specific CD8+ T cell.[11] This reduced immune theory is also implicated in the risk for hepatitis B reactivation in patients with HCV treated with DAAs. In addition, an abstract at the 2017 European Association for the Study of the Liver (EASL) meeting described HCV's epigenetic effects, noting that the marked changes in histone modifications by chronic HCV are only partially reversed by viral cure by DAAs.[12] This abnormal transcription may drive HCC tumorigenesis.

Lower HCC Risk

The initial data supporting a decreased HCC risk in patients with HCV treated with DAA were presented at the 2016 annual meeting of the American Association for the Study of Liver Diseases by Romano and colleagues.[13] These researchers found incidence rates that ranged from 0.23 per 100 person-years (PY) in patients without cirrhosis to 1.93 per 100 PY for those with cirrhosis, as compared with 2.8 per 100 PY among untreated HCV patients. However, they also found that those who developed incident cancers demonstrated aggressive infiltrative and multinodular phenotypes.

The most robust data to support a lower rate of HCC come from the French ANRS (France Recherche Nord & sud Sida-HIV Hépatites) agency. This group assessed the risk for HCC recurrence in three large prospective studies (HEPATHER, CirVir, and CULPIT) of HCV-infected patients treated with DAAs.[14] Patients were treated for HCC using curative procedures (hepatic resection, percutaneous ablation, or liver transplantation); those treated with chemoembolization were excluded. There was a lower rate of HCC recurrence in patients treated with DAAs compared with those not treated with DAAs among those in the HEPATHER (N = 267) cohort (24/189 [12.7%] vs 16/78 [20.5%]) and in the CirVir (N = 79) cohort (1/13 [7.7%] vs 31/66 [47%]). In the CUPILT (N = 314) cohort, the recurrence rate was 7/314 (2.2%). This study was limited by the registry nature of the cohort, which reduces its ability to identify relevant confounders.

Further analysis of the CirVir cohort was provided in an update at EASL 2017, which concluded that factors other than DAA treatment predict HCC recurrence, such as age older than 50 years, heavy alcohol use, and decompensated liver disease.[15]

These and other reports[13,16] demonstrate that the strongest risk for HCC development is advanced disease. Patients who were not candidates for interferon therapy previously because of advanced liver disease can now be treated with DAA therapy. This cohort, however, probably has a higher risk for HCC.

An additional study demonstrates a low risk of dropout due to neoplastic progression in HCV-HCC patients treated with DAAs who are awaiting liver transplant.[17]

Several other abstracts presented at EASL 2017 supported a lower incidence of liver cancer among people treated with DAAs compared with interferon treatment.[18,19,20,21] These data are limited by their retrospective single-center design, small sample sizes, and short follow-up.

In a regression meta-analysis of 26 studies of HCC occurrence and 15 studies of HCC recurrence (3875 participants overall), the adjusted relative risk of DAA versus interferon treatment was 0.75 for initial HCC occurrence and 0.62 for HCC recurrence.[22]

Retrospective data from large Veterans Health Administration hospital populations also support a reduced rate of HCC in patients with HCV treated with DAAs. In one report, among 22,500 (39% with cirrhosis) patients from 129 institutions, patients with SVR had a significantly reduced risk for HCC (0.90 HCC/100 PY) compared with patients without SVR (3.45 HCC/100 PY; hazard ratio, 0.28; 95% CI, 0.22-0.36).[23] In another report of 15,059 patients with advanced liver disease, HCC rates were 1.9 HCC/100 PY for patients with SVR and 11.5 HCC/100 PY for patients without SVR, an 83.5% reduction (P < .001).[24]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.