HIV and Hepatitis Coinfection
Hepatitis C and hepatitis B are viral infections of the liver; over time they can lead to serious consequences including liver cirrhosis and liver cancer. Most studies show that HIV infection leads to more aggressive hepatitis C or hepatitis B and a higher risk of liver damage. Studies of how HCV and HBV affect HIV disease are less clear. Most research shows that HCV does not accelerate HIV disease progression, but HIV/HCV coinfection may impair immune system recovery after starting antiretroviral therapy.
Coinfection can complicate treatment. People with liver damage due to chronic hepatitis are more likely to experience hepatotoxicity (liver toxicity) related to anti-HIV drugs. In addition, drugs used to treat HIV and hepatitis can interact and side effects may be exacerbated. Most experts recommend that HIV should be controlled first before a person begins HCV treatment. With careful management, most people with HIV/HCV or HIV/HBV coinfection can be successfully treated for both diseases. In fact, several recent studies suggest that HIV/HCV-coinfected people with well-controlled HIV disease and relatively high CD4 cell counts may do as well as those with HCV alone.
HIV/HCV coinfection is increasingly recognized as a growing public health problem. Early in the HIV/AIDS epidemic most people with HIV were expected to die from AIDS, and less attention was devoted to other long-term conditions. Because chronic hepatitis C progresses so slowly, many HIV positive people who were infected with HCV in the 1970s or 1980s are only now beginning to develop advanced liver disease. At several recent conferences a number of presentations were devoted to HIV/HCV coinfection. David Thomas, M.D., from Johns Hopkins University in Baltimore, Maryland, reviewed the current state of knowledge about HIV/HCV coinfection at a June 2002 meeting convened by the National Institutes of Health (NIH) to generate a new consensus statement about HCV treatment, care, prevention, and future research.
As improved HIV treatment has reduced mortality due to opportunistic illnesses (OIs), liver failure -- often related to chronic viral hepatitis -- has become a major cause of hospitalization and death in people with HIV/AIDS. In some recent studies liver failure due to HCV was the leading cause of death.
Several studies at the 9th Conference on Retroviruses and Opportunistic Infections (CROI) held in February 2002 looked at illness and death in HIV/HCV-coinfected people. Ronald Reisler, M.D., M.P.H., of the NIH and colleagues reported that the rates of severe (grade 4) adverse events and death were higher in HIV positive people coinfected with HCV or HBV than in those with HIV alone. David Rimland, M.D., of the Atlanta Veterans' Administration Medical Center and colleagues showed that coinfected people had shorter survival times after an HIV or AIDS diagnosis than those with HIV alone. Kelly Gebo, M.D., of Johns Hopkins and colleagues found that HIV/HCV coinfection substantially increased the likelihood of hospitalization compared with those who had only HIV. On the other hand, Ellen Tedaldi, M.D., of Temple University in Philadelphia and colleagues reported that after controlling for CD4 cell count, survival rates were comparable in HIV/HCV-coinfected people and those with HIV alone, suggesting that effective HIV treatment can minimize the detrimental effects of HCV coinfection.
Because the presence of HIV accelerates the progression of hepatitis C, HCV is thought of as an OI in people with HIV; however, it is not considered an AIDS-defining illness.
Along with these similarities, there are also several differences between the two viruses. HCV, unlike HIV, does not integrate into human cells and is thus easier to eradicate. HCV is less likely than HIV to be transmitted sexually or from mother to child during pregnancy, birth, or breast-feeding. According to the Centers for Disease Control and Prevention (CDC), people who contracted HIV through sexual activity have HCV infection rates similar to those of the adult population as a whole (estimated at under 3% for people in monogamous heterosexual relationships, but somewhat higher among gay men and people with multiple sex partners).
However, studies show that the risk of sexual or perinatal transmission of HCV is greater if a person also has HIV -- possibly due to the fact that HIV/HCV-coinfected people tend to have higher HCV viral loads. Recent data reported by researchers from Chelsea and Westminster Hospital in London suggest that sexual transmission is responsible for an increasing proportion of HCV infections among people with HIV. And while 5% or less of HCV-infected mothers without HIV transmit HCV to their infants, among HIV/HCV-coinfected mothers the transmission rate may be three times as high.
