| |
Screening for and Treating Hepatitis B Virus in Patients with HIV Infection EDITORIAL
|
| |
| |
Clinical Infectious Diseases Sept 1, 2007
Hans L. Tillmann
Medical Department II, University of Leipzig, Leipzig, Germany
(See the article by Gaglio et al. on pages 618-23 and the article by Miailhes et al. on pages 624-32)
Why it is so important to be diligent with regard to screening for hepatitis B virus (HBV) in the context of HIV infection? Treating HBV infection, if necessary, makes a difference. When I started my professional life caring for HBV-infected patients in the context of liver transplantation, outside of clinical trials, liver transplantation in HBV-infected patients who had HBV early antigen (HBeAg) in serum or an HBV DNA level of about 280,000 copies/mL was prohibited because of its bad prognosis [1]. Fortunately for my patients at that time, the first antivirals that were effective against HBV infection, such as famciclovir, were being tested, leading to some improvement [2], followed by more potent substances, such as lamivudine and adefovir [3, 4]. These achievements not only increased the survival rate after liver transplantation, but also frequently circumvented the need for transplantation [5]. Thus, treating HBV infection can make a difference, but to treat it, you must be aware of it. However, does this hold true for patients with HIV coinfection?
Although hepatitis C virus (HCV) is mostly linked to injection drug use, HBV infection is present in the chronic form in almost 10% of HIV-infected patients in the Western world, irrespective of the risk group. Indicators of past infection have been more prevalent, and the rate of coinfection with HBV and HIV is higher in areas where the rate of chronic HBV infection is high, such as certain areas in Africa, where recently, a 50% prevalence of HBV infection was reported in Nigerian HIV-positive male patients, compared with 12.5% in HIV-negative male individuals [6]. The risk of developing a chronic HBV infection in cases of acute hepatitis B was reported to be 5-6-times higher in HIV-positive patients than in HIV-negative patients [7]; in addition, seroreversion has been seen in the context of HIV infection [8]. Data on hepatitis D virus, which is estimated to be present in about 5% of HBV-infected patients [9], are less well-documented, but this may be because of the limited treatment options [9], leading to a worse prognosis for patients coinfected with hepatitis D virus than for patients with HIV and HBV coinfection only, as reported previously [10].
In the public awareness, HCV infection seems to be more recognized than HBV infection. However, although HCV infection usually shows an accelerated course only in relation to immunodeficiency and failure to control HIV infection [11, 12], HBV infection seems to be a risk factor for mortality, even in the presence of a still functioning immune system [11]; mortality increases further in the case of immunodeficiency in the absence of effective antivirals [13]. Although treatment of HCV infection in HIV-infected patients is still limited, with only about 40% clearance rates of HCV elimination [14], HBV infection cannot be cleared but can be halted in most HIV-positive patients [15]; some patients might become negative for HBV surface antigen (HBsAg) [16]. In addition, it has been shown that treatment of HBV infection improves prognosis [17, 18], which is still questioned in the case of HCV infection [19].
In this issue of Clinical Infectious Diseases, 2 articles [16, 20] review recent data on HBV infection in HIV-infected patients. In 1 study [20], treatment centers were surveyed about diagnosis criteria, treatment practices, and recommendations for therapy using a questionnaire sent to 161 sites, of which approximately one-half responded. The other article focuses on the impact of HAART on HBsAg and HBeAg clearance [16].
The survey appears to be very promising, because it represents 52,000 HIV-infected patients. However, the survey was not performed on a patient basis but rather on a practice basis. Thus, I believe the validity of the survey should have been tested by randomly assessing individual patient data for at least a subgroup of sites. What would one expect to find if one surveyed well-trained physicians participating in a program such as the Terry Beirn Community Programs for Clinical Research on AIDS who aim at optimal treatment for their patients? By asking physicians how they would manage HBV infection in relation to HIV infection, you only can be certain that they report what they should do, not what they actually do in the "real world." However, it is reassuring that most physicians indicated they do what they should do; they screen all HIV-infected patients for HBV infection. However, whether they screen HBsAg-positive patients for hepatitis D virus is not mentioned but should be emphasized. Most doctors provide vaccination against HBV infection to patients who are not immune or infected. Many doctors indicated that the decision to treat HBV-HIV-coinfected patients was based on elevated liver enzyme levels, which was a criterium that was "state of the art" at the time of the study and is still recommended by American Association for the Study of Liver Disease practice guidelines [21]. Since the survey was performed, however, data have emerged suggesting that an elevated viral load alone may be an indication for treatment [22, 23], which will be reflected in such documents as the updated German guidelines on HBV infection [24]. Overall, the good concordance of the clinical management, according to the survey, with the guidelines is especially reassuring, because the survey was completed in December 2004 [20], prior to the publication of the European Association for the Study of the Liver consensus statement on HIV-HBV coinfection in May 2005 [25].
