November 27, 2006
Table of Contents
Chronic infection with hepatitis B virus (HBV) or hepatitis C virus (HCV) constitutes two of the leading causes of chronic hepatitis and cirrhosis worldwide.1-3 Because HIV, HBV and HCV share common transmission routes involving blood -- for example, via injection drug use (IDU), sexual contact, or from mother to child during pregnancy or birth -- individuals at high risk for HIV infection are also likely to be at risk for HBV or HCV coinfection.
Overall, approximately 25% of patients in the United States with HIV are also infected with HCV.4 In contrast, the prevalence of HBV in HIV-infected individuals has been estimated at only 9% by one European study.5
The complexity of treating a patient with HIV is compounded by coinfection with HBV or HCV. Now that people with HIV are living longer due to advancements in HIV therapy, aggressive management of HBV and/or HCV coinfection is critical to avoid progression to end-stage liver disease (ESLD), which has emerged as a leading cause of death in HIV-infected patients in the era of highly active antiretroviral therapy (HAART).6
To understand why HIV/HCV coinfection is so challenging to treat, it is important to keep in mind that the rate of HCV disease progression is accelerated by HIV. HCV progression is, of course, also influenced by a range of other factors, including the duration of infection, older age at the time of infection, alcohol use and concurrent HBV infection.3,7
Cirrhosis remains a considerable risk for HIV-infected patients who are coinfected with chronic HCV. Whereas only about 20% of HIV-uninfected patients with chronic HCV infection develop complicated cirrhosis,8 HIV/HCV-coinfected patients have a two- and six-fold greater risk of developing cirrhosis and ESLD, respectively, according to a meta-analysis conducted by Graham and colleagues.9 In some studies, mainly from Europe, coinfection with HCV has been associated with impaired CD4+ cell count recovery and with a higher rate of HIV-related deaths than among those not coinfected with HCV.10-12 However, recent reports pointed to a significantly improved prognosis as far as liver disease is concerned in the last 10 years.13-15
Because the risk of HCV causing serious liver damage and liver failure is greater in people with HIV, it's critical to:
The latter is usually determined by liver biopsy. Surrogate markers have been developed, but none are able to correctly identify the different stages of fibrosis.17 Limited application of some of these tests will be able to decrease the need for liver biopsy in the future.
There are relatively few data on the long-term prognosis of established cirrhosis in HIV-infected patients with HCV and/or HBV coinfection. One Italian study examined five-year outcomes in a cohort of 384 HIV-uninfected patients with compensated HCV-related, biopsy-confirmed cirrhosis.18 Patients were not infected with HBV or hepatitis A virus (HAV). The risk of decompensation was 18%, and 7% of patients developed hepatocellular carcinoma at five years. After decompensation, the actuarial survival probabilities were 80% and 60% at one and three years, respectively. In comparison, 153 HIV/HCV-coinfected patients with decompensated liver disease were followed prospectively by Merchante.19 A total of 62% of the patients died during the follow-up period; the median survival time was 13 months. Significant predictors of mortality, according to multivariate analysis, included a high Child-Pugh score (> 7), a CD4+ cell count below 100 cells/mm3 at the time of decompensation, and the presence of hepatic encephalopathy at the time of hepatic decompensation. The cumulative probability of survival in patients receiving antiretroviral therapy (ART) was 60% and 40% at one and three years, respectively, compared with 38% and 18%, respectively, in patients who were not receiving ART (hazard ratio [HR] = 0.5; 95% confidence interval [CI], 0.3-0.9; P < .0001). Thus, available data suggest that HIV/HCV-coinfected patients with cirrhosis are more likely to experience hepatic decompensation than HCV-infected patients without HIV.
The role of HCV treatment in patients with HIV and ESLD remains controversial. HCV treatment is contraindicated in patients with HIV who have advanced liver disease due to the possibility of developing clinical decompensation, including death. This complication has been observed in about 10% of patients in the APRICOT (AIDS Pegasys Ribavirin International Co-infection Trial) study20 and in 2% of RIBAVIC (i.e., ANRS HC02) study patients21 -- both studies examined HCV treatment in HIV/HCV-coinfected individuals.
Whereas compensated cirrhotic patients with HIV should be strongly considered for HCV treatment, patients with decompensated disease are not candidates for interferon-based therapy and should be referred for orthotopic liver transplantation (OLT). As such, treatment of decompensated cases should be reserved for clinical centers that perform OLT.
One of the challenges of providing ART to HIV-infected patients with advanced liver disease is that these patients often have a diminished ability to metabolize drugs due to hepatic and sometimes renal insufficiency. Plasma concentrations of non-nucleoside reverse transcriptase inhibitors (NNRTIs) and protease inhibitors (PIs) may be increased as a result of synthetic liver dysfunction, because these antiretroviral classes are extensively metabolized by cytochrome P450 enzymes. With the exception of abacavir (ABC, Ziagen), nucleoside reverse transcriptase inhibitors (NRTIs) are not extensively metabolized by the liver and are not expected to be negatively affected by hepatic insufficiency.22 Some nucleoside agents, such as didanosine (ddI, Videx) and stavudine (d4T, Zerit), may lead to hepatic mitochondrial toxicity and should be used with caution in the setting of advanced liver disease.20 Ribavirin (Copegus, Rebetol, Virazole) increases didanosine levels in vitro. Since didanosine is also associated with hepatic decompensation in cirrhotics, it is my practice to avoid didanosine in coinfected patients who are to start combination therapy.