Because many people are coinfected with HCV and HIV, the U.S. Public Health Service and the Infectious Disease Society of America recommend that all people with HIV should be screened for HCV. Detecting HCV in people whose immune systems are severely compromised can be difficult because they may not produce enough antibodies to show up on a test. In HIV positive people with a CD4 cell count over 200 cells/mm3, a standard HCV antibody test is usually sufficient; if the CD4 cell count is below 200 cells/mm3, an HCV RNA viral load test may be necessary to diagnose hepatitis C.
Most people with acute or chronic HCV have no symptoms. Those that do may experience fatigue, nausea, loss of appetite, abdominal pain, and a flu-like feeling. An estimated 10-25% of people with chronic HCV develop severe liver disease -- usually after 10-40 years -- which may include liver inflammation, fibrosis (the development of tough, stringy tissue in the liver), cirrhosis (scarring), hepatocellular carcinoma (liver cancer), and liver failure. A minority may develop jaundice (yellowing of the skin and whites of the eyes). When people develop decompensated cirrhosis, scar tissue blocks the flow of blood through the liver and the organ is no longer able to function properly. This can lead to serious conditions such as bleeding veins (varices) in the esophagus or stomach, abdominal swelling (ascites), and brain dysfunction (hepatic encephalopathy). Liver failure due to HCV is the leading reason for liver transplants in the U.S.
Antibody tests (ELISA and RIBA) are used to detect whether a person has been infected with HCV. Genotype tests are used to determine what strain of HCV a person has. There are six known HCV genotypes; 1a and 1b, which are most common in the U.S., are more difficult to treat. Liver enzymes, in particular alanine transaminase (ALT) and aspartate transaminase (AST), are measured as an indication of liver inflammation. Many -- but not all -- people with chronic hepatitis have elevated liver enzyme levels. Viral load tests (PCR, bDNA, and TMA) measure the amount of HCV genetic material (RNA) in the blood, and can help indicate whether treatment is working. In contrast with HIV, HCV viral load is not correlated with disease severity.
Liver biopsy, in which a small sample of tissue is withdrawn using a needle and examined under a microscope, is considered the "gold standard" for gauging the extent of liver damage. Biopsies are used to help make decisions about whether treatment is needed. Liver tissue damage is graded on a scale of 0-4. Although several tests are under study, there currently is no reliable noninvasive means of detecting liver fibrosis.
Not everyone with HCV needs to be treated. Many different factors -- such as a person's age, how long he or she has been infected, HCV genotype, and extent of existing liver damage -- should be taken into account when deciding whether to treat. The usual treatment for HCV is a combination of interferon-alpha (Intron-A or Roferon-A) plus ribavirin (Rebetol, Copegus). Pegylated interferon (Peg-Intron or Pegasys) is a new, chemically altered form of interferon that lasts longer in the body and appears to work better than standard interferon.
Recent studies show that combination therapy with pegylated interferon plus ribavirin can clear HCV in about 50% of HIV negative people with genotype 1 and about 80% of those with genotypes 2 or 3. As discussed below, treatment response rates tend to be lower in people coinfected with HIV. Traditionally, HCV therapy is administered for a specified period of time (usually 6-24 months) and discontinued if HCV viral load does not decrease. However, experts increasingly believe that treatment may reduce liver damage even if HCV viral load does not become undetectable. A trial called HALT-C is now underway to study the possible benefits of long-term HCV maintenance therapy.
Side effects of interferon are common, and may include fever, fatigue, headaches, flu-like symptoms, muscle aches, low blood cell counts, and irritability or depression. Ribavirin may cause hemolytic anemia (destruction of red blood cells) and birth defects.
Much of the liver damage related to hepatitis C is caused not by the virus itself, but rather by the immune system's response to HCV. Thus, it might be expected that people with compromised immune systems would mount a weaker immune response that causes less liver tissue damage. However, research indicates that the opposite seems to be the case.
Most studies show that HCV disease progression is more rapid in HIV/HCV-coinfected people and is more likely to lead to severe liver damage. Many studies show higher rates of cirrhosis in coinfected individuals. For example, at the November 2001 meeting of the American Association for the Study of Liver Disease, Vincent di Martino, M.D., and colleagues with the French MULTIVIRC study team reported that coinfected people died earlier than those with only HCV because they progressed to cirrhosis sooner. At the XIV International AIDS Conference held in Barcelona this past July, A.H. Mohsen from St. Thomas School of Medicine in London and colleagues estimated that the average time between HCV infection and the development of cirrhosis is 22 years in coinfected people compared with 33 years in those with HCV alone -- a 1.5-fold increase in the rate of liver disease progression.
