CROI 2015: Deferring Hepatitis C Treatment Can Lead to Liver Cancer and Death


HIV/HCV coinfected people who delay hepatitis C treatment remain at risk for liver failure, hepatocellular carcinoma and liver-related death even after being cured -- with outcomes worsening the longer it is put off -- indicating that treatment should not be deferred until advanced disease, according to a presentation at the 2015 Conference on Retroviruses and Opportunistic Infections (CROI) last week in Seattle. Treating only after progression to cirrhosis increased the risk of liver-related death by more than 5-fold and the duration of infectiousness by 4-fold.

Over years or decades chronic hepatitis C virus (HCV) infection can lead to advanced liver disease including cirrhosis (scarring), hepatocellular carcinoma (HCC; a type of primary liver cancer), and end-stage liver failure. People with HIV/HCV coinfection experience faster disease progression, on average, than those with HCV alone. Successful hepatitis C treatment reduces -- but does not eliminate -- the risk.

When the standard of care for chronic hepatitis C was interferon-based therapy -- which was poorly tolerated, lasted up to a year, and only cured about half of people with HCV genotype 1 -- experts generally recommended that patients put off treatment until they experienced liver disease progression. Now that highly effective and well-tolerated direct-acting antiviral (DAA) regimens are available, many providers and advocates believe that everyone living with hepatitis C should be eligible for treatment. But the high cost of the new drugs has led to restrictions on their use, with some insurers and public payers only covering treatment for the sickest patients.

Cindy Zahnd from the University of Bern and fellow investigators with the Swiss HIV and Swiss Hepatitis C Cohort Studies used a mathematical model to predict the impact of different HCV treatment strategies on liver fibrosis progression among HIV/HCV coinfected people.

The researchers fed the model data on new HCV infections among gay and bisexual men in the Swiss HIV Cohort Study, who are screened annually for HCV antibodies and ALT liver enzyme levels. The model simulated patients from the time of acute HCV infection through all stages of liver disease: absent fibrosis (Metavir stage F0), mild fibrosis (F1), moderate fibrosis (F2), severe or advanced fibrosis (F3), and liver cirrhosis (F4), as well as decompensated liver disease, liver cancer, and death. They assumed that age at the time of HCV infection and alcohol consumption influenced liver disease progression.

In the initial scenario, before the advent of DAAs, the researchers assumed that 60% of patients were treated with pegylated interferon plus ribavirin, with about a 40% sustained response rate (although much higher cure rates have been seen with interferon-based therapy started during acute HCV infection). They then ran a scenario using interferon-free DAA regimens, assuming 100% of patients were treated with an efficacy of about 90%.

The researchers then looked at the model's predictions about occurrence of liver-related events and death, as well as the duration of infectiousness (i.e., having detectable HCV RNA), comparing the strategies of treating all patients 1 month after HCV diagnosis (not necessarily 1 month after actual time of HCV infection), 1 year after diagnosis, or as they reached liver disease stages F2, F3, or F4. They did not take into account the potential for HCV reinfection after successful treatment, which can be substantial in some high-risk groups.


Starting treating after 1 month versus 1 year "did not really make a difference," Zahnd explained, but whether treatment is started at stage F2, F3, or F4 "really matters."

Importantly, the risk of progression to decompensation, liver cancer, or liver-related death did not fall to zero even after successful treatment. Some of these events occurred after HCV clearance, with post-clearance proportions increasing as treatment was further delayed. In fact, if therapy was deferred to stages F3 or F4, the majority of liver-related deaths occurred after being cured.

The remaining risk of liver complications and liver-related death after successful therapy indicates that people with HIV/HCV coinfection often "have other factors that keep them at risk of liver disease even after HCV clearance," Zahnd suggested. These may include antiretroviral drug toxicity, other coinfections, or metabolic liver disease.

"Our model suggests that timely treatment of HCV infection is important," the researchers concluded. "Patients can progress to end-stage liver disease after HCV clearance if treatment is delayed until later stages of liver disease due to imperfect treatment responses and residual fibrosis progression in HIV-infected patients."

These findings provide further support for a "different mental model" of hepatitis C treatment," David Thomas of Johns Hopkins said at a CROI press conference at which Zahnd and others described their hepatitis C studies. "We used to wait until [stage] F3 or F4 before treating. But that is changing."

After Zahnd's presentation her colleagueHans-Jakob Furrer from the Swiss HIV Cohort Study team noted that providing evidence for earlier treatment is why they did the study. "We're using this data for discussions with governments," he said. "It is making negotiations easier -- the data clearly support earlier [treatment] rollout."

Hepatitis C patients, providers, and advocates in the U.S. also have an arduous task, as instead of negotiating with a single national health plan they must contend with various government entities such as Medicaid, Medicare, the Veterans Health Administration, and state prison officials, as well as the insurance companies and pharmacy benefits managers that determine who should be treated and when.



C Zahnd, LP Salazar-Vizcaya, J-F Dufour, et al (Swiss HIV and Hepatitis C Cohort Studies Team). Impact of Deferring HCV Treatment on Liver-Related Events in HIV+ Patients. 2015 Conference on Retroviruses and Opportunistic Infections. Seattle, February 23-24, 2015. Abstract 150.