Dosing adjustments in patients with hepatic insufficiency are recommended for both atazanavir (ATV, Reyataz) and amprenavir (APV, Agenerase). Zidovudine/lamivudine (AZT/3TC, Combivir) and zidovudine/lamivudine/abacavir (AZT/3TC/ABC, Trizivir) are not recommended in cirrhotic patients.
Only about 30% of HIV/HCV-coinfected patients with genotype 1 or 4 are estimated to experience successful HCV treatment. The number doubles to 60% with HIV-infected patients with genotypes 2 or 3.23 Unfortunately, HCV genotype 1 is the most common subtype found in the United States, representing approximately 75% to 80% of all HCV infections.24 However, HIV-infected patients with HCV genotype 1 and an HCV viremia of less than 800,000 IU/mL, have a 60% chance of sustained virologic response.20 In addition, an ongoing international, multisite, prospective study known as the Paradigm study compares U.S. Food and Drug Administration (FDA)-approved ribavirin 800 mg daily to ribavirin 1,000 or 1,200 mg once daily. This study will establish whether administering a higher, weight-adjusted ribavirin dose will improve sustained virologic response with reasonable safety.
The decision whether to treat the HIV or HCV first depends on many factors. The ideal time to start HCV therapy in relation to starting HIV antiretrovirals is still a subject of debate. A liver biopsy should be performed to aid in decision making. The need for and type of treatment should be chosen based on the extent of liver damage seen in both the liver biopsy and baseline assays. Treatment should be provided to HIV/HCV-coinfected patients when the fibrosis score is F2 to F4, as the risk for developing serious liver disease outweighs the risk of morbidity due to the adverse effects of therapy. Treating patients earlier is considered on a case-by-case basis. Before HCV treatment is initiated, the following tests should be performed: baseline liver chemistry, complete blood counts, thyroid-stimulating hormone, glucose, quantitative HCV RNA in serum, HCV genotype, and alpha-fetoprotein (only if cirrhotic). For females, it is important to perform a pregnancy test, because of the potential teratogenic effects of HCV treatment.
Patient response to HCV therapy appears to be best at higher CD4+ cell counts, and sustained clearance of HCV may be better achieved with good immune status. However, in the APRICOT study, CD4+ cell counts or percentages were not clearly associated with chance of sustained virologic response.23
The goal of HCV treatment is to achieve a sustained virologic response, which is defined as the suppression of HCV RNA to undetectable levels six months post-treatment. In the APRICOT study, the largest pivotal study on HIV/HCV-coinfected patients treated with pegylated interferon-based regimens versus standard interferon-based regimens, peginterferon alfa-2a (Pegasys) 180 µg/week plus ribavirin 400 mg twice daily has been shown to result in the following sustained virologic response rates: 29% for HCV genotype 1 and 62% for HCV genotypes 2 and 3. In HIV-infected patients coinfected with HCV genotype 1 with a low baseline HCV RNA load (< 800,000 IU/mL),23 a sustained virologic response was observed in 61%. Based on data from monoinfected patients, a sustained virologic response is likely to result in a cure in the vast majority of patients.
In the end, therapy initiation involves weighing the risks of an expected CD4+ cell count decline while on pegylated interferon, with the benefits of improving liver disease and tolerance to HIV medications if HCV is treated first.
Results from the largest HIV/HCV-coinfection trials are summarized in the following table.21,23,25,26 Since the studies mentioned in the table were carried out in vastly different populations and in different research units, it is impossible to attempt to compare across trials, particularly about the relative effectiveness of the two major interferons.
|Table 1. Treatment of HCV Infection in Patients With HIV|
|289||15||520 + 277||Peg-IFN
|800||Genotype 1: 48
Genotype 2/3: 48
|800||Genotype 1: 48
Genotype 2/3: 48
Genotype 4: 48
2/3: 24 or
* Mean + standard deviation, or median (range).|
** Intent-to-treat analysis.
AEs = adverse events; IFN = interferon; SVR = sustained virologic response.