At the June NIH consensus meeting Dr. Thomas cited a meta-analysis showing that HIV/HCV-coinfected people had a two-fold greater risk of progression to cirrhosis and a six-fold greater chance of developing end-stage liver disease. Javier Garcia-Samaniego from Madrid, Spain, and colleagues reported in the January 2001 issue of the American Journal of Gastroenterology that HIV/HCV-coinfected individuals had a higher rate of hepatocellular carcinoma than people with HCV alone. In addition, most research indicates that HIV/HCV-coinfected people typically have higher HCV viral loads than those with only HCV.
Importantly, the longer-term studies of HIV/HCV coinfection were initiated prior to the widespread use of highly active antiretroviral therapy (HAART), and many of the participants had low CD4 cell counts. More recent research suggests that the differences in HCV disease progression between coinfected people and those with HCV alone may not hold for people who have well-controlled HIV and whose immune systems remain relatively intact with high CD4 cell counts. For example, Yves Benhamou, M.D., and colleagues, also with the MULTI-VIRC team, reported in the October 1999 issue of Hepatology that HIV/HCV coinfection, having a CD4 cell count below 200 cells/mm3, and heavy alcohol consumption were all associated with a greater likelihood of HCV disease progression. But HIV/HCV-coinfected participants with over 200 cells/mm3 had hepatitis C progression rates similar to those of people with HCV alone.
Massimo Puoti, M.D., and colleagues with the HIV-HCV Coinfection Study Group reported in the January 1, 2001 issue of the Journal of Infectious Diseases that the presence of fibrosis was associated with CD4 cell counts below 500 cells/mm3, whether or not a person had HIV. And at the Barcelona AIDS conference C. Arizcorreta and colleagues from Cadiz, Spain, presented research showing that both lower CD4 cell counts and higher HIV viral loads were associated with worse liver damage in coinfected individuals.
In a study published in the August 2001 issue of Hepatology, Dr. Benhamou's team reported that people receiving antiretroviral regimens that included a protease inhibitor (PI) had lower fibrosis scores and lower rates of progression to cirrhosis than those who had not taken PIs (2% vs. 5% progression rate at five years, 5% vs. 18% at 15 years, and 9% vs. 27% at 25 years). This and other research suggests that by keeping their HIV under control, HIV/HCV-coinfected people may do nearly as well as people with HCV alone.
Similarly, Dr. Rimland's team reported in the July 1999 issue of Clinical Infectious Diseases that HIV/HCV coinfection appeared to have no effect on the progression of HIV disease or survival in 100 coinfected people treated between January 1992 and May 1997. Dr. Rimland concluded that there was "absolutely no difference" in HIV disease progression and survival -- as assessed by time from HIV diagnosis to AIDS diagnosis, from HIV diagnosis to death, or from AIDS diagnosis to death -- in HIV/HCV-coinfected people compared with those who had HIV alone.
However, not all research confirms that HCV has no effect on HIV progression. Some studies suggest that infection with certain HCV genotypes may be associated with more rapid progression to AIDS or death. Dr. Sherman's team found that HCV genotype 1 is more common in people with HIV/HCV coinfection (about 83%) than in people with HCV alone (about 70%), which may contribute to more aggressive hepatitis C. In the October 14, 2002 issue of the Archives of Internal Medicine, Andrea de Luca, M.D., and colleagues with the Italian Cohort Naive Antiretrovirals Study Group reported that coinfection with HCV (but not HBV) was associated with increased risk of progression to AIDS-defining illness and death. In addition, coinfected people in this study experienced "consistently reduced recovery" of CD4 cells when treated with HAART compared with those who had HIV alone.
Several other studies confirm that immune recovery after starting HAART may be impaired in HIV/HCV-coinfected people. For example, in the November 25, 2000 issue of The Lancet Gilbert Greub, M.D., and colleagues with the Swiss HIV Cohort Study reported on response to HIV treatment in 3,111 HIV positive participants, 1,157 of whom also had HCV. Both the coinfected participants and those with HIV alone were equally likely to achieve HIV viral loads below 400 copies/mL after starting HIV treatment. But the coinfected people were less likely than those with HIV alone to experience a CD4 cell count increase of at least 50 cells/mm3 (75% vs. 84%). By the end of follow-up, about 8% of the coinfected participants had developed an OI compared with about 5% of those with HIV alone, and the death rate due to all causes was more than twice as high among the coinfected participants.
