icon- folder.gif   Conference Reports for NATAP  
  16th CROI
Conference on Retroviruses and Opportunistic Infections Montreal, Canada
February 8-11, 2009
Back grey_arrow_rt.gif
Summary from CROI for Hepatitis Coinfection - What are the challenges and how can we improve outcome of Hepatitis coinfection in HIV-Infected Patients?
  Jurgen K. Rockstroh M.D., Professor of Medicine
University of Bonn, Germany
Hepatitis coinfection has evolved over the last conferences as an increasingly recognized research area underlining the clinical challenges which come along with this comorbidity. Indeed several important studies related to hepatitis B and C coinfection were presented at the 16th Conference on Retroviruses and Opportunistic Infections in Montreal. This includes studies on therapy of hepatitis B as well as cohort analysis addressing the outcome of hepatitis C in HIV/HCV-coinfected individuals as well as treatment outcome in HIV/HCV-coinfected individuals.
Treatment of hepatitis B
Several trials addressed the issues of ongoing HBV replication under HBV-active therapy with tenofovir for treatment of hepatitis B in HIV-coinfection (1,2). Despite the lack of not having been able to identify detectable resistance under HBV treatment with tenofovir so far, clearly the fear remains that over time HBV-relevant resistance development in patients with ongoing HBV replication under therapy may emerge thereby possibly dramatically limiting further treatment options for hepatitis B. Katherine Lacombe from Paris presented work on HBV blippers, rebounders and non-responders under treatment with tenofovir in HIV/HBV-coinfection (1). Within this study HIV/HBV-coinfected patients within a large prospective French cohort study, were included if they had received tenofovir for at least six months. Non-responders were defined as patients with an HBV-DNA viral load persistently > 2,000 IU/ml. Blippers were patients displaying viral loads rebounds > 2,000 IU/ml but then suppressing again with HBV virus load < 2,000 IU/ml and rebounders were patients who once displayed HBV viral load < 2,000 IU/ml followed by an HBV viral load consistently > 2,000 IU/ml. 308 HIV/HBV coinfected patients were included into the cohort study, of those 180 initiated TDF treatment. 168 patients were eligible with more than six months of treatment with tenofovir, from which 165 patients had at least two available viral loads more than six months apart for analysis. Table 1 summarizes the baseline characteristics of the study population upon tenofovir initiation.
Table 1:


Mean duration of tenofovir therapy was 31.5 months (IQR 17.6), median time to HBV-DNA < 2,000 IU/ml was 8.1 months (IQR 7.6). Ten patients stopped tenofovir after a mean delay of 13.5 months (7.6-28.6). Of the 165 patients with at least two available viral loads more than six months apart, 148 controlled HBV continuously < 2,000 IU/ml. Eight patients were identified as blippers, 6 as rebounders and 3 were true non-responders. In blippers, rebounders and non-responders, plasma tenofovir levels were determined. Interestingly, none of the non-responders revealed tenofovir exposure suggesting compliance problems as the reason for persistent viremia. No tenofovir exposure was also detected in 4 of 8 blippers as well as in 2 of 6 rebounders. All 4 blippers with measurable TDF levels were subsequently sequenced. No specific HBV mutation was found to be associated with HBV blips but the L217R mutation considered to be an HBV genotype A2 polymorphism was found in 3 of 4 patients. Two of these 3 patients did not harbor the corresponding mutation at baseline before tenofovir initiation. The 4th patient developed an R274W polymorphism which not has been associated with HBV nucleos(t)ide therapy so far. In both rebounders the L217R mutation was already present at baseline before tenofovir initiation and was also detectable upon rebound. In one of the patients upon rebound an additional S219A mutation was found, again a mutation which has not been described in HBV infected patients with anti-HBV failing regimens before. The L217R mutation has been previously described in HIV/HBV-coinfected patients failing under adefovir therapy but achieving HBV virological response to tenofovir thereafter (3). Similar results have been reported from a recent study in HBV genotype A2 mono- and coinfected patients who again showed suboptimal response to tenofovir under this mutation (4). This however has not been confirmed in a recent study evaluating phenotypic assays where no decrease in sensitivity to adefovir in monoinfected patients was found in the presence of the L217R mutation (5).
