This year's conferences covering HIV/AIDS and liver disease featured numerous presentations concerning hepatitis B virus (HBV) and hepatitis C virus (HCV) infection in people with HIV. Below are highlights from several of the most interesting reports.
Numerous studies looked at hepatitis C treatment in people coinfected with HIV and HCV. HCV treatment has improved in recent years, but as many as half of all treated patients do not achieve a sustained virological response (SVR) -- an undetectable HCV viral load six months after finishing treatment -- which is usually considered a cure. Plus, response rates are lower for coinfected people as compared to those with HCV alone. Many presentations focused on individually tailored treatment -- including higher doses of peg-interferon or ribavirin and longer treatment length -- as well as looking at which factors predict response to treatment.
The Spanish PRESCO trial studied one brand of peg-interferon (Pegasys) plus ribavirin in 389 people coinfected with HCV and HIV. Those with hard-to-treat HCV genotypes (1 or 4 -- about 60% of the total) received treatment for either the standard duration of 48 weeks or for 72 weeks. Those with genotypes 2 or 3 (which respond better to treatment) received either the standard 24 weeks or 48 weeks. At the 14th Conference on Retroviruses and Opportunistic Infections (CROI) in February, researchers reported that the overall SVR rates were 35% for people with genotype 1 or 4 and 72% for those with genotype 2 or 3, but extending the length of treatment did not reduce the risk of relapse for any of the genotypes.
The higher SVR rate in this study than in earlier ones (for example, the pivotal APRICOT trial reported SVR rates of 29% for genotype 1 and 62% for genotypes 2 or 3) may be due to the fact that people in PRESCO had higher CD4 counts (at least 300) and were using higher doses of ribavirin, based on body weight. In addition, they were less likely to use Videx or Retrovir (AZT), both of which can lead to worse side effects when taken with ribavirin.
The researchers found that an undetectable HCV viral load four weeks after starting treatment predicted eventual SVR. As is the case for HIV-negative people, those who did not demonstrate a four-week response were unlikely to later achieve an SVR, suggesting that the four-week mark can guide decisions about whether to stop treatment early. Interferon can cause difficult side effects, so many people prefer to stop treatment that is not likely to produce a cure. Starting treatment with a lower HCV viral load also predicted a higher chance of SVR in people with genotype 1.
At the Digestive Disease Week (DDW) meeting in May, Alex Monto reported surprisingly good outcomes in a study of 21 HIV-positive and 112 HIV-negative patients with hepatitis C. Coinfected people actually had a slightly higher SVR rate than those with HCV alone (62% vs. 47%), although the difference was not large enough to reach statistical significance. However, the coinfected patients were about twice as likely (48% vs. 23%) to need EPO injections to manage anemia due to ribavirin. Coinfected people taking "friendly" nucleosides (any other than Retrovir, Videx, or Zerit) or none at all were more likely to respond to interferon treatment.
In another study, presented at the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention (IAS) in July, Michael Polis and colleagues treated 11 coinfected patients with Pegasys either once weekly for 48 weeks or twice weekly for four weeks followed by once weekly for an additional 44 weeks. Everyone also took ribavirin. After one week, 72% in the double-dose arm achieved more than a one log reduction in HCV viral load, compared with 9% in the standard-dose arm. Side effects were similar in both groups, but people in the double-dose arm were more likely to achieve an SVR (40% vs. 18%).
Research has shown that people who achieve an SVR experience slower liver disease progression, and may even have a reduction in the degree of fibrosis or cirrhosis (scarring of the liver). In promising news, Anna De Bona reported that interferon appears to reduce the rate of cirrhosis progression even in coinfected people who do not achieve an SVR. In a look back at 25 untreated patients and 25 who were treated but did not achieve an SVR, half the untreated patients saw their cirrhosis progress, compared to just 16% of the non-responders.
Other encouraging news came from Mark Swain, who reported at the DDW meeting that people who achieve an SVR after 24 weeks of treatment have a very good chance of remaining permanently free of HCV. In a study of HIV-negative patients who achieved an SVR, 989 out of 997 -- or 99% -- still had undetectable HCV viral load five years later. Further research is needed to find out if this will also hold true for coinfected individuals.