More recent studies offer similar results related to immune recovery. At the 8th Retrovirus conference in February 2001, J. Martin and colleagues from Madrid reported that among 902 study participants with HIV -- 72% of whom were coinfected with HCV -- responses to HAART differed dramatically. Participants with HIV alone experienced an average HIV viral load decrease of over 5,700 copies/mL and an average CD4 cell count increase of 111 cells/mm3. In contrast, the HIV/HCV-coinfected participants experienced an HIV viral load decrease of only 606 copies/mL and a CD4 cell count increase of just 53 cells/mm3. At the Barcelona AIDS conference Juan Antonio Pineda and colleagues from Seville, Spain, presented evidence showing that CD4 cell recovery after starting HAART is slower in HIV/HCV-coinfected people compared with those who have HIV alone. Likewise, Maria Dorrucci, M.D., and colleagues from Rome also reported at the same conference that coinfected people had a poorer response to HAART.
The question of how HCV affects HIV disease remains unsettled. As HIV/HCV-coinfected people live longer, more data will become available that should shed light on how HCV infection influences the long-term natural history of HIV disease.
The effect of HCV on HIV disease is less clear, but most studies show that HCV does not increase HIV viral load or directly accelerate HIV disease progression.
Coinfection with HCV does appear to impair immune recovery after starting HAART.
People with HIV who are diagnosed with hepatitis C should be evaluated and considered for HCV treatment. Many people with chronic hepatitis C -- coinfected or not -- do not experience symptoms for decades and may never develop serious liver disease. Most experts recommend against treatment for people who are asymptomatic, have normal liver enzyme levels, and have minimal existing liver fibrosis; many believe that for such people, watchful waiting with regular monitoring is a better option. For people with moderate to severe liver fibrosis, most physicians recommend that hepatitis C should be treated.
Treatment is most successful in people who have low HCV viral loads, are under age 40, are female, have HCV genotypes other than 1, and have not yet sustained extensive liver damage. Interferon is usually not recommended for people with decompensated cirrhosis, although some may be treated in clinical trials. Treatment decisions should be guided by biopsy results, not liver enzyme levels alone, since some people maintain persistently normal ALT levels despite progressive liver damage. HCV viral load is likewise not a good marker of disease progression. However, reductions in ALT and HCV viral load can be useful indications that treatment is working.
In general, guidelines for deciding whether to treat hepatitis C apply to coinfected people. However, some experts believe that it is best to treat HCV early in people with HIV, while their immune systems are still functioning well and their CD4 cell counts are still high. Studies have shown that coinfected people with higher CD4 cell counts respond better to HCV treatment. This may be because interferon works in part by boosting immune activity, and the immune systems of people with advanced HIV disease may be too damaged to respond. In addition, interferon can sometimes cause a decrease in CD4 cell count (although the CD4 percentage may remain the same). Thus, it may be appropriate to treat some asymptomatic HIV/HCV-coinfected people who might otherwise not be considered candidates for treatment if they had HCV alone.
Hepatitis C treatment generally is not recommended for people with advanced HIV disease and severe immunosuppression. Those with CD4 cell counts below 200 cells/mm3 or a concurrent OI are usually not considered good candidates for HCV treatment. Therefore, OIs should be treated and antiretroviral therapy should be used to control HIV and raise the CD4 cell count before starting HCV treatment.
However, while experts generally recommend that HIV be treated first before beginning HCV therapy, there are some cases in which the opposite approach may be preferable. In people who are newly infected with HIV and have minimal immune system damage and high CD4 cell counts, but advanced hepatitis C, HCV treatment might be started first to control liver damage and improve the chances of tolerating anti-HIV drugs in the future. Treatment decisions should be made on an individual basis. Once treatment is underway, regular monitoring should be done to assess liver disease progression. Dr. Thomas suggests that liver biopsies be repeated perhaps every three years in coinfected people who are not receiving HCV therapy.
Much remains to be learned about how HCV treatment impacts HIV disease and vice versa. Some researchers have reported that interferon therapy leads to decreased CD4 cell counts, while others have found no such effect. HCV treatment does not appear to directly affect HIV viral load. Similarly, anti-HIV drugs do not appear to have a direct effect on HCV. While a majority of researchers have reported that HAART does not affect HCV viral load, others have seen HCV RNA increases in people starting antiretroviral therapy for HIV. In some cases HAART can produce an increase in liver inflammation (a "flare") as anti-HIV drugs improve immune system function. Clearly, more research is needed in this area.