HBV is associated with markedly increased rates of hepatocellular carcinoma (100-fold higher compared with that of non-HBV-infected individuals) and liver mortality.27 HIV coinfection further compounds the risk of hepatic decompensation and mortality.13-15 The ongoing prospective study of liver-related mortality among homosexual men who are at risk for HIV and HBV infections, known as the Multicenter AIDS Cohort Study (MACS), reported that liver-attributable mortality rates per 1,000 person-years of observation were 1.7 in HIV-infected individuals, 0.8 in individuals with chronic HBV infection, and 14.2 (P < .001) in HIV/HBV-coinfected patients.13,28 A study from Taiwan noted a 10-fold increase in liver decompensation in HIV/HBV-coinfected compared to HIV-monoinfected patients.29 These findings emphasize the fact that HIV coinfection worsens liver disease in patients with HBV. The relationship between these two viral infections and the risk of progression to liver disease were strongest in the setting of low nadir CD4+ cell counts.
On the other hand, no study has yet definitively shown that HBV coinfection influences the rate of HIV disease progression. CD4+ cell count recovery appears to be similar with or without HBV coinfection. In a European study, all-cause mortality and liver mortality (as expected) was higher in HIV/HBV-coinfected patients compared with HIV-monoinfected patients, but no difference was found in AIDS-related mortality or clinical parameters between the two groups of patients. Both the numbers of patients able to achieve CD4+ cell count recovery and a viral load below 400 copies/mL were equivalent in HIV/HBV-coinfected and HIV-monoinfected patients after multiple adjustments. Interestingly, an increase in CD4+ cell count by 50% from baseline reduced liver mortality by 40% in HIV/HBV-coinfected patients.5 HIV/HBV-coinfected patients did have several baseline factors that differed significantly from that of HIV-monoinfected patients, including a higher rate of HCV coinfection, a longer duration of HIV infection, lower CD4+ cell counts and higher viral loads. Another study has shown that failure to suppress viral load, but not CD4+ cell count recovery, was more common in HIV/HBV-coinfected patients compared with HIV-monoinfected patients.29
Another possible complication for HBV-infected patients is coinfection with hepatitis delta, or D, virus (HDV), which only infects chronic HBV carriers. Patients with HBV and HDV generally have worse liver outcomes than those with only HBV.30 In addition, there is a trend for more virologic failure in HBV/HDV-coinfected patients compared with HIV/HBV-coinfected patients lacking HDV infection. Only a very few studies have investigated HIV/HBV/HDV coinfection in patients. One study assessing HBV/HCV/HDV multiple infection in HIV-infected individuals found that HDV appears to be the dominant virus, suppressing both HBV and HCV replication.30
In HIV/HBV-coinfected patients, HAART is usually well tolerated and several antiretroviral combinations suppress both HIV and HBV.31 However, patients should be monitored carefully during the first few weeks after HIV treatment initiation; liver enzyme elevations are estimated to occur in 5% to 10% of patients with only HIV infection, while patients coinfected with HBV demonstrate higher rates.32 Practitioners should be particularly cautious with coinfected patients beginning nevirapine (NVP, Viramune), efavirenz (EFV, Sustiva, Stocrin), or full-dose ritonavir (RTV, Norvir).
Measurement of serum HBV DNA, hepatitis B surface antigen (HBsAg), hepatitis B e antigen (HBeAg), alanine aminotransferase (ALT) levels, and liver histology are used to assess HBV-related liver damage and monitor the response to HBV treatment. HDV needs to be ruled out with antibody screening tests (IgM and IgG). In general, individuals with elevated ALT levels, HBeAg positivity, an HBV DNA count greater than 100,000 copies/mL, and/or the presence of necrosis and inflammation according to liver histology are candidates for HBV therapy.33 Patients with HBeAg-negative chronic HBV need to be considered earlier (DNA > 4 log copies or 2,000 IU/mL). Drugs currently used to treat coinfected patients include pegylated interferon alfa; the nucleoside analogs lamivudine (3TC, Epivir-HBV), emtricitabine (FTC, Emtriva), entecavir (Baraclude), and telbivudine (Tyzeka); and the nucleotide analogs adefovir dipivoxil (Hepsera) and tenofovir (TDF, Viread) (see table).
|Table 2. Treatment Regimens for Hepatitis B|
|Adefovir dipivoxil (Hepsera)**||10 mg daily: treatment duration unknown*||
|Emtricitabine (FTC, Emtriva)**||200 mg daily: treatment duration unknown*||
|Entecavir (Baraclude)||0.5-1.0 mg daily: treatment duration unknown*||
|Lamivudine (3TC, Epivir)**||150 mg twice daily or 300 mg daily (dosage as part of ART regimen) for 1 year or more*||
|Peginterferon alfa-2a (Pegasys)||180 µg per week for 4-6 months*||
|Tenofovir (TDF, Viread)**||300 mg daily: treatment duration unknown*||
|Telbivudine (Tyzeka)||600 mg daily||
* The duration and expected efficacy of treatment vary according to the treatment strategy and the individual patient characteristics.|
** Agents are active against both HIV and HBV.
Fourteen percent to 32% of patients develop resistance to lamivudine within a year (66% after four years) when it is used as a single agent without other HBV treatment.34 Emtricitabine is also associated with mutations that occur in the same HBV polymerase domains as with lamivudine. Resistance that develops in the presence of adefovir occurs much less frequently than with lamivudine, yet such resistance can still lead to virologic failure. Adefovir was originally developed for HIV treatment, but the doses required to suppress HIV frequently cause kidney toxicity. At lower doses, however, adefovir has been shown to be effective against HBV. The rate of resistance to adefovir increases to 18% at year 4 and to more than 28% at year 5 in non-HIV-infected patients.33 Because of the high probability of resistance development, HBV monotherapy with these agents is not recommended in HIV/HBV-coinfected patients.