In conclusion, viral suppression in HIV/HBV-coinfected patients treated with tenofovir is rapid (median time to virological control = 8.1 months) and sustained (median follow-up = 31 months). More than 98 % of the patients show a viral load < 2,000 IU/ml at the end of follow-up. Overall, only 4 blippers (2.6 %) and 2 (1.3 %) rebounders were identified in these small number of patients. In these patients HBV polymorphisms seem to occur, which however, have not yet been clearly associated with failure to nucleoside or nucleotide treatment of HBV. Clearly further sequencing and surveillance monitoring in patients with remaining HBV replication under selective pressure of anti-HBV drugs is warranted.
An interesting case report was presented from the Chelsea Westminster group in London reporting on a 45 year old individual with HIV/HBV coinfection, who in February 2006 was noted to have reverted hepatitis status from HBe-Ag negative and anti-HBV positive to HBe-Ag positive and anti-HBV negative with a corresponding serum HBV-DNA rise from 8,820 copies/ml to 33,6 Mio. copies/ml (6). At this time the patient was naive for HIV/HBV therapy. HIV disease was stable with a CD4-count of 613 cells/ml (36 %) and an HIV viral load of 14,462 copies/ml. In January 2008, Fibroscan demonstrated significant liver stiffness of 12.8 kPa (>F2 fibrosis) and a serum HBV-DNA of 662 Mio. copies/ml. The patient was offered treatment with either triple therapy for both HIV and HBV (tenofovir/3TC or FTC and efavirenz) or dual therapy for HBV alone (adefovir and telbivudine). HIV was still stable with a CD4-count of 640 cells/ml (36 %) and an HIV viral load of 8,650 copies/ml. With regard to his stable HIV-infection, the decision was made to start with dual HBV therapy only.
Two months after initiation of therapy, the individual had experienced a fall in serum HBV-DNA to 2,782 copies/ml and a suppressed HIV viral load to < 50 copies/ml. This was confirmed on repeated blood testing four weeks later. Telbivudine was discontinued five months after commencement, while adefovir was continued. At the time of discontinuation of telbivudine the HIV viral load had increased to 127 copies/ml rebounding to 3,903 copies/ml one month later. Three months following discontinuation, the patient agreed to be reexposed with telbivudine for a two week duration. On the day of starting telbivudine, his HIV viral load was 1,074 copies/ml, one week later 177 copies/ml and two weeks later 71 copies/ml. These surprising findings were explained by a possible activity of telbivudine against HIV-1. The possibility of adefovir having an effect was considered unlikely due to the fact that low-dose adefovir does not seem to have significant anti-HIV activity.
The finding of this case report, however, was contrasted by the results from a late breaker poster next to the case report (7). Within this poster researchers from Novartis summarized several in vitro experiments to investigate possible anti-HIV activity of telbivudine against strains BH-10 and NL4-3 in cells treated for 3-7 days using efavirenz as a control. Anti-HIV-1 activity was also determined against the drug sensitive reference HIV-1 strain (CNDO) and multi-drug resistance strain (MDRC4), 8 different wild type HIV clinical isolates representing different geographic locations with viral subtypes (A, B, BF, CND) and to HIV multi-drug resistant isolates to either telbivudine or entecavir (as a control) using the finalsenseTM HIV assay (VR001 Monogram Sciences) by measuring viral replication 48 hours after infection. Most interestingly, telbivudine did not exhibit any in vitro anti-HIV activity (IC50 values > 600 mmol) against any ever selected laboratory or clinical HIV isolates using a validated sensitive HIV phenotypic assay. In contrast both efavirenz and entecavir exhibited antiviral activity against the clinical HIV isolates. For efavirenz IC50 values range from 0.0009-0.0021 mmol and for entecavir from 7.62-15.09 mmol with the IC50 increased > 8-fold and HIV isolates harboring the M184V mutation. Clearly, the very comprehensive lab work does not suggest any signal for anti-HIV activity for telbivudine. Nevertheless taking into account the few number of patients treated with this drug and the lack of in vivo data, close surveillance of patients on sole anti-HBV agents with underlying HIV-infection is needed. In the case of unexpected declines in HIV-RNA or rebounds genotypic resistance testing is recommended.