Studies have shown that combination therapy with interferon-alpha plus the antiviral drug ribavirin is more effective than interferon monotherapy. Ribavirin is a nucleoside reverse transcriptase inhibitor (NRTI) like many anti-HIV medications. (In fact, ribavirin was studied as an HIV treatment but did not show much promise and was never approved for this indication.) As is the case with therapy for HIV, using multiple drugs that work by different mechanisms appears to be the most effective approach to treating hepatitis C.
As mentioned above, pegylated interferon is a recently developed long-acting formulation of interferon that can be injected less often, typically once per week. Pegylation is a process in which polyethylene glycol is attached to a protein to extend its activity in the body. Schering-Plough's Peg-Intron brand of pegylated interferon-alpha-2b was approved in 2001; Roche's Pegasys brand of pegylated interferon-alpha-2a was approved this past October. Most studies show that treatment response rates for combination therapy with pegylated interferon plus ribavirin are higher than those for standard interferon plus ribavirin.
To date, no treatments have been specifically approved for use by individuals with HIV/HCV coinfection, but most physicians use the same regimens for coinfected people as they do for those with HCV alone. In general, research has shown that treatment response rates for all types of HCV therapy are somewhat lower for HIV/HCV-coinfected people than for those with HCV alone. However, some studies suggest that people with well-controlled HIV disease and high CD4 cell counts may respond nearly as well to HCV treatment as those without HIV. Whether a person is coinfected or not, the relative effectiveness of the various regimens appears to be the same: pegylated interferon plus ribavirin is generally superior to standard interferon plus ribavirin, which in turn is more effective than interferon monotherapy. Likewise, treatment response rates are consistently higher for people with HCV genotypes 2 or 3 compared with those who have genotypes 1 or 4, whether or not they are coinfected with HIV.
Numerous studies have looked at the safety and efficacy of standard interferon with or without ribavirin in HIV/HCV-coinfected people. But the most recent results presented at conferences concern pegylated interferon, which is increasingly considered the standard of care for chronic hepatitis C.
The most promising results to date are from AIDS Clinical Trials Group (ACTG) study 5071. In a late-breaker presentation at the February 2002 Retrovirus conference, Raymond Chung, M.D., from Massachusetts General Hospital in Boston reported on an analysis of 133 coinfected participants treated with ribavirin plus either Pegasys or standard interferon. After 24 weeks, 44% of those in the Pegasys arm achieved an undetectable HCV viral load compared with 15% of those in the standard interferon arm using an intent-to-treat (ITT) analysis (in which all participants were analyzed whether or not they completed therapy).
Among those with HCV genotype 1, the response rates were 33% for Pegasys and 7% for standard interferon; among those with genotypes 2 or 3, the respective response rates were 80% and 40%. These response rates -- especially for genotype 1 -- are somewhat lower than those seen in studies of people with HCV alone. At the time of this presentation, sustained virological response rates (undetectable HCV viral load six months after the end of treatment) were not yet available and the study is continuing. Severe adverse side effects were more common in the pegylated interferon group (17 events) than in the standard interferon group (5 events), but dropout rates were similar in both groups (15% and 12%, respectively).
At the 42nd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) this past September, Margaret Hoffman-Terry, M.D., of Lehigh Valley Hospital in Allentown, Pennsylvania, and colleagues reported results from 119 coinfected people (most with HCV genotype 1) treated with Pegasys with or without ribavirin. After 24 weeks, 39% experienced at least a 2 log decrease in HCV viral load.
Results from other studies do not look so good, however. A trial by Christian Perronne, M.D., and colleagues from Paris (the RIBAVIC study) -- also reported at the September ICAAC -- compared Peg-Intron plus ribavirin against standard interferon plus ribavirin in 416 HCV-treatment-naive coinfected individuals, most of whom were taking HAART. After 48 weeks, 38% in the Peg-Intron group and 24% in the standard interferon group achieved an undetectable HCV viral load using an ITT analysis. Using a less rigorous as-treated analysis (in which only those who completed therapy were analyzed), the respective response rates were 51% and 31%. By genotype, about 40% of those with genotypes 2 or 3 and about 25% of those with genotypes 1 or 4 achieved an undetectable viral load in the Peg-Intron group. The response rate in the Peg-Intron group (25%) was better than that seen in the standard interferon group (10%) for those with genotype 1 or 4, but the difference was not significant for those with genotypes 2 or 3.
