| |
Coinfection with HIV and HCV in Injection Drug Users and Minority Populations
|
| |
| |
Clinical Infectious Diseases, July 1, 2005
...Two large studies involving patients coinfected with HIV and HCV demonstrated a 15%-25% prevalence of cirrhosis within 10-15 years of HCV infection.... hepatocellular carcinoma developed at a mean age of 42 years in one study of patients coinfected with HIV and HCV, whereas it occurred at a mean age of 69 years among the control population...(note from Jules Levin: several recently published studies suggest that when HIV is well or fully suppressed with good CD4 counts HCV disease progression may not be any more accelerated than in HCV monoinfection, but I think its preliminary to believe these study results are widely applicable and bear in mind a significant proportion of coinfected patients do not have fully suppressed HIV)...
Doris B. Strader
Department of Medicine, Division of Gastroenterology/Hepatology, Fletcher Allen Health Care, University of Vermont College of Medicine, Burlington
ABSTRACT
Coinfection with human immunodeficiency virus (HIV) and hepatitis C virus (HCV) is common. In the United States, it has been estimated that 25% of persons infected with HIV are also infected with HCV. The prevalence of coinfection with HIV and HCV is highest among those infected via percutaneous routes. In fact, in urban areas in the United States, 50%-90% of persons infected with HIV via injection drug use are coinfected with HCV.
In addition, limited data from drug treatment centers in these urban areas suggest that the prevalence of coinfection with HIV and HCV may be highest among African Americans and Hispanics.
Little information is available with regard to the epidemiology of coinfection with HIV and HCV among injection drug users (IDUs) or minority populations. Likewise, although there is a growing body of data on the potential complexities of treating HCV among IDUs and the poor response to current anti-HCV treatment among African Americans, few data address the therapy of coinfection with HIV and HCV among IDUs and minority populations.
INTRODUCTION
Hepatitis C virus (HCV) infection and HIV infection are major public health problems worldwide. In the United States, the estimated prevalence of serum HCV antibody is 1.8%, with 2.7 million persons having HCV RNA detected in blood, which is indicative of ongoing infection [1]. The prevalence of HIV is <1%, and the virus infects an estimated 800,000 persons [2]. Because of the shared routes of transmission, many persons with HIV infection are also infected with HCV [3]. Before the introduction of HAART in 1996, most persons with HIV infection died of HIV-related opportunistic infections. Since the widespread use of HAART began, the morbidity and mortality associated with HIV infection have decreased, and patients coinfected with HIV and HCV are now at increased risk of HCV-related hepatic disease [4-6]. The majority of patients coinfected with HIV and HCV are injection drug users (IDUs), and emerging data suggest that, in the United States, these patients tend to be poor, urban, and African American or Hispanic [3]. Below is a summary of the epidemiology, natural history, and response to therapy among persons coinfected with HIV and HCV in the United States, with focus on currently available data regarding this infection in IDUs and minority populations.
EPIDEMIOLOGY OF COINFECTION WITH HIV AND HCV
Coinfection with HIV and HCV is common. The US Adult AIDS Clinical Trial Group reported that 16.1% of 213 persons with HIV also had serum antibody to HCV [3]. In the Johns Hopkins HIV Clinical Cohort, the prevalence of HCV coinfection among HIV-positive patients was 44.6% [7]. Of interest, the prevalence of coinfection with HIV and HCV differs markedly depending on HIV risk group. Although both HIV and HCV are efficiently transmitted via percutaneous exposure, HCV is 〜10-fold more easily transmitted via small-volume percutaneous exposure [8, 9]. The estimated transmission risk of HCV after accidental needlestick is 1.5%-3.0%, compared with 0.3% for HIV [8]. By contrast, HIV is more efficiently transmitted via sexual or mother-to-infant routes [10-13]. As a consequence, accumulating data suggest that more than one-half of the persons infected with HIV via percutaneous exposures, including injection drug use and transfusion of clotting factors, are also infected with HCV. Several studies in urban areas of the United States indicate that the prevalence of coinfection with HIV and HCV among HIV-positive IDUs is 50%-90%, compared with <20% among those who were infected with HIV via sexual exposure [5, 14]. In some western European countries, the prevalence of coinfection with HIV and HCV among IDUs is as high as 70% and 90% [15, 16]. Greub et al. [17] reported rates of coinfection with HIV and HCV of 87.7%, 6%, and 3.7% among a cohort of Swiss IDU, heterosexual, and homosexual subjects, respectively. Likewise, a community-based survey among Chinese IDUs in Sichuan Province documented an 11.3% rate of coinfection with HIV and HCV, with HCV infection found in 100% of HIV-positive IDUs [18]. By contrast, persons infected with HIV via sexual routes are much less likely to be coinfected with HCV. In one study, 13% of female sex partners of hemophiliac men coinfected with HIV and HCV were found to be HIV positive, whereas only 3% were HCV positive [11]. Likewise, data show that homosexual and heterosexual HIV-positive persons who are not IDUs are much less likely to be infected with HCV [3, 7, 19].