Tenofovir has emerged as an extremely effective treatment for controlling HBV in HIV-infected patients, particularly those with lamivudine-resistant HBV.35 The first randomized study to compare tenofovir to adefovir in the treatment of lamivudine-resistant HBV in coinfected patients was AIDS Clinical Trials Group (ACTG) 5127.36 This trial demonstrated that tenofovir was noninferior to adefovir, in fact, it was better, with HBV DNA being reduced by 4.4 log10 IU/mL in the tenofovir group and by 3.2 log10 IU/mL in the adefovir group. In another coinfected study with 80% of patients lamivudine-experienced, tenofovir reduced HBV DNA by 4.56 log10 IU/mL in 52 HBeAg-positive patients, with 30% having undetectable levels (< 200 IU/mL), and by 2.53 log10 IU/mL in 13 HBeAg-negative patients, with 82% having undetectable levels.37
Entecavir is the sixth FDA-approved medication for the treatment of HBeAg-positive, HBeAg-negative, and lamivudine-resistant HBV infection. It has no activity against the HIV reverse transcriptase.38,39 Additionally, entecavir has no effect on CYP450 and does not interfere with HAART. In a randomized multicenter trial in lamivudine-experienced, HIV/HBV-coinfected patients conducted by Pessoa et al, the addition of entecavir 1.0 mg to lamivudine produced a 3.66 log10 IU/mL decrease in HBV DNA at week 24 compared with no change in patients who had placebo added to lamivudine.40 There is a lower threshold for entecavir resistance in patients with pre-existing lamivudine resistance. In the Pessoa study, mutations associated with entecavir resistance were found in two of 48 patients (4%), but no phenotypic evidence of resistance was observed at 48 weeks.41
Goal of therapy:
However, coinfected patients need HIV meds, so they are kept on medications long term, beyond endpoints. If the endpoints are not met, a reasonable endpoint is to keep HBV DNA below 2,000 IU/mL (or 10,000 copies/mL).
Treatment guidelines from the HIV-HBV International Panel recommend that all HIV-infected patients receive an assessment of HBV status prior to the initiation of ART.33 For HIV/HBV-coinfected patients already on an antiretroviral regimen that contains lamivudine, a reasonable approach is to add tenofovir 300 mg/day to the regimen. In HIV/HBV-coinfected patients who have not started ART and who have detectable HBV DNA, institution of ART containing tenofovir plus lamivudine or emtricitabine is the recommended approach to provide activity against both HBV and HIV. For those in whom antiretroviral therapy for HIV can be deferred, the use of entecavir 0.5 mg/day is a reasonable choice.
Diagnosis of Hepatic Fibrosis and Cirrhosis by Transient Elastography in HIV/HCV-Coinfected Patients
As mentioned previously, the development of liver fibrosis in patients with HIV/HCV coinfection is typically assessed by liver biopsy. Efforts have been made to evaluate the stage of liver fibrosis using alternative, noninvasive methods. One such method is the measurement of liver stiffness using transient elastography. Elastometry involves the application of ultrasonic waves that traverse the liver parenchyma and return to the measuring device with a speed that appears to be proportional to the stiffness of the liver, which is associated with fibrosis.
This prospective study assessed the accuracy of liver stiffness measurement for the detection of fibrosis and cirrhosis in 72 HIV/HCV-coinfected patients.42 The investigators evaluated HIV/HCV-coinfected patients who had undergone adequate liver biopsy and subsequently took at least five liver stiffness measurements, expressed in kilopascal (kPa). They found that liver stiffness significantly correlated with the fibrosis stage (Kendall [tau]-b = 0.48; P < .0001). Exact fibrosis staging was not possible in all cases, primarily due to a significant overlap in stiffness for patients with fibrosis stages F1-F3. In these cases, the correlation between liver stiffness and fibrosis was modest (R = 0.48). A liver stiffness cutoff below 11.8 kPa correctly identified 90% of patients with stage F4, or cirrhotic, disease. However, above 11.8 kPa, the positive predictive value for F4 disease was only 81%. A higher cutoff of 14.8 kPa correctly identified 95% of patients with stage F4 disease and was predicted to decrease the number of false positives from four to two (out of 17 total), but missed two of 17 histological cirrhosis cases. Also of note, transient elastography was significantly better at diagnosing cirrhosis than several non-invasive biochemical markers sometimes used for this purpose, such as the platelet count, the aspartate aminotransferase (AST):ALT ratio, and the AST:platelet ratio index (APRI).