In both treatment groups, participants with CD4 cell counts above 500 cells/mm3 had better response rates. The incidence of serious side effects was high (about 25% in the Peg-Intron group and 21% in the standard interferon group), and about one-third of participants discontinued treatment. The lower response rates in this study compared with ACTG 5071 may reflect a difference between Peg-Intron and Pegasys; the two brands of pegylated interferon have not yet been compared in head-to-head trials.
Pegylated interferon also offers promise for HIV/HCV-coinfected people who have not responded to previous HCV therapy. Also at ICAAC, Maribel Rodrigues-Torres, M.D., of the University of Puerto Rico in Rio Piedras and colleagues reported results of a study in which 75 coinfected people who had not responded to prior treatment with standard interferon monotherapy were given either Pegasys or Pegasys plus ribavirin. After 24 weeks, 27% of participants in both groups experienced at least a 2 log decrease in HCV RNA and 30% achieved an undetectable HCV viral load. These response rates are higher than those seen in other studies of previous nonresponders.
While some early research suggested that response to HCV therapy tended to be slower in coinfected people, recent studies using pegylated interferon plus ribavirin suggest that coinfected people are as likely as those with only HCV to achieve an early treatment response (although response is slower in people with HCV genotypes 1 or 4 compared with those who have genotypes 2 or 3, regardless of HIV status). In terms of treatment length, researchers from the European Benelux Study presented evidence at the May 2001 Digestive Disease Week conference showing that 18 months of HCV treatment -- rather than the typical 12 months -- may be more effective for coinfected people.
Most studies of HCV treatment have looked at virological response rates. But a few have focused on histological response (improved liver health, including decreased inflammation and fibrosis). In a study published in the February 15, 2002 issue of AIDS, Dr. di Martino and colleagues reported that while sustained virological response rates in coinfected people receiving standard interferon plus ribavirin were lower than those in people with HCV alone, histological response rates were similar in the two groups.
These results and others suggest that HCV treatment may improve liver health even if undetectable HCV viral loads are not achieved. According to Douglas Dieterich, M.D., of New York University School of Medicine, "even in so-called 'failures' interferon therapy decreases fibrosis, increases T-cell responsiveness to hepatitis C antigens, and decreases the rate of fatal hepatomas [liver cancer]. ... Lowering the [HCV] viral load can only help the ultimate outcome of the liver disease, and may permit [coinfected people] to take protease inhibitors that will certainly prolong life."
Side effects of HCV treatment are common in people with or without HIV coinfection. While many studies show that coinfected people are less able to tolerate HCV treatment and have higher discontinuation rates, others have found similar side effect and dropout rates. More research is needed in this area. Another issue requiring further study is whether pegylated interferon is associated with more or less frequent and severe adverse effects than standard interferon; here again, study results are conflicting. HIV/HCV-infected people should report all new or changed symptoms to their physicians. Healthcare providers, support groups, and HCV advocacy organizations can offer help in learning to manage and live with HCV treatment side effects.
Most experts recommend that HIV disease and OIs should be controlled first before starting HCV treatment.
In people with little immune system impairment but advanced hepatitis C, starting HCV treatment first may help them tolerate anti-HIV drugs later.
HCV treatment response rates tend to be lower in coinfected people than in those with HCV alone.
Pegylated interferon plus ribavirin is currently the most effective HCV treatment option.
Drug-related liver injury is often signaled by increased levels of liver enzymes, in particular ALT and AST. In fact, when combination antiretroviral therapy was first used many physicians noticed dramatic increases in their patients' liver enzyme levels and began testing them for hepatitis C -- thus revealing that HIV/HCV coinfection was more common than previously suspected. Mild-to-moderate hepatotoxicity is usually asymptomatic, but some people may experience nausea, fatigue, itching, or elevated bilirubin (bile pigment) levels leading to jaundice. Elevated liver enzyme levels are most common soon after starting a new drug and typically stabilize over time, but in some cases hepatotoxicity develops after a longer period on therapy.