Hepatitis C
With regard to hepatitis C interesting data was presented with regard to the impact of hepatitis C viremia and genotype on all cause mortality and liver-related death in patients with HIV and hepatitis C coinfection (8). So far most cohorts looking at the outcome of hepatitis C in HIV-coinfection were restricted by defining chronic hepatitis C soley by the presence of positive anti-HCV antibodies. EuroSIDA is the first cohort which has obtained hepatitis C viral load and genotype results within the cohort. Overall 1952 HIV/HCV coinfected patients were included into this analysis. 415 (21 %) of the patients were aviremic (HCV-RNA level < 650 U/ml), 716 (37 %) revealed low viremia with HCV-RNA reading > 650 but < 500,000 U/ml and 821 persons (42 %) were in the high viral load group with a HCV-RNA > 500,000 U/ml. In the multivariate analysis patients with a high hepatitis C viral load had a 1,77 times higher risk of death than people with a low HCV viral load (95 % CI: 1.11-2.82, p = 0.016). In contrast, HCV viral load had no major impact on response to HIV therapy. With regard to the genotype results, of the 1,537 people who could be genotyped 800 (52 %) had genotype 1, 218 (14 %) had genotype 2, 466 (30 %) had genotype 3 and 53 (3 %) showed genotype 4. Risk of death was higher for patients with genotype 1 compared to patients with genotype 3. In contrast response to antiviral therapy was significantly worse for the 2-4 genotypes although this difference reached statistical significance only for genotype 4 when response was measured as an HIV viral load < 500 copies (p = 0.011) or 50 % gain in CD4-cells (p = 0.010). Clearly this genotype analysis is limited by the relative low numbers for the more rare genotypes.
The different disease characteristics of HCV coinfection in patients with underlying HIV namely faster progression of liver fibrosis, higher mortality due to liver disease, and higher levels of HCV viremia particularly in patients prior to developing liver failure, may help to explain why HCV viral load could play a role in coinfection whereas so this association has not been observed in monoinfected hepatitis C cohorts. The unfavorable association between higher hepatitis C viral loads and risk for further liver disease events could have clinical implications and may underline, why hepatitis C specific treatment may be of particular importance in this patient group. As high hepatitis C viral load, however, is an unfavorable factor for achieving sustained virological response under interferon/ribavirin combination therapy, the need for the development of better acting alternative HCV drugs in the near future is emphasized.
With regard to a change of hepatitis C under highly active antiviral therapy a very interesting study was presented from the Swiss colleagues (9). T-cell responses against HCV core proteins were measured before and after initiating HAART in HIV/HCV coinfected individuals. Most interestingly a subgroup of patients could be identified with no HCV T-cell specific response prior to HAART initiation but clear signals of T-cell activity after extended HAART exposure. In these patients with improved immunologic response against HCV, a decline in HCV viral loads was noted over time. Under consideration that lower hepatitis C viral loads may not only have a more favorable outcome of liver disease but also are an independent predictor for better treatment response to interferon/ribavirin this could be of utmost clinical importance highlighting the need of HAART initiation prior to delivering HCV combination therapy.