Theoretically, a cutoff that predicts stage F4 disease with 95% certainty is an interesting tool, as it would obviate the need for liver biopsies in cirrhotics -- patients who tend to have poor coagulation parameters. Unfortunately, the prediction of fibrosis at lower stages is still imprecise with transient elastography. In addition, it must be remembered that a liver biopsy cannot be replaced in some cases, for example, when histological assessment of liver iron content or a liver steatosis score is needed. More studies are needed to assess the place of this noninvasive diagnostic tool in the management of patients with HCV, with or without HIV coinfection.
Treatment of HCV in HIV-Infected Patients: A Meta-Analysis
Since HIV/HCV-coinfected individuals are now living longer thanks to the introduction of HAART, long-term issues such as durable HCV control and the safety of HCV treatments are becoming increasingly important for optimal patient management. Toward this end, Kim and colleagues undertook a study to better delineate therapeutic improvements and safety concerns associated with various combinations of pegylated interferon, interferon and ribavirin based on data obtained from prospective, randomized, controlled trials.43 The investigators collected information from all trials published in paper or abstract form between January 2001 and August 2005 that compared the use of pegylated interferon + ribavirin versus pegylated interferon monotherapy or interferon + ribavirin.
Six randomized, controlled trials were identified that randomly assigned a total of 1,756 HIV/HCV-coinfected patients to various forms of HCV treatment. Patients treated with pegylated interferon + ribavirin were more likely to achieve a sustained virologic response compared with patients treated with standard interferon + ribavirin (odds ratio [OR]: 3.00; 95% CI: 2.27-3.96), and this held true for patients with HCV genotypes 1 or 4 (OR: 4.40; 95% CI: 2.75-7.03) and genotypes 2 or 3 (OR: 2.56; 95% CI: 1.71-3.85). Higher sustained virologic response rates were also achieved with pegylated interferon + ribavirin versus pegylated interferon monotherapy (OR: 2.60; 95% CI: 1.84-3.67), making pegylated interferon + ribavirin the overall winner in terms of efficacy.
The study did not provide comparative data on pegylated interferon monotherapy versus interferon + ribavirin. These data would have been useful to know since pegylated interferon monotherapy is used in special patient populations, such as patients with kidney failure who did not tolerate ribavirin.
The treatment withdrawal rates were very similar between patients treated with pegylated interferon + ribavirin and those treated with interferon + ribavirin, primarily due to a similar incidence of severe side effects and nonresponse. This meta-analysis is also useful in that it identified the percentage of patients who developed significant adverse events. In the pegylated interferon + ribavirin cohorts, 1% to 1.5% of patients died, 1% to 2% developed hepatic decompensation and 4% developed mitochondriopathy.
Sustained Virologic Response in HIV/HCV-Coinfected Patients With HCV Genotype 1 Infection Who Have a Rapid Virologic Response at Week 4 of Treatment With Peginterferon Alfa-2a Plus Ribavirin
Early virologic response at week 12 is often used to predict the chance of achieving a sustained virologic response. Earlier viral load assessment at week 4 (rapid virologic response, defined as HCV RNA negative at week 4) of treatment may also have predictive value. To assess this hypothesis, Dieterich and colleagues retrospectively reviewed data from the APRICOT trial in which HIV/HCV-coinfected adults with compensated liver disease and stable HIV disease were randomized to receive pegylated interferon + ribavirin, pegylated interferon monotherapy, or interferon + ribavirin.44 Week 4 response data were obtained for 172 individuals with HCV genotype 1 who received treatment with pegylated interferon + ribavirin. Of these patients, 22 (13%) achieved a rapid virologic response, the vast majority of whom (18/22; 82%) had a low baseline HCV RNA level (< 800,000 IU/mL). A total of 82% of patients who achieved a rapid virologic response went on to attain a sustained virologic response, and the sustained virologic response rate was comparable for patients with low and high baseline HCV RNA levels (83% and 75%, respectively). Relapse was very uncommon among individuals with a rapid virologic response. In HCV genotype 1 patients who failed to achieve a rapid virologic response, those with a low baseline HCV RNA load still had a 46% chance of achieving a sustained virologic response, compared with only a 16% chance for individuals with a high baseline HCV RNA level.
As stated by the authors, assessing for a rapid virologic response at week 4 may be used as a "motivator" to encourage patients to adhere to treatment, especially those individuals who are experiencing substantial treatment-related side effects. What would make the 4-week data worthwhile is determining whether patients who achieve a rapid virologic response can be treated for shorter periods without sacrificing sustained virologic response. Although testing for a rapid virologic response may be worthwhile for patients with a low baseline HCV RNA load (< 800,000 IU/mL), the use of the 4-week data in the high baseline HCV RNA group seems pointless since these individuals have only a 3% chance (4/126) of having a rapid virologic response.