No published data currently exist on the rate of coinfection with HIV and HCV among the 2 largest minority groups in the United States, African Americans and Hispanics. However, given epidemiological data on HCV, HIV, and injection drug use in the United States, it is clear that the burden of coinfection with HIV and HCV among minority IDUs is high. Data from the National Health and Nutrition Examination Survey III indicate that the prevalence of HCV infection is 3.2% among African Americans, 2.1% among Mexican Americans, and 2.9% among those giving their race/ethnicity as "other" [1]. An epidemiological study of HIV, hepatitis B virus (HBV), HCV, and tuberculosis among IDUs found that African Americans account for the highest incidence of HIV and AIDS both in the general population and among IDUs [20]. Of all cases reported to the Centers for Disease Control and Prevention through June 2000, one-third of AIDS cases and 46% of cases of HIV infection occurred in African American adolescent boys and men. Among IDUs, African American men accounted for 60% of cases of HIV infection. African American women accounted for 57% of AIDS cases and 68% of cases of HIV infection, and, among IDUs, 60% of cases of HIV infection occurred among African American women [20]. Because of the heterogeneity among Hispanic populations, exact numbers are somewhat difficult to estimate. Nevertheless, rates of HIV infection in Hispanic populations appear to be increasing. Between 1988 and 1999, cases of AIDS among Hispanics related to injection drug use increased by 127% [20]. The prevalence of HIV infection due to injection drug use was 35% among Hispanic men born in the United States, 27% among men born in the Dominican Republic, 61% among men born in Puerto Rico, and 6.2% among men born in Mexico [20]. These data, together with the above-mentioned 50%-90% rate of coinfection with HIV and HCV among HIV-positive IDUs, suggests that one-third to one-half of IDUs coinfected with HIV and HCV in the United States may be African American or Hispanic.
NATURAL HISTORY OF COINFECTION WITH HIV AND HCV
Current data show that HIV infection adversely affects the natural history of HCV infection in several ways. First, spontaneous recovery from HCV infection occurs in only 5%-10% of persons coinfected with HIV and HCV, compared with 15%-35% of those monoinfected with HCV [21-24]. Second, coinfection with HIV and HCV is associated with more-rapid progression of cirrhosis, liver failure, and hepatocellular carcinoma [14, 25-28]. A case control study of patients coinfected with HIV and HCV via injection drug use revealed an increase in liver fibrosis, compared with matched HCV-infected patients without HIV infection [29]. A retrospective study of hemophiliac patients coinfected with HIV and HCV revealed a median time from HCV infection to clinical cirrhosis of only 16.5 years, compared with 20-30 years in patients monoinfected with HCV [30]. Likewise, Eyster et al. [31] reported an acceleration of decompensated liver disease among coinfected hemophiliac patients. Two large studies involving patients coinfected with HIV and HCV demonstrated a 15%-25% prevalence of cirrhosis within 10-15 years of HCV infection [28, 32]. Finally, hepatocellular carcinoma developed at a mean age of 42 years in one study of patients coinfected with HIV and HCV, whereas it occurred at a mean age of 69 years among the control population [33].
It should be noted that, although most studies show an increase in progression of HCV-related liver disease among patients coinfected with HIV, the effect of HCV on the progression of HIV infection appears to be minimal to modest. In the majority of instances, the effect of HCV on HIV disease is related to the influence of HCV on the tolerability and response to HAART [17, 34, 35]. In one study of >3000 HIV-infected persons receiving HAART, HCV infection was associated with a decrease in the magnitude of the CD4+ cell response and a modest increase in the risk of developing an AIDS-defining illness [17]. By contrast, another study of patients in an HIV clinic in Baltimore did not find accelerated progression to AIDS after adjustment for HAART [7]. Likewise, a study of 298 patients coinfected with HIV and HCV was unable to identify any significant hepatotoxicity associated with HAART [34]. Although controversy exists, at present, it is thought that the contribution of HCV to accelerating the progression of HIV is negligible.