Finally, a study on HCV treatment was reported from the American SLAM C (Sustained Long-term Antiviral Maintenance with pegylated interferon in HIV/HCV-coinfected patients trial, also known as ACTG A5178) (10). In this step 3 of the study patients who after initiating pegylated interferon-a + weight-adapted ribavirin who had achieved an early virological response defined as an at least 2 log decrease in hepatitis C viral load at week 12 after interferon/ribavirin combination therapy were followed-up for prolonged pegylated interferon + ribavirin for a total duration of 72 weeks. Previous non-responders had dropped ribavirin and had been randomized to either continuing the same dose of pegylated interferon monotherapy to test the hypothesis of interferon maintenance therapy as a possible intervention to prevent further fibrosis progression versus no therapy at all. These results were reported at last yearfs Retroviruses Conference and had shown no difference in fibrosis progression making this approach of unlikely benefit. In the 183 patients entering step 3, 169 opted to continue combination therapy for 72 weeks. Information about sustained virological response (SVR) was, however, only available for 146 participants at this point of time. Overall, 51 % of the patients who had achieved an early virological response went on to achieve SVR. The SVR rate was about twice as high among patients with HCV genotype 2 or 3 compared with genotype 1 or 4 (82 vs 42 %, respectively). Participants with previous unsuccessful interferon/ribavirin treatment attempts were half as likely to achieve SVR as treatment naive patients (30 vs 60 %, respectively). Furthermore, participants who had achieved complete EVR (negative HCV-PCR week 12) were nearly 4-times more likely to achieve SVR than those with partial EVR (62 vs 17 %, respectively). No difference in outcome was noted with regard to different ethnic or racial background. This study reinforces that treatment outcome in HIV/HCV coinfected patients is lower than in monoinfection and reconfirms the predictive value of early complete virological response as well as HCV genotype on SVR rates.
Summary for hepatitis B and C
• HIV/HBV coinfected patients receiving a tenofovir-containing regimen are most likely to become undetectable for HBV-replication (>98%)
• In the few HIV/HBV coinfected patients with ongoing HBV replication despite prolonged tenofovir therapy no clear HBV relevant mutation has been described so far. The development of polymorphisms not associated with HBV resistance however, has been observed
• A first case report suggests anti-HIV activity of telbivudine which however has not been confirmed in comprehensive in vitro studies. Close monitoring of telbivudine treated HIV/HBV patients not on HAART is warranted
• First cohort data suggests that higher HCV viral load levels may have a significant negative impact on overall survival and liver disease related death
• Although these findings need to be confirmed clinicians should pay attention to HCV viral load in the setting of HIV/HCV coinfection
• In a subset of patients with HIV/HCV coinfection successful HAART is accompanied by an improved HCV specific T-cell response which may lead to lower HCV viral load levels
• Early virological response (negative HCV-RNA at week 12) is associated with the best treatment outcome under peg-IFN/RBV therapy
1. Lacombe K, Gozlan J, Boyd A et al: HBV blippers and rebounders under treatment with tenofovir in HIV/HBV-coinfection. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 100
2. Sitou H, Thibault V, Valantin M et al: HBV resistance to tenofovir in HIV/HBV Co-infection. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 817
3. Schildgen O, Schewe CK, Vogel M et al: Successful therapy of hepatitis B with tenofovir in HIV-infected patients failing previous adefovir and lamivudine treatment. AIDS 2004; 8: 2325-2327
4. Bottecchia M, Madejon A, Sheldon J et al: Hepatitis B virus genotype A2 harbors and L217R polymorphism which may account for a low response to adefovir. J Antimicrobial Chemother 2008; 62: 626-627
5. Carrouee-Durantel S, Durantel D et al: Suboptimal response to adefovir dipivoxil therapy for chronic hepatitis B in nucleoside naive patients is not due to preexisting drug resistant mutants. Antiviral Ther 2008; 13: 381-388
6. Low E, Cox A, Atkins M, Nelson M: Telbivudine has activity against HIV. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 813a
7. Avila C, Karwowska S, Lai C, Evans T: Telbivudine has no in vitro activity against laboratory and clinical HIV-1 including five clades and drug-resistant clinical isolates. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 813b
8. Rockstroh JK, Peters L, Soriano V et al: High hepatitis C viremia is associated with an increased risk for mortality in HIV/HCV-coinfected individuals. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 101
9. Rohrbach J, Harcourt G, Gaudieri et al: Successful ART is associated with increasing HCV-specific T-cell responses. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 105
10. Chung R, Umbleja T, Butt A et al and the ACTG 5178 Team SLAM C (ACTG A5178): Role of early virological response in extended therapy with PEG-Interferon/weight-based ribavirin in HCV/HIV coinfection. 16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009; abstract 103