ALT Increase After Initiation of PI-Containing ART in HIV/HCV-Coinfected Patients Is Not Related to High PI Plasma Concentrations
Assessing the cause of liver injury after HAART initiation is complex. ALT increases have been reported to occur in 5% to 12% of HIV/HCV-coinfected patients after the initiation of ART, but the pathophysiological mechanisms underlying these events are not well understood.
Ritonavir has been associated with high rates of abnormal liver tests, but not at the doses currently used to boost PIs.45,46 Nevirapine also has been associated with significant liver injury, but close enzyme monitoring can prevent severe outcomes.47 In some instances, higher drug levels (e.g., with nevirapine) have been associated with higher liver enzymes. In a number of cases, transaminase elevations may not be secondary to drug hepatotoxicity, but may instead be associated with viral diseases (HAV, HBV or HCV), alcohol use, biliary disease, over-the-counter analgesics and complementary medicines, immune reconstitution and steatosis, among many other factors.48
To better understand the mechanisms involved in ALT elevations, Gervais et al compared 27 HIV/HCV-coinfected patients with ALT levels five-fold above the baseline level at four years after the initiation of PI-based ART (or four-fold above normal, if the initial ALT value was normal) with 78 age- and sex-matched controls with no ALT change. The investigators found no differences between cases and controls with regard to excessive alcohol consumption, the use of medications, and hepatitis G virus or HBV coinfection. Moreover, no differences were found between cases and controls with regard to the type of PI included in the regimen when ALT values peaked, nor with regard to variations in the therapeutic ranges for each PI. More cases than controls had a below-normal (< 19 kg/m2) or above-normal (> 25 kg/m2) body mass index (44% versus 31%). Also more cases were infected with HCV genotype 3 (30% versus 13%) and abused drugs (81% versus 60%; P = .04). Despite this, none of the above factors were found to correlate with increased ALT levels according to multivariate analysis, in part because too few events were studied.49
In contrast to prior speculation, ALT elevations do not appear to be associated with high PI concentrations in HIV/HCV-coinfected patients, at least according to this study. Multiple mechanisms that might contribute to increases in ALT levels were postulated -- immune reconstitution, varied PI use, toxicity due to other medications, metabolic changes and drug abuse.
Keys to minimizing elevated ALT and AST levels before the start of HAART include:
Severe Liver Disease Associated With Prolonged Exposure to Antiretroviral Drugs
Although liver damage in HIV-infected individuals typically stems from HBV and/or HCV coinfection, alcohol abuse, medications and other factors, the etiology of liver disease remains unknown for some individuals. The presence of persistently elevated aminotransferase levels in the absence of common causes of liver disease is referred to as cryptogenic liver disease (CLD).
Maida and colleagues retrospectively reviewed the medical records of all HIV-infected individuals seen at two large HIV clinics in Spain in 2004 to identify those with CLD. Out of 3,200 patients included in the analysis, CLD was diagnosed in 17 (0.5%) patients, of whom more than half (9/17; 52.9%) developed signs and symptoms indicative of portal hypertension. In addition, five of 17 patients were either over- or underweight -- conditions associated with nonalcoholic fatty liver disease. The mean time from HIV diagnosis to the development of CLD was greater than 15 years, and all patients had been exposed to ART, particularly nevirapine, stavudine and didanosine. None had liver abnormalities prior to the initiation of ART. To assess which factors were associated with CLD, HIV-infected controls without CLD were matched to cases according to age, gender and CD4+ cell count status. Prolonged didanosine exposure emerged as the only independent predictor of CLD development in this population.50
The lack of biopsies in the majority of patients included in this study is a drawback. The multivariable analysis is difficult to evaluate since the number of cases was small. Additionally, increased ART duration was associated with protection against CLD, but the use of didanosine was associated with CLD, an apparent contradiction.
A subset of patients who stopped didanosine experienced normalization of liver enzyme levels, but the data presented were thin. Taken together, with information from other studies, these findings suggest that didanosine or the combination of didanosine + stavudine should be avoided in HIV-infected patients due to the potential risk of acute or chronic hepatic complications.
Progression and Severity of Liver Fibrosis in HIV/HCV-Coinfected Patients With Normal CD4+ Cell Count: Comparison With HCV-Infected Patients
Retrospective analyses have suggested that the course of HCV infection is accelerated in the presence of HIV coinfection, leading to faster HCV disease progression to cirrhosis, along with more severe fibrosis. Roulot and colleagues performed a prospective study to verify this observation by comparing liver fibrosis scores (based on liver biopsy) and HCV replication levels in 33 immunocompetent HIV/HCV-coinfected patients with those from 33 matched HCV-monoinfected patients.51 Most patients (70%) were on suppressive ART, all were nonalcoholic, all had a known date of HCV infection, all had detectable HCV RNA levels, all had a CD4+ cell count above 250 cells/mm3 and all were naive to HCV treatment.