HEPATOTOXICITY OF HAART IN PATIENTS COINFECTED WITH HIV AND HCV
Of somewhat greater concern is the potential hepatotoxicity of HAART. It has been suggested that HAART increases HCV RNA levels and that certain antiretrovirals can increase necroinflammatory activity and progression of HCV disease. Several studies show an increase in HCV RNA level of 0.2-0.48 log10 IU/mL in patients coinfected with HIV and HCV who were treated with HAART [36-38]. Conversely, a retrospective analysis of 65 coinfected patients receiving HAART did not show any significant changes in HCV RNA or aminotransferase levels [39]. In 2 separate studies of either HBV- or HCV-coinfected patients, Sulkowski et al. [40] noted severe elevations in transaminase levels among 15.6% of patients receiving nevirapine, 8% of patients receiving efavirenz, and 10.4% of patients receiving [full dose] ritonavir. A large retrospective study of HIV-infected patients receiving HAART reported a 6.3% overall incidence of elevations in aminotransferase levels directly associated with higher baseline aminotransferase levels, chronic infection with HBV or HCV, and nevirapine- or ritonavir-based therapy [41]. Recently, severe liver toxicity was noted in 4% of patients coinfected with HIV and HCV after 12 months of therapy with the new combination drug Kaletra (lopinavir-ritonavir; Abbott Laboratories) [42]. In addition, studies implicating the nucleoside reverse-transcriptase inhibitors (NRTIs) didanosine, stavudine, and zidovudine in hepatotoxicity have been published [43-45]. (note from Jules Levin: a recent study from Sulkowski shows the rate of abnormally elevated liver enzymes were the same for various boosted PI regimens including Kaletra & nelfinavir, and the use of 100mg of ritonavir to boost a PI did not cause additional liver enzyme elevations).
The mechanism by which HAART-related hepatotoxicity occurs with increased frequency in HBV- or HCV-infected persons is unclear. It has been hypothesized that idiosyncratic immune-mediated adverse events may be observed with all NRTIs, the protease inhibitor amprenavir, and the nucleoside analogue abacavir [46, 47]. By contrast, others have suggested that intrinsic toxic events occur with all protease inhibitors, some NRTIs, and nevirapine [48]. These events include inhibition of human mitochondrial DNA polymerase gamma (NRTIs), increased levels of hepatic lipid synthesis (ritonavir), and induction of cytokine-mediated hepatic inflammation [46-50].
Despite these potential hepatotoxic events, overwhelming data show the substantial survival benefit of HAART. In its 1999 US Public Health Service/Infectious Diseases Association of America guidelines on the prevention of opportunistic infection in persons with HIV, the Centers for Disease Control and Prevention does not recommend withholding HAART from persons coinfected with HIV and HCV because of concerns about hepatotoxicity [51]. Recent studies reveal minimal hepatotoxicity in appropriately monitored patients. In one study of patients coinfected with HIV and HCV treated with HAART, severe hepatotoxicity was noted in only 12% of patients, and no patient had significant or irreversible hepatotoxicity [34]. A recent retrospective analysis of patients coinfected with HIV and HCV (81% with hemophilia) identified HAART and any antiretroviral therapy as independent predictors of liver-related survival [52]. Although severe elevations in aminotransferase levels did occur in patients undergoing HAART, no deaths were reported. These data suggest that HAART is associated with hepatotoxicity, but, overall, the survival benefit afforded by the therapy outweighs the potential untoward effects of liver-related injury in patients coinfected with HIV and HCV.
TREATMENT OF HCV INFECTION AMONG DRUG USERS AND MINORITY POPULATIONS COINFECTED WITH HIV AND HCV
By far, the most imprecise issue for physicians who treat IDUs coinfected with HIV and HCV is the appropriate treatment of HCV infection. There are no medications approved by the US Food and Drug Administration for the treatment of HCV infection in coinfected patients (note from Jules Levin: just recently, after submission of this article, Pegasys was approved for HCV treatment in HIV/HCV coinfected patients). In fact, the initial trials of treatment for HCV infection specifically excluded patients with concomitant HIV infection and patients who were actively using injection drugs from participating. As a result, most available data on the treatment of HCV in HIV-coinfected patients are based on practices used to treat HCV-monoinfected persons.