In recent years, outbreaks of HCV that appears to be sexually transmitted have been reported among gay and bisexual men in large European cities, most of whom also have HIV. To date, nearly 400 such cases have been reported in London and Brighton in the U.K., with smaller clusters in France, Germany, and the Netherlands. Similar small outbreaks have also been reported in Australia and the U.S.
At the IAS conference, Mark Danta reported on a study exploring how these outbreaks are related. The researchers constructed "family trees" of HCV taken from 107 men who have sex with men (MSM) in the U.K., 51 in the Netherlands, 24 in Germany, and eight in France. Although half the men had HCV genotype 1a, which is common in Europe, an unusually high number had genotype 4d, the predominant type in Africa and the Middle East. The analysis revealed ten clusters of related HCV strains which accounted for 88% of all infections. Seven clusters contained HCV strains from more than one country, while four included HCV from more than two countries. These findings suggest a large HCV transmission network among HIV-positive MSM in Europe, likely due to travel between cities. (Powerpoint)
Axel Schmidt studied 22 coinfected MSM in Germany with no history of injection drug use and 44 people with HIV but no HCV. Risk factors for HCV included use of intranasal drugs like cocaine, a history of major surgery, group sex, more than five episodes of unprotected anal sex within the past year, fisting, and use of Viagra. In a multi-factor analysis, however, only intranasal drug use and anal injuries during sex remained statistically significant. Julie Fox reported on 12 MSM with recent HIV infection in London who also became infected with HCV. The median time between HIV and HCV infection was 17 months. Although there was only one new HCV diagnosis in 2003 and 2004, the number rose to six in 2005 and four in 2006. All the men reported unprotected anal sex and use of recreational drugs (including intranasal use).
Martin Fisher from Brighton reported at CROI that sexual transmission of HCV also occurs among HIV-negative MSM. After screening more than 7,000 patients at an HIV/STD clinic since 2000, the researchers identified 25 new HCV infections: 16 in HIV-positive men, five in HIV-negative men, and four in men of unknown HIV status. The number increased from zero cases in 2000-2002 to 13 cases in 2006. New HCV diagnosis was linked to fisting, unprotected anal sex, multiple partners, other STDs, and non-injection drug use (including intranasal use). Although HIV-positive men were about 13 times more likely to be diagnosed with hepatitis C, this study shows that HIV-negative MSM are also at risk.
Together, these studies emphasize the need for increased education and prevention efforts, and suggest that at-risk MSM -- both HIV-positive and -negative -- should be routinely screened for hepatitis C.
Structured HIV treatment interruption is a controversial issue, with several recent studies showing that certain types of treatment breaks encourage disease progression. The largest of these, the SMART trial, found that stopping HAART (highly active antiretroviral therapy) when the CD4 count fell below 350 and restarting when it reached 250 led not only to more HIV-related opportunistic illnesses, but also to a higher rate of cardiovascular, kidney. and liver complications compared with continuous HAART.
Two studies at the IAS conference looked at people with hepatitis B or C in SMART. Of the 5,472 people with HIV in the study, 2% had HBV, 14.6% had HCV, and 0.25% had both HBV and HCV. There were few deaths due to opportunistic illnesses in either the continuous treatment or the interruption group, and overall death rates were comparable.
But Ellen Tedaldi reported that people coinfected with HBV or HCV had nearly four times the risk of death due to non-opportunistic illnesses. They made up just 17% of all participants, but they accounted for almost 50% of non-opportunistic deaths. Although the most common non-opportunistic causes of death among the coinfected patients were related to substance abuse (overdose, for example) and non-AIDS defining cancers, coinfected people were more likely to die of liver or kidney disease than people with HIV alone, and more often died from unknown causes. The researchers concluded that HIV treatment interruptions may be particularly harmful for people with HBV or HCV, who already have a higher underlying risk of death. (Powerpoint)
In the second SMART analysis, Greg Dore and colleagues studied whether coinfected patients were more likely than those with HIV alone to have restarted HAART because their CD4 count fell below 250. Since some HIV meds are active against both HIV and HBV, people with both viruses who are taking these drugs face an increased risk of progression of both diseases when they interrupt therapy (study investigators discouraged enrollment of people who needed antiretroviral drugs to control HBV). Indeed, the researchers found that HIV/HBV coinfected participants were significantly more likely to have restarted HAART, and did so sooner after treatment interruption, than either HIV/HCV coinfected patients or those with HIV alone. Although the number of HIV/HBV coinfected participants in SMART was small, these findings suggest they may be at risk for faster HIV disease progression after interrupting HAART. (Powerpoint)
Past research has shown that people with both HIV and HCV tend to experience more rapid liver disease progression than people who have HCV alone, suggesting that they may benefit from earlier hepatitis C treatment. Some recent studies, however, suggest that faster progression is linked to CD4 counts rather than HIV infection itself.