Coinfected and monoinfected patients had a comparable median age at the time of the biopsy (44 + 10 versus 42 + 5 years; P > .05) and median duration of HCV infection (20.0 + 6.7 versus 21.5 + 4.0 years; P > .05). Interestingly, all liver fibrosis analyses and HCV replication measurements were similar between the two groups of patients (see table). According to multivariate analysis, only the histological activity score significantly correlated with more severe fibrosis.
|Table 3. Liver and HCV Outcomes in HIV/HCV-Coinfected Versus HCV-Monoinfected Patients|
Patients (n = 33)
Patients (n = 33)
|Histologic activity > A2-A3, %||39||36|
|Fibrosis score > F2, %||48||52|
|Rate of fibrosis progression||0.085||0.084|
|HCV RNA, log10 IU/mL
6.00 + 0.77
3.83 + 1.00
5.77 + 0.57
3.75 + 0.75
These data add to the large body of evidence showing that patients with well-controlled HIV disease have an HCV prognosis similar to that observed in non-HIV-infected patients with HCV, and that higher CD4+ cell counts are associated with decreased mortality from liver disease in both HIV/HCV- and HIV/HBV-coinfected patients. The corollary is that clinicians should now expect to find lower stages of fibrosis in biopsies taken from HIV/HCV-coinfected patients. For individuals with early fibrosis (F0, F1), treatment of HCV should take a back seat to intensive HIV therapy.
Critical Role of Ribavirin for the Achievement of Early Virologic Response to HCV Therapy in HCV/HIV-Coinfected Patients
HCV/HIV-coinfected individuals typically demonstrate a poorer response to HCV treatment than HCV-monoinfected individuals. In the standard HCV treatment with pegylated interferon + ribavirin, ribavirin is given at a fixed, FDA-approved dose of 800 mg/day. Ramos et al hypothesized that the use of weight-based ribavirin dosing (1,000 mg/day if under 75 kg, 1,200 mg/day if more than 75 kg) might improve the response to HCV therapy in coinfected patients.52
To test their hypothesis, the investigators assessed the rapid virologic response rate and the early virologic response rate (defined as an HCV RNA decline of more than 2 log10 IU/mL at week 12) among individuals enrolled in the PRESCO study, a multicenter, ongoing trial administering pegylated interferon plus weight-based ribavirin to HIV/HCV-coinfected patients. These data were compared with historical data from the APRICOT study, in which HIV/HCV-coinfected patients received pegylated interferon plus fixed-dose ribavirin (800 mg/day), as well as with data from the Fried et al study, in which HCV-monoinfected patients received pegylated interferon plus weight-based ribavirin (1,000-1,200 mg/day).
The results revealed that for HIV/HCV-coinfected individuals, PRESCO patients receiving weight-based ribavirin achieved higher rapid virologic response and early virologic response rates than APRICOT patients receiving fixed-dose ribavirin (see table). Moreover, the rates of rapid virologic response and early virologic response for PRESCO patients were comparable to those achieved by HCV-monoinfected patients who were also receiving weight-based ribavirin. Both higher ribavirin plasma levels (P = .007) and infection with HCV genotype 2/3 (P = .001) were independent predictors of a rapid virologic response and an early virologic response.
|HCV genotype 1||N = 176||N = 298||N = 94|
|Negative HCV-RNA week 4||13%||31.2%||33.8%|
|HCV-RNA drop > 2 log week 12||63%||80.5%||77.7%|
|HCV genotypes 2 & 3||N = 95||N = 140||N = 70|
|Negative HCV-RNA week 4||37%||84.2%||85.4%|
|HCV-RNA drop > 2 log week 12||88%||97.1%||94.3%|
Although it makes sense that higher ribavirin doses lead to better response rates, a caveat is in order. Patients coinfected with HIV may be more likely to develop symptomatic anemia with higher ribavirin doses. The final sustained virologic response data from PRESCO have indicated that weight-based ribavirin use results in a sustained virologic response that appears superior to the response seen in a historical pivotal study using fixed-dose ribavirin. The large, multinational, prospective Paradigm study, is currently enrolling HIV/HCV-coinfected patients to more precisely assess the effect of higher ribavirin dosing based on weight. To do this, a fixed-dose comparator arm (800 mg/day) will be used -- a much better option than comparing current data with those from historical cohorts.
Risk Factors for Bacterial Infections in HIV/HCV-Coinfected Patients During Interferon Plus Ribavirin-Based Therapy
Complications secondary to ESLD account for a significant proportion of deaths in HIV/HCV-coinfected patients. Bacterial infections constitute one such complication, even though they are, in part, preventable. To identify the risk factors associated with bacterial infections in HIV/HCV-coinfected patients, Bani-Sadr et al reviewed the medical records of 383 individuals receiving pegylated interferon + ribavirin in the RIBAVIC trial.53 A total of 18 cases (5%) of bacterial infection were identified in 17 patients. Among the most common types of infection, three were urinary infections, three were upper respiratory tract infections, two involved the skin, and two involved sepsis. A median time of 20 weeks (range: two to 40 weeks) lapsed between the beginning of HCV therapy and the onset of infection.