(note from Jules LevIn; obviously this article did not include the latest & most well known studies of treatment with peginterferon plus ribavirin in coinfected patients. The APRICOT Study treated patients with Pegasys plus 800 mg ribavirin/ daily and 40% overall of the coinfected patients achieved a Sustained Viral Response; PegIntron was used to treat coinfected patients along with ribavirin in the RIBAVIC Study where the SVR was 27%. The study authors report the study population in RIBAVIC was a more difficult to treat population than in APRICOT. The third important coinfection study was ACTG 5071, where 27% achieved an SVR. The import of these study results were limited by several issues of study design: ribavirin dosing was dose escalated over the early part of the study weeks, and it is well accepted that full dosing and adherence of ribavirin in early stages after starting therapy is key to achieveing a Sustained Viral Response, and there was a high percent of African-Americans & patients with cirrhosis in this study; it is well known the African-Americans have a lower response to therapy; cirrhosis can reduce response rates. Patients with genotype 2 or 3 have significantly higher Sustained Viral Response rates compared with patients with genotype 1. Adherence study in HCV monoinfection showed that SVR rates increased by up to 50% with full adherence. When patients have insulin resistance or diabetes or are overweight or have fatty liver, these factors can reduce SVR rates.).
Initial trials of IFN monotherapy for the treatment of patients coinfected with HIV and HCV who had CD4+ cell counts of >500 cells/mm3 revealed overall response rates similar to those of patients without HIV infection [53, 54]. Similar to data for HCV-monoinfected patients, the use of combination therapy with IFN and ribavirin improved response rates in coinfected persons. Several small cohort studies of 〜20 coinfected patients treated with IFN plus ribavirin observed rates of sustained virological response of 18%-40% [55-57]. In a larger study of 110 patients coinfected with HIV and HCV, HCV RNA was undetectable at 12 weeks in 23% of patients receiving combination therapy, compared with 5% of patients receiving IFN alone [58]. Data about sustained virological response rates are not yet available. Recently, reports on the use of therapy with pegylated IFN plus ribavirin to treat coinfected patients have been published. In a randomized, open-label study of 133 coinfected patients, Chung et al. [59] reported a virological response rate of 44% at week 24 among patients receiving combination therapy with pegylated IFN plus ribavirin, compared with 15% among patients receiving standard IFN plus ribavirin. Rates of grade 4 toxicity were greater among patients receiving pegylated IFN, but the rates of discontinuation of treatment were not statistically different. By contrast, other reports suggest that the efficacy of pegylated IFN in patients coinfected with HIV and HCV is limited by its poor tolerability. Nine of 23 coinfected patients in one study discontinued treatment because of severe adverse events (anemia, pancreatitis, and depression), and a sustained virological response was noted in only 3 patients [55]. In another report, a sustained virological response was noted in 35% of coinfected patients, but drug was discontinued because of adverse events in 15% [56]. A recently published treatment trial of IFN plus full-dose ribavirin versus IFN plus ribavirin delayed for 16 weeks observed that 55 of 107 coinfected patients discontinued therapy prematurely. Twenty-three percent of these patients experienced adverse events, including anemia, depression, anxiety, and fatigue [60]. In addition, dose reductions were necessary for 22% of patients because of neutropenia, neuropsychiatric symptoms, or fatigue. Patients receiving zidovudine had more ribavirin-induced anemia. Previous data have suggested that ribavirin-related hemolysis and IFN-induced suppression of hematopoiesis may be worse in HIV-coinfected persons [61, 62]. In addition, the drug-drug interaction between ribavirin and didanosine, which leads to increased activity of this NRTI and increased mitochondrial toxicity, has led to great concern about the use of ribavirin among patients taking zidovudine or didanosine [63-66]. Although it is clear that acceptable rates of sustained virological response to anti-HCV therapy may be achieved in HIV-coinfected patients, these data suggest that the risk of potential adverse events, particularly among patients undergoing HAART who are receiving ribavirin therapy, requires close monitoring by experienced physicians.