Different studies have shown that antiretroviral regimens on the whole are associated with a two-fold to five-fold increased risk of hepatotoxicity. All classes of anti-HIV drugs have been linked with liver toxicity. Most studies show that the non-nucleoside reverse transcriptase inhibitor (NNRTI) nevirapine and the PI ritonavir (Norvir) at full dosage are the worst offenders; the small amounts of ritonavir sometimes added to regimens to "boost" blood levels of other PIs are less likely to be problematic. Other anti-HIV drugs associated with liver toxicity include the NRTIs AZT (zidovudine, Retrovir), ddI (didanosine, Videx), ddC (zalcitabine, Hivid), d4T (stavudine, Zerit), and abacavir (Ziagen); the NNRTIs delavirdine (Rescriptor) and efavirenz; and the PIs indinavir (Crixivan) and saquinavir (Fortovase). 3TC (lamivudine, Epivir) causes minimal liver toxicity. Studies have shown that severe hepatotoxicity related to the PIs nelfinavir (Viracept) and lopinavir/ritonavir (Kaletra) is uncommon, and recent research suggests that the new PI atazanavir (Zrivada) may have a similar low liver toxicity rate.
As with many aspects of HIV/HCV coinfection, researchers do not all agree that antiretroviral drugs lead to higher rates of liver toxicity in people with viral hepatitis. Some studies have shown low rates of hepatotoxicity even for drugs that have proven most problematic in other trials. For example, Dr. Dieterich and colleagues reported in the April 1, 2002 issue of the Journal of AIDS that they saw severe (grade 3 or 4) ALT or AST elevations in just 1.1% of their cohort of HIV positive men taking efavirenz, nevirapine, or delavirdine; none of these severe events occurred among the subset of 40 men coinfected with HCV. And in the May 1, 2002 issue of Clinical Infectious Diseases, Curtis Cooper, M.D., from the University of Ottawa and colleagues reported similar rates of liver toxicity in people taking ritonavir and those on ritonavir-sparing regimens.
Even mild liver enzyme elevations should be taken seriously, especially in people with chronic hepatitis. At the Barcelona AIDS conference Amy Justice, M.D., from the University of Pittsburgh and colleagues reported that in their population of 5,700 participants from two observational studies, elevations of ALT or AST were associated with a higher risk of death. Those with mild-to-moderate liver enzyme elevations had a 1.7-fold increased risk of death, while those with liver enzyme levels two or more times normal had a five-fold increased mortality rate. According to Dr. Justice, "Our study shows that even patients whose elevations are mild to moderate have a death rate that is nearly twice that of patients with mid-range normal levels. This association with increased mortality suggests that any elevation in ALT and AST should be addressed." [For Dr. Justice's perspective on liver health concerns specific to women, see "In Their Own Words: The Current State of Women and HIV" in this issue.]
In related research, Corinne Rancinan, M.D., and colleagues with the French Aquitaine study group reported in the July 5, 2002 issue of AIDS that HCV coinfection itself was not associated with a greater risk of death in people with HIV unless they had elevated liver enzyme levels.
NRTI drugs are known to cause bone marrow suppression leading to low blood cell counts, and HIV treatment regimens almost always include at least one of these drugs. Likewise, interferon -- especially pegylated interferon -- often causes neutropenia (low white blood cell count). Ribavirin causes hemolytic anemia in 20% or more of patients. Using certain anti-HIV drugs (in particular, AZT) and HCV therapies together can potentially lead to dose-limiting blood cell deficiencies.
Mitochondrial toxicity refers to drug-induced damage to small organelles within cells that are involved in energy production. It is associated with elevated levels of lactic acid in the blood (hyperlactatemia or lactic acidosis). Symptoms may include fatigue and muscle weakness. Mitochondrial toxicity is a known side effect of NRTI drugs, especially ddI and d4T. It is also a possible side effect of ribavirin, and the risk is compounded when people take ribavirin and ddI and/or d4T together. In the RIBAVIC study, for example, Dr. Perronne and colleagues detected evidence of mitochondrial toxicity in 22% of those taking ribavirin, ddI, and d4T; in 7% of those taking ribavirin and ddI but not d4T; in 1% of those taking ribavirin and d4T but not ddI; and in 1% of those taking ribavirin with neither ddI nor d4T.
This past September the FDA announced that the ddI label had been revised to include a warning about use of the drug with ribavirin. The warning states that ddI blood levels may increase when taken with ribavirin, and recommends that people taking both drugs should be monitored carefully and that ddI should be discontinued if signs of pancreatitis (inflammation of the pancreas), symptomatic hyperlactatemia, or lactic acidosis develop.