The rate of bacterial infections was found to be higher for HIV/HCV-coinfected patients when compared with historical data from HCV-monoinfected patients (54 versus 24 per 1,000 patient-years). Four factors were found to be independently and significantly associated with bacterial infections according to multivariate analysis:
Baseline or on-therapy neutropenia was not found to be a risk factor for the development of bacterial infections. Although it is desirable to keep the absolute neutrophil count above 1,000 cells/mm3, the absolute number is not a good predictor of bacterial infections. Perhaps the duration of neutropenia is a more significant predictor for such events. The majority of infections occur after week 2 when the nadir of bone marrow suppression is expected.
Liver Transplantation of HIV/HCV- and HIV/HBV-Coinfected Patients: A Large Experience in a Single Center
Although HCV- or HBV-infected individuals with HIV coinfection are not typically considered to be ideal candidates for liver transplantation, this procedure can be highly successful in these patients. The investigators from one treatment center described the five-year outcomes following liver transplantation for ESLD in 41 coinfected patients (HIV/HCV: n = 36; HIV/HBV: n = 5, with two also coinfected with HDV).54 All patients were on suppressive HAART, and liver biopsies were systematically performed following liver transplantation to track liver disease status.
After a median follow-up of 18 months (range: two to 70 months), seven patients, all with HCV, had died at two, three, four, 11, 22 and 24 months after liver transplantation. The cause of death was severe HCV recurrence and mitochondrial toxicity in three, cerebral hemorrhage in one, acute pancreatitis in one, sepsis in one and pancreatic carcinoma in one. The two-year survival rates were 70% and 100% for HIV/HCV- and HIV/HBV-coinfected patients, respectively.
Two of 13 HIV/HCV-coinfected patients treated with pegylated interferon + ribavirin achieved a sustained virologic response. No patients experienced HBV or HDV reinfection following liver transplantation due to the administration of antivirals and hepatitis B immunoglobulin maintenance therapy. HIV RNA remained undetectable in all but one patient who experienced a transient viral rebound in conjunction with a moderate esophageal Candida infection.
These findings support the notion that patients with HIV infection and cirrhosis should be evaluated for liver transplantation. HIV/HBV- and HIV/HBV/HDV-coinfected patients demonstrate better long-term transplant outcomes than HIV/HCV-coinfected patients. Experiences in other centers also suggest that patients with HBV, cryptogenic or alcoholic liver disease have a better prognosis after liver transplantation compared with HIV/HCV-coinfected patients. Typical cutoffs used for referral of cirrhotic patients for liver transplantation involve a Child-Pugh-Turcotte score of 7 or greater, or a Model for End-Stage Liver Disease (MELD) score greater than 10.
Impact of HCV Coinfection on Response to HAART and Outcome in HIV-Infected Individuals: A Nationwide Cohort Study
HCV coinfection has been associated with impaired immunologic improvement after the initiation of HAART. However, adverse outcomes in terms of AIDS-defining illnesses have not been uniformly demonstrated. In particular, U.S. cohorts have shown similar HAART outcomes in HIV/HCV-coinfected patients compared with HIV-monoinfected patients. Weis and colleagues sought to determine the impact of HCV infection on the response to HAART among Danish patients with HIV-1 infection. All Danish adults with HIV-1 infection who started HAART from January 1995 to January 2004 were prospectively followed. A total of 443 patients (16%) were HCV infected, whereas 2,183 (80%) were HCV uninfected and 108 (4%) had an unknown HCV status.10
Overall mortality was significantly higher for HIV/HCV-coinfected patients versus HIV-monoinfected patients (mortality rate risk: 2.4; 95% CI: 1.9-3.0), as was liver disease-related mortality (mortality rate risk: 16; 95% CI: 7.2-33). HIV/HCV-coinfected patients also had a higher risk of dying from an AIDS-related disease if they had a prothrombin time below 0.3, or if they had a history of alcohol abuse. No differences in HIV RNA levels were observed between HCV-infected and HCV-uninfected groups, but the HCV-infected group did have a marginally lower absolute CD4+ cell count following HAART initiation.
This Danish study points to a lower CD4+ cell count improvement and significantly worse AIDS-related mortality following the initiation of HAART in HIV/HCV-coinfected versus HIV-monoinfected individuals. Consequently, consensus on the impact of HCV coinfection on the response to HAART is, as of yet, unsettled.
While therapy of HCV in HIV-infected patients is now fairly well established, the issue of weight-based ribavirin still demands an answer. Avoidance of liver biopsy, using non-invasive testing, may become a reality in the next few years. Some patients may still need liver sampling to measure steatosis or iron stores, for example. Drug-induced liver injury, its presentation and clinical associations are being studied in a large national cohort (the Drug-Induced Liver Injury Network). Answers from this study are likely to help clinicians grappling with this issue in HIV-infected patients. In patients with HIV and cirrhosis, liver transplantation is a definite option in carefully selected patients.
Please note: Knowledge about HIV changes rapidly. Note the date of this article, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this article.