At the IAS meeting, Hla Thein presented results of an analysis of 17 studies of liver fibrosis progression in HIV/HCV coinfected individuals. The overall probability of developing liver cirrhosis was 24% after 20 years and 75% after 40 years -- higher than the rate seen in HIV-negative people with HCV, which is around 14% after 20 years. Use of HAART, however, was linked to slowed liver disease progression. (Powerpoint)
In a study presented at the 42nd European Association for the Study of the Liver (EASL) meeting in April, Marianne Ziol compared liver fibrosis progression and immune responses in 33 HIV-positive and 33 HIV-negative people with HCV. Everyone with HIV had a CD4 count above 250 and none in either group had been treated for hepatitis C. The researchers found that both groups had similar fibrosis scores (2.69 vs. 2.72 on the Ishak scale of 0-4) and a similar fibrosis progression rate. The frequency and strength of HCV-specific immune responses were also similar. Regardless of HIV status, faster fibrosis progression was associated with reduced HCV-specific immune response.
Two studies presented at CROI also support a link between immune function and liver disease progression. In a study of 1,101 HIV/HCV coinfected Spanish patients, Jose Garcia-Garcia reported that the rate of end-stage liver disease was relatively low: about 6% developed decompensated cirrhosis and 3% died of liver-related causes. But complications like abdominal swelling (ascites), enlarged liver, and death due to liver disease were more common among people who had smaller CD4 cell gains (less than 100) after starting HAART. Looking at coinfected participants in the Swiss HIV Cohort, Andri Rauch and colleagues found that those with low CD4 counts had significantly higher HCV viral loads, though what this means in terms of disease progression is not known.
Another study presented at the same conference suggested that HIV-positive people newly infected with HCV may progress faster to advanced liver disease. Daniel Fierer and colleagues performed liver biopsies on five HIV-positive New York City men with acute HCV infection. Despite being recently diagnosed with hepatitis C, four already had moderate (stage 2) fibrosis and moderate or severe liver inflammation. Typically, it takes years or decades for fibrosis to progress to this extent. All five men had no other contributing factors (like heavy alcohol use), and all had relatively high CD4 cell counts (from about 250 to 700). All but one were on HAART. Although this study was small, it suggests that liver disease progression may be accelerated in people who already have HIV at the time of HCV infection, underlining the need for further research.
Looking at HBV/HIV coinfection, Chloe Thio reported at CROI that, contrary to previous reports, the hepatitis B drug Baraclude (entecavir) is also active against HIV and may lead to HIV drug resistance.
Several drugs (Epivir, Emtriva, and Viread) are active against both HIV and HBV, and experts advise that HIV/HBV coinfected people should include at least one of these in their antiretroviral regimens. However, Baraclude was often recommended for coinfected patients being treated for hepatitis B who did not yet require HAART, since using a single HIV drug leads to drug resistance.
Thio's team identified three HIV/HBV coinfected patients not on HAART whose HIV viral load dropped by at least a log after starting Baraclude. They then confirmed the activity of Baraclude against HIV in laboratory tests, and found that it led to the emergence of the M184V mutation, which causes resistance to Epivir and Emtriva. This contradicts extensive prior testing by the drug's maker, Bristol-Myers Squibb, showing that Baraclude was not active against HIV. At the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) in September, the same team reported that Baraclude only partially inhibits HIV, and that the M184V mutation emerges slowly, and not in all cases.
Based on these findings, Bristol-Myers Squibb and the FDA revised the Baraclude product label, and treatment guidelines were updated to recommend that the drug not be used alone by HIV/HBV coinfected patients who are not also taking HAART. This leaves Tyzeka (telbivudine) as the only approved HBV treatment (besides interferon) that appears to have no activity against HIV.
Links to more abstracts are available at the following conference websites:
Conference on Retroviruses and Opportunistic Infections (CROI):
European Association for the Study of the Liver (EASL):
International AIDS Society (IAS):
Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC):
Liz Highleyman is a San Francisco-based freelance medical writer.