Interestingly, high blood fat levels -- one of the most worrisome side effects associated with anti-HIV drugs -- appear less likely to occur in HIV/HCV-coinfected individuals. For example, Jack Stapleton, M.D., and colleagues from the Veterans Administration Medical Center in Iowa City, Iowa, and colleagues presented data at the Barcelona AIDS conference showing that HIV/HCV-coinfected people taking HAART were significantly less likely to have high cholesterol levels -- both total cholesterol and low-density lipoprotein (LDL) "bad" cholesterol -- than those with HIV alone. The researchers suggested that the difference may be related to the way HCV binds to lipoproteins or due to altered liver metabolism of fats and glucose.
These results are not isolated. Julio Collazos, M.D., and colleagues from Vizcaya, Spain, presented data at the September 2002 ICAAC showing that HIV/HCV-coinfected people taking HAART had hyperlipidemia (high blood fat) rates similar to those of people not receiving HIV therapy. However, L. Casado and colleagues from Ovideo, Spain, reported at the same conference that they saw similar cholesterol and triglyceride levels in HAART-treated coinfected individuals and those with HIV alone; moreover, in their study lipodystrophy (body fat irregularities) was more common in those with both HCV and HIV.
To some extent, careful drug selection and dose modification can help prevent or reduce hepatotoxicity and intensified side effects. Since there are many more anti-HIV drugs than medications for HCV, there is more leeway for adjustment of HIV regimens.
Among the NRTIs, AZT is most often associated with low blood cell counts, and ddI and d4T are most often linked to mitochondrial toxicity. Many experts recommend that people being treated for HCV use NRTIs other than AZT, ddI, or d4T.
Among the PIs, ritonavir and saquinavir are most often associated with severe liver toxicity. Nelfinavir and the new PI atazanavir may be the best options for coinfected people.
Among the NNRTIs, nevirapine has often been linked to hepatotoxicity, so one of the other NNRTI drugs -- or a regimen that does not include an NNRTI -- may be better choices.
Despite these suggestions, it is important to construct an anti-HIV regimen that has adequate potency. If drugs associated with greater toxicity must be used, frequent and careful monitoring is essential.
While hepatotoxicity and other side effects should be minimized if possible, studies show that a majority of people with HIV/HCV coinfection do not experience serious adverse events, regardless of what drugs they use. Discussing his July 2002 JAMA article, Dr. Sulkowski stated that HAART was safe for people with HCV and that HIV/HCV-coinfected people do "just as well" on antiretroviral therapy as those with HIV alone.
Although some physicians are daunted by the many possible drug side effects and interactions -- and thus may be hesitant to treat coinfected people for either HCV or HIV -- such reluctance does not seem warranted. Again, many people with HIV and HCV can be successfully treated for both diseases. Effective control of HIV can slow HCV disease progression and lessen the risk of severe liver damage. And treatment for HCV can be beneficial even in the absence of a sustained virological response.
A variety of therapies can be used to help manage the side effects of anti-HIV and HCV drugs. For example, erythropoietin (Epogen or Procrit) may be used to stimulate red blood cell production and granulocyte colony-stimulating factor (Neupogen) can promote white blood cell proliferation. Antidepressants are often used to manage the psychological side effects of interferon therapy. In addition, there are many practical, supportive, and self-help measures that can help people cope with adverse effects, such as injecting interferon at bedtime to sleep through the worst symptoms and avoiding spicy or greasy foods to lessen nausea.
The key to successful therapy for people with HIV/HCV coinfection is careful monitoring. ALT, AST, and bilirubin levels should be measured regularly for signs of hepatotoxicity. Blood cell counts should be monitored for indications of neutropenia or anemia. Most side effects are worse when a new drug is first started and often improve over time. Therefore, monitoring is especially important when beginning a new medication. However, some side effects -- including hepatotoxicity and mitochondrial toxicity -- may develop over time. Monitoring should not stop just because a person is currently doing well.
Since management of HIV/HCV coinfection can be complex, the care of coinfected people ideally should be managed by physicians who have experience with both diseases or by teams that include both a hepatologist (liver disease specialist) and an infectious disease expert.
Hepatotoxicity and drug interactions can complicate HIV treatment in coinfected individuals.
Medication regimens often can be adjusted to minimize adverse events; do not adjust doses or change drugs without medical advice. Report all new or worsening symptoms to a healthcare provider.
Liz Highleyman (firstname.lastname@example.org) is a freelance medical writer and editor based in San Francisco.
This article was provided by San Francisco AIDS Foundation. It is a part of the publication Bulletin of Experimental Treatments for AIDS. Visit San Francisco AIDS Foundation's Web site to find out more about their activities, publications and services.