As mentioned in the study by Gaglio et al. [20], it is not surprising to see differences among the sites with regard to indication for liver biopsy and at what HBV DNA level to initiate HBV-specific therapy, given the inconsistencies with regard to these aspects among different guidelines. The threshold of the HBV DNA level to start HBV-specific therapy ranges from 104 copies/mL to 105 copies/mL (2000-20,000 IU/mL), because there is an increased risk associated with an HBV DNA level >104 copies/mL, compared with lower levels; the more pronounced increased risk is associated with an HBV DNA level >105 copies/mL [22, 23]. Currently, elevated transaminase levels and an HBV DNA level >105 copies/mL or >104 copies/mL persisting over 6 months are indications for antiviral therapy [21]. Thus, in contrast to a liver biopsy for patients with elevated transaminase levels and elevated HBV DNA levels, I would favor the biopsy for those with elevated HBV DNA levels but normal transaminase levels or, perhaps, for those with some level of viral replication to prove that antiviral therapy can be safely withheld in the absence of significant liver fibrosis.
However, there are 2 concerns. First, there are data that show that HBV DNA level testing is performed significantly less frequently in the "real world" [26] than is recommended by guidelines of the American Association for the Study of Liver Disease, European Association for the Study of the Liver, and the Asian Pacific Association for the Study of the Liver [21, 25, 27, 28], which all recommend determination of the HBV DNA level every 3-6 months, illustrating that what doctors do might not always be the same as what doctors know they should do. The point of low compliance with regular viral load monitoring was emphasized recently again by the demonstration of the higher frequency of HBV testing among HIV-negative than among HIV-positive patients with HBV infection, with the conclusion that assessment of HBV infection in patients coinfected with HIV needs to be improved [29]. This example, perhaps, also argues that these patients should be referred to a liver disease specialist, such as a gastroenterologist.
The fact that a dual treatment strategy with lamivudine plus tenofovir was administered at 88% of the sites to patients who had not yet had an indication to receive HAART is also worrisome. This treatment would be antiretroviral therapy, instead of HAART, for which I would see no indication in the HAART era. Even though doctors might administer lopinavir plus ritonavir "monotherapy" [30] to spare the adverse effects of nucleoside reverse transcriptase inhibitors [31], there are no data supporting dual "nucs" without a nonnucleoside reverse transcriptase inhibitor or protease inhibitor.
It is not surprising that no clear rule for when to discontinue anti-HBV therapy emerged from this survey, because there was no clear rule at that time, whereas doctors are aware of the risk of "flares" in transaminase levels after HBV-specific therapy is discontinued, which can occur even when treatment is stopped after HBsAg seroconversion only [32]. In HIV-negative, HBeAg-positive patients, therapy can be discontinued when seroconversion to antibody to HBeAg has been achieved for 6-12 months, whereas in HIV-negative, HBeAg-negative patients, discontinuation is recommended after HBsAg clearance only. However, continued monitoring is requiring for all patients [21, 27]. The same approach might be applied for HIV-positive patients [21], even though data are less clear and recommendations are not uniform [25].