Unfortunately, few data are available regarding therapy for HCV among HIV-coinfected active IDUs. However, several studies of treatment of HCV among patients enrolled in substance abuse programs have been published. In a study from Munich, 34 of 50 heroin injectors entering a detoxification program received IFN monotherapy, and 16 received combination therapy for HCV infection [67]. The overall rate of sustained virological response was 36%, despite the fact that 80% of patients relapsed to injection drug use during the study [67]. The strongest predictor of response was adherence to weekly clinic appointments, with 45% of those who reported for two-thirds of their appointments achieving a sustained virological response, compared with only 8% of those who failed to keep at least two-thirds of clinic appointments [67]. In another study in which patients were selected for anti-HCV therapy on the basis of demonstrated motivation, as evidenced by participation in pretreatment education sessions, 29% of 66 patients undergoing maintenance therapy with methadone achieved a sustained virological response [68]. The response rate was not significantly associated with the level of abstinence from illicit drugs during therapy. Concerns about HCV reinfection among active IDUs being considered for therapy have been used as a justification for denying therapy. However, limited available data suggest that reinfection is rare in active IDUs if efforts to minimize risk (i.e., avoiding shared needles) are observed [67, 69]. In the above-mentioned study conducted in Munich, 2 of 12 active IDUs counseled about safe injection practices had evidence of reinfection during the follow-up period [67]. Both were infected with the same pretreatment genotype. Similar data were reported in a 5-year follow-up study of 27 IDUs who achieved a sustained virological response [69]. In that study, 9 patients (33%) relapsed to drug use; only 1 became reinfected with HCV [69]. It should be noted that few patients in those 2 studies were coinfected with HIV. It is unclear whether HIV coinfection increases the risk of reinfection with HCV among patients who were actively injecting illicit drugs.
To my knowledge, no data exist regarding anti-HCV therapy and subsequent response to treatment among ethnic minority IDUs coinfected with HIV. Available data suggest that there may be ethnic differences in sustained virological response among patients with HCV monoinfection. The largest multicenter trial of anti-HCV therapy in which differences in response by race could be measured revealed an overall rate of sustained response of 27% among whites, 11% among African Americans, 16 among Hispanics, and 44% among Asians receiving standard IFN plus ribavirin therapy [70]. When a subset of African Americans and whites infected with genotype 1 were compared, racial differences in response rate disappeared (23% vs. 22%) [70]. However, the small number of minority patients participating in the study (<3% in each minority group) cast some doubt on these findings. Several subsequent small studies comparing response rates between African Americans and whites have been performed. Contradictory results have been obtained, with some confirming the above findings and others maintaining that racial differences in response to anti-HCV therapy with standard IFN persist despite correction for genotype [71-74]. A recent prospective, multicenter, open-label trial evaluated treatment for 48 weeks with pegylated IFN plus ribavirin therapy among 78 African American and 28 white patients infected with HCV genotype 1. The sustained virological response among African Americans was 26%, compared with 39% among whites [75]. Severe neutropenia necessitating dose modification occurred more frequently among African Americans, but treatment discontinuation was less frequent than it was among whites [75]. Currently, a large, multicenter, 4-year trial sponsored by the National Institutes of Health is comparing the response to pegylated IFN plus ribavirin among 200 African American and 200 white patients infected with HCV genotype 1. Although these patients are not coinfected with HIV and are unlikely to be active IDUs, the data obtained may help guide therapy in the future.
FUTURE RESEARCH
Coinfection with HIV and HCV is widespread, and injection drug use is the most frequently associated risk factor. In the United States, coinfection with HIV and HCV appears to be concentrated in urban areas, institutions, and prisons, often disproportionately affecting ethnic minority populations. In addition, many of these persons are ongoing IDUs. Current data regarding the management of coinfection with HIV and HCV in IDUs and minority populations is sparse. No drugs are approved by the US Food and Drug Administration for the treatment of coinfection with HIV and HCV, and most treatment protocols exclude patients with ongoing injection drug use. Of critical importance is an accurate estimate of the numbers of IDUs infected with HIV and HCV. The current methods used to collect data on the prevalence of HIV and HCV may underestimate the true numbers of persons coinfected with HIV and HCV, because the homeless, institutionalized, or imprisoned populations are not surveyed. Research to further understand the viral interactions between HIV and HCV and the effect of ongoing illicit injection drug use on these interactions is essential. Current data do not suggest that ongoing illicit drug use necessarily limits response to anti-HCV therapy among patients without concurrent HIV infection. Trials should be done to evaluate whether current combination therapy in IDUs coinfected with HIV and HCV is effective, particularly among African Americans, among whom current therapy in HCV-monoinfected patients appears to be suboptimal. In addition, treatment trials directed specifically at defining the optimal approach to treating IDUs, with emphasis on the timing of anti-HCV therapy (before, during, or after therapy for substance abuse), whether abstinence is essential for response to therapy, and the management of relapse to substance abuse, are needed. Finally, effective strategies are needed to identify patients who are at high risk of initiating injection drug use and to prevent this activity, to educate ongoing IDUs about the avoidance of practices that lead to infection with HIV and HCV, and to assist patients with sufficient motivation in discontinuing injection drug use.
|
|
| |
| |
|
|
|
