Co-infection with HIV and HCV does not increase the risk of end-stage liver disease or liver cancer

People with HIV and hepatitis C are no longer at higher risk of end-stage liver disease than people with hepatitis C alone, and the trend is probably associated with the improved effectiveness of antiretroviral treatment, a French study has reported in the journal Hepatology.

The study also found that people with HIV and hepatitis C virus (HCV) co-infection do not have a raised risk of liver cancer compared to people with HCV alone.

Past studies have shown that people with HIV and HCV experience much faster progression of liver fibrosis to cirrhosis and end-stage liver disease than people with HCV alone. This accelerated rate of disease progression is due to dampened immune responses to HCV in people with HIV, together with much higher levels of HCV replication and liver inflammation. Older antiretroviral drugs may also have contributed to liver damage.

The most recent meta-analysis, published in 2009, took data from 29 studies and calculated that people with HIV/HCV co-infection were five times more likely to develop decompensated cirrhosis and three and a half times more likely to die than their counterparts with HCV alone.

Most of those studies were carried out prior to the widespread uptake of early antiretroviral treatment, however, and prior to the availability of direct-acting antiviral treatment to cure hepatitis C.

French researchers wanted to know whether the prognosis of people with HIV/HCV co-infection has changed as treatment patterns for HIV and HCV have changed.

Using data from the HEPAVIH and CirVir cohorts, they investigated rates of decompensated cirrhosis, liver cancer and death in people with HIV and HCV compared to HCV alone.

Liver cancer and decompensated cirrhosis – the onset of end-stage liver disease when the liver ceases to compensate for damage caused by HCV – were chosen as endpoints in this study because they are unambiguous clinical events. Previous studies assessing the impact of HIV on HCV disease progression have used varying measures of fibrosis and cirrhosis, creating difficulties in comparing between studies and patients.

The study population consisted of people with early-stage (Child-Pugh A) cirrhosis confirmed by biopsy. The HEPAVIH cohort contributed 175 people with HIV/HCV co-infection and cirrhosis and the CirVir cohort contributed 1253 people with hepatitis C and cirrhosis. People also with hepatitis B co-infection were not included in this analysis, nor were people with a prior history of liver complications.

The cohorts were followed for just under five years in the case of the HCV cohort and four and a half years for the co-infected cohort, from the date of inclusion in the cohort to which they belonged or biopsy-proven diagnosis of cirrhosis, whichever was later.

At baseline those with co-infection were younger (47 years vs 56 years, p < 0.001), more likely to have genotype 3 infection (25% vs 15%, p < 0.001) and were more likely to be current alcohol users (40 vs 25%, P < 0.001). 19.9% per cent of the HCV cohort and 16.6% of the co-infected cohort had already been cured of hepatitis C at baseline.

Almost all people with co-infection (92%) were already on antiretroviral treatment at baseline and 80% had an undetectable viral load.

At the end of the follow-up period the investigators found:

• No difference in cure rate between the two cohorts (47% HIV vs 52% HCV).
• No significant difference between cohorts in the incidence of hepatocellular carcinoma (HCC) (liver cancer) (14% HCV vs 7.4% HIV), although liver cancer was more severe at diagnosis in people with co-infection despite comparable frequency of monitoring for HCC.
• No difference in the proportion who suffered at least one episode of decompensation (liver decompensation can be reversed) (16.4% HCV vs 10.9% HIV).
• No difference in the interval between baseline and first occurrence of decompensation.
• No difference in the death rate between the two cohorts (13% in each) although people with HIV died a median of ten years younger, had a higher risk of death overall and a higher risk of non-liver-related death (SHR = 2.44, 95% CI 1.18-5.07, p = 0.016).

Multivariate analysis showed that decompensation or death were each associated with a lack of hepatitis C cure or greater severity of cirrhosis, as measured by platelets < 100,000/mm³ and albumin < 35 g/L. Death was also associated with older age.

It was not possible to analyse the effect of antiretroviral treatment on the study outcomes within the HIV cohort because the proportion of people already receiving treatment at baseline was so high, but the study authors argue that their results indicate that antiretroviral treatment has altered the course of hepatitis C co-infection in people living with HIV.

They also argue that the abandonment of drugs such as stavudine and didanosine has reduced the incidence of liver damage in people living with HIV.

They caution that despite these improvements, liver cancer remains more aggressive in people with co-infection, underscoring the importance of curing hepatitis C in people living with HIV.