Thus, does HBeAg and/or HBsAg clearance occur, and does it occur in HIV-infected patients with HBV coinfection? Among immunocompetent patients, the annual HBsAg clearance rate was about 0.5% (19 [1.9%] of 998 patients over a mean duration [±SD] of 4 ± 2.3 years) among symptomatic Asian patients, 0.8% (35 [2.2%] of 1598 patients over a mean duration [±SD] of 2.7 ± 1.4 years) among asymptomatic Asian patients [33], and 0.7% among 317 asymptomatic Canadian patients [34]. Similar data were reported among children (HBsAg clearance rate, 1.16%; 15 [16.9%] of 89 children over a mean duration [±SD] of 14.5 ± 6.1 years) [35]. A report from Brazil showed slightly higher clearance rates of about 9.3% over a mean duration (±SD) of 5.5 ± 4.5 years (annual clearance rate, 1.7%), with no difference in relation to HIV infection (7 of 79 HIV-negative patients and 2 of 17 HIV-positive patients) [36]. Interestingly, a recent study reported an increase in the rate of HBsAg clearance correlated with an increase in age, with annual rates of clearance of 0.77% in patients aged <30 years and 1.83% in patients aged >50 years [37].
In this issue of Clinical Infectious Diseases, Miailhes et al. [16] describe their experience with 92 HIV-coinfected patients. Of these patients, only 10 did not receive antiretroviral therapy; thus, their aim to describe the natural history of coinfection was limited. Ten (16.9%) of 59 HBeAg-positive patients cleared HBeAg and 5 (6.1%) of 82 cleared HBsAg during treatment. These results are similar to those of another French study that found a slightly higher rate of clearance of HBeAg or HBsAg during bitherapy than during monotherapy [29] but are somewhat, although not significantly, lower than those of a recent study involving HIV-infected patients from Spain in which clearance of HBeAg occurred in 10 (28%) of 36 HBeAg-positive patients and clearance of HBsAg occurred in 10 (13.3%) of 75 treated patients [38].
The study by Miailhes et al. [16] reveals additional interesting data worth noting. These researchers found a greater increase in CD4 cell count in HIV-positive patients displaying an HBV response. Thus, is the greater increase in CD4 cell count a result of the HBV response, or is the restoration of the immune system required for good control of HBV infection? Considering that no difference in response to HAART in relation to HBsAg positivity was seen [39], I would favor the latter. More interesting is the statement by Miailhes et al. [16] that a shorter course of antiretroviral therapy was associated with a higher rate of HBeAg and HBsAg seroconversion. This finding was probably because of the censoring of the patients once an HBV response was achieved. Thus, this might be interpreted as patients not achieving response early after initiation of HAART being unlikely to do so in the long term.
It is also surprising that the rate of seroconversion was higher in patients with a low nadir CD4 cell count and in patients with AIDS than in those in earlier stages of HIV disease. An explanation might be that these patients may have experienced controlled HBV infection earlier in life, and HBV infection might have reactivated when HIV disease progressed and the patient was in a more immunocompromised state. It would be interesting, in the future, to study the HBV T cell function of such patients to see whether they might still have an immune memory against HBV infection. Certainly, one would not recommend waiting until the CD4 cell count decreased too low, because HBV infection usually takes an accelerated course in the presence of immunodeficiency [11, 13]. The findings of Miailhes et al. [16] of a role of both increase in CD4 cell count and control of HIV replication is in accordance with a Spanish study [38]. This seems to be different in the case of HCV infection, in the context of which the nadir CD4 cell count has been shown to be the most important factor for controlling the disease, supporting the different nature of these 2 hepatitis virus infections. However, it also supports the fundamental role of the immune system in controlling HBV infection and shows that response to antivirals is dependent on not only the antiviral's potency, but also the patient's immunocompetence.
Another interesting aspect that has not been examined in earlier studies involving either HIV-infected or HIV-uninfected patients is the baseline viral load threshold associated with absence of viral resistance. None of the patients with a baseline viral load up to 5.5 log copies/mL (300,000 copies/mL; about 60,000 IU/mL) developed resistance. It would be interesting to see if this holds true for more patients, with or without HIV infection, receiving treatment with different antivirals, because it might indicate that some patients might not need regular follow-up measurements of HBV DNA levels.
Finally, preventing HBV infection is better than treating it. Thus, every HIV-infected patient should be screened for HBV infection and offered vaccination in the instance of negative results for HBV markers [21].
Financial support. Federal Ministry for Education and Research (01KG0507 and 01KI0435).
Potential conflicts of interest. H.L.T. has received honoraria from and has participated on the advisory boards for Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, Novarits/Idenix, and Roche; has received research funding from Roche; and has served as a consultant for Bristol-Myers Squibb, Novartis/Idenix, and GlaxoSmithKline.
|
|
| |
| |
